{
  "best_global_step": 26856,
  "best_metric": 1.0765974521636963,
  "best_model_checkpoint": "saves_multiple/lntuning/llama-3-8b-instruct/train_math_qa_42_1760637608/checkpoint-26856",
  "epoch": 20.0,
  "eval_steps": 6714,
  "global_step": 134280,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0007447125409591898,
      "grad_norm": 10.875,
      "learning_rate": 1.4894250819183796e-08,
      "loss": 1.0387,
      "num_input_tokens_seen": 2688,
      "step": 5
    },
    {
      "epoch": 0.0014894250819183796,
      "grad_norm": 14.1875,
      "learning_rate": 3.351206434316354e-08,
      "loss": 1.449,
      "num_input_tokens_seen": 5856,
      "step": 10
    },
    {
      "epoch": 0.002234137622877569,
      "grad_norm": 23.875,
      "learning_rate": 5.2129877867143284e-08,
      "loss": 1.161,
      "num_input_tokens_seen": 8512,
      "step": 15
    },
    {
      "epoch": 0.002978850163836759,
      "grad_norm": 15.625,
      "learning_rate": 7.074769139112303e-08,
      "loss": 1.1005,
      "num_input_tokens_seen": 11200,
      "step": 20
    },
    {
      "epoch": 0.0037235627047959487,
      "grad_norm": 9.9375,
      "learning_rate": 8.936550491510277e-08,
      "loss": 1.2429,
      "num_input_tokens_seen": 14336,
      "step": 25
    },
    {
      "epoch": 0.004468275245755138,
      "grad_norm": 12.375,
      "learning_rate": 1.0798331843908253e-07,
      "loss": 1.108,
      "num_input_tokens_seen": 17536,
      "step": 30
    },
    {
      "epoch": 0.005212987786714328,
      "grad_norm": 10.8125,
      "learning_rate": 1.2660113196306226e-07,
      "loss": 1.291,
      "num_input_tokens_seen": 20704,
      "step": 35
    },
    {
      "epoch": 0.005957700327673518,
      "grad_norm": 11.8125,
      "learning_rate": 1.45218945487042e-07,
      "loss": 0.998,
      "num_input_tokens_seen": 23552,
      "step": 40
    },
    {
      "epoch": 0.006702412868632708,
      "grad_norm": 8.0,
      "learning_rate": 1.6383675901102174e-07,
      "loss": 1.0665,
      "num_input_tokens_seen": 26528,
      "step": 45
    },
    {
      "epoch": 0.0074471254095918975,
      "grad_norm": 24.875,
      "learning_rate": 1.824545725350015e-07,
      "loss": 1.4834,
      "num_input_tokens_seen": 29152,
      "step": 50
    },
    {
      "epoch": 0.008191837950551088,
      "grad_norm": 17.0,
      "learning_rate": 2.0107238605898125e-07,
      "loss": 1.3611,
      "num_input_tokens_seen": 31872,
      "step": 55
    },
    {
      "epoch": 0.008936550491510277,
      "grad_norm": 11.4375,
      "learning_rate": 2.1969019958296101e-07,
      "loss": 1.0154,
      "num_input_tokens_seen": 34784,
      "step": 60
    },
    {
      "epoch": 0.009681263032469467,
      "grad_norm": 9.25,
      "learning_rate": 2.3830801310694073e-07,
      "loss": 1.2252,
      "num_input_tokens_seen": 37664,
      "step": 65
    },
    {
      "epoch": 0.010425975573428656,
      "grad_norm": 23.875,
      "learning_rate": 2.569258266309205e-07,
      "loss": 1.745,
      "num_input_tokens_seen": 40512,
      "step": 70
    },
    {
      "epoch": 0.011170688114387846,
      "grad_norm": 9.4375,
      "learning_rate": 2.755436401549002e-07,
      "loss": 1.3445,
      "num_input_tokens_seen": 43424,
      "step": 75
    },
    {
      "epoch": 0.011915400655347037,
      "grad_norm": 9.9375,
      "learning_rate": 2.9416145367888e-07,
      "loss": 1.1779,
      "num_input_tokens_seen": 46208,
      "step": 80
    },
    {
      "epoch": 0.012660113196306225,
      "grad_norm": 9.8125,
      "learning_rate": 3.1277926720285975e-07,
      "loss": 1.1807,
      "num_input_tokens_seen": 48832,
      "step": 85
    },
    {
      "epoch": 0.013404825737265416,
      "grad_norm": 13.25,
      "learning_rate": 3.3139708072683946e-07,
      "loss": 1.05,
      "num_input_tokens_seen": 51584,
      "step": 90
    },
    {
      "epoch": 0.014149538278224605,
      "grad_norm": 13.375,
      "learning_rate": 3.5001489425081923e-07,
      "loss": 1.0668,
      "num_input_tokens_seen": 54368,
      "step": 95
    },
    {
      "epoch": 0.014894250819183795,
      "grad_norm": 11.9375,
      "learning_rate": 3.6863270777479894e-07,
      "loss": 1.0834,
      "num_input_tokens_seen": 57248,
      "step": 100
    },
    {
      "epoch": 0.015638963360142984,
      "grad_norm": 19.25,
      "learning_rate": 3.872505212987787e-07,
      "loss": 1.3761,
      "num_input_tokens_seen": 60352,
      "step": 105
    },
    {
      "epoch": 0.016383675901102176,
      "grad_norm": 13.0,
      "learning_rate": 4.058683348227585e-07,
      "loss": 1.2995,
      "num_input_tokens_seen": 63072,
      "step": 110
    },
    {
      "epoch": 0.017128388442061365,
      "grad_norm": 24.75,
      "learning_rate": 4.244861483467382e-07,
      "loss": 1.4206,
      "num_input_tokens_seen": 65952,
      "step": 115
    },
    {
      "epoch": 0.017873100983020553,
      "grad_norm": 10.375,
      "learning_rate": 4.431039618707179e-07,
      "loss": 0.9397,
      "num_input_tokens_seen": 68672,
      "step": 120
    },
    {
      "epoch": 0.018617813523979745,
      "grad_norm": 13.4375,
      "learning_rate": 4.617217753946977e-07,
      "loss": 1.2722,
      "num_input_tokens_seen": 71488,
      "step": 125
    },
    {
      "epoch": 0.019362526064938934,
      "grad_norm": 19.125,
      "learning_rate": 4.803395889186774e-07,
      "loss": 1.1991,
      "num_input_tokens_seen": 74304,
      "step": 130
    },
    {
      "epoch": 0.020107238605898123,
      "grad_norm": 21.875,
      "learning_rate": 4.989574024426572e-07,
      "loss": 1.3744,
      "num_input_tokens_seen": 77120,
      "step": 135
    },
    {
      "epoch": 0.02085195114685731,
      "grad_norm": 14.625,
      "learning_rate": 5.17575215966637e-07,
      "loss": 1.1466,
      "num_input_tokens_seen": 80000,
      "step": 140
    },
    {
      "epoch": 0.021596663687816504,
      "grad_norm": 23.25,
      "learning_rate": 5.361930294906167e-07,
      "loss": 1.3971,
      "num_input_tokens_seen": 82912,
      "step": 145
    },
    {
      "epoch": 0.022341376228775692,
      "grad_norm": 16.75,
      "learning_rate": 5.548108430145964e-07,
      "loss": 1.1791,
      "num_input_tokens_seen": 85824,
      "step": 150
    },
    {
      "epoch": 0.02308608876973488,
      "grad_norm": 15.3125,
      "learning_rate": 5.734286565385761e-07,
      "loss": 1.3559,
      "num_input_tokens_seen": 88640,
      "step": 155
    },
    {
      "epoch": 0.023830801310694073,
      "grad_norm": 20.5,
      "learning_rate": 5.920464700625559e-07,
      "loss": 1.3246,
      "num_input_tokens_seen": 91584,
      "step": 160
    },
    {
      "epoch": 0.024575513851653262,
      "grad_norm": 9.3125,
      "learning_rate": 6.106642835865357e-07,
      "loss": 1.3633,
      "num_input_tokens_seen": 94592,
      "step": 165
    },
    {
      "epoch": 0.02532022639261245,
      "grad_norm": 20.125,
      "learning_rate": 6.292820971105154e-07,
      "loss": 1.4197,
      "num_input_tokens_seen": 97760,
      "step": 170
    },
    {
      "epoch": 0.026064938933571643,
      "grad_norm": 12.4375,
      "learning_rate": 6.478999106344952e-07,
      "loss": 1.3163,
      "num_input_tokens_seen": 100960,
      "step": 175
    },
    {
      "epoch": 0.02680965147453083,
      "grad_norm": 7.03125,
      "learning_rate": 6.665177241584749e-07,
      "loss": 0.9659,
      "num_input_tokens_seen": 104320,
      "step": 180
    },
    {
      "epoch": 0.02755436401549002,
      "grad_norm": 14.75,
      "learning_rate": 6.851355376824546e-07,
      "loss": 1.0073,
      "num_input_tokens_seen": 107008,
      "step": 185
    },
    {
      "epoch": 0.02829907655644921,
      "grad_norm": 20.875,
      "learning_rate": 7.037533512064343e-07,
      "loss": 1.2512,
      "num_input_tokens_seen": 110048,
      "step": 190
    },
    {
      "epoch": 0.0290437890974084,
      "grad_norm": 13.0625,
      "learning_rate": 7.223711647304142e-07,
      "loss": 1.0458,
      "num_input_tokens_seen": 112896,
      "step": 195
    },
    {
      "epoch": 0.02978850163836759,
      "grad_norm": 9.9375,
      "learning_rate": 7.409889782543939e-07,
      "loss": 1.0537,
      "num_input_tokens_seen": 116192,
      "step": 200
    },
    {
      "epoch": 0.03053321417932678,
      "grad_norm": 7.8125,
      "learning_rate": 7.596067917783736e-07,
      "loss": 1.2154,
      "num_input_tokens_seen": 119200,
      "step": 205
    },
    {
      "epoch": 0.03127792672028597,
      "grad_norm": 17.125,
      "learning_rate": 7.782246053023533e-07,
      "loss": 1.2229,
      "num_input_tokens_seen": 122144,
      "step": 210
    },
    {
      "epoch": 0.032022639261245156,
      "grad_norm": 17.0,
      "learning_rate": 7.96842418826333e-07,
      "loss": 1.4295,
      "num_input_tokens_seen": 125312,
      "step": 215
    },
    {
      "epoch": 0.03276735180220435,
      "grad_norm": 8.6875,
      "learning_rate": 8.154602323503128e-07,
      "loss": 1.2853,
      "num_input_tokens_seen": 128160,
      "step": 220
    },
    {
      "epoch": 0.03351206434316354,
      "grad_norm": 20.875,
      "learning_rate": 8.340780458742925e-07,
      "loss": 1.3309,
      "num_input_tokens_seen": 131168,
      "step": 225
    },
    {
      "epoch": 0.03425677688412273,
      "grad_norm": 18.0,
      "learning_rate": 8.526958593982724e-07,
      "loss": 1.1224,
      "num_input_tokens_seen": 133888,
      "step": 230
    },
    {
      "epoch": 0.03500148942508192,
      "grad_norm": 9.625,
      "learning_rate": 8.713136729222521e-07,
      "loss": 1.0293,
      "num_input_tokens_seen": 136608,
      "step": 235
    },
    {
      "epoch": 0.035746201966041107,
      "grad_norm": 11.3125,
      "learning_rate": 8.899314864462318e-07,
      "loss": 1.2389,
      "num_input_tokens_seen": 139840,
      "step": 240
    },
    {
      "epoch": 0.036490914507000295,
      "grad_norm": 12.4375,
      "learning_rate": 9.085492999702115e-07,
      "loss": 1.2513,
      "num_input_tokens_seen": 143072,
      "step": 245
    },
    {
      "epoch": 0.03723562704795949,
      "grad_norm": 14.375,
      "learning_rate": 9.271671134941912e-07,
      "loss": 0.7685,
      "num_input_tokens_seen": 145920,
      "step": 250
    },
    {
      "epoch": 0.03798033958891868,
      "grad_norm": 13.0625,
      "learning_rate": 9.457849270181709e-07,
      "loss": 1.0393,
      "num_input_tokens_seen": 149088,
      "step": 255
    },
    {
      "epoch": 0.03872505212987787,
      "grad_norm": 11.5,
      "learning_rate": 9.644027405421507e-07,
      "loss": 1.3198,
      "num_input_tokens_seen": 152704,
      "step": 260
    },
    {
      "epoch": 0.03946976467083706,
      "grad_norm": 8.1875,
      "learning_rate": 9.830205540661306e-07,
      "loss": 1.0541,
      "num_input_tokens_seen": 155552,
      "step": 265
    },
    {
      "epoch": 0.040214477211796246,
      "grad_norm": 17.625,
      "learning_rate": 1.0016383675901103e-06,
      "loss": 1.4659,
      "num_input_tokens_seen": 158496,
      "step": 270
    },
    {
      "epoch": 0.040959189752755434,
      "grad_norm": 10.0,
      "learning_rate": 1.02025618111409e-06,
      "loss": 1.2722,
      "num_input_tokens_seen": 161536,
      "step": 275
    },
    {
      "epoch": 0.04170390229371462,
      "grad_norm": 18.875,
      "learning_rate": 1.0388739946380697e-06,
      "loss": 1.26,
      "num_input_tokens_seen": 164384,
      "step": 280
    },
    {
      "epoch": 0.04244861483467382,
      "grad_norm": 12.0625,
      "learning_rate": 1.0574918081620494e-06,
      "loss": 1.2212,
      "num_input_tokens_seen": 167200,
      "step": 285
    },
    {
      "epoch": 0.04319332737563301,
      "grad_norm": 13.75,
      "learning_rate": 1.0761096216860292e-06,
      "loss": 1.3081,
      "num_input_tokens_seen": 170208,
      "step": 290
    },
    {
      "epoch": 0.043938039916592196,
      "grad_norm": 15.0,
      "learning_rate": 1.0947274352100089e-06,
      "loss": 1.1689,
      "num_input_tokens_seen": 173408,
      "step": 295
    },
    {
      "epoch": 0.044682752457551385,
      "grad_norm": 17.375,
      "learning_rate": 1.1133452487339888e-06,
      "loss": 1.1064,
      "num_input_tokens_seen": 176064,
      "step": 300
    },
    {
      "epoch": 0.045427464998510574,
      "grad_norm": 19.0,
      "learning_rate": 1.1319630622579685e-06,
      "loss": 1.109,
      "num_input_tokens_seen": 178848,
      "step": 305
    },
    {
      "epoch": 0.04617217753946976,
      "grad_norm": 18.375,
      "learning_rate": 1.1505808757819482e-06,
      "loss": 1.2065,
      "num_input_tokens_seen": 181856,
      "step": 310
    },
    {
      "epoch": 0.04691689008042895,
      "grad_norm": 9.4375,
      "learning_rate": 1.169198689305928e-06,
      "loss": 1.0128,
      "num_input_tokens_seen": 184896,
      "step": 315
    },
    {
      "epoch": 0.04766160262138815,
      "grad_norm": 16.625,
      "learning_rate": 1.1878165028299077e-06,
      "loss": 1.146,
      "num_input_tokens_seen": 187808,
      "step": 320
    },
    {
      "epoch": 0.048406315162347335,
      "grad_norm": 9.0625,
      "learning_rate": 1.2064343163538874e-06,
      "loss": 0.8166,
      "num_input_tokens_seen": 190880,
      "step": 325
    },
    {
      "epoch": 0.049151027703306524,
      "grad_norm": 13.8125,
      "learning_rate": 1.2250521298778673e-06,
      "loss": 1.3636,
      "num_input_tokens_seen": 193536,
      "step": 330
    },
    {
      "epoch": 0.04989574024426571,
      "grad_norm": 18.5,
      "learning_rate": 1.243669943401847e-06,
      "loss": 1.4758,
      "num_input_tokens_seen": 196448,
      "step": 335
    },
    {
      "epoch": 0.0506404527852249,
      "grad_norm": 16.5,
      "learning_rate": 1.2622877569258267e-06,
      "loss": 0.9189,
      "num_input_tokens_seen": 199136,
      "step": 340
    },
    {
      "epoch": 0.05138516532618409,
      "grad_norm": 15.125,
      "learning_rate": 1.2809055704498064e-06,
      "loss": 1.4634,
      "num_input_tokens_seen": 202240,
      "step": 345
    },
    {
      "epoch": 0.052129877867143286,
      "grad_norm": 16.25,
      "learning_rate": 1.2995233839737862e-06,
      "loss": 1.166,
      "num_input_tokens_seen": 204896,
      "step": 350
    },
    {
      "epoch": 0.052874590408102475,
      "grad_norm": 11.0625,
      "learning_rate": 1.3181411974977659e-06,
      "loss": 1.1617,
      "num_input_tokens_seen": 207552,
      "step": 355
    },
    {
      "epoch": 0.05361930294906166,
      "grad_norm": 9.75,
      "learning_rate": 1.3367590110217456e-06,
      "loss": 1.0501,
      "num_input_tokens_seen": 210240,
      "step": 360
    },
    {
      "epoch": 0.05436401549002085,
      "grad_norm": 13.875,
      "learning_rate": 1.3553768245457255e-06,
      "loss": 0.9849,
      "num_input_tokens_seen": 213024,
      "step": 365
    },
    {
      "epoch": 0.05510872803098004,
      "grad_norm": 9.9375,
      "learning_rate": 1.3739946380697052e-06,
      "loss": 0.9696,
      "num_input_tokens_seen": 215840,
      "step": 370
    },
    {
      "epoch": 0.05585344057193923,
      "grad_norm": 18.375,
      "learning_rate": 1.392612451593685e-06,
      "loss": 1.0323,
      "num_input_tokens_seen": 218528,
      "step": 375
    },
    {
      "epoch": 0.05659815311289842,
      "grad_norm": 15.625,
      "learning_rate": 1.4112302651176647e-06,
      "loss": 1.3477,
      "num_input_tokens_seen": 221824,
      "step": 380
    },
    {
      "epoch": 0.057342865653857614,
      "grad_norm": 8.625,
      "learning_rate": 1.4298480786416444e-06,
      "loss": 0.9682,
      "num_input_tokens_seen": 224480,
      "step": 385
    },
    {
      "epoch": 0.0580875781948168,
      "grad_norm": 13.0625,
      "learning_rate": 1.448465892165624e-06,
      "loss": 0.8177,
      "num_input_tokens_seen": 227296,
      "step": 390
    },
    {
      "epoch": 0.05883229073577599,
      "grad_norm": 19.375,
      "learning_rate": 1.4670837056896038e-06,
      "loss": 1.3916,
      "num_input_tokens_seen": 229920,
      "step": 395
    },
    {
      "epoch": 0.05957700327673518,
      "grad_norm": 8.9375,
      "learning_rate": 1.4857015192135837e-06,
      "loss": 0.9775,
      "num_input_tokens_seen": 232512,
      "step": 400
    },
    {
      "epoch": 0.06032171581769437,
      "grad_norm": 11.375,
      "learning_rate": 1.5043193327375634e-06,
      "loss": 1.1244,
      "num_input_tokens_seen": 235584,
      "step": 405
    },
    {
      "epoch": 0.06106642835865356,
      "grad_norm": 14.3125,
      "learning_rate": 1.5229371462615432e-06,
      "loss": 1.4792,
      "num_input_tokens_seen": 238496,
      "step": 410
    },
    {
      "epoch": 0.06181114089961275,
      "grad_norm": 10.1875,
      "learning_rate": 1.5415549597855229e-06,
      "loss": 1.1367,
      "num_input_tokens_seen": 241376,
      "step": 415
    },
    {
      "epoch": 0.06255585344057193,
      "grad_norm": 13.0,
      "learning_rate": 1.5601727733095026e-06,
      "loss": 1.0573,
      "num_input_tokens_seen": 244320,
      "step": 420
    },
    {
      "epoch": 0.06330056598153112,
      "grad_norm": 13.3125,
      "learning_rate": 1.5787905868334823e-06,
      "loss": 1.5503,
      "num_input_tokens_seen": 247488,
      "step": 425
    },
    {
      "epoch": 0.06404527852249031,
      "grad_norm": 13.1875,
      "learning_rate": 1.597408400357462e-06,
      "loss": 1.6603,
      "num_input_tokens_seen": 250208,
      "step": 430
    },
    {
      "epoch": 0.06478999106344951,
      "grad_norm": 11.1875,
      "learning_rate": 1.6160262138814417e-06,
      "loss": 1.3359,
      "num_input_tokens_seen": 253120,
      "step": 435
    },
    {
      "epoch": 0.0655347036044087,
      "grad_norm": 27.375,
      "learning_rate": 1.6346440274054214e-06,
      "loss": 1.0745,
      "num_input_tokens_seen": 256032,
      "step": 440
    },
    {
      "epoch": 0.06627941614536789,
      "grad_norm": 10.625,
      "learning_rate": 1.6532618409294012e-06,
      "loss": 1.4314,
      "num_input_tokens_seen": 258912,
      "step": 445
    },
    {
      "epoch": 0.06702412868632708,
      "grad_norm": 9.8125,
      "learning_rate": 1.6718796544533813e-06,
      "loss": 1.1041,
      "num_input_tokens_seen": 261728,
      "step": 450
    },
    {
      "epoch": 0.06776884122728627,
      "grad_norm": 19.625,
      "learning_rate": 1.690497467977361e-06,
      "loss": 0.8706,
      "num_input_tokens_seen": 264608,
      "step": 455
    },
    {
      "epoch": 0.06851355376824546,
      "grad_norm": 13.8125,
      "learning_rate": 1.7091152815013407e-06,
      "loss": 1.1232,
      "num_input_tokens_seen": 267744,
      "step": 460
    },
    {
      "epoch": 0.06925826630920465,
      "grad_norm": 11.25,
      "learning_rate": 1.7277330950253204e-06,
      "loss": 1.2115,
      "num_input_tokens_seen": 270912,
      "step": 465
    },
    {
      "epoch": 0.07000297885016384,
      "grad_norm": 26.375,
      "learning_rate": 1.7463509085493002e-06,
      "loss": 1.7157,
      "num_input_tokens_seen": 273824,
      "step": 470
    },
    {
      "epoch": 0.07074769139112302,
      "grad_norm": 25.625,
      "learning_rate": 1.7649687220732799e-06,
      "loss": 1.2459,
      "num_input_tokens_seen": 276960,
      "step": 475
    },
    {
      "epoch": 0.07149240393208221,
      "grad_norm": 8.3125,
      "learning_rate": 1.7835865355972596e-06,
      "loss": 0.8739,
      "num_input_tokens_seen": 279712,
      "step": 480
    },
    {
      "epoch": 0.0722371164730414,
      "grad_norm": 33.0,
      "learning_rate": 1.8022043491212393e-06,
      "loss": 1.3137,
      "num_input_tokens_seen": 282592,
      "step": 485
    },
    {
      "epoch": 0.07298182901400059,
      "grad_norm": 18.375,
      "learning_rate": 1.820822162645219e-06,
      "loss": 1.7034,
      "num_input_tokens_seen": 285472,
      "step": 490
    },
    {
      "epoch": 0.07372654155495978,
      "grad_norm": 13.3125,
      "learning_rate": 1.8394399761691987e-06,
      "loss": 1.133,
      "num_input_tokens_seen": 288544,
      "step": 495
    },
    {
      "epoch": 0.07447125409591898,
      "grad_norm": 12.9375,
      "learning_rate": 1.8580577896931784e-06,
      "loss": 1.036,
      "num_input_tokens_seen": 291552,
      "step": 500
    },
    {
      "epoch": 0.07521596663687817,
      "grad_norm": 15.875,
      "learning_rate": 1.8766756032171582e-06,
      "loss": 1.0359,
      "num_input_tokens_seen": 294432,
      "step": 505
    },
    {
      "epoch": 0.07596067917783736,
      "grad_norm": 11.1875,
      "learning_rate": 1.8952934167411379e-06,
      "loss": 1.2926,
      "num_input_tokens_seen": 297344,
      "step": 510
    },
    {
      "epoch": 0.07670539171879655,
      "grad_norm": 20.625,
      "learning_rate": 1.913911230265118e-06,
      "loss": 1.3271,
      "num_input_tokens_seen": 300416,
      "step": 515
    },
    {
      "epoch": 0.07745010425975574,
      "grad_norm": 4.15625,
      "learning_rate": 1.9325290437890977e-06,
      "loss": 1.1026,
      "num_input_tokens_seen": 303520,
      "step": 520
    },
    {
      "epoch": 0.07819481680071493,
      "grad_norm": 17.875,
      "learning_rate": 1.9511468573130772e-06,
      "loss": 1.2202,
      "num_input_tokens_seen": 306208,
      "step": 525
    },
    {
      "epoch": 0.07893952934167411,
      "grad_norm": 15.8125,
      "learning_rate": 1.969764670837057e-06,
      "loss": 1.2928,
      "num_input_tokens_seen": 309504,
      "step": 530
    },
    {
      "epoch": 0.0796842418826333,
      "grad_norm": 14.375,
      "learning_rate": 1.9883824843610367e-06,
      "loss": 1.2325,
      "num_input_tokens_seen": 312384,
      "step": 535
    },
    {
      "epoch": 0.08042895442359249,
      "grad_norm": 9.5,
      "learning_rate": 2.0070002978850166e-06,
      "loss": 1.4715,
      "num_input_tokens_seen": 315072,
      "step": 540
    },
    {
      "epoch": 0.08117366696455168,
      "grad_norm": 22.375,
      "learning_rate": 2.025618111408996e-06,
      "loss": 1.3269,
      "num_input_tokens_seen": 318080,
      "step": 545
    },
    {
      "epoch": 0.08191837950551087,
      "grad_norm": 11.5,
      "learning_rate": 2.044235924932976e-06,
      "loss": 1.0733,
      "num_input_tokens_seen": 320960,
      "step": 550
    },
    {
      "epoch": 0.08266309204647006,
      "grad_norm": 18.0,
      "learning_rate": 2.0628537384569555e-06,
      "loss": 1.1503,
      "num_input_tokens_seen": 324160,
      "step": 555
    },
    {
      "epoch": 0.08340780458742925,
      "grad_norm": 17.25,
      "learning_rate": 2.0814715519809354e-06,
      "loss": 1.1256,
      "num_input_tokens_seen": 326784,
      "step": 560
    },
    {
      "epoch": 0.08415251712838845,
      "grad_norm": 7.0625,
      "learning_rate": 2.100089365504915e-06,
      "loss": 0.7965,
      "num_input_tokens_seen": 329696,
      "step": 565
    },
    {
      "epoch": 0.08489722966934764,
      "grad_norm": 15.4375,
      "learning_rate": 2.118707179028895e-06,
      "loss": 1.3919,
      "num_input_tokens_seen": 333280,
      "step": 570
    },
    {
      "epoch": 0.08564194221030683,
      "grad_norm": 7.65625,
      "learning_rate": 2.1373249925528744e-06,
      "loss": 1.2749,
      "num_input_tokens_seen": 336160,
      "step": 575
    },
    {
      "epoch": 0.08638665475126601,
      "grad_norm": 24.75,
      "learning_rate": 2.1559428060768547e-06,
      "loss": 1.153,
      "num_input_tokens_seen": 338880,
      "step": 580
    },
    {
      "epoch": 0.0871313672922252,
      "grad_norm": 13.4375,
      "learning_rate": 2.1745606196008342e-06,
      "loss": 1.2488,
      "num_input_tokens_seen": 341824,
      "step": 585
    },
    {
      "epoch": 0.08787607983318439,
      "grad_norm": 16.25,
      "learning_rate": 2.193178433124814e-06,
      "loss": 1.1094,
      "num_input_tokens_seen": 344672,
      "step": 590
    },
    {
      "epoch": 0.08862079237414358,
      "grad_norm": 10.9375,
      "learning_rate": 2.2117962466487937e-06,
      "loss": 0.9609,
      "num_input_tokens_seen": 347424,
      "step": 595
    },
    {
      "epoch": 0.08936550491510277,
      "grad_norm": 7.21875,
      "learning_rate": 2.2304140601727736e-06,
      "loss": 1.1394,
      "num_input_tokens_seen": 350272,
      "step": 600
    },
    {
      "epoch": 0.09011021745606196,
      "grad_norm": 13.875,
      "learning_rate": 2.249031873696753e-06,
      "loss": 0.9438,
      "num_input_tokens_seen": 353248,
      "step": 605
    },
    {
      "epoch": 0.09085492999702115,
      "grad_norm": 19.625,
      "learning_rate": 2.267649687220733e-06,
      "loss": 1.2439,
      "num_input_tokens_seen": 356096,
      "step": 610
    },
    {
      "epoch": 0.09159964253798034,
      "grad_norm": 7.3125,
      "learning_rate": 2.2862675007447125e-06,
      "loss": 1.5649,
      "num_input_tokens_seen": 358784,
      "step": 615
    },
    {
      "epoch": 0.09234435507893952,
      "grad_norm": 14.3125,
      "learning_rate": 2.3048853142686924e-06,
      "loss": 1.306,
      "num_input_tokens_seen": 361536,
      "step": 620
    },
    {
      "epoch": 0.09308906761989871,
      "grad_norm": 19.75,
      "learning_rate": 2.323503127792672e-06,
      "loss": 1.3571,
      "num_input_tokens_seen": 364512,
      "step": 625
    },
    {
      "epoch": 0.0938337801608579,
      "grad_norm": 12.6875,
      "learning_rate": 2.342120941316652e-06,
      "loss": 0.8935,
      "num_input_tokens_seen": 367424,
      "step": 630
    },
    {
      "epoch": 0.0945784927018171,
      "grad_norm": 9.125,
      "learning_rate": 2.3607387548406314e-06,
      "loss": 1.1146,
      "num_input_tokens_seen": 370208,
      "step": 635
    },
    {
      "epoch": 0.0953232052427763,
      "grad_norm": 6.8125,
      "learning_rate": 2.3793565683646113e-06,
      "loss": 1.3821,
      "num_input_tokens_seen": 373248,
      "step": 640
    },
    {
      "epoch": 0.09606791778373548,
      "grad_norm": 14.75,
      "learning_rate": 2.3979743818885912e-06,
      "loss": 1.4203,
      "num_input_tokens_seen": 376192,
      "step": 645
    },
    {
      "epoch": 0.09681263032469467,
      "grad_norm": 10.5,
      "learning_rate": 2.416592195412571e-06,
      "loss": 1.1066,
      "num_input_tokens_seen": 378912,
      "step": 650
    },
    {
      "epoch": 0.09755734286565386,
      "grad_norm": 18.5,
      "learning_rate": 2.4352100089365507e-06,
      "loss": 1.24,
      "num_input_tokens_seen": 381984,
      "step": 655
    },
    {
      "epoch": 0.09830205540661305,
      "grad_norm": 21.625,
      "learning_rate": 2.4538278224605306e-06,
      "loss": 1.844,
      "num_input_tokens_seen": 384800,
      "step": 660
    },
    {
      "epoch": 0.09904676794757224,
      "grad_norm": 13.0,
      "learning_rate": 2.47244563598451e-06,
      "loss": 1.1967,
      "num_input_tokens_seen": 387712,
      "step": 665
    },
    {
      "epoch": 0.09979148048853143,
      "grad_norm": 15.75,
      "learning_rate": 2.49106344950849e-06,
      "loss": 0.8664,
      "num_input_tokens_seen": 390656,
      "step": 670
    },
    {
      "epoch": 0.10053619302949061,
      "grad_norm": 8.8125,
      "learning_rate": 2.5096812630324695e-06,
      "loss": 1.6785,
      "num_input_tokens_seen": 393504,
      "step": 675
    },
    {
      "epoch": 0.1012809055704498,
      "grad_norm": 10.625,
      "learning_rate": 2.5282990765564494e-06,
      "loss": 0.881,
      "num_input_tokens_seen": 396256,
      "step": 680
    },
    {
      "epoch": 0.10202561811140899,
      "grad_norm": 15.0,
      "learning_rate": 2.546916890080429e-06,
      "loss": 0.9887,
      "num_input_tokens_seen": 399328,
      "step": 685
    },
    {
      "epoch": 0.10277033065236818,
      "grad_norm": 11.1875,
      "learning_rate": 2.565534703604409e-06,
      "loss": 1.1556,
      "num_input_tokens_seen": 402080,
      "step": 690
    },
    {
      "epoch": 0.10351504319332737,
      "grad_norm": 12.3125,
      "learning_rate": 2.5841525171283884e-06,
      "loss": 1.1915,
      "num_input_tokens_seen": 404896,
      "step": 695
    },
    {
      "epoch": 0.10425975573428657,
      "grad_norm": 13.4375,
      "learning_rate": 2.6027703306523683e-06,
      "loss": 1.5508,
      "num_input_tokens_seen": 407552,
      "step": 700
    },
    {
      "epoch": 0.10500446827524576,
      "grad_norm": 25.375,
      "learning_rate": 2.621388144176348e-06,
      "loss": 1.589,
      "num_input_tokens_seen": 410592,
      "step": 705
    },
    {
      "epoch": 0.10574918081620495,
      "grad_norm": 17.125,
      "learning_rate": 2.6400059577003277e-06,
      "loss": 1.2631,
      "num_input_tokens_seen": 413376,
      "step": 710
    },
    {
      "epoch": 0.10649389335716414,
      "grad_norm": 11.125,
      "learning_rate": 2.6586237712243077e-06,
      "loss": 0.9704,
      "num_input_tokens_seen": 416160,
      "step": 715
    },
    {
      "epoch": 0.10723860589812333,
      "grad_norm": 20.0,
      "learning_rate": 2.6772415847482876e-06,
      "loss": 1.3365,
      "num_input_tokens_seen": 419168,
      "step": 720
    },
    {
      "epoch": 0.10798331843908252,
      "grad_norm": 13.1875,
      "learning_rate": 2.695859398272267e-06,
      "loss": 1.294,
      "num_input_tokens_seen": 422112,
      "step": 725
    },
    {
      "epoch": 0.1087280309800417,
      "grad_norm": 21.25,
      "learning_rate": 2.714477211796247e-06,
      "loss": 1.7873,
      "num_input_tokens_seen": 424960,
      "step": 730
    },
    {
      "epoch": 0.10947274352100089,
      "grad_norm": 5.65625,
      "learning_rate": 2.7330950253202265e-06,
      "loss": 1.2275,
      "num_input_tokens_seen": 427872,
      "step": 735
    },
    {
      "epoch": 0.11021745606196008,
      "grad_norm": 8.0625,
      "learning_rate": 2.7517128388442064e-06,
      "loss": 1.0824,
      "num_input_tokens_seen": 431008,
      "step": 740
    },
    {
      "epoch": 0.11096216860291927,
      "grad_norm": 16.125,
      "learning_rate": 2.770330652368186e-06,
      "loss": 1.166,
      "num_input_tokens_seen": 434016,
      "step": 745
    },
    {
      "epoch": 0.11170688114387846,
      "grad_norm": 10.25,
      "learning_rate": 2.788948465892166e-06,
      "loss": 0.9849,
      "num_input_tokens_seen": 436736,
      "step": 750
    },
    {
      "epoch": 0.11245159368483765,
      "grad_norm": 11.125,
      "learning_rate": 2.8075662794161454e-06,
      "loss": 1.0808,
      "num_input_tokens_seen": 439776,
      "step": 755
    },
    {
      "epoch": 0.11319630622579684,
      "grad_norm": 18.5,
      "learning_rate": 2.8261840929401253e-06,
      "loss": 1.2257,
      "num_input_tokens_seen": 442592,
      "step": 760
    },
    {
      "epoch": 0.11394101876675604,
      "grad_norm": 11.625,
      "learning_rate": 2.844801906464105e-06,
      "loss": 1.0267,
      "num_input_tokens_seen": 445568,
      "step": 765
    },
    {
      "epoch": 0.11468573130771523,
      "grad_norm": 19.0,
      "learning_rate": 2.8634197199880847e-06,
      "loss": 1.1727,
      "num_input_tokens_seen": 448480,
      "step": 770
    },
    {
      "epoch": 0.11543044384867442,
      "grad_norm": 15.375,
      "learning_rate": 2.8820375335120642e-06,
      "loss": 1.126,
      "num_input_tokens_seen": 451168,
      "step": 775
    },
    {
      "epoch": 0.1161751563896336,
      "grad_norm": 16.0,
      "learning_rate": 2.9006553470360446e-06,
      "loss": 1.1043,
      "num_input_tokens_seen": 454496,
      "step": 780
    },
    {
      "epoch": 0.1169198689305928,
      "grad_norm": 8.125,
      "learning_rate": 2.919273160560024e-06,
      "loss": 1.2237,
      "num_input_tokens_seen": 457824,
      "step": 785
    },
    {
      "epoch": 0.11766458147155198,
      "grad_norm": 9.4375,
      "learning_rate": 2.937890974084004e-06,
      "loss": 1.2009,
      "num_input_tokens_seen": 460768,
      "step": 790
    },
    {
      "epoch": 0.11840929401251117,
      "grad_norm": 6.375,
      "learning_rate": 2.9565087876079835e-06,
      "loss": 1.0531,
      "num_input_tokens_seen": 463680,
      "step": 795
    },
    {
      "epoch": 0.11915400655347036,
      "grad_norm": 11.75,
      "learning_rate": 2.9751266011319634e-06,
      "loss": 1.2919,
      "num_input_tokens_seen": 466624,
      "step": 800
    },
    {
      "epoch": 0.11989871909442955,
      "grad_norm": 14.625,
      "learning_rate": 2.993744414655943e-06,
      "loss": 1.3731,
      "num_input_tokens_seen": 469568,
      "step": 805
    },
    {
      "epoch": 0.12064343163538874,
      "grad_norm": 8.8125,
      "learning_rate": 3.012362228179923e-06,
      "loss": 1.1633,
      "num_input_tokens_seen": 472352,
      "step": 810
    },
    {
      "epoch": 0.12138814417634793,
      "grad_norm": 15.5,
      "learning_rate": 3.0309800417039024e-06,
      "loss": 1.5032,
      "num_input_tokens_seen": 475872,
      "step": 815
    },
    {
      "epoch": 0.12213285671730711,
      "grad_norm": 12.625,
      "learning_rate": 3.0495978552278823e-06,
      "loss": 0.928,
      "num_input_tokens_seen": 478912,
      "step": 820
    },
    {
      "epoch": 0.1228775692582663,
      "grad_norm": 14.25,
      "learning_rate": 3.068215668751862e-06,
      "loss": 1.1487,
      "num_input_tokens_seen": 481600,
      "step": 825
    },
    {
      "epoch": 0.1236222817992255,
      "grad_norm": 16.25,
      "learning_rate": 3.0868334822758417e-06,
      "loss": 1.2298,
      "num_input_tokens_seen": 484576,
      "step": 830
    },
    {
      "epoch": 0.1243669943401847,
      "grad_norm": 8.5,
      "learning_rate": 3.1054512957998212e-06,
      "loss": 1.017,
      "num_input_tokens_seen": 487520,
      "step": 835
    },
    {
      "epoch": 0.12511170688114387,
      "grad_norm": 13.4375,
      "learning_rate": 3.124069109323801e-06,
      "loss": 0.884,
      "num_input_tokens_seen": 490496,
      "step": 840
    },
    {
      "epoch": 0.12585641942210307,
      "grad_norm": 11.5625,
      "learning_rate": 3.142686922847781e-06,
      "loss": 1.1836,
      "num_input_tokens_seen": 493408,
      "step": 845
    },
    {
      "epoch": 0.12660113196306225,
      "grad_norm": 11.8125,
      "learning_rate": 3.1613047363717606e-06,
      "loss": 1.504,
      "num_input_tokens_seen": 496096,
      "step": 850
    },
    {
      "epoch": 0.12734584450402145,
      "grad_norm": 11.4375,
      "learning_rate": 3.1799225498957405e-06,
      "loss": 0.9927,
      "num_input_tokens_seen": 498912,
      "step": 855
    },
    {
      "epoch": 0.12809055704498062,
      "grad_norm": 18.25,
      "learning_rate": 3.19854036341972e-06,
      "loss": 1.1825,
      "num_input_tokens_seen": 501856,
      "step": 860
    },
    {
      "epoch": 0.12883526958593983,
      "grad_norm": 14.875,
      "learning_rate": 3.2171581769437e-06,
      "loss": 1.3061,
      "num_input_tokens_seen": 504736,
      "step": 865
    },
    {
      "epoch": 0.12957998212689903,
      "grad_norm": 13.5625,
      "learning_rate": 3.2357759904676794e-06,
      "loss": 1.2222,
      "num_input_tokens_seen": 507456,
      "step": 870
    },
    {
      "epoch": 0.1303246946678582,
      "grad_norm": 26.875,
      "learning_rate": 3.2543938039916594e-06,
      "loss": 1.4822,
      "num_input_tokens_seen": 510432,
      "step": 875
    },
    {
      "epoch": 0.1310694072088174,
      "grad_norm": 14.375,
      "learning_rate": 3.2730116175156393e-06,
      "loss": 1.2065,
      "num_input_tokens_seen": 513216,
      "step": 880
    },
    {
      "epoch": 0.13181411974977658,
      "grad_norm": 13.1875,
      "learning_rate": 3.291629431039619e-06,
      "loss": 1.1111,
      "num_input_tokens_seen": 516256,
      "step": 885
    },
    {
      "epoch": 0.13255883229073578,
      "grad_norm": 10.875,
      "learning_rate": 3.3102472445635987e-06,
      "loss": 1.2789,
      "num_input_tokens_seen": 519296,
      "step": 890
    },
    {
      "epoch": 0.13330354483169496,
      "grad_norm": 14.3125,
      "learning_rate": 3.3288650580875782e-06,
      "loss": 0.9573,
      "num_input_tokens_seen": 522112,
      "step": 895
    },
    {
      "epoch": 0.13404825737265416,
      "grad_norm": 9.125,
      "learning_rate": 3.347482871611558e-06,
      "loss": 1.2193,
      "num_input_tokens_seen": 525376,
      "step": 900
    },
    {
      "epoch": 0.13479296991361334,
      "grad_norm": 8.3125,
      "learning_rate": 3.3661006851355377e-06,
      "loss": 1.1665,
      "num_input_tokens_seen": 528192,
      "step": 905
    },
    {
      "epoch": 0.13553768245457254,
      "grad_norm": 16.0,
      "learning_rate": 3.3847184986595176e-06,
      "loss": 1.5013,
      "num_input_tokens_seen": 531136,
      "step": 910
    },
    {
      "epoch": 0.1362823949955317,
      "grad_norm": 16.25,
      "learning_rate": 3.403336312183497e-06,
      "loss": 1.5285,
      "num_input_tokens_seen": 534144,
      "step": 915
    },
    {
      "epoch": 0.13702710753649092,
      "grad_norm": 15.1875,
      "learning_rate": 3.421954125707477e-06,
      "loss": 1.2905,
      "num_input_tokens_seen": 536832,
      "step": 920
    },
    {
      "epoch": 0.1377718200774501,
      "grad_norm": 24.375,
      "learning_rate": 3.4405719392314565e-06,
      "loss": 1.1833,
      "num_input_tokens_seen": 539328,
      "step": 925
    },
    {
      "epoch": 0.1385165326184093,
      "grad_norm": 15.3125,
      "learning_rate": 3.4591897527554364e-06,
      "loss": 1.1748,
      "num_input_tokens_seen": 542208,
      "step": 930
    },
    {
      "epoch": 0.1392612451593685,
      "grad_norm": 9.6875,
      "learning_rate": 3.477807566279416e-06,
      "loss": 0.758,
      "num_input_tokens_seen": 545088,
      "step": 935
    },
    {
      "epoch": 0.14000595770032767,
      "grad_norm": 16.5,
      "learning_rate": 3.496425379803396e-06,
      "loss": 1.1103,
      "num_input_tokens_seen": 547776,
      "step": 940
    },
    {
      "epoch": 0.14075067024128687,
      "grad_norm": 10.5625,
      "learning_rate": 3.5150431933273762e-06,
      "loss": 1.2078,
      "num_input_tokens_seen": 550400,
      "step": 945
    },
    {
      "epoch": 0.14149538278224605,
      "grad_norm": 11.0,
      "learning_rate": 3.5336610068513553e-06,
      "loss": 1.0798,
      "num_input_tokens_seen": 553184,
      "step": 950
    },
    {
      "epoch": 0.14224009532320525,
      "grad_norm": 31.125,
      "learning_rate": 3.5522788203753356e-06,
      "loss": 1.3325,
      "num_input_tokens_seen": 556096,
      "step": 955
    },
    {
      "epoch": 0.14298480786416443,
      "grad_norm": 14.4375,
      "learning_rate": 3.570896633899315e-06,
      "loss": 1.1119,
      "num_input_tokens_seen": 558784,
      "step": 960
    },
    {
      "epoch": 0.14372952040512363,
      "grad_norm": 10.375,
      "learning_rate": 3.589514447423295e-06,
      "loss": 1.5069,
      "num_input_tokens_seen": 561760,
      "step": 965
    },
    {
      "epoch": 0.1444742329460828,
      "grad_norm": 18.0,
      "learning_rate": 3.6081322609472746e-06,
      "loss": 1.1191,
      "num_input_tokens_seen": 564544,
      "step": 970
    },
    {
      "epoch": 0.145218945487042,
      "grad_norm": 13.3125,
      "learning_rate": 3.6267500744712545e-06,
      "loss": 0.8828,
      "num_input_tokens_seen": 567200,
      "step": 975
    },
    {
      "epoch": 0.14596365802800118,
      "grad_norm": 12.625,
      "learning_rate": 3.645367887995234e-06,
      "loss": 0.8606,
      "num_input_tokens_seen": 570240,
      "step": 980
    },
    {
      "epoch": 0.14670837056896038,
      "grad_norm": 13.25,
      "learning_rate": 3.663985701519214e-06,
      "loss": 1.2092,
      "num_input_tokens_seen": 572896,
      "step": 985
    },
    {
      "epoch": 0.14745308310991956,
      "grad_norm": 18.125,
      "learning_rate": 3.6826035150431934e-06,
      "loss": 1.1601,
      "num_input_tokens_seen": 575744,
      "step": 990
    },
    {
      "epoch": 0.14819779565087876,
      "grad_norm": 4.3125,
      "learning_rate": 3.7012213285671734e-06,
      "loss": 1.0113,
      "num_input_tokens_seen": 578848,
      "step": 995
    },
    {
      "epoch": 0.14894250819183796,
      "grad_norm": 10.625,
      "learning_rate": 3.719839142091153e-06,
      "loss": 1.5754,
      "num_input_tokens_seen": 581504,
      "step": 1000
    },
    {
      "epoch": 0.14968722073279714,
      "grad_norm": 16.875,
      "learning_rate": 3.738456955615133e-06,
      "loss": 1.2812,
      "num_input_tokens_seen": 584256,
      "step": 1005
    },
    {
      "epoch": 0.15043193327375634,
      "grad_norm": 19.875,
      "learning_rate": 3.7570747691391127e-06,
      "loss": 1.2594,
      "num_input_tokens_seen": 586912,
      "step": 1010
    },
    {
      "epoch": 0.15117664581471552,
      "grad_norm": 15.25,
      "learning_rate": 3.7756925826630922e-06,
      "loss": 1.0633,
      "num_input_tokens_seen": 589920,
      "step": 1015
    },
    {
      "epoch": 0.15192135835567472,
      "grad_norm": 14.375,
      "learning_rate": 3.794310396187072e-06,
      "loss": 1.1833,
      "num_input_tokens_seen": 592992,
      "step": 1020
    },
    {
      "epoch": 0.1526660708966339,
      "grad_norm": 9.5625,
      "learning_rate": 3.8129282097110517e-06,
      "loss": 1.0523,
      "num_input_tokens_seen": 595872,
      "step": 1025
    },
    {
      "epoch": 0.1534107834375931,
      "grad_norm": 19.0,
      "learning_rate": 3.831546023235032e-06,
      "loss": 1.2385,
      "num_input_tokens_seen": 599008,
      "step": 1030
    },
    {
      "epoch": 0.15415549597855227,
      "grad_norm": 8.875,
      "learning_rate": 3.850163836759011e-06,
      "loss": 1.1273,
      "num_input_tokens_seen": 601792,
      "step": 1035
    },
    {
      "epoch": 0.15490020851951147,
      "grad_norm": 16.625,
      "learning_rate": 3.8687816502829914e-06,
      "loss": 0.7435,
      "num_input_tokens_seen": 604608,
      "step": 1040
    },
    {
      "epoch": 0.15564492106047065,
      "grad_norm": 9.0,
      "learning_rate": 3.8873994638069705e-06,
      "loss": 1.5025,
      "num_input_tokens_seen": 607168,
      "step": 1045
    },
    {
      "epoch": 0.15638963360142985,
      "grad_norm": 10.9375,
      "learning_rate": 3.9060172773309504e-06,
      "loss": 1.0795,
      "num_input_tokens_seen": 610048,
      "step": 1050
    },
    {
      "epoch": 0.15713434614238903,
      "grad_norm": 9.5625,
      "learning_rate": 3.92463509085493e-06,
      "loss": 1.0493,
      "num_input_tokens_seen": 612864,
      "step": 1055
    },
    {
      "epoch": 0.15787905868334823,
      "grad_norm": 9.6875,
      "learning_rate": 3.94325290437891e-06,
      "loss": 1.1304,
      "num_input_tokens_seen": 616096,
      "step": 1060
    },
    {
      "epoch": 0.15862377122430743,
      "grad_norm": 17.25,
      "learning_rate": 3.961870717902889e-06,
      "loss": 1.2251,
      "num_input_tokens_seen": 618784,
      "step": 1065
    },
    {
      "epoch": 0.1593684837652666,
      "grad_norm": 23.125,
      "learning_rate": 3.980488531426869e-06,
      "loss": 1.1559,
      "num_input_tokens_seen": 621600,
      "step": 1070
    },
    {
      "epoch": 0.1601131963062258,
      "grad_norm": 17.625,
      "learning_rate": 3.999106344950849e-06,
      "loss": 1.3093,
      "num_input_tokens_seen": 624704,
      "step": 1075
    },
    {
      "epoch": 0.16085790884718498,
      "grad_norm": 11.4375,
      "learning_rate": 4.017724158474829e-06,
      "loss": 1.5255,
      "num_input_tokens_seen": 627808,
      "step": 1080
    },
    {
      "epoch": 0.16160262138814419,
      "grad_norm": 18.375,
      "learning_rate": 4.036341971998809e-06,
      "loss": 1.3271,
      "num_input_tokens_seen": 630688,
      "step": 1085
    },
    {
      "epoch": 0.16234733392910336,
      "grad_norm": 13.375,
      "learning_rate": 4.054959785522788e-06,
      "loss": 1.2033,
      "num_input_tokens_seen": 633600,
      "step": 1090
    },
    {
      "epoch": 0.16309204647006256,
      "grad_norm": 17.125,
      "learning_rate": 4.073577599046768e-06,
      "loss": 1.4179,
      "num_input_tokens_seen": 636672,
      "step": 1095
    },
    {
      "epoch": 0.16383675901102174,
      "grad_norm": 27.625,
      "learning_rate": 4.092195412570748e-06,
      "loss": 1.3813,
      "num_input_tokens_seen": 639168,
      "step": 1100
    },
    {
      "epoch": 0.16458147155198094,
      "grad_norm": 10.9375,
      "learning_rate": 4.110813226094728e-06,
      "loss": 1.2305,
      "num_input_tokens_seen": 641984,
      "step": 1105
    },
    {
      "epoch": 0.16532618409294011,
      "grad_norm": 8.75,
      "learning_rate": 4.129431039618707e-06,
      "loss": 0.9364,
      "num_input_tokens_seen": 645408,
      "step": 1110
    },
    {
      "epoch": 0.16607089663389932,
      "grad_norm": 9.3125,
      "learning_rate": 4.148048853142687e-06,
      "loss": 1.4128,
      "num_input_tokens_seen": 648032,
      "step": 1115
    },
    {
      "epoch": 0.1668156091748585,
      "grad_norm": 5.1875,
      "learning_rate": 4.166666666666667e-06,
      "loss": 1.0463,
      "num_input_tokens_seen": 651232,
      "step": 1120
    },
    {
      "epoch": 0.1675603217158177,
      "grad_norm": 9.3125,
      "learning_rate": 4.185284480190647e-06,
      "loss": 1.0353,
      "num_input_tokens_seen": 654112,
      "step": 1125
    },
    {
      "epoch": 0.1683050342567769,
      "grad_norm": 17.25,
      "learning_rate": 4.203902293714626e-06,
      "loss": 1.2192,
      "num_input_tokens_seen": 657184,
      "step": 1130
    },
    {
      "epoch": 0.16904974679773607,
      "grad_norm": 10.875,
      "learning_rate": 4.222520107238606e-06,
      "loss": 0.983,
      "num_input_tokens_seen": 659936,
      "step": 1135
    },
    {
      "epoch": 0.16979445933869527,
      "grad_norm": 18.625,
      "learning_rate": 4.241137920762586e-06,
      "loss": 1.1447,
      "num_input_tokens_seen": 662880,
      "step": 1140
    },
    {
      "epoch": 0.17053917187965445,
      "grad_norm": 18.0,
      "learning_rate": 4.259755734286566e-06,
      "loss": 1.4121,
      "num_input_tokens_seen": 665760,
      "step": 1145
    },
    {
      "epoch": 0.17128388442061365,
      "grad_norm": 9.375,
      "learning_rate": 4.278373547810546e-06,
      "loss": 1.2326,
      "num_input_tokens_seen": 668672,
      "step": 1150
    },
    {
      "epoch": 0.17202859696157283,
      "grad_norm": 21.5,
      "learning_rate": 4.296991361334525e-06,
      "loss": 1.5139,
      "num_input_tokens_seen": 671424,
      "step": 1155
    },
    {
      "epoch": 0.17277330950253203,
      "grad_norm": 16.125,
      "learning_rate": 4.3156091748585054e-06,
      "loss": 1.5474,
      "num_input_tokens_seen": 674176,
      "step": 1160
    },
    {
      "epoch": 0.1735180220434912,
      "grad_norm": 15.875,
      "learning_rate": 4.3342269883824845e-06,
      "loss": 1.1601,
      "num_input_tokens_seen": 676992,
      "step": 1165
    },
    {
      "epoch": 0.1742627345844504,
      "grad_norm": 9.875,
      "learning_rate": 4.3528448019064644e-06,
      "loss": 1.134,
      "num_input_tokens_seen": 679936,
      "step": 1170
    },
    {
      "epoch": 0.17500744712540958,
      "grad_norm": 12.6875,
      "learning_rate": 4.3714626154304435e-06,
      "loss": 1.1072,
      "num_input_tokens_seen": 682656,
      "step": 1175
    },
    {
      "epoch": 0.17575215966636878,
      "grad_norm": 12.875,
      "learning_rate": 4.390080428954424e-06,
      "loss": 1.52,
      "num_input_tokens_seen": 685376,
      "step": 1180
    },
    {
      "epoch": 0.17649687220732796,
      "grad_norm": 10.0625,
      "learning_rate": 4.408698242478403e-06,
      "loss": 1.3607,
      "num_input_tokens_seen": 688000,
      "step": 1185
    },
    {
      "epoch": 0.17724158474828716,
      "grad_norm": 9.9375,
      "learning_rate": 4.427316056002383e-06,
      "loss": 0.9941,
      "num_input_tokens_seen": 690752,
      "step": 1190
    },
    {
      "epoch": 0.17798629728924636,
      "grad_norm": 17.0,
      "learning_rate": 4.445933869526362e-06,
      "loss": 1.2957,
      "num_input_tokens_seen": 693696,
      "step": 1195
    },
    {
      "epoch": 0.17873100983020554,
      "grad_norm": 9.875,
      "learning_rate": 4.464551683050343e-06,
      "loss": 1.5978,
      "num_input_tokens_seen": 696832,
      "step": 1200
    },
    {
      "epoch": 0.17947572237116474,
      "grad_norm": 9.875,
      "learning_rate": 4.483169496574322e-06,
      "loss": 1.2028,
      "num_input_tokens_seen": 700192,
      "step": 1205
    },
    {
      "epoch": 0.18022043491212392,
      "grad_norm": 19.125,
      "learning_rate": 4.501787310098302e-06,
      "loss": 0.979,
      "num_input_tokens_seen": 702976,
      "step": 1210
    },
    {
      "epoch": 0.18096514745308312,
      "grad_norm": 16.5,
      "learning_rate": 4.520405123622282e-06,
      "loss": 0.9658,
      "num_input_tokens_seen": 705696,
      "step": 1215
    },
    {
      "epoch": 0.1817098599940423,
      "grad_norm": 10.875,
      "learning_rate": 4.539022937146262e-06,
      "loss": 1.3398,
      "num_input_tokens_seen": 708544,
      "step": 1220
    },
    {
      "epoch": 0.1824545725350015,
      "grad_norm": 12.8125,
      "learning_rate": 4.557640750670242e-06,
      "loss": 1.0629,
      "num_input_tokens_seen": 711168,
      "step": 1225
    },
    {
      "epoch": 0.18319928507596067,
      "grad_norm": 26.375,
      "learning_rate": 4.576258564194221e-06,
      "loss": 1.2787,
      "num_input_tokens_seen": 714176,
      "step": 1230
    },
    {
      "epoch": 0.18394399761691987,
      "grad_norm": 20.375,
      "learning_rate": 4.594876377718201e-06,
      "loss": 1.2135,
      "num_input_tokens_seen": 716800,
      "step": 1235
    },
    {
      "epoch": 0.18468871015787905,
      "grad_norm": 9.625,
      "learning_rate": 4.613494191242181e-06,
      "loss": 1.189,
      "num_input_tokens_seen": 719712,
      "step": 1240
    },
    {
      "epoch": 0.18543342269883825,
      "grad_norm": 15.25,
      "learning_rate": 4.632112004766161e-06,
      "loss": 1.1534,
      "num_input_tokens_seen": 722720,
      "step": 1245
    },
    {
      "epoch": 0.18617813523979743,
      "grad_norm": 20.875,
      "learning_rate": 4.65072981829014e-06,
      "loss": 1.4004,
      "num_input_tokens_seen": 725248,
      "step": 1250
    },
    {
      "epoch": 0.18692284778075663,
      "grad_norm": 6.59375,
      "learning_rate": 4.66934763181412e-06,
      "loss": 1.0414,
      "num_input_tokens_seen": 728160,
      "step": 1255
    },
    {
      "epoch": 0.1876675603217158,
      "grad_norm": 11.875,
      "learning_rate": 4.6879654453381e-06,
      "loss": 1.0234,
      "num_input_tokens_seen": 731008,
      "step": 1260
    },
    {
      "epoch": 0.188412272862675,
      "grad_norm": 14.375,
      "learning_rate": 4.70658325886208e-06,
      "loss": 0.9435,
      "num_input_tokens_seen": 733824,
      "step": 1265
    },
    {
      "epoch": 0.1891569854036342,
      "grad_norm": 10.6875,
      "learning_rate": 4.725201072386059e-06,
      "loss": 0.9215,
      "num_input_tokens_seen": 736640,
      "step": 1270
    },
    {
      "epoch": 0.18990169794459338,
      "grad_norm": 9.6875,
      "learning_rate": 4.743818885910039e-06,
      "loss": 1.2009,
      "num_input_tokens_seen": 739488,
      "step": 1275
    },
    {
      "epoch": 0.1906464104855526,
      "grad_norm": 21.375,
      "learning_rate": 4.7624366994340194e-06,
      "loss": 1.1568,
      "num_input_tokens_seen": 742176,
      "step": 1280
    },
    {
      "epoch": 0.19139112302651176,
      "grad_norm": 4.34375,
      "learning_rate": 4.7810545129579985e-06,
      "loss": 0.8609,
      "num_input_tokens_seen": 745184,
      "step": 1285
    },
    {
      "epoch": 0.19213583556747096,
      "grad_norm": 11.8125,
      "learning_rate": 4.7996723264819784e-06,
      "loss": 0.8855,
      "num_input_tokens_seen": 748224,
      "step": 1290
    },
    {
      "epoch": 0.19288054810843014,
      "grad_norm": 14.6875,
      "learning_rate": 4.8182901400059575e-06,
      "loss": 1.059,
      "num_input_tokens_seen": 750944,
      "step": 1295
    },
    {
      "epoch": 0.19362526064938934,
      "grad_norm": 9.625,
      "learning_rate": 4.836907953529938e-06,
      "loss": 1.1457,
      "num_input_tokens_seen": 753888,
      "step": 1300
    },
    {
      "epoch": 0.19436997319034852,
      "grad_norm": 12.75,
      "learning_rate": 4.855525767053917e-06,
      "loss": 1.6421,
      "num_input_tokens_seen": 756704,
      "step": 1305
    },
    {
      "epoch": 0.19511468573130772,
      "grad_norm": 11.9375,
      "learning_rate": 4.874143580577897e-06,
      "loss": 0.9046,
      "num_input_tokens_seen": 759904,
      "step": 1310
    },
    {
      "epoch": 0.1958593982722669,
      "grad_norm": 21.875,
      "learning_rate": 4.892761394101876e-06,
      "loss": 1.1109,
      "num_input_tokens_seen": 762848,
      "step": 1315
    },
    {
      "epoch": 0.1966041108132261,
      "grad_norm": 10.0,
      "learning_rate": 4.911379207625857e-06,
      "loss": 1.0261,
      "num_input_tokens_seen": 765536,
      "step": 1320
    },
    {
      "epoch": 0.19734882335418527,
      "grad_norm": 20.125,
      "learning_rate": 4.929997021149836e-06,
      "loss": 1.188,
      "num_input_tokens_seen": 768384,
      "step": 1325
    },
    {
      "epoch": 0.19809353589514447,
      "grad_norm": 11.375,
      "learning_rate": 4.948614834673816e-06,
      "loss": 1.0571,
      "num_input_tokens_seen": 771552,
      "step": 1330
    },
    {
      "epoch": 0.19883824843610368,
      "grad_norm": 5.0625,
      "learning_rate": 4.967232648197795e-06,
      "loss": 1.1863,
      "num_input_tokens_seen": 774336,
      "step": 1335
    },
    {
      "epoch": 0.19958296097706285,
      "grad_norm": 8.6875,
      "learning_rate": 4.985850461721776e-06,
      "loss": 0.9061,
      "num_input_tokens_seen": 777312,
      "step": 1340
    },
    {
      "epoch": 0.20032767351802205,
      "grad_norm": 10.0,
      "learning_rate": 5.004468275245756e-06,
      "loss": 1.524,
      "num_input_tokens_seen": 780032,
      "step": 1345
    },
    {
      "epoch": 0.20107238605898123,
      "grad_norm": 8.8125,
      "learning_rate": 5.023086088769735e-06,
      "loss": 0.9122,
      "num_input_tokens_seen": 782720,
      "step": 1350
    },
    {
      "epoch": 0.20181709859994043,
      "grad_norm": 11.875,
      "learning_rate": 5.041703902293715e-06,
      "loss": 1.0421,
      "num_input_tokens_seen": 785760,
      "step": 1355
    },
    {
      "epoch": 0.2025618111408996,
      "grad_norm": 8.8125,
      "learning_rate": 5.060321715817695e-06,
      "loss": 1.0322,
      "num_input_tokens_seen": 788608,
      "step": 1360
    },
    {
      "epoch": 0.2033065236818588,
      "grad_norm": 12.875,
      "learning_rate": 5.078939529341675e-06,
      "loss": 1.1729,
      "num_input_tokens_seen": 791872,
      "step": 1365
    },
    {
      "epoch": 0.20405123622281798,
      "grad_norm": 12.4375,
      "learning_rate": 5.097557342865654e-06,
      "loss": 0.9203,
      "num_input_tokens_seen": 795136,
      "step": 1370
    },
    {
      "epoch": 0.20479594876377719,
      "grad_norm": 21.375,
      "learning_rate": 5.116175156389634e-06,
      "loss": 1.172,
      "num_input_tokens_seen": 798048,
      "step": 1375
    },
    {
      "epoch": 0.20554066130473636,
      "grad_norm": 12.4375,
      "learning_rate": 5.134792969913614e-06,
      "loss": 1.1312,
      "num_input_tokens_seen": 800576,
      "step": 1380
    },
    {
      "epoch": 0.20628537384569556,
      "grad_norm": 12.8125,
      "learning_rate": 5.153410783437594e-06,
      "loss": 1.1132,
      "num_input_tokens_seen": 803616,
      "step": 1385
    },
    {
      "epoch": 0.20703008638665474,
      "grad_norm": 13.9375,
      "learning_rate": 5.172028596961573e-06,
      "loss": 1.5241,
      "num_input_tokens_seen": 806304,
      "step": 1390
    },
    {
      "epoch": 0.20777479892761394,
      "grad_norm": 10.375,
      "learning_rate": 5.190646410485553e-06,
      "loss": 1.317,
      "num_input_tokens_seen": 808992,
      "step": 1395
    },
    {
      "epoch": 0.20851951146857314,
      "grad_norm": 10.4375,
      "learning_rate": 5.209264224009533e-06,
      "loss": 1.1336,
      "num_input_tokens_seen": 811904,
      "step": 1400
    },
    {
      "epoch": 0.20926422400953232,
      "grad_norm": 20.625,
      "learning_rate": 5.2278820375335125e-06,
      "loss": 1.4046,
      "num_input_tokens_seen": 815072,
      "step": 1405
    },
    {
      "epoch": 0.21000893655049152,
      "grad_norm": 25.25,
      "learning_rate": 5.2464998510574924e-06,
      "loss": 1.7311,
      "num_input_tokens_seen": 818016,
      "step": 1410
    },
    {
      "epoch": 0.2107536490914507,
      "grad_norm": 16.75,
      "learning_rate": 5.2651176645814715e-06,
      "loss": 1.1083,
      "num_input_tokens_seen": 820896,
      "step": 1415
    },
    {
      "epoch": 0.2114983616324099,
      "grad_norm": 16.75,
      "learning_rate": 5.2837354781054514e-06,
      "loss": 1.1319,
      "num_input_tokens_seen": 823520,
      "step": 1420
    },
    {
      "epoch": 0.21224307417336907,
      "grad_norm": 11.625,
      "learning_rate": 5.302353291629431e-06,
      "loss": 1.1473,
      "num_input_tokens_seen": 826144,
      "step": 1425
    },
    {
      "epoch": 0.21298778671432828,
      "grad_norm": 14.1875,
      "learning_rate": 5.320971105153411e-06,
      "loss": 1.1721,
      "num_input_tokens_seen": 828768,
      "step": 1430
    },
    {
      "epoch": 0.21373249925528745,
      "grad_norm": 9.75,
      "learning_rate": 5.33958891867739e-06,
      "loss": 1.1465,
      "num_input_tokens_seen": 831936,
      "step": 1435
    },
    {
      "epoch": 0.21447721179624665,
      "grad_norm": 10.5,
      "learning_rate": 5.35820673220137e-06,
      "loss": 0.9861,
      "num_input_tokens_seen": 835008,
      "step": 1440
    },
    {
      "epoch": 0.21522192433720583,
      "grad_norm": 18.5,
      "learning_rate": 5.37682454572535e-06,
      "loss": 1.3671,
      "num_input_tokens_seen": 837600,
      "step": 1445
    },
    {
      "epoch": 0.21596663687816503,
      "grad_norm": 19.375,
      "learning_rate": 5.39544235924933e-06,
      "loss": 1.4792,
      "num_input_tokens_seen": 840480,
      "step": 1450
    },
    {
      "epoch": 0.2167113494191242,
      "grad_norm": 16.625,
      "learning_rate": 5.414060172773309e-06,
      "loss": 1.1415,
      "num_input_tokens_seen": 843424,
      "step": 1455
    },
    {
      "epoch": 0.2174560619600834,
      "grad_norm": 18.75,
      "learning_rate": 5.43267798629729e-06,
      "loss": 1.2107,
      "num_input_tokens_seen": 846272,
      "step": 1460
    },
    {
      "epoch": 0.2182007745010426,
      "grad_norm": 14.0625,
      "learning_rate": 5.451295799821269e-06,
      "loss": 1.3623,
      "num_input_tokens_seen": 849216,
      "step": 1465
    },
    {
      "epoch": 0.21894548704200179,
      "grad_norm": 7.875,
      "learning_rate": 5.469913613345249e-06,
      "loss": 1.416,
      "num_input_tokens_seen": 852032,
      "step": 1470
    },
    {
      "epoch": 0.219690199582961,
      "grad_norm": 11.625,
      "learning_rate": 5.488531426869229e-06,
      "loss": 1.0778,
      "num_input_tokens_seen": 855072,
      "step": 1475
    },
    {
      "epoch": 0.22043491212392016,
      "grad_norm": 16.875,
      "learning_rate": 5.507149240393209e-06,
      "loss": 1.3676,
      "num_input_tokens_seen": 857760,
      "step": 1480
    },
    {
      "epoch": 0.22117962466487937,
      "grad_norm": 15.8125,
      "learning_rate": 5.525767053917189e-06,
      "loss": 1.5085,
      "num_input_tokens_seen": 860928,
      "step": 1485
    },
    {
      "epoch": 0.22192433720583854,
      "grad_norm": 8.0,
      "learning_rate": 5.544384867441168e-06,
      "loss": 1.2315,
      "num_input_tokens_seen": 864032,
      "step": 1490
    },
    {
      "epoch": 0.22266904974679774,
      "grad_norm": 18.5,
      "learning_rate": 5.563002680965148e-06,
      "loss": 1.6683,
      "num_input_tokens_seen": 867040,
      "step": 1495
    },
    {
      "epoch": 0.22341376228775692,
      "grad_norm": 8.375,
      "learning_rate": 5.581620494489128e-06,
      "loss": 1.1951,
      "num_input_tokens_seen": 870048,
      "step": 1500
    },
    {
      "epoch": 0.22415847482871612,
      "grad_norm": 28.0,
      "learning_rate": 5.600238308013108e-06,
      "loss": 1.7779,
      "num_input_tokens_seen": 873632,
      "step": 1505
    },
    {
      "epoch": 0.2249031873696753,
      "grad_norm": 11.125,
      "learning_rate": 5.618856121537087e-06,
      "loss": 1.2465,
      "num_input_tokens_seen": 876608,
      "step": 1510
    },
    {
      "epoch": 0.2256478999106345,
      "grad_norm": 10.0,
      "learning_rate": 5.637473935061067e-06,
      "loss": 1.2311,
      "num_input_tokens_seen": 879424,
      "step": 1515
    },
    {
      "epoch": 0.22639261245159367,
      "grad_norm": 17.25,
      "learning_rate": 5.656091748585047e-06,
      "loss": 1.0784,
      "num_input_tokens_seen": 882496,
      "step": 1520
    },
    {
      "epoch": 0.22713732499255287,
      "grad_norm": 13.5,
      "learning_rate": 5.6747095621090265e-06,
      "loss": 1.1215,
      "num_input_tokens_seen": 885632,
      "step": 1525
    },
    {
      "epoch": 0.22788203753351208,
      "grad_norm": 9.0625,
      "learning_rate": 5.693327375633006e-06,
      "loss": 0.8395,
      "num_input_tokens_seen": 888480,
      "step": 1530
    },
    {
      "epoch": 0.22862675007447125,
      "grad_norm": 8.9375,
      "learning_rate": 5.7119451891569855e-06,
      "loss": 0.9134,
      "num_input_tokens_seen": 891392,
      "step": 1535
    },
    {
      "epoch": 0.22937146261543045,
      "grad_norm": 8.0625,
      "learning_rate": 5.7305630026809654e-06,
      "loss": 0.9142,
      "num_input_tokens_seen": 894336,
      "step": 1540
    },
    {
      "epoch": 0.23011617515638963,
      "grad_norm": 12.75,
      "learning_rate": 5.749180816204945e-06,
      "loss": 0.9339,
      "num_input_tokens_seen": 897280,
      "step": 1545
    },
    {
      "epoch": 0.23086088769734883,
      "grad_norm": 19.875,
      "learning_rate": 5.767798629728925e-06,
      "loss": 1.6353,
      "num_input_tokens_seen": 900192,
      "step": 1550
    },
    {
      "epoch": 0.231605600238308,
      "grad_norm": 12.5,
      "learning_rate": 5.786416443252904e-06,
      "loss": 1.197,
      "num_input_tokens_seen": 903232,
      "step": 1555
    },
    {
      "epoch": 0.2323503127792672,
      "grad_norm": 9.8125,
      "learning_rate": 5.805034256776884e-06,
      "loss": 1.1489,
      "num_input_tokens_seen": 906080,
      "step": 1560
    },
    {
      "epoch": 0.23309502532022638,
      "grad_norm": 4.5,
      "learning_rate": 5.823652070300864e-06,
      "loss": 0.9824,
      "num_input_tokens_seen": 908896,
      "step": 1565
    },
    {
      "epoch": 0.2338397378611856,
      "grad_norm": 7.65625,
      "learning_rate": 5.842269883824844e-06,
      "loss": 1.2908,
      "num_input_tokens_seen": 911840,
      "step": 1570
    },
    {
      "epoch": 0.23458445040214476,
      "grad_norm": 8.0625,
      "learning_rate": 5.860887697348823e-06,
      "loss": 1.2591,
      "num_input_tokens_seen": 914848,
      "step": 1575
    },
    {
      "epoch": 0.23532916294310396,
      "grad_norm": 18.0,
      "learning_rate": 5.879505510872803e-06,
      "loss": 1.3147,
      "num_input_tokens_seen": 917792,
      "step": 1580
    },
    {
      "epoch": 0.23607387548406314,
      "grad_norm": 12.625,
      "learning_rate": 5.898123324396783e-06,
      "loss": 1.5856,
      "num_input_tokens_seen": 920608,
      "step": 1585
    },
    {
      "epoch": 0.23681858802502234,
      "grad_norm": 12.125,
      "learning_rate": 5.916741137920763e-06,
      "loss": 1.5757,
      "num_input_tokens_seen": 923648,
      "step": 1590
    },
    {
      "epoch": 0.23756330056598154,
      "grad_norm": 10.1875,
      "learning_rate": 5.935358951444742e-06,
      "loss": 1.0628,
      "num_input_tokens_seen": 926528,
      "step": 1595
    },
    {
      "epoch": 0.23830801310694072,
      "grad_norm": 10.0625,
      "learning_rate": 5.953976764968722e-06,
      "loss": 0.8467,
      "num_input_tokens_seen": 929184,
      "step": 1600
    },
    {
      "epoch": 0.23905272564789992,
      "grad_norm": 9.6875,
      "learning_rate": 5.972594578492702e-06,
      "loss": 1.2703,
      "num_input_tokens_seen": 931872,
      "step": 1605
    },
    {
      "epoch": 0.2397974381888591,
      "grad_norm": 13.125,
      "learning_rate": 5.991212392016682e-06,
      "loss": 1.1206,
      "num_input_tokens_seen": 934688,
      "step": 1610
    },
    {
      "epoch": 0.2405421507298183,
      "grad_norm": 8.1875,
      "learning_rate": 6.009830205540662e-06,
      "loss": 0.8453,
      "num_input_tokens_seen": 938048,
      "step": 1615
    },
    {
      "epoch": 0.24128686327077747,
      "grad_norm": 7.28125,
      "learning_rate": 6.028448019064641e-06,
      "loss": 1.203,
      "num_input_tokens_seen": 940864,
      "step": 1620
    },
    {
      "epoch": 0.24203157581173668,
      "grad_norm": 5.75,
      "learning_rate": 6.047065832588622e-06,
      "loss": 1.2016,
      "num_input_tokens_seen": 943808,
      "step": 1625
    },
    {
      "epoch": 0.24277628835269585,
      "grad_norm": 13.875,
      "learning_rate": 6.065683646112601e-06,
      "loss": 0.815,
      "num_input_tokens_seen": 946976,
      "step": 1630
    },
    {
      "epoch": 0.24352100089365505,
      "grad_norm": 8.9375,
      "learning_rate": 6.084301459636581e-06,
      "loss": 0.8614,
      "num_input_tokens_seen": 949696,
      "step": 1635
    },
    {
      "epoch": 0.24426571343461423,
      "grad_norm": 10.6875,
      "learning_rate": 6.10291927316056e-06,
      "loss": 1.3035,
      "num_input_tokens_seen": 952512,
      "step": 1640
    },
    {
      "epoch": 0.24501042597557343,
      "grad_norm": 10.4375,
      "learning_rate": 6.1215370866845405e-06,
      "loss": 1.0513,
      "num_input_tokens_seen": 955200,
      "step": 1645
    },
    {
      "epoch": 0.2457551385165326,
      "grad_norm": 9.4375,
      "learning_rate": 6.14015490020852e-06,
      "loss": 0.7605,
      "num_input_tokens_seen": 958048,
      "step": 1650
    },
    {
      "epoch": 0.2464998510574918,
      "grad_norm": 21.5,
      "learning_rate": 6.1587727137324995e-06,
      "loss": 1.0489,
      "num_input_tokens_seen": 961152,
      "step": 1655
    },
    {
      "epoch": 0.247244563598451,
      "grad_norm": 14.5625,
      "learning_rate": 6.1773905272564794e-06,
      "loss": 0.9087,
      "num_input_tokens_seen": 964000,
      "step": 1660
    },
    {
      "epoch": 0.2479892761394102,
      "grad_norm": 16.25,
      "learning_rate": 6.196008340780459e-06,
      "loss": 1.0836,
      "num_input_tokens_seen": 966656,
      "step": 1665
    },
    {
      "epoch": 0.2487339886803694,
      "grad_norm": 13.0,
      "learning_rate": 6.2146261543044384e-06,
      "loss": 1.2783,
      "num_input_tokens_seen": 969760,
      "step": 1670
    },
    {
      "epoch": 0.24947870122132856,
      "grad_norm": 10.875,
      "learning_rate": 6.233243967828418e-06,
      "loss": 1.2603,
      "num_input_tokens_seen": 972672,
      "step": 1675
    },
    {
      "epoch": 0.25022341376228774,
      "grad_norm": 15.9375,
      "learning_rate": 6.251861781352398e-06,
      "loss": 1.1839,
      "num_input_tokens_seen": 975552,
      "step": 1680
    },
    {
      "epoch": 0.25096812630324694,
      "grad_norm": 11.8125,
      "learning_rate": 6.270479594876379e-06,
      "loss": 0.76,
      "num_input_tokens_seen": 978304,
      "step": 1685
    },
    {
      "epoch": 0.25171283884420614,
      "grad_norm": 9.8125,
      "learning_rate": 6.289097408400357e-06,
      "loss": 0.9434,
      "num_input_tokens_seen": 981088,
      "step": 1690
    },
    {
      "epoch": 0.25245755138516535,
      "grad_norm": 9.5,
      "learning_rate": 6.307715221924337e-06,
      "loss": 1.108,
      "num_input_tokens_seen": 983712,
      "step": 1695
    },
    {
      "epoch": 0.2532022639261245,
      "grad_norm": 11.75,
      "learning_rate": 6.326333035448317e-06,
      "loss": 1.2514,
      "num_input_tokens_seen": 986784,
      "step": 1700
    },
    {
      "epoch": 0.2539469764670837,
      "grad_norm": 11.8125,
      "learning_rate": 6.344950848972298e-06,
      "loss": 1.3359,
      "num_input_tokens_seen": 989984,
      "step": 1705
    },
    {
      "epoch": 0.2546916890080429,
      "grad_norm": 12.5625,
      "learning_rate": 6.363568662496276e-06,
      "loss": 0.9731,
      "num_input_tokens_seen": 992896,
      "step": 1710
    },
    {
      "epoch": 0.2554364015490021,
      "grad_norm": 17.375,
      "learning_rate": 6.382186476020256e-06,
      "loss": 1.3026,
      "num_input_tokens_seen": 995744,
      "step": 1715
    },
    {
      "epoch": 0.25618111408996125,
      "grad_norm": 14.375,
      "learning_rate": 6.400804289544236e-06,
      "loss": 1.2512,
      "num_input_tokens_seen": 998752,
      "step": 1720
    },
    {
      "epoch": 0.25692582663092045,
      "grad_norm": 11.875,
      "learning_rate": 6.419422103068217e-06,
      "loss": 1.5161,
      "num_input_tokens_seen": 1001792,
      "step": 1725
    },
    {
      "epoch": 0.25767053917187965,
      "grad_norm": 16.875,
      "learning_rate": 6.438039916592195e-06,
      "loss": 1.2143,
      "num_input_tokens_seen": 1004608,
      "step": 1730
    },
    {
      "epoch": 0.25841525171283886,
      "grad_norm": 10.3125,
      "learning_rate": 6.456657730116175e-06,
      "loss": 1.2736,
      "num_input_tokens_seen": 1007232,
      "step": 1735
    },
    {
      "epoch": 0.25915996425379806,
      "grad_norm": 13.9375,
      "learning_rate": 6.475275543640155e-06,
      "loss": 1.4187,
      "num_input_tokens_seen": 1010272,
      "step": 1740
    },
    {
      "epoch": 0.2599046767947572,
      "grad_norm": 24.375,
      "learning_rate": 6.493893357164136e-06,
      "loss": 1.387,
      "num_input_tokens_seen": 1013024,
      "step": 1745
    },
    {
      "epoch": 0.2606493893357164,
      "grad_norm": 14.5625,
      "learning_rate": 6.5125111706881156e-06,
      "loss": 1.0493,
      "num_input_tokens_seen": 1016256,
      "step": 1750
    },
    {
      "epoch": 0.2613941018766756,
      "grad_norm": 14.875,
      "learning_rate": 6.531128984212094e-06,
      "loss": 0.8787,
      "num_input_tokens_seen": 1019200,
      "step": 1755
    },
    {
      "epoch": 0.2621388144176348,
      "grad_norm": 6.9375,
      "learning_rate": 6.549746797736074e-06,
      "loss": 1.4109,
      "num_input_tokens_seen": 1021888,
      "step": 1760
    },
    {
      "epoch": 0.26288352695859396,
      "grad_norm": 22.875,
      "learning_rate": 6.5683646112600545e-06,
      "loss": 1.1945,
      "num_input_tokens_seen": 1024864,
      "step": 1765
    },
    {
      "epoch": 0.26362823949955316,
      "grad_norm": 10.375,
      "learning_rate": 6.5869824247840344e-06,
      "loss": 0.9039,
      "num_input_tokens_seen": 1027648,
      "step": 1770
    },
    {
      "epoch": 0.26437295204051237,
      "grad_norm": 13.0,
      "learning_rate": 6.605600238308013e-06,
      "loss": 1.3837,
      "num_input_tokens_seen": 1030496,
      "step": 1775
    },
    {
      "epoch": 0.26511766458147157,
      "grad_norm": 13.0,
      "learning_rate": 6.624218051831993e-06,
      "loss": 1.0766,
      "num_input_tokens_seen": 1033728,
      "step": 1780
    },
    {
      "epoch": 0.2658623771224307,
      "grad_norm": 18.75,
      "learning_rate": 6.642835865355973e-06,
      "loss": 1.5532,
      "num_input_tokens_seen": 1036672,
      "step": 1785
    },
    {
      "epoch": 0.2666070896633899,
      "grad_norm": 13.6875,
      "learning_rate": 6.661453678879953e-06,
      "loss": 1.0965,
      "num_input_tokens_seen": 1039808,
      "step": 1790
    },
    {
      "epoch": 0.2673518022043491,
      "grad_norm": 7.5625,
      "learning_rate": 6.6800714924039315e-06,
      "loss": 0.8325,
      "num_input_tokens_seen": 1042720,
      "step": 1795
    },
    {
      "epoch": 0.2680965147453083,
      "grad_norm": 5.625,
      "learning_rate": 6.6986893059279114e-06,
      "loss": 1.0332,
      "num_input_tokens_seen": 1045536,
      "step": 1800
    },
    {
      "epoch": 0.2688412272862675,
      "grad_norm": 14.125,
      "learning_rate": 6.717307119451892e-06,
      "loss": 1.0726,
      "num_input_tokens_seen": 1048384,
      "step": 1805
    },
    {
      "epoch": 0.2695859398272267,
      "grad_norm": 10.875,
      "learning_rate": 6.735924932975872e-06,
      "loss": 1.4317,
      "num_input_tokens_seen": 1051168,
      "step": 1810
    },
    {
      "epoch": 0.2703306523681859,
      "grad_norm": 11.25,
      "learning_rate": 6.754542746499852e-06,
      "loss": 1.2546,
      "num_input_tokens_seen": 1053952,
      "step": 1815
    },
    {
      "epoch": 0.2710753649091451,
      "grad_norm": 12.375,
      "learning_rate": 6.77316056002383e-06,
      "loss": 1.0726,
      "num_input_tokens_seen": 1057056,
      "step": 1820
    },
    {
      "epoch": 0.2718200774501043,
      "grad_norm": 12.25,
      "learning_rate": 6.791778373547811e-06,
      "loss": 1.1937,
      "num_input_tokens_seen": 1059968,
      "step": 1825
    },
    {
      "epoch": 0.2725647899910634,
      "grad_norm": 19.25,
      "learning_rate": 6.810396187071791e-06,
      "loss": 1.2519,
      "num_input_tokens_seen": 1063008,
      "step": 1830
    },
    {
      "epoch": 0.27330950253202263,
      "grad_norm": 13.8125,
      "learning_rate": 6.829014000595771e-06,
      "loss": 0.9272,
      "num_input_tokens_seen": 1065728,
      "step": 1835
    },
    {
      "epoch": 0.27405421507298183,
      "grad_norm": 10.1875,
      "learning_rate": 6.847631814119749e-06,
      "loss": 1.0655,
      "num_input_tokens_seen": 1068608,
      "step": 1840
    },
    {
      "epoch": 0.27479892761394104,
      "grad_norm": 18.5,
      "learning_rate": 6.86624962764373e-06,
      "loss": 1.1292,
      "num_input_tokens_seen": 1071552,
      "step": 1845
    },
    {
      "epoch": 0.2755436401549002,
      "grad_norm": 12.5625,
      "learning_rate": 6.88486744116771e-06,
      "loss": 1.2588,
      "num_input_tokens_seen": 1074400,
      "step": 1850
    },
    {
      "epoch": 0.2762883526958594,
      "grad_norm": 9.9375,
      "learning_rate": 6.90348525469169e-06,
      "loss": 1.0004,
      "num_input_tokens_seen": 1077440,
      "step": 1855
    },
    {
      "epoch": 0.2770330652368186,
      "grad_norm": 16.625,
      "learning_rate": 6.922103068215669e-06,
      "loss": 1.1437,
      "num_input_tokens_seen": 1080576,
      "step": 1860
    },
    {
      "epoch": 0.2777777777777778,
      "grad_norm": 15.1875,
      "learning_rate": 6.940720881739649e-06,
      "loss": 1.0173,
      "num_input_tokens_seen": 1083616,
      "step": 1865
    },
    {
      "epoch": 0.278522490318737,
      "grad_norm": 13.0,
      "learning_rate": 6.959338695263629e-06,
      "loss": 1.5135,
      "num_input_tokens_seen": 1086624,
      "step": 1870
    },
    {
      "epoch": 0.27926720285969614,
      "grad_norm": 11.5625,
      "learning_rate": 6.977956508787609e-06,
      "loss": 1.441,
      "num_input_tokens_seen": 1089632,
      "step": 1875
    },
    {
      "epoch": 0.28001191540065534,
      "grad_norm": 6.125,
      "learning_rate": 6.9965743223115886e-06,
      "loss": 0.9625,
      "num_input_tokens_seen": 1092512,
      "step": 1880
    },
    {
      "epoch": 0.28075662794161454,
      "grad_norm": 15.75,
      "learning_rate": 7.015192135835568e-06,
      "loss": 1.1698,
      "num_input_tokens_seen": 1095488,
      "step": 1885
    },
    {
      "epoch": 0.28150134048257375,
      "grad_norm": 24.0,
      "learning_rate": 7.033809949359548e-06,
      "loss": 1.2673,
      "num_input_tokens_seen": 1098176,
      "step": 1890
    },
    {
      "epoch": 0.2822460530235329,
      "grad_norm": 10.1875,
      "learning_rate": 7.0524277628835275e-06,
      "loss": 0.9935,
      "num_input_tokens_seen": 1101312,
      "step": 1895
    },
    {
      "epoch": 0.2829907655644921,
      "grad_norm": 14.25,
      "learning_rate": 7.0710455764075074e-06,
      "loss": 0.9321,
      "num_input_tokens_seen": 1104480,
      "step": 1900
    },
    {
      "epoch": 0.2837354781054513,
      "grad_norm": 13.8125,
      "learning_rate": 7.0896633899314865e-06,
      "loss": 1.0822,
      "num_input_tokens_seen": 1107424,
      "step": 1905
    },
    {
      "epoch": 0.2844801906464105,
      "grad_norm": 13.875,
      "learning_rate": 7.1082812034554664e-06,
      "loss": 1.5198,
      "num_input_tokens_seen": 1110112,
      "step": 1910
    },
    {
      "epoch": 0.28522490318736965,
      "grad_norm": 29.625,
      "learning_rate": 7.126899016979446e-06,
      "loss": 1.5888,
      "num_input_tokens_seen": 1113120,
      "step": 1915
    },
    {
      "epoch": 0.28596961572832885,
      "grad_norm": 12.875,
      "learning_rate": 7.145516830503426e-06,
      "loss": 1.3021,
      "num_input_tokens_seen": 1115904,
      "step": 1920
    },
    {
      "epoch": 0.28671432826928805,
      "grad_norm": 15.8125,
      "learning_rate": 7.164134644027405e-06,
      "loss": 1.004,
      "num_input_tokens_seen": 1118880,
      "step": 1925
    },
    {
      "epoch": 0.28745904081024726,
      "grad_norm": 9.3125,
      "learning_rate": 7.182752457551385e-06,
      "loss": 1.2074,
      "num_input_tokens_seen": 1121952,
      "step": 1930
    },
    {
      "epoch": 0.28820375335120646,
      "grad_norm": 8.9375,
      "learning_rate": 7.201370271075365e-06,
      "loss": 0.8638,
      "num_input_tokens_seen": 1124704,
      "step": 1935
    },
    {
      "epoch": 0.2889484658921656,
      "grad_norm": 15.0,
      "learning_rate": 7.219988084599345e-06,
      "loss": 1.3215,
      "num_input_tokens_seen": 1127456,
      "step": 1940
    },
    {
      "epoch": 0.2896931784331248,
      "grad_norm": 22.5,
      "learning_rate": 7.238605898123325e-06,
      "loss": 1.3651,
      "num_input_tokens_seen": 1130784,
      "step": 1945
    },
    {
      "epoch": 0.290437890974084,
      "grad_norm": 10.1875,
      "learning_rate": 7.257223711647304e-06,
      "loss": 1.0343,
      "num_input_tokens_seen": 1133760,
      "step": 1950
    },
    {
      "epoch": 0.2911826035150432,
      "grad_norm": 8.25,
      "learning_rate": 7.275841525171284e-06,
      "loss": 1.2154,
      "num_input_tokens_seen": 1136640,
      "step": 1955
    },
    {
      "epoch": 0.29192731605600236,
      "grad_norm": 18.625,
      "learning_rate": 7.294459338695264e-06,
      "loss": 1.0295,
      "num_input_tokens_seen": 1139744,
      "step": 1960
    },
    {
      "epoch": 0.29267202859696156,
      "grad_norm": 12.6875,
      "learning_rate": 7.313077152219244e-06,
      "loss": 1.2983,
      "num_input_tokens_seen": 1142688,
      "step": 1965
    },
    {
      "epoch": 0.29341674113792077,
      "grad_norm": 9.75,
      "learning_rate": 7.331694965743223e-06,
      "loss": 1.3751,
      "num_input_tokens_seen": 1145824,
      "step": 1970
    },
    {
      "epoch": 0.29416145367887997,
      "grad_norm": 14.375,
      "learning_rate": 7.350312779267203e-06,
      "loss": 1.0497,
      "num_input_tokens_seen": 1148544,
      "step": 1975
    },
    {
      "epoch": 0.2949061662198391,
      "grad_norm": 21.25,
      "learning_rate": 7.368930592791183e-06,
      "loss": 1.617,
      "num_input_tokens_seen": 1151328,
      "step": 1980
    },
    {
      "epoch": 0.2956508787607983,
      "grad_norm": 16.625,
      "learning_rate": 7.387548406315163e-06,
      "loss": 0.9439,
      "num_input_tokens_seen": 1153952,
      "step": 1985
    },
    {
      "epoch": 0.2963955913017575,
      "grad_norm": 16.625,
      "learning_rate": 7.406166219839142e-06,
      "loss": 1.2023,
      "num_input_tokens_seen": 1156800,
      "step": 1990
    },
    {
      "epoch": 0.2971403038427167,
      "grad_norm": 22.625,
      "learning_rate": 7.424784033363122e-06,
      "loss": 1.2718,
      "num_input_tokens_seen": 1159808,
      "step": 1995
    },
    {
      "epoch": 0.2978850163836759,
      "grad_norm": 10.625,
      "learning_rate": 7.443401846887102e-06,
      "loss": 1.0962,
      "num_input_tokens_seen": 1162848,
      "step": 2000
    },
    {
      "epoch": 0.2986297289246351,
      "grad_norm": 10.1875,
      "learning_rate": 7.462019660411082e-06,
      "loss": 1.4088,
      "num_input_tokens_seen": 1166208,
      "step": 2005
    },
    {
      "epoch": 0.2993744414655943,
      "grad_norm": 10.9375,
      "learning_rate": 7.480637473935062e-06,
      "loss": 1.0125,
      "num_input_tokens_seen": 1168928,
      "step": 2010
    },
    {
      "epoch": 0.3001191540065535,
      "grad_norm": 12.4375,
      "learning_rate": 7.499255287459041e-06,
      "loss": 1.3303,
      "num_input_tokens_seen": 1171712,
      "step": 2015
    },
    {
      "epoch": 0.3008638665475127,
      "grad_norm": 8.1875,
      "learning_rate": 7.517873100983021e-06,
      "loss": 1.1505,
      "num_input_tokens_seen": 1174368,
      "step": 2020
    },
    {
      "epoch": 0.30160857908847183,
      "grad_norm": 10.8125,
      "learning_rate": 7.5364909145070005e-06,
      "loss": 0.8916,
      "num_input_tokens_seen": 1177184,
      "step": 2025
    },
    {
      "epoch": 0.30235329162943103,
      "grad_norm": 10.6875,
      "learning_rate": 7.555108728030981e-06,
      "loss": 1.2394,
      "num_input_tokens_seen": 1179872,
      "step": 2030
    },
    {
      "epoch": 0.30309800417039023,
      "grad_norm": 19.5,
      "learning_rate": 7.5737265415549595e-06,
      "loss": 1.3911,
      "num_input_tokens_seen": 1182464,
      "step": 2035
    },
    {
      "epoch": 0.30384271671134944,
      "grad_norm": 12.125,
      "learning_rate": 7.5923443550789394e-06,
      "loss": 1.2193,
      "num_input_tokens_seen": 1185632,
      "step": 2040
    },
    {
      "epoch": 0.3045874292523086,
      "grad_norm": 8.3125,
      "learning_rate": 7.610962168602919e-06,
      "loss": 0.9996,
      "num_input_tokens_seen": 1188416,
      "step": 2045
    },
    {
      "epoch": 0.3053321417932678,
      "grad_norm": 21.125,
      "learning_rate": 7.6295799821269e-06,
      "loss": 1.202,
      "num_input_tokens_seen": 1191648,
      "step": 2050
    },
    {
      "epoch": 0.306076854334227,
      "grad_norm": 11.6875,
      "learning_rate": 7.64819779565088e-06,
      "loss": 1.506,
      "num_input_tokens_seen": 1194432,
      "step": 2055
    },
    {
      "epoch": 0.3068215668751862,
      "grad_norm": 10.75,
      "learning_rate": 7.666815609174859e-06,
      "loss": 1.2976,
      "num_input_tokens_seen": 1197248,
      "step": 2060
    },
    {
      "epoch": 0.3075662794161454,
      "grad_norm": 13.4375,
      "learning_rate": 7.685433422698839e-06,
      "loss": 1.3177,
      "num_input_tokens_seen": 1200256,
      "step": 2065
    },
    {
      "epoch": 0.30831099195710454,
      "grad_norm": 11.125,
      "learning_rate": 7.704051236222819e-06,
      "loss": 1.8297,
      "num_input_tokens_seen": 1203040,
      "step": 2070
    },
    {
      "epoch": 0.30905570449806374,
      "grad_norm": 10.0625,
      "learning_rate": 7.722669049746799e-06,
      "loss": 1.1106,
      "num_input_tokens_seen": 1206080,
      "step": 2075
    },
    {
      "epoch": 0.30980041703902295,
      "grad_norm": 12.6875,
      "learning_rate": 7.741286863270777e-06,
      "loss": 0.9714,
      "num_input_tokens_seen": 1208704,
      "step": 2080
    },
    {
      "epoch": 0.31054512957998215,
      "grad_norm": 14.875,
      "learning_rate": 7.759904676794757e-06,
      "loss": 1.2002,
      "num_input_tokens_seen": 1211424,
      "step": 2085
    },
    {
      "epoch": 0.3112898421209413,
      "grad_norm": 6.34375,
      "learning_rate": 7.778522490318737e-06,
      "loss": 0.9219,
      "num_input_tokens_seen": 1214272,
      "step": 2090
    },
    {
      "epoch": 0.3120345546619005,
      "grad_norm": 15.875,
      "learning_rate": 7.797140303842717e-06,
      "loss": 1.519,
      "num_input_tokens_seen": 1217280,
      "step": 2095
    },
    {
      "epoch": 0.3127792672028597,
      "grad_norm": 26.75,
      "learning_rate": 7.815758117366697e-06,
      "loss": 0.9518,
      "num_input_tokens_seen": 1220064,
      "step": 2100
    },
    {
      "epoch": 0.3135239797438189,
      "grad_norm": 13.8125,
      "learning_rate": 7.834375930890677e-06,
      "loss": 0.8415,
      "num_input_tokens_seen": 1222848,
      "step": 2105
    },
    {
      "epoch": 0.31426869228477805,
      "grad_norm": 5.15625,
      "learning_rate": 7.852993744414657e-06,
      "loss": 0.9891,
      "num_input_tokens_seen": 1225760,
      "step": 2110
    },
    {
      "epoch": 0.31501340482573725,
      "grad_norm": 14.75,
      "learning_rate": 7.871611557938637e-06,
      "loss": 1.3537,
      "num_input_tokens_seen": 1228864,
      "step": 2115
    },
    {
      "epoch": 0.31575811736669646,
      "grad_norm": 19.0,
      "learning_rate": 7.890229371462615e-06,
      "loss": 0.9858,
      "num_input_tokens_seen": 1231712,
      "step": 2120
    },
    {
      "epoch": 0.31650282990765566,
      "grad_norm": 21.0,
      "learning_rate": 7.908847184986595e-06,
      "loss": 1.003,
      "num_input_tokens_seen": 1234720,
      "step": 2125
    },
    {
      "epoch": 0.31724754244861486,
      "grad_norm": 15.25,
      "learning_rate": 7.927464998510575e-06,
      "loss": 1.043,
      "num_input_tokens_seen": 1237504,
      "step": 2130
    },
    {
      "epoch": 0.317992254989574,
      "grad_norm": 11.6875,
      "learning_rate": 7.946082812034555e-06,
      "loss": 1.1006,
      "num_input_tokens_seen": 1240256,
      "step": 2135
    },
    {
      "epoch": 0.3187369675305332,
      "grad_norm": 11.5625,
      "learning_rate": 7.964700625558536e-06,
      "loss": 1.2911,
      "num_input_tokens_seen": 1243264,
      "step": 2140
    },
    {
      "epoch": 0.3194816800714924,
      "grad_norm": 14.875,
      "learning_rate": 7.983318439082515e-06,
      "loss": 1.2312,
      "num_input_tokens_seen": 1246304,
      "step": 2145
    },
    {
      "epoch": 0.3202263926124516,
      "grad_norm": 13.1875,
      "learning_rate": 8.001936252606494e-06,
      "loss": 1.1923,
      "num_input_tokens_seen": 1249184,
      "step": 2150
    },
    {
      "epoch": 0.32097110515341076,
      "grad_norm": 29.0,
      "learning_rate": 8.020554066130474e-06,
      "loss": 1.5031,
      "num_input_tokens_seen": 1252064,
      "step": 2155
    },
    {
      "epoch": 0.32171581769436997,
      "grad_norm": 13.3125,
      "learning_rate": 8.039171879654454e-06,
      "loss": 1.1498,
      "num_input_tokens_seen": 1255200,
      "step": 2160
    },
    {
      "epoch": 0.32246053023532917,
      "grad_norm": 9.875,
      "learning_rate": 8.057789693178433e-06,
      "loss": 1.3571,
      "num_input_tokens_seen": 1258048,
      "step": 2165
    },
    {
      "epoch": 0.32320524277628837,
      "grad_norm": 16.0,
      "learning_rate": 8.076407506702412e-06,
      "loss": 1.1438,
      "num_input_tokens_seen": 1261024,
      "step": 2170
    },
    {
      "epoch": 0.3239499553172475,
      "grad_norm": 8.6875,
      "learning_rate": 8.095025320226392e-06,
      "loss": 1.2469,
      "num_input_tokens_seen": 1263872,
      "step": 2175
    },
    {
      "epoch": 0.3246946678582067,
      "grad_norm": 14.4375,
      "learning_rate": 8.113643133750374e-06,
      "loss": 0.9568,
      "num_input_tokens_seen": 1266912,
      "step": 2180
    },
    {
      "epoch": 0.3254393803991659,
      "grad_norm": 25.5,
      "learning_rate": 8.132260947274352e-06,
      "loss": 1.0968,
      "num_input_tokens_seen": 1269728,
      "step": 2185
    },
    {
      "epoch": 0.3261840929401251,
      "grad_norm": 19.125,
      "learning_rate": 8.150878760798332e-06,
      "loss": 1.2276,
      "num_input_tokens_seen": 1272416,
      "step": 2190
    },
    {
      "epoch": 0.32692880548108433,
      "grad_norm": 10.5625,
      "learning_rate": 8.169496574322312e-06,
      "loss": 1.1608,
      "num_input_tokens_seen": 1275552,
      "step": 2195
    },
    {
      "epoch": 0.3276735180220435,
      "grad_norm": 14.1875,
      "learning_rate": 8.188114387846292e-06,
      "loss": 0.9361,
      "num_input_tokens_seen": 1278752,
      "step": 2200
    },
    {
      "epoch": 0.3284182305630027,
      "grad_norm": 17.125,
      "learning_rate": 8.206732201370272e-06,
      "loss": 1.6388,
      "num_input_tokens_seen": 1281728,
      "step": 2205
    },
    {
      "epoch": 0.3291629431039619,
      "grad_norm": 9.375,
      "learning_rate": 8.22535001489425e-06,
      "loss": 1.1216,
      "num_input_tokens_seen": 1284608,
      "step": 2210
    },
    {
      "epoch": 0.3299076556449211,
      "grad_norm": 24.125,
      "learning_rate": 8.24396782841823e-06,
      "loss": 1.2825,
      "num_input_tokens_seen": 1287648,
      "step": 2215
    },
    {
      "epoch": 0.33065236818588023,
      "grad_norm": 16.125,
      "learning_rate": 8.262585641942212e-06,
      "loss": 1.1945,
      "num_input_tokens_seen": 1290496,
      "step": 2220
    },
    {
      "epoch": 0.33139708072683943,
      "grad_norm": 14.5625,
      "learning_rate": 8.281203455466192e-06,
      "loss": 0.8266,
      "num_input_tokens_seen": 1293344,
      "step": 2225
    },
    {
      "epoch": 0.33214179326779864,
      "grad_norm": 20.375,
      "learning_rate": 8.29982126899017e-06,
      "loss": 1.0535,
      "num_input_tokens_seen": 1296192,
      "step": 2230
    },
    {
      "epoch": 0.33288650580875784,
      "grad_norm": 11.75,
      "learning_rate": 8.31843908251415e-06,
      "loss": 1.208,
      "num_input_tokens_seen": 1298848,
      "step": 2235
    },
    {
      "epoch": 0.333631218349717,
      "grad_norm": 18.125,
      "learning_rate": 8.33705689603813e-06,
      "loss": 1.2211,
      "num_input_tokens_seen": 1301536,
      "step": 2240
    },
    {
      "epoch": 0.3343759308906762,
      "grad_norm": 19.5,
      "learning_rate": 8.35567470956211e-06,
      "loss": 1.4194,
      "num_input_tokens_seen": 1304448,
      "step": 2245
    },
    {
      "epoch": 0.3351206434316354,
      "grad_norm": 9.0,
      "learning_rate": 8.374292523086088e-06,
      "loss": 1.2034,
      "num_input_tokens_seen": 1307328,
      "step": 2250
    },
    {
      "epoch": 0.3358653559725946,
      "grad_norm": 6.46875,
      "learning_rate": 8.392910336610068e-06,
      "loss": 0.7829,
      "num_input_tokens_seen": 1310368,
      "step": 2255
    },
    {
      "epoch": 0.3366100685135538,
      "grad_norm": 15.625,
      "learning_rate": 8.41152815013405e-06,
      "loss": 1.07,
      "num_input_tokens_seen": 1313216,
      "step": 2260
    },
    {
      "epoch": 0.33735478105451294,
      "grad_norm": 17.75,
      "learning_rate": 8.43014596365803e-06,
      "loss": 1.2678,
      "num_input_tokens_seen": 1316256,
      "step": 2265
    },
    {
      "epoch": 0.33809949359547214,
      "grad_norm": 10.75,
      "learning_rate": 8.44876377718201e-06,
      "loss": 1.1672,
      "num_input_tokens_seen": 1318976,
      "step": 2270
    },
    {
      "epoch": 0.33884420613643135,
      "grad_norm": 17.75,
      "learning_rate": 8.467381590705988e-06,
      "loss": 1.1669,
      "num_input_tokens_seen": 1321632,
      "step": 2275
    },
    {
      "epoch": 0.33958891867739055,
      "grad_norm": 12.5,
      "learning_rate": 8.485999404229967e-06,
      "loss": 1.0159,
      "num_input_tokens_seen": 1324256,
      "step": 2280
    },
    {
      "epoch": 0.3403336312183497,
      "grad_norm": 12.375,
      "learning_rate": 8.504617217753947e-06,
      "loss": 1.358,
      "num_input_tokens_seen": 1327008,
      "step": 2285
    },
    {
      "epoch": 0.3410783437593089,
      "grad_norm": 15.375,
      "learning_rate": 8.523235031277927e-06,
      "loss": 1.033,
      "num_input_tokens_seen": 1330016,
      "step": 2290
    },
    {
      "epoch": 0.3418230563002681,
      "grad_norm": 18.25,
      "learning_rate": 8.541852844801907e-06,
      "loss": 1.2132,
      "num_input_tokens_seen": 1333024,
      "step": 2295
    },
    {
      "epoch": 0.3425677688412273,
      "grad_norm": 16.75,
      "learning_rate": 8.560470658325887e-06,
      "loss": 1.117,
      "num_input_tokens_seen": 1335872,
      "step": 2300
    },
    {
      "epoch": 0.34331248138218645,
      "grad_norm": 16.5,
      "learning_rate": 8.579088471849867e-06,
      "loss": 1.5749,
      "num_input_tokens_seen": 1338560,
      "step": 2305
    },
    {
      "epoch": 0.34405719392314565,
      "grad_norm": 6.5,
      "learning_rate": 8.597706285373847e-06,
      "loss": 0.9675,
      "num_input_tokens_seen": 1341248,
      "step": 2310
    },
    {
      "epoch": 0.34480190646410486,
      "grad_norm": 24.375,
      "learning_rate": 8.616324098897825e-06,
      "loss": 1.2109,
      "num_input_tokens_seen": 1343936,
      "step": 2315
    },
    {
      "epoch": 0.34554661900506406,
      "grad_norm": 9.625,
      "learning_rate": 8.634941912421805e-06,
      "loss": 1.1754,
      "num_input_tokens_seen": 1346880,
      "step": 2320
    },
    {
      "epoch": 0.34629133154602326,
      "grad_norm": 17.75,
      "learning_rate": 8.653559725945785e-06,
      "loss": 1.404,
      "num_input_tokens_seen": 1349824,
      "step": 2325
    },
    {
      "epoch": 0.3470360440869824,
      "grad_norm": 17.75,
      "learning_rate": 8.672177539469765e-06,
      "loss": 1.5986,
      "num_input_tokens_seen": 1352576,
      "step": 2330
    },
    {
      "epoch": 0.3477807566279416,
      "grad_norm": 10.3125,
      "learning_rate": 8.690795352993745e-06,
      "loss": 0.9659,
      "num_input_tokens_seen": 1355872,
      "step": 2335
    },
    {
      "epoch": 0.3485254691689008,
      "grad_norm": 11.0,
      "learning_rate": 8.709413166517725e-06,
      "loss": 0.9842,
      "num_input_tokens_seen": 1358816,
      "step": 2340
    },
    {
      "epoch": 0.34927018170986,
      "grad_norm": 21.75,
      "learning_rate": 8.728030980041705e-06,
      "loss": 1.2947,
      "num_input_tokens_seen": 1362144,
      "step": 2345
    },
    {
      "epoch": 0.35001489425081916,
      "grad_norm": 13.5625,
      "learning_rate": 8.746648793565685e-06,
      "loss": 1.2017,
      "num_input_tokens_seen": 1365120,
      "step": 2350
    },
    {
      "epoch": 0.35075960679177837,
      "grad_norm": 19.5,
      "learning_rate": 8.765266607089665e-06,
      "loss": 1.2536,
      "num_input_tokens_seen": 1368192,
      "step": 2355
    },
    {
      "epoch": 0.35150431933273757,
      "grad_norm": 10.5625,
      "learning_rate": 8.783884420613643e-06,
      "loss": 1.1812,
      "num_input_tokens_seen": 1371200,
      "step": 2360
    },
    {
      "epoch": 0.35224903187369677,
      "grad_norm": 10.5625,
      "learning_rate": 8.802502234137623e-06,
      "loss": 1.1266,
      "num_input_tokens_seen": 1374112,
      "step": 2365
    },
    {
      "epoch": 0.3529937444146559,
      "grad_norm": 14.0625,
      "learning_rate": 8.821120047661603e-06,
      "loss": 1.6997,
      "num_input_tokens_seen": 1376640,
      "step": 2370
    },
    {
      "epoch": 0.3537384569556151,
      "grad_norm": 12.125,
      "learning_rate": 8.839737861185583e-06,
      "loss": 1.3726,
      "num_input_tokens_seen": 1379616,
      "step": 2375
    },
    {
      "epoch": 0.3544831694965743,
      "grad_norm": 16.875,
      "learning_rate": 8.858355674709563e-06,
      "loss": 1.5228,
      "num_input_tokens_seen": 1382304,
      "step": 2380
    },
    {
      "epoch": 0.3552278820375335,
      "grad_norm": 13.125,
      "learning_rate": 8.876973488233543e-06,
      "loss": 1.2451,
      "num_input_tokens_seen": 1385184,
      "step": 2385
    },
    {
      "epoch": 0.35597259457849273,
      "grad_norm": 9.375,
      "learning_rate": 8.895591301757522e-06,
      "loss": 0.8602,
      "num_input_tokens_seen": 1388224,
      "step": 2390
    },
    {
      "epoch": 0.3567173071194519,
      "grad_norm": 11.25,
      "learning_rate": 8.914209115281502e-06,
      "loss": 0.9647,
      "num_input_tokens_seen": 1391040,
      "step": 2395
    },
    {
      "epoch": 0.3574620196604111,
      "grad_norm": 9.75,
      "learning_rate": 8.93282692880548e-06,
      "loss": 1.2537,
      "num_input_tokens_seen": 1393792,
      "step": 2400
    },
    {
      "epoch": 0.3582067322013703,
      "grad_norm": 11.6875,
      "learning_rate": 8.95144474232946e-06,
      "loss": 1.1287,
      "num_input_tokens_seen": 1396576,
      "step": 2405
    },
    {
      "epoch": 0.3589514447423295,
      "grad_norm": 11.0625,
      "learning_rate": 8.97006255585344e-06,
      "loss": 1.1396,
      "num_input_tokens_seen": 1399552,
      "step": 2410
    },
    {
      "epoch": 0.35969615728328863,
      "grad_norm": 20.875,
      "learning_rate": 8.98868036937742e-06,
      "loss": 1.1688,
      "num_input_tokens_seen": 1402368,
      "step": 2415
    },
    {
      "epoch": 0.36044086982424783,
      "grad_norm": 14.5,
      "learning_rate": 9.0072981829014e-06,
      "loss": 1.5739,
      "num_input_tokens_seen": 1405344,
      "step": 2420
    },
    {
      "epoch": 0.36118558236520704,
      "grad_norm": 17.375,
      "learning_rate": 9.02591599642538e-06,
      "loss": 1.3424,
      "num_input_tokens_seen": 1408064,
      "step": 2425
    },
    {
      "epoch": 0.36193029490616624,
      "grad_norm": 19.375,
      "learning_rate": 9.04453380994936e-06,
      "loss": 1.2722,
      "num_input_tokens_seen": 1410944,
      "step": 2430
    },
    {
      "epoch": 0.3626750074471254,
      "grad_norm": 13.125,
      "learning_rate": 9.06315162347334e-06,
      "loss": 1.3758,
      "num_input_tokens_seen": 1413856,
      "step": 2435
    },
    {
      "epoch": 0.3634197199880846,
      "grad_norm": 14.9375,
      "learning_rate": 9.08176943699732e-06,
      "loss": 0.9152,
      "num_input_tokens_seen": 1416864,
      "step": 2440
    },
    {
      "epoch": 0.3641644325290438,
      "grad_norm": 15.6875,
      "learning_rate": 9.100387250521298e-06,
      "loss": 1.3194,
      "num_input_tokens_seen": 1419680,
      "step": 2445
    },
    {
      "epoch": 0.364909145070003,
      "grad_norm": 9.375,
      "learning_rate": 9.119005064045278e-06,
      "loss": 0.8869,
      "num_input_tokens_seen": 1422656,
      "step": 2450
    },
    {
      "epoch": 0.3656538576109622,
      "grad_norm": 13.875,
      "learning_rate": 9.137622877569258e-06,
      "loss": 1.2531,
      "num_input_tokens_seen": 1425504,
      "step": 2455
    },
    {
      "epoch": 0.36639857015192134,
      "grad_norm": 16.125,
      "learning_rate": 9.156240691093238e-06,
      "loss": 1.2913,
      "num_input_tokens_seen": 1428704,
      "step": 2460
    },
    {
      "epoch": 0.36714328269288055,
      "grad_norm": 12.6875,
      "learning_rate": 9.174858504617218e-06,
      "loss": 1.2746,
      "num_input_tokens_seen": 1431744,
      "step": 2465
    },
    {
      "epoch": 0.36788799523383975,
      "grad_norm": 9.875,
      "learning_rate": 9.193476318141198e-06,
      "loss": 1.1912,
      "num_input_tokens_seen": 1434880,
      "step": 2470
    },
    {
      "epoch": 0.36863270777479895,
      "grad_norm": 5.53125,
      "learning_rate": 9.212094131665178e-06,
      "loss": 0.894,
      "num_input_tokens_seen": 1437728,
      "step": 2475
    },
    {
      "epoch": 0.3693774203157581,
      "grad_norm": 17.75,
      "learning_rate": 9.230711945189158e-06,
      "loss": 1.1068,
      "num_input_tokens_seen": 1440672,
      "step": 2480
    },
    {
      "epoch": 0.3701221328567173,
      "grad_norm": 6.90625,
      "learning_rate": 9.249329758713138e-06,
      "loss": 0.817,
      "num_input_tokens_seen": 1443424,
      "step": 2485
    },
    {
      "epoch": 0.3708668453976765,
      "grad_norm": 15.3125,
      "learning_rate": 9.267947572237116e-06,
      "loss": 1.1557,
      "num_input_tokens_seen": 1446368,
      "step": 2490
    },
    {
      "epoch": 0.3716115579386357,
      "grad_norm": 7.09375,
      "learning_rate": 9.286565385761096e-06,
      "loss": 0.9719,
      "num_input_tokens_seen": 1449184,
      "step": 2495
    },
    {
      "epoch": 0.37235627047959485,
      "grad_norm": 25.5,
      "learning_rate": 9.305183199285077e-06,
      "loss": 1.2246,
      "num_input_tokens_seen": 1452096,
      "step": 2500
    },
    {
      "epoch": 0.37310098302055406,
      "grad_norm": 21.875,
      "learning_rate": 9.323801012809057e-06,
      "loss": 1.4629,
      "num_input_tokens_seen": 1455040,
      "step": 2505
    },
    {
      "epoch": 0.37384569556151326,
      "grad_norm": 9.75,
      "learning_rate": 9.342418826333036e-06,
      "loss": 0.8399,
      "num_input_tokens_seen": 1457984,
      "step": 2510
    },
    {
      "epoch": 0.37459040810247246,
      "grad_norm": 9.9375,
      "learning_rate": 9.361036639857016e-06,
      "loss": 0.9391,
      "num_input_tokens_seen": 1460896,
      "step": 2515
    },
    {
      "epoch": 0.3753351206434316,
      "grad_norm": 11.375,
      "learning_rate": 9.379654453380995e-06,
      "loss": 0.921,
      "num_input_tokens_seen": 1463776,
      "step": 2520
    },
    {
      "epoch": 0.3760798331843908,
      "grad_norm": 14.25,
      "learning_rate": 9.398272266904975e-06,
      "loss": 1.5743,
      "num_input_tokens_seen": 1466464,
      "step": 2525
    },
    {
      "epoch": 0.37682454572535,
      "grad_norm": 15.625,
      "learning_rate": 9.416890080428954e-06,
      "loss": 1.107,
      "num_input_tokens_seen": 1469472,
      "step": 2530
    },
    {
      "epoch": 0.3775692582663092,
      "grad_norm": 12.3125,
      "learning_rate": 9.435507893952934e-06,
      "loss": 1.1676,
      "num_input_tokens_seen": 1472256,
      "step": 2535
    },
    {
      "epoch": 0.3783139708072684,
      "grad_norm": 17.75,
      "learning_rate": 9.454125707476915e-06,
      "loss": 1.0931,
      "num_input_tokens_seen": 1475392,
      "step": 2540
    },
    {
      "epoch": 0.37905868334822757,
      "grad_norm": 7.34375,
      "learning_rate": 9.472743521000895e-06,
      "loss": 0.977,
      "num_input_tokens_seen": 1477856,
      "step": 2545
    },
    {
      "epoch": 0.37980339588918677,
      "grad_norm": 10.8125,
      "learning_rate": 9.491361334524875e-06,
      "loss": 1.2531,
      "num_input_tokens_seen": 1480736,
      "step": 2550
    },
    {
      "epoch": 0.38054810843014597,
      "grad_norm": 5.96875,
      "learning_rate": 9.509979148048853e-06,
      "loss": 1.2117,
      "num_input_tokens_seen": 1483680,
      "step": 2555
    },
    {
      "epoch": 0.3812928209711052,
      "grad_norm": 21.25,
      "learning_rate": 9.528596961572833e-06,
      "loss": 1.0791,
      "num_input_tokens_seen": 1486560,
      "step": 2560
    },
    {
      "epoch": 0.3820375335120643,
      "grad_norm": 15.3125,
      "learning_rate": 9.547214775096813e-06,
      "loss": 1.1541,
      "num_input_tokens_seen": 1489376,
      "step": 2565
    },
    {
      "epoch": 0.3827822460530235,
      "grad_norm": 5.65625,
      "learning_rate": 9.565832588620793e-06,
      "loss": 0.9481,
      "num_input_tokens_seen": 1492448,
      "step": 2570
    },
    {
      "epoch": 0.3835269585939827,
      "grad_norm": 16.25,
      "learning_rate": 9.584450402144771e-06,
      "loss": 1.446,
      "num_input_tokens_seen": 1495424,
      "step": 2575
    },
    {
      "epoch": 0.38427167113494193,
      "grad_norm": 12.6875,
      "learning_rate": 9.603068215668753e-06,
      "loss": 0.9572,
      "num_input_tokens_seen": 1498368,
      "step": 2580
    },
    {
      "epoch": 0.3850163836759011,
      "grad_norm": 14.75,
      "learning_rate": 9.621686029192733e-06,
      "loss": 1.3153,
      "num_input_tokens_seen": 1501152,
      "step": 2585
    },
    {
      "epoch": 0.3857610962168603,
      "grad_norm": 19.25,
      "learning_rate": 9.640303842716713e-06,
      "loss": 1.4003,
      "num_input_tokens_seen": 1504096,
      "step": 2590
    },
    {
      "epoch": 0.3865058087578195,
      "grad_norm": 12.9375,
      "learning_rate": 9.658921656240691e-06,
      "loss": 1.1629,
      "num_input_tokens_seen": 1507136,
      "step": 2595
    },
    {
      "epoch": 0.3872505212987787,
      "grad_norm": 9.3125,
      "learning_rate": 9.677539469764671e-06,
      "loss": 1.022,
      "num_input_tokens_seen": 1509888,
      "step": 2600
    },
    {
      "epoch": 0.3879952338397379,
      "grad_norm": 8.3125,
      "learning_rate": 9.69615728328865e-06,
      "loss": 1.1015,
      "num_input_tokens_seen": 1512960,
      "step": 2605
    },
    {
      "epoch": 0.38873994638069703,
      "grad_norm": 10.0,
      "learning_rate": 9.71477509681263e-06,
      "loss": 1.4068,
      "num_input_tokens_seen": 1515648,
      "step": 2610
    },
    {
      "epoch": 0.38948465892165623,
      "grad_norm": 5.6875,
      "learning_rate": 9.73339291033661e-06,
      "loss": 1.1436,
      "num_input_tokens_seen": 1518432,
      "step": 2615
    },
    {
      "epoch": 0.39022937146261544,
      "grad_norm": 10.9375,
      "learning_rate": 9.75201072386059e-06,
      "loss": 1.7331,
      "num_input_tokens_seen": 1521152,
      "step": 2620
    },
    {
      "epoch": 0.39097408400357464,
      "grad_norm": 16.5,
      "learning_rate": 9.77062853738457e-06,
      "loss": 1.3405,
      "num_input_tokens_seen": 1523744,
      "step": 2625
    },
    {
      "epoch": 0.3917187965445338,
      "grad_norm": 10.9375,
      "learning_rate": 9.78924635090855e-06,
      "loss": 1.0438,
      "num_input_tokens_seen": 1526720,
      "step": 2630
    },
    {
      "epoch": 0.392463509085493,
      "grad_norm": 17.75,
      "learning_rate": 9.80786416443253e-06,
      "loss": 0.9498,
      "num_input_tokens_seen": 1529792,
      "step": 2635
    },
    {
      "epoch": 0.3932082216264522,
      "grad_norm": 16.875,
      "learning_rate": 9.826481977956509e-06,
      "loss": 1.3155,
      "num_input_tokens_seen": 1532544,
      "step": 2640
    },
    {
      "epoch": 0.3939529341674114,
      "grad_norm": 7.21875,
      "learning_rate": 9.845099791480489e-06,
      "loss": 1.3256,
      "num_input_tokens_seen": 1535616,
      "step": 2645
    },
    {
      "epoch": 0.39469764670837054,
      "grad_norm": 10.5,
      "learning_rate": 9.863717605004468e-06,
      "loss": 1.0696,
      "num_input_tokens_seen": 1539008,
      "step": 2650
    },
    {
      "epoch": 0.39544235924932974,
      "grad_norm": 8.875,
      "learning_rate": 9.882335418528448e-06,
      "loss": 1.1243,
      "num_input_tokens_seen": 1541792,
      "step": 2655
    },
    {
      "epoch": 0.39618707179028895,
      "grad_norm": 17.125,
      "learning_rate": 9.900953232052428e-06,
      "loss": 1.3793,
      "num_input_tokens_seen": 1544864,
      "step": 2660
    },
    {
      "epoch": 0.39693178433124815,
      "grad_norm": 4.40625,
      "learning_rate": 9.919571045576408e-06,
      "loss": 1.0973,
      "num_input_tokens_seen": 1547904,
      "step": 2665
    },
    {
      "epoch": 0.39767649687220735,
      "grad_norm": 18.25,
      "learning_rate": 9.938188859100388e-06,
      "loss": 0.9827,
      "num_input_tokens_seen": 1550688,
      "step": 2670
    },
    {
      "epoch": 0.3984212094131665,
      "grad_norm": 14.875,
      "learning_rate": 9.956806672624368e-06,
      "loss": 1.1913,
      "num_input_tokens_seen": 1553824,
      "step": 2675
    },
    {
      "epoch": 0.3991659219541257,
      "grad_norm": 17.375,
      "learning_rate": 9.975424486148348e-06,
      "loss": 0.8289,
      "num_input_tokens_seen": 1556576,
      "step": 2680
    },
    {
      "epoch": 0.3999106344950849,
      "grad_norm": 17.0,
      "learning_rate": 9.994042299672326e-06,
      "loss": 1.2393,
      "num_input_tokens_seen": 1559296,
      "step": 2685
    },
    {
      "epoch": 0.4006553470360441,
      "grad_norm": 12.875,
      "learning_rate": 1.0012660113196306e-05,
      "loss": 1.2376,
      "num_input_tokens_seen": 1562240,
      "step": 2690
    },
    {
      "epoch": 0.40140005957700325,
      "grad_norm": 19.25,
      "learning_rate": 1.0031277926720286e-05,
      "loss": 1.0056,
      "num_input_tokens_seen": 1565088,
      "step": 2695
    },
    {
      "epoch": 0.40214477211796246,
      "grad_norm": 12.875,
      "learning_rate": 1.0049895740244266e-05,
      "loss": 1.0616,
      "num_input_tokens_seen": 1568000,
      "step": 2700
    },
    {
      "epoch": 0.40288948465892166,
      "grad_norm": 10.1875,
      "learning_rate": 1.0068513553768246e-05,
      "loss": 1.0627,
      "num_input_tokens_seen": 1570912,
      "step": 2705
    },
    {
      "epoch": 0.40363419719988086,
      "grad_norm": 12.5625,
      "learning_rate": 1.0087131367292226e-05,
      "loss": 1.3034,
      "num_input_tokens_seen": 1573888,
      "step": 2710
    },
    {
      "epoch": 0.40437890974084,
      "grad_norm": 14.4375,
      "learning_rate": 1.0105749180816206e-05,
      "loss": 1.2514,
      "num_input_tokens_seen": 1576672,
      "step": 2715
    },
    {
      "epoch": 0.4051236222817992,
      "grad_norm": 17.625,
      "learning_rate": 1.0124366994340186e-05,
      "loss": 1.2703,
      "num_input_tokens_seen": 1579456,
      "step": 2720
    },
    {
      "epoch": 0.4058683348227584,
      "grad_norm": 14.625,
      "learning_rate": 1.0142984807864164e-05,
      "loss": 0.8661,
      "num_input_tokens_seen": 1581952,
      "step": 2725
    },
    {
      "epoch": 0.4066130473637176,
      "grad_norm": 16.375,
      "learning_rate": 1.0161602621388144e-05,
      "loss": 1.3153,
      "num_input_tokens_seen": 1584864,
      "step": 2730
    },
    {
      "epoch": 0.4073577599046768,
      "grad_norm": 11.6875,
      "learning_rate": 1.0180220434912124e-05,
      "loss": 1.1266,
      "num_input_tokens_seen": 1588160,
      "step": 2735
    },
    {
      "epoch": 0.40810247244563597,
      "grad_norm": 15.75,
      "learning_rate": 1.0198838248436104e-05,
      "loss": 1.6869,
      "num_input_tokens_seen": 1590880,
      "step": 2740
    },
    {
      "epoch": 0.40884718498659517,
      "grad_norm": 11.5625,
      "learning_rate": 1.0217456061960085e-05,
      "loss": 1.0561,
      "num_input_tokens_seen": 1593600,
      "step": 2745
    },
    {
      "epoch": 0.40959189752755437,
      "grad_norm": 16.625,
      "learning_rate": 1.0236073875484064e-05,
      "loss": 1.3334,
      "num_input_tokens_seen": 1596512,
      "step": 2750
    },
    {
      "epoch": 0.4103366100685136,
      "grad_norm": 16.125,
      "learning_rate": 1.0254691689008044e-05,
      "loss": 1.2198,
      "num_input_tokens_seen": 1599296,
      "step": 2755
    },
    {
      "epoch": 0.4110813226094727,
      "grad_norm": 15.0625,
      "learning_rate": 1.0273309502532023e-05,
      "loss": 1.0951,
      "num_input_tokens_seen": 1602208,
      "step": 2760
    },
    {
      "epoch": 0.4118260351504319,
      "grad_norm": 12.125,
      "learning_rate": 1.0291927316056003e-05,
      "loss": 1.399,
      "num_input_tokens_seen": 1605184,
      "step": 2765
    },
    {
      "epoch": 0.4125707476913911,
      "grad_norm": 9.4375,
      "learning_rate": 1.0310545129579982e-05,
      "loss": 0.964,
      "num_input_tokens_seen": 1608032,
      "step": 2770
    },
    {
      "epoch": 0.41331546023235033,
      "grad_norm": 11.125,
      "learning_rate": 1.0329162943103962e-05,
      "loss": 1.2925,
      "num_input_tokens_seen": 1611104,
      "step": 2775
    },
    {
      "epoch": 0.4140601727733095,
      "grad_norm": 13.8125,
      "learning_rate": 1.0347780756627941e-05,
      "loss": 1.4255,
      "num_input_tokens_seen": 1614048,
      "step": 2780
    },
    {
      "epoch": 0.4148048853142687,
      "grad_norm": 11.625,
      "learning_rate": 1.0366398570151923e-05,
      "loss": 1.3745,
      "num_input_tokens_seen": 1617056,
      "step": 2785
    },
    {
      "epoch": 0.4155495978552279,
      "grad_norm": 19.25,
      "learning_rate": 1.0385016383675901e-05,
      "loss": 1.3973,
      "num_input_tokens_seen": 1619680,
      "step": 2790
    },
    {
      "epoch": 0.4162943103961871,
      "grad_norm": 7.65625,
      "learning_rate": 1.0403634197199881e-05,
      "loss": 0.9695,
      "num_input_tokens_seen": 1622816,
      "step": 2795
    },
    {
      "epoch": 0.4170390229371463,
      "grad_norm": 12.375,
      "learning_rate": 1.0422252010723861e-05,
      "loss": 1.1875,
      "num_input_tokens_seen": 1625664,
      "step": 2800
    },
    {
      "epoch": 0.41778373547810543,
      "grad_norm": 10.6875,
      "learning_rate": 1.0440869824247841e-05,
      "loss": 1.1626,
      "num_input_tokens_seen": 1628384,
      "step": 2805
    },
    {
      "epoch": 0.41852844801906464,
      "grad_norm": 10.6875,
      "learning_rate": 1.0459487637771821e-05,
      "loss": 1.0823,
      "num_input_tokens_seen": 1631616,
      "step": 2810
    },
    {
      "epoch": 0.41927316056002384,
      "grad_norm": 9.1875,
      "learning_rate": 1.04781054512958e-05,
      "loss": 1.1197,
      "num_input_tokens_seen": 1634304,
      "step": 2815
    },
    {
      "epoch": 0.42001787310098304,
      "grad_norm": 17.125,
      "learning_rate": 1.049672326481978e-05,
      "loss": 1.2683,
      "num_input_tokens_seen": 1637312,
      "step": 2820
    },
    {
      "epoch": 0.4207625856419422,
      "grad_norm": 19.0,
      "learning_rate": 1.051534107834376e-05,
      "loss": 1.3119,
      "num_input_tokens_seen": 1640192,
      "step": 2825
    },
    {
      "epoch": 0.4215072981829014,
      "grad_norm": 10.375,
      "learning_rate": 1.053395889186774e-05,
      "loss": 1.1338,
      "num_input_tokens_seen": 1643008,
      "step": 2830
    },
    {
      "epoch": 0.4222520107238606,
      "grad_norm": 8.5,
      "learning_rate": 1.0552576705391719e-05,
      "loss": 1.3639,
      "num_input_tokens_seen": 1645696,
      "step": 2835
    },
    {
      "epoch": 0.4229967232648198,
      "grad_norm": 9.25,
      "learning_rate": 1.0571194518915699e-05,
      "loss": 1.235,
      "num_input_tokens_seen": 1648448,
      "step": 2840
    },
    {
      "epoch": 0.42374143580577894,
      "grad_norm": 10.8125,
      "learning_rate": 1.0589812332439679e-05,
      "loss": 1.4424,
      "num_input_tokens_seen": 1651392,
      "step": 2845
    },
    {
      "epoch": 0.42448614834673815,
      "grad_norm": 21.75,
      "learning_rate": 1.0608430145963659e-05,
      "loss": 1.3133,
      "num_input_tokens_seen": 1654240,
      "step": 2850
    },
    {
      "epoch": 0.42523086088769735,
      "grad_norm": 14.4375,
      "learning_rate": 1.0627047959487637e-05,
      "loss": 1.0425,
      "num_input_tokens_seen": 1657088,
      "step": 2855
    },
    {
      "epoch": 0.42597557342865655,
      "grad_norm": 9.3125,
      "learning_rate": 1.0645665773011617e-05,
      "loss": 1.4214,
      "num_input_tokens_seen": 1659744,
      "step": 2860
    },
    {
      "epoch": 0.42672028596961575,
      "grad_norm": 15.25,
      "learning_rate": 1.0664283586535598e-05,
      "loss": 1.0266,
      "num_input_tokens_seen": 1662656,
      "step": 2865
    },
    {
      "epoch": 0.4274649985105749,
      "grad_norm": 13.375,
      "learning_rate": 1.0682901400059578e-05,
      "loss": 0.9659,
      "num_input_tokens_seen": 1665408,
      "step": 2870
    },
    {
      "epoch": 0.4282097110515341,
      "grad_norm": 11.5,
      "learning_rate": 1.0701519213583558e-05,
      "loss": 0.9026,
      "num_input_tokens_seen": 1668160,
      "step": 2875
    },
    {
      "epoch": 0.4289544235924933,
      "grad_norm": 5.4375,
      "learning_rate": 1.0720137027107537e-05,
      "loss": 1.18,
      "num_input_tokens_seen": 1671232,
      "step": 2880
    },
    {
      "epoch": 0.4296991361334525,
      "grad_norm": 21.625,
      "learning_rate": 1.0738754840631517e-05,
      "loss": 1.0969,
      "num_input_tokens_seen": 1674336,
      "step": 2885
    },
    {
      "epoch": 0.43044384867441166,
      "grad_norm": 8.25,
      "learning_rate": 1.0757372654155496e-05,
      "loss": 1.2101,
      "num_input_tokens_seen": 1677184,
      "step": 2890
    },
    {
      "epoch": 0.43118856121537086,
      "grad_norm": 17.75,
      "learning_rate": 1.0775990467679476e-05,
      "loss": 1.3418,
      "num_input_tokens_seen": 1679904,
      "step": 2895
    },
    {
      "epoch": 0.43193327375633006,
      "grad_norm": 9.5,
      "learning_rate": 1.0794608281203456e-05,
      "loss": 0.9383,
      "num_input_tokens_seen": 1682656,
      "step": 2900
    },
    {
      "epoch": 0.43267798629728926,
      "grad_norm": 11.1875,
      "learning_rate": 1.0813226094727436e-05,
      "loss": 1.1994,
      "num_input_tokens_seen": 1685440,
      "step": 2905
    },
    {
      "epoch": 0.4334226988382484,
      "grad_norm": 4.125,
      "learning_rate": 1.0831843908251416e-05,
      "loss": 1.1167,
      "num_input_tokens_seen": 1688288,
      "step": 2910
    },
    {
      "epoch": 0.4341674113792076,
      "grad_norm": 9.75,
      "learning_rate": 1.0850461721775396e-05,
      "loss": 0.9297,
      "num_input_tokens_seen": 1691264,
      "step": 2915
    },
    {
      "epoch": 0.4349121239201668,
      "grad_norm": 10.5,
      "learning_rate": 1.0869079535299374e-05,
      "loss": 1.1942,
      "num_input_tokens_seen": 1694240,
      "step": 2920
    },
    {
      "epoch": 0.435656836461126,
      "grad_norm": 20.875,
      "learning_rate": 1.0887697348823354e-05,
      "loss": 1.3393,
      "num_input_tokens_seen": 1697152,
      "step": 2925
    },
    {
      "epoch": 0.4364015490020852,
      "grad_norm": 12.125,
      "learning_rate": 1.0906315162347334e-05,
      "loss": 1.154,
      "num_input_tokens_seen": 1699936,
      "step": 2930
    },
    {
      "epoch": 0.43714626154304437,
      "grad_norm": 12.9375,
      "learning_rate": 1.0924932975871314e-05,
      "loss": 1.0362,
      "num_input_tokens_seen": 1702880,
      "step": 2935
    },
    {
      "epoch": 0.43789097408400357,
      "grad_norm": 15.25,
      "learning_rate": 1.0943550789395294e-05,
      "loss": 0.92,
      "num_input_tokens_seen": 1705664,
      "step": 2940
    },
    {
      "epoch": 0.4386356866249628,
      "grad_norm": 18.875,
      "learning_rate": 1.0962168602919274e-05,
      "loss": 0.9735,
      "num_input_tokens_seen": 1708416,
      "step": 2945
    },
    {
      "epoch": 0.439380399165922,
      "grad_norm": 7.75,
      "learning_rate": 1.0980786416443254e-05,
      "loss": 0.9251,
      "num_input_tokens_seen": 1711424,
      "step": 2950
    },
    {
      "epoch": 0.4401251117068811,
      "grad_norm": 12.125,
      "learning_rate": 1.0999404229967234e-05,
      "loss": 1.7399,
      "num_input_tokens_seen": 1714208,
      "step": 2955
    },
    {
      "epoch": 0.4408698242478403,
      "grad_norm": 11.75,
      "learning_rate": 1.1018022043491214e-05,
      "loss": 1.5214,
      "num_input_tokens_seen": 1716992,
      "step": 2960
    },
    {
      "epoch": 0.4416145367887995,
      "grad_norm": 20.25,
      "learning_rate": 1.1036639857015192e-05,
      "loss": 1.4687,
      "num_input_tokens_seen": 1719840,
      "step": 2965
    },
    {
      "epoch": 0.44235924932975873,
      "grad_norm": 10.1875,
      "learning_rate": 1.1055257670539172e-05,
      "loss": 1.3326,
      "num_input_tokens_seen": 1722464,
      "step": 2970
    },
    {
      "epoch": 0.4431039618707179,
      "grad_norm": 9.375,
      "learning_rate": 1.1073875484063152e-05,
      "loss": 1.0533,
      "num_input_tokens_seen": 1725216,
      "step": 2975
    },
    {
      "epoch": 0.4438486744116771,
      "grad_norm": 7.09375,
      "learning_rate": 1.1092493297587132e-05,
      "loss": 1.0537,
      "num_input_tokens_seen": 1728000,
      "step": 2980
    },
    {
      "epoch": 0.4445933869526363,
      "grad_norm": 10.5625,
      "learning_rate": 1.1111111111111112e-05,
      "loss": 1.4953,
      "num_input_tokens_seen": 1730912,
      "step": 2985
    },
    {
      "epoch": 0.4453380994935955,
      "grad_norm": 14.3125,
      "learning_rate": 1.1129728924635092e-05,
      "loss": 1.181,
      "num_input_tokens_seen": 1733600,
      "step": 2990
    },
    {
      "epoch": 0.4460828120345547,
      "grad_norm": 15.0625,
      "learning_rate": 1.1148346738159071e-05,
      "loss": 0.75,
      "num_input_tokens_seen": 1736480,
      "step": 2995
    },
    {
      "epoch": 0.44682752457551383,
      "grad_norm": 14.25,
      "learning_rate": 1.1166964551683051e-05,
      "loss": 1.0888,
      "num_input_tokens_seen": 1739264,
      "step": 3000
    },
    {
      "epoch": 0.44757223711647304,
      "grad_norm": 13.875,
      "learning_rate": 1.1185582365207031e-05,
      "loss": 1.1269,
      "num_input_tokens_seen": 1742336,
      "step": 3005
    },
    {
      "epoch": 0.44831694965743224,
      "grad_norm": 9.5,
      "learning_rate": 1.120420017873101e-05,
      "loss": 1.0109,
      "num_input_tokens_seen": 1745216,
      "step": 3010
    },
    {
      "epoch": 0.44906166219839144,
      "grad_norm": 14.6875,
      "learning_rate": 1.122281799225499e-05,
      "loss": 1.3103,
      "num_input_tokens_seen": 1748192,
      "step": 3015
    },
    {
      "epoch": 0.4498063747393506,
      "grad_norm": 13.6875,
      "learning_rate": 1.124143580577897e-05,
      "loss": 1.3702,
      "num_input_tokens_seen": 1750944,
      "step": 3020
    },
    {
      "epoch": 0.4505510872803098,
      "grad_norm": 21.375,
      "learning_rate": 1.126005361930295e-05,
      "loss": 0.993,
      "num_input_tokens_seen": 1753856,
      "step": 3025
    },
    {
      "epoch": 0.451295799821269,
      "grad_norm": 13.875,
      "learning_rate": 1.127867143282693e-05,
      "loss": 0.8851,
      "num_input_tokens_seen": 1756896,
      "step": 3030
    },
    {
      "epoch": 0.4520405123622282,
      "grad_norm": 13.0,
      "learning_rate": 1.129728924635091e-05,
      "loss": 1.2222,
      "num_input_tokens_seen": 1760032,
      "step": 3035
    },
    {
      "epoch": 0.45278522490318734,
      "grad_norm": 17.0,
      "learning_rate": 1.1315907059874889e-05,
      "loss": 1.1823,
      "num_input_tokens_seen": 1762848,
      "step": 3040
    },
    {
      "epoch": 0.45352993744414655,
      "grad_norm": 11.1875,
      "learning_rate": 1.1334524873398869e-05,
      "loss": 1.0465,
      "num_input_tokens_seen": 1765760,
      "step": 3045
    },
    {
      "epoch": 0.45427464998510575,
      "grad_norm": 10.25,
      "learning_rate": 1.1353142686922847e-05,
      "loss": 1.272,
      "num_input_tokens_seen": 1768608,
      "step": 3050
    },
    {
      "epoch": 0.45501936252606495,
      "grad_norm": 16.625,
      "learning_rate": 1.1371760500446827e-05,
      "loss": 1.0945,
      "num_input_tokens_seen": 1771808,
      "step": 3055
    },
    {
      "epoch": 0.45576407506702415,
      "grad_norm": 17.5,
      "learning_rate": 1.1390378313970807e-05,
      "loss": 1.4014,
      "num_input_tokens_seen": 1774656,
      "step": 3060
    },
    {
      "epoch": 0.4565087876079833,
      "grad_norm": 13.375,
      "learning_rate": 1.1408996127494787e-05,
      "loss": 1.1736,
      "num_input_tokens_seen": 1777568,
      "step": 3065
    },
    {
      "epoch": 0.4572535001489425,
      "grad_norm": 13.625,
      "learning_rate": 1.1427613941018769e-05,
      "loss": 1.2081,
      "num_input_tokens_seen": 1780672,
      "step": 3070
    },
    {
      "epoch": 0.4579982126899017,
      "grad_norm": 11.8125,
      "learning_rate": 1.1446231754542747e-05,
      "loss": 0.7847,
      "num_input_tokens_seen": 1783424,
      "step": 3075
    },
    {
      "epoch": 0.4587429252308609,
      "grad_norm": 17.25,
      "learning_rate": 1.1464849568066727e-05,
      "loss": 1.257,
      "num_input_tokens_seen": 1786176,
      "step": 3080
    },
    {
      "epoch": 0.45948763777182006,
      "grad_norm": 21.25,
      "learning_rate": 1.1483467381590707e-05,
      "loss": 1.5973,
      "num_input_tokens_seen": 1788896,
      "step": 3085
    },
    {
      "epoch": 0.46023235031277926,
      "grad_norm": 10.1875,
      "learning_rate": 1.1502085195114687e-05,
      "loss": 1.2456,
      "num_input_tokens_seen": 1791840,
      "step": 3090
    },
    {
      "epoch": 0.46097706285373846,
      "grad_norm": 12.25,
      "learning_rate": 1.1520703008638665e-05,
      "loss": 0.855,
      "num_input_tokens_seen": 1794752,
      "step": 3095
    },
    {
      "epoch": 0.46172177539469766,
      "grad_norm": 20.125,
      "learning_rate": 1.1539320822162645e-05,
      "loss": 1.1262,
      "num_input_tokens_seen": 1797568,
      "step": 3100
    },
    {
      "epoch": 0.4624664879356568,
      "grad_norm": 15.3125,
      "learning_rate": 1.1557938635686626e-05,
      "loss": 1.3594,
      "num_input_tokens_seen": 1800768,
      "step": 3105
    },
    {
      "epoch": 0.463211200476616,
      "grad_norm": 12.6875,
      "learning_rate": 1.1576556449210606e-05,
      "loss": 0.989,
      "num_input_tokens_seen": 1803744,
      "step": 3110
    },
    {
      "epoch": 0.4639559130175752,
      "grad_norm": 18.625,
      "learning_rate": 1.1595174262734585e-05,
      "loss": 1.0659,
      "num_input_tokens_seen": 1806624,
      "step": 3115
    },
    {
      "epoch": 0.4647006255585344,
      "grad_norm": 18.625,
      "learning_rate": 1.1613792076258565e-05,
      "loss": 1.2459,
      "num_input_tokens_seen": 1809600,
      "step": 3120
    },
    {
      "epoch": 0.4654453380994936,
      "grad_norm": 13.375,
      "learning_rate": 1.1632409889782545e-05,
      "loss": 1.1389,
      "num_input_tokens_seen": 1812480,
      "step": 3125
    },
    {
      "epoch": 0.46619005064045277,
      "grad_norm": 15.5,
      "learning_rate": 1.1651027703306524e-05,
      "loss": 0.8374,
      "num_input_tokens_seen": 1815328,
      "step": 3130
    },
    {
      "epoch": 0.46693476318141197,
      "grad_norm": 10.125,
      "learning_rate": 1.1669645516830504e-05,
      "loss": 1.1532,
      "num_input_tokens_seen": 1818368,
      "step": 3135
    },
    {
      "epoch": 0.4676794757223712,
      "grad_norm": 11.0625,
      "learning_rate": 1.1688263330354483e-05,
      "loss": 1.1341,
      "num_input_tokens_seen": 1821312,
      "step": 3140
    },
    {
      "epoch": 0.4684241882633304,
      "grad_norm": 11.375,
      "learning_rate": 1.1706881143878464e-05,
      "loss": 1.2266,
      "num_input_tokens_seen": 1824640,
      "step": 3145
    },
    {
      "epoch": 0.4691689008042895,
      "grad_norm": 6.625,
      "learning_rate": 1.1725498957402444e-05,
      "loss": 1.2865,
      "num_input_tokens_seen": 1827392,
      "step": 3150
    },
    {
      "epoch": 0.4699136133452487,
      "grad_norm": 12.5625,
      "learning_rate": 1.1744116770926424e-05,
      "loss": 1.3586,
      "num_input_tokens_seen": 1830080,
      "step": 3155
    },
    {
      "epoch": 0.47065832588620793,
      "grad_norm": 15.0,
      "learning_rate": 1.1762734584450402e-05,
      "loss": 1.2733,
      "num_input_tokens_seen": 1833024,
      "step": 3160
    },
    {
      "epoch": 0.47140303842716713,
      "grad_norm": 16.875,
      "learning_rate": 1.1781352397974382e-05,
      "loss": 1.4041,
      "num_input_tokens_seen": 1835680,
      "step": 3165
    },
    {
      "epoch": 0.4721477509681263,
      "grad_norm": 12.3125,
      "learning_rate": 1.1799970211498362e-05,
      "loss": 1.0917,
      "num_input_tokens_seen": 1838464,
      "step": 3170
    },
    {
      "epoch": 0.4728924635090855,
      "grad_norm": 8.0625,
      "learning_rate": 1.1818588025022342e-05,
      "loss": 1.0583,
      "num_input_tokens_seen": 1841728,
      "step": 3175
    },
    {
      "epoch": 0.4736371760500447,
      "grad_norm": 11.875,
      "learning_rate": 1.183720583854632e-05,
      "loss": 1.1763,
      "num_input_tokens_seen": 1844576,
      "step": 3180
    },
    {
      "epoch": 0.4743818885910039,
      "grad_norm": 12.75,
      "learning_rate": 1.1855823652070302e-05,
      "loss": 1.0962,
      "num_input_tokens_seen": 1847328,
      "step": 3185
    },
    {
      "epoch": 0.4751266011319631,
      "grad_norm": 13.4375,
      "learning_rate": 1.1874441465594282e-05,
      "loss": 1.2437,
      "num_input_tokens_seen": 1850240,
      "step": 3190
    },
    {
      "epoch": 0.47587131367292224,
      "grad_norm": 4.125,
      "learning_rate": 1.1893059279118262e-05,
      "loss": 1.3076,
      "num_input_tokens_seen": 1853216,
      "step": 3195
    },
    {
      "epoch": 0.47661602621388144,
      "grad_norm": 5.71875,
      "learning_rate": 1.1911677092642242e-05,
      "loss": 0.868,
      "num_input_tokens_seen": 1856000,
      "step": 3200
    },
    {
      "epoch": 0.47736073875484064,
      "grad_norm": 12.3125,
      "learning_rate": 1.193029490616622e-05,
      "loss": 1.198,
      "num_input_tokens_seen": 1858912,
      "step": 3205
    },
    {
      "epoch": 0.47810545129579984,
      "grad_norm": 19.875,
      "learning_rate": 1.19489127196902e-05,
      "loss": 1.3195,
      "num_input_tokens_seen": 1862240,
      "step": 3210
    },
    {
      "epoch": 0.478850163836759,
      "grad_norm": 19.25,
      "learning_rate": 1.196753053321418e-05,
      "loss": 1.4116,
      "num_input_tokens_seen": 1864896,
      "step": 3215
    },
    {
      "epoch": 0.4795948763777182,
      "grad_norm": 6.71875,
      "learning_rate": 1.198614834673816e-05,
      "loss": 1.0648,
      "num_input_tokens_seen": 1867904,
      "step": 3220
    },
    {
      "epoch": 0.4803395889186774,
      "grad_norm": 12.6875,
      "learning_rate": 1.200476616026214e-05,
      "loss": 1.1334,
      "num_input_tokens_seen": 1870496,
      "step": 3225
    },
    {
      "epoch": 0.4810843014596366,
      "grad_norm": 12.5625,
      "learning_rate": 1.202338397378612e-05,
      "loss": 1.1,
      "num_input_tokens_seen": 1873152,
      "step": 3230
    },
    {
      "epoch": 0.48182901400059575,
      "grad_norm": 14.5,
      "learning_rate": 1.20420017873101e-05,
      "loss": 1.2429,
      "num_input_tokens_seen": 1876096,
      "step": 3235
    },
    {
      "epoch": 0.48257372654155495,
      "grad_norm": 9.5,
      "learning_rate": 1.206061960083408e-05,
      "loss": 1.0765,
      "num_input_tokens_seen": 1879168,
      "step": 3240
    },
    {
      "epoch": 0.48331843908251415,
      "grad_norm": 9.1875,
      "learning_rate": 1.2079237414358058e-05,
      "loss": 1.2444,
      "num_input_tokens_seen": 1882304,
      "step": 3245
    },
    {
      "epoch": 0.48406315162347335,
      "grad_norm": 14.5625,
      "learning_rate": 1.2097855227882038e-05,
      "loss": 1.0105,
      "num_input_tokens_seen": 1885120,
      "step": 3250
    },
    {
      "epoch": 0.48480786416443256,
      "grad_norm": 5.4375,
      "learning_rate": 1.2116473041406018e-05,
      "loss": 0.8762,
      "num_input_tokens_seen": 1887968,
      "step": 3255
    },
    {
      "epoch": 0.4855525767053917,
      "grad_norm": 14.1875,
      "learning_rate": 1.2135090854929997e-05,
      "loss": 1.006,
      "num_input_tokens_seen": 1890688,
      "step": 3260
    },
    {
      "epoch": 0.4862972892463509,
      "grad_norm": 9.875,
      "learning_rate": 1.2153708668453977e-05,
      "loss": 0.8895,
      "num_input_tokens_seen": 1893408,
      "step": 3265
    },
    {
      "epoch": 0.4870420017873101,
      "grad_norm": 9.9375,
      "learning_rate": 1.2172326481977957e-05,
      "loss": 0.8851,
      "num_input_tokens_seen": 1896288,
      "step": 3270
    },
    {
      "epoch": 0.4877867143282693,
      "grad_norm": 10.5,
      "learning_rate": 1.2190944295501937e-05,
      "loss": 1.3372,
      "num_input_tokens_seen": 1899296,
      "step": 3275
    },
    {
      "epoch": 0.48853142686922846,
      "grad_norm": 13.125,
      "learning_rate": 1.2209562109025917e-05,
      "loss": 1.1979,
      "num_input_tokens_seen": 1902176,
      "step": 3280
    },
    {
      "epoch": 0.48927613941018766,
      "grad_norm": 17.375,
      "learning_rate": 1.2228179922549897e-05,
      "loss": 0.9872,
      "num_input_tokens_seen": 1904992,
      "step": 3285
    },
    {
      "epoch": 0.49002085195114686,
      "grad_norm": 25.5,
      "learning_rate": 1.2246797736073875e-05,
      "loss": 1.5835,
      "num_input_tokens_seen": 1907936,
      "step": 3290
    },
    {
      "epoch": 0.49076556449210607,
      "grad_norm": 19.125,
      "learning_rate": 1.2265415549597855e-05,
      "loss": 1.0763,
      "num_input_tokens_seen": 1911072,
      "step": 3295
    },
    {
      "epoch": 0.4915102770330652,
      "grad_norm": 6.75,
      "learning_rate": 1.2284033363121835e-05,
      "loss": 1.3223,
      "num_input_tokens_seen": 1913792,
      "step": 3300
    },
    {
      "epoch": 0.4922549895740244,
      "grad_norm": 12.5625,
      "learning_rate": 1.2302651176645815e-05,
      "loss": 1.2279,
      "num_input_tokens_seen": 1916864,
      "step": 3305
    },
    {
      "epoch": 0.4929997021149836,
      "grad_norm": 8.1875,
      "learning_rate": 1.2321268990169795e-05,
      "loss": 0.9272,
      "num_input_tokens_seen": 1920128,
      "step": 3310
    },
    {
      "epoch": 0.4937444146559428,
      "grad_norm": 17.75,
      "learning_rate": 1.2339886803693775e-05,
      "loss": 1.0581,
      "num_input_tokens_seen": 1922976,
      "step": 3315
    },
    {
      "epoch": 0.494489127196902,
      "grad_norm": 13.25,
      "learning_rate": 1.2358504617217755e-05,
      "loss": 0.9866,
      "num_input_tokens_seen": 1925952,
      "step": 3320
    },
    {
      "epoch": 0.49523383973786117,
      "grad_norm": 12.4375,
      "learning_rate": 1.2377122430741735e-05,
      "loss": 1.0547,
      "num_input_tokens_seen": 1928928,
      "step": 3325
    },
    {
      "epoch": 0.4959785522788204,
      "grad_norm": 17.25,
      "learning_rate": 1.2395740244265713e-05,
      "loss": 1.1007,
      "num_input_tokens_seen": 1931808,
      "step": 3330
    },
    {
      "epoch": 0.4967232648197796,
      "grad_norm": 8.4375,
      "learning_rate": 1.2414358057789693e-05,
      "loss": 1.2388,
      "num_input_tokens_seen": 1934528,
      "step": 3335
    },
    {
      "epoch": 0.4974679773607388,
      "grad_norm": 19.25,
      "learning_rate": 1.2432975871313673e-05,
      "loss": 1.2548,
      "num_input_tokens_seen": 1937568,
      "step": 3340
    },
    {
      "epoch": 0.4982126899016979,
      "grad_norm": 8.4375,
      "learning_rate": 1.2451593684837653e-05,
      "loss": 0.9238,
      "num_input_tokens_seen": 1940832,
      "step": 3345
    },
    {
      "epoch": 0.4989574024426571,
      "grad_norm": 16.25,
      "learning_rate": 1.2470211498361634e-05,
      "loss": 1.243,
      "num_input_tokens_seen": 1943520,
      "step": 3350
    },
    {
      "epoch": 0.49970211498361633,
      "grad_norm": 13.625,
      "learning_rate": 1.2488829311885613e-05,
      "loss": 1.2174,
      "num_input_tokens_seen": 1946240,
      "step": 3355
    },
    {
      "epoch": 0.5004468275245755,
      "grad_norm": 14.625,
      "learning_rate": 1.2507447125409594e-05,
      "loss": 0.8694,
      "num_input_tokens_seen": 1949248,
      "step": 3360
    },
    {
      "epoch": 0.5011915400655347,
      "grad_norm": 18.625,
      "learning_rate": 1.252606493893357e-05,
      "loss": 1.0929,
      "num_input_tokens_seen": 1952160,
      "step": 3365
    },
    {
      "epoch": 0.5019362526064939,
      "grad_norm": 16.5,
      "learning_rate": 1.254468275245755e-05,
      "loss": 1.1387,
      "num_input_tokens_seen": 1955040,
      "step": 3370
    },
    {
      "epoch": 0.5026809651474531,
      "grad_norm": 6.125,
      "learning_rate": 1.256330056598153e-05,
      "loss": 1.4159,
      "num_input_tokens_seen": 1957888,
      "step": 3375
    },
    {
      "epoch": 0.5034256776884123,
      "grad_norm": 8.1875,
      "learning_rate": 1.258191837950551e-05,
      "loss": 1.0235,
      "num_input_tokens_seen": 1960992,
      "step": 3380
    },
    {
      "epoch": 0.5041703902293715,
      "grad_norm": 16.0,
      "learning_rate": 1.260053619302949e-05,
      "loss": 0.8223,
      "num_input_tokens_seen": 1963968,
      "step": 3385
    },
    {
      "epoch": 0.5049151027703307,
      "grad_norm": 19.75,
      "learning_rate": 1.2619154006553472e-05,
      "loss": 1.5764,
      "num_input_tokens_seen": 1966848,
      "step": 3390
    },
    {
      "epoch": 0.5056598153112899,
      "grad_norm": 9.75,
      "learning_rate": 1.2637771820077452e-05,
      "loss": 0.9438,
      "num_input_tokens_seen": 1969536,
      "step": 3395
    },
    {
      "epoch": 0.506404527852249,
      "grad_norm": 9.375,
      "learning_rate": 1.2656389633601432e-05,
      "loss": 1.3489,
      "num_input_tokens_seen": 1972704,
      "step": 3400
    },
    {
      "epoch": 0.5071492403932082,
      "grad_norm": 6.59375,
      "learning_rate": 1.2675007447125412e-05,
      "loss": 1.0529,
      "num_input_tokens_seen": 1975552,
      "step": 3405
    },
    {
      "epoch": 0.5078939529341674,
      "grad_norm": 11.3125,
      "learning_rate": 1.2693625260649388e-05,
      "loss": 1.3158,
      "num_input_tokens_seen": 1978624,
      "step": 3410
    },
    {
      "epoch": 0.5086386654751266,
      "grad_norm": 12.75,
      "learning_rate": 1.2712243074173368e-05,
      "loss": 1.2727,
      "num_input_tokens_seen": 1981824,
      "step": 3415
    },
    {
      "epoch": 0.5093833780160858,
      "grad_norm": 11.375,
      "learning_rate": 1.2730860887697348e-05,
      "loss": 1.1667,
      "num_input_tokens_seen": 1984736,
      "step": 3420
    },
    {
      "epoch": 0.510128090557045,
      "grad_norm": 14.375,
      "learning_rate": 1.2749478701221328e-05,
      "loss": 1.2221,
      "num_input_tokens_seen": 1987552,
      "step": 3425
    },
    {
      "epoch": 0.5108728030980042,
      "grad_norm": 22.875,
      "learning_rate": 1.276809651474531e-05,
      "loss": 1.5333,
      "num_input_tokens_seen": 1990880,
      "step": 3430
    },
    {
      "epoch": 0.5116175156389634,
      "grad_norm": 10.875,
      "learning_rate": 1.278671432826929e-05,
      "loss": 0.9079,
      "num_input_tokens_seen": 1994048,
      "step": 3435
    },
    {
      "epoch": 0.5123622281799225,
      "grad_norm": 14.75,
      "learning_rate": 1.280533214179327e-05,
      "loss": 1.2128,
      "num_input_tokens_seen": 1997792,
      "step": 3440
    },
    {
      "epoch": 0.5131069407208817,
      "grad_norm": 7.03125,
      "learning_rate": 1.282394995531725e-05,
      "loss": 1.1852,
      "num_input_tokens_seen": 2000800,
      "step": 3445
    },
    {
      "epoch": 0.5138516532618409,
      "grad_norm": 18.5,
      "learning_rate": 1.2842567768841226e-05,
      "loss": 1.2785,
      "num_input_tokens_seen": 2003904,
      "step": 3450
    },
    {
      "epoch": 0.5145963658028001,
      "grad_norm": 7.03125,
      "learning_rate": 1.2861185582365206e-05,
      "loss": 1.2998,
      "num_input_tokens_seen": 2006944,
      "step": 3455
    },
    {
      "epoch": 0.5153410783437593,
      "grad_norm": 10.25,
      "learning_rate": 1.2879803395889186e-05,
      "loss": 1.1958,
      "num_input_tokens_seen": 2010496,
      "step": 3460
    },
    {
      "epoch": 0.5160857908847185,
      "grad_norm": 9.0625,
      "learning_rate": 1.2898421209413166e-05,
      "loss": 1.3593,
      "num_input_tokens_seen": 2013408,
      "step": 3465
    },
    {
      "epoch": 0.5168305034256777,
      "grad_norm": 24.75,
      "learning_rate": 1.2917039022937148e-05,
      "loss": 1.2313,
      "num_input_tokens_seen": 2016032,
      "step": 3470
    },
    {
      "epoch": 0.5175752159666369,
      "grad_norm": 12.125,
      "learning_rate": 1.2935656836461127e-05,
      "loss": 1.1295,
      "num_input_tokens_seen": 2018752,
      "step": 3475
    },
    {
      "epoch": 0.5183199285075961,
      "grad_norm": 18.5,
      "learning_rate": 1.2954274649985107e-05,
      "loss": 1.4243,
      "num_input_tokens_seen": 2021600,
      "step": 3480
    },
    {
      "epoch": 0.5190646410485552,
      "grad_norm": 13.6875,
      "learning_rate": 1.2972892463509087e-05,
      "loss": 0.9988,
      "num_input_tokens_seen": 2024544,
      "step": 3485
    },
    {
      "epoch": 0.5198093535895144,
      "grad_norm": 15.0,
      "learning_rate": 1.2991510277033067e-05,
      "loss": 1.8328,
      "num_input_tokens_seen": 2027552,
      "step": 3490
    },
    {
      "epoch": 0.5205540661304736,
      "grad_norm": 11.375,
      "learning_rate": 1.3010128090557044e-05,
      "loss": 0.726,
      "num_input_tokens_seen": 2030656,
      "step": 3495
    },
    {
      "epoch": 0.5212987786714328,
      "grad_norm": 8.1875,
      "learning_rate": 1.3028745904081024e-05,
      "loss": 0.9894,
      "num_input_tokens_seen": 2033600,
      "step": 3500
    },
    {
      "epoch": 0.522043491212392,
      "grad_norm": 10.4375,
      "learning_rate": 1.3047363717605005e-05,
      "loss": 1.1389,
      "num_input_tokens_seen": 2036544,
      "step": 3505
    },
    {
      "epoch": 0.5227882037533512,
      "grad_norm": 6.40625,
      "learning_rate": 1.3065981531128985e-05,
      "loss": 1.2424,
      "num_input_tokens_seen": 2039392,
      "step": 3510
    },
    {
      "epoch": 0.5235329162943104,
      "grad_norm": 10.375,
      "learning_rate": 1.3084599344652965e-05,
      "loss": 1.2039,
      "num_input_tokens_seen": 2042080,
      "step": 3515
    },
    {
      "epoch": 0.5242776288352696,
      "grad_norm": 10.6875,
      "learning_rate": 1.3103217158176945e-05,
      "loss": 1.2525,
      "num_input_tokens_seen": 2044992,
      "step": 3520
    },
    {
      "epoch": 0.5250223413762288,
      "grad_norm": 9.0,
      "learning_rate": 1.3121834971700925e-05,
      "loss": 1.3487,
      "num_input_tokens_seen": 2047776,
      "step": 3525
    },
    {
      "epoch": 0.5257670539171879,
      "grad_norm": 9.1875,
      "learning_rate": 1.3140452785224905e-05,
      "loss": 1.1043,
      "num_input_tokens_seen": 2050816,
      "step": 3530
    },
    {
      "epoch": 0.5265117664581471,
      "grad_norm": 15.5,
      "learning_rate": 1.3159070598748885e-05,
      "loss": 1.1917,
      "num_input_tokens_seen": 2053952,
      "step": 3535
    },
    {
      "epoch": 0.5272564789991063,
      "grad_norm": 5.0,
      "learning_rate": 1.3177688412272861e-05,
      "loss": 0.977,
      "num_input_tokens_seen": 2056896,
      "step": 3540
    },
    {
      "epoch": 0.5280011915400655,
      "grad_norm": 13.0,
      "learning_rate": 1.3196306225796843e-05,
      "loss": 1.1576,
      "num_input_tokens_seen": 2059776,
      "step": 3545
    },
    {
      "epoch": 0.5287459040810247,
      "grad_norm": 11.375,
      "learning_rate": 1.3214924039320823e-05,
      "loss": 0.9239,
      "num_input_tokens_seen": 2062880,
      "step": 3550
    },
    {
      "epoch": 0.5294906166219839,
      "grad_norm": 7.6875,
      "learning_rate": 1.3233541852844803e-05,
      "loss": 1.3035,
      "num_input_tokens_seen": 2065856,
      "step": 3555
    },
    {
      "epoch": 0.5302353291629431,
      "grad_norm": 18.375,
      "learning_rate": 1.3252159666368783e-05,
      "loss": 1.1442,
      "num_input_tokens_seen": 2068704,
      "step": 3560
    },
    {
      "epoch": 0.5309800417039023,
      "grad_norm": 5.78125,
      "learning_rate": 1.3270777479892763e-05,
      "loss": 0.9701,
      "num_input_tokens_seen": 2071488,
      "step": 3565
    },
    {
      "epoch": 0.5317247542448614,
      "grad_norm": 17.375,
      "learning_rate": 1.3289395293416743e-05,
      "loss": 1.1609,
      "num_input_tokens_seen": 2074432,
      "step": 3570
    },
    {
      "epoch": 0.5324694667858206,
      "grad_norm": 14.25,
      "learning_rate": 1.3308013106940723e-05,
      "loss": 1.0491,
      "num_input_tokens_seen": 2077120,
      "step": 3575
    },
    {
      "epoch": 0.5332141793267798,
      "grad_norm": 6.75,
      "learning_rate": 1.33266309204647e-05,
      "loss": 1.1728,
      "num_input_tokens_seen": 2079968,
      "step": 3580
    },
    {
      "epoch": 0.533958891867739,
      "grad_norm": 8.3125,
      "learning_rate": 1.334524873398868e-05,
      "loss": 1.1545,
      "num_input_tokens_seen": 2082784,
      "step": 3585
    },
    {
      "epoch": 0.5347036044086982,
      "grad_norm": 9.0,
      "learning_rate": 1.336386654751266e-05,
      "loss": 1.178,
      "num_input_tokens_seen": 2085920,
      "step": 3590
    },
    {
      "epoch": 0.5354483169496574,
      "grad_norm": 11.875,
      "learning_rate": 1.338248436103664e-05,
      "loss": 0.7975,
      "num_input_tokens_seen": 2088608,
      "step": 3595
    },
    {
      "epoch": 0.5361930294906166,
      "grad_norm": 9.9375,
      "learning_rate": 1.340110217456062e-05,
      "loss": 0.8525,
      "num_input_tokens_seen": 2091584,
      "step": 3600
    },
    {
      "epoch": 0.5369377420315758,
      "grad_norm": 11.25,
      "learning_rate": 1.34197199880846e-05,
      "loss": 0.9965,
      "num_input_tokens_seen": 2094272,
      "step": 3605
    },
    {
      "epoch": 0.537682454572535,
      "grad_norm": 9.5,
      "learning_rate": 1.343833780160858e-05,
      "loss": 1.0044,
      "num_input_tokens_seen": 2097344,
      "step": 3610
    },
    {
      "epoch": 0.5384271671134941,
      "grad_norm": 8.875,
      "learning_rate": 1.345695561513256e-05,
      "loss": 1.2268,
      "num_input_tokens_seen": 2100192,
      "step": 3615
    },
    {
      "epoch": 0.5391718796544533,
      "grad_norm": 21.375,
      "learning_rate": 1.347557342865654e-05,
      "loss": 1.1581,
      "num_input_tokens_seen": 2103136,
      "step": 3620
    },
    {
      "epoch": 0.5399165921954125,
      "grad_norm": 8.3125,
      "learning_rate": 1.3494191242180519e-05,
      "loss": 0.9863,
      "num_input_tokens_seen": 2106272,
      "step": 3625
    },
    {
      "epoch": 0.5406613047363718,
      "grad_norm": 12.375,
      "learning_rate": 1.3512809055704498e-05,
      "loss": 0.9228,
      "num_input_tokens_seen": 2109440,
      "step": 3630
    },
    {
      "epoch": 0.541406017277331,
      "grad_norm": 11.3125,
      "learning_rate": 1.3531426869228478e-05,
      "loss": 1.0942,
      "num_input_tokens_seen": 2112352,
      "step": 3635
    },
    {
      "epoch": 0.5421507298182902,
      "grad_norm": 11.3125,
      "learning_rate": 1.3550044682752458e-05,
      "loss": 0.9714,
      "num_input_tokens_seen": 2115200,
      "step": 3640
    },
    {
      "epoch": 0.5428954423592494,
      "grad_norm": 20.5,
      "learning_rate": 1.3568662496276438e-05,
      "loss": 1.2014,
      "num_input_tokens_seen": 2118112,
      "step": 3645
    },
    {
      "epoch": 0.5436401549002086,
      "grad_norm": 14.8125,
      "learning_rate": 1.3587280309800418e-05,
      "loss": 1.1973,
      "num_input_tokens_seen": 2121024,
      "step": 3650
    },
    {
      "epoch": 0.5443848674411678,
      "grad_norm": 8.0625,
      "learning_rate": 1.3605898123324398e-05,
      "loss": 1.0261,
      "num_input_tokens_seen": 2123680,
      "step": 3655
    },
    {
      "epoch": 0.5451295799821269,
      "grad_norm": 5.375,
      "learning_rate": 1.3624515936848378e-05,
      "loss": 1.2017,
      "num_input_tokens_seen": 2126784,
      "step": 3660
    },
    {
      "epoch": 0.5458742925230861,
      "grad_norm": 15.6875,
      "learning_rate": 1.3643133750372358e-05,
      "loss": 1.3608,
      "num_input_tokens_seen": 2129504,
      "step": 3665
    },
    {
      "epoch": 0.5466190050640453,
      "grad_norm": 13.9375,
      "learning_rate": 1.3661751563896336e-05,
      "loss": 1.3224,
      "num_input_tokens_seen": 2132416,
      "step": 3670
    },
    {
      "epoch": 0.5473637176050045,
      "grad_norm": 5.09375,
      "learning_rate": 1.3680369377420316e-05,
      "loss": 1.049,
      "num_input_tokens_seen": 2135264,
      "step": 3675
    },
    {
      "epoch": 0.5481084301459637,
      "grad_norm": 8.75,
      "learning_rate": 1.3698987190944296e-05,
      "loss": 1.4889,
      "num_input_tokens_seen": 2138496,
      "step": 3680
    },
    {
      "epoch": 0.5488531426869229,
      "grad_norm": 11.9375,
      "learning_rate": 1.3717605004468276e-05,
      "loss": 1.2615,
      "num_input_tokens_seen": 2141344,
      "step": 3685
    },
    {
      "epoch": 0.5495978552278821,
      "grad_norm": 6.4375,
      "learning_rate": 1.3736222817992256e-05,
      "loss": 1.3274,
      "num_input_tokens_seen": 2144256,
      "step": 3690
    },
    {
      "epoch": 0.5503425677688413,
      "grad_norm": 11.3125,
      "learning_rate": 1.3754840631516236e-05,
      "loss": 0.9958,
      "num_input_tokens_seen": 2147040,
      "step": 3695
    },
    {
      "epoch": 0.5510872803098004,
      "grad_norm": 12.0,
      "learning_rate": 1.3773458445040216e-05,
      "loss": 1.185,
      "num_input_tokens_seen": 2150144,
      "step": 3700
    },
    {
      "epoch": 0.5518319928507596,
      "grad_norm": 12.0625,
      "learning_rate": 1.3792076258564196e-05,
      "loss": 0.9504,
      "num_input_tokens_seen": 2152864,
      "step": 3705
    },
    {
      "epoch": 0.5525767053917188,
      "grad_norm": 5.78125,
      "learning_rate": 1.3810694072088174e-05,
      "loss": 1.2541,
      "num_input_tokens_seen": 2156032,
      "step": 3710
    },
    {
      "epoch": 0.553321417932678,
      "grad_norm": 8.3125,
      "learning_rate": 1.3829311885612154e-05,
      "loss": 0.9254,
      "num_input_tokens_seen": 2158848,
      "step": 3715
    },
    {
      "epoch": 0.5540661304736372,
      "grad_norm": 13.125,
      "learning_rate": 1.3847929699136134e-05,
      "loss": 1.3726,
      "num_input_tokens_seen": 2161664,
      "step": 3720
    },
    {
      "epoch": 0.5548108430145964,
      "grad_norm": 11.625,
      "learning_rate": 1.3866547512660114e-05,
      "loss": 1.1822,
      "num_input_tokens_seen": 2164544,
      "step": 3725
    },
    {
      "epoch": 0.5555555555555556,
      "grad_norm": 9.75,
      "learning_rate": 1.3885165326184094e-05,
      "loss": 1.2072,
      "num_input_tokens_seen": 2167168,
      "step": 3730
    },
    {
      "epoch": 0.5563002680965148,
      "grad_norm": 14.25,
      "learning_rate": 1.3903783139708073e-05,
      "loss": 1.298,
      "num_input_tokens_seen": 2169952,
      "step": 3735
    },
    {
      "epoch": 0.557044980637474,
      "grad_norm": 11.8125,
      "learning_rate": 1.3922400953232053e-05,
      "loss": 1.0231,
      "num_input_tokens_seen": 2173376,
      "step": 3740
    },
    {
      "epoch": 0.5577896931784331,
      "grad_norm": 16.875,
      "learning_rate": 1.3941018766756033e-05,
      "loss": 1.3828,
      "num_input_tokens_seen": 2176032,
      "step": 3745
    },
    {
      "epoch": 0.5585344057193923,
      "grad_norm": 13.5,
      "learning_rate": 1.3959636580280013e-05,
      "loss": 0.826,
      "num_input_tokens_seen": 2179136,
      "step": 3750
    },
    {
      "epoch": 0.5592791182603515,
      "grad_norm": 10.3125,
      "learning_rate": 1.3978254393803992e-05,
      "loss": 1.0916,
      "num_input_tokens_seen": 2181920,
      "step": 3755
    },
    {
      "epoch": 0.5600238308013107,
      "grad_norm": 11.4375,
      "learning_rate": 1.3996872207327971e-05,
      "loss": 1.0181,
      "num_input_tokens_seen": 2186080,
      "step": 3760
    },
    {
      "epoch": 0.5607685433422699,
      "grad_norm": 9.75,
      "learning_rate": 1.4015490020851951e-05,
      "loss": 1.1169,
      "num_input_tokens_seen": 2189120,
      "step": 3765
    },
    {
      "epoch": 0.5615132558832291,
      "grad_norm": 8.6875,
      "learning_rate": 1.4034107834375931e-05,
      "loss": 0.9967,
      "num_input_tokens_seen": 2191744,
      "step": 3770
    },
    {
      "epoch": 0.5622579684241883,
      "grad_norm": 9.875,
      "learning_rate": 1.4052725647899911e-05,
      "loss": 1.4337,
      "num_input_tokens_seen": 2194816,
      "step": 3775
    },
    {
      "epoch": 0.5630026809651475,
      "grad_norm": 19.75,
      "learning_rate": 1.4071343461423891e-05,
      "loss": 1.3415,
      "num_input_tokens_seen": 2197312,
      "step": 3780
    },
    {
      "epoch": 0.5637473935061067,
      "grad_norm": 11.0625,
      "learning_rate": 1.4089961274947871e-05,
      "loss": 1.0682,
      "num_input_tokens_seen": 2200448,
      "step": 3785
    },
    {
      "epoch": 0.5644921060470658,
      "grad_norm": 12.875,
      "learning_rate": 1.4108579088471851e-05,
      "loss": 1.3873,
      "num_input_tokens_seen": 2203392,
      "step": 3790
    },
    {
      "epoch": 0.565236818588025,
      "grad_norm": 6.84375,
      "learning_rate": 1.412719690199583e-05,
      "loss": 1.1177,
      "num_input_tokens_seen": 2206112,
      "step": 3795
    },
    {
      "epoch": 0.5659815311289842,
      "grad_norm": 17.5,
      "learning_rate": 1.4145814715519809e-05,
      "loss": 1.5312,
      "num_input_tokens_seen": 2209216,
      "step": 3800
    },
    {
      "epoch": 0.5667262436699434,
      "grad_norm": 19.25,
      "learning_rate": 1.4164432529043789e-05,
      "loss": 1.1877,
      "num_input_tokens_seen": 2211840,
      "step": 3805
    },
    {
      "epoch": 0.5674709562109026,
      "grad_norm": 10.5625,
      "learning_rate": 1.4183050342567769e-05,
      "loss": 0.9568,
      "num_input_tokens_seen": 2214848,
      "step": 3810
    },
    {
      "epoch": 0.5682156687518618,
      "grad_norm": 11.375,
      "learning_rate": 1.4201668156091749e-05,
      "loss": 1.0177,
      "num_input_tokens_seen": 2217696,
      "step": 3815
    },
    {
      "epoch": 0.568960381292821,
      "grad_norm": 11.5625,
      "learning_rate": 1.4220285969615729e-05,
      "loss": 1.127,
      "num_input_tokens_seen": 2220480,
      "step": 3820
    },
    {
      "epoch": 0.5697050938337802,
      "grad_norm": 14.4375,
      "learning_rate": 1.4238903783139709e-05,
      "loss": 1.2553,
      "num_input_tokens_seen": 2223200,
      "step": 3825
    },
    {
      "epoch": 0.5704498063747393,
      "grad_norm": 8.9375,
      "learning_rate": 1.4257521596663689e-05,
      "loss": 1.702,
      "num_input_tokens_seen": 2226048,
      "step": 3830
    },
    {
      "epoch": 0.5711945189156985,
      "grad_norm": 11.5,
      "learning_rate": 1.4276139410187669e-05,
      "loss": 1.2666,
      "num_input_tokens_seen": 2229024,
      "step": 3835
    },
    {
      "epoch": 0.5719392314566577,
      "grad_norm": 10.0,
      "learning_rate": 1.4294757223711647e-05,
      "loss": 1.636,
      "num_input_tokens_seen": 2232032,
      "step": 3840
    },
    {
      "epoch": 0.5726839439976169,
      "grad_norm": 8.3125,
      "learning_rate": 1.4313375037235627e-05,
      "loss": 1.2575,
      "num_input_tokens_seen": 2234752,
      "step": 3845
    },
    {
      "epoch": 0.5734286565385761,
      "grad_norm": 9.75,
      "learning_rate": 1.4331992850759607e-05,
      "loss": 0.9922,
      "num_input_tokens_seen": 2237664,
      "step": 3850
    },
    {
      "epoch": 0.5741733690795353,
      "grad_norm": 13.3125,
      "learning_rate": 1.4350610664283587e-05,
      "loss": 1.3979,
      "num_input_tokens_seen": 2240480,
      "step": 3855
    },
    {
      "epoch": 0.5749180816204945,
      "grad_norm": 18.625,
      "learning_rate": 1.4369228477807567e-05,
      "loss": 0.9095,
      "num_input_tokens_seen": 2243264,
      "step": 3860
    },
    {
      "epoch": 0.5756627941614537,
      "grad_norm": 10.8125,
      "learning_rate": 1.4387846291331546e-05,
      "loss": 1.1011,
      "num_input_tokens_seen": 2246304,
      "step": 3865
    },
    {
      "epoch": 0.5764075067024129,
      "grad_norm": 15.4375,
      "learning_rate": 1.4406464104855526e-05,
      "loss": 0.9723,
      "num_input_tokens_seen": 2249216,
      "step": 3870
    },
    {
      "epoch": 0.577152219243372,
      "grad_norm": 21.125,
      "learning_rate": 1.4425081918379506e-05,
      "loss": 1.3616,
      "num_input_tokens_seen": 2251904,
      "step": 3875
    },
    {
      "epoch": 0.5778969317843312,
      "grad_norm": 10.9375,
      "learning_rate": 1.4443699731903488e-05,
      "loss": 1.3944,
      "num_input_tokens_seen": 2254880,
      "step": 3880
    },
    {
      "epoch": 0.5786416443252904,
      "grad_norm": 8.5,
      "learning_rate": 1.4462317545427465e-05,
      "loss": 1.1738,
      "num_input_tokens_seen": 2257888,
      "step": 3885
    },
    {
      "epoch": 0.5793863568662496,
      "grad_norm": 11.5,
      "learning_rate": 1.4480935358951444e-05,
      "loss": 1.1388,
      "num_input_tokens_seen": 2260736,
      "step": 3890
    },
    {
      "epoch": 0.5801310694072088,
      "grad_norm": 12.75,
      "learning_rate": 1.4499553172475424e-05,
      "loss": 1.2156,
      "num_input_tokens_seen": 2263968,
      "step": 3895
    },
    {
      "epoch": 0.580875781948168,
      "grad_norm": 4.9375,
      "learning_rate": 1.4518170985999404e-05,
      "loss": 0.9131,
      "num_input_tokens_seen": 2266560,
      "step": 3900
    },
    {
      "epoch": 0.5816204944891272,
      "grad_norm": 18.625,
      "learning_rate": 1.4536788799523384e-05,
      "loss": 1.2891,
      "num_input_tokens_seen": 2269376,
      "step": 3905
    },
    {
      "epoch": 0.5823652070300864,
      "grad_norm": 12.4375,
      "learning_rate": 1.4555406613047364e-05,
      "loss": 1.3656,
      "num_input_tokens_seen": 2272320,
      "step": 3910
    },
    {
      "epoch": 0.5831099195710456,
      "grad_norm": 11.8125,
      "learning_rate": 1.4574024426571346e-05,
      "loss": 1.5457,
      "num_input_tokens_seen": 2275328,
      "step": 3915
    },
    {
      "epoch": 0.5838546321120047,
      "grad_norm": 14.25,
      "learning_rate": 1.4592642240095326e-05,
      "loss": 1.2066,
      "num_input_tokens_seen": 2278240,
      "step": 3920
    },
    {
      "epoch": 0.5845993446529639,
      "grad_norm": 15.4375,
      "learning_rate": 1.4611260053619302e-05,
      "loss": 1.1734,
      "num_input_tokens_seen": 2281504,
      "step": 3925
    },
    {
      "epoch": 0.5853440571939231,
      "grad_norm": 17.5,
      "learning_rate": 1.4629877867143282e-05,
      "loss": 1.4122,
      "num_input_tokens_seen": 2284320,
      "step": 3930
    },
    {
      "epoch": 0.5860887697348823,
      "grad_norm": 9.9375,
      "learning_rate": 1.4648495680667262e-05,
      "loss": 1.1354,
      "num_input_tokens_seen": 2287232,
      "step": 3935
    },
    {
      "epoch": 0.5868334822758415,
      "grad_norm": 19.0,
      "learning_rate": 1.4667113494191242e-05,
      "loss": 1.4029,
      "num_input_tokens_seen": 2290176,
      "step": 3940
    },
    {
      "epoch": 0.5875781948168007,
      "grad_norm": 15.75,
      "learning_rate": 1.4685731307715222e-05,
      "loss": 1.2401,
      "num_input_tokens_seen": 2292832,
      "step": 3945
    },
    {
      "epoch": 0.5883229073577599,
      "grad_norm": 11.8125,
      "learning_rate": 1.4704349121239202e-05,
      "loss": 1.3188,
      "num_input_tokens_seen": 2295712,
      "step": 3950
    },
    {
      "epoch": 0.5890676198987191,
      "grad_norm": 19.125,
      "learning_rate": 1.4722966934763183e-05,
      "loss": 1.3884,
      "num_input_tokens_seen": 2298432,
      "step": 3955
    },
    {
      "epoch": 0.5898123324396782,
      "grad_norm": 15.375,
      "learning_rate": 1.4741584748287163e-05,
      "loss": 1.1405,
      "num_input_tokens_seen": 2301408,
      "step": 3960
    },
    {
      "epoch": 0.5905570449806374,
      "grad_norm": 8.625,
      "learning_rate": 1.4760202561811143e-05,
      "loss": 1.0642,
      "num_input_tokens_seen": 2304512,
      "step": 3965
    },
    {
      "epoch": 0.5913017575215966,
      "grad_norm": 24.875,
      "learning_rate": 1.477882037533512e-05,
      "loss": 1.1839,
      "num_input_tokens_seen": 2307584,
      "step": 3970
    },
    {
      "epoch": 0.5920464700625558,
      "grad_norm": 5.875,
      "learning_rate": 1.47974381888591e-05,
      "loss": 1.1373,
      "num_input_tokens_seen": 2310336,
      "step": 3975
    },
    {
      "epoch": 0.592791182603515,
      "grad_norm": 9.8125,
      "learning_rate": 1.481605600238308e-05,
      "loss": 0.7803,
      "num_input_tokens_seen": 2313600,
      "step": 3980
    },
    {
      "epoch": 0.5935358951444742,
      "grad_norm": 15.3125,
      "learning_rate": 1.483467381590706e-05,
      "loss": 1.116,
      "num_input_tokens_seen": 2316448,
      "step": 3985
    },
    {
      "epoch": 0.5942806076854334,
      "grad_norm": 19.875,
      "learning_rate": 1.485329162943104e-05,
      "loss": 1.7349,
      "num_input_tokens_seen": 2319648,
      "step": 3990
    },
    {
      "epoch": 0.5950253202263927,
      "grad_norm": 9.5625,
      "learning_rate": 1.4871909442955021e-05,
      "loss": 1.353,
      "num_input_tokens_seen": 2322496,
      "step": 3995
    },
    {
      "epoch": 0.5957700327673519,
      "grad_norm": 14.75,
      "learning_rate": 1.4890527256479001e-05,
      "loss": 1.0562,
      "num_input_tokens_seen": 2325568,
      "step": 4000
    },
    {
      "epoch": 0.596514745308311,
      "grad_norm": 16.5,
      "learning_rate": 1.4909145070002981e-05,
      "loss": 1.1808,
      "num_input_tokens_seen": 2328512,
      "step": 4005
    },
    {
      "epoch": 0.5972594578492701,
      "grad_norm": 10.5,
      "learning_rate": 1.4927762883526961e-05,
      "loss": 1.2714,
      "num_input_tokens_seen": 2331584,
      "step": 4010
    },
    {
      "epoch": 0.5980041703902294,
      "grad_norm": 18.625,
      "learning_rate": 1.4946380697050938e-05,
      "loss": 1.2666,
      "num_input_tokens_seen": 2334528,
      "step": 4015
    },
    {
      "epoch": 0.5987488829311886,
      "grad_norm": 14.0625,
      "learning_rate": 1.4964998510574917e-05,
      "loss": 1.7099,
      "num_input_tokens_seen": 2337600,
      "step": 4020
    },
    {
      "epoch": 0.5994935954721478,
      "grad_norm": 12.8125,
      "learning_rate": 1.4983616324098897e-05,
      "loss": 1.2106,
      "num_input_tokens_seen": 2340480,
      "step": 4025
    },
    {
      "epoch": 0.600238308013107,
      "grad_norm": 12.6875,
      "learning_rate": 1.5002234137622877e-05,
      "loss": 1.3037,
      "num_input_tokens_seen": 2343232,
      "step": 4030
    },
    {
      "epoch": 0.6009830205540662,
      "grad_norm": 16.625,
      "learning_rate": 1.5020851951146859e-05,
      "loss": 1.0223,
      "num_input_tokens_seen": 2345920,
      "step": 4035
    },
    {
      "epoch": 0.6017277330950254,
      "grad_norm": 11.8125,
      "learning_rate": 1.5039469764670839e-05,
      "loss": 1.2161,
      "num_input_tokens_seen": 2349248,
      "step": 4040
    },
    {
      "epoch": 0.6024724456359845,
      "grad_norm": 24.375,
      "learning_rate": 1.5058087578194819e-05,
      "loss": 1.2265,
      "num_input_tokens_seen": 2352224,
      "step": 4045
    },
    {
      "epoch": 0.6032171581769437,
      "grad_norm": 8.9375,
      "learning_rate": 1.5076705391718799e-05,
      "loss": 0.8631,
      "num_input_tokens_seen": 2355040,
      "step": 4050
    },
    {
      "epoch": 0.6039618707179029,
      "grad_norm": 6.75,
      "learning_rate": 1.5095323205242775e-05,
      "loss": 1.2102,
      "num_input_tokens_seen": 2357728,
      "step": 4055
    },
    {
      "epoch": 0.6047065832588621,
      "grad_norm": 6.21875,
      "learning_rate": 1.5113941018766755e-05,
      "loss": 1.0701,
      "num_input_tokens_seen": 2360672,
      "step": 4060
    },
    {
      "epoch": 0.6054512957998213,
      "grad_norm": 14.125,
      "learning_rate": 1.5132558832290735e-05,
      "loss": 1.4014,
      "num_input_tokens_seen": 2363424,
      "step": 4065
    },
    {
      "epoch": 0.6061960083407805,
      "grad_norm": 14.1875,
      "learning_rate": 1.5151176645814715e-05,
      "loss": 1.368,
      "num_input_tokens_seen": 2366368,
      "step": 4070
    },
    {
      "epoch": 0.6069407208817397,
      "grad_norm": 18.125,
      "learning_rate": 1.5169794459338697e-05,
      "loss": 1.181,
      "num_input_tokens_seen": 2369312,
      "step": 4075
    },
    {
      "epoch": 0.6076854334226989,
      "grad_norm": 8.875,
      "learning_rate": 1.5188412272862677e-05,
      "loss": 1.0349,
      "num_input_tokens_seen": 2372352,
      "step": 4080
    },
    {
      "epoch": 0.6084301459636581,
      "grad_norm": 17.875,
      "learning_rate": 1.5207030086386656e-05,
      "loss": 1.0838,
      "num_input_tokens_seen": 2375264,
      "step": 4085
    },
    {
      "epoch": 0.6091748585046172,
      "grad_norm": 11.125,
      "learning_rate": 1.5225647899910636e-05,
      "loss": 1.1117,
      "num_input_tokens_seen": 2377984,
      "step": 4090
    },
    {
      "epoch": 0.6099195710455764,
      "grad_norm": 12.875,
      "learning_rate": 1.5244265713434616e-05,
      "loss": 1.116,
      "num_input_tokens_seen": 2381120,
      "step": 4095
    },
    {
      "epoch": 0.6106642835865356,
      "grad_norm": 17.5,
      "learning_rate": 1.5262883526958593e-05,
      "loss": 1.7865,
      "num_input_tokens_seen": 2383744,
      "step": 4100
    },
    {
      "epoch": 0.6114089961274948,
      "grad_norm": 10.5625,
      "learning_rate": 1.5281501340482574e-05,
      "loss": 1.2341,
      "num_input_tokens_seen": 2386432,
      "step": 4105
    },
    {
      "epoch": 0.612153708668454,
      "grad_norm": 15.9375,
      "learning_rate": 1.5300119154006553e-05,
      "loss": 1.3609,
      "num_input_tokens_seen": 2389312,
      "step": 4110
    },
    {
      "epoch": 0.6128984212094132,
      "grad_norm": 10.0625,
      "learning_rate": 1.5318736967530534e-05,
      "loss": 0.9984,
      "num_input_tokens_seen": 2392608,
      "step": 4115
    },
    {
      "epoch": 0.6136431337503724,
      "grad_norm": 8.4375,
      "learning_rate": 1.5337354781054513e-05,
      "loss": 1.0549,
      "num_input_tokens_seen": 2395296,
      "step": 4120
    },
    {
      "epoch": 0.6143878462913316,
      "grad_norm": 15.0625,
      "learning_rate": 1.5355972594578494e-05,
      "loss": 1.3223,
      "num_input_tokens_seen": 2398112,
      "step": 4125
    },
    {
      "epoch": 0.6151325588322908,
      "grad_norm": 5.875,
      "learning_rate": 1.5374590408102472e-05,
      "loss": 1.0636,
      "num_input_tokens_seen": 2400704,
      "step": 4130
    },
    {
      "epoch": 0.6158772713732499,
      "grad_norm": 11.125,
      "learning_rate": 1.5393208221626454e-05,
      "loss": 0.9107,
      "num_input_tokens_seen": 2403392,
      "step": 4135
    },
    {
      "epoch": 0.6166219839142091,
      "grad_norm": 6.375,
      "learning_rate": 1.5411826035150436e-05,
      "loss": 1.0405,
      "num_input_tokens_seen": 2406432,
      "step": 4140
    },
    {
      "epoch": 0.6173666964551683,
      "grad_norm": 8.625,
      "learning_rate": 1.543044384867441e-05,
      "loss": 0.9188,
      "num_input_tokens_seen": 2409376,
      "step": 4145
    },
    {
      "epoch": 0.6181114089961275,
      "grad_norm": 18.75,
      "learning_rate": 1.5449061662198392e-05,
      "loss": 1.3028,
      "num_input_tokens_seen": 2412864,
      "step": 4150
    },
    {
      "epoch": 0.6188561215370867,
      "grad_norm": 25.5,
      "learning_rate": 1.546767947572237e-05,
      "loss": 1.8477,
      "num_input_tokens_seen": 2415744,
      "step": 4155
    },
    {
      "epoch": 0.6196008340780459,
      "grad_norm": 24.625,
      "learning_rate": 1.5486297289246352e-05,
      "loss": 1.74,
      "num_input_tokens_seen": 2418560,
      "step": 4160
    },
    {
      "epoch": 0.6203455466190051,
      "grad_norm": 16.375,
      "learning_rate": 1.550491510277033e-05,
      "loss": 1.0361,
      "num_input_tokens_seen": 2421440,
      "step": 4165
    },
    {
      "epoch": 0.6210902591599643,
      "grad_norm": 8.25,
      "learning_rate": 1.5523532916294312e-05,
      "loss": 1.624,
      "num_input_tokens_seen": 2424640,
      "step": 4170
    },
    {
      "epoch": 0.6218349717009234,
      "grad_norm": 14.5625,
      "learning_rate": 1.554215072981829e-05,
      "loss": 1.3517,
      "num_input_tokens_seen": 2427488,
      "step": 4175
    },
    {
      "epoch": 0.6225796842418826,
      "grad_norm": 9.4375,
      "learning_rate": 1.5560768543342272e-05,
      "loss": 0.8761,
      "num_input_tokens_seen": 2430208,
      "step": 4180
    },
    {
      "epoch": 0.6233243967828418,
      "grad_norm": 12.8125,
      "learning_rate": 1.557938635686625e-05,
      "loss": 1.3473,
      "num_input_tokens_seen": 2432864,
      "step": 4185
    },
    {
      "epoch": 0.624069109323801,
      "grad_norm": 16.0,
      "learning_rate": 1.5598004170390228e-05,
      "loss": 1.328,
      "num_input_tokens_seen": 2435840,
      "step": 4190
    },
    {
      "epoch": 0.6248138218647602,
      "grad_norm": 17.625,
      "learning_rate": 1.561662198391421e-05,
      "loss": 1.4324,
      "num_input_tokens_seen": 2438560,
      "step": 4195
    },
    {
      "epoch": 0.6255585344057194,
      "grad_norm": 11.3125,
      "learning_rate": 1.5635239797438188e-05,
      "loss": 1.3031,
      "num_input_tokens_seen": 2441376,
      "step": 4200
    },
    {
      "epoch": 0.6263032469466786,
      "grad_norm": 13.125,
      "learning_rate": 1.565385761096217e-05,
      "loss": 1.0165,
      "num_input_tokens_seen": 2444288,
      "step": 4205
    },
    {
      "epoch": 0.6270479594876378,
      "grad_norm": 9.125,
      "learning_rate": 1.5672475424486148e-05,
      "loss": 1.2511,
      "num_input_tokens_seen": 2447168,
      "step": 4210
    },
    {
      "epoch": 0.627792672028597,
      "grad_norm": 9.3125,
      "learning_rate": 1.569109323801013e-05,
      "loss": 0.9771,
      "num_input_tokens_seen": 2450240,
      "step": 4215
    },
    {
      "epoch": 0.6285373845695561,
      "grad_norm": 10.4375,
      "learning_rate": 1.570971105153411e-05,
      "loss": 1.2339,
      "num_input_tokens_seen": 2453024,
      "step": 4220
    },
    {
      "epoch": 0.6292820971105153,
      "grad_norm": 10.25,
      "learning_rate": 1.572832886505809e-05,
      "loss": 0.8637,
      "num_input_tokens_seen": 2455872,
      "step": 4225
    },
    {
      "epoch": 0.6300268096514745,
      "grad_norm": 13.75,
      "learning_rate": 1.5746946678582068e-05,
      "loss": 1.3326,
      "num_input_tokens_seen": 2458624,
      "step": 4230
    },
    {
      "epoch": 0.6307715221924337,
      "grad_norm": 12.6875,
      "learning_rate": 1.5765564492106046e-05,
      "loss": 1.331,
      "num_input_tokens_seen": 2461312,
      "step": 4235
    },
    {
      "epoch": 0.6315162347333929,
      "grad_norm": 15.5625,
      "learning_rate": 1.5784182305630027e-05,
      "loss": 1.4073,
      "num_input_tokens_seen": 2464352,
      "step": 4240
    },
    {
      "epoch": 0.6322609472743521,
      "grad_norm": 15.875,
      "learning_rate": 1.5802800119154006e-05,
      "loss": 1.5776,
      "num_input_tokens_seen": 2467040,
      "step": 4245
    },
    {
      "epoch": 0.6330056598153113,
      "grad_norm": 25.25,
      "learning_rate": 1.5821417932677987e-05,
      "loss": 1.2092,
      "num_input_tokens_seen": 2470112,
      "step": 4250
    },
    {
      "epoch": 0.6337503723562705,
      "grad_norm": 10.9375,
      "learning_rate": 1.5840035746201966e-05,
      "loss": 1.3129,
      "num_input_tokens_seen": 2472768,
      "step": 4255
    },
    {
      "epoch": 0.6344950848972297,
      "grad_norm": 12.8125,
      "learning_rate": 1.5858653559725947e-05,
      "loss": 1.2722,
      "num_input_tokens_seen": 2475936,
      "step": 4260
    },
    {
      "epoch": 0.6352397974381888,
      "grad_norm": 17.625,
      "learning_rate": 1.587727137324993e-05,
      "loss": 1.2167,
      "num_input_tokens_seen": 2478688,
      "step": 4265
    },
    {
      "epoch": 0.635984509979148,
      "grad_norm": 10.0,
      "learning_rate": 1.5895889186773907e-05,
      "loss": 0.9456,
      "num_input_tokens_seen": 2481728,
      "step": 4270
    },
    {
      "epoch": 0.6367292225201072,
      "grad_norm": 13.9375,
      "learning_rate": 1.5914507000297885e-05,
      "loss": 1.0961,
      "num_input_tokens_seen": 2484416,
      "step": 4275
    },
    {
      "epoch": 0.6374739350610664,
      "grad_norm": 19.625,
      "learning_rate": 1.5933124813821863e-05,
      "loss": 1.2926,
      "num_input_tokens_seen": 2487296,
      "step": 4280
    },
    {
      "epoch": 0.6382186476020256,
      "grad_norm": 12.6875,
      "learning_rate": 1.5951742627345845e-05,
      "loss": 1.411,
      "num_input_tokens_seen": 2490336,
      "step": 4285
    },
    {
      "epoch": 0.6389633601429848,
      "grad_norm": 12.9375,
      "learning_rate": 1.5970360440869823e-05,
      "loss": 1.1459,
      "num_input_tokens_seen": 2493600,
      "step": 4290
    },
    {
      "epoch": 0.639708072683944,
      "grad_norm": 9.25,
      "learning_rate": 1.5988978254393805e-05,
      "loss": 1.1375,
      "num_input_tokens_seen": 2496352,
      "step": 4295
    },
    {
      "epoch": 0.6404527852249032,
      "grad_norm": 16.875,
      "learning_rate": 1.6007596067917787e-05,
      "loss": 1.4896,
      "num_input_tokens_seen": 2499296,
      "step": 4300
    },
    {
      "epoch": 0.6411974977658623,
      "grad_norm": 8.75,
      "learning_rate": 1.6026213881441765e-05,
      "loss": 0.9617,
      "num_input_tokens_seen": 2502080,
      "step": 4305
    },
    {
      "epoch": 0.6419422103068215,
      "grad_norm": 11.75,
      "learning_rate": 1.6044831694965746e-05,
      "loss": 0.9484,
      "num_input_tokens_seen": 2504832,
      "step": 4310
    },
    {
      "epoch": 0.6426869228477807,
      "grad_norm": 4.625,
      "learning_rate": 1.606344950848972e-05,
      "loss": 0.9852,
      "num_input_tokens_seen": 2507776,
      "step": 4315
    },
    {
      "epoch": 0.6434316353887399,
      "grad_norm": 13.8125,
      "learning_rate": 1.6082067322013703e-05,
      "loss": 1.4213,
      "num_input_tokens_seen": 2510720,
      "step": 4320
    },
    {
      "epoch": 0.6441763479296991,
      "grad_norm": 15.1875,
      "learning_rate": 1.610068513553768e-05,
      "loss": 0.942,
      "num_input_tokens_seen": 2513600,
      "step": 4325
    },
    {
      "epoch": 0.6449210604706583,
      "grad_norm": 11.375,
      "learning_rate": 1.6119302949061663e-05,
      "loss": 1.2184,
      "num_input_tokens_seen": 2516576,
      "step": 4330
    },
    {
      "epoch": 0.6456657730116175,
      "grad_norm": 11.6875,
      "learning_rate": 1.6137920762585644e-05,
      "loss": 1.2122,
      "num_input_tokens_seen": 2519296,
      "step": 4335
    },
    {
      "epoch": 0.6464104855525767,
      "grad_norm": 20.0,
      "learning_rate": 1.6156538576109623e-05,
      "loss": 1.1515,
      "num_input_tokens_seen": 2522112,
      "step": 4340
    },
    {
      "epoch": 0.6471551980935359,
      "grad_norm": 10.8125,
      "learning_rate": 1.6175156389633604e-05,
      "loss": 1.1925,
      "num_input_tokens_seen": 2525280,
      "step": 4345
    },
    {
      "epoch": 0.647899910634495,
      "grad_norm": 13.25,
      "learning_rate": 1.6193774203157582e-05,
      "loss": 1.1419,
      "num_input_tokens_seen": 2528288,
      "step": 4350
    },
    {
      "epoch": 0.6486446231754542,
      "grad_norm": 17.875,
      "learning_rate": 1.6212392016681564e-05,
      "loss": 1.5911,
      "num_input_tokens_seen": 2531040,
      "step": 4355
    },
    {
      "epoch": 0.6493893357164134,
      "grad_norm": 6.78125,
      "learning_rate": 1.623100983020554e-05,
      "loss": 0.9501,
      "num_input_tokens_seen": 2533952,
      "step": 4360
    },
    {
      "epoch": 0.6501340482573726,
      "grad_norm": 5.0,
      "learning_rate": 1.624962764372952e-05,
      "loss": 1.3048,
      "num_input_tokens_seen": 2536864,
      "step": 4365
    },
    {
      "epoch": 0.6508787607983318,
      "grad_norm": 12.0,
      "learning_rate": 1.62682454572535e-05,
      "loss": 0.8969,
      "num_input_tokens_seen": 2539552,
      "step": 4370
    },
    {
      "epoch": 0.651623473339291,
      "grad_norm": 17.25,
      "learning_rate": 1.628686327077748e-05,
      "loss": 1.2031,
      "num_input_tokens_seen": 2542208,
      "step": 4375
    },
    {
      "epoch": 0.6523681858802503,
      "grad_norm": 17.375,
      "learning_rate": 1.6305481084301462e-05,
      "loss": 1.4816,
      "num_input_tokens_seen": 2545152,
      "step": 4380
    },
    {
      "epoch": 0.6531128984212095,
      "grad_norm": 18.75,
      "learning_rate": 1.632409889782544e-05,
      "loss": 1.5802,
      "num_input_tokens_seen": 2548032,
      "step": 4385
    },
    {
      "epoch": 0.6538576109621687,
      "grad_norm": 20.0,
      "learning_rate": 1.6342716711349422e-05,
      "loss": 0.9993,
      "num_input_tokens_seen": 2550976,
      "step": 4390
    },
    {
      "epoch": 0.6546023235031277,
      "grad_norm": 6.40625,
      "learning_rate": 1.63613345248734e-05,
      "loss": 1.1599,
      "num_input_tokens_seen": 2554336,
      "step": 4395
    },
    {
      "epoch": 0.655347036044087,
      "grad_norm": 10.5,
      "learning_rate": 1.637995233839738e-05,
      "loss": 1.0785,
      "num_input_tokens_seen": 2557504,
      "step": 4400
    },
    {
      "epoch": 0.6560917485850462,
      "grad_norm": 18.75,
      "learning_rate": 1.6398570151921357e-05,
      "loss": 1.1537,
      "num_input_tokens_seen": 2560288,
      "step": 4405
    },
    {
      "epoch": 0.6568364611260054,
      "grad_norm": 20.625,
      "learning_rate": 1.6417187965445338e-05,
      "loss": 1.3917,
      "num_input_tokens_seen": 2563136,
      "step": 4410
    },
    {
      "epoch": 0.6575811736669646,
      "grad_norm": 8.1875,
      "learning_rate": 1.643580577896932e-05,
      "loss": 0.7703,
      "num_input_tokens_seen": 2566144,
      "step": 4415
    },
    {
      "epoch": 0.6583258862079238,
      "grad_norm": 16.375,
      "learning_rate": 1.6454423592493298e-05,
      "loss": 1.2149,
      "num_input_tokens_seen": 2568800,
      "step": 4420
    },
    {
      "epoch": 0.659070598748883,
      "grad_norm": 17.25,
      "learning_rate": 1.647304140601728e-05,
      "loss": 1.119,
      "num_input_tokens_seen": 2571840,
      "step": 4425
    },
    {
      "epoch": 0.6598153112898422,
      "grad_norm": 12.4375,
      "learning_rate": 1.6491659219541258e-05,
      "loss": 1.2227,
      "num_input_tokens_seen": 2574688,
      "step": 4430
    },
    {
      "epoch": 0.6605600238308013,
      "grad_norm": 17.0,
      "learning_rate": 1.651027703306524e-05,
      "loss": 0.9153,
      "num_input_tokens_seen": 2577376,
      "step": 4435
    },
    {
      "epoch": 0.6613047363717605,
      "grad_norm": 20.875,
      "learning_rate": 1.6528894846589218e-05,
      "loss": 1.2269,
      "num_input_tokens_seen": 2580448,
      "step": 4440
    },
    {
      "epoch": 0.6620494489127197,
      "grad_norm": 12.875,
      "learning_rate": 1.6547512660113196e-05,
      "loss": 1.0186,
      "num_input_tokens_seen": 2583168,
      "step": 4445
    },
    {
      "epoch": 0.6627941614536789,
      "grad_norm": 13.8125,
      "learning_rate": 1.6566130473637174e-05,
      "loss": 0.9662,
      "num_input_tokens_seen": 2585984,
      "step": 4450
    },
    {
      "epoch": 0.6635388739946381,
      "grad_norm": 12.5,
      "learning_rate": 1.6584748287161156e-05,
      "loss": 1.6783,
      "num_input_tokens_seen": 2588928,
      "step": 4455
    },
    {
      "epoch": 0.6642835865355973,
      "grad_norm": 13.875,
      "learning_rate": 1.6603366100685137e-05,
      "loss": 1.019,
      "num_input_tokens_seen": 2591584,
      "step": 4460
    },
    {
      "epoch": 0.6650282990765565,
      "grad_norm": 16.375,
      "learning_rate": 1.6621983914209116e-05,
      "loss": 1.0045,
      "num_input_tokens_seen": 2594496,
      "step": 4465
    },
    {
      "epoch": 0.6657730116175157,
      "grad_norm": 18.5,
      "learning_rate": 1.6640601727733097e-05,
      "loss": 1.3045,
      "num_input_tokens_seen": 2597408,
      "step": 4470
    },
    {
      "epoch": 0.6665177241584749,
      "grad_norm": 11.6875,
      "learning_rate": 1.6659219541257075e-05,
      "loss": 1.0828,
      "num_input_tokens_seen": 2600320,
      "step": 4475
    },
    {
      "epoch": 0.667262436699434,
      "grad_norm": 6.53125,
      "learning_rate": 1.6677837354781057e-05,
      "loss": 1.0405,
      "num_input_tokens_seen": 2603072,
      "step": 4480
    },
    {
      "epoch": 0.6680071492403932,
      "grad_norm": 11.0,
      "learning_rate": 1.6696455168305035e-05,
      "loss": 1.2953,
      "num_input_tokens_seen": 2605728,
      "step": 4485
    },
    {
      "epoch": 0.6687518617813524,
      "grad_norm": 6.9375,
      "learning_rate": 1.6715072981829014e-05,
      "loss": 1.0262,
      "num_input_tokens_seen": 2608320,
      "step": 4490
    },
    {
      "epoch": 0.6694965743223116,
      "grad_norm": 11.4375,
      "learning_rate": 1.6733690795352995e-05,
      "loss": 1.1376,
      "num_input_tokens_seen": 2611168,
      "step": 4495
    },
    {
      "epoch": 0.6702412868632708,
      "grad_norm": 16.375,
      "learning_rate": 1.6752308608876973e-05,
      "loss": 1.1313,
      "num_input_tokens_seen": 2614016,
      "step": 4500
    },
    {
      "epoch": 0.67098599940423,
      "grad_norm": 13.875,
      "learning_rate": 1.6770926422400955e-05,
      "loss": 1.3165,
      "num_input_tokens_seen": 2616864,
      "step": 4505
    },
    {
      "epoch": 0.6717307119451892,
      "grad_norm": 16.0,
      "learning_rate": 1.6789544235924933e-05,
      "loss": 1.585,
      "num_input_tokens_seen": 2619808,
      "step": 4510
    },
    {
      "epoch": 0.6724754244861484,
      "grad_norm": 22.625,
      "learning_rate": 1.6808162049448915e-05,
      "loss": 1.028,
      "num_input_tokens_seen": 2622464,
      "step": 4515
    },
    {
      "epoch": 0.6732201370271076,
      "grad_norm": 10.25,
      "learning_rate": 1.6826779862972893e-05,
      "loss": 0.9953,
      "num_input_tokens_seen": 2625216,
      "step": 4520
    },
    {
      "epoch": 0.6739648495680667,
      "grad_norm": 7.25,
      "learning_rate": 1.6845397676496875e-05,
      "loss": 0.9925,
      "num_input_tokens_seen": 2628096,
      "step": 4525
    },
    {
      "epoch": 0.6747095621090259,
      "grad_norm": 14.1875,
      "learning_rate": 1.6864015490020853e-05,
      "loss": 1.0648,
      "num_input_tokens_seen": 2631072,
      "step": 4530
    },
    {
      "epoch": 0.6754542746499851,
      "grad_norm": 15.3125,
      "learning_rate": 1.688263330354483e-05,
      "loss": 1.1637,
      "num_input_tokens_seen": 2634016,
      "step": 4535
    },
    {
      "epoch": 0.6761989871909443,
      "grad_norm": 9.8125,
      "learning_rate": 1.6901251117068813e-05,
      "loss": 1.0923,
      "num_input_tokens_seen": 2637376,
      "step": 4540
    },
    {
      "epoch": 0.6769436997319035,
      "grad_norm": 7.40625,
      "learning_rate": 1.691986893059279e-05,
      "loss": 0.9375,
      "num_input_tokens_seen": 2640192,
      "step": 4545
    },
    {
      "epoch": 0.6776884122728627,
      "grad_norm": 15.8125,
      "learning_rate": 1.6938486744116773e-05,
      "loss": 1.3025,
      "num_input_tokens_seen": 2643040,
      "step": 4550
    },
    {
      "epoch": 0.6784331248138219,
      "grad_norm": 11.8125,
      "learning_rate": 1.695710455764075e-05,
      "loss": 0.7952,
      "num_input_tokens_seen": 2645856,
      "step": 4555
    },
    {
      "epoch": 0.6791778373547811,
      "grad_norm": 15.125,
      "learning_rate": 1.6975722371164733e-05,
      "loss": 1.5888,
      "num_input_tokens_seen": 2648928,
      "step": 4560
    },
    {
      "epoch": 0.6799225498957402,
      "grad_norm": 14.8125,
      "learning_rate": 1.699434018468871e-05,
      "loss": 1.1287,
      "num_input_tokens_seen": 2651648,
      "step": 4565
    },
    {
      "epoch": 0.6806672624366994,
      "grad_norm": 11.0,
      "learning_rate": 1.7012957998212692e-05,
      "loss": 1.4573,
      "num_input_tokens_seen": 2654528,
      "step": 4570
    },
    {
      "epoch": 0.6814119749776586,
      "grad_norm": 8.0,
      "learning_rate": 1.703157581173667e-05,
      "loss": 1.2286,
      "num_input_tokens_seen": 2657376,
      "step": 4575
    },
    {
      "epoch": 0.6821566875186178,
      "grad_norm": 11.8125,
      "learning_rate": 1.705019362526065e-05,
      "loss": 0.9924,
      "num_input_tokens_seen": 2660160,
      "step": 4580
    },
    {
      "epoch": 0.682901400059577,
      "grad_norm": 13.4375,
      "learning_rate": 1.706881143878463e-05,
      "loss": 1.1938,
      "num_input_tokens_seen": 2663520,
      "step": 4585
    },
    {
      "epoch": 0.6836461126005362,
      "grad_norm": 21.125,
      "learning_rate": 1.708742925230861e-05,
      "loss": 1.5845,
      "num_input_tokens_seen": 2666272,
      "step": 4590
    },
    {
      "epoch": 0.6843908251414954,
      "grad_norm": 19.625,
      "learning_rate": 1.710604706583259e-05,
      "loss": 1.0194,
      "num_input_tokens_seen": 2669248,
      "step": 4595
    },
    {
      "epoch": 0.6851355376824546,
      "grad_norm": 17.375,
      "learning_rate": 1.712466487935657e-05,
      "loss": 1.1956,
      "num_input_tokens_seen": 2672064,
      "step": 4600
    },
    {
      "epoch": 0.6858802502234138,
      "grad_norm": 12.5625,
      "learning_rate": 1.714328269288055e-05,
      "loss": 1.39,
      "num_input_tokens_seen": 2675008,
      "step": 4605
    },
    {
      "epoch": 0.6866249627643729,
      "grad_norm": 10.1875,
      "learning_rate": 1.716190050640453e-05,
      "loss": 1.084,
      "num_input_tokens_seen": 2678080,
      "step": 4610
    },
    {
      "epoch": 0.6873696753053321,
      "grad_norm": 10.875,
      "learning_rate": 1.718051831992851e-05,
      "loss": 1.217,
      "num_input_tokens_seen": 2680960,
      "step": 4615
    },
    {
      "epoch": 0.6881143878462913,
      "grad_norm": 10.125,
      "learning_rate": 1.7199136133452488e-05,
      "loss": 1.2268,
      "num_input_tokens_seen": 2683904,
      "step": 4620
    },
    {
      "epoch": 0.6888591003872505,
      "grad_norm": 21.25,
      "learning_rate": 1.7217753946976467e-05,
      "loss": 1.4294,
      "num_input_tokens_seen": 2686880,
      "step": 4625
    },
    {
      "epoch": 0.6896038129282097,
      "grad_norm": 19.75,
      "learning_rate": 1.7236371760500448e-05,
      "loss": 1.2029,
      "num_input_tokens_seen": 2689536,
      "step": 4630
    },
    {
      "epoch": 0.6903485254691689,
      "grad_norm": 17.125,
      "learning_rate": 1.7254989574024426e-05,
      "loss": 1.2141,
      "num_input_tokens_seen": 2692384,
      "step": 4635
    },
    {
      "epoch": 0.6910932380101281,
      "grad_norm": 8.6875,
      "learning_rate": 1.7273607387548408e-05,
      "loss": 1.2945,
      "num_input_tokens_seen": 2695424,
      "step": 4640
    },
    {
      "epoch": 0.6918379505510873,
      "grad_norm": 15.9375,
      "learning_rate": 1.7292225201072386e-05,
      "loss": 1.0696,
      "num_input_tokens_seen": 2698464,
      "step": 4645
    },
    {
      "epoch": 0.6925826630920465,
      "grad_norm": 16.375,
      "learning_rate": 1.7310843014596368e-05,
      "loss": 0.8657,
      "num_input_tokens_seen": 2701184,
      "step": 4650
    },
    {
      "epoch": 0.6933273756330056,
      "grad_norm": 9.0,
      "learning_rate": 1.7329460828120346e-05,
      "loss": 0.9871,
      "num_input_tokens_seen": 2704160,
      "step": 4655
    },
    {
      "epoch": 0.6940720881739648,
      "grad_norm": 13.3125,
      "learning_rate": 1.7348078641644328e-05,
      "loss": 1.0602,
      "num_input_tokens_seen": 2706944,
      "step": 4660
    },
    {
      "epoch": 0.694816800714924,
      "grad_norm": 14.6875,
      "learning_rate": 1.7366696455168306e-05,
      "loss": 1.3195,
      "num_input_tokens_seen": 2709824,
      "step": 4665
    },
    {
      "epoch": 0.6955615132558832,
      "grad_norm": 15.4375,
      "learning_rate": 1.7385314268692284e-05,
      "loss": 1.3388,
      "num_input_tokens_seen": 2712448,
      "step": 4670
    },
    {
      "epoch": 0.6963062257968424,
      "grad_norm": 5.59375,
      "learning_rate": 1.7403932082216266e-05,
      "loss": 0.7673,
      "num_input_tokens_seen": 2715456,
      "step": 4675
    },
    {
      "epoch": 0.6970509383378016,
      "grad_norm": 12.75,
      "learning_rate": 1.7422549895740244e-05,
      "loss": 1.2854,
      "num_input_tokens_seen": 2718080,
      "step": 4680
    },
    {
      "epoch": 0.6977956508787608,
      "grad_norm": 12.625,
      "learning_rate": 1.7441167709264226e-05,
      "loss": 1.0026,
      "num_input_tokens_seen": 2720928,
      "step": 4685
    },
    {
      "epoch": 0.69854036341972,
      "grad_norm": 15.625,
      "learning_rate": 1.7459785522788204e-05,
      "loss": 1.2461,
      "num_input_tokens_seen": 2724096,
      "step": 4690
    },
    {
      "epoch": 0.6992850759606791,
      "grad_norm": 11.125,
      "learning_rate": 1.7478403336312185e-05,
      "loss": 1.0114,
      "num_input_tokens_seen": 2727264,
      "step": 4695
    },
    {
      "epoch": 0.7000297885016383,
      "grad_norm": 14.875,
      "learning_rate": 1.7497021149836164e-05,
      "loss": 1.1284,
      "num_input_tokens_seen": 2729760,
      "step": 4700
    },
    {
      "epoch": 0.7007745010425975,
      "grad_norm": 14.4375,
      "learning_rate": 1.7515638963360142e-05,
      "loss": 1.1619,
      "num_input_tokens_seen": 2732416,
      "step": 4705
    },
    {
      "epoch": 0.7015192135835567,
      "grad_norm": 15.0,
      "learning_rate": 1.7534256776884124e-05,
      "loss": 0.9171,
      "num_input_tokens_seen": 2735072,
      "step": 4710
    },
    {
      "epoch": 0.7022639261245159,
      "grad_norm": 14.6875,
      "learning_rate": 1.7552874590408102e-05,
      "loss": 1.3304,
      "num_input_tokens_seen": 2737792,
      "step": 4715
    },
    {
      "epoch": 0.7030086386654751,
      "grad_norm": 15.0,
      "learning_rate": 1.7571492403932083e-05,
      "loss": 1.2396,
      "num_input_tokens_seen": 2740704,
      "step": 4720
    },
    {
      "epoch": 0.7037533512064343,
      "grad_norm": 12.9375,
      "learning_rate": 1.759011021745606e-05,
      "loss": 1.1757,
      "num_input_tokens_seen": 2743520,
      "step": 4725
    },
    {
      "epoch": 0.7044980637473935,
      "grad_norm": 16.5,
      "learning_rate": 1.7608728030980043e-05,
      "loss": 0.823,
      "num_input_tokens_seen": 2746368,
      "step": 4730
    },
    {
      "epoch": 0.7052427762883527,
      "grad_norm": 18.75,
      "learning_rate": 1.762734584450402e-05,
      "loss": 0.99,
      "num_input_tokens_seen": 2749184,
      "step": 4735
    },
    {
      "epoch": 0.7059874888293118,
      "grad_norm": 16.125,
      "learning_rate": 1.7645963658028003e-05,
      "loss": 1.1239,
      "num_input_tokens_seen": 2752000,
      "step": 4740
    },
    {
      "epoch": 0.706732201370271,
      "grad_norm": 7.0,
      "learning_rate": 1.7664581471551985e-05,
      "loss": 1.1337,
      "num_input_tokens_seen": 2754688,
      "step": 4745
    },
    {
      "epoch": 0.7074769139112302,
      "grad_norm": 8.25,
      "learning_rate": 1.768319928507596e-05,
      "loss": 1.3207,
      "num_input_tokens_seen": 2757472,
      "step": 4750
    },
    {
      "epoch": 0.7082216264521894,
      "grad_norm": 16.5,
      "learning_rate": 1.770181709859994e-05,
      "loss": 1.0851,
      "num_input_tokens_seen": 2760288,
      "step": 4755
    },
    {
      "epoch": 0.7089663389931486,
      "grad_norm": 12.125,
      "learning_rate": 1.772043491212392e-05,
      "loss": 1.0767,
      "num_input_tokens_seen": 2763040,
      "step": 4760
    },
    {
      "epoch": 0.7097110515341079,
      "grad_norm": 15.875,
      "learning_rate": 1.77390527256479e-05,
      "loss": 1.1685,
      "num_input_tokens_seen": 2766176,
      "step": 4765
    },
    {
      "epoch": 0.710455764075067,
      "grad_norm": 9.0,
      "learning_rate": 1.775767053917188e-05,
      "loss": 1.0113,
      "num_input_tokens_seen": 2769408,
      "step": 4770
    },
    {
      "epoch": 0.7112004766160263,
      "grad_norm": 11.25,
      "learning_rate": 1.777628835269586e-05,
      "loss": 1.2574,
      "num_input_tokens_seen": 2772256,
      "step": 4775
    },
    {
      "epoch": 0.7119451891569855,
      "grad_norm": 9.8125,
      "learning_rate": 1.779490616621984e-05,
      "loss": 1.2515,
      "num_input_tokens_seen": 2774912,
      "step": 4780
    },
    {
      "epoch": 0.7126899016979446,
      "grad_norm": 10.0,
      "learning_rate": 1.781352397974382e-05,
      "loss": 1.084,
      "num_input_tokens_seen": 2777600,
      "step": 4785
    },
    {
      "epoch": 0.7134346142389038,
      "grad_norm": 8.0625,
      "learning_rate": 1.78321417932678e-05,
      "loss": 0.8989,
      "num_input_tokens_seen": 2780480,
      "step": 4790
    },
    {
      "epoch": 0.714179326779863,
      "grad_norm": 13.4375,
      "learning_rate": 1.7850759606791777e-05,
      "loss": 1.3449,
      "num_input_tokens_seen": 2783648,
      "step": 4795
    },
    {
      "epoch": 0.7149240393208222,
      "grad_norm": 13.125,
      "learning_rate": 1.786937742031576e-05,
      "loss": 1.088,
      "num_input_tokens_seen": 2786880,
      "step": 4800
    },
    {
      "epoch": 0.7156687518617814,
      "grad_norm": 9.25,
      "learning_rate": 1.7887995233839737e-05,
      "loss": 0.8539,
      "num_input_tokens_seen": 2789440,
      "step": 4805
    },
    {
      "epoch": 0.7164134644027406,
      "grad_norm": 11.0,
      "learning_rate": 1.790661304736372e-05,
      "loss": 1.2079,
      "num_input_tokens_seen": 2792640,
      "step": 4810
    },
    {
      "epoch": 0.7171581769436998,
      "grad_norm": 14.0,
      "learning_rate": 1.7925230860887697e-05,
      "loss": 0.8332,
      "num_input_tokens_seen": 2795616,
      "step": 4815
    },
    {
      "epoch": 0.717902889484659,
      "grad_norm": 18.125,
      "learning_rate": 1.794384867441168e-05,
      "loss": 1.1951,
      "num_input_tokens_seen": 2798752,
      "step": 4820
    },
    {
      "epoch": 0.7186476020256181,
      "grad_norm": 16.0,
      "learning_rate": 1.796246648793566e-05,
      "loss": 1.3752,
      "num_input_tokens_seen": 2801344,
      "step": 4825
    },
    {
      "epoch": 0.7193923145665773,
      "grad_norm": 5.96875,
      "learning_rate": 1.798108430145964e-05,
      "loss": 1.2215,
      "num_input_tokens_seen": 2804384,
      "step": 4830
    },
    {
      "epoch": 0.7201370271075365,
      "grad_norm": 15.125,
      "learning_rate": 1.7999702114983617e-05,
      "loss": 1.0032,
      "num_input_tokens_seen": 2807456,
      "step": 4835
    },
    {
      "epoch": 0.7208817396484957,
      "grad_norm": 15.25,
      "learning_rate": 1.8018319928507595e-05,
      "loss": 1.296,
      "num_input_tokens_seen": 2810400,
      "step": 4840
    },
    {
      "epoch": 0.7216264521894549,
      "grad_norm": 10.4375,
      "learning_rate": 1.8036937742031576e-05,
      "loss": 0.7379,
      "num_input_tokens_seen": 2813408,
      "step": 4845
    },
    {
      "epoch": 0.7223711647304141,
      "grad_norm": 9.9375,
      "learning_rate": 1.8055555555555555e-05,
      "loss": 1.2818,
      "num_input_tokens_seen": 2816384,
      "step": 4850
    },
    {
      "epoch": 0.7231158772713733,
      "grad_norm": 11.875,
      "learning_rate": 1.8074173369079536e-05,
      "loss": 1.2918,
      "num_input_tokens_seen": 2819552,
      "step": 4855
    },
    {
      "epoch": 0.7238605898123325,
      "grad_norm": 13.0,
      "learning_rate": 1.8092791182603515e-05,
      "loss": 1.0089,
      "num_input_tokens_seen": 2822496,
      "step": 4860
    },
    {
      "epoch": 0.7246053023532917,
      "grad_norm": 18.5,
      "learning_rate": 1.8111408996127496e-05,
      "loss": 1.1496,
      "num_input_tokens_seen": 2825472,
      "step": 4865
    },
    {
      "epoch": 0.7253500148942508,
      "grad_norm": 14.375,
      "learning_rate": 1.8130026809651478e-05,
      "loss": 0.9073,
      "num_input_tokens_seen": 2828288,
      "step": 4870
    },
    {
      "epoch": 0.72609472743521,
      "grad_norm": 10.6875,
      "learning_rate": 1.8148644623175456e-05,
      "loss": 1.2221,
      "num_input_tokens_seen": 2831232,
      "step": 4875
    },
    {
      "epoch": 0.7268394399761692,
      "grad_norm": 16.125,
      "learning_rate": 1.8167262436699434e-05,
      "loss": 0.9704,
      "num_input_tokens_seen": 2834208,
      "step": 4880
    },
    {
      "epoch": 0.7275841525171284,
      "grad_norm": 7.875,
      "learning_rate": 1.8185880250223413e-05,
      "loss": 1.0514,
      "num_input_tokens_seen": 2836832,
      "step": 4885
    },
    {
      "epoch": 0.7283288650580876,
      "grad_norm": 19.5,
      "learning_rate": 1.8204498063747394e-05,
      "loss": 1.2171,
      "num_input_tokens_seen": 2839296,
      "step": 4890
    },
    {
      "epoch": 0.7290735775990468,
      "grad_norm": 12.3125,
      "learning_rate": 1.8223115877271372e-05,
      "loss": 1.1758,
      "num_input_tokens_seen": 2842048,
      "step": 4895
    },
    {
      "epoch": 0.729818290140006,
      "grad_norm": 14.5,
      "learning_rate": 1.8241733690795354e-05,
      "loss": 1.4879,
      "num_input_tokens_seen": 2844896,
      "step": 4900
    },
    {
      "epoch": 0.7305630026809652,
      "grad_norm": 10.6875,
      "learning_rate": 1.8260351504319336e-05,
      "loss": 1.1588,
      "num_input_tokens_seen": 2847552,
      "step": 4905
    },
    {
      "epoch": 0.7313077152219244,
      "grad_norm": 7.1875,
      "learning_rate": 1.8278969317843314e-05,
      "loss": 0.7118,
      "num_input_tokens_seen": 2850368,
      "step": 4910
    },
    {
      "epoch": 0.7320524277628835,
      "grad_norm": 15.25,
      "learning_rate": 1.8297587131367295e-05,
      "loss": 1.1099,
      "num_input_tokens_seen": 2853344,
      "step": 4915
    },
    {
      "epoch": 0.7327971403038427,
      "grad_norm": 10.875,
      "learning_rate": 1.831620494489127e-05,
      "loss": 1.4519,
      "num_input_tokens_seen": 2856416,
      "step": 4920
    },
    {
      "epoch": 0.7335418528448019,
      "grad_norm": 18.0,
      "learning_rate": 1.8334822758415252e-05,
      "loss": 1.3286,
      "num_input_tokens_seen": 2859360,
      "step": 4925
    },
    {
      "epoch": 0.7342865653857611,
      "grad_norm": 9.375,
      "learning_rate": 1.835344057193923e-05,
      "loss": 1.3145,
      "num_input_tokens_seen": 2862368,
      "step": 4930
    },
    {
      "epoch": 0.7350312779267203,
      "grad_norm": 15.875,
      "learning_rate": 1.8372058385463212e-05,
      "loss": 1.4988,
      "num_input_tokens_seen": 2865120,
      "step": 4935
    },
    {
      "epoch": 0.7357759904676795,
      "grad_norm": 12.25,
      "learning_rate": 1.8390676198987193e-05,
      "loss": 1.4853,
      "num_input_tokens_seen": 2868224,
      "step": 4940
    },
    {
      "epoch": 0.7365207030086387,
      "grad_norm": 6.75,
      "learning_rate": 1.840929401251117e-05,
      "loss": 1.1801,
      "num_input_tokens_seen": 2871360,
      "step": 4945
    },
    {
      "epoch": 0.7372654155495979,
      "grad_norm": 10.6875,
      "learning_rate": 1.8427911826035153e-05,
      "loss": 0.9588,
      "num_input_tokens_seen": 2874496,
      "step": 4950
    },
    {
      "epoch": 0.738010128090557,
      "grad_norm": 14.75,
      "learning_rate": 1.844652963955913e-05,
      "loss": 1.2177,
      "num_input_tokens_seen": 2877248,
      "step": 4955
    },
    {
      "epoch": 0.7387548406315162,
      "grad_norm": 13.3125,
      "learning_rate": 1.8465147453083113e-05,
      "loss": 0.8132,
      "num_input_tokens_seen": 2880128,
      "step": 4960
    },
    {
      "epoch": 0.7394995531724754,
      "grad_norm": 14.875,
      "learning_rate": 1.8483765266607088e-05,
      "loss": 1.1337,
      "num_input_tokens_seen": 2882944,
      "step": 4965
    },
    {
      "epoch": 0.7402442657134346,
      "grad_norm": 12.125,
      "learning_rate": 1.850238308013107e-05,
      "loss": 1.473,
      "num_input_tokens_seen": 2885856,
      "step": 4970
    },
    {
      "epoch": 0.7409889782543938,
      "grad_norm": 21.75,
      "learning_rate": 1.8521000893655048e-05,
      "loss": 1.5026,
      "num_input_tokens_seen": 2888768,
      "step": 4975
    },
    {
      "epoch": 0.741733690795353,
      "grad_norm": 5.96875,
      "learning_rate": 1.853961870717903e-05,
      "loss": 1.0372,
      "num_input_tokens_seen": 2891648,
      "step": 4980
    },
    {
      "epoch": 0.7424784033363122,
      "grad_norm": 14.3125,
      "learning_rate": 1.855823652070301e-05,
      "loss": 1.6106,
      "num_input_tokens_seen": 2894464,
      "step": 4985
    },
    {
      "epoch": 0.7432231158772714,
      "grad_norm": 12.9375,
      "learning_rate": 1.857685433422699e-05,
      "loss": 1.0421,
      "num_input_tokens_seen": 2897088,
      "step": 4990
    },
    {
      "epoch": 0.7439678284182306,
      "grad_norm": 28.875,
      "learning_rate": 1.859547214775097e-05,
      "loss": 1.4757,
      "num_input_tokens_seen": 2900032,
      "step": 4995
    },
    {
      "epoch": 0.7447125409591897,
      "grad_norm": 11.6875,
      "learning_rate": 1.861408996127495e-05,
      "loss": 1.2612,
      "num_input_tokens_seen": 2902944,
      "step": 5000
    },
    {
      "epoch": 0.7454572535001489,
      "grad_norm": 17.625,
      "learning_rate": 1.863270777479893e-05,
      "loss": 1.261,
      "num_input_tokens_seen": 2905856,
      "step": 5005
    },
    {
      "epoch": 0.7462019660411081,
      "grad_norm": 7.9375,
      "learning_rate": 1.8651325588322906e-05,
      "loss": 1.1032,
      "num_input_tokens_seen": 2908544,
      "step": 5010
    },
    {
      "epoch": 0.7469466785820673,
      "grad_norm": 6.90625,
      "learning_rate": 1.8669943401846887e-05,
      "loss": 1.6839,
      "num_input_tokens_seen": 2911584,
      "step": 5015
    },
    {
      "epoch": 0.7476913911230265,
      "grad_norm": 14.625,
      "learning_rate": 1.868856121537087e-05,
      "loss": 1.0103,
      "num_input_tokens_seen": 2914752,
      "step": 5020
    },
    {
      "epoch": 0.7484361036639857,
      "grad_norm": 12.9375,
      "learning_rate": 1.8707179028894847e-05,
      "loss": 1.2756,
      "num_input_tokens_seen": 2917472,
      "step": 5025
    },
    {
      "epoch": 0.7491808162049449,
      "grad_norm": 10.625,
      "learning_rate": 1.872579684241883e-05,
      "loss": 1.4888,
      "num_input_tokens_seen": 2920224,
      "step": 5030
    },
    {
      "epoch": 0.7499255287459041,
      "grad_norm": 13.3125,
      "learning_rate": 1.8744414655942807e-05,
      "loss": 0.9365,
      "num_input_tokens_seen": 2923136,
      "step": 5035
    },
    {
      "epoch": 0.7506702412868632,
      "grad_norm": 11.625,
      "learning_rate": 1.876303246946679e-05,
      "loss": 0.9503,
      "num_input_tokens_seen": 2926816,
      "step": 5040
    },
    {
      "epoch": 0.7514149538278224,
      "grad_norm": 15.5625,
      "learning_rate": 1.8781650282990767e-05,
      "loss": 1.5346,
      "num_input_tokens_seen": 2929504,
      "step": 5045
    },
    {
      "epoch": 0.7521596663687816,
      "grad_norm": 13.6875,
      "learning_rate": 1.8800268096514745e-05,
      "loss": 1.2784,
      "num_input_tokens_seen": 2932448,
      "step": 5050
    },
    {
      "epoch": 0.7529043789097408,
      "grad_norm": 12.875,
      "learning_rate": 1.8818885910038723e-05,
      "loss": 1.4429,
      "num_input_tokens_seen": 2935648,
      "step": 5055
    },
    {
      "epoch": 0.7536490914507,
      "grad_norm": 11.875,
      "learning_rate": 1.8837503723562705e-05,
      "loss": 1.3516,
      "num_input_tokens_seen": 2938496,
      "step": 5060
    },
    {
      "epoch": 0.7543938039916592,
      "grad_norm": 20.125,
      "learning_rate": 1.8856121537086686e-05,
      "loss": 0.9542,
      "num_input_tokens_seen": 2941536,
      "step": 5065
    },
    {
      "epoch": 0.7551385165326184,
      "grad_norm": 14.375,
      "learning_rate": 1.8874739350610665e-05,
      "loss": 1.2874,
      "num_input_tokens_seen": 2944736,
      "step": 5070
    },
    {
      "epoch": 0.7558832290735776,
      "grad_norm": 8.1875,
      "learning_rate": 1.8893357164134646e-05,
      "loss": 0.908,
      "num_input_tokens_seen": 2947552,
      "step": 5075
    },
    {
      "epoch": 0.7566279416145368,
      "grad_norm": 6.96875,
      "learning_rate": 1.8911974977658625e-05,
      "loss": 1.075,
      "num_input_tokens_seen": 2950592,
      "step": 5080
    },
    {
      "epoch": 0.7573726541554959,
      "grad_norm": 13.0,
      "learning_rate": 1.8930592791182606e-05,
      "loss": 1.2682,
      "num_input_tokens_seen": 2953408,
      "step": 5085
    },
    {
      "epoch": 0.7581173666964551,
      "grad_norm": 8.5,
      "learning_rate": 1.8949210604706584e-05,
      "loss": 0.7641,
      "num_input_tokens_seen": 2956064,
      "step": 5090
    },
    {
      "epoch": 0.7588620792374143,
      "grad_norm": 11.4375,
      "learning_rate": 1.8967828418230563e-05,
      "loss": 1.2206,
      "num_input_tokens_seen": 2959392,
      "step": 5095
    },
    {
      "epoch": 0.7596067917783735,
      "grad_norm": 13.4375,
      "learning_rate": 1.8986446231754544e-05,
      "loss": 1.2782,
      "num_input_tokens_seen": 2962240,
      "step": 5100
    },
    {
      "epoch": 0.7603515043193327,
      "grad_norm": 6.40625,
      "learning_rate": 1.9005064045278523e-05,
      "loss": 1.238,
      "num_input_tokens_seen": 2965152,
      "step": 5105
    },
    {
      "epoch": 0.7610962168602919,
      "grad_norm": 10.3125,
      "learning_rate": 1.9023681858802504e-05,
      "loss": 1.1743,
      "num_input_tokens_seen": 2968064,
      "step": 5110
    },
    {
      "epoch": 0.7618409294012511,
      "grad_norm": 12.6875,
      "learning_rate": 1.9042299672326482e-05,
      "loss": 1.2896,
      "num_input_tokens_seen": 2970752,
      "step": 5115
    },
    {
      "epoch": 0.7625856419422103,
      "grad_norm": 16.625,
      "learning_rate": 1.9060917485850464e-05,
      "loss": 0.998,
      "num_input_tokens_seen": 2973664,
      "step": 5120
    },
    {
      "epoch": 0.7633303544831695,
      "grad_norm": 7.6875,
      "learning_rate": 1.9079535299374442e-05,
      "loss": 1.0979,
      "num_input_tokens_seen": 2976704,
      "step": 5125
    },
    {
      "epoch": 0.7640750670241286,
      "grad_norm": 13.1875,
      "learning_rate": 1.9098153112898424e-05,
      "loss": 1.0637,
      "num_input_tokens_seen": 2979680,
      "step": 5130
    },
    {
      "epoch": 0.7648197795650878,
      "grad_norm": 8.4375,
      "learning_rate": 1.9116770926422402e-05,
      "loss": 0.9462,
      "num_input_tokens_seen": 2982464,
      "step": 5135
    },
    {
      "epoch": 0.765564492106047,
      "grad_norm": 13.6875,
      "learning_rate": 1.913538873994638e-05,
      "loss": 0.735,
      "num_input_tokens_seen": 2985216,
      "step": 5140
    },
    {
      "epoch": 0.7663092046470062,
      "grad_norm": 17.375,
      "learning_rate": 1.9154006553470362e-05,
      "loss": 1.3453,
      "num_input_tokens_seen": 2988000,
      "step": 5145
    },
    {
      "epoch": 0.7670539171879655,
      "grad_norm": 16.125,
      "learning_rate": 1.917262436699434e-05,
      "loss": 1.0689,
      "num_input_tokens_seen": 2990752,
      "step": 5150
    },
    {
      "epoch": 0.7677986297289247,
      "grad_norm": 9.375,
      "learning_rate": 1.9191242180518322e-05,
      "loss": 1.087,
      "num_input_tokens_seen": 2993664,
      "step": 5155
    },
    {
      "epoch": 0.7685433422698839,
      "grad_norm": 14.125,
      "learning_rate": 1.92098599940423e-05,
      "loss": 1.1609,
      "num_input_tokens_seen": 2996480,
      "step": 5160
    },
    {
      "epoch": 0.7692880548108431,
      "grad_norm": 9.75,
      "learning_rate": 1.922847780756628e-05,
      "loss": 1.0131,
      "num_input_tokens_seen": 2999520,
      "step": 5165
    },
    {
      "epoch": 0.7700327673518021,
      "grad_norm": 15.75,
      "learning_rate": 1.924709562109026e-05,
      "loss": 1.0264,
      "num_input_tokens_seen": 3002496,
      "step": 5170
    },
    {
      "epoch": 0.7707774798927614,
      "grad_norm": 10.25,
      "learning_rate": 1.926571343461424e-05,
      "loss": 0.893,
      "num_input_tokens_seen": 3005312,
      "step": 5175
    },
    {
      "epoch": 0.7715221924337206,
      "grad_norm": 23.75,
      "learning_rate": 1.928433124813822e-05,
      "loss": 1.3159,
      "num_input_tokens_seen": 3008128,
      "step": 5180
    },
    {
      "epoch": 0.7722669049746798,
      "grad_norm": 15.1875,
      "learning_rate": 1.9302949061662198e-05,
      "loss": 1.0939,
      "num_input_tokens_seen": 3011072,
      "step": 5185
    },
    {
      "epoch": 0.773011617515639,
      "grad_norm": 12.625,
      "learning_rate": 1.932156687518618e-05,
      "loss": 1.262,
      "num_input_tokens_seen": 3014272,
      "step": 5190
    },
    {
      "epoch": 0.7737563300565982,
      "grad_norm": 12.1875,
      "learning_rate": 1.9340184688710158e-05,
      "loss": 1.1392,
      "num_input_tokens_seen": 3017248,
      "step": 5195
    },
    {
      "epoch": 0.7745010425975574,
      "grad_norm": 19.875,
      "learning_rate": 1.935880250223414e-05,
      "loss": 1.1778,
      "num_input_tokens_seen": 3020160,
      "step": 5200
    },
    {
      "epoch": 0.7752457551385166,
      "grad_norm": 15.9375,
      "learning_rate": 1.9377420315758118e-05,
      "loss": 1.2589,
      "num_input_tokens_seen": 3023200,
      "step": 5205
    },
    {
      "epoch": 0.7759904676794758,
      "grad_norm": 11.0625,
      "learning_rate": 1.93960381292821e-05,
      "loss": 0.993,
      "num_input_tokens_seen": 3026240,
      "step": 5210
    },
    {
      "epoch": 0.7767351802204349,
      "grad_norm": 8.3125,
      "learning_rate": 1.9414655942806077e-05,
      "loss": 1.2303,
      "num_input_tokens_seen": 3028928,
      "step": 5215
    },
    {
      "epoch": 0.7774798927613941,
      "grad_norm": 9.6875,
      "learning_rate": 1.943327375633006e-05,
      "loss": 1.2175,
      "num_input_tokens_seen": 3031808,
      "step": 5220
    },
    {
      "epoch": 0.7782246053023533,
      "grad_norm": 12.125,
      "learning_rate": 1.9451891569854037e-05,
      "loss": 1.4367,
      "num_input_tokens_seen": 3034656,
      "step": 5225
    },
    {
      "epoch": 0.7789693178433125,
      "grad_norm": 12.5625,
      "learning_rate": 1.9470509383378016e-05,
      "loss": 1.2995,
      "num_input_tokens_seen": 3037760,
      "step": 5230
    },
    {
      "epoch": 0.7797140303842717,
      "grad_norm": 23.125,
      "learning_rate": 1.9489127196901997e-05,
      "loss": 1.3506,
      "num_input_tokens_seen": 3040672,
      "step": 5235
    },
    {
      "epoch": 0.7804587429252309,
      "grad_norm": 9.625,
      "learning_rate": 1.9507745010425975e-05,
      "loss": 0.9964,
      "num_input_tokens_seen": 3043552,
      "step": 5240
    },
    {
      "epoch": 0.7812034554661901,
      "grad_norm": 13.875,
      "learning_rate": 1.9526362823949957e-05,
      "loss": 1.2249,
      "num_input_tokens_seen": 3046944,
      "step": 5245
    },
    {
      "epoch": 0.7819481680071493,
      "grad_norm": 16.375,
      "learning_rate": 1.9544980637473935e-05,
      "loss": 1.0096,
      "num_input_tokens_seen": 3049952,
      "step": 5250
    },
    {
      "epoch": 0.7826928805481085,
      "grad_norm": 8.5625,
      "learning_rate": 1.9563598450997917e-05,
      "loss": 1.1432,
      "num_input_tokens_seen": 3052576,
      "step": 5255
    },
    {
      "epoch": 0.7834375930890676,
      "grad_norm": 14.75,
      "learning_rate": 1.9582216264521895e-05,
      "loss": 0.9314,
      "num_input_tokens_seen": 3055520,
      "step": 5260
    },
    {
      "epoch": 0.7841823056300268,
      "grad_norm": 11.625,
      "learning_rate": 1.9600834078045877e-05,
      "loss": 1.1619,
      "num_input_tokens_seen": 3058112,
      "step": 5265
    },
    {
      "epoch": 0.784927018170986,
      "grad_norm": 15.3125,
      "learning_rate": 1.9619451891569855e-05,
      "loss": 1.3148,
      "num_input_tokens_seen": 3061024,
      "step": 5270
    },
    {
      "epoch": 0.7856717307119452,
      "grad_norm": 16.875,
      "learning_rate": 1.9638069705093833e-05,
      "loss": 1.199,
      "num_input_tokens_seen": 3064160,
      "step": 5275
    },
    {
      "epoch": 0.7864164432529044,
      "grad_norm": 18.75,
      "learning_rate": 1.9656687518617815e-05,
      "loss": 0.865,
      "num_input_tokens_seen": 3066912,
      "step": 5280
    },
    {
      "epoch": 0.7871611557938636,
      "grad_norm": 5.59375,
      "learning_rate": 1.9675305332141793e-05,
      "loss": 1.0499,
      "num_input_tokens_seen": 3069856,
      "step": 5285
    },
    {
      "epoch": 0.7879058683348228,
      "grad_norm": 12.1875,
      "learning_rate": 1.9693923145665775e-05,
      "loss": 0.7818,
      "num_input_tokens_seen": 3072448,
      "step": 5290
    },
    {
      "epoch": 0.788650580875782,
      "grad_norm": 20.75,
      "learning_rate": 1.9712540959189753e-05,
      "loss": 1.2611,
      "num_input_tokens_seen": 3075392,
      "step": 5295
    },
    {
      "epoch": 0.7893952934167411,
      "grad_norm": 16.5,
      "learning_rate": 1.9731158772713735e-05,
      "loss": 1.1956,
      "num_input_tokens_seen": 3078176,
      "step": 5300
    },
    {
      "epoch": 0.7901400059577003,
      "grad_norm": 11.4375,
      "learning_rate": 1.9749776586237713e-05,
      "loss": 1.126,
      "num_input_tokens_seen": 3081024,
      "step": 5305
    },
    {
      "epoch": 0.7908847184986595,
      "grad_norm": 10.9375,
      "learning_rate": 1.976839439976169e-05,
      "loss": 0.9746,
      "num_input_tokens_seen": 3083840,
      "step": 5310
    },
    {
      "epoch": 0.7916294310396187,
      "grad_norm": 15.4375,
      "learning_rate": 1.9787012213285673e-05,
      "loss": 1.4692,
      "num_input_tokens_seen": 3086560,
      "step": 5315
    },
    {
      "epoch": 0.7923741435805779,
      "grad_norm": 20.25,
      "learning_rate": 1.980563002680965e-05,
      "loss": 1.3984,
      "num_input_tokens_seen": 3089440,
      "step": 5320
    },
    {
      "epoch": 0.7931188561215371,
      "grad_norm": 10.8125,
      "learning_rate": 1.9824247840333632e-05,
      "loss": 1.2305,
      "num_input_tokens_seen": 3092192,
      "step": 5325
    },
    {
      "epoch": 0.7938635686624963,
      "grad_norm": 11.1875,
      "learning_rate": 1.984286565385761e-05,
      "loss": 0.8675,
      "num_input_tokens_seen": 3095264,
      "step": 5330
    },
    {
      "epoch": 0.7946082812034555,
      "grad_norm": 12.875,
      "learning_rate": 1.9861483467381592e-05,
      "loss": 1.0203,
      "num_input_tokens_seen": 3098272,
      "step": 5335
    },
    {
      "epoch": 0.7953529937444147,
      "grad_norm": 7.90625,
      "learning_rate": 1.988010128090557e-05,
      "loss": 1.2597,
      "num_input_tokens_seen": 3101216,
      "step": 5340
    },
    {
      "epoch": 0.7960977062853738,
      "grad_norm": 14.625,
      "learning_rate": 1.9898719094429552e-05,
      "loss": 1.0859,
      "num_input_tokens_seen": 3104032,
      "step": 5345
    },
    {
      "epoch": 0.796842418826333,
      "grad_norm": 15.1875,
      "learning_rate": 1.9917336907953534e-05,
      "loss": 1.3082,
      "num_input_tokens_seen": 3106784,
      "step": 5350
    },
    {
      "epoch": 0.7975871313672922,
      "grad_norm": 9.4375,
      "learning_rate": 1.993595472147751e-05,
      "loss": 1.3438,
      "num_input_tokens_seen": 3109440,
      "step": 5355
    },
    {
      "epoch": 0.7983318439082514,
      "grad_norm": 9.0,
      "learning_rate": 1.995457253500149e-05,
      "loss": 0.941,
      "num_input_tokens_seen": 3112448,
      "step": 5360
    },
    {
      "epoch": 0.7990765564492106,
      "grad_norm": 18.75,
      "learning_rate": 1.997319034852547e-05,
      "loss": 1.0593,
      "num_input_tokens_seen": 3115264,
      "step": 5365
    },
    {
      "epoch": 0.7998212689901698,
      "grad_norm": 10.5625,
      "learning_rate": 1.999180816204945e-05,
      "loss": 1.0925,
      "num_input_tokens_seen": 3118016,
      "step": 5370
    },
    {
      "epoch": 0.800565981531129,
      "grad_norm": 9.0,
      "learning_rate": 2.001042597557343e-05,
      "loss": 0.9301,
      "num_input_tokens_seen": 3121056,
      "step": 5375
    },
    {
      "epoch": 0.8013106940720882,
      "grad_norm": 13.1875,
      "learning_rate": 2.002904378909741e-05,
      "loss": 0.977,
      "num_input_tokens_seen": 3123936,
      "step": 5380
    },
    {
      "epoch": 0.8020554066130474,
      "grad_norm": 9.0,
      "learning_rate": 2.0047661602621388e-05,
      "loss": 1.3659,
      "num_input_tokens_seen": 3126816,
      "step": 5385
    },
    {
      "epoch": 0.8028001191540065,
      "grad_norm": 5.46875,
      "learning_rate": 2.006627941614537e-05,
      "loss": 0.85,
      "num_input_tokens_seen": 3129792,
      "step": 5390
    },
    {
      "epoch": 0.8035448316949657,
      "grad_norm": 5.84375,
      "learning_rate": 2.008489722966935e-05,
      "loss": 1.3051,
      "num_input_tokens_seen": 3132832,
      "step": 5395
    },
    {
      "epoch": 0.8042895442359249,
      "grad_norm": 12.9375,
      "learning_rate": 2.0103515043193326e-05,
      "loss": 0.8619,
      "num_input_tokens_seen": 3135968,
      "step": 5400
    },
    {
      "epoch": 0.8050342567768841,
      "grad_norm": 12.1875,
      "learning_rate": 2.0122132856717308e-05,
      "loss": 1.102,
      "num_input_tokens_seen": 3138752,
      "step": 5405
    },
    {
      "epoch": 0.8057789693178433,
      "grad_norm": 10.75,
      "learning_rate": 2.0140750670241286e-05,
      "loss": 0.8481,
      "num_input_tokens_seen": 3141696,
      "step": 5410
    },
    {
      "epoch": 0.8065236818588025,
      "grad_norm": 7.75,
      "learning_rate": 2.0159368483765268e-05,
      "loss": 1.0635,
      "num_input_tokens_seen": 3144352,
      "step": 5415
    },
    {
      "epoch": 0.8072683943997617,
      "grad_norm": 17.0,
      "learning_rate": 2.0177986297289246e-05,
      "loss": 1.4272,
      "num_input_tokens_seen": 3147040,
      "step": 5420
    },
    {
      "epoch": 0.8080131069407209,
      "grad_norm": 21.75,
      "learning_rate": 2.0196604110813228e-05,
      "loss": 1.7409,
      "num_input_tokens_seen": 3150080,
      "step": 5425
    },
    {
      "epoch": 0.80875781948168,
      "grad_norm": 8.0625,
      "learning_rate": 2.021522192433721e-05,
      "loss": 0.7821,
      "num_input_tokens_seen": 3152992,
      "step": 5430
    },
    {
      "epoch": 0.8095025320226392,
      "grad_norm": 6.75,
      "learning_rate": 2.0233839737861187e-05,
      "loss": 1.345,
      "num_input_tokens_seen": 3155936,
      "step": 5435
    },
    {
      "epoch": 0.8102472445635984,
      "grad_norm": 10.0,
      "learning_rate": 2.0252457551385166e-05,
      "loss": 0.9609,
      "num_input_tokens_seen": 3158912,
      "step": 5440
    },
    {
      "epoch": 0.8109919571045576,
      "grad_norm": 6.75,
      "learning_rate": 2.0271075364909144e-05,
      "loss": 1.0215,
      "num_input_tokens_seen": 3161920,
      "step": 5445
    },
    {
      "epoch": 0.8117366696455168,
      "grad_norm": 9.375,
      "learning_rate": 2.0289693178433126e-05,
      "loss": 1.0472,
      "num_input_tokens_seen": 3165184,
      "step": 5450
    },
    {
      "epoch": 0.812481382186476,
      "grad_norm": 9.9375,
      "learning_rate": 2.0308310991957104e-05,
      "loss": 1.3466,
      "num_input_tokens_seen": 3168512,
      "step": 5455
    },
    {
      "epoch": 0.8132260947274352,
      "grad_norm": 15.375,
      "learning_rate": 2.0326928805481085e-05,
      "loss": 1.0532,
      "num_input_tokens_seen": 3171168,
      "step": 5460
    },
    {
      "epoch": 0.8139708072683944,
      "grad_norm": 12.125,
      "learning_rate": 2.0345546619005064e-05,
      "loss": 1.2387,
      "num_input_tokens_seen": 3173888,
      "step": 5465
    },
    {
      "epoch": 0.8147155198093536,
      "grad_norm": 13.3125,
      "learning_rate": 2.0364164432529045e-05,
      "loss": 1.3619,
      "num_input_tokens_seen": 3176800,
      "step": 5470
    },
    {
      "epoch": 0.8154602323503127,
      "grad_norm": 10.1875,
      "learning_rate": 2.0382782246053027e-05,
      "loss": 1.2755,
      "num_input_tokens_seen": 3179552,
      "step": 5475
    },
    {
      "epoch": 0.8162049448912719,
      "grad_norm": 7.21875,
      "learning_rate": 2.0401400059577005e-05,
      "loss": 0.9489,
      "num_input_tokens_seen": 3182400,
      "step": 5480
    },
    {
      "epoch": 0.8169496574322311,
      "grad_norm": 10.8125,
      "learning_rate": 2.0420017873100983e-05,
      "loss": 1.0418,
      "num_input_tokens_seen": 3185280,
      "step": 5485
    },
    {
      "epoch": 0.8176943699731903,
      "grad_norm": 9.9375,
      "learning_rate": 2.043863568662496e-05,
      "loss": 0.7888,
      "num_input_tokens_seen": 3188128,
      "step": 5490
    },
    {
      "epoch": 0.8184390825141495,
      "grad_norm": 10.0625,
      "learning_rate": 2.0457253500148943e-05,
      "loss": 1.1943,
      "num_input_tokens_seen": 3191072,
      "step": 5495
    },
    {
      "epoch": 0.8191837950551087,
      "grad_norm": 16.375,
      "learning_rate": 2.047587131367292e-05,
      "loss": 1.2848,
      "num_input_tokens_seen": 3194272,
      "step": 5500
    },
    {
      "epoch": 0.819928507596068,
      "grad_norm": 16.75,
      "learning_rate": 2.0494489127196903e-05,
      "loss": 1.2493,
      "num_input_tokens_seen": 3196992,
      "step": 5505
    },
    {
      "epoch": 0.8206732201370271,
      "grad_norm": 11.875,
      "learning_rate": 2.0513106940720885e-05,
      "loss": 0.8635,
      "num_input_tokens_seen": 3200096,
      "step": 5510
    },
    {
      "epoch": 0.8214179326779864,
      "grad_norm": 20.75,
      "learning_rate": 2.0531724754244863e-05,
      "loss": 1.1656,
      "num_input_tokens_seen": 3203104,
      "step": 5515
    },
    {
      "epoch": 0.8221626452189454,
      "grad_norm": 24.375,
      "learning_rate": 2.0550342567768845e-05,
      "loss": 1.2716,
      "num_input_tokens_seen": 3205728,
      "step": 5520
    },
    {
      "epoch": 0.8229073577599046,
      "grad_norm": 23.875,
      "learning_rate": 2.0568960381292823e-05,
      "loss": 1.3269,
      "num_input_tokens_seen": 3208320,
      "step": 5525
    },
    {
      "epoch": 0.8236520703008638,
      "grad_norm": 18.625,
      "learning_rate": 2.05875781948168e-05,
      "loss": 1.5944,
      "num_input_tokens_seen": 3211072,
      "step": 5530
    },
    {
      "epoch": 0.824396782841823,
      "grad_norm": 11.5,
      "learning_rate": 2.060619600834078e-05,
      "loss": 1.1731,
      "num_input_tokens_seen": 3214144,
      "step": 5535
    },
    {
      "epoch": 0.8251414953827823,
      "grad_norm": 9.875,
      "learning_rate": 2.062481382186476e-05,
      "loss": 1.2749,
      "num_input_tokens_seen": 3216992,
      "step": 5540
    },
    {
      "epoch": 0.8258862079237415,
      "grad_norm": 13.0625,
      "learning_rate": 2.0643431635388742e-05,
      "loss": 1.3706,
      "num_input_tokens_seen": 3220064,
      "step": 5545
    },
    {
      "epoch": 0.8266309204647007,
      "grad_norm": 10.8125,
      "learning_rate": 2.066204944891272e-05,
      "loss": 0.9366,
      "num_input_tokens_seen": 3222880,
      "step": 5550
    },
    {
      "epoch": 0.8273756330056599,
      "grad_norm": 12.9375,
      "learning_rate": 2.0680667262436702e-05,
      "loss": 0.9876,
      "num_input_tokens_seen": 3225952,
      "step": 5555
    },
    {
      "epoch": 0.828120345546619,
      "grad_norm": 14.3125,
      "learning_rate": 2.069928507596068e-05,
      "loss": 1.1288,
      "num_input_tokens_seen": 3228672,
      "step": 5560
    },
    {
      "epoch": 0.8288650580875782,
      "grad_norm": 8.0,
      "learning_rate": 2.0717902889484662e-05,
      "loss": 1.132,
      "num_input_tokens_seen": 3231392,
      "step": 5565
    },
    {
      "epoch": 0.8296097706285374,
      "grad_norm": 9.6875,
      "learning_rate": 2.0736520703008637e-05,
      "loss": 1.3763,
      "num_input_tokens_seen": 3234272,
      "step": 5570
    },
    {
      "epoch": 0.8303544831694966,
      "grad_norm": 8.0625,
      "learning_rate": 2.075513851653262e-05,
      "loss": 1.0994,
      "num_input_tokens_seen": 3237056,
      "step": 5575
    },
    {
      "epoch": 0.8310991957104558,
      "grad_norm": 16.125,
      "learning_rate": 2.0773756330056597e-05,
      "loss": 1.1813,
      "num_input_tokens_seen": 3240064,
      "step": 5580
    },
    {
      "epoch": 0.831843908251415,
      "grad_norm": 7.4375,
      "learning_rate": 2.079237414358058e-05,
      "loss": 1.146,
      "num_input_tokens_seen": 3243072,
      "step": 5585
    },
    {
      "epoch": 0.8325886207923742,
      "grad_norm": 14.375,
      "learning_rate": 2.081099195710456e-05,
      "loss": 1.1313,
      "num_input_tokens_seen": 3245920,
      "step": 5590
    },
    {
      "epoch": 0.8333333333333334,
      "grad_norm": 13.75,
      "learning_rate": 2.082960977062854e-05,
      "loss": 1.03,
      "num_input_tokens_seen": 3249280,
      "step": 5595
    },
    {
      "epoch": 0.8340780458742926,
      "grad_norm": 8.6875,
      "learning_rate": 2.084822758415252e-05,
      "loss": 1.1506,
      "num_input_tokens_seen": 3252320,
      "step": 5600
    },
    {
      "epoch": 0.8348227584152517,
      "grad_norm": 11.125,
      "learning_rate": 2.0866845397676498e-05,
      "loss": 1.4277,
      "num_input_tokens_seen": 3255456,
      "step": 5605
    },
    {
      "epoch": 0.8355674709562109,
      "grad_norm": 12.75,
      "learning_rate": 2.088546321120048e-05,
      "loss": 1.3355,
      "num_input_tokens_seen": 3258144,
      "step": 5610
    },
    {
      "epoch": 0.8363121834971701,
      "grad_norm": 23.5,
      "learning_rate": 2.0904081024724455e-05,
      "loss": 1.3513,
      "num_input_tokens_seen": 3260832,
      "step": 5615
    },
    {
      "epoch": 0.8370568960381293,
      "grad_norm": 9.375,
      "learning_rate": 2.0922698838248436e-05,
      "loss": 1.0738,
      "num_input_tokens_seen": 3263712,
      "step": 5620
    },
    {
      "epoch": 0.8378016085790885,
      "grad_norm": 7.75,
      "learning_rate": 2.0941316651772418e-05,
      "loss": 1.1091,
      "num_input_tokens_seen": 3266688,
      "step": 5625
    },
    {
      "epoch": 0.8385463211200477,
      "grad_norm": 9.75,
      "learning_rate": 2.0959934465296396e-05,
      "loss": 1.083,
      "num_input_tokens_seen": 3269504,
      "step": 5630
    },
    {
      "epoch": 0.8392910336610069,
      "grad_norm": 4.9375,
      "learning_rate": 2.0978552278820378e-05,
      "loss": 0.8238,
      "num_input_tokens_seen": 3272608,
      "step": 5635
    },
    {
      "epoch": 0.8400357462019661,
      "grad_norm": 11.8125,
      "learning_rate": 2.0997170092344356e-05,
      "loss": 1.1453,
      "num_input_tokens_seen": 3275584,
      "step": 5640
    },
    {
      "epoch": 0.8407804587429253,
      "grad_norm": 7.8125,
      "learning_rate": 2.1015787905868338e-05,
      "loss": 1.1333,
      "num_input_tokens_seen": 3278368,
      "step": 5645
    },
    {
      "epoch": 0.8415251712838844,
      "grad_norm": 17.375,
      "learning_rate": 2.1034405719392316e-05,
      "loss": 0.9227,
      "num_input_tokens_seen": 3281312,
      "step": 5650
    },
    {
      "epoch": 0.8422698838248436,
      "grad_norm": 12.1875,
      "learning_rate": 2.1053023532916297e-05,
      "loss": 0.8709,
      "num_input_tokens_seen": 3284128,
      "step": 5655
    },
    {
      "epoch": 0.8430145963658028,
      "grad_norm": 8.625,
      "learning_rate": 2.1071641346440272e-05,
      "loss": 1.0749,
      "num_input_tokens_seen": 3286816,
      "step": 5660
    },
    {
      "epoch": 0.843759308906762,
      "grad_norm": 11.4375,
      "learning_rate": 2.1090259159964254e-05,
      "loss": 1.1023,
      "num_input_tokens_seen": 3289824,
      "step": 5665
    },
    {
      "epoch": 0.8445040214477212,
      "grad_norm": 13.625,
      "learning_rate": 2.1108876973488236e-05,
      "loss": 0.8717,
      "num_input_tokens_seen": 3293152,
      "step": 5670
    },
    {
      "epoch": 0.8452487339886804,
      "grad_norm": 17.75,
      "learning_rate": 2.1127494787012214e-05,
      "loss": 1.2019,
      "num_input_tokens_seen": 3295968,
      "step": 5675
    },
    {
      "epoch": 0.8459934465296396,
      "grad_norm": 10.5,
      "learning_rate": 2.1146112600536195e-05,
      "loss": 1.0536,
      "num_input_tokens_seen": 3298528,
      "step": 5680
    },
    {
      "epoch": 0.8467381590705988,
      "grad_norm": 10.5625,
      "learning_rate": 2.1164730414060174e-05,
      "loss": 0.9789,
      "num_input_tokens_seen": 3301696,
      "step": 5685
    },
    {
      "epoch": 0.8474828716115579,
      "grad_norm": 9.9375,
      "learning_rate": 2.1183348227584155e-05,
      "loss": 0.9719,
      "num_input_tokens_seen": 3304672,
      "step": 5690
    },
    {
      "epoch": 0.8482275841525171,
      "grad_norm": 11.3125,
      "learning_rate": 2.1201966041108133e-05,
      "loss": 1.1002,
      "num_input_tokens_seen": 3307168,
      "step": 5695
    },
    {
      "epoch": 0.8489722966934763,
      "grad_norm": 18.125,
      "learning_rate": 2.1220583854632112e-05,
      "loss": 1.3135,
      "num_input_tokens_seen": 3309952,
      "step": 5700
    },
    {
      "epoch": 0.8497170092344355,
      "grad_norm": 19.5,
      "learning_rate": 2.1239201668156093e-05,
      "loss": 1.3587,
      "num_input_tokens_seen": 3312800,
      "step": 5705
    },
    {
      "epoch": 0.8504617217753947,
      "grad_norm": 12.6875,
      "learning_rate": 2.125781948168007e-05,
      "loss": 1.0663,
      "num_input_tokens_seen": 3315808,
      "step": 5710
    },
    {
      "epoch": 0.8512064343163539,
      "grad_norm": 11.25,
      "learning_rate": 2.1276437295204053e-05,
      "loss": 0.9332,
      "num_input_tokens_seen": 3318720,
      "step": 5715
    },
    {
      "epoch": 0.8519511468573131,
      "grad_norm": 7.28125,
      "learning_rate": 2.129505510872803e-05,
      "loss": 0.9878,
      "num_input_tokens_seen": 3321728,
      "step": 5720
    },
    {
      "epoch": 0.8526958593982723,
      "grad_norm": 11.5,
      "learning_rate": 2.1313672922252013e-05,
      "loss": 1.1802,
      "num_input_tokens_seen": 3324640,
      "step": 5725
    },
    {
      "epoch": 0.8534405719392315,
      "grad_norm": 7.375,
      "learning_rate": 2.133229073577599e-05,
      "loss": 1.1607,
      "num_input_tokens_seen": 3327456,
      "step": 5730
    },
    {
      "epoch": 0.8541852844801906,
      "grad_norm": 13.5,
      "learning_rate": 2.1350908549299973e-05,
      "loss": 1.1824,
      "num_input_tokens_seen": 3330560,
      "step": 5735
    },
    {
      "epoch": 0.8549299970211498,
      "grad_norm": 15.25,
      "learning_rate": 2.136952636282395e-05,
      "loss": 0.9259,
      "num_input_tokens_seen": 3333600,
      "step": 5740
    },
    {
      "epoch": 0.855674709562109,
      "grad_norm": 10.625,
      "learning_rate": 2.138814417634793e-05,
      "loss": 1.0042,
      "num_input_tokens_seen": 3336800,
      "step": 5745
    },
    {
      "epoch": 0.8564194221030682,
      "grad_norm": 12.25,
      "learning_rate": 2.140676198987191e-05,
      "loss": 0.9506,
      "num_input_tokens_seen": 3339552,
      "step": 5750
    },
    {
      "epoch": 0.8571641346440274,
      "grad_norm": 15.8125,
      "learning_rate": 2.142537980339589e-05,
      "loss": 0.9482,
      "num_input_tokens_seen": 3342624,
      "step": 5755
    },
    {
      "epoch": 0.8579088471849866,
      "grad_norm": 11.5625,
      "learning_rate": 2.144399761691987e-05,
      "loss": 0.9248,
      "num_input_tokens_seen": 3345984,
      "step": 5760
    },
    {
      "epoch": 0.8586535597259458,
      "grad_norm": 8.6875,
      "learning_rate": 2.146261543044385e-05,
      "loss": 1.0749,
      "num_input_tokens_seen": 3348864,
      "step": 5765
    },
    {
      "epoch": 0.859398272266905,
      "grad_norm": 17.5,
      "learning_rate": 2.148123324396783e-05,
      "loss": 0.9907,
      "num_input_tokens_seen": 3351744,
      "step": 5770
    },
    {
      "epoch": 0.8601429848078642,
      "grad_norm": 13.4375,
      "learning_rate": 2.149985105749181e-05,
      "loss": 1.2084,
      "num_input_tokens_seen": 3354528,
      "step": 5775
    },
    {
      "epoch": 0.8608876973488233,
      "grad_norm": 14.3125,
      "learning_rate": 2.151846887101579e-05,
      "loss": 1.4181,
      "num_input_tokens_seen": 3357376,
      "step": 5780
    },
    {
      "epoch": 0.8616324098897825,
      "grad_norm": 15.9375,
      "learning_rate": 2.153708668453977e-05,
      "loss": 1.4317,
      "num_input_tokens_seen": 3360224,
      "step": 5785
    },
    {
      "epoch": 0.8623771224307417,
      "grad_norm": 7.125,
      "learning_rate": 2.1555704498063747e-05,
      "loss": 0.9613,
      "num_input_tokens_seen": 3362816,
      "step": 5790
    },
    {
      "epoch": 0.8631218349717009,
      "grad_norm": 8.9375,
      "learning_rate": 2.157432231158773e-05,
      "loss": 1.054,
      "num_input_tokens_seen": 3365728,
      "step": 5795
    },
    {
      "epoch": 0.8638665475126601,
      "grad_norm": 7.875,
      "learning_rate": 2.1592940125111707e-05,
      "loss": 0.8017,
      "num_input_tokens_seen": 3368448,
      "step": 5800
    },
    {
      "epoch": 0.8646112600536193,
      "grad_norm": 6.75,
      "learning_rate": 2.161155793863569e-05,
      "loss": 1.299,
      "num_input_tokens_seen": 3371392,
      "step": 5805
    },
    {
      "epoch": 0.8653559725945785,
      "grad_norm": 16.25,
      "learning_rate": 2.1630175752159667e-05,
      "loss": 0.8656,
      "num_input_tokens_seen": 3373952,
      "step": 5810
    },
    {
      "epoch": 0.8661006851355377,
      "grad_norm": 11.3125,
      "learning_rate": 2.164879356568365e-05,
      "loss": 1.2235,
      "num_input_tokens_seen": 3376672,
      "step": 5815
    },
    {
      "epoch": 0.8668453976764968,
      "grad_norm": 18.0,
      "learning_rate": 2.1667411379207627e-05,
      "loss": 1.3896,
      "num_input_tokens_seen": 3379840,
      "step": 5820
    },
    {
      "epoch": 0.867590110217456,
      "grad_norm": 4.9375,
      "learning_rate": 2.1686029192731608e-05,
      "loss": 0.933,
      "num_input_tokens_seen": 3382592,
      "step": 5825
    },
    {
      "epoch": 0.8683348227584152,
      "grad_norm": 8.0625,
      "learning_rate": 2.1704647006255586e-05,
      "loss": 1.3356,
      "num_input_tokens_seen": 3385248,
      "step": 5830
    },
    {
      "epoch": 0.8690795352993744,
      "grad_norm": 6.90625,
      "learning_rate": 2.1723264819779565e-05,
      "loss": 1.0992,
      "num_input_tokens_seen": 3388288,
      "step": 5835
    },
    {
      "epoch": 0.8698242478403336,
      "grad_norm": 11.5,
      "learning_rate": 2.1741882633303546e-05,
      "loss": 0.9772,
      "num_input_tokens_seen": 3391008,
      "step": 5840
    },
    {
      "epoch": 0.8705689603812928,
      "grad_norm": 8.875,
      "learning_rate": 2.1760500446827525e-05,
      "loss": 0.9754,
      "num_input_tokens_seen": 3393920,
      "step": 5845
    },
    {
      "epoch": 0.871313672922252,
      "grad_norm": 13.125,
      "learning_rate": 2.1779118260351506e-05,
      "loss": 1.1087,
      "num_input_tokens_seen": 3396768,
      "step": 5850
    },
    {
      "epoch": 0.8720583854632112,
      "grad_norm": 13.375,
      "learning_rate": 2.1797736073875484e-05,
      "loss": 1.3484,
      "num_input_tokens_seen": 3399680,
      "step": 5855
    },
    {
      "epoch": 0.8728030980041704,
      "grad_norm": 12.6875,
      "learning_rate": 2.1816353887399466e-05,
      "loss": 0.961,
      "num_input_tokens_seen": 3402528,
      "step": 5860
    },
    {
      "epoch": 0.8735478105451295,
      "grad_norm": 6.59375,
      "learning_rate": 2.1834971700923444e-05,
      "loss": 1.0566,
      "num_input_tokens_seen": 3405248,
      "step": 5865
    },
    {
      "epoch": 0.8742925230860887,
      "grad_norm": 14.875,
      "learning_rate": 2.1853589514447426e-05,
      "loss": 1.1124,
      "num_input_tokens_seen": 3408448,
      "step": 5870
    },
    {
      "epoch": 0.8750372356270479,
      "grad_norm": 7.9375,
      "learning_rate": 2.1872207327971404e-05,
      "loss": 1.0581,
      "num_input_tokens_seen": 3411520,
      "step": 5875
    },
    {
      "epoch": 0.8757819481680071,
      "grad_norm": 7.0625,
      "learning_rate": 2.1890825141495382e-05,
      "loss": 0.9097,
      "num_input_tokens_seen": 3414528,
      "step": 5880
    },
    {
      "epoch": 0.8765266607089663,
      "grad_norm": 9.125,
      "learning_rate": 2.1909442955019364e-05,
      "loss": 1.392,
      "num_input_tokens_seen": 3417280,
      "step": 5885
    },
    {
      "epoch": 0.8772713732499255,
      "grad_norm": 17.375,
      "learning_rate": 2.1928060768543342e-05,
      "loss": 1.1364,
      "num_input_tokens_seen": 3420416,
      "step": 5890
    },
    {
      "epoch": 0.8780160857908847,
      "grad_norm": 14.75,
      "learning_rate": 2.1946678582067324e-05,
      "loss": 1.1579,
      "num_input_tokens_seen": 3423104,
      "step": 5895
    },
    {
      "epoch": 0.878760798331844,
      "grad_norm": 15.125,
      "learning_rate": 2.1965296395591302e-05,
      "loss": 0.9072,
      "num_input_tokens_seen": 3426080,
      "step": 5900
    },
    {
      "epoch": 0.8795055108728032,
      "grad_norm": 8.875,
      "learning_rate": 2.1983914209115284e-05,
      "loss": 0.9904,
      "num_input_tokens_seen": 3429152,
      "step": 5905
    },
    {
      "epoch": 0.8802502234137622,
      "grad_norm": 15.4375,
      "learning_rate": 2.2002532022639262e-05,
      "loss": 1.4401,
      "num_input_tokens_seen": 3431904,
      "step": 5910
    },
    {
      "epoch": 0.8809949359547214,
      "grad_norm": 10.8125,
      "learning_rate": 2.202114983616324e-05,
      "loss": 0.9855,
      "num_input_tokens_seen": 3434880,
      "step": 5915
    },
    {
      "epoch": 0.8817396484956807,
      "grad_norm": 8.5,
      "learning_rate": 2.2039767649687222e-05,
      "loss": 1.1022,
      "num_input_tokens_seen": 3437728,
      "step": 5920
    },
    {
      "epoch": 0.8824843610366399,
      "grad_norm": 15.1875,
      "learning_rate": 2.20583854632112e-05,
      "loss": 0.9831,
      "num_input_tokens_seen": 3440640,
      "step": 5925
    },
    {
      "epoch": 0.883229073577599,
      "grad_norm": 4.125,
      "learning_rate": 2.207700327673518e-05,
      "loss": 1.1346,
      "num_input_tokens_seen": 3443488,
      "step": 5930
    },
    {
      "epoch": 0.8839737861185583,
      "grad_norm": 12.1875,
      "learning_rate": 2.209562109025916e-05,
      "loss": 1.1445,
      "num_input_tokens_seen": 3446176,
      "step": 5935
    },
    {
      "epoch": 0.8847184986595175,
      "grad_norm": 13.5625,
      "learning_rate": 2.211423890378314e-05,
      "loss": 1.1661,
      "num_input_tokens_seen": 3448960,
      "step": 5940
    },
    {
      "epoch": 0.8854632112004767,
      "grad_norm": 5.8125,
      "learning_rate": 2.213285671730712e-05,
      "loss": 0.7783,
      "num_input_tokens_seen": 3451648,
      "step": 5945
    },
    {
      "epoch": 0.8862079237414358,
      "grad_norm": 19.75,
      "learning_rate": 2.21514745308311e-05,
      "loss": 1.72,
      "num_input_tokens_seen": 3454464,
      "step": 5950
    },
    {
      "epoch": 0.886952636282395,
      "grad_norm": 9.1875,
      "learning_rate": 2.2170092344355083e-05,
      "loss": 0.8226,
      "num_input_tokens_seen": 3457248,
      "step": 5955
    },
    {
      "epoch": 0.8876973488233542,
      "grad_norm": 10.75,
      "learning_rate": 2.2188710157879058e-05,
      "loss": 1.3589,
      "num_input_tokens_seen": 3459872,
      "step": 5960
    },
    {
      "epoch": 0.8884420613643134,
      "grad_norm": 13.1875,
      "learning_rate": 2.220732797140304e-05,
      "loss": 0.7945,
      "num_input_tokens_seen": 3462880,
      "step": 5965
    },
    {
      "epoch": 0.8891867739052726,
      "grad_norm": 8.3125,
      "learning_rate": 2.2225945784927018e-05,
      "loss": 1.0095,
      "num_input_tokens_seen": 3465792,
      "step": 5970
    },
    {
      "epoch": 0.8899314864462318,
      "grad_norm": 12.9375,
      "learning_rate": 2.2244563598451e-05,
      "loss": 1.4397,
      "num_input_tokens_seen": 3468608,
      "step": 5975
    },
    {
      "epoch": 0.890676198987191,
      "grad_norm": 20.375,
      "learning_rate": 2.2263181411974977e-05,
      "loss": 1.2702,
      "num_input_tokens_seen": 3471520,
      "step": 5980
    },
    {
      "epoch": 0.8914209115281502,
      "grad_norm": 14.25,
      "learning_rate": 2.228179922549896e-05,
      "loss": 0.9959,
      "num_input_tokens_seen": 3474432,
      "step": 5985
    },
    {
      "epoch": 0.8921656240691094,
      "grad_norm": 5.9375,
      "learning_rate": 2.2300417039022937e-05,
      "loss": 1.0258,
      "num_input_tokens_seen": 3477216,
      "step": 5990
    },
    {
      "epoch": 0.8929103366100685,
      "grad_norm": 7.1875,
      "learning_rate": 2.231903485254692e-05,
      "loss": 1.1307,
      "num_input_tokens_seen": 3480160,
      "step": 5995
    },
    {
      "epoch": 0.8936550491510277,
      "grad_norm": 18.375,
      "learning_rate": 2.23376526660709e-05,
      "loss": 1.2264,
      "num_input_tokens_seen": 3483040,
      "step": 6000
    },
    {
      "epoch": 0.8943997616919869,
      "grad_norm": 11.25,
      "learning_rate": 2.2356270479594875e-05,
      "loss": 1.1253,
      "num_input_tokens_seen": 3485824,
      "step": 6005
    },
    {
      "epoch": 0.8951444742329461,
      "grad_norm": 21.0,
      "learning_rate": 2.2374888293118857e-05,
      "loss": 1.1174,
      "num_input_tokens_seen": 3488544,
      "step": 6010
    },
    {
      "epoch": 0.8958891867739053,
      "grad_norm": 9.8125,
      "learning_rate": 2.2393506106642835e-05,
      "loss": 1.0614,
      "num_input_tokens_seen": 3491680,
      "step": 6015
    },
    {
      "epoch": 0.8966338993148645,
      "grad_norm": 11.5,
      "learning_rate": 2.2412123920166817e-05,
      "loss": 1.2677,
      "num_input_tokens_seen": 3494784,
      "step": 6020
    },
    {
      "epoch": 0.8973786118558237,
      "grad_norm": 11.25,
      "learning_rate": 2.2430741733690795e-05,
      "loss": 0.9868,
      "num_input_tokens_seen": 3497504,
      "step": 6025
    },
    {
      "epoch": 0.8981233243967829,
      "grad_norm": 11.6875,
      "learning_rate": 2.2449359547214777e-05,
      "loss": 1.168,
      "num_input_tokens_seen": 3500416,
      "step": 6030
    },
    {
      "epoch": 0.898868036937742,
      "grad_norm": 19.0,
      "learning_rate": 2.246797736073876e-05,
      "loss": 1.3475,
      "num_input_tokens_seen": 3503168,
      "step": 6035
    },
    {
      "epoch": 0.8996127494787012,
      "grad_norm": 12.5,
      "learning_rate": 2.2486595174262737e-05,
      "loss": 1.1507,
      "num_input_tokens_seen": 3506080,
      "step": 6040
    },
    {
      "epoch": 0.9003574620196604,
      "grad_norm": 6.375,
      "learning_rate": 2.2505212987786715e-05,
      "loss": 1.08,
      "num_input_tokens_seen": 3509216,
      "step": 6045
    },
    {
      "epoch": 0.9011021745606196,
      "grad_norm": 23.625,
      "learning_rate": 2.2523830801310693e-05,
      "loss": 1.7862,
      "num_input_tokens_seen": 3512064,
      "step": 6050
    },
    {
      "epoch": 0.9018468871015788,
      "grad_norm": 14.875,
      "learning_rate": 2.2542448614834675e-05,
      "loss": 1.2566,
      "num_input_tokens_seen": 3514784,
      "step": 6055
    },
    {
      "epoch": 0.902591599642538,
      "grad_norm": 5.8125,
      "learning_rate": 2.2561066428358653e-05,
      "loss": 0.8891,
      "num_input_tokens_seen": 3518048,
      "step": 6060
    },
    {
      "epoch": 0.9033363121834972,
      "grad_norm": 5.6875,
      "learning_rate": 2.2579684241882634e-05,
      "loss": 0.7652,
      "num_input_tokens_seen": 3521184,
      "step": 6065
    },
    {
      "epoch": 0.9040810247244564,
      "grad_norm": 9.0625,
      "learning_rate": 2.2598302055406613e-05,
      "loss": 1.1871,
      "num_input_tokens_seen": 3524224,
      "step": 6070
    },
    {
      "epoch": 0.9048257372654156,
      "grad_norm": 16.0,
      "learning_rate": 2.2616919868930594e-05,
      "loss": 1.0905,
      "num_input_tokens_seen": 3527104,
      "step": 6075
    },
    {
      "epoch": 0.9055704498063747,
      "grad_norm": 17.125,
      "learning_rate": 2.2635537682454576e-05,
      "loss": 1.1588,
      "num_input_tokens_seen": 3529824,
      "step": 6080
    },
    {
      "epoch": 0.9063151623473339,
      "grad_norm": 9.5625,
      "learning_rate": 2.2654155495978554e-05,
      "loss": 1.3454,
      "num_input_tokens_seen": 3532768,
      "step": 6085
    },
    {
      "epoch": 0.9070598748882931,
      "grad_norm": 14.4375,
      "learning_rate": 2.2672773309502532e-05,
      "loss": 1.3689,
      "num_input_tokens_seen": 3535552,
      "step": 6090
    },
    {
      "epoch": 0.9078045874292523,
      "grad_norm": 11.0,
      "learning_rate": 2.269139112302651e-05,
      "loss": 0.9496,
      "num_input_tokens_seen": 3538336,
      "step": 6095
    },
    {
      "epoch": 0.9085492999702115,
      "grad_norm": 16.125,
      "learning_rate": 2.2710008936550492e-05,
      "loss": 1.431,
      "num_input_tokens_seen": 3540928,
      "step": 6100
    },
    {
      "epoch": 0.9092940125111707,
      "grad_norm": 18.125,
      "learning_rate": 2.272862675007447e-05,
      "loss": 1.0781,
      "num_input_tokens_seen": 3544000,
      "step": 6105
    },
    {
      "epoch": 0.9100387250521299,
      "grad_norm": 8.25,
      "learning_rate": 2.2747244563598452e-05,
      "loss": 1.1214,
      "num_input_tokens_seen": 3546784,
      "step": 6110
    },
    {
      "epoch": 0.9107834375930891,
      "grad_norm": 11.5625,
      "learning_rate": 2.2765862377122434e-05,
      "loss": 1.1106,
      "num_input_tokens_seen": 3549696,
      "step": 6115
    },
    {
      "epoch": 0.9115281501340483,
      "grad_norm": 6.4375,
      "learning_rate": 2.2784480190646412e-05,
      "loss": 0.91,
      "num_input_tokens_seen": 3552480,
      "step": 6120
    },
    {
      "epoch": 0.9122728626750074,
      "grad_norm": 13.4375,
      "learning_rate": 2.2803098004170394e-05,
      "loss": 0.9494,
      "num_input_tokens_seen": 3555360,
      "step": 6125
    },
    {
      "epoch": 0.9130175752159666,
      "grad_norm": 11.75,
      "learning_rate": 2.2821715817694372e-05,
      "loss": 0.8835,
      "num_input_tokens_seen": 3558528,
      "step": 6130
    },
    {
      "epoch": 0.9137622877569258,
      "grad_norm": 17.25,
      "learning_rate": 2.284033363121835e-05,
      "loss": 1.0724,
      "num_input_tokens_seen": 3561280,
      "step": 6135
    },
    {
      "epoch": 0.914507000297885,
      "grad_norm": 9.1875,
      "learning_rate": 2.2858951444742328e-05,
      "loss": 0.9177,
      "num_input_tokens_seen": 3564544,
      "step": 6140
    },
    {
      "epoch": 0.9152517128388442,
      "grad_norm": 14.5625,
      "learning_rate": 2.287756925826631e-05,
      "loss": 1.5115,
      "num_input_tokens_seen": 3567104,
      "step": 6145
    },
    {
      "epoch": 0.9159964253798034,
      "grad_norm": 13.625,
      "learning_rate": 2.289618707179029e-05,
      "loss": 1.2348,
      "num_input_tokens_seen": 3570080,
      "step": 6150
    },
    {
      "epoch": 0.9167411379207626,
      "grad_norm": 13.125,
      "learning_rate": 2.291480488531427e-05,
      "loss": 1.0252,
      "num_input_tokens_seen": 3573056,
      "step": 6155
    },
    {
      "epoch": 0.9174858504617218,
      "grad_norm": 8.0625,
      "learning_rate": 2.293342269883825e-05,
      "loss": 0.9346,
      "num_input_tokens_seen": 3576064,
      "step": 6160
    },
    {
      "epoch": 0.9182305630026809,
      "grad_norm": 11.0625,
      "learning_rate": 2.295204051236223e-05,
      "loss": 0.9287,
      "num_input_tokens_seen": 3579040,
      "step": 6165
    },
    {
      "epoch": 0.9189752755436401,
      "grad_norm": 12.75,
      "learning_rate": 2.297065832588621e-05,
      "loss": 0.9731,
      "num_input_tokens_seen": 3581984,
      "step": 6170
    },
    {
      "epoch": 0.9197199880845993,
      "grad_norm": 21.625,
      "learning_rate": 2.2989276139410186e-05,
      "loss": 1.5972,
      "num_input_tokens_seen": 3584960,
      "step": 6175
    },
    {
      "epoch": 0.9204647006255585,
      "grad_norm": 17.0,
      "learning_rate": 2.3007893952934168e-05,
      "loss": 0.9784,
      "num_input_tokens_seen": 3587808,
      "step": 6180
    },
    {
      "epoch": 0.9212094131665177,
      "grad_norm": 10.625,
      "learning_rate": 2.3026511766458146e-05,
      "loss": 1.0413,
      "num_input_tokens_seen": 3590656,
      "step": 6185
    },
    {
      "epoch": 0.9219541257074769,
      "grad_norm": 13.875,
      "learning_rate": 2.3045129579982128e-05,
      "loss": 1.3169,
      "num_input_tokens_seen": 3593248,
      "step": 6190
    },
    {
      "epoch": 0.9226988382484361,
      "grad_norm": 6.71875,
      "learning_rate": 2.306374739350611e-05,
      "loss": 0.8518,
      "num_input_tokens_seen": 3596192,
      "step": 6195
    },
    {
      "epoch": 0.9234435507893953,
      "grad_norm": 9.125,
      "learning_rate": 2.3082365207030087e-05,
      "loss": 1.3013,
      "num_input_tokens_seen": 3599264,
      "step": 6200
    },
    {
      "epoch": 0.9241882633303545,
      "grad_norm": 11.75,
      "learning_rate": 2.310098302055407e-05,
      "loss": 0.8337,
      "num_input_tokens_seen": 3602304,
      "step": 6205
    },
    {
      "epoch": 0.9249329758713136,
      "grad_norm": 11.125,
      "learning_rate": 2.3119600834078047e-05,
      "loss": 1.1089,
      "num_input_tokens_seen": 3605056,
      "step": 6210
    },
    {
      "epoch": 0.9256776884122728,
      "grad_norm": 16.125,
      "learning_rate": 2.313821864760203e-05,
      "loss": 0.8837,
      "num_input_tokens_seen": 3607904,
      "step": 6215
    },
    {
      "epoch": 0.926422400953232,
      "grad_norm": 12.6875,
      "learning_rate": 2.3156836461126004e-05,
      "loss": 1.1434,
      "num_input_tokens_seen": 3611008,
      "step": 6220
    },
    {
      "epoch": 0.9271671134941912,
      "grad_norm": 24.875,
      "learning_rate": 2.3175454274649985e-05,
      "loss": 1.6011,
      "num_input_tokens_seen": 3613920,
      "step": 6225
    },
    {
      "epoch": 0.9279118260351504,
      "grad_norm": 10.6875,
      "learning_rate": 2.3194072088173967e-05,
      "loss": 0.9226,
      "num_input_tokens_seen": 3616608,
      "step": 6230
    },
    {
      "epoch": 0.9286565385761096,
      "grad_norm": 15.0625,
      "learning_rate": 2.3212689901697945e-05,
      "loss": 1.0785,
      "num_input_tokens_seen": 3619776,
      "step": 6235
    },
    {
      "epoch": 0.9294012511170688,
      "grad_norm": 15.1875,
      "learning_rate": 2.3231307715221927e-05,
      "loss": 0.9458,
      "num_input_tokens_seen": 3622624,
      "step": 6240
    },
    {
      "epoch": 0.930145963658028,
      "grad_norm": 20.25,
      "learning_rate": 2.3249925528745905e-05,
      "loss": 0.9619,
      "num_input_tokens_seen": 3625536,
      "step": 6245
    },
    {
      "epoch": 0.9308906761989872,
      "grad_norm": 17.0,
      "learning_rate": 2.3268543342269887e-05,
      "loss": 1.0369,
      "num_input_tokens_seen": 3628288,
      "step": 6250
    },
    {
      "epoch": 0.9316353887399463,
      "grad_norm": 9.5625,
      "learning_rate": 2.3287161155793865e-05,
      "loss": 1.2866,
      "num_input_tokens_seen": 3631200,
      "step": 6255
    },
    {
      "epoch": 0.9323801012809055,
      "grad_norm": 7.875,
      "learning_rate": 2.3305778969317847e-05,
      "loss": 1.591,
      "num_input_tokens_seen": 3634144,
      "step": 6260
    },
    {
      "epoch": 0.9331248138218647,
      "grad_norm": 15.375,
      "learning_rate": 2.332439678284182e-05,
      "loss": 1.1295,
      "num_input_tokens_seen": 3637024,
      "step": 6265
    },
    {
      "epoch": 0.9338695263628239,
      "grad_norm": 12.125,
      "learning_rate": 2.3343014596365803e-05,
      "loss": 1.2598,
      "num_input_tokens_seen": 3640000,
      "step": 6270
    },
    {
      "epoch": 0.9346142389037831,
      "grad_norm": 13.3125,
      "learning_rate": 2.3361632409889785e-05,
      "loss": 1.2484,
      "num_input_tokens_seen": 3643168,
      "step": 6275
    },
    {
      "epoch": 0.9353589514447423,
      "grad_norm": 6.4375,
      "learning_rate": 2.3380250223413763e-05,
      "loss": 0.8942,
      "num_input_tokens_seen": 3645856,
      "step": 6280
    },
    {
      "epoch": 0.9361036639857016,
      "grad_norm": 16.0,
      "learning_rate": 2.3398868036937744e-05,
      "loss": 1.3582,
      "num_input_tokens_seen": 3648480,
      "step": 6285
    },
    {
      "epoch": 0.9368483765266608,
      "grad_norm": 7.9375,
      "learning_rate": 2.3417485850461723e-05,
      "loss": 1.5821,
      "num_input_tokens_seen": 3651168,
      "step": 6290
    },
    {
      "epoch": 0.9375930890676198,
      "grad_norm": 17.0,
      "learning_rate": 2.3436103663985704e-05,
      "loss": 0.9834,
      "num_input_tokens_seen": 3654144,
      "step": 6295
    },
    {
      "epoch": 0.938337801608579,
      "grad_norm": 15.5,
      "learning_rate": 2.3454721477509683e-05,
      "loss": 1.3373,
      "num_input_tokens_seen": 3657056,
      "step": 6300
    },
    {
      "epoch": 0.9390825141495382,
      "grad_norm": 9.5,
      "learning_rate": 2.347333929103366e-05,
      "loss": 1.1536,
      "num_input_tokens_seen": 3660192,
      "step": 6305
    },
    {
      "epoch": 0.9398272266904975,
      "grad_norm": 10.5,
      "learning_rate": 2.3491957104557642e-05,
      "loss": 1.0936,
      "num_input_tokens_seen": 3663008,
      "step": 6310
    },
    {
      "epoch": 0.9405719392314567,
      "grad_norm": 26.5,
      "learning_rate": 2.351057491808162e-05,
      "loss": 1.4829,
      "num_input_tokens_seen": 3665984,
      "step": 6315
    },
    {
      "epoch": 0.9413166517724159,
      "grad_norm": 15.3125,
      "learning_rate": 2.3529192731605602e-05,
      "loss": 1.4833,
      "num_input_tokens_seen": 3668896,
      "step": 6320
    },
    {
      "epoch": 0.9420613643133751,
      "grad_norm": 12.0,
      "learning_rate": 2.354781054512958e-05,
      "loss": 1.0451,
      "num_input_tokens_seen": 3671552,
      "step": 6325
    },
    {
      "epoch": 0.9428060768543343,
      "grad_norm": 9.875,
      "learning_rate": 2.3566428358653562e-05,
      "loss": 1.3182,
      "num_input_tokens_seen": 3674528,
      "step": 6330
    },
    {
      "epoch": 0.9435507893952935,
      "grad_norm": 15.6875,
      "learning_rate": 2.358504617217754e-05,
      "loss": 1.2991,
      "num_input_tokens_seen": 3677184,
      "step": 6335
    },
    {
      "epoch": 0.9442955019362526,
      "grad_norm": 16.0,
      "learning_rate": 2.3603663985701522e-05,
      "loss": 1.3698,
      "num_input_tokens_seen": 3679744,
      "step": 6340
    },
    {
      "epoch": 0.9450402144772118,
      "grad_norm": 13.875,
      "learning_rate": 2.36222817992255e-05,
      "loss": 1.3446,
      "num_input_tokens_seen": 3682624,
      "step": 6345
    },
    {
      "epoch": 0.945784927018171,
      "grad_norm": 8.1875,
      "learning_rate": 2.364089961274948e-05,
      "loss": 1.2631,
      "num_input_tokens_seen": 3685984,
      "step": 6350
    },
    {
      "epoch": 0.9465296395591302,
      "grad_norm": 11.875,
      "learning_rate": 2.365951742627346e-05,
      "loss": 1.1254,
      "num_input_tokens_seen": 3688800,
      "step": 6355
    },
    {
      "epoch": 0.9472743521000894,
      "grad_norm": 26.125,
      "learning_rate": 2.3678135239797438e-05,
      "loss": 1.2644,
      "num_input_tokens_seen": 3691840,
      "step": 6360
    },
    {
      "epoch": 0.9480190646410486,
      "grad_norm": 13.9375,
      "learning_rate": 2.369675305332142e-05,
      "loss": 1.0261,
      "num_input_tokens_seen": 3694720,
      "step": 6365
    },
    {
      "epoch": 0.9487637771820078,
      "grad_norm": 12.0,
      "learning_rate": 2.3715370866845398e-05,
      "loss": 1.0979,
      "num_input_tokens_seen": 3697568,
      "step": 6370
    },
    {
      "epoch": 0.949508489722967,
      "grad_norm": 14.25,
      "learning_rate": 2.373398868036938e-05,
      "loss": 1.1721,
      "num_input_tokens_seen": 3700416,
      "step": 6375
    },
    {
      "epoch": 0.9502532022639262,
      "grad_norm": 20.0,
      "learning_rate": 2.3752606493893358e-05,
      "loss": 1.1092,
      "num_input_tokens_seen": 3703456,
      "step": 6380
    },
    {
      "epoch": 0.9509979148048853,
      "grad_norm": 13.0625,
      "learning_rate": 2.377122430741734e-05,
      "loss": 1.0167,
      "num_input_tokens_seen": 3706048,
      "step": 6385
    },
    {
      "epoch": 0.9517426273458445,
      "grad_norm": 8.8125,
      "learning_rate": 2.3789842120941318e-05,
      "loss": 1.1195,
      "num_input_tokens_seen": 3708736,
      "step": 6390
    },
    {
      "epoch": 0.9524873398868037,
      "grad_norm": 8.375,
      "learning_rate": 2.3808459934465296e-05,
      "loss": 0.8666,
      "num_input_tokens_seen": 3711392,
      "step": 6395
    },
    {
      "epoch": 0.9532320524277629,
      "grad_norm": 8.375,
      "learning_rate": 2.3827077747989278e-05,
      "loss": 1.4694,
      "num_input_tokens_seen": 3714304,
      "step": 6400
    },
    {
      "epoch": 0.9539767649687221,
      "grad_norm": 18.875,
      "learning_rate": 2.3845695561513256e-05,
      "loss": 1.2461,
      "num_input_tokens_seen": 3717120,
      "step": 6405
    },
    {
      "epoch": 0.9547214775096813,
      "grad_norm": 11.25,
      "learning_rate": 2.3864313375037238e-05,
      "loss": 1.2239,
      "num_input_tokens_seen": 3719648,
      "step": 6410
    },
    {
      "epoch": 0.9554661900506405,
      "grad_norm": 17.625,
      "learning_rate": 2.3882931188561216e-05,
      "loss": 1.2492,
      "num_input_tokens_seen": 3722688,
      "step": 6415
    },
    {
      "epoch": 0.9562109025915997,
      "grad_norm": 10.25,
      "learning_rate": 2.3901549002085197e-05,
      "loss": 1.2986,
      "num_input_tokens_seen": 3725440,
      "step": 6420
    },
    {
      "epoch": 0.9569556151325588,
      "grad_norm": 6.625,
      "learning_rate": 2.3920166815609176e-05,
      "loss": 0.7604,
      "num_input_tokens_seen": 3728320,
      "step": 6425
    },
    {
      "epoch": 0.957700327673518,
      "grad_norm": 18.625,
      "learning_rate": 2.3938784629133157e-05,
      "loss": 1.1521,
      "num_input_tokens_seen": 3731360,
      "step": 6430
    },
    {
      "epoch": 0.9584450402144772,
      "grad_norm": 13.8125,
      "learning_rate": 2.3957402442657135e-05,
      "loss": 1.0009,
      "num_input_tokens_seen": 3733984,
      "step": 6435
    },
    {
      "epoch": 0.9591897527554364,
      "grad_norm": 14.125,
      "learning_rate": 2.3976020256181114e-05,
      "loss": 1.1811,
      "num_input_tokens_seen": 3736832,
      "step": 6440
    },
    {
      "epoch": 0.9599344652963956,
      "grad_norm": 11.1875,
      "learning_rate": 2.3994638069705095e-05,
      "loss": 1.0515,
      "num_input_tokens_seen": 3739808,
      "step": 6445
    },
    {
      "epoch": 0.9606791778373548,
      "grad_norm": 5.125,
      "learning_rate": 2.4013255883229074e-05,
      "loss": 1.2035,
      "num_input_tokens_seen": 3742816,
      "step": 6450
    },
    {
      "epoch": 0.961423890378314,
      "grad_norm": 19.375,
      "learning_rate": 2.4031873696753055e-05,
      "loss": 1.1354,
      "num_input_tokens_seen": 3745760,
      "step": 6455
    },
    {
      "epoch": 0.9621686029192732,
      "grad_norm": 15.875,
      "learning_rate": 2.4050491510277033e-05,
      "loss": 1.2468,
      "num_input_tokens_seen": 3748512,
      "step": 6460
    },
    {
      "epoch": 0.9629133154602324,
      "grad_norm": 20.75,
      "learning_rate": 2.4069109323801015e-05,
      "loss": 1.4358,
      "num_input_tokens_seen": 3751520,
      "step": 6465
    },
    {
      "epoch": 0.9636580280011915,
      "grad_norm": 16.0,
      "learning_rate": 2.4087727137324993e-05,
      "loss": 1.1572,
      "num_input_tokens_seen": 3754368,
      "step": 6470
    },
    {
      "epoch": 0.9644027405421507,
      "grad_norm": 10.3125,
      "learning_rate": 2.4106344950848975e-05,
      "loss": 1.2651,
      "num_input_tokens_seen": 3757536,
      "step": 6475
    },
    {
      "epoch": 0.9651474530831099,
      "grad_norm": 9.625,
      "learning_rate": 2.4124962764372953e-05,
      "loss": 1.2453,
      "num_input_tokens_seen": 3760768,
      "step": 6480
    },
    {
      "epoch": 0.9658921656240691,
      "grad_norm": 7.84375,
      "learning_rate": 2.414358057789693e-05,
      "loss": 0.9094,
      "num_input_tokens_seen": 3763584,
      "step": 6485
    },
    {
      "epoch": 0.9666368781650283,
      "grad_norm": 14.5,
      "learning_rate": 2.4162198391420913e-05,
      "loss": 1.3793,
      "num_input_tokens_seen": 3766560,
      "step": 6490
    },
    {
      "epoch": 0.9673815907059875,
      "grad_norm": 10.5625,
      "learning_rate": 2.418081620494489e-05,
      "loss": 0.9849,
      "num_input_tokens_seen": 3769920,
      "step": 6495
    },
    {
      "epoch": 0.9681263032469467,
      "grad_norm": 12.9375,
      "learning_rate": 2.4199434018468873e-05,
      "loss": 1.3054,
      "num_input_tokens_seen": 3772800,
      "step": 6500
    },
    {
      "epoch": 0.9688710157879059,
      "grad_norm": 7.6875,
      "learning_rate": 2.421805183199285e-05,
      "loss": 1.0655,
      "num_input_tokens_seen": 3775712,
      "step": 6505
    },
    {
      "epoch": 0.9696157283288651,
      "grad_norm": 11.25,
      "learning_rate": 2.4236669645516833e-05,
      "loss": 0.8521,
      "num_input_tokens_seen": 3778496,
      "step": 6510
    },
    {
      "epoch": 0.9703604408698242,
      "grad_norm": 10.1875,
      "learning_rate": 2.425528745904081e-05,
      "loss": 1.2131,
      "num_input_tokens_seen": 3781664,
      "step": 6515
    },
    {
      "epoch": 0.9711051534107834,
      "grad_norm": 10.25,
      "learning_rate": 2.4273905272564793e-05,
      "loss": 1.0687,
      "num_input_tokens_seen": 3784416,
      "step": 6520
    },
    {
      "epoch": 0.9718498659517426,
      "grad_norm": 6.84375,
      "learning_rate": 2.429252308608877e-05,
      "loss": 0.9021,
      "num_input_tokens_seen": 3787168,
      "step": 6525
    },
    {
      "epoch": 0.9725945784927018,
      "grad_norm": 22.375,
      "learning_rate": 2.431114089961275e-05,
      "loss": 1.2497,
      "num_input_tokens_seen": 3790144,
      "step": 6530
    },
    {
      "epoch": 0.973339291033661,
      "grad_norm": 11.6875,
      "learning_rate": 2.432975871313673e-05,
      "loss": 0.7878,
      "num_input_tokens_seen": 3792992,
      "step": 6535
    },
    {
      "epoch": 0.9740840035746202,
      "grad_norm": 13.8125,
      "learning_rate": 2.434837652666071e-05,
      "loss": 1.2559,
      "num_input_tokens_seen": 3795840,
      "step": 6540
    },
    {
      "epoch": 0.9748287161155794,
      "grad_norm": 16.625,
      "learning_rate": 2.436699434018469e-05,
      "loss": 1.4853,
      "num_input_tokens_seen": 3798464,
      "step": 6545
    },
    {
      "epoch": 0.9755734286565386,
      "grad_norm": 9.4375,
      "learning_rate": 2.438561215370867e-05,
      "loss": 0.971,
      "num_input_tokens_seen": 3801472,
      "step": 6550
    },
    {
      "epoch": 0.9763181411974977,
      "grad_norm": 16.375,
      "learning_rate": 2.440422996723265e-05,
      "loss": 1.2135,
      "num_input_tokens_seen": 3804320,
      "step": 6555
    },
    {
      "epoch": 0.9770628537384569,
      "grad_norm": 13.5,
      "learning_rate": 2.4422847780756632e-05,
      "loss": 1.1204,
      "num_input_tokens_seen": 3807072,
      "step": 6560
    },
    {
      "epoch": 0.9778075662794161,
      "grad_norm": 15.125,
      "learning_rate": 2.4441465594280607e-05,
      "loss": 1.3555,
      "num_input_tokens_seen": 3809984,
      "step": 6565
    },
    {
      "epoch": 0.9785522788203753,
      "grad_norm": 11.875,
      "learning_rate": 2.446008340780459e-05,
      "loss": 1.2064,
      "num_input_tokens_seen": 3813024,
      "step": 6570
    },
    {
      "epoch": 0.9792969913613345,
      "grad_norm": 6.6875,
      "learning_rate": 2.4478701221328567e-05,
      "loss": 1.1637,
      "num_input_tokens_seen": 3815584,
      "step": 6575
    },
    {
      "epoch": 0.9800417039022937,
      "grad_norm": 9.375,
      "learning_rate": 2.4497319034852548e-05,
      "loss": 0.9453,
      "num_input_tokens_seen": 3818272,
      "step": 6580
    },
    {
      "epoch": 0.9807864164432529,
      "grad_norm": 9.4375,
      "learning_rate": 2.4515936848376527e-05,
      "loss": 0.966,
      "num_input_tokens_seen": 3821088,
      "step": 6585
    },
    {
      "epoch": 0.9815311289842121,
      "grad_norm": 9.75,
      "learning_rate": 2.4534554661900508e-05,
      "loss": 0.9714,
      "num_input_tokens_seen": 3823872,
      "step": 6590
    },
    {
      "epoch": 0.9822758415251713,
      "grad_norm": 12.25,
      "learning_rate": 2.4553172475424486e-05,
      "loss": 1.2448,
      "num_input_tokens_seen": 3826752,
      "step": 6595
    },
    {
      "epoch": 0.9830205540661304,
      "grad_norm": 12.125,
      "learning_rate": 2.4571790288948468e-05,
      "loss": 1.0822,
      "num_input_tokens_seen": 3829504,
      "step": 6600
    },
    {
      "epoch": 0.9837652666070896,
      "grad_norm": 18.375,
      "learning_rate": 2.459040810247245e-05,
      "loss": 1.1615,
      "num_input_tokens_seen": 3832448,
      "step": 6605
    },
    {
      "epoch": 0.9845099791480488,
      "grad_norm": 10.3125,
      "learning_rate": 2.4609025915996424e-05,
      "loss": 1.1623,
      "num_input_tokens_seen": 3835456,
      "step": 6610
    },
    {
      "epoch": 0.985254691689008,
      "grad_norm": 9.1875,
      "learning_rate": 2.4627643729520406e-05,
      "loss": 0.9736,
      "num_input_tokens_seen": 3838560,
      "step": 6615
    },
    {
      "epoch": 0.9859994042299672,
      "grad_norm": 11.75,
      "learning_rate": 2.4646261543044384e-05,
      "loss": 1.172,
      "num_input_tokens_seen": 3841568,
      "step": 6620
    },
    {
      "epoch": 0.9867441167709264,
      "grad_norm": 10.8125,
      "learning_rate": 2.4664879356568366e-05,
      "loss": 1.2493,
      "num_input_tokens_seen": 3844384,
      "step": 6625
    },
    {
      "epoch": 0.9874888293118856,
      "grad_norm": 11.9375,
      "learning_rate": 2.4683497170092344e-05,
      "loss": 1.2396,
      "num_input_tokens_seen": 3847616,
      "step": 6630
    },
    {
      "epoch": 0.9882335418528448,
      "grad_norm": 7.28125,
      "learning_rate": 2.4702114983616326e-05,
      "loss": 1.3367,
      "num_input_tokens_seen": 3850496,
      "step": 6635
    },
    {
      "epoch": 0.988978254393804,
      "grad_norm": 14.5625,
      "learning_rate": 2.4720732797140307e-05,
      "loss": 1.089,
      "num_input_tokens_seen": 3853600,
      "step": 6640
    },
    {
      "epoch": 0.9897229669347631,
      "grad_norm": 14.875,
      "learning_rate": 2.4739350610664286e-05,
      "loss": 1.33,
      "num_input_tokens_seen": 3856416,
      "step": 6645
    },
    {
      "epoch": 0.9904676794757223,
      "grad_norm": 7.8125,
      "learning_rate": 2.4757968424188267e-05,
      "loss": 1.1547,
      "num_input_tokens_seen": 3858976,
      "step": 6650
    },
    {
      "epoch": 0.9912123920166815,
      "grad_norm": 12.5,
      "learning_rate": 2.4776586237712242e-05,
      "loss": 1.2022,
      "num_input_tokens_seen": 3861984,
      "step": 6655
    },
    {
      "epoch": 0.9919571045576407,
      "grad_norm": 13.5625,
      "learning_rate": 2.4795204051236224e-05,
      "loss": 1.3487,
      "num_input_tokens_seen": 3864768,
      "step": 6660
    },
    {
      "epoch": 0.9927018170986,
      "grad_norm": 6.125,
      "learning_rate": 2.4813821864760202e-05,
      "loss": 0.8489,
      "num_input_tokens_seen": 3867680,
      "step": 6665
    },
    {
      "epoch": 0.9934465296395592,
      "grad_norm": 8.5,
      "learning_rate": 2.4832439678284184e-05,
      "loss": 1.1237,
      "num_input_tokens_seen": 3870560,
      "step": 6670
    },
    {
      "epoch": 0.9941912421805184,
      "grad_norm": 12.8125,
      "learning_rate": 2.4851057491808162e-05,
      "loss": 0.9041,
      "num_input_tokens_seen": 3873408,
      "step": 6675
    },
    {
      "epoch": 0.9949359547214776,
      "grad_norm": 17.125,
      "learning_rate": 2.4869675305332143e-05,
      "loss": 1.4361,
      "num_input_tokens_seen": 3876320,
      "step": 6680
    },
    {
      "epoch": 0.9956806672624366,
      "grad_norm": 17.5,
      "learning_rate": 2.4888293118856125e-05,
      "loss": 1.1408,
      "num_input_tokens_seen": 3879008,
      "step": 6685
    },
    {
      "epoch": 0.9964253798033958,
      "grad_norm": 11.625,
      "learning_rate": 2.4906910932380103e-05,
      "loss": 1.2623,
      "num_input_tokens_seen": 3881760,
      "step": 6690
    },
    {
      "epoch": 0.997170092344355,
      "grad_norm": 13.125,
      "learning_rate": 2.492552874590408e-05,
      "loss": 1.0238,
      "num_input_tokens_seen": 3884672,
      "step": 6695
    },
    {
      "epoch": 0.9979148048853143,
      "grad_norm": 4.0625,
      "learning_rate": 2.494414655942806e-05,
      "loss": 0.6763,
      "num_input_tokens_seen": 3887360,
      "step": 6700
    },
    {
      "epoch": 0.9986595174262735,
      "grad_norm": 18.25,
      "learning_rate": 2.496276437295204e-05,
      "loss": 1.5089,
      "num_input_tokens_seen": 3889920,
      "step": 6705
    },
    {
      "epoch": 0.9994042299672327,
      "grad_norm": 11.0,
      "learning_rate": 2.498138218647602e-05,
      "loss": 1.0655,
      "num_input_tokens_seen": 3892704,
      "step": 6710
    },
    {
      "epoch": 1.0,
      "eval_loss": 1.1415724754333496,
      "eval_runtime": 43.7352,
      "eval_samples_per_second": 68.229,
      "eval_steps_per_second": 17.057,
      "num_input_tokens_seen": 3894552,
      "step": 6714
    },
    {
      "epoch": 1.0001489425081918,
      "grad_norm": 12.0625,
      "learning_rate": 2.5e-05,
      "loss": 0.8641,
      "num_input_tokens_seen": 3895096,
      "step": 6715
    },
    {
      "epoch": 1.000893655049151,
      "grad_norm": 18.875,
      "learning_rate": 2.5018617813523983e-05,
      "loss": 0.882,
      "num_input_tokens_seen": 3898200,
      "step": 6720
    },
    {
      "epoch": 1.0016383675901102,
      "grad_norm": 9.5,
      "learning_rate": 2.503723562704796e-05,
      "loss": 1.3913,
      "num_input_tokens_seen": 3900984,
      "step": 6725
    },
    {
      "epoch": 1.0023830801310694,
      "grad_norm": 11.875,
      "learning_rate": 2.5055853440571943e-05,
      "loss": 1.0411,
      "num_input_tokens_seen": 3903928,
      "step": 6730
    },
    {
      "epoch": 1.0031277926720286,
      "grad_norm": 12.1875,
      "learning_rate": 2.507447125409592e-05,
      "loss": 1.1688,
      "num_input_tokens_seen": 3906840,
      "step": 6735
    },
    {
      "epoch": 1.0038725052129878,
      "grad_norm": 15.0,
      "learning_rate": 2.5093089067619903e-05,
      "loss": 1.2366,
      "num_input_tokens_seen": 3909880,
      "step": 6740
    },
    {
      "epoch": 1.004617217753947,
      "grad_norm": 13.4375,
      "learning_rate": 2.511170688114388e-05,
      "loss": 0.9745,
      "num_input_tokens_seen": 3912952,
      "step": 6745
    },
    {
      "epoch": 1.0053619302949062,
      "grad_norm": 18.25,
      "learning_rate": 2.5130324694667862e-05,
      "loss": 1.2724,
      "num_input_tokens_seen": 3916312,
      "step": 6750
    },
    {
      "epoch": 1.0061066428358654,
      "grad_norm": 16.75,
      "learning_rate": 2.514894250819184e-05,
      "loss": 1.0517,
      "num_input_tokens_seen": 3919576,
      "step": 6755
    },
    {
      "epoch": 1.0068513553768246,
      "grad_norm": 12.1875,
      "learning_rate": 2.5167560321715815e-05,
      "loss": 1.1391,
      "num_input_tokens_seen": 3922680,
      "step": 6760
    },
    {
      "epoch": 1.0075960679177838,
      "grad_norm": 19.0,
      "learning_rate": 2.5186178135239797e-05,
      "loss": 1.4484,
      "num_input_tokens_seen": 3925560,
      "step": 6765
    },
    {
      "epoch": 1.008340780458743,
      "grad_norm": 9.4375,
      "learning_rate": 2.5204795948763775e-05,
      "loss": 1.057,
      "num_input_tokens_seen": 3928248,
      "step": 6770
    },
    {
      "epoch": 1.0090854929997022,
      "grad_norm": 20.375,
      "learning_rate": 2.5223413762287757e-05,
      "loss": 1.157,
      "num_input_tokens_seen": 3931160,
      "step": 6775
    },
    {
      "epoch": 1.0098302055406614,
      "grad_norm": 14.1875,
      "learning_rate": 2.5242031575811735e-05,
      "loss": 1.2176,
      "num_input_tokens_seen": 3933720,
      "step": 6780
    },
    {
      "epoch": 1.0105749180816206,
      "grad_norm": 15.25,
      "learning_rate": 2.5260649389335717e-05,
      "loss": 1.264,
      "num_input_tokens_seen": 3936728,
      "step": 6785
    },
    {
      "epoch": 1.0113196306225798,
      "grad_norm": 8.6875,
      "learning_rate": 2.5279267202859695e-05,
      "loss": 0.9626,
      "num_input_tokens_seen": 3939832,
      "step": 6790
    },
    {
      "epoch": 1.0120643431635388,
      "grad_norm": 8.3125,
      "learning_rate": 2.5297885016383677e-05,
      "loss": 0.8386,
      "num_input_tokens_seen": 3942616,
      "step": 6795
    },
    {
      "epoch": 1.012809055704498,
      "grad_norm": 7.53125,
      "learning_rate": 2.5316502829907658e-05,
      "loss": 0.913,
      "num_input_tokens_seen": 3945464,
      "step": 6800
    },
    {
      "epoch": 1.0135537682454572,
      "grad_norm": 10.5625,
      "learning_rate": 2.5335120643431636e-05,
      "loss": 1.5296,
      "num_input_tokens_seen": 3948536,
      "step": 6805
    },
    {
      "epoch": 1.0142984807864164,
      "grad_norm": 5.96875,
      "learning_rate": 2.5353738456955618e-05,
      "loss": 1.3449,
      "num_input_tokens_seen": 3951288,
      "step": 6810
    },
    {
      "epoch": 1.0150431933273756,
      "grad_norm": 12.6875,
      "learning_rate": 2.5372356270479596e-05,
      "loss": 1.0556,
      "num_input_tokens_seen": 3954072,
      "step": 6815
    },
    {
      "epoch": 1.0157879058683348,
      "grad_norm": 21.875,
      "learning_rate": 2.5390974084003578e-05,
      "loss": 1.2689,
      "num_input_tokens_seen": 3956824,
      "step": 6820
    },
    {
      "epoch": 1.016532618409294,
      "grad_norm": 8.6875,
      "learning_rate": 2.5409591897527556e-05,
      "loss": 1.057,
      "num_input_tokens_seen": 3959608,
      "step": 6825
    },
    {
      "epoch": 1.0172773309502532,
      "grad_norm": 11.6875,
      "learning_rate": 2.5428209711051538e-05,
      "loss": 1.0402,
      "num_input_tokens_seen": 3962104,
      "step": 6830
    },
    {
      "epoch": 1.0180220434912124,
      "grad_norm": 12.875,
      "learning_rate": 2.5446827524575516e-05,
      "loss": 1.0697,
      "num_input_tokens_seen": 3964984,
      "step": 6835
    },
    {
      "epoch": 1.0187667560321716,
      "grad_norm": 6.4375,
      "learning_rate": 2.5465445338099498e-05,
      "loss": 1.2641,
      "num_input_tokens_seen": 3967800,
      "step": 6840
    },
    {
      "epoch": 1.0195114685731308,
      "grad_norm": 10.75,
      "learning_rate": 2.5484063151623473e-05,
      "loss": 0.9188,
      "num_input_tokens_seen": 3970712,
      "step": 6845
    },
    {
      "epoch": 1.02025618111409,
      "grad_norm": 7.40625,
      "learning_rate": 2.550268096514745e-05,
      "loss": 1.0488,
      "num_input_tokens_seen": 3973624,
      "step": 6850
    },
    {
      "epoch": 1.0210008936550492,
      "grad_norm": 8.75,
      "learning_rate": 2.5521298778671432e-05,
      "loss": 1.4859,
      "num_input_tokens_seen": 3976792,
      "step": 6855
    },
    {
      "epoch": 1.0217456061960084,
      "grad_norm": 14.125,
      "learning_rate": 2.553991659219541e-05,
      "loss": 1.2022,
      "num_input_tokens_seen": 3979512,
      "step": 6860
    },
    {
      "epoch": 1.0224903187369676,
      "grad_norm": 16.25,
      "learning_rate": 2.5558534405719392e-05,
      "loss": 1.3871,
      "num_input_tokens_seen": 3982360,
      "step": 6865
    },
    {
      "epoch": 1.0232350312779268,
      "grad_norm": 18.75,
      "learning_rate": 2.557715221924337e-05,
      "loss": 1.1531,
      "num_input_tokens_seen": 3985240,
      "step": 6870
    },
    {
      "epoch": 1.023979743818886,
      "grad_norm": 15.375,
      "learning_rate": 2.5595770032767352e-05,
      "loss": 1.1485,
      "num_input_tokens_seen": 3987960,
      "step": 6875
    },
    {
      "epoch": 1.024724456359845,
      "grad_norm": 14.875,
      "learning_rate": 2.5614387846291334e-05,
      "loss": 0.846,
      "num_input_tokens_seen": 3990712,
      "step": 6880
    },
    {
      "epoch": 1.0254691689008042,
      "grad_norm": 11.8125,
      "learning_rate": 2.5633005659815312e-05,
      "loss": 1.3969,
      "num_input_tokens_seen": 3993592,
      "step": 6885
    },
    {
      "epoch": 1.0262138814417634,
      "grad_norm": 8.5625,
      "learning_rate": 2.5651623473339294e-05,
      "loss": 1.1046,
      "num_input_tokens_seen": 3996792,
      "step": 6890
    },
    {
      "epoch": 1.0269585939827226,
      "grad_norm": 6.15625,
      "learning_rate": 2.5670241286863272e-05,
      "loss": 1.3134,
      "num_input_tokens_seen": 3999512,
      "step": 6895
    },
    {
      "epoch": 1.0277033065236818,
      "grad_norm": 25.25,
      "learning_rate": 2.5688859100387253e-05,
      "loss": 1.1721,
      "num_input_tokens_seen": 4002488,
      "step": 6900
    },
    {
      "epoch": 1.028448019064641,
      "grad_norm": 11.0,
      "learning_rate": 2.570747691391123e-05,
      "loss": 1.3146,
      "num_input_tokens_seen": 4005336,
      "step": 6905
    },
    {
      "epoch": 1.0291927316056002,
      "grad_norm": 13.5625,
      "learning_rate": 2.5726094727435213e-05,
      "loss": 1.3967,
      "num_input_tokens_seen": 4008184,
      "step": 6910
    },
    {
      "epoch": 1.0299374441465594,
      "grad_norm": 9.25,
      "learning_rate": 2.574471254095919e-05,
      "loss": 1.1942,
      "num_input_tokens_seen": 4011128,
      "step": 6915
    },
    {
      "epoch": 1.0306821566875186,
      "grad_norm": 17.875,
      "learning_rate": 2.5763330354483173e-05,
      "loss": 1.1613,
      "num_input_tokens_seen": 4014168,
      "step": 6920
    },
    {
      "epoch": 1.0314268692284778,
      "grad_norm": 7.90625,
      "learning_rate": 2.578194816800715e-05,
      "loss": 1.0773,
      "num_input_tokens_seen": 4017144,
      "step": 6925
    },
    {
      "epoch": 1.032171581769437,
      "grad_norm": 9.625,
      "learning_rate": 2.5800565981531133e-05,
      "loss": 1.1354,
      "num_input_tokens_seen": 4020024,
      "step": 6930
    },
    {
      "epoch": 1.0329162943103962,
      "grad_norm": 8.8125,
      "learning_rate": 2.5819183795055108e-05,
      "loss": 0.9239,
      "num_input_tokens_seen": 4022904,
      "step": 6935
    },
    {
      "epoch": 1.0336610068513554,
      "grad_norm": 14.9375,
      "learning_rate": 2.5837801608579086e-05,
      "loss": 1.1325,
      "num_input_tokens_seen": 4025656,
      "step": 6940
    },
    {
      "epoch": 1.0344057193923146,
      "grad_norm": 15.6875,
      "learning_rate": 2.5856419422103068e-05,
      "loss": 1.3539,
      "num_input_tokens_seen": 4028632,
      "step": 6945
    },
    {
      "epoch": 1.0351504319332738,
      "grad_norm": 8.875,
      "learning_rate": 2.587503723562705e-05,
      "loss": 1.0727,
      "num_input_tokens_seen": 4031480,
      "step": 6950
    },
    {
      "epoch": 1.035895144474233,
      "grad_norm": 14.3125,
      "learning_rate": 2.5893655049151027e-05,
      "loss": 1.3682,
      "num_input_tokens_seen": 4034424,
      "step": 6955
    },
    {
      "epoch": 1.0366398570151922,
      "grad_norm": 14.1875,
      "learning_rate": 2.591227286267501e-05,
      "loss": 1.1879,
      "num_input_tokens_seen": 4037272,
      "step": 6960
    },
    {
      "epoch": 1.0373845695561514,
      "grad_norm": 7.0,
      "learning_rate": 2.5930890676198987e-05,
      "loss": 1.1047,
      "num_input_tokens_seen": 4040312,
      "step": 6965
    },
    {
      "epoch": 1.0381292820971104,
      "grad_norm": 14.875,
      "learning_rate": 2.594950848972297e-05,
      "loss": 1.1179,
      "num_input_tokens_seen": 4043224,
      "step": 6970
    },
    {
      "epoch": 1.0388739946380696,
      "grad_norm": 12.6875,
      "learning_rate": 2.5968126303246947e-05,
      "loss": 1.1959,
      "num_input_tokens_seen": 4046168,
      "step": 6975
    },
    {
      "epoch": 1.0396187071790288,
      "grad_norm": 11.375,
      "learning_rate": 2.598674411677093e-05,
      "loss": 1.4404,
      "num_input_tokens_seen": 4049080,
      "step": 6980
    },
    {
      "epoch": 1.040363419719988,
      "grad_norm": 8.9375,
      "learning_rate": 2.6005361930294907e-05,
      "loss": 1.0251,
      "num_input_tokens_seen": 4052312,
      "step": 6985
    },
    {
      "epoch": 1.0411081322609472,
      "grad_norm": 18.25,
      "learning_rate": 2.602397974381889e-05,
      "loss": 1.0305,
      "num_input_tokens_seen": 4055000,
      "step": 6990
    },
    {
      "epoch": 1.0418528448019064,
      "grad_norm": 15.4375,
      "learning_rate": 2.6042597557342867e-05,
      "loss": 1.3576,
      "num_input_tokens_seen": 4058136,
      "step": 6995
    },
    {
      "epoch": 1.0425975573428656,
      "grad_norm": 13.625,
      "learning_rate": 2.606121537086685e-05,
      "loss": 1.3406,
      "num_input_tokens_seen": 4061176,
      "step": 7000
    },
    {
      "epoch": 1.0433422698838248,
      "grad_norm": 6.34375,
      "learning_rate": 2.6079833184390827e-05,
      "loss": 0.9705,
      "num_input_tokens_seen": 4063960,
      "step": 7005
    },
    {
      "epoch": 1.044086982424784,
      "grad_norm": 8.0625,
      "learning_rate": 2.609845099791481e-05,
      "loss": 0.6851,
      "num_input_tokens_seen": 4066872,
      "step": 7010
    },
    {
      "epoch": 1.0448316949657432,
      "grad_norm": 10.4375,
      "learning_rate": 2.611706881143879e-05,
      "loss": 1.3842,
      "num_input_tokens_seen": 4069688,
      "step": 7015
    },
    {
      "epoch": 1.0455764075067024,
      "grad_norm": 14.875,
      "learning_rate": 2.613568662496276e-05,
      "loss": 1.3092,
      "num_input_tokens_seen": 4072504,
      "step": 7020
    },
    {
      "epoch": 1.0463211200476616,
      "grad_norm": 18.75,
      "learning_rate": 2.6154304438486743e-05,
      "loss": 1.1865,
      "num_input_tokens_seen": 4075448,
      "step": 7025
    },
    {
      "epoch": 1.0470658325886208,
      "grad_norm": 10.3125,
      "learning_rate": 2.6172922252010725e-05,
      "loss": 1.4792,
      "num_input_tokens_seen": 4078168,
      "step": 7030
    },
    {
      "epoch": 1.04781054512958,
      "grad_norm": 16.625,
      "learning_rate": 2.6191540065534703e-05,
      "loss": 1.6919,
      "num_input_tokens_seen": 4081048,
      "step": 7035
    },
    {
      "epoch": 1.0485552576705393,
      "grad_norm": 9.5,
      "learning_rate": 2.6210157879058685e-05,
      "loss": 1.3264,
      "num_input_tokens_seen": 4083928,
      "step": 7040
    },
    {
      "epoch": 1.0492999702114985,
      "grad_norm": 13.8125,
      "learning_rate": 2.6228775692582663e-05,
      "loss": 1.095,
      "num_input_tokens_seen": 4086680,
      "step": 7045
    },
    {
      "epoch": 1.0500446827524577,
      "grad_norm": 4.71875,
      "learning_rate": 2.6247393506106644e-05,
      "loss": 1.0018,
      "num_input_tokens_seen": 4089592,
      "step": 7050
    },
    {
      "epoch": 1.0507893952934166,
      "grad_norm": 13.625,
      "learning_rate": 2.6266011319630623e-05,
      "loss": 1.1848,
      "num_input_tokens_seen": 4092504,
      "step": 7055
    },
    {
      "epoch": 1.0515341078343758,
      "grad_norm": 5.84375,
      "learning_rate": 2.6284629133154604e-05,
      "loss": 0.9983,
      "num_input_tokens_seen": 4095576,
      "step": 7060
    },
    {
      "epoch": 1.052278820375335,
      "grad_norm": 9.8125,
      "learning_rate": 2.6303246946678582e-05,
      "loss": 1.1027,
      "num_input_tokens_seen": 4098744,
      "step": 7065
    },
    {
      "epoch": 1.0530235329162942,
      "grad_norm": 12.75,
      "learning_rate": 2.6321864760202564e-05,
      "loss": 1.3407,
      "num_input_tokens_seen": 4101624,
      "step": 7070
    },
    {
      "epoch": 1.0537682454572534,
      "grad_norm": 13.625,
      "learning_rate": 2.6340482573726542e-05,
      "loss": 1.1458,
      "num_input_tokens_seen": 4104856,
      "step": 7075
    },
    {
      "epoch": 1.0545129579982127,
      "grad_norm": 10.5625,
      "learning_rate": 2.6359100387250524e-05,
      "loss": 1.2882,
      "num_input_tokens_seen": 4107800,
      "step": 7080
    },
    {
      "epoch": 1.0552576705391719,
      "grad_norm": 7.0,
      "learning_rate": 2.6377718200774502e-05,
      "loss": 0.8222,
      "num_input_tokens_seen": 4110744,
      "step": 7085
    },
    {
      "epoch": 1.056002383080131,
      "grad_norm": 5.28125,
      "learning_rate": 2.6396336014298484e-05,
      "loss": 1.0874,
      "num_input_tokens_seen": 4113592,
      "step": 7090
    },
    {
      "epoch": 1.0567470956210903,
      "grad_norm": 11.6875,
      "learning_rate": 2.6414953827822465e-05,
      "loss": 1.1064,
      "num_input_tokens_seen": 4116472,
      "step": 7095
    },
    {
      "epoch": 1.0574918081620495,
      "grad_norm": 16.0,
      "learning_rate": 2.6433571641346444e-05,
      "loss": 1.229,
      "num_input_tokens_seen": 4119288,
      "step": 7100
    },
    {
      "epoch": 1.0582365207030087,
      "grad_norm": 11.3125,
      "learning_rate": 2.645218945487042e-05,
      "loss": 1.4081,
      "num_input_tokens_seen": 4122488,
      "step": 7105
    },
    {
      "epoch": 1.0589812332439679,
      "grad_norm": 10.9375,
      "learning_rate": 2.64708072683944e-05,
      "loss": 1.3501,
      "num_input_tokens_seen": 4125432,
      "step": 7110
    },
    {
      "epoch": 1.059725945784927,
      "grad_norm": 13.4375,
      "learning_rate": 2.648942508191838e-05,
      "loss": 0.8918,
      "num_input_tokens_seen": 4128248,
      "step": 7115
    },
    {
      "epoch": 1.0604706583258863,
      "grad_norm": 16.75,
      "learning_rate": 2.650804289544236e-05,
      "loss": 1.3122,
      "num_input_tokens_seen": 4131256,
      "step": 7120
    },
    {
      "epoch": 1.0612153708668455,
      "grad_norm": 10.5,
      "learning_rate": 2.6526660708966338e-05,
      "loss": 1.3997,
      "num_input_tokens_seen": 4134296,
      "step": 7125
    },
    {
      "epoch": 1.0619600834078047,
      "grad_norm": 11.375,
      "learning_rate": 2.654527852249032e-05,
      "loss": 1.1815,
      "num_input_tokens_seen": 4137112,
      "step": 7130
    },
    {
      "epoch": 1.0627047959487639,
      "grad_norm": 12.75,
      "learning_rate": 2.6563896336014298e-05,
      "loss": 1.1263,
      "num_input_tokens_seen": 4140024,
      "step": 7135
    },
    {
      "epoch": 1.063449508489723,
      "grad_norm": 10.625,
      "learning_rate": 2.658251414953828e-05,
      "loss": 0.7251,
      "num_input_tokens_seen": 4142968,
      "step": 7140
    },
    {
      "epoch": 1.064194221030682,
      "grad_norm": 10.0625,
      "learning_rate": 2.6601131963062258e-05,
      "loss": 0.9917,
      "num_input_tokens_seen": 4145784,
      "step": 7145
    },
    {
      "epoch": 1.0649389335716413,
      "grad_norm": 10.5,
      "learning_rate": 2.661974977658624e-05,
      "loss": 1.0607,
      "num_input_tokens_seen": 4148408,
      "step": 7150
    },
    {
      "epoch": 1.0656836461126005,
      "grad_norm": 8.8125,
      "learning_rate": 2.6638367590110218e-05,
      "loss": 1.0785,
      "num_input_tokens_seen": 4151352,
      "step": 7155
    },
    {
      "epoch": 1.0664283586535597,
      "grad_norm": 18.125,
      "learning_rate": 2.66569854036342e-05,
      "loss": 1.1361,
      "num_input_tokens_seen": 4154168,
      "step": 7160
    },
    {
      "epoch": 1.0671730711945189,
      "grad_norm": 19.375,
      "learning_rate": 2.667560321715818e-05,
      "loss": 0.9165,
      "num_input_tokens_seen": 4157240,
      "step": 7165
    },
    {
      "epoch": 1.067917783735478,
      "grad_norm": 11.4375,
      "learning_rate": 2.669422103068216e-05,
      "loss": 1.1676,
      "num_input_tokens_seen": 4160216,
      "step": 7170
    },
    {
      "epoch": 1.0686624962764373,
      "grad_norm": 9.4375,
      "learning_rate": 2.671283884420614e-05,
      "loss": 0.7905,
      "num_input_tokens_seen": 4162968,
      "step": 7175
    },
    {
      "epoch": 1.0694072088173965,
      "grad_norm": 6.875,
      "learning_rate": 2.673145665773012e-05,
      "loss": 1.0986,
      "num_input_tokens_seen": 4166232,
      "step": 7180
    },
    {
      "epoch": 1.0701519213583557,
      "grad_norm": 12.375,
      "learning_rate": 2.67500744712541e-05,
      "loss": 0.9552,
      "num_input_tokens_seen": 4169368,
      "step": 7185
    },
    {
      "epoch": 1.0708966338993149,
      "grad_norm": 15.8125,
      "learning_rate": 2.676869228477808e-05,
      "loss": 1.4121,
      "num_input_tokens_seen": 4172344,
      "step": 7190
    },
    {
      "epoch": 1.071641346440274,
      "grad_norm": 7.53125,
      "learning_rate": 2.6787310098302054e-05,
      "loss": 1.0431,
      "num_input_tokens_seen": 4175064,
      "step": 7195
    },
    {
      "epoch": 1.0723860589812333,
      "grad_norm": 16.5,
      "learning_rate": 2.6805927911826035e-05,
      "loss": 0.9022,
      "num_input_tokens_seen": 4177752,
      "step": 7200
    },
    {
      "epoch": 1.0731307715221925,
      "grad_norm": 17.5,
      "learning_rate": 2.6824545725350014e-05,
      "loss": 1.2245,
      "num_input_tokens_seen": 4180408,
      "step": 7205
    },
    {
      "epoch": 1.0738754840631517,
      "grad_norm": 12.6875,
      "learning_rate": 2.6843163538873995e-05,
      "loss": 1.1857,
      "num_input_tokens_seen": 4183480,
      "step": 7210
    },
    {
      "epoch": 1.074620196604111,
      "grad_norm": 14.1875,
      "learning_rate": 2.6861781352397974e-05,
      "loss": 1.131,
      "num_input_tokens_seen": 4186200,
      "step": 7215
    },
    {
      "epoch": 1.07536490914507,
      "grad_norm": 13.125,
      "learning_rate": 2.6880399165921955e-05,
      "loss": 1.2183,
      "num_input_tokens_seen": 4189112,
      "step": 7220
    },
    {
      "epoch": 1.076109621686029,
      "grad_norm": 8.75,
      "learning_rate": 2.6899016979445933e-05,
      "loss": 1.3248,
      "num_input_tokens_seen": 4192024,
      "step": 7225
    },
    {
      "epoch": 1.0768543342269883,
      "grad_norm": 17.0,
      "learning_rate": 2.6917634792969915e-05,
      "loss": 1.449,
      "num_input_tokens_seen": 4194776,
      "step": 7230
    },
    {
      "epoch": 1.0775990467679475,
      "grad_norm": 20.125,
      "learning_rate": 2.6936252606493893e-05,
      "loss": 1.785,
      "num_input_tokens_seen": 4197880,
      "step": 7235
    },
    {
      "epoch": 1.0783437593089067,
      "grad_norm": 9.75,
      "learning_rate": 2.6954870420017875e-05,
      "loss": 1.5447,
      "num_input_tokens_seen": 4200920,
      "step": 7240
    },
    {
      "epoch": 1.079088471849866,
      "grad_norm": 11.25,
      "learning_rate": 2.6973488233541856e-05,
      "loss": 1.1083,
      "num_input_tokens_seen": 4203832,
      "step": 7245
    },
    {
      "epoch": 1.079833184390825,
      "grad_norm": 11.75,
      "learning_rate": 2.6992106047065835e-05,
      "loss": 1.4019,
      "num_input_tokens_seen": 4206872,
      "step": 7250
    },
    {
      "epoch": 1.0805778969317843,
      "grad_norm": 8.25,
      "learning_rate": 2.7010723860589816e-05,
      "loss": 0.8754,
      "num_input_tokens_seen": 4209688,
      "step": 7255
    },
    {
      "epoch": 1.0813226094727435,
      "grad_norm": 22.625,
      "learning_rate": 2.7029341674113795e-05,
      "loss": 1.4282,
      "num_input_tokens_seen": 4212600,
      "step": 7260
    },
    {
      "epoch": 1.0820673220137027,
      "grad_norm": 9.625,
      "learning_rate": 2.7047959487637776e-05,
      "loss": 1.2222,
      "num_input_tokens_seen": 4215480,
      "step": 7265
    },
    {
      "epoch": 1.082812034554662,
      "grad_norm": 10.25,
      "learning_rate": 2.7066577301161754e-05,
      "loss": 0.9147,
      "num_input_tokens_seen": 4218360,
      "step": 7270
    },
    {
      "epoch": 1.083556747095621,
      "grad_norm": 9.625,
      "learning_rate": 2.7085195114685736e-05,
      "loss": 1.237,
      "num_input_tokens_seen": 4221528,
      "step": 7275
    },
    {
      "epoch": 1.0843014596365803,
      "grad_norm": 10.5,
      "learning_rate": 2.710381292820971e-05,
      "loss": 1.0596,
      "num_input_tokens_seen": 4224120,
      "step": 7280
    },
    {
      "epoch": 1.0850461721775395,
      "grad_norm": 9.3125,
      "learning_rate": 2.712243074173369e-05,
      "loss": 0.9886,
      "num_input_tokens_seen": 4226840,
      "step": 7285
    },
    {
      "epoch": 1.0857908847184987,
      "grad_norm": 12.0,
      "learning_rate": 2.714104855525767e-05,
      "loss": 1.1407,
      "num_input_tokens_seen": 4229976,
      "step": 7290
    },
    {
      "epoch": 1.086535597259458,
      "grad_norm": 23.5,
      "learning_rate": 2.715966636878165e-05,
      "loss": 1.724,
      "num_input_tokens_seen": 4233080,
      "step": 7295
    },
    {
      "epoch": 1.0872803098004171,
      "grad_norm": 8.4375,
      "learning_rate": 2.717828418230563e-05,
      "loss": 1.0278,
      "num_input_tokens_seen": 4235800,
      "step": 7300
    },
    {
      "epoch": 1.0880250223413763,
      "grad_norm": 12.875,
      "learning_rate": 2.719690199582961e-05,
      "loss": 1.3566,
      "num_input_tokens_seen": 4238552,
      "step": 7305
    },
    {
      "epoch": 1.0887697348823355,
      "grad_norm": 12.75,
      "learning_rate": 2.721551980935359e-05,
      "loss": 0.9764,
      "num_input_tokens_seen": 4241624,
      "step": 7310
    },
    {
      "epoch": 1.0895144474232945,
      "grad_norm": 15.3125,
      "learning_rate": 2.723413762287757e-05,
      "loss": 1.2055,
      "num_input_tokens_seen": 4244664,
      "step": 7315
    },
    {
      "epoch": 1.0902591599642537,
      "grad_norm": 15.375,
      "learning_rate": 2.725275543640155e-05,
      "loss": 1.4984,
      "num_input_tokens_seen": 4247512,
      "step": 7320
    },
    {
      "epoch": 1.091003872505213,
      "grad_norm": 15.9375,
      "learning_rate": 2.7271373249925532e-05,
      "loss": 1.1056,
      "num_input_tokens_seen": 4250392,
      "step": 7325
    },
    {
      "epoch": 1.0917485850461721,
      "grad_norm": 13.0625,
      "learning_rate": 2.728999106344951e-05,
      "loss": 1.0482,
      "num_input_tokens_seen": 4253496,
      "step": 7330
    },
    {
      "epoch": 1.0924932975871313,
      "grad_norm": 11.6875,
      "learning_rate": 2.7308608876973492e-05,
      "loss": 0.6989,
      "num_input_tokens_seen": 4256472,
      "step": 7335
    },
    {
      "epoch": 1.0932380101280905,
      "grad_norm": 24.375,
      "learning_rate": 2.732722669049747e-05,
      "loss": 1.4111,
      "num_input_tokens_seen": 4259256,
      "step": 7340
    },
    {
      "epoch": 1.0939827226690497,
      "grad_norm": 8.1875,
      "learning_rate": 2.734584450402145e-05,
      "loss": 0.841,
      "num_input_tokens_seen": 4261816,
      "step": 7345
    },
    {
      "epoch": 1.094727435210009,
      "grad_norm": 15.8125,
      "learning_rate": 2.736446231754543e-05,
      "loss": 1.1068,
      "num_input_tokens_seen": 4265048,
      "step": 7350
    },
    {
      "epoch": 1.0954721477509681,
      "grad_norm": 10.6875,
      "learning_rate": 2.738308013106941e-05,
      "loss": 1.1184,
      "num_input_tokens_seen": 4267864,
      "step": 7355
    },
    {
      "epoch": 1.0962168602919273,
      "grad_norm": 10.375,
      "learning_rate": 2.740169794459339e-05,
      "loss": 0.9694,
      "num_input_tokens_seen": 4270776,
      "step": 7360
    },
    {
      "epoch": 1.0969615728328865,
      "grad_norm": 10.0625,
      "learning_rate": 2.7420315758117365e-05,
      "loss": 1.1744,
      "num_input_tokens_seen": 4273656,
      "step": 7365
    },
    {
      "epoch": 1.0977062853738457,
      "grad_norm": 12.5625,
      "learning_rate": 2.7438933571641346e-05,
      "loss": 0.8347,
      "num_input_tokens_seen": 4276408,
      "step": 7370
    },
    {
      "epoch": 1.098450997914805,
      "grad_norm": 10.125,
      "learning_rate": 2.7457551385165324e-05,
      "loss": 1.1343,
      "num_input_tokens_seen": 4279768,
      "step": 7375
    },
    {
      "epoch": 1.0991957104557641,
      "grad_norm": 15.625,
      "learning_rate": 2.7476169198689306e-05,
      "loss": 1.2115,
      "num_input_tokens_seen": 4282712,
      "step": 7380
    },
    {
      "epoch": 1.0999404229967233,
      "grad_norm": 11.6875,
      "learning_rate": 2.7494787012213284e-05,
      "loss": 1.2601,
      "num_input_tokens_seen": 4285784,
      "step": 7385
    },
    {
      "epoch": 1.1006851355376825,
      "grad_norm": 9.5,
      "learning_rate": 2.7513404825737266e-05,
      "loss": 0.9304,
      "num_input_tokens_seen": 4288824,
      "step": 7390
    },
    {
      "epoch": 1.1014298480786415,
      "grad_norm": 13.6875,
      "learning_rate": 2.7532022639261244e-05,
      "loss": 1.2237,
      "num_input_tokens_seen": 4291384,
      "step": 7395
    },
    {
      "epoch": 1.1021745606196007,
      "grad_norm": 16.75,
      "learning_rate": 2.7550640452785226e-05,
      "loss": 1.2455,
      "num_input_tokens_seen": 4294200,
      "step": 7400
    },
    {
      "epoch": 1.10291927316056,
      "grad_norm": 10.0625,
      "learning_rate": 2.7569258266309207e-05,
      "loss": 1.0604,
      "num_input_tokens_seen": 4297176,
      "step": 7405
    },
    {
      "epoch": 1.1036639857015191,
      "grad_norm": 10.625,
      "learning_rate": 2.7587876079833186e-05,
      "loss": 1.2732,
      "num_input_tokens_seen": 4299992,
      "step": 7410
    },
    {
      "epoch": 1.1044086982424783,
      "grad_norm": 11.9375,
      "learning_rate": 2.7606493893357167e-05,
      "loss": 0.9371,
      "num_input_tokens_seen": 4303320,
      "step": 7415
    },
    {
      "epoch": 1.1051534107834375,
      "grad_norm": 17.75,
      "learning_rate": 2.7625111706881145e-05,
      "loss": 1.0383,
      "num_input_tokens_seen": 4306232,
      "step": 7420
    },
    {
      "epoch": 1.1058981233243967,
      "grad_norm": 8.1875,
      "learning_rate": 2.7643729520405127e-05,
      "loss": 1.2508,
      "num_input_tokens_seen": 4309400,
      "step": 7425
    },
    {
      "epoch": 1.106642835865356,
      "grad_norm": 10.625,
      "learning_rate": 2.7662347333929105e-05,
      "loss": 1.3366,
      "num_input_tokens_seen": 4312376,
      "step": 7430
    },
    {
      "epoch": 1.1073875484063151,
      "grad_norm": 8.375,
      "learning_rate": 2.7680965147453087e-05,
      "loss": 0.9845,
      "num_input_tokens_seen": 4316280,
      "step": 7435
    },
    {
      "epoch": 1.1081322609472744,
      "grad_norm": 12.4375,
      "learning_rate": 2.7699582960977065e-05,
      "loss": 1.2408,
      "num_input_tokens_seen": 4319000,
      "step": 7440
    },
    {
      "epoch": 1.1088769734882336,
      "grad_norm": 7.8125,
      "learning_rate": 2.7718200774501047e-05,
      "loss": 1.2817,
      "num_input_tokens_seen": 4321912,
      "step": 7445
    },
    {
      "epoch": 1.1096216860291928,
      "grad_norm": 7.34375,
      "learning_rate": 2.7736818588025025e-05,
      "loss": 1.0105,
      "num_input_tokens_seen": 4325304,
      "step": 7450
    },
    {
      "epoch": 1.110366398570152,
      "grad_norm": 9.6875,
      "learning_rate": 2.7755436401549e-05,
      "loss": 0.9474,
      "num_input_tokens_seen": 4328280,
      "step": 7455
    },
    {
      "epoch": 1.1111111111111112,
      "grad_norm": 11.9375,
      "learning_rate": 2.777405421507298e-05,
      "loss": 1.0122,
      "num_input_tokens_seen": 4331032,
      "step": 7460
    },
    {
      "epoch": 1.1118558236520704,
      "grad_norm": 15.875,
      "learning_rate": 2.779267202859696e-05,
      "loss": 1.0888,
      "num_input_tokens_seen": 4334104,
      "step": 7465
    },
    {
      "epoch": 1.1126005361930296,
      "grad_norm": 21.0,
      "learning_rate": 2.781128984212094e-05,
      "loss": 1.4163,
      "num_input_tokens_seen": 4336952,
      "step": 7470
    },
    {
      "epoch": 1.1133452487339888,
      "grad_norm": 16.375,
      "learning_rate": 2.782990765564492e-05,
      "loss": 1.1969,
      "num_input_tokens_seen": 4340088,
      "step": 7475
    },
    {
      "epoch": 1.114089961274948,
      "grad_norm": 6.03125,
      "learning_rate": 2.78485254691689e-05,
      "loss": 0.7177,
      "num_input_tokens_seen": 4342872,
      "step": 7480
    },
    {
      "epoch": 1.1148346738159072,
      "grad_norm": 8.75,
      "learning_rate": 2.7867143282692883e-05,
      "loss": 1.3334,
      "num_input_tokens_seen": 4345752,
      "step": 7485
    },
    {
      "epoch": 1.1155793863568662,
      "grad_norm": 10.75,
      "learning_rate": 2.788576109621686e-05,
      "loss": 1.2926,
      "num_input_tokens_seen": 4348376,
      "step": 7490
    },
    {
      "epoch": 1.1163240988978254,
      "grad_norm": 9.625,
      "learning_rate": 2.7904378909740843e-05,
      "loss": 1.2398,
      "num_input_tokens_seen": 4351256,
      "step": 7495
    },
    {
      "epoch": 1.1170688114387846,
      "grad_norm": 11.875,
      "learning_rate": 2.792299672326482e-05,
      "loss": 0.9275,
      "num_input_tokens_seen": 4354232,
      "step": 7500
    },
    {
      "epoch": 1.1178135239797438,
      "grad_norm": 13.25,
      "learning_rate": 2.7941614536788802e-05,
      "loss": 1.3127,
      "num_input_tokens_seen": 4356920,
      "step": 7505
    },
    {
      "epoch": 1.118558236520703,
      "grad_norm": 10.875,
      "learning_rate": 2.796023235031278e-05,
      "loss": 1.0228,
      "num_input_tokens_seen": 4359928,
      "step": 7510
    },
    {
      "epoch": 1.1193029490616622,
      "grad_norm": 9.1875,
      "learning_rate": 2.7978850163836762e-05,
      "loss": 1.2931,
      "num_input_tokens_seen": 4362840,
      "step": 7515
    },
    {
      "epoch": 1.1200476616026214,
      "grad_norm": 9.125,
      "learning_rate": 2.799746797736074e-05,
      "loss": 1.4212,
      "num_input_tokens_seen": 4365624,
      "step": 7520
    },
    {
      "epoch": 1.1207923741435806,
      "grad_norm": 11.375,
      "learning_rate": 2.8016085790884722e-05,
      "loss": 1.1061,
      "num_input_tokens_seen": 4368664,
      "step": 7525
    },
    {
      "epoch": 1.1215370866845398,
      "grad_norm": 15.5,
      "learning_rate": 2.80347036044087e-05,
      "loss": 1.3144,
      "num_input_tokens_seen": 4371736,
      "step": 7530
    },
    {
      "epoch": 1.122281799225499,
      "grad_norm": 7.1875,
      "learning_rate": 2.8053321417932682e-05,
      "loss": 0.9617,
      "num_input_tokens_seen": 4374808,
      "step": 7535
    },
    {
      "epoch": 1.1230265117664582,
      "grad_norm": 6.65625,
      "learning_rate": 2.8071939231456657e-05,
      "loss": 0.8935,
      "num_input_tokens_seen": 4377560,
      "step": 7540
    },
    {
      "epoch": 1.1237712243074174,
      "grad_norm": 5.0625,
      "learning_rate": 2.8090557044980635e-05,
      "loss": 0.9018,
      "num_input_tokens_seen": 4380472,
      "step": 7545
    },
    {
      "epoch": 1.1245159368483766,
      "grad_norm": 11.5625,
      "learning_rate": 2.8109174858504617e-05,
      "loss": 0.9044,
      "num_input_tokens_seen": 4383352,
      "step": 7550
    },
    {
      "epoch": 1.1252606493893358,
      "grad_norm": 18.625,
      "learning_rate": 2.81277926720286e-05,
      "loss": 0.8572,
      "num_input_tokens_seen": 4386072,
      "step": 7555
    },
    {
      "epoch": 1.126005361930295,
      "grad_norm": 14.875,
      "learning_rate": 2.8146410485552577e-05,
      "loss": 1.0171,
      "num_input_tokens_seen": 4388920,
      "step": 7560
    },
    {
      "epoch": 1.1267500744712542,
      "grad_norm": 8.875,
      "learning_rate": 2.8165028299076558e-05,
      "loss": 0.9568,
      "num_input_tokens_seen": 4391864,
      "step": 7565
    },
    {
      "epoch": 1.1274947870122132,
      "grad_norm": 12.375,
      "learning_rate": 2.8183646112600536e-05,
      "loss": 0.9584,
      "num_input_tokens_seen": 4394808,
      "step": 7570
    },
    {
      "epoch": 1.1282394995531724,
      "grad_norm": 11.1875,
      "learning_rate": 2.8202263926124518e-05,
      "loss": 1.0422,
      "num_input_tokens_seen": 4397592,
      "step": 7575
    },
    {
      "epoch": 1.1289842120941316,
      "grad_norm": 12.8125,
      "learning_rate": 2.8220881739648496e-05,
      "loss": 1.6524,
      "num_input_tokens_seen": 4400760,
      "step": 7580
    },
    {
      "epoch": 1.1297289246350908,
      "grad_norm": 9.0,
      "learning_rate": 2.8239499553172478e-05,
      "loss": 0.8202,
      "num_input_tokens_seen": 4403832,
      "step": 7585
    },
    {
      "epoch": 1.13047363717605,
      "grad_norm": 19.125,
      "learning_rate": 2.8258117366696456e-05,
      "loss": 1.2422,
      "num_input_tokens_seen": 4406552,
      "step": 7590
    },
    {
      "epoch": 1.1312183497170092,
      "grad_norm": 9.9375,
      "learning_rate": 2.8276735180220438e-05,
      "loss": 1.0473,
      "num_input_tokens_seen": 4409400,
      "step": 7595
    },
    {
      "epoch": 1.1319630622579684,
      "grad_norm": 6.5,
      "learning_rate": 2.8295352993744416e-05,
      "loss": 1.0403,
      "num_input_tokens_seen": 4412248,
      "step": 7600
    },
    {
      "epoch": 1.1327077747989276,
      "grad_norm": 12.125,
      "learning_rate": 2.8313970807268398e-05,
      "loss": 1.063,
      "num_input_tokens_seen": 4415160,
      "step": 7605
    },
    {
      "epoch": 1.1334524873398868,
      "grad_norm": 20.5,
      "learning_rate": 2.8332588620792376e-05,
      "loss": 1.5821,
      "num_input_tokens_seen": 4418104,
      "step": 7610
    },
    {
      "epoch": 1.134197199880846,
      "grad_norm": 12.0625,
      "learning_rate": 2.8351206434316357e-05,
      "loss": 1.1326,
      "num_input_tokens_seen": 4420952,
      "step": 7615
    },
    {
      "epoch": 1.1349419124218052,
      "grad_norm": 26.25,
      "learning_rate": 2.836982424784034e-05,
      "loss": 1.3437,
      "num_input_tokens_seen": 4423896,
      "step": 7620
    },
    {
      "epoch": 1.1356866249627644,
      "grad_norm": 8.1875,
      "learning_rate": 2.838844206136431e-05,
      "loss": 1.3294,
      "num_input_tokens_seen": 4427000,
      "step": 7625
    },
    {
      "epoch": 1.1364313375037236,
      "grad_norm": 17.375,
      "learning_rate": 2.8407059874888292e-05,
      "loss": 1.2492,
      "num_input_tokens_seen": 4429880,
      "step": 7630
    },
    {
      "epoch": 1.1371760500446828,
      "grad_norm": 19.5,
      "learning_rate": 2.8425677688412274e-05,
      "loss": 1.0252,
      "num_input_tokens_seen": 4433016,
      "step": 7635
    },
    {
      "epoch": 1.137920762585642,
      "grad_norm": 19.375,
      "learning_rate": 2.8444295501936252e-05,
      "loss": 1.3764,
      "num_input_tokens_seen": 4436184,
      "step": 7640
    },
    {
      "epoch": 1.1386654751266012,
      "grad_norm": 11.5,
      "learning_rate": 2.8462913315460234e-05,
      "loss": 1.2976,
      "num_input_tokens_seen": 4439256,
      "step": 7645
    },
    {
      "epoch": 1.1394101876675604,
      "grad_norm": 6.6875,
      "learning_rate": 2.8481531128984212e-05,
      "loss": 1.0167,
      "num_input_tokens_seen": 4442584,
      "step": 7650
    },
    {
      "epoch": 1.1401549002085196,
      "grad_norm": 8.125,
      "learning_rate": 2.8500148942508193e-05,
      "loss": 0.866,
      "num_input_tokens_seen": 4445464,
      "step": 7655
    },
    {
      "epoch": 1.1408996127494788,
      "grad_norm": 12.0,
      "learning_rate": 2.8518766756032172e-05,
      "loss": 0.8685,
      "num_input_tokens_seen": 4448184,
      "step": 7660
    },
    {
      "epoch": 1.1416443252904378,
      "grad_norm": 18.375,
      "learning_rate": 2.8537384569556153e-05,
      "loss": 1.147,
      "num_input_tokens_seen": 4451608,
      "step": 7665
    },
    {
      "epoch": 1.142389037831397,
      "grad_norm": 8.125,
      "learning_rate": 2.855600238308013e-05,
      "loss": 0.967,
      "num_input_tokens_seen": 4454296,
      "step": 7670
    },
    {
      "epoch": 1.1431337503723562,
      "grad_norm": 15.1875,
      "learning_rate": 2.8574620196604113e-05,
      "loss": 0.9968,
      "num_input_tokens_seen": 4457208,
      "step": 7675
    },
    {
      "epoch": 1.1438784629133154,
      "grad_norm": 13.125,
      "learning_rate": 2.859323801012809e-05,
      "loss": 0.8863,
      "num_input_tokens_seen": 4460504,
      "step": 7680
    },
    {
      "epoch": 1.1446231754542746,
      "grad_norm": 15.5625,
      "learning_rate": 2.8611855823652073e-05,
      "loss": 1.2215,
      "num_input_tokens_seen": 4463320,
      "step": 7685
    },
    {
      "epoch": 1.1453678879952338,
      "grad_norm": 11.8125,
      "learning_rate": 2.863047363717605e-05,
      "loss": 1.0976,
      "num_input_tokens_seen": 4466232,
      "step": 7690
    },
    {
      "epoch": 1.146112600536193,
      "grad_norm": 18.875,
      "learning_rate": 2.8649091450700033e-05,
      "loss": 1.162,
      "num_input_tokens_seen": 4469240,
      "step": 7695
    },
    {
      "epoch": 1.1468573130771522,
      "grad_norm": 11.125,
      "learning_rate": 2.8667709264224015e-05,
      "loss": 0.7708,
      "num_input_tokens_seen": 4472024,
      "step": 7700
    },
    {
      "epoch": 1.1476020256181114,
      "grad_norm": 10.0625,
      "learning_rate": 2.8686327077747993e-05,
      "loss": 0.8463,
      "num_input_tokens_seen": 4475128,
      "step": 7705
    },
    {
      "epoch": 1.1483467381590706,
      "grad_norm": 19.25,
      "learning_rate": 2.8704944891271968e-05,
      "loss": 1.1647,
      "num_input_tokens_seen": 4478168,
      "step": 7710
    },
    {
      "epoch": 1.1490914507000298,
      "grad_norm": 10.0,
      "learning_rate": 2.872356270479595e-05,
      "loss": 0.9836,
      "num_input_tokens_seen": 4481080,
      "step": 7715
    },
    {
      "epoch": 1.149836163240989,
      "grad_norm": 8.75,
      "learning_rate": 2.8742180518319927e-05,
      "loss": 1.1937,
      "num_input_tokens_seen": 4483992,
      "step": 7720
    },
    {
      "epoch": 1.1505808757819482,
      "grad_norm": 11.1875,
      "learning_rate": 2.876079833184391e-05,
      "loss": 1.2176,
      "num_input_tokens_seen": 4487192,
      "step": 7725
    },
    {
      "epoch": 1.1513255883229074,
      "grad_norm": 14.8125,
      "learning_rate": 2.8779416145367887e-05,
      "loss": 1.1343,
      "num_input_tokens_seen": 4490552,
      "step": 7730
    },
    {
      "epoch": 1.1520703008638666,
      "grad_norm": 7.15625,
      "learning_rate": 2.879803395889187e-05,
      "loss": 1.112,
      "num_input_tokens_seen": 4493464,
      "step": 7735
    },
    {
      "epoch": 1.1528150134048256,
      "grad_norm": 9.25,
      "learning_rate": 2.8816651772415847e-05,
      "loss": 1.1278,
      "num_input_tokens_seen": 4496216,
      "step": 7740
    },
    {
      "epoch": 1.1535597259457848,
      "grad_norm": 14.75,
      "learning_rate": 2.883526958593983e-05,
      "loss": 1.0333,
      "num_input_tokens_seen": 4499160,
      "step": 7745
    },
    {
      "epoch": 1.154304438486744,
      "grad_norm": 8.875,
      "learning_rate": 2.8853887399463807e-05,
      "loss": 1.1554,
      "num_input_tokens_seen": 4501912,
      "step": 7750
    },
    {
      "epoch": 1.1550491510277032,
      "grad_norm": 15.0625,
      "learning_rate": 2.887250521298779e-05,
      "loss": 1.1282,
      "num_input_tokens_seen": 4504568,
      "step": 7755
    },
    {
      "epoch": 1.1557938635686624,
      "grad_norm": 11.0,
      "learning_rate": 2.8891123026511767e-05,
      "loss": 1.3064,
      "num_input_tokens_seen": 4507320,
      "step": 7760
    },
    {
      "epoch": 1.1565385761096216,
      "grad_norm": 16.875,
      "learning_rate": 2.890974084003575e-05,
      "loss": 1.0316,
      "num_input_tokens_seen": 4509976,
      "step": 7765
    },
    {
      "epoch": 1.1572832886505808,
      "grad_norm": 9.375,
      "learning_rate": 2.892835865355973e-05,
      "loss": 0.991,
      "num_input_tokens_seen": 4512760,
      "step": 7770
    },
    {
      "epoch": 1.15802800119154,
      "grad_norm": 11.75,
      "learning_rate": 2.894697646708371e-05,
      "loss": 1.2466,
      "num_input_tokens_seen": 4515928,
      "step": 7775
    },
    {
      "epoch": 1.1587727137324992,
      "grad_norm": 21.375,
      "learning_rate": 2.896559428060769e-05,
      "loss": 1.4356,
      "num_input_tokens_seen": 4518872,
      "step": 7780
    },
    {
      "epoch": 1.1595174262734584,
      "grad_norm": 6.96875,
      "learning_rate": 2.8984212094131668e-05,
      "loss": 1.2628,
      "num_input_tokens_seen": 4521784,
      "step": 7785
    },
    {
      "epoch": 1.1602621388144176,
      "grad_norm": 11.6875,
      "learning_rate": 2.900282990765565e-05,
      "loss": 1.148,
      "num_input_tokens_seen": 4524824,
      "step": 7790
    },
    {
      "epoch": 1.1610068513553768,
      "grad_norm": 3.578125,
      "learning_rate": 2.9021447721179628e-05,
      "loss": 0.9215,
      "num_input_tokens_seen": 4527640,
      "step": 7795
    },
    {
      "epoch": 1.161751563896336,
      "grad_norm": 8.625,
      "learning_rate": 2.9040065534703603e-05,
      "loss": 1.267,
      "num_input_tokens_seen": 4530392,
      "step": 7800
    },
    {
      "epoch": 1.1624962764372953,
      "grad_norm": 14.8125,
      "learning_rate": 2.9058683348227584e-05,
      "loss": 0.8691,
      "num_input_tokens_seen": 4533016,
      "step": 7805
    },
    {
      "epoch": 1.1632409889782545,
      "grad_norm": 12.3125,
      "learning_rate": 2.9077301161751563e-05,
      "loss": 0.9959,
      "num_input_tokens_seen": 4536120,
      "step": 7810
    },
    {
      "epoch": 1.1639857015192137,
      "grad_norm": 14.125,
      "learning_rate": 2.9095918975275544e-05,
      "loss": 1.0868,
      "num_input_tokens_seen": 4539160,
      "step": 7815
    },
    {
      "epoch": 1.1647304140601729,
      "grad_norm": 13.8125,
      "learning_rate": 2.9114536788799523e-05,
      "loss": 1.0995,
      "num_input_tokens_seen": 4542232,
      "step": 7820
    },
    {
      "epoch": 1.165475126601132,
      "grad_norm": 18.25,
      "learning_rate": 2.9133154602323504e-05,
      "loss": 1.0032,
      "num_input_tokens_seen": 4545176,
      "step": 7825
    },
    {
      "epoch": 1.1662198391420913,
      "grad_norm": 10.375,
      "learning_rate": 2.9151772415847482e-05,
      "loss": 1.0152,
      "num_input_tokens_seen": 4548248,
      "step": 7830
    },
    {
      "epoch": 1.1669645516830505,
      "grad_norm": 15.1875,
      "learning_rate": 2.9170390229371464e-05,
      "loss": 0.9009,
      "num_input_tokens_seen": 4551224,
      "step": 7835
    },
    {
      "epoch": 1.1677092642240094,
      "grad_norm": 9.75,
      "learning_rate": 2.9189008042895442e-05,
      "loss": 0.9141,
      "num_input_tokens_seen": 4553944,
      "step": 7840
    },
    {
      "epoch": 1.1684539767649686,
      "grad_norm": 10.9375,
      "learning_rate": 2.9207625856419424e-05,
      "loss": 1.1882,
      "num_input_tokens_seen": 4556888,
      "step": 7845
    },
    {
      "epoch": 1.1691986893059279,
      "grad_norm": 13.8125,
      "learning_rate": 2.9226243669943406e-05,
      "loss": 1.3283,
      "num_input_tokens_seen": 4559704,
      "step": 7850
    },
    {
      "epoch": 1.169943401846887,
      "grad_norm": 10.9375,
      "learning_rate": 2.9244861483467384e-05,
      "loss": 1.1313,
      "num_input_tokens_seen": 4562456,
      "step": 7855
    },
    {
      "epoch": 1.1706881143878463,
      "grad_norm": 11.75,
      "learning_rate": 2.9263479296991365e-05,
      "loss": 1.0842,
      "num_input_tokens_seen": 4565240,
      "step": 7860
    },
    {
      "epoch": 1.1714328269288055,
      "grad_norm": 10.0625,
      "learning_rate": 2.9282097110515344e-05,
      "loss": 0.9168,
      "num_input_tokens_seen": 4568440,
      "step": 7865
    },
    {
      "epoch": 1.1721775394697647,
      "grad_norm": 17.375,
      "learning_rate": 2.9300714924039325e-05,
      "loss": 1.0489,
      "num_input_tokens_seen": 4571512,
      "step": 7870
    },
    {
      "epoch": 1.1729222520107239,
      "grad_norm": 10.6875,
      "learning_rate": 2.9319332737563303e-05,
      "loss": 1.227,
      "num_input_tokens_seen": 4574392,
      "step": 7875
    },
    {
      "epoch": 1.173666964551683,
      "grad_norm": 11.75,
      "learning_rate": 2.9337950551087285e-05,
      "loss": 0.8589,
      "num_input_tokens_seen": 4577304,
      "step": 7880
    },
    {
      "epoch": 1.1744116770926423,
      "grad_norm": 13.875,
      "learning_rate": 2.935656836461126e-05,
      "loss": 1.1366,
      "num_input_tokens_seen": 4580440,
      "step": 7885
    },
    {
      "epoch": 1.1751563896336015,
      "grad_norm": 12.375,
      "learning_rate": 2.9375186178135238e-05,
      "loss": 1.3813,
      "num_input_tokens_seen": 4583320,
      "step": 7890
    },
    {
      "epoch": 1.1759011021745607,
      "grad_norm": 12.0,
      "learning_rate": 2.939380399165922e-05,
      "loss": 1.1611,
      "num_input_tokens_seen": 4586296,
      "step": 7895
    },
    {
      "epoch": 1.1766458147155199,
      "grad_norm": 11.3125,
      "learning_rate": 2.9412421805183198e-05,
      "loss": 1.3475,
      "num_input_tokens_seen": 4589240,
      "step": 7900
    },
    {
      "epoch": 1.177390527256479,
      "grad_norm": 9.3125,
      "learning_rate": 2.943103961870718e-05,
      "loss": 1.1715,
      "num_input_tokens_seen": 4592248,
      "step": 7905
    },
    {
      "epoch": 1.1781352397974383,
      "grad_norm": 12.125,
      "learning_rate": 2.9449657432231158e-05,
      "loss": 0.816,
      "num_input_tokens_seen": 4595192,
      "step": 7910
    },
    {
      "epoch": 1.1788799523383973,
      "grad_norm": 13.0625,
      "learning_rate": 2.946827524575514e-05,
      "loss": 1.1514,
      "num_input_tokens_seen": 4597912,
      "step": 7915
    },
    {
      "epoch": 1.1796246648793565,
      "grad_norm": 14.5,
      "learning_rate": 2.9486893059279118e-05,
      "loss": 0.9244,
      "num_input_tokens_seen": 4600696,
      "step": 7920
    },
    {
      "epoch": 1.1803693774203157,
      "grad_norm": 8.1875,
      "learning_rate": 2.95055108728031e-05,
      "loss": 1.2915,
      "num_input_tokens_seen": 4603704,
      "step": 7925
    },
    {
      "epoch": 1.1811140899612749,
      "grad_norm": 5.90625,
      "learning_rate": 2.952412868632708e-05,
      "loss": 0.8808,
      "num_input_tokens_seen": 4606424,
      "step": 7930
    },
    {
      "epoch": 1.181858802502234,
      "grad_norm": 7.125,
      "learning_rate": 2.954274649985106e-05,
      "loss": 1.0033,
      "num_input_tokens_seen": 4609240,
      "step": 7935
    },
    {
      "epoch": 1.1826035150431933,
      "grad_norm": 10.9375,
      "learning_rate": 2.956136431337504e-05,
      "loss": 1.0103,
      "num_input_tokens_seen": 4612088,
      "step": 7940
    },
    {
      "epoch": 1.1833482275841525,
      "grad_norm": 11.375,
      "learning_rate": 2.957998212689902e-05,
      "loss": 0.9963,
      "num_input_tokens_seen": 4614744,
      "step": 7945
    },
    {
      "epoch": 1.1840929401251117,
      "grad_norm": 10.25,
      "learning_rate": 2.9598599940423e-05,
      "loss": 1.1037,
      "num_input_tokens_seen": 4617400,
      "step": 7950
    },
    {
      "epoch": 1.1848376526660709,
      "grad_norm": 14.125,
      "learning_rate": 2.961721775394698e-05,
      "loss": 0.7543,
      "num_input_tokens_seen": 4620344,
      "step": 7955
    },
    {
      "epoch": 1.18558236520703,
      "grad_norm": 27.875,
      "learning_rate": 2.963583556747096e-05,
      "loss": 1.2082,
      "num_input_tokens_seen": 4623384,
      "step": 7960
    },
    {
      "epoch": 1.1863270777479893,
      "grad_norm": 22.125,
      "learning_rate": 2.965445338099494e-05,
      "loss": 1.3112,
      "num_input_tokens_seen": 4626424,
      "step": 7965
    },
    {
      "epoch": 1.1870717902889485,
      "grad_norm": 16.0,
      "learning_rate": 2.9673071194518914e-05,
      "loss": 1.3613,
      "num_input_tokens_seen": 4628984,
      "step": 7970
    },
    {
      "epoch": 1.1878165028299077,
      "grad_norm": 6.65625,
      "learning_rate": 2.9691689008042895e-05,
      "loss": 0.7097,
      "num_input_tokens_seen": 4631864,
      "step": 7975
    },
    {
      "epoch": 1.188561215370867,
      "grad_norm": 8.125,
      "learning_rate": 2.9710306821566873e-05,
      "loss": 1.0532,
      "num_input_tokens_seen": 4634808,
      "step": 7980
    },
    {
      "epoch": 1.189305927911826,
      "grad_norm": 10.6875,
      "learning_rate": 2.9728924635090855e-05,
      "loss": 1.449,
      "num_input_tokens_seen": 4637560,
      "step": 7985
    },
    {
      "epoch": 1.1900506404527853,
      "grad_norm": 9.625,
      "learning_rate": 2.9747542448614833e-05,
      "loss": 1.0194,
      "num_input_tokens_seen": 4640728,
      "step": 7990
    },
    {
      "epoch": 1.1907953529937445,
      "grad_norm": 13.375,
      "learning_rate": 2.9766160262138815e-05,
      "loss": 0.9525,
      "num_input_tokens_seen": 4643384,
      "step": 7995
    },
    {
      "epoch": 1.1915400655347037,
      "grad_norm": 18.875,
      "learning_rate": 2.9784778075662793e-05,
      "loss": 1.1864,
      "num_input_tokens_seen": 4646040,
      "step": 8000
    },
    {
      "epoch": 1.192284778075663,
      "grad_norm": 11.6875,
      "learning_rate": 2.9803395889186775e-05,
      "loss": 1.157,
      "num_input_tokens_seen": 4648728,
      "step": 8005
    },
    {
      "epoch": 1.193029490616622,
      "grad_norm": 7.78125,
      "learning_rate": 2.9822013702710756e-05,
      "loss": 0.9542,
      "num_input_tokens_seen": 4651672,
      "step": 8010
    },
    {
      "epoch": 1.193774203157581,
      "grad_norm": 17.125,
      "learning_rate": 2.9840631516234735e-05,
      "loss": 0.8647,
      "num_input_tokens_seen": 4654520,
      "step": 8015
    },
    {
      "epoch": 1.1945189156985403,
      "grad_norm": 7.59375,
      "learning_rate": 2.9859249329758716e-05,
      "loss": 1.1893,
      "num_input_tokens_seen": 4657432,
      "step": 8020
    },
    {
      "epoch": 1.1952636282394995,
      "grad_norm": 7.8125,
      "learning_rate": 2.9877867143282694e-05,
      "loss": 0.9956,
      "num_input_tokens_seen": 4660280,
      "step": 8025
    },
    {
      "epoch": 1.1960083407804587,
      "grad_norm": 16.625,
      "learning_rate": 2.9896484956806676e-05,
      "loss": 1.0244,
      "num_input_tokens_seen": 4663032,
      "step": 8030
    },
    {
      "epoch": 1.196753053321418,
      "grad_norm": 12.5625,
      "learning_rate": 2.9915102770330654e-05,
      "loss": 1.0459,
      "num_input_tokens_seen": 4666136,
      "step": 8035
    },
    {
      "epoch": 1.197497765862377,
      "grad_norm": 8.3125,
      "learning_rate": 2.9933720583854636e-05,
      "loss": 1.1533,
      "num_input_tokens_seen": 4668952,
      "step": 8040
    },
    {
      "epoch": 1.1982424784033363,
      "grad_norm": 5.1875,
      "learning_rate": 2.9952338397378614e-05,
      "loss": 1.2836,
      "num_input_tokens_seen": 4671928,
      "step": 8045
    },
    {
      "epoch": 1.1989871909442955,
      "grad_norm": 6.21875,
      "learning_rate": 2.9970956210902596e-05,
      "loss": 1.1179,
      "num_input_tokens_seen": 4674616,
      "step": 8050
    },
    {
      "epoch": 1.1997319034852547,
      "grad_norm": 9.875,
      "learning_rate": 2.9989574024426574e-05,
      "loss": 1.1995,
      "num_input_tokens_seen": 4677368,
      "step": 8055
    },
    {
      "epoch": 1.200476616026214,
      "grad_norm": 17.875,
      "learning_rate": 3.000819183795055e-05,
      "loss": 0.7215,
      "num_input_tokens_seen": 4679992,
      "step": 8060
    },
    {
      "epoch": 1.2012213285671731,
      "grad_norm": 11.5625,
      "learning_rate": 3.002680965147453e-05,
      "loss": 1.2214,
      "num_input_tokens_seen": 4683000,
      "step": 8065
    },
    {
      "epoch": 1.2019660411081323,
      "grad_norm": 16.625,
      "learning_rate": 3.004542746499851e-05,
      "loss": 1.0861,
      "num_input_tokens_seen": 4685848,
      "step": 8070
    },
    {
      "epoch": 1.2027107536490915,
      "grad_norm": 16.5,
      "learning_rate": 3.006404527852249e-05,
      "loss": 1.1577,
      "num_input_tokens_seen": 4688664,
      "step": 8075
    },
    {
      "epoch": 1.2034554661900507,
      "grad_norm": 15.5,
      "learning_rate": 3.008266309204647e-05,
      "loss": 0.9382,
      "num_input_tokens_seen": 4691640,
      "step": 8080
    },
    {
      "epoch": 1.2042001787310097,
      "grad_norm": 23.25,
      "learning_rate": 3.010128090557045e-05,
      "loss": 1.6578,
      "num_input_tokens_seen": 4694520,
      "step": 8085
    },
    {
      "epoch": 1.204944891271969,
      "grad_norm": 6.625,
      "learning_rate": 3.0119898719094432e-05,
      "loss": 0.987,
      "num_input_tokens_seen": 4697208,
      "step": 8090
    },
    {
      "epoch": 1.2056896038129281,
      "grad_norm": 9.375,
      "learning_rate": 3.013851653261841e-05,
      "loss": 0.9464,
      "num_input_tokens_seen": 4700184,
      "step": 8095
    },
    {
      "epoch": 1.2064343163538873,
      "grad_norm": 8.25,
      "learning_rate": 3.015713434614239e-05,
      "loss": 0.9351,
      "num_input_tokens_seen": 4703224,
      "step": 8100
    },
    {
      "epoch": 1.2071790288948465,
      "grad_norm": 6.53125,
      "learning_rate": 3.017575215966637e-05,
      "loss": 1.3051,
      "num_input_tokens_seen": 4705816,
      "step": 8105
    },
    {
      "epoch": 1.2079237414358057,
      "grad_norm": 9.5625,
      "learning_rate": 3.019436997319035e-05,
      "loss": 1.0498,
      "num_input_tokens_seen": 4708600,
      "step": 8110
    },
    {
      "epoch": 1.208668453976765,
      "grad_norm": 15.625,
      "learning_rate": 3.021298778671433e-05,
      "loss": 1.3938,
      "num_input_tokens_seen": 4711416,
      "step": 8115
    },
    {
      "epoch": 1.2094131665177241,
      "grad_norm": 14.5625,
      "learning_rate": 3.023160560023831e-05,
      "loss": 0.9807,
      "num_input_tokens_seen": 4714392,
      "step": 8120
    },
    {
      "epoch": 1.2101578790586833,
      "grad_norm": 15.875,
      "learning_rate": 3.025022341376229e-05,
      "loss": 1.072,
      "num_input_tokens_seen": 4717432,
      "step": 8125
    },
    {
      "epoch": 1.2109025915996425,
      "grad_norm": 14.0625,
      "learning_rate": 3.026884122728627e-05,
      "loss": 0.9257,
      "num_input_tokens_seen": 4720632,
      "step": 8130
    },
    {
      "epoch": 1.2116473041406017,
      "grad_norm": 7.5625,
      "learning_rate": 3.028745904081025e-05,
      "loss": 1.2595,
      "num_input_tokens_seen": 4723384,
      "step": 8135
    },
    {
      "epoch": 1.212392016681561,
      "grad_norm": 8.5,
      "learning_rate": 3.030607685433423e-05,
      "loss": 1.4622,
      "num_input_tokens_seen": 4726232,
      "step": 8140
    },
    {
      "epoch": 1.2131367292225201,
      "grad_norm": 13.3125,
      "learning_rate": 3.0324694667858206e-05,
      "loss": 0.9629,
      "num_input_tokens_seen": 4728952,
      "step": 8145
    },
    {
      "epoch": 1.2138814417634793,
      "grad_norm": 25.75,
      "learning_rate": 3.0343312481382184e-05,
      "loss": 1.4242,
      "num_input_tokens_seen": 4731672,
      "step": 8150
    },
    {
      "epoch": 1.2146261543044385,
      "grad_norm": 12.25,
      "learning_rate": 3.0361930294906166e-05,
      "loss": 1.385,
      "num_input_tokens_seen": 4734776,
      "step": 8155
    },
    {
      "epoch": 1.2153708668453977,
      "grad_norm": 30.5,
      "learning_rate": 3.0380548108430147e-05,
      "loss": 1.3401,
      "num_input_tokens_seen": 4737592,
      "step": 8160
    },
    {
      "epoch": 1.216115579386357,
      "grad_norm": 16.375,
      "learning_rate": 3.0399165921954126e-05,
      "loss": 0.9031,
      "num_input_tokens_seen": 4740568,
      "step": 8165
    },
    {
      "epoch": 1.2168602919273162,
      "grad_norm": 6.65625,
      "learning_rate": 3.0417783735478107e-05,
      "loss": 1.2131,
      "num_input_tokens_seen": 4743736,
      "step": 8170
    },
    {
      "epoch": 1.2176050044682754,
      "grad_norm": 15.375,
      "learning_rate": 3.0436401549002085e-05,
      "loss": 1.2684,
      "num_input_tokens_seen": 4746616,
      "step": 8175
    },
    {
      "epoch": 1.2183497170092346,
      "grad_norm": 17.125,
      "learning_rate": 3.0455019362526067e-05,
      "loss": 1.3338,
      "num_input_tokens_seen": 4749432,
      "step": 8180
    },
    {
      "epoch": 1.2190944295501935,
      "grad_norm": 10.0,
      "learning_rate": 3.0473637176050045e-05,
      "loss": 1.1184,
      "num_input_tokens_seen": 4752376,
      "step": 8185
    },
    {
      "epoch": 1.2198391420911527,
      "grad_norm": 7.84375,
      "learning_rate": 3.0492254989574027e-05,
      "loss": 0.8768,
      "num_input_tokens_seen": 4755192,
      "step": 8190
    },
    {
      "epoch": 1.220583854632112,
      "grad_norm": 5.71875,
      "learning_rate": 3.0510872803098005e-05,
      "loss": 1.2955,
      "num_input_tokens_seen": 4757944,
      "step": 8195
    },
    {
      "epoch": 1.2213285671730711,
      "grad_norm": 17.875,
      "learning_rate": 3.052949061662199e-05,
      "loss": 1.1303,
      "num_input_tokens_seen": 4760792,
      "step": 8200
    },
    {
      "epoch": 1.2220732797140303,
      "grad_norm": 5.6875,
      "learning_rate": 3.0548108430145965e-05,
      "loss": 1.0419,
      "num_input_tokens_seen": 4763768,
      "step": 8205
    },
    {
      "epoch": 1.2228179922549895,
      "grad_norm": 12.0,
      "learning_rate": 3.056672624366994e-05,
      "loss": 1.2327,
      "num_input_tokens_seen": 4766520,
      "step": 8210
    },
    {
      "epoch": 1.2235627047959488,
      "grad_norm": 24.75,
      "learning_rate": 3.058534405719393e-05,
      "loss": 0.7241,
      "num_input_tokens_seen": 4769464,
      "step": 8215
    },
    {
      "epoch": 1.224307417336908,
      "grad_norm": 5.625,
      "learning_rate": 3.0603961870717907e-05,
      "loss": 0.8173,
      "num_input_tokens_seen": 4772184,
      "step": 8220
    },
    {
      "epoch": 1.2250521298778672,
      "grad_norm": 19.625,
      "learning_rate": 3.0622579684241885e-05,
      "loss": 1.3909,
      "num_input_tokens_seen": 4775096,
      "step": 8225
    },
    {
      "epoch": 1.2257968424188264,
      "grad_norm": 8.875,
      "learning_rate": 3.064119749776586e-05,
      "loss": 1.2673,
      "num_input_tokens_seen": 4777976,
      "step": 8230
    },
    {
      "epoch": 1.2265415549597856,
      "grad_norm": 15.6875,
      "learning_rate": 3.065981531128984e-05,
      "loss": 1.1078,
      "num_input_tokens_seen": 4781208,
      "step": 8235
    },
    {
      "epoch": 1.2272862675007448,
      "grad_norm": 13.375,
      "learning_rate": 3.067843312481382e-05,
      "loss": 1.1733,
      "num_input_tokens_seen": 4783992,
      "step": 8240
    },
    {
      "epoch": 1.228030980041704,
      "grad_norm": 10.625,
      "learning_rate": 3.0697050938337804e-05,
      "loss": 1.3081,
      "num_input_tokens_seen": 4787032,
      "step": 8245
    },
    {
      "epoch": 1.2287756925826632,
      "grad_norm": 7.4375,
      "learning_rate": 3.071566875186178e-05,
      "loss": 1.1546,
      "num_input_tokens_seen": 4789976,
      "step": 8250
    },
    {
      "epoch": 1.2295204051236224,
      "grad_norm": 18.125,
      "learning_rate": 3.073428656538576e-05,
      "loss": 1.2974,
      "num_input_tokens_seen": 4792696,
      "step": 8255
    },
    {
      "epoch": 1.2302651176645814,
      "grad_norm": 8.875,
      "learning_rate": 3.075290437890974e-05,
      "loss": 0.8592,
      "num_input_tokens_seen": 4795608,
      "step": 8260
    },
    {
      "epoch": 1.2310098302055406,
      "grad_norm": 12.0625,
      "learning_rate": 3.0771522192433724e-05,
      "loss": 1.081,
      "num_input_tokens_seen": 4798520,
      "step": 8265
    },
    {
      "epoch": 1.2317545427464998,
      "grad_norm": 8.5,
      "learning_rate": 3.07901400059577e-05,
      "loss": 1.2043,
      "num_input_tokens_seen": 4801624,
      "step": 8270
    },
    {
      "epoch": 1.232499255287459,
      "grad_norm": 14.0,
      "learning_rate": 3.080875781948168e-05,
      "loss": 1.2685,
      "num_input_tokens_seen": 4804952,
      "step": 8275
    },
    {
      "epoch": 1.2332439678284182,
      "grad_norm": 7.21875,
      "learning_rate": 3.082737563300566e-05,
      "loss": 1.0139,
      "num_input_tokens_seen": 4807736,
      "step": 8280
    },
    {
      "epoch": 1.2339886803693774,
      "grad_norm": 11.0625,
      "learning_rate": 3.0845993446529644e-05,
      "loss": 1.4345,
      "num_input_tokens_seen": 4810424,
      "step": 8285
    },
    {
      "epoch": 1.2347333929103366,
      "grad_norm": 7.0,
      "learning_rate": 3.086461126005362e-05,
      "loss": 1.1841,
      "num_input_tokens_seen": 4813720,
      "step": 8290
    },
    {
      "epoch": 1.2354781054512958,
      "grad_norm": 12.125,
      "learning_rate": 3.08832290735776e-05,
      "loss": 1.2302,
      "num_input_tokens_seen": 4816664,
      "step": 8295
    },
    {
      "epoch": 1.236222817992255,
      "grad_norm": 11.0,
      "learning_rate": 3.0901846887101585e-05,
      "loss": 1.2807,
      "num_input_tokens_seen": 4819544,
      "step": 8300
    },
    {
      "epoch": 1.2369675305332142,
      "grad_norm": 11.6875,
      "learning_rate": 3.0920464700625564e-05,
      "loss": 1.0809,
      "num_input_tokens_seen": 4822616,
      "step": 8305
    },
    {
      "epoch": 1.2377122430741734,
      "grad_norm": 17.875,
      "learning_rate": 3.093908251414954e-05,
      "loss": 1.8887,
      "num_input_tokens_seen": 4825752,
      "step": 8310
    },
    {
      "epoch": 1.2384569556151326,
      "grad_norm": 7.46875,
      "learning_rate": 3.095770032767352e-05,
      "loss": 0.7813,
      "num_input_tokens_seen": 4828920,
      "step": 8315
    },
    {
      "epoch": 1.2392016681560918,
      "grad_norm": 16.25,
      "learning_rate": 3.09763181411975e-05,
      "loss": 1.2334,
      "num_input_tokens_seen": 4831672,
      "step": 8320
    },
    {
      "epoch": 1.239946380697051,
      "grad_norm": 8.3125,
      "learning_rate": 3.0994935954721477e-05,
      "loss": 0.9903,
      "num_input_tokens_seen": 4834424,
      "step": 8325
    },
    {
      "epoch": 1.2406910932380102,
      "grad_norm": 13.1875,
      "learning_rate": 3.1013553768245455e-05,
      "loss": 1.2342,
      "num_input_tokens_seen": 4837496,
      "step": 8330
    },
    {
      "epoch": 1.2414358057789694,
      "grad_norm": 10.0625,
      "learning_rate": 3.103217158176944e-05,
      "loss": 1.4873,
      "num_input_tokens_seen": 4840440,
      "step": 8335
    },
    {
      "epoch": 1.2421805183199286,
      "grad_norm": 14.5,
      "learning_rate": 3.105078939529342e-05,
      "loss": 1.2938,
      "num_input_tokens_seen": 4843032,
      "step": 8340
    },
    {
      "epoch": 1.2429252308608878,
      "grad_norm": 9.625,
      "learning_rate": 3.1069407208817396e-05,
      "loss": 1.1523,
      "num_input_tokens_seen": 4845848,
      "step": 8345
    },
    {
      "epoch": 1.243669943401847,
      "grad_norm": 8.0625,
      "learning_rate": 3.1088025022341374e-05,
      "loss": 0.6831,
      "num_input_tokens_seen": 4848824,
      "step": 8350
    },
    {
      "epoch": 1.244414655942806,
      "grad_norm": 12.8125,
      "learning_rate": 3.110664283586536e-05,
      "loss": 1.0126,
      "num_input_tokens_seen": 4851608,
      "step": 8355
    },
    {
      "epoch": 1.2451593684837652,
      "grad_norm": 12.125,
      "learning_rate": 3.112526064938934e-05,
      "loss": 1.2354,
      "num_input_tokens_seen": 4854840,
      "step": 8360
    },
    {
      "epoch": 1.2459040810247244,
      "grad_norm": 5.53125,
      "learning_rate": 3.1143878462913316e-05,
      "loss": 1.2661,
      "num_input_tokens_seen": 4857912,
      "step": 8365
    },
    {
      "epoch": 1.2466487935656836,
      "grad_norm": 11.8125,
      "learning_rate": 3.1162496276437294e-05,
      "loss": 1.2791,
      "num_input_tokens_seen": 4860696,
      "step": 8370
    },
    {
      "epoch": 1.2473935061066428,
      "grad_norm": 12.4375,
      "learning_rate": 3.118111408996128e-05,
      "loss": 1.2249,
      "num_input_tokens_seen": 4863736,
      "step": 8375
    },
    {
      "epoch": 1.248138218647602,
      "grad_norm": 6.125,
      "learning_rate": 3.119973190348526e-05,
      "loss": 0.9405,
      "num_input_tokens_seen": 4866488,
      "step": 8380
    },
    {
      "epoch": 1.2488829311885612,
      "grad_norm": 14.75,
      "learning_rate": 3.1218349717009236e-05,
      "loss": 1.3764,
      "num_input_tokens_seen": 4869720,
      "step": 8385
    },
    {
      "epoch": 1.2496276437295204,
      "grad_norm": 14.625,
      "learning_rate": 3.123696753053322e-05,
      "loss": 1.1117,
      "num_input_tokens_seen": 4872792,
      "step": 8390
    },
    {
      "epoch": 1.2503723562704796,
      "grad_norm": 10.5625,
      "learning_rate": 3.12555853440572e-05,
      "loss": 1.2537,
      "num_input_tokens_seen": 4876120,
      "step": 8395
    },
    {
      "epoch": 1.2511170688114388,
      "grad_norm": 5.65625,
      "learning_rate": 3.127420315758118e-05,
      "loss": 0.7991,
      "num_input_tokens_seen": 4879000,
      "step": 8400
    },
    {
      "epoch": 1.251861781352398,
      "grad_norm": 17.875,
      "learning_rate": 3.1292820971105155e-05,
      "loss": 0.9032,
      "num_input_tokens_seen": 4881560,
      "step": 8405
    },
    {
      "epoch": 1.2526064938933572,
      "grad_norm": 14.375,
      "learning_rate": 3.1311438784629134e-05,
      "loss": 1.1191,
      "num_input_tokens_seen": 4884440,
      "step": 8410
    },
    {
      "epoch": 1.2533512064343164,
      "grad_norm": 9.375,
      "learning_rate": 3.133005659815311e-05,
      "loss": 1.3341,
      "num_input_tokens_seen": 4887160,
      "step": 8415
    },
    {
      "epoch": 1.2540959189752756,
      "grad_norm": 18.125,
      "learning_rate": 3.134867441167709e-05,
      "loss": 1.2421,
      "num_input_tokens_seen": 4890424,
      "step": 8420
    },
    {
      "epoch": 1.2548406315162346,
      "grad_norm": 13.125,
      "learning_rate": 3.1367292225201075e-05,
      "loss": 1.039,
      "num_input_tokens_seen": 4893144,
      "step": 8425
    },
    {
      "epoch": 1.2555853440571938,
      "grad_norm": 10.25,
      "learning_rate": 3.138591003872505e-05,
      "loss": 1.146,
      "num_input_tokens_seen": 4896120,
      "step": 8430
    },
    {
      "epoch": 1.256330056598153,
      "grad_norm": 13.125,
      "learning_rate": 3.140452785224903e-05,
      "loss": 1.5552,
      "num_input_tokens_seen": 4898968,
      "step": 8435
    },
    {
      "epoch": 1.2570747691391122,
      "grad_norm": 7.1875,
      "learning_rate": 3.142314566577301e-05,
      "loss": 1.5403,
      "num_input_tokens_seen": 4901880,
      "step": 8440
    },
    {
      "epoch": 1.2578194816800714,
      "grad_norm": 13.1875,
      "learning_rate": 3.1441763479296995e-05,
      "loss": 1.0262,
      "num_input_tokens_seen": 4904568,
      "step": 8445
    },
    {
      "epoch": 1.2585641942210306,
      "grad_norm": 7.90625,
      "learning_rate": 3.146038129282097e-05,
      "loss": 0.6137,
      "num_input_tokens_seen": 4907480,
      "step": 8450
    },
    {
      "epoch": 1.2593089067619898,
      "grad_norm": 13.0625,
      "learning_rate": 3.147899910634495e-05,
      "loss": 0.9668,
      "num_input_tokens_seen": 4910744,
      "step": 8455
    },
    {
      "epoch": 1.260053619302949,
      "grad_norm": 12.375,
      "learning_rate": 3.1497616919868936e-05,
      "loss": 0.834,
      "num_input_tokens_seen": 4913720,
      "step": 8460
    },
    {
      "epoch": 1.2607983318439082,
      "grad_norm": 11.0,
      "learning_rate": 3.1516234733392914e-05,
      "loss": 1.2176,
      "num_input_tokens_seen": 4916568,
      "step": 8465
    },
    {
      "epoch": 1.2615430443848674,
      "grad_norm": 12.8125,
      "learning_rate": 3.153485254691689e-05,
      "loss": 1.1115,
      "num_input_tokens_seen": 4919224,
      "step": 8470
    },
    {
      "epoch": 1.2622877569258266,
      "grad_norm": 16.375,
      "learning_rate": 3.155347036044087e-05,
      "loss": 1.2485,
      "num_input_tokens_seen": 4922168,
      "step": 8475
    },
    {
      "epoch": 1.2630324694667858,
      "grad_norm": 19.375,
      "learning_rate": 3.1572088173964856e-05,
      "loss": 0.9674,
      "num_input_tokens_seen": 4924824,
      "step": 8480
    },
    {
      "epoch": 1.263777182007745,
      "grad_norm": 11.375,
      "learning_rate": 3.1590705987488834e-05,
      "loss": 0.9949,
      "num_input_tokens_seen": 4927864,
      "step": 8485
    },
    {
      "epoch": 1.2645218945487042,
      "grad_norm": 13.375,
      "learning_rate": 3.1609323801012806e-05,
      "loss": 1.1221,
      "num_input_tokens_seen": 4930648,
      "step": 8490
    },
    {
      "epoch": 1.2652666070896634,
      "grad_norm": 16.0,
      "learning_rate": 3.162794161453679e-05,
      "loss": 1.2526,
      "num_input_tokens_seen": 4933944,
      "step": 8495
    },
    {
      "epoch": 1.2660113196306226,
      "grad_norm": 9.9375,
      "learning_rate": 3.164655942806077e-05,
      "loss": 1.3627,
      "num_input_tokens_seen": 4936600,
      "step": 8500
    },
    {
      "epoch": 1.2667560321715818,
      "grad_norm": 11.0,
      "learning_rate": 3.166517724158475e-05,
      "loss": 1.0826,
      "num_input_tokens_seen": 4939192,
      "step": 8505
    },
    {
      "epoch": 1.267500744712541,
      "grad_norm": 9.5,
      "learning_rate": 3.1683795055108725e-05,
      "loss": 0.9995,
      "num_input_tokens_seen": 4942232,
      "step": 8510
    },
    {
      "epoch": 1.2682454572535002,
      "grad_norm": 12.4375,
      "learning_rate": 3.170241286863271e-05,
      "loss": 1.4233,
      "num_input_tokens_seen": 4945240,
      "step": 8515
    },
    {
      "epoch": 1.2689901697944594,
      "grad_norm": 8.375,
      "learning_rate": 3.172103068215669e-05,
      "loss": 1.0345,
      "num_input_tokens_seen": 4948248,
      "step": 8520
    },
    {
      "epoch": 1.2697348823354186,
      "grad_norm": 11.875,
      "learning_rate": 3.173964849568067e-05,
      "loss": 0.9924,
      "num_input_tokens_seen": 4951224,
      "step": 8525
    },
    {
      "epoch": 1.2704795948763778,
      "grad_norm": 17.625,
      "learning_rate": 3.1758266309204645e-05,
      "loss": 0.8737,
      "num_input_tokens_seen": 4954168,
      "step": 8530
    },
    {
      "epoch": 1.2712243074173368,
      "grad_norm": 15.125,
      "learning_rate": 3.177688412272863e-05,
      "loss": 1.1406,
      "num_input_tokens_seen": 4956952,
      "step": 8535
    },
    {
      "epoch": 1.271969019958296,
      "grad_norm": 18.375,
      "learning_rate": 3.179550193625261e-05,
      "loss": 1.1621,
      "num_input_tokens_seen": 4959512,
      "step": 8540
    },
    {
      "epoch": 1.2727137324992552,
      "grad_norm": 7.8125,
      "learning_rate": 3.1814119749776586e-05,
      "loss": 1.2196,
      "num_input_tokens_seen": 4962296,
      "step": 8545
    },
    {
      "epoch": 1.2734584450402144,
      "grad_norm": 16.125,
      "learning_rate": 3.183273756330057e-05,
      "loss": 1.3434,
      "num_input_tokens_seen": 4965048,
      "step": 8550
    },
    {
      "epoch": 1.2742031575811736,
      "grad_norm": 11.75,
      "learning_rate": 3.185135537682455e-05,
      "loss": 1.3777,
      "num_input_tokens_seen": 4967672,
      "step": 8555
    },
    {
      "epoch": 1.2749478701221328,
      "grad_norm": 11.5625,
      "learning_rate": 3.186997319034853e-05,
      "loss": 1.2772,
      "num_input_tokens_seen": 4970744,
      "step": 8560
    },
    {
      "epoch": 1.275692582663092,
      "grad_norm": 8.125,
      "learning_rate": 3.1888591003872506e-05,
      "loss": 1.1019,
      "num_input_tokens_seen": 4973432,
      "step": 8565
    },
    {
      "epoch": 1.2764372952040512,
      "grad_norm": 28.75,
      "learning_rate": 3.190720881739649e-05,
      "loss": 1.204,
      "num_input_tokens_seen": 4976568,
      "step": 8570
    },
    {
      "epoch": 1.2771820077450105,
      "grad_norm": 19.5,
      "learning_rate": 3.192582663092047e-05,
      "loss": 1.4367,
      "num_input_tokens_seen": 4979320,
      "step": 8575
    },
    {
      "epoch": 1.2779267202859697,
      "grad_norm": 8.6875,
      "learning_rate": 3.194444444444444e-05,
      "loss": 0.9118,
      "num_input_tokens_seen": 4981976,
      "step": 8580
    },
    {
      "epoch": 1.2786714328269289,
      "grad_norm": 9.25,
      "learning_rate": 3.1963062257968426e-05,
      "loss": 1.2747,
      "num_input_tokens_seen": 4984856,
      "step": 8585
    },
    {
      "epoch": 1.279416145367888,
      "grad_norm": 15.5625,
      "learning_rate": 3.1981680071492404e-05,
      "loss": 1.3299,
      "num_input_tokens_seen": 4987928,
      "step": 8590
    },
    {
      "epoch": 1.2801608579088473,
      "grad_norm": 9.4375,
      "learning_rate": 3.200029788501638e-05,
      "loss": 0.9903,
      "num_input_tokens_seen": 4990936,
      "step": 8595
    },
    {
      "epoch": 1.2809055704498062,
      "grad_norm": 13.6875,
      "learning_rate": 3.201891569854036e-05,
      "loss": 1.1968,
      "num_input_tokens_seen": 4993464,
      "step": 8600
    },
    {
      "epoch": 1.2816502829907654,
      "grad_norm": 5.9375,
      "learning_rate": 3.2037533512064346e-05,
      "loss": 1.0754,
      "num_input_tokens_seen": 4996248,
      "step": 8605
    },
    {
      "epoch": 1.2823949955317246,
      "grad_norm": 15.9375,
      "learning_rate": 3.2056151325588324e-05,
      "loss": 1.3406,
      "num_input_tokens_seen": 4999128,
      "step": 8610
    },
    {
      "epoch": 1.2831397080726838,
      "grad_norm": 18.625,
      "learning_rate": 3.20747691391123e-05,
      "loss": 1.1257,
      "num_input_tokens_seen": 5001624,
      "step": 8615
    },
    {
      "epoch": 1.283884420613643,
      "grad_norm": 20.0,
      "learning_rate": 3.209338695263629e-05,
      "loss": 0.9847,
      "num_input_tokens_seen": 5004248,
      "step": 8620
    },
    {
      "epoch": 1.2846291331546023,
      "grad_norm": 12.75,
      "learning_rate": 3.2112004766160265e-05,
      "loss": 1.2523,
      "num_input_tokens_seen": 5007032,
      "step": 8625
    },
    {
      "epoch": 1.2853738456955615,
      "grad_norm": 7.71875,
      "learning_rate": 3.2130622579684244e-05,
      "loss": 1.1728,
      "num_input_tokens_seen": 5009912,
      "step": 8630
    },
    {
      "epoch": 1.2861185582365207,
      "grad_norm": 21.375,
      "learning_rate": 3.214924039320822e-05,
      "loss": 1.3178,
      "num_input_tokens_seen": 5012728,
      "step": 8635
    },
    {
      "epoch": 1.2868632707774799,
      "grad_norm": 9.4375,
      "learning_rate": 3.216785820673221e-05,
      "loss": 0.9177,
      "num_input_tokens_seen": 5015576,
      "step": 8640
    },
    {
      "epoch": 1.287607983318439,
      "grad_norm": 6.28125,
      "learning_rate": 3.2186476020256185e-05,
      "loss": 0.9648,
      "num_input_tokens_seen": 5018296,
      "step": 8645
    },
    {
      "epoch": 1.2883526958593983,
      "grad_norm": 16.375,
      "learning_rate": 3.220509383378016e-05,
      "loss": 1.0369,
      "num_input_tokens_seen": 5020824,
      "step": 8650
    },
    {
      "epoch": 1.2890974084003575,
      "grad_norm": 15.3125,
      "learning_rate": 3.222371164730414e-05,
      "loss": 0.825,
      "num_input_tokens_seen": 5023512,
      "step": 8655
    },
    {
      "epoch": 1.2898421209413167,
      "grad_norm": 16.5,
      "learning_rate": 3.2242329460828126e-05,
      "loss": 1.4232,
      "num_input_tokens_seen": 5026232,
      "step": 8660
    },
    {
      "epoch": 1.2905868334822759,
      "grad_norm": 5.21875,
      "learning_rate": 3.22609472743521e-05,
      "loss": 0.9092,
      "num_input_tokens_seen": 5029304,
      "step": 8665
    },
    {
      "epoch": 1.291331546023235,
      "grad_norm": 8.75,
      "learning_rate": 3.2279565087876076e-05,
      "loss": 0.9897,
      "num_input_tokens_seen": 5032056,
      "step": 8670
    },
    {
      "epoch": 1.2920762585641943,
      "grad_norm": 11.6875,
      "learning_rate": 3.229818290140006e-05,
      "loss": 1.1138,
      "num_input_tokens_seen": 5034936,
      "step": 8675
    },
    {
      "epoch": 1.2928209711051535,
      "grad_norm": 13.6875,
      "learning_rate": 3.231680071492404e-05,
      "loss": 1.0426,
      "num_input_tokens_seen": 5037944,
      "step": 8680
    },
    {
      "epoch": 1.2935656836461127,
      "grad_norm": 16.125,
      "learning_rate": 3.233541852844802e-05,
      "loss": 1.0562,
      "num_input_tokens_seen": 5041080,
      "step": 8685
    },
    {
      "epoch": 1.2943103961870719,
      "grad_norm": 10.625,
      "learning_rate": 3.2354036341972e-05,
      "loss": 1.2372,
      "num_input_tokens_seen": 5043896,
      "step": 8690
    },
    {
      "epoch": 1.295055108728031,
      "grad_norm": 12.375,
      "learning_rate": 3.237265415549598e-05,
      "loss": 0.8941,
      "num_input_tokens_seen": 5046680,
      "step": 8695
    },
    {
      "epoch": 1.2957998212689903,
      "grad_norm": 22.375,
      "learning_rate": 3.239127196901996e-05,
      "loss": 1.342,
      "num_input_tokens_seen": 5049464,
      "step": 8700
    },
    {
      "epoch": 1.2965445338099495,
      "grad_norm": 10.625,
      "learning_rate": 3.240988978254394e-05,
      "loss": 1.1294,
      "num_input_tokens_seen": 5052568,
      "step": 8705
    },
    {
      "epoch": 1.2972892463509085,
      "grad_norm": 16.875,
      "learning_rate": 3.242850759606792e-05,
      "loss": 1.4394,
      "num_input_tokens_seen": 5055256,
      "step": 8710
    },
    {
      "epoch": 1.2980339588918677,
      "grad_norm": 9.9375,
      "learning_rate": 3.24471254095919e-05,
      "loss": 1.0739,
      "num_input_tokens_seen": 5058264,
      "step": 8715
    },
    {
      "epoch": 1.2987786714328269,
      "grad_norm": 16.0,
      "learning_rate": 3.246574322311588e-05,
      "loss": 1.2209,
      "num_input_tokens_seen": 5061048,
      "step": 8720
    },
    {
      "epoch": 1.299523383973786,
      "grad_norm": 11.8125,
      "learning_rate": 3.248436103663986e-05,
      "loss": 0.8368,
      "num_input_tokens_seen": 5063864,
      "step": 8725
    },
    {
      "epoch": 1.3002680965147453,
      "grad_norm": 8.75,
      "learning_rate": 3.250297885016384e-05,
      "loss": 1.1951,
      "num_input_tokens_seen": 5066552,
      "step": 8730
    },
    {
      "epoch": 1.3010128090557045,
      "grad_norm": 17.125,
      "learning_rate": 3.252159666368782e-05,
      "loss": 1.2864,
      "num_input_tokens_seen": 5069368,
      "step": 8735
    },
    {
      "epoch": 1.3017575215966637,
      "grad_norm": 14.75,
      "learning_rate": 3.25402144772118e-05,
      "loss": 0.9899,
      "num_input_tokens_seen": 5072248,
      "step": 8740
    },
    {
      "epoch": 1.302502234137623,
      "grad_norm": 8.4375,
      "learning_rate": 3.255883229073578e-05,
      "loss": 0.8606,
      "num_input_tokens_seen": 5075192,
      "step": 8745
    },
    {
      "epoch": 1.303246946678582,
      "grad_norm": 6.125,
      "learning_rate": 3.2577450104259755e-05,
      "loss": 1.1297,
      "num_input_tokens_seen": 5078008,
      "step": 8750
    },
    {
      "epoch": 1.3039916592195413,
      "grad_norm": 13.125,
      "learning_rate": 3.259606791778373e-05,
      "loss": 1.2625,
      "num_input_tokens_seen": 5081112,
      "step": 8755
    },
    {
      "epoch": 1.3047363717605005,
      "grad_norm": 9.1875,
      "learning_rate": 3.261468573130771e-05,
      "loss": 0.9066,
      "num_input_tokens_seen": 5083992,
      "step": 8760
    },
    {
      "epoch": 1.3054810843014597,
      "grad_norm": 10.25,
      "learning_rate": 3.2633303544831696e-05,
      "loss": 0.9102,
      "num_input_tokens_seen": 5086936,
      "step": 8765
    },
    {
      "epoch": 1.306225796842419,
      "grad_norm": 14.25,
      "learning_rate": 3.2651921358355675e-05,
      "loss": 1.074,
      "num_input_tokens_seen": 5090008,
      "step": 8770
    },
    {
      "epoch": 1.3069705093833779,
      "grad_norm": 28.25,
      "learning_rate": 3.267053917187965e-05,
      "loss": 1.6805,
      "num_input_tokens_seen": 5093144,
      "step": 8775
    },
    {
      "epoch": 1.307715221924337,
      "grad_norm": 8.4375,
      "learning_rate": 3.268915698540364e-05,
      "loss": 1.022,
      "num_input_tokens_seen": 5096312,
      "step": 8780
    },
    {
      "epoch": 1.3084599344652963,
      "grad_norm": 16.5,
      "learning_rate": 3.2707774798927616e-05,
      "loss": 1.1937,
      "num_input_tokens_seen": 5099192,
      "step": 8785
    },
    {
      "epoch": 1.3092046470062555,
      "grad_norm": 8.9375,
      "learning_rate": 3.2726392612451594e-05,
      "loss": 0.9947,
      "num_input_tokens_seen": 5101848,
      "step": 8790
    },
    {
      "epoch": 1.3099493595472147,
      "grad_norm": 11.625,
      "learning_rate": 3.274501042597557e-05,
      "loss": 1.219,
      "num_input_tokens_seen": 5104760,
      "step": 8795
    },
    {
      "epoch": 1.310694072088174,
      "grad_norm": 11.0625,
      "learning_rate": 3.276362823949956e-05,
      "loss": 1.3419,
      "num_input_tokens_seen": 5107544,
      "step": 8800
    },
    {
      "epoch": 1.311438784629133,
      "grad_norm": 11.6875,
      "learning_rate": 3.2782246053023536e-05,
      "loss": 0.9165,
      "num_input_tokens_seen": 5110488,
      "step": 8805
    },
    {
      "epoch": 1.3121834971700923,
      "grad_norm": 11.25,
      "learning_rate": 3.2800863866547514e-05,
      "loss": 1.2973,
      "num_input_tokens_seen": 5113400,
      "step": 8810
    },
    {
      "epoch": 1.3129282097110515,
      "grad_norm": 16.125,
      "learning_rate": 3.281948168007149e-05,
      "loss": 1.188,
      "num_input_tokens_seen": 5116600,
      "step": 8815
    },
    {
      "epoch": 1.3136729222520107,
      "grad_norm": 10.875,
      "learning_rate": 3.283809949359548e-05,
      "loss": 1.1666,
      "num_input_tokens_seen": 5119512,
      "step": 8820
    },
    {
      "epoch": 1.31441763479297,
      "grad_norm": 7.71875,
      "learning_rate": 3.2856717307119456e-05,
      "loss": 0.9034,
      "num_input_tokens_seen": 5122360,
      "step": 8825
    },
    {
      "epoch": 1.3151623473339291,
      "grad_norm": 11.625,
      "learning_rate": 3.2875335120643434e-05,
      "loss": 1.0972,
      "num_input_tokens_seen": 5125176,
      "step": 8830
    },
    {
      "epoch": 1.3159070598748883,
      "grad_norm": 16.25,
      "learning_rate": 3.289395293416741e-05,
      "loss": 1.4169,
      "num_input_tokens_seen": 5127960,
      "step": 8835
    },
    {
      "epoch": 1.3166517724158475,
      "grad_norm": 8.4375,
      "learning_rate": 3.291257074769139e-05,
      "loss": 1.1416,
      "num_input_tokens_seen": 5131000,
      "step": 8840
    },
    {
      "epoch": 1.3173964849568067,
      "grad_norm": 6.625,
      "learning_rate": 3.293118856121537e-05,
      "loss": 0.863,
      "num_input_tokens_seen": 5133976,
      "step": 8845
    },
    {
      "epoch": 1.318141197497766,
      "grad_norm": 9.6875,
      "learning_rate": 3.2949806374739354e-05,
      "loss": 1.1531,
      "num_input_tokens_seen": 5137112,
      "step": 8850
    },
    {
      "epoch": 1.3188859100387251,
      "grad_norm": 19.0,
      "learning_rate": 3.296842418826333e-05,
      "loss": 1.4552,
      "num_input_tokens_seen": 5140152,
      "step": 8855
    },
    {
      "epoch": 1.3196306225796843,
      "grad_norm": 5.78125,
      "learning_rate": 3.298704200178731e-05,
      "loss": 1.3694,
      "num_input_tokens_seen": 5143288,
      "step": 8860
    },
    {
      "epoch": 1.3203753351206435,
      "grad_norm": 9.0,
      "learning_rate": 3.300565981531129e-05,
      "loss": 0.9017,
      "num_input_tokens_seen": 5145976,
      "step": 8865
    },
    {
      "epoch": 1.3211200476616027,
      "grad_norm": 8.6875,
      "learning_rate": 3.302427762883527e-05,
      "loss": 0.9072,
      "num_input_tokens_seen": 5148792,
      "step": 8870
    },
    {
      "epoch": 1.321864760202562,
      "grad_norm": 6.3125,
      "learning_rate": 3.304289544235925e-05,
      "loss": 0.7676,
      "num_input_tokens_seen": 5151384,
      "step": 8875
    },
    {
      "epoch": 1.322609472743521,
      "grad_norm": 5.875,
      "learning_rate": 3.306151325588323e-05,
      "loss": 1.0393,
      "num_input_tokens_seen": 5154168,
      "step": 8880
    },
    {
      "epoch": 1.3233541852844801,
      "grad_norm": 20.125,
      "learning_rate": 3.308013106940721e-05,
      "loss": 1.2882,
      "num_input_tokens_seen": 5157432,
      "step": 8885
    },
    {
      "epoch": 1.3240988978254393,
      "grad_norm": 14.4375,
      "learning_rate": 3.309874888293119e-05,
      "loss": 1.1993,
      "num_input_tokens_seen": 5160216,
      "step": 8890
    },
    {
      "epoch": 1.3248436103663985,
      "grad_norm": 6.9375,
      "learning_rate": 3.311736669645517e-05,
      "loss": 0.9205,
      "num_input_tokens_seen": 5162936,
      "step": 8895
    },
    {
      "epoch": 1.3255883229073577,
      "grad_norm": 7.4375,
      "learning_rate": 3.313598450997915e-05,
      "loss": 1.1781,
      "num_input_tokens_seen": 5165848,
      "step": 8900
    },
    {
      "epoch": 1.326333035448317,
      "grad_norm": 8.4375,
      "learning_rate": 3.3154602323503134e-05,
      "loss": 1.2113,
      "num_input_tokens_seen": 5168760,
      "step": 8905
    },
    {
      "epoch": 1.3270777479892761,
      "grad_norm": 10.625,
      "learning_rate": 3.317322013702711e-05,
      "loss": 1.22,
      "num_input_tokens_seen": 5171576,
      "step": 8910
    },
    {
      "epoch": 1.3278224605302353,
      "grad_norm": 19.25,
      "learning_rate": 3.319183795055109e-05,
      "loss": 1.3895,
      "num_input_tokens_seen": 5174456,
      "step": 8915
    },
    {
      "epoch": 1.3285671730711945,
      "grad_norm": 10.4375,
      "learning_rate": 3.321045576407507e-05,
      "loss": 0.9537,
      "num_input_tokens_seen": 5177240,
      "step": 8920
    },
    {
      "epoch": 1.3293118856121537,
      "grad_norm": 8.9375,
      "learning_rate": 3.322907357759905e-05,
      "loss": 0.8266,
      "num_input_tokens_seen": 5180216,
      "step": 8925
    },
    {
      "epoch": 1.330056598153113,
      "grad_norm": 18.25,
      "learning_rate": 3.3247691391123026e-05,
      "loss": 1.4773,
      "num_input_tokens_seen": 5183224,
      "step": 8930
    },
    {
      "epoch": 1.3308013106940721,
      "grad_norm": 13.875,
      "learning_rate": 3.3266309204647004e-05,
      "loss": 1.2567,
      "num_input_tokens_seen": 5186072,
      "step": 8935
    },
    {
      "epoch": 1.3315460232350314,
      "grad_norm": 11.625,
      "learning_rate": 3.328492701817099e-05,
      "loss": 0.9867,
      "num_input_tokens_seen": 5189464,
      "step": 8940
    },
    {
      "epoch": 1.3322907357759903,
      "grad_norm": 9.9375,
      "learning_rate": 3.330354483169497e-05,
      "loss": 0.9544,
      "num_input_tokens_seen": 5192440,
      "step": 8945
    },
    {
      "epoch": 1.3330354483169495,
      "grad_norm": 8.125,
      "learning_rate": 3.3322162645218945e-05,
      "loss": 1.3963,
      "num_input_tokens_seen": 5195352,
      "step": 8950
    },
    {
      "epoch": 1.3337801608579087,
      "grad_norm": 9.6875,
      "learning_rate": 3.3340780458742924e-05,
      "loss": 0.8251,
      "num_input_tokens_seen": 5198328,
      "step": 8955
    },
    {
      "epoch": 1.334524873398868,
      "grad_norm": 11.0625,
      "learning_rate": 3.335939827226691e-05,
      "loss": 1.1567,
      "num_input_tokens_seen": 5201432,
      "step": 8960
    },
    {
      "epoch": 1.3352695859398271,
      "grad_norm": 14.1875,
      "learning_rate": 3.337801608579089e-05,
      "loss": 1.2148,
      "num_input_tokens_seen": 5204344,
      "step": 8965
    },
    {
      "epoch": 1.3360142984807863,
      "grad_norm": 13.4375,
      "learning_rate": 3.3396633899314865e-05,
      "loss": 1.2352,
      "num_input_tokens_seen": 5207672,
      "step": 8970
    },
    {
      "epoch": 1.3367590110217455,
      "grad_norm": 9.0,
      "learning_rate": 3.341525171283884e-05,
      "loss": 1.1679,
      "num_input_tokens_seen": 5210552,
      "step": 8975
    },
    {
      "epoch": 1.3375037235627047,
      "grad_norm": 10.0625,
      "learning_rate": 3.343386952636283e-05,
      "loss": 1.2236,
      "num_input_tokens_seen": 5213496,
      "step": 8980
    },
    {
      "epoch": 1.338248436103664,
      "grad_norm": 8.875,
      "learning_rate": 3.3452487339886806e-05,
      "loss": 1.0915,
      "num_input_tokens_seen": 5215992,
      "step": 8985
    },
    {
      "epoch": 1.3389931486446232,
      "grad_norm": 9.75,
      "learning_rate": 3.3471105153410785e-05,
      "loss": 1.3584,
      "num_input_tokens_seen": 5218584,
      "step": 8990
    },
    {
      "epoch": 1.3397378611855824,
      "grad_norm": 7.53125,
      "learning_rate": 3.348972296693477e-05,
      "loss": 1.0286,
      "num_input_tokens_seen": 5221592,
      "step": 8995
    },
    {
      "epoch": 1.3404825737265416,
      "grad_norm": 14.8125,
      "learning_rate": 3.350834078045875e-05,
      "loss": 1.2921,
      "num_input_tokens_seen": 5224504,
      "step": 9000
    },
    {
      "epoch": 1.3412272862675008,
      "grad_norm": 6.59375,
      "learning_rate": 3.3526958593982726e-05,
      "loss": 0.9775,
      "num_input_tokens_seen": 5227512,
      "step": 9005
    },
    {
      "epoch": 1.34197199880846,
      "grad_norm": 18.625,
      "learning_rate": 3.3545576407506704e-05,
      "loss": 1.3044,
      "num_input_tokens_seen": 5230328,
      "step": 9010
    },
    {
      "epoch": 1.3427167113494192,
      "grad_norm": 8.875,
      "learning_rate": 3.356419422103068e-05,
      "loss": 1.3115,
      "num_input_tokens_seen": 5233240,
      "step": 9015
    },
    {
      "epoch": 1.3434614238903784,
      "grad_norm": 6.53125,
      "learning_rate": 3.358281203455466e-05,
      "loss": 0.826,
      "num_input_tokens_seen": 5236312,
      "step": 9020
    },
    {
      "epoch": 1.3442061364313376,
      "grad_norm": 29.0,
      "learning_rate": 3.360142984807864e-05,
      "loss": 1.5007,
      "num_input_tokens_seen": 5239544,
      "step": 9025
    },
    {
      "epoch": 1.3449508489722968,
      "grad_norm": 7.59375,
      "learning_rate": 3.3620047661602624e-05,
      "loss": 0.9943,
      "num_input_tokens_seen": 5242808,
      "step": 9030
    },
    {
      "epoch": 1.345695561513256,
      "grad_norm": 12.25,
      "learning_rate": 3.36386654751266e-05,
      "loss": 1.0911,
      "num_input_tokens_seen": 5245560,
      "step": 9035
    },
    {
      "epoch": 1.3464402740542152,
      "grad_norm": 8.5,
      "learning_rate": 3.365728328865058e-05,
      "loss": 0.8796,
      "num_input_tokens_seen": 5248408,
      "step": 9040
    },
    {
      "epoch": 1.3471849865951744,
      "grad_norm": 12.5625,
      "learning_rate": 3.367590110217456e-05,
      "loss": 1.1694,
      "num_input_tokens_seen": 5251352,
      "step": 9045
    },
    {
      "epoch": 1.3479296991361336,
      "grad_norm": 20.75,
      "learning_rate": 3.3694518915698544e-05,
      "loss": 1.146,
      "num_input_tokens_seen": 5254200,
      "step": 9050
    },
    {
      "epoch": 1.3486744116770926,
      "grad_norm": 12.1875,
      "learning_rate": 3.371313672922252e-05,
      "loss": 1.2621,
      "num_input_tokens_seen": 5257144,
      "step": 9055
    },
    {
      "epoch": 1.3494191242180518,
      "grad_norm": 18.5,
      "learning_rate": 3.37317545427465e-05,
      "loss": 1.1159,
      "num_input_tokens_seen": 5260120,
      "step": 9060
    },
    {
      "epoch": 1.350163836759011,
      "grad_norm": 10.875,
      "learning_rate": 3.3750372356270485e-05,
      "loss": 1.3631,
      "num_input_tokens_seen": 5262872,
      "step": 9065
    },
    {
      "epoch": 1.3509085492999702,
      "grad_norm": 11.625,
      "learning_rate": 3.3768990169794464e-05,
      "loss": 1.1594,
      "num_input_tokens_seen": 5265944,
      "step": 9070
    },
    {
      "epoch": 1.3516532618409294,
      "grad_norm": 10.0625,
      "learning_rate": 3.378760798331844e-05,
      "loss": 0.9631,
      "num_input_tokens_seen": 5268920,
      "step": 9075
    },
    {
      "epoch": 1.3523979743818886,
      "grad_norm": 12.9375,
      "learning_rate": 3.380622579684242e-05,
      "loss": 0.951,
      "num_input_tokens_seen": 5272024,
      "step": 9080
    },
    {
      "epoch": 1.3531426869228478,
      "grad_norm": 6.71875,
      "learning_rate": 3.3824843610366405e-05,
      "loss": 1.0983,
      "num_input_tokens_seen": 5274968,
      "step": 9085
    },
    {
      "epoch": 1.353887399463807,
      "grad_norm": 12.1875,
      "learning_rate": 3.384346142389038e-05,
      "loss": 1.167,
      "num_input_tokens_seen": 5277944,
      "step": 9090
    },
    {
      "epoch": 1.3546321120047662,
      "grad_norm": 9.75,
      "learning_rate": 3.3862079237414355e-05,
      "loss": 0.8075,
      "num_input_tokens_seen": 5280984,
      "step": 9095
    },
    {
      "epoch": 1.3553768245457254,
      "grad_norm": 10.375,
      "learning_rate": 3.388069705093834e-05,
      "loss": 1.3366,
      "num_input_tokens_seen": 5283800,
      "step": 9100
    },
    {
      "epoch": 1.3561215370866846,
      "grad_norm": 9.125,
      "learning_rate": 3.389931486446232e-05,
      "loss": 1.0483,
      "num_input_tokens_seen": 5286744,
      "step": 9105
    },
    {
      "epoch": 1.3568662496276438,
      "grad_norm": 6.84375,
      "learning_rate": 3.3917932677986296e-05,
      "loss": 0.9698,
      "num_input_tokens_seen": 5289304,
      "step": 9110
    },
    {
      "epoch": 1.357610962168603,
      "grad_norm": 11.1875,
      "learning_rate": 3.3936550491510274e-05,
      "loss": 1.4989,
      "num_input_tokens_seen": 5292088,
      "step": 9115
    },
    {
      "epoch": 1.358355674709562,
      "grad_norm": 8.9375,
      "learning_rate": 3.395516830503426e-05,
      "loss": 1.2347,
      "num_input_tokens_seen": 5295096,
      "step": 9120
    },
    {
      "epoch": 1.3591003872505212,
      "grad_norm": 5.6875,
      "learning_rate": 3.397378611855824e-05,
      "loss": 1.2371,
      "num_input_tokens_seen": 5298104,
      "step": 9125
    },
    {
      "epoch": 1.3598450997914804,
      "grad_norm": 17.0,
      "learning_rate": 3.3992403932082216e-05,
      "loss": 1.4447,
      "num_input_tokens_seen": 5301016,
      "step": 9130
    },
    {
      "epoch": 1.3605898123324396,
      "grad_norm": 5.6875,
      "learning_rate": 3.4011021745606194e-05,
      "loss": 0.8945,
      "num_input_tokens_seen": 5303960,
      "step": 9135
    },
    {
      "epoch": 1.3613345248733988,
      "grad_norm": 16.375,
      "learning_rate": 3.402963955913018e-05,
      "loss": 1.3943,
      "num_input_tokens_seen": 5306680,
      "step": 9140
    },
    {
      "epoch": 1.362079237414358,
      "grad_norm": 17.0,
      "learning_rate": 3.404825737265416e-05,
      "loss": 1.1922,
      "num_input_tokens_seen": 5309336,
      "step": 9145
    },
    {
      "epoch": 1.3628239499553172,
      "grad_norm": 9.3125,
      "learning_rate": 3.4066875186178136e-05,
      "loss": 0.8774,
      "num_input_tokens_seen": 5312216,
      "step": 9150
    },
    {
      "epoch": 1.3635686624962764,
      "grad_norm": 20.375,
      "learning_rate": 3.408549299970212e-05,
      "loss": 1.1275,
      "num_input_tokens_seen": 5315480,
      "step": 9155
    },
    {
      "epoch": 1.3643133750372356,
      "grad_norm": 5.53125,
      "learning_rate": 3.41041108132261e-05,
      "loss": 0.9545,
      "num_input_tokens_seen": 5318392,
      "step": 9160
    },
    {
      "epoch": 1.3650580875781948,
      "grad_norm": 18.5,
      "learning_rate": 3.412272862675008e-05,
      "loss": 1.1068,
      "num_input_tokens_seen": 5321336,
      "step": 9165
    },
    {
      "epoch": 1.365802800119154,
      "grad_norm": 13.3125,
      "learning_rate": 3.4141346440274055e-05,
      "loss": 1.221,
      "num_input_tokens_seen": 5324024,
      "step": 9170
    },
    {
      "epoch": 1.3665475126601132,
      "grad_norm": 27.5,
      "learning_rate": 3.415996425379804e-05,
      "loss": 1.1158,
      "num_input_tokens_seen": 5327000,
      "step": 9175
    },
    {
      "epoch": 1.3672922252010724,
      "grad_norm": 8.75,
      "learning_rate": 3.417858206732202e-05,
      "loss": 1.2127,
      "num_input_tokens_seen": 5329944,
      "step": 9180
    },
    {
      "epoch": 1.3680369377420316,
      "grad_norm": 12.0625,
      "learning_rate": 3.419719988084599e-05,
      "loss": 1.1935,
      "num_input_tokens_seen": 5332760,
      "step": 9185
    },
    {
      "epoch": 1.3687816502829908,
      "grad_norm": 17.875,
      "learning_rate": 3.4215817694369975e-05,
      "loss": 1.2628,
      "num_input_tokens_seen": 5335928,
      "step": 9190
    },
    {
      "epoch": 1.36952636282395,
      "grad_norm": 15.8125,
      "learning_rate": 3.423443550789395e-05,
      "loss": 1.0791,
      "num_input_tokens_seen": 5338904,
      "step": 9195
    },
    {
      "epoch": 1.3702710753649092,
      "grad_norm": 13.875,
      "learning_rate": 3.425305332141793e-05,
      "loss": 1.1127,
      "num_input_tokens_seen": 5341816,
      "step": 9200
    },
    {
      "epoch": 1.3710157879058684,
      "grad_norm": 8.375,
      "learning_rate": 3.427167113494191e-05,
      "loss": 0.9405,
      "num_input_tokens_seen": 5344696,
      "step": 9205
    },
    {
      "epoch": 1.3717605004468276,
      "grad_norm": 13.0,
      "learning_rate": 3.4290288948465895e-05,
      "loss": 1.4565,
      "num_input_tokens_seen": 5347832,
      "step": 9210
    },
    {
      "epoch": 1.3725052129877868,
      "grad_norm": 17.125,
      "learning_rate": 3.430890676198987e-05,
      "loss": 1.2262,
      "num_input_tokens_seen": 5350584,
      "step": 9215
    },
    {
      "epoch": 1.373249925528746,
      "grad_norm": 13.25,
      "learning_rate": 3.432752457551385e-05,
      "loss": 1.0374,
      "num_input_tokens_seen": 5353656,
      "step": 9220
    },
    {
      "epoch": 1.3739946380697052,
      "grad_norm": 15.25,
      "learning_rate": 3.4346142389037836e-05,
      "loss": 0.8073,
      "num_input_tokens_seen": 5356728,
      "step": 9225
    },
    {
      "epoch": 1.3747393506106642,
      "grad_norm": 8.8125,
      "learning_rate": 3.4364760202561814e-05,
      "loss": 1.1038,
      "num_input_tokens_seen": 5359768,
      "step": 9230
    },
    {
      "epoch": 1.3754840631516234,
      "grad_norm": 16.75,
      "learning_rate": 3.438337801608579e-05,
      "loss": 1.2412,
      "num_input_tokens_seen": 5362904,
      "step": 9235
    },
    {
      "epoch": 1.3762287756925826,
      "grad_norm": 12.25,
      "learning_rate": 3.440199582960977e-05,
      "loss": 1.0203,
      "num_input_tokens_seen": 5366008,
      "step": 9240
    },
    {
      "epoch": 1.3769734882335418,
      "grad_norm": 9.75,
      "learning_rate": 3.4420613643133756e-05,
      "loss": 0.8545,
      "num_input_tokens_seen": 5369016,
      "step": 9245
    },
    {
      "epoch": 1.377718200774501,
      "grad_norm": 6.125,
      "learning_rate": 3.4439231456657734e-05,
      "loss": 1.2657,
      "num_input_tokens_seen": 5371928,
      "step": 9250
    },
    {
      "epoch": 1.3784629133154602,
      "grad_norm": 13.125,
      "learning_rate": 3.445784927018171e-05,
      "loss": 0.8815,
      "num_input_tokens_seen": 5374520,
      "step": 9255
    },
    {
      "epoch": 1.3792076258564194,
      "grad_norm": 8.625,
      "learning_rate": 3.447646708370569e-05,
      "loss": 0.8205,
      "num_input_tokens_seen": 5377336,
      "step": 9260
    },
    {
      "epoch": 1.3799523383973786,
      "grad_norm": 6.6875,
      "learning_rate": 3.4495084897229676e-05,
      "loss": 0.8687,
      "num_input_tokens_seen": 5380504,
      "step": 9265
    },
    {
      "epoch": 1.3806970509383378,
      "grad_norm": 7.4375,
      "learning_rate": 3.451370271075365e-05,
      "loss": 0.9709,
      "num_input_tokens_seen": 5383256,
      "step": 9270
    },
    {
      "epoch": 1.381441763479297,
      "grad_norm": 9.0625,
      "learning_rate": 3.4532320524277625e-05,
      "loss": 1.0722,
      "num_input_tokens_seen": 5386168,
      "step": 9275
    },
    {
      "epoch": 1.3821864760202562,
      "grad_norm": 19.875,
      "learning_rate": 3.455093833780161e-05,
      "loss": 1.0579,
      "num_input_tokens_seen": 5388984,
      "step": 9280
    },
    {
      "epoch": 1.3829311885612154,
      "grad_norm": 12.1875,
      "learning_rate": 3.456955615132559e-05,
      "loss": 1.2192,
      "num_input_tokens_seen": 5391608,
      "step": 9285
    },
    {
      "epoch": 1.3836759011021746,
      "grad_norm": 8.9375,
      "learning_rate": 3.458817396484957e-05,
      "loss": 1.2992,
      "num_input_tokens_seen": 5394360,
      "step": 9290
    },
    {
      "epoch": 1.3844206136431336,
      "grad_norm": 9.9375,
      "learning_rate": 3.460679177837355e-05,
      "loss": 1.4826,
      "num_input_tokens_seen": 5397272,
      "step": 9295
    },
    {
      "epoch": 1.3851653261840928,
      "grad_norm": 12.6875,
      "learning_rate": 3.462540959189753e-05,
      "loss": 1.1749,
      "num_input_tokens_seen": 5400024,
      "step": 9300
    },
    {
      "epoch": 1.385910038725052,
      "grad_norm": 7.65625,
      "learning_rate": 3.464402740542151e-05,
      "loss": 0.9213,
      "num_input_tokens_seen": 5402872,
      "step": 9305
    },
    {
      "epoch": 1.3866547512660112,
      "grad_norm": 14.375,
      "learning_rate": 3.4662645218945486e-05,
      "loss": 1.3424,
      "num_input_tokens_seen": 5405656,
      "step": 9310
    },
    {
      "epoch": 1.3873994638069704,
      "grad_norm": 7.21875,
      "learning_rate": 3.468126303246947e-05,
      "loss": 0.9012,
      "num_input_tokens_seen": 5408696,
      "step": 9315
    },
    {
      "epoch": 1.3881441763479296,
      "grad_norm": 10.9375,
      "learning_rate": 3.469988084599345e-05,
      "loss": 0.7198,
      "num_input_tokens_seen": 5411608,
      "step": 9320
    },
    {
      "epoch": 1.3888888888888888,
      "grad_norm": 9.375,
      "learning_rate": 3.471849865951743e-05,
      "loss": 1.1828,
      "num_input_tokens_seen": 5414936,
      "step": 9325
    },
    {
      "epoch": 1.389633601429848,
      "grad_norm": 12.75,
      "learning_rate": 3.4737116473041406e-05,
      "loss": 1.2661,
      "num_input_tokens_seen": 5417592,
      "step": 9330
    },
    {
      "epoch": 1.3903783139708072,
      "grad_norm": 11.125,
      "learning_rate": 3.475573428656539e-05,
      "loss": 1.5815,
      "num_input_tokens_seen": 5420344,
      "step": 9335
    },
    {
      "epoch": 1.3911230265117664,
      "grad_norm": 19.375,
      "learning_rate": 3.477435210008937e-05,
      "loss": 0.9726,
      "num_input_tokens_seen": 5423416,
      "step": 9340
    },
    {
      "epoch": 1.3918677390527256,
      "grad_norm": 11.8125,
      "learning_rate": 3.479296991361335e-05,
      "loss": 1.6856,
      "num_input_tokens_seen": 5426264,
      "step": 9345
    },
    {
      "epoch": 1.3926124515936849,
      "grad_norm": 14.625,
      "learning_rate": 3.4811587727137326e-05,
      "loss": 1.0483,
      "num_input_tokens_seen": 5429496,
      "step": 9350
    },
    {
      "epoch": 1.393357164134644,
      "grad_norm": 13.9375,
      "learning_rate": 3.4830205540661304e-05,
      "loss": 1.2351,
      "num_input_tokens_seen": 5432376,
      "step": 9355
    },
    {
      "epoch": 1.3941018766756033,
      "grad_norm": 12.1875,
      "learning_rate": 3.484882335418528e-05,
      "loss": 1.232,
      "num_input_tokens_seen": 5435288,
      "step": 9360
    },
    {
      "epoch": 1.3948465892165625,
      "grad_norm": 10.1875,
      "learning_rate": 3.486744116770926e-05,
      "loss": 1.0247,
      "num_input_tokens_seen": 5438264,
      "step": 9365
    },
    {
      "epoch": 1.3955913017575217,
      "grad_norm": 13.125,
      "learning_rate": 3.4886058981233246e-05,
      "loss": 1.1267,
      "num_input_tokens_seen": 5441208,
      "step": 9370
    },
    {
      "epoch": 1.3963360142984809,
      "grad_norm": 12.625,
      "learning_rate": 3.4904676794757224e-05,
      "loss": 0.7143,
      "num_input_tokens_seen": 5444120,
      "step": 9375
    },
    {
      "epoch": 1.39708072683944,
      "grad_norm": 16.625,
      "learning_rate": 3.49232946082812e-05,
      "loss": 1.2688,
      "num_input_tokens_seen": 5446904,
      "step": 9380
    },
    {
      "epoch": 1.3978254393803993,
      "grad_norm": 10.5625,
      "learning_rate": 3.494191242180519e-05,
      "loss": 1.0875,
      "num_input_tokens_seen": 5449720,
      "step": 9385
    },
    {
      "epoch": 1.3985701519213585,
      "grad_norm": 8.25,
      "learning_rate": 3.4960530235329165e-05,
      "loss": 1.4443,
      "num_input_tokens_seen": 5452536,
      "step": 9390
    },
    {
      "epoch": 1.3993148644623177,
      "grad_norm": 15.8125,
      "learning_rate": 3.4979148048853143e-05,
      "loss": 1.049,
      "num_input_tokens_seen": 5455224,
      "step": 9395
    },
    {
      "epoch": 1.4000595770032767,
      "grad_norm": 15.9375,
      "learning_rate": 3.499776586237712e-05,
      "loss": 1.2016,
      "num_input_tokens_seen": 5458264,
      "step": 9400
    },
    {
      "epoch": 1.4008042895442359,
      "grad_norm": 11.1875,
      "learning_rate": 3.501638367590111e-05,
      "loss": 1.3522,
      "num_input_tokens_seen": 5460952,
      "step": 9405
    },
    {
      "epoch": 1.401549002085195,
      "grad_norm": 14.0625,
      "learning_rate": 3.5035001489425085e-05,
      "loss": 1.2243,
      "num_input_tokens_seen": 5463576,
      "step": 9410
    },
    {
      "epoch": 1.4022937146261543,
      "grad_norm": 9.875,
      "learning_rate": 3.505361930294906e-05,
      "loss": 1.2327,
      "num_input_tokens_seen": 5466296,
      "step": 9415
    },
    {
      "epoch": 1.4030384271671135,
      "grad_norm": 12.0,
      "learning_rate": 3.507223711647304e-05,
      "loss": 1.3586,
      "num_input_tokens_seen": 5469112,
      "step": 9420
    },
    {
      "epoch": 1.4037831397080727,
      "grad_norm": 18.875,
      "learning_rate": 3.5090854929997026e-05,
      "loss": 1.2673,
      "num_input_tokens_seen": 5471864,
      "step": 9425
    },
    {
      "epoch": 1.4045278522490319,
      "grad_norm": 10.4375,
      "learning_rate": 3.5109472743521005e-05,
      "loss": 0.8715,
      "num_input_tokens_seen": 5474712,
      "step": 9430
    },
    {
      "epoch": 1.405272564789991,
      "grad_norm": 18.875,
      "learning_rate": 3.512809055704498e-05,
      "loss": 1.0484,
      "num_input_tokens_seen": 5477336,
      "step": 9435
    },
    {
      "epoch": 1.4060172773309503,
      "grad_norm": 19.375,
      "learning_rate": 3.514670837056897e-05,
      "loss": 1.2136,
      "num_input_tokens_seen": 5480440,
      "step": 9440
    },
    {
      "epoch": 1.4067619898719095,
      "grad_norm": 19.625,
      "learning_rate": 3.516532618409294e-05,
      "loss": 0.989,
      "num_input_tokens_seen": 5483640,
      "step": 9445
    },
    {
      "epoch": 1.4075067024128687,
      "grad_norm": 13.25,
      "learning_rate": 3.518394399761692e-05,
      "loss": 1.3218,
      "num_input_tokens_seen": 5486520,
      "step": 9450
    },
    {
      "epoch": 1.4082514149538279,
      "grad_norm": 7.28125,
      "learning_rate": 3.52025618111409e-05,
      "loss": 0.8392,
      "num_input_tokens_seen": 5489272,
      "step": 9455
    },
    {
      "epoch": 1.408996127494787,
      "grad_norm": 7.875,
      "learning_rate": 3.522117962466488e-05,
      "loss": 1.2816,
      "num_input_tokens_seen": 5492632,
      "step": 9460
    },
    {
      "epoch": 1.409740840035746,
      "grad_norm": 5.4375,
      "learning_rate": 3.523979743818886e-05,
      "loss": 0.9352,
      "num_input_tokens_seen": 5495480,
      "step": 9465
    },
    {
      "epoch": 1.4104855525767053,
      "grad_norm": 9.625,
      "learning_rate": 3.525841525171284e-05,
      "loss": 1.0888,
      "num_input_tokens_seen": 5498264,
      "step": 9470
    },
    {
      "epoch": 1.4112302651176645,
      "grad_norm": 8.3125,
      "learning_rate": 3.527703306523682e-05,
      "loss": 1.0761,
      "num_input_tokens_seen": 5501080,
      "step": 9475
    },
    {
      "epoch": 1.4119749776586237,
      "grad_norm": 10.5,
      "learning_rate": 3.52956508787608e-05,
      "loss": 1.3279,
      "num_input_tokens_seen": 5503896,
      "step": 9480
    },
    {
      "epoch": 1.4127196901995829,
      "grad_norm": 8.4375,
      "learning_rate": 3.531426869228478e-05,
      "loss": 1.0781,
      "num_input_tokens_seen": 5507032,
      "step": 9485
    },
    {
      "epoch": 1.413464402740542,
      "grad_norm": 12.5,
      "learning_rate": 3.533288650580876e-05,
      "loss": 1.1313,
      "num_input_tokens_seen": 5509944,
      "step": 9490
    },
    {
      "epoch": 1.4142091152815013,
      "grad_norm": 8.3125,
      "learning_rate": 3.535150431933274e-05,
      "loss": 1.0186,
      "num_input_tokens_seen": 5512920,
      "step": 9495
    },
    {
      "epoch": 1.4149538278224605,
      "grad_norm": 12.1875,
      "learning_rate": 3.537012213285672e-05,
      "loss": 1.2544,
      "num_input_tokens_seen": 5515960,
      "step": 9500
    },
    {
      "epoch": 1.4156985403634197,
      "grad_norm": 15.4375,
      "learning_rate": 3.53887399463807e-05,
      "loss": 1.0881,
      "num_input_tokens_seen": 5518968,
      "step": 9505
    },
    {
      "epoch": 1.416443252904379,
      "grad_norm": 18.625,
      "learning_rate": 3.5407357759904683e-05,
      "loss": 1.1376,
      "num_input_tokens_seen": 5521816,
      "step": 9510
    },
    {
      "epoch": 1.417187965445338,
      "grad_norm": 15.0625,
      "learning_rate": 3.542597557342866e-05,
      "loss": 0.9614,
      "num_input_tokens_seen": 5524504,
      "step": 9515
    },
    {
      "epoch": 1.4179326779862973,
      "grad_norm": 8.8125,
      "learning_rate": 3.544459338695264e-05,
      "loss": 1.2785,
      "num_input_tokens_seen": 5527320,
      "step": 9520
    },
    {
      "epoch": 1.4186773905272565,
      "grad_norm": 14.25,
      "learning_rate": 3.546321120047662e-05,
      "loss": 1.2274,
      "num_input_tokens_seen": 5530040,
      "step": 9525
    },
    {
      "epoch": 1.4194221030682157,
      "grad_norm": 7.3125,
      "learning_rate": 3.5481829014000596e-05,
      "loss": 1.1645,
      "num_input_tokens_seen": 5532888,
      "step": 9530
    },
    {
      "epoch": 1.420166815609175,
      "grad_norm": 10.3125,
      "learning_rate": 3.5500446827524575e-05,
      "loss": 1.1347,
      "num_input_tokens_seen": 5535992,
      "step": 9535
    },
    {
      "epoch": 1.420911528150134,
      "grad_norm": 6.3125,
      "learning_rate": 3.551906464104855e-05,
      "loss": 1.0684,
      "num_input_tokens_seen": 5538872,
      "step": 9540
    },
    {
      "epoch": 1.4216562406910933,
      "grad_norm": 12.625,
      "learning_rate": 3.553768245457254e-05,
      "loss": 1.1537,
      "num_input_tokens_seen": 5541720,
      "step": 9545
    },
    {
      "epoch": 1.4224009532320525,
      "grad_norm": 11.4375,
      "learning_rate": 3.5556300268096516e-05,
      "loss": 1.4267,
      "num_input_tokens_seen": 5544920,
      "step": 9550
    },
    {
      "epoch": 1.4231456657730117,
      "grad_norm": 20.375,
      "learning_rate": 3.5574918081620494e-05,
      "loss": 1.2798,
      "num_input_tokens_seen": 5547864,
      "step": 9555
    },
    {
      "epoch": 1.423890378313971,
      "grad_norm": 13.625,
      "learning_rate": 3.559353589514447e-05,
      "loss": 1.1633,
      "num_input_tokens_seen": 5550744,
      "step": 9560
    },
    {
      "epoch": 1.4246350908549301,
      "grad_norm": 16.375,
      "learning_rate": 3.561215370866846e-05,
      "loss": 1.2817,
      "num_input_tokens_seen": 5553560,
      "step": 9565
    },
    {
      "epoch": 1.4253798033958893,
      "grad_norm": 10.25,
      "learning_rate": 3.5630771522192436e-05,
      "loss": 0.9688,
      "num_input_tokens_seen": 5556408,
      "step": 9570
    },
    {
      "epoch": 1.4261245159368483,
      "grad_norm": 12.6875,
      "learning_rate": 3.5649389335716414e-05,
      "loss": 1.3765,
      "num_input_tokens_seen": 5559256,
      "step": 9575
    },
    {
      "epoch": 1.4268692284778075,
      "grad_norm": 12.1875,
      "learning_rate": 3.566800714924039e-05,
      "loss": 0.7981,
      "num_input_tokens_seen": 5562040,
      "step": 9580
    },
    {
      "epoch": 1.4276139410187667,
      "grad_norm": 7.59375,
      "learning_rate": 3.568662496276438e-05,
      "loss": 1.0076,
      "num_input_tokens_seen": 5564984,
      "step": 9585
    },
    {
      "epoch": 1.428358653559726,
      "grad_norm": 24.75,
      "learning_rate": 3.5705242776288356e-05,
      "loss": 1.4287,
      "num_input_tokens_seen": 5567640,
      "step": 9590
    },
    {
      "epoch": 1.4291033661006851,
      "grad_norm": 18.875,
      "learning_rate": 3.5723860589812334e-05,
      "loss": 1.2565,
      "num_input_tokens_seen": 5570456,
      "step": 9595
    },
    {
      "epoch": 1.4298480786416443,
      "grad_norm": 11.0625,
      "learning_rate": 3.574247840333632e-05,
      "loss": 1.1769,
      "num_input_tokens_seen": 5573048,
      "step": 9600
    },
    {
      "epoch": 1.4305927911826035,
      "grad_norm": 9.5625,
      "learning_rate": 3.57610962168603e-05,
      "loss": 0.8997,
      "num_input_tokens_seen": 5575832,
      "step": 9605
    },
    {
      "epoch": 1.4313375037235627,
      "grad_norm": 15.9375,
      "learning_rate": 3.5779714030384275e-05,
      "loss": 1.1643,
      "num_input_tokens_seen": 5578456,
      "step": 9610
    },
    {
      "epoch": 1.432082216264522,
      "grad_norm": 9.25,
      "learning_rate": 3.5798331843908253e-05,
      "loss": 0.9832,
      "num_input_tokens_seen": 5581560,
      "step": 9615
    },
    {
      "epoch": 1.4328269288054811,
      "grad_norm": 10.9375,
      "learning_rate": 3.581694965743223e-05,
      "loss": 0.9753,
      "num_input_tokens_seen": 5584280,
      "step": 9620
    },
    {
      "epoch": 1.4335716413464403,
      "grad_norm": 15.125,
      "learning_rate": 3.583556747095621e-05,
      "loss": 1.0727,
      "num_input_tokens_seen": 5586968,
      "step": 9625
    },
    {
      "epoch": 1.4343163538873995,
      "grad_norm": 6.15625,
      "learning_rate": 3.585418528448019e-05,
      "loss": 0.9122,
      "num_input_tokens_seen": 5589688,
      "step": 9630
    },
    {
      "epoch": 1.4350610664283587,
      "grad_norm": 10.8125,
      "learning_rate": 3.587280309800417e-05,
      "loss": 1.3157,
      "num_input_tokens_seen": 5592504,
      "step": 9635
    },
    {
      "epoch": 1.4358057789693177,
      "grad_norm": 12.9375,
      "learning_rate": 3.589142091152815e-05,
      "loss": 0.9948,
      "num_input_tokens_seen": 5595320,
      "step": 9640
    },
    {
      "epoch": 1.436550491510277,
      "grad_norm": 15.25,
      "learning_rate": 3.591003872505213e-05,
      "loss": 1.1212,
      "num_input_tokens_seen": 5598296,
      "step": 9645
    },
    {
      "epoch": 1.4372952040512361,
      "grad_norm": 10.875,
      "learning_rate": 3.592865653857611e-05,
      "loss": 0.8598,
      "num_input_tokens_seen": 5601240,
      "step": 9650
    },
    {
      "epoch": 1.4380399165921953,
      "grad_norm": 11.75,
      "learning_rate": 3.594727435210009e-05,
      "loss": 1.1645,
      "num_input_tokens_seen": 5604152,
      "step": 9655
    },
    {
      "epoch": 1.4387846291331545,
      "grad_norm": 19.0,
      "learning_rate": 3.596589216562407e-05,
      "loss": 0.9102,
      "num_input_tokens_seen": 5607000,
      "step": 9660
    },
    {
      "epoch": 1.4395293416741137,
      "grad_norm": 20.75,
      "learning_rate": 3.598450997914805e-05,
      "loss": 0.9127,
      "num_input_tokens_seen": 5609720,
      "step": 9665
    },
    {
      "epoch": 1.440274054215073,
      "grad_norm": 8.5,
      "learning_rate": 3.6003127792672034e-05,
      "loss": 1.1335,
      "num_input_tokens_seen": 5612632,
      "step": 9670
    },
    {
      "epoch": 1.4410187667560321,
      "grad_norm": 12.4375,
      "learning_rate": 3.602174560619601e-05,
      "loss": 0.9571,
      "num_input_tokens_seen": 5615480,
      "step": 9675
    },
    {
      "epoch": 1.4417634792969913,
      "grad_norm": 8.875,
      "learning_rate": 3.604036341971999e-05,
      "loss": 1.1189,
      "num_input_tokens_seen": 5618200,
      "step": 9680
    },
    {
      "epoch": 1.4425081918379505,
      "grad_norm": 13.625,
      "learning_rate": 3.605898123324397e-05,
      "loss": 1.1142,
      "num_input_tokens_seen": 5621048,
      "step": 9685
    },
    {
      "epoch": 1.4432529043789097,
      "grad_norm": 10.0625,
      "learning_rate": 3.6077599046767954e-05,
      "loss": 1.3345,
      "num_input_tokens_seen": 5624088,
      "step": 9690
    },
    {
      "epoch": 1.443997616919869,
      "grad_norm": 11.125,
      "learning_rate": 3.609621686029193e-05,
      "loss": 0.9635,
      "num_input_tokens_seen": 5627064,
      "step": 9695
    },
    {
      "epoch": 1.4447423294608281,
      "grad_norm": 11.3125,
      "learning_rate": 3.6114834673815904e-05,
      "loss": 1.1461,
      "num_input_tokens_seen": 5630456,
      "step": 9700
    },
    {
      "epoch": 1.4454870420017873,
      "grad_norm": 23.75,
      "learning_rate": 3.613345248733989e-05,
      "loss": 1.2235,
      "num_input_tokens_seen": 5633464,
      "step": 9705
    },
    {
      "epoch": 1.4462317545427466,
      "grad_norm": 7.1875,
      "learning_rate": 3.615207030086387e-05,
      "loss": 1.0716,
      "num_input_tokens_seen": 5636376,
      "step": 9710
    },
    {
      "epoch": 1.4469764670837058,
      "grad_norm": 20.25,
      "learning_rate": 3.6170688114387845e-05,
      "loss": 1.5764,
      "num_input_tokens_seen": 5639256,
      "step": 9715
    },
    {
      "epoch": 1.447721179624665,
      "grad_norm": 8.625,
      "learning_rate": 3.6189305927911823e-05,
      "loss": 1.0906,
      "num_input_tokens_seen": 5641944,
      "step": 9720
    },
    {
      "epoch": 1.4484658921656242,
      "grad_norm": 7.9375,
      "learning_rate": 3.620792374143581e-05,
      "loss": 1.1246,
      "num_input_tokens_seen": 5644696,
      "step": 9725
    },
    {
      "epoch": 1.4492106047065834,
      "grad_norm": 10.1875,
      "learning_rate": 3.622654155495979e-05,
      "loss": 1.2078,
      "num_input_tokens_seen": 5647800,
      "step": 9730
    },
    {
      "epoch": 1.4499553172475426,
      "grad_norm": 10.5,
      "learning_rate": 3.6245159368483765e-05,
      "loss": 0.8114,
      "num_input_tokens_seen": 5650776,
      "step": 9735
    },
    {
      "epoch": 1.4507000297885018,
      "grad_norm": 10.5625,
      "learning_rate": 3.626377718200774e-05,
      "loss": 0.9589,
      "num_input_tokens_seen": 5653592,
      "step": 9740
    },
    {
      "epoch": 1.4514447423294607,
      "grad_norm": 5.5,
      "learning_rate": 3.628239499553173e-05,
      "loss": 1.0589,
      "num_input_tokens_seen": 5656568,
      "step": 9745
    },
    {
      "epoch": 1.45218945487042,
      "grad_norm": 17.375,
      "learning_rate": 3.6301012809055706e-05,
      "loss": 1.2711,
      "num_input_tokens_seen": 5659256,
      "step": 9750
    },
    {
      "epoch": 1.4529341674113792,
      "grad_norm": 9.875,
      "learning_rate": 3.6319630622579685e-05,
      "loss": 1.2978,
      "num_input_tokens_seen": 5662392,
      "step": 9755
    },
    {
      "epoch": 1.4536788799523384,
      "grad_norm": 9.625,
      "learning_rate": 3.633824843610367e-05,
      "loss": 0.9974,
      "num_input_tokens_seen": 5665112,
      "step": 9760
    },
    {
      "epoch": 1.4544235924932976,
      "grad_norm": 10.125,
      "learning_rate": 3.635686624962765e-05,
      "loss": 1.1609,
      "num_input_tokens_seen": 5667928,
      "step": 9765
    },
    {
      "epoch": 1.4551683050342568,
      "grad_norm": 21.75,
      "learning_rate": 3.6375484063151626e-05,
      "loss": 1.3465,
      "num_input_tokens_seen": 5670968,
      "step": 9770
    },
    {
      "epoch": 1.455913017575216,
      "grad_norm": 19.25,
      "learning_rate": 3.6394101876675604e-05,
      "loss": 1.303,
      "num_input_tokens_seen": 5674008,
      "step": 9775
    },
    {
      "epoch": 1.4566577301161752,
      "grad_norm": 6.9375,
      "learning_rate": 3.641271969019959e-05,
      "loss": 1.118,
      "num_input_tokens_seen": 5677016,
      "step": 9780
    },
    {
      "epoch": 1.4574024426571344,
      "grad_norm": 20.75,
      "learning_rate": 3.643133750372357e-05,
      "loss": 1.6007,
      "num_input_tokens_seen": 5680216,
      "step": 9785
    },
    {
      "epoch": 1.4581471551980936,
      "grad_norm": 10.25,
      "learning_rate": 3.644995531724754e-05,
      "loss": 1.4847,
      "num_input_tokens_seen": 5683320,
      "step": 9790
    },
    {
      "epoch": 1.4588918677390528,
      "grad_norm": 14.4375,
      "learning_rate": 3.6468573130771524e-05,
      "loss": 1.0973,
      "num_input_tokens_seen": 5685880,
      "step": 9795
    },
    {
      "epoch": 1.459636580280012,
      "grad_norm": 20.375,
      "learning_rate": 3.64871909442955e-05,
      "loss": 1.3092,
      "num_input_tokens_seen": 5688792,
      "step": 9800
    },
    {
      "epoch": 1.4603812928209712,
      "grad_norm": 11.1875,
      "learning_rate": 3.650580875781948e-05,
      "loss": 1.2531,
      "num_input_tokens_seen": 5691608,
      "step": 9805
    },
    {
      "epoch": 1.4611260053619302,
      "grad_norm": 6.65625,
      "learning_rate": 3.652442657134346e-05,
      "loss": 1.3537,
      "num_input_tokens_seen": 5694456,
      "step": 9810
    },
    {
      "epoch": 1.4618707179028894,
      "grad_norm": 9.1875,
      "learning_rate": 3.6543044384867444e-05,
      "loss": 1.177,
      "num_input_tokens_seen": 5697496,
      "step": 9815
    },
    {
      "epoch": 1.4626154304438486,
      "grad_norm": 14.25,
      "learning_rate": 3.656166219839142e-05,
      "loss": 1.3924,
      "num_input_tokens_seen": 5700440,
      "step": 9820
    },
    {
      "epoch": 1.4633601429848078,
      "grad_norm": 13.375,
      "learning_rate": 3.65802800119154e-05,
      "loss": 1.3916,
      "num_input_tokens_seen": 5703032,
      "step": 9825
    },
    {
      "epoch": 1.464104855525767,
      "grad_norm": 5.90625,
      "learning_rate": 3.6598897825439385e-05,
      "loss": 1.336,
      "num_input_tokens_seen": 5705784,
      "step": 9830
    },
    {
      "epoch": 1.4648495680667262,
      "grad_norm": 11.0,
      "learning_rate": 3.6617515638963363e-05,
      "loss": 0.9319,
      "num_input_tokens_seen": 5708504,
      "step": 9835
    },
    {
      "epoch": 1.4655942806076854,
      "grad_norm": 14.75,
      "learning_rate": 3.663613345248734e-05,
      "loss": 1.1559,
      "num_input_tokens_seen": 5711352,
      "step": 9840
    },
    {
      "epoch": 1.4663389931486446,
      "grad_norm": 9.1875,
      "learning_rate": 3.665475126601132e-05,
      "loss": 0.9455,
      "num_input_tokens_seen": 5714072,
      "step": 9845
    },
    {
      "epoch": 1.4670837056896038,
      "grad_norm": 8.0625,
      "learning_rate": 3.6673369079535305e-05,
      "loss": 1.2885,
      "num_input_tokens_seen": 5717432,
      "step": 9850
    },
    {
      "epoch": 1.467828418230563,
      "grad_norm": 12.25,
      "learning_rate": 3.669198689305928e-05,
      "loss": 1.3348,
      "num_input_tokens_seen": 5720120,
      "step": 9855
    },
    {
      "epoch": 1.4685731307715222,
      "grad_norm": 7.25,
      "learning_rate": 3.671060470658326e-05,
      "loss": 1.19,
      "num_input_tokens_seen": 5723096,
      "step": 9860
    },
    {
      "epoch": 1.4693178433124814,
      "grad_norm": 5.75,
      "learning_rate": 3.672922252010724e-05,
      "loss": 0.9529,
      "num_input_tokens_seen": 5726040,
      "step": 9865
    },
    {
      "epoch": 1.4700625558534406,
      "grad_norm": 6.15625,
      "learning_rate": 3.6747840333631225e-05,
      "loss": 1.1534,
      "num_input_tokens_seen": 5729240,
      "step": 9870
    },
    {
      "epoch": 1.4708072683943998,
      "grad_norm": 5.6875,
      "learning_rate": 3.6766458147155196e-05,
      "loss": 1.0677,
      "num_input_tokens_seen": 5732248,
      "step": 9875
    },
    {
      "epoch": 1.471551980935359,
      "grad_norm": 15.8125,
      "learning_rate": 3.6785075960679174e-05,
      "loss": 1.1157,
      "num_input_tokens_seen": 5735288,
      "step": 9880
    },
    {
      "epoch": 1.4722966934763182,
      "grad_norm": 8.6875,
      "learning_rate": 3.680369377420316e-05,
      "loss": 1.0138,
      "num_input_tokens_seen": 5738264,
      "step": 9885
    },
    {
      "epoch": 1.4730414060172774,
      "grad_norm": 16.375,
      "learning_rate": 3.682231158772714e-05,
      "loss": 1.1884,
      "num_input_tokens_seen": 5741432,
      "step": 9890
    },
    {
      "epoch": 1.4737861185582366,
      "grad_norm": 25.25,
      "learning_rate": 3.6840929401251116e-05,
      "loss": 1.5389,
      "num_input_tokens_seen": 5744344,
      "step": 9895
    },
    {
      "epoch": 1.4745308310991958,
      "grad_norm": 9.8125,
      "learning_rate": 3.68595472147751e-05,
      "loss": 1.3275,
      "num_input_tokens_seen": 5747416,
      "step": 9900
    },
    {
      "epoch": 1.475275543640155,
      "grad_norm": 5.40625,
      "learning_rate": 3.687816502829908e-05,
      "loss": 0.6777,
      "num_input_tokens_seen": 5750040,
      "step": 9905
    },
    {
      "epoch": 1.4760202561811142,
      "grad_norm": 9.375,
      "learning_rate": 3.689678284182306e-05,
      "loss": 0.9677,
      "num_input_tokens_seen": 5752856,
      "step": 9910
    },
    {
      "epoch": 1.4767649687220734,
      "grad_norm": 13.625,
      "learning_rate": 3.6915400655347035e-05,
      "loss": 1.1734,
      "num_input_tokens_seen": 5755864,
      "step": 9915
    },
    {
      "epoch": 1.4775096812630324,
      "grad_norm": 10.25,
      "learning_rate": 3.693401846887102e-05,
      "loss": 1.0797,
      "num_input_tokens_seen": 5758584,
      "step": 9920
    },
    {
      "epoch": 1.4782543938039916,
      "grad_norm": 9.3125,
      "learning_rate": 3.6952636282395e-05,
      "loss": 0.8041,
      "num_input_tokens_seen": 5761464,
      "step": 9925
    },
    {
      "epoch": 1.4789991063449508,
      "grad_norm": 15.375,
      "learning_rate": 3.697125409591898e-05,
      "loss": 1.3413,
      "num_input_tokens_seen": 5764120,
      "step": 9930
    },
    {
      "epoch": 1.47974381888591,
      "grad_norm": 10.875,
      "learning_rate": 3.6989871909442955e-05,
      "loss": 1.0224,
      "num_input_tokens_seen": 5766840,
      "step": 9935
    },
    {
      "epoch": 1.4804885314268692,
      "grad_norm": 6.78125,
      "learning_rate": 3.700848972296694e-05,
      "loss": 0.8128,
      "num_input_tokens_seen": 5770296,
      "step": 9940
    },
    {
      "epoch": 1.4812332439678284,
      "grad_norm": 9.125,
      "learning_rate": 3.702710753649092e-05,
      "loss": 1.069,
      "num_input_tokens_seen": 5772984,
      "step": 9945
    },
    {
      "epoch": 1.4819779565087876,
      "grad_norm": 10.375,
      "learning_rate": 3.70457253500149e-05,
      "loss": 1.0282,
      "num_input_tokens_seen": 5775800,
      "step": 9950
    },
    {
      "epoch": 1.4827226690497468,
      "grad_norm": 13.625,
      "learning_rate": 3.7064343163538875e-05,
      "loss": 0.7102,
      "num_input_tokens_seen": 5778712,
      "step": 9955
    },
    {
      "epoch": 1.483467381590706,
      "grad_norm": 19.0,
      "learning_rate": 3.708296097706285e-05,
      "loss": 1.2844,
      "num_input_tokens_seen": 5781528,
      "step": 9960
    },
    {
      "epoch": 1.4842120941316652,
      "grad_norm": 5.15625,
      "learning_rate": 3.710157879058683e-05,
      "loss": 1.1686,
      "num_input_tokens_seen": 5784728,
      "step": 9965
    },
    {
      "epoch": 1.4849568066726244,
      "grad_norm": 8.0625,
      "learning_rate": 3.712019660411081e-05,
      "loss": 0.9944,
      "num_input_tokens_seen": 5787800,
      "step": 9970
    },
    {
      "epoch": 1.4857015192135836,
      "grad_norm": 6.25,
      "learning_rate": 3.7138814417634795e-05,
      "loss": 0.8803,
      "num_input_tokens_seen": 5791064,
      "step": 9975
    },
    {
      "epoch": 1.4864462317545428,
      "grad_norm": 13.25,
      "learning_rate": 3.715743223115877e-05,
      "loss": 1.0336,
      "num_input_tokens_seen": 5793880,
      "step": 9980
    },
    {
      "epoch": 1.4871909442955018,
      "grad_norm": 10.5,
      "learning_rate": 3.717605004468275e-05,
      "loss": 1.3717,
      "num_input_tokens_seen": 5796696,
      "step": 9985
    },
    {
      "epoch": 1.487935656836461,
      "grad_norm": 4.28125,
      "learning_rate": 3.7194667858206736e-05,
      "loss": 1.2194,
      "num_input_tokens_seen": 5799768,
      "step": 9990
    },
    {
      "epoch": 1.4886803693774202,
      "grad_norm": 17.5,
      "learning_rate": 3.7213285671730714e-05,
      "loss": 1.2605,
      "num_input_tokens_seen": 5802488,
      "step": 9995
    },
    {
      "epoch": 1.4894250819183794,
      "grad_norm": 11.25,
      "learning_rate": 3.723190348525469e-05,
      "loss": 0.904,
      "num_input_tokens_seen": 5805560,
      "step": 10000
    },
    {
      "epoch": 1.4901697944593386,
      "grad_norm": 11.4375,
      "learning_rate": 3.725052129877867e-05,
      "loss": 1.1862,
      "num_input_tokens_seen": 5808504,
      "step": 10005
    },
    {
      "epoch": 1.4909145070002978,
      "grad_norm": 8.875,
      "learning_rate": 3.7269139112302656e-05,
      "loss": 1.2042,
      "num_input_tokens_seen": 5811000,
      "step": 10010
    },
    {
      "epoch": 1.491659219541257,
      "grad_norm": 11.8125,
      "learning_rate": 3.7287756925826634e-05,
      "loss": 1.0614,
      "num_input_tokens_seen": 5813720,
      "step": 10015
    },
    {
      "epoch": 1.4924039320822162,
      "grad_norm": 10.125,
      "learning_rate": 3.730637473935061e-05,
      "loss": 1.0959,
      "num_input_tokens_seen": 5816792,
      "step": 10020
    },
    {
      "epoch": 1.4931486446231754,
      "grad_norm": 10.75,
      "learning_rate": 3.732499255287459e-05,
      "loss": 1.1506,
      "num_input_tokens_seen": 5819544,
      "step": 10025
    },
    {
      "epoch": 1.4938933571641346,
      "grad_norm": 9.5,
      "learning_rate": 3.7343610366398575e-05,
      "loss": 1.2078,
      "num_input_tokens_seen": 5822936,
      "step": 10030
    },
    {
      "epoch": 1.4946380697050938,
      "grad_norm": 6.875,
      "learning_rate": 3.7362228179922554e-05,
      "loss": 1.1126,
      "num_input_tokens_seen": 5825720,
      "step": 10035
    },
    {
      "epoch": 1.495382782246053,
      "grad_norm": 9.5,
      "learning_rate": 3.738084599344653e-05,
      "loss": 1.4645,
      "num_input_tokens_seen": 5829240,
      "step": 10040
    },
    {
      "epoch": 1.4961274947870122,
      "grad_norm": 18.25,
      "learning_rate": 3.739946380697052e-05,
      "loss": 1.0572,
      "num_input_tokens_seen": 5832280,
      "step": 10045
    },
    {
      "epoch": 1.4968722073279714,
      "grad_norm": 19.0,
      "learning_rate": 3.741808162049449e-05,
      "loss": 1.306,
      "num_input_tokens_seen": 5835160,
      "step": 10050
    },
    {
      "epoch": 1.4976169198689306,
      "grad_norm": 33.25,
      "learning_rate": 3.743669943401847e-05,
      "loss": 0.9859,
      "num_input_tokens_seen": 5838072,
      "step": 10055
    },
    {
      "epoch": 1.4983616324098898,
      "grad_norm": 9.5,
      "learning_rate": 3.745531724754245e-05,
      "loss": 0.8197,
      "num_input_tokens_seen": 5840888,
      "step": 10060
    },
    {
      "epoch": 1.499106344950849,
      "grad_norm": 18.125,
      "learning_rate": 3.747393506106643e-05,
      "loss": 1.2511,
      "num_input_tokens_seen": 5843864,
      "step": 10065
    },
    {
      "epoch": 1.4998510574918082,
      "grad_norm": 11.0,
      "learning_rate": 3.749255287459041e-05,
      "loss": 1.5245,
      "num_input_tokens_seen": 5846360,
      "step": 10070
    },
    {
      "epoch": 1.5005957700327675,
      "grad_norm": 13.0625,
      "learning_rate": 3.7511170688114386e-05,
      "loss": 1.1077,
      "num_input_tokens_seen": 5849304,
      "step": 10075
    },
    {
      "epoch": 1.5013404825737267,
      "grad_norm": 11.375,
      "learning_rate": 3.752978850163837e-05,
      "loss": 1.0552,
      "num_input_tokens_seen": 5852408,
      "step": 10080
    },
    {
      "epoch": 1.5020851951146859,
      "grad_norm": 18.25,
      "learning_rate": 3.754840631516235e-05,
      "loss": 1.2561,
      "num_input_tokens_seen": 5855576,
      "step": 10085
    },
    {
      "epoch": 1.502829907655645,
      "grad_norm": 10.6875,
      "learning_rate": 3.756702412868633e-05,
      "loss": 1.0703,
      "num_input_tokens_seen": 5858520,
      "step": 10090
    },
    {
      "epoch": 1.5035746201966043,
      "grad_norm": 8.8125,
      "learning_rate": 3.7585641942210306e-05,
      "loss": 0.8829,
      "num_input_tokens_seen": 5861304,
      "step": 10095
    },
    {
      "epoch": 1.5043193327375635,
      "grad_norm": 11.125,
      "learning_rate": 3.760425975573429e-05,
      "loss": 1.0509,
      "num_input_tokens_seen": 5864184,
      "step": 10100
    },
    {
      "epoch": 1.5050640452785224,
      "grad_norm": 10.0625,
      "learning_rate": 3.762287756925827e-05,
      "loss": 1.0346,
      "num_input_tokens_seen": 5866968,
      "step": 10105
    },
    {
      "epoch": 1.5058087578194816,
      "grad_norm": 15.4375,
      "learning_rate": 3.764149538278225e-05,
      "loss": 0.9829,
      "num_input_tokens_seen": 5869912,
      "step": 10110
    },
    {
      "epoch": 1.5065534703604408,
      "grad_norm": 17.0,
      "learning_rate": 3.766011319630623e-05,
      "loss": 1.234,
      "num_input_tokens_seen": 5872984,
      "step": 10115
    },
    {
      "epoch": 1.5072981829014,
      "grad_norm": 10.875,
      "learning_rate": 3.767873100983021e-05,
      "loss": 1.071,
      "num_input_tokens_seen": 5875960,
      "step": 10120
    },
    {
      "epoch": 1.5080428954423593,
      "grad_norm": 13.5625,
      "learning_rate": 3.769734882335419e-05,
      "loss": 0.9912,
      "num_input_tokens_seen": 5879128,
      "step": 10125
    },
    {
      "epoch": 1.5087876079833185,
      "grad_norm": 21.5,
      "learning_rate": 3.771596663687817e-05,
      "loss": 0.9512,
      "num_input_tokens_seen": 5882040,
      "step": 10130
    },
    {
      "epoch": 1.5095323205242777,
      "grad_norm": 13.4375,
      "learning_rate": 3.7734584450402145e-05,
      "loss": 1.2628,
      "num_input_tokens_seen": 5884536,
      "step": 10135
    },
    {
      "epoch": 1.5102770330652369,
      "grad_norm": 10.375,
      "learning_rate": 3.7753202263926124e-05,
      "loss": 1.1704,
      "num_input_tokens_seen": 5887608,
      "step": 10140
    },
    {
      "epoch": 1.5110217456061958,
      "grad_norm": 11.4375,
      "learning_rate": 3.77718200774501e-05,
      "loss": 1.0175,
      "num_input_tokens_seen": 5890392,
      "step": 10145
    },
    {
      "epoch": 1.511766458147155,
      "grad_norm": 8.625,
      "learning_rate": 3.779043789097409e-05,
      "loss": 1.2188,
      "num_input_tokens_seen": 5893400,
      "step": 10150
    },
    {
      "epoch": 1.5125111706881142,
      "grad_norm": 14.875,
      "learning_rate": 3.7809055704498065e-05,
      "loss": 1.1651,
      "num_input_tokens_seen": 5896280,
      "step": 10155
    },
    {
      "epoch": 1.5132558832290735,
      "grad_norm": 14.8125,
      "learning_rate": 3.7827673518022043e-05,
      "loss": 1.1645,
      "num_input_tokens_seen": 5899000,
      "step": 10160
    },
    {
      "epoch": 1.5140005957700327,
      "grad_norm": 5.0625,
      "learning_rate": 3.784629133154602e-05,
      "loss": 1.1172,
      "num_input_tokens_seen": 5901720,
      "step": 10165
    },
    {
      "epoch": 1.5147453083109919,
      "grad_norm": 20.75,
      "learning_rate": 3.786490914507001e-05,
      "loss": 1.054,
      "num_input_tokens_seen": 5904568,
      "step": 10170
    },
    {
      "epoch": 1.515490020851951,
      "grad_norm": 10.9375,
      "learning_rate": 3.7883526958593985e-05,
      "loss": 1.2048,
      "num_input_tokens_seen": 5907416,
      "step": 10175
    },
    {
      "epoch": 1.5162347333929103,
      "grad_norm": 12.625,
      "learning_rate": 3.790214477211796e-05,
      "loss": 1.1687,
      "num_input_tokens_seen": 5910360,
      "step": 10180
    },
    {
      "epoch": 1.5169794459338695,
      "grad_norm": 17.875,
      "learning_rate": 3.792076258564194e-05,
      "loss": 0.9904,
      "num_input_tokens_seen": 5913240,
      "step": 10185
    },
    {
      "epoch": 1.5177241584748287,
      "grad_norm": 7.875,
      "learning_rate": 3.7939380399165926e-05,
      "loss": 0.9854,
      "num_input_tokens_seen": 5915800,
      "step": 10190
    },
    {
      "epoch": 1.5184688710157879,
      "grad_norm": 4.5625,
      "learning_rate": 3.7957998212689905e-05,
      "loss": 1.2032,
      "num_input_tokens_seen": 5918552,
      "step": 10195
    },
    {
      "epoch": 1.519213583556747,
      "grad_norm": 8.9375,
      "learning_rate": 3.797661602621388e-05,
      "loss": 1.2075,
      "num_input_tokens_seen": 5921720,
      "step": 10200
    },
    {
      "epoch": 1.5199582960977063,
      "grad_norm": 9.0,
      "learning_rate": 3.799523383973787e-05,
      "loss": 0.9384,
      "num_input_tokens_seen": 5924504,
      "step": 10205
    },
    {
      "epoch": 1.5207030086386655,
      "grad_norm": 9.5625,
      "learning_rate": 3.8013851653261846e-05,
      "loss": 1.0537,
      "num_input_tokens_seen": 5927384,
      "step": 10210
    },
    {
      "epoch": 1.5214477211796247,
      "grad_norm": 8.625,
      "learning_rate": 3.8032469466785824e-05,
      "loss": 0.8727,
      "num_input_tokens_seen": 5930008,
      "step": 10215
    },
    {
      "epoch": 1.5221924337205839,
      "grad_norm": 14.625,
      "learning_rate": 3.80510872803098e-05,
      "loss": 1.3927,
      "num_input_tokens_seen": 5933016,
      "step": 10220
    },
    {
      "epoch": 1.522937146261543,
      "grad_norm": 7.0625,
      "learning_rate": 3.806970509383378e-05,
      "loss": 1.0453,
      "num_input_tokens_seen": 5936088,
      "step": 10225
    },
    {
      "epoch": 1.5236818588025023,
      "grad_norm": 11.75,
      "learning_rate": 3.808832290735776e-05,
      "loss": 1.09,
      "num_input_tokens_seen": 5939032,
      "step": 10230
    },
    {
      "epoch": 1.5244265713434615,
      "grad_norm": 14.5625,
      "learning_rate": 3.810694072088174e-05,
      "loss": 1.4305,
      "num_input_tokens_seen": 5941912,
      "step": 10235
    },
    {
      "epoch": 1.5251712838844207,
      "grad_norm": 14.9375,
      "learning_rate": 3.812555853440572e-05,
      "loss": 1.4016,
      "num_input_tokens_seen": 5945080,
      "step": 10240
    },
    {
      "epoch": 1.52591599642538,
      "grad_norm": 8.6875,
      "learning_rate": 3.81441763479297e-05,
      "loss": 0.9312,
      "num_input_tokens_seen": 5947832,
      "step": 10245
    },
    {
      "epoch": 1.526660708966339,
      "grad_norm": 10.6875,
      "learning_rate": 3.816279416145368e-05,
      "loss": 0.931,
      "num_input_tokens_seen": 5950808,
      "step": 10250
    },
    {
      "epoch": 1.5274054215072983,
      "grad_norm": 15.9375,
      "learning_rate": 3.818141197497766e-05,
      "loss": 1.2516,
      "num_input_tokens_seen": 5953528,
      "step": 10255
    },
    {
      "epoch": 1.5281501340482575,
      "grad_norm": 9.4375,
      "learning_rate": 3.820002978850164e-05,
      "loss": 1.0988,
      "num_input_tokens_seen": 5956216,
      "step": 10260
    },
    {
      "epoch": 1.5288948465892167,
      "grad_norm": 14.8125,
      "learning_rate": 3.821864760202562e-05,
      "loss": 1.1601,
      "num_input_tokens_seen": 5958840,
      "step": 10265
    },
    {
      "epoch": 1.529639559130176,
      "grad_norm": 11.0625,
      "learning_rate": 3.82372654155496e-05,
      "loss": 0.9042,
      "num_input_tokens_seen": 5961912,
      "step": 10270
    },
    {
      "epoch": 1.5303842716711349,
      "grad_norm": 11.3125,
      "learning_rate": 3.8255883229073583e-05,
      "loss": 1.5789,
      "num_input_tokens_seen": 5964696,
      "step": 10275
    },
    {
      "epoch": 1.531128984212094,
      "grad_norm": 10.1875,
      "learning_rate": 3.827450104259756e-05,
      "loss": 1.1619,
      "num_input_tokens_seen": 5967576,
      "step": 10280
    },
    {
      "epoch": 1.5318736967530533,
      "grad_norm": 10.5,
      "learning_rate": 3.829311885612154e-05,
      "loss": 1.047,
      "num_input_tokens_seen": 5970360,
      "step": 10285
    },
    {
      "epoch": 1.5326184092940125,
      "grad_norm": 20.125,
      "learning_rate": 3.831173666964552e-05,
      "loss": 1.2013,
      "num_input_tokens_seen": 5973016,
      "step": 10290
    },
    {
      "epoch": 1.5333631218349717,
      "grad_norm": 26.625,
      "learning_rate": 3.83303544831695e-05,
      "loss": 1.3881,
      "num_input_tokens_seen": 5976216,
      "step": 10295
    },
    {
      "epoch": 1.534107834375931,
      "grad_norm": 22.375,
      "learning_rate": 3.834897229669348e-05,
      "loss": 1.0446,
      "num_input_tokens_seen": 5979160,
      "step": 10300
    },
    {
      "epoch": 1.53485254691689,
      "grad_norm": 10.0,
      "learning_rate": 3.836759011021746e-05,
      "loss": 1.6413,
      "num_input_tokens_seen": 5982008,
      "step": 10305
    },
    {
      "epoch": 1.5355972594578493,
      "grad_norm": 23.375,
      "learning_rate": 3.838620792374144e-05,
      "loss": 1.5323,
      "num_input_tokens_seen": 5984920,
      "step": 10310
    },
    {
      "epoch": 1.5363419719988085,
      "grad_norm": 5.53125,
      "learning_rate": 3.8404825737265416e-05,
      "loss": 1.1868,
      "num_input_tokens_seen": 5987864,
      "step": 10315
    },
    {
      "epoch": 1.5370866845397675,
      "grad_norm": 10.375,
      "learning_rate": 3.8423443550789394e-05,
      "loss": 0.7158,
      "num_input_tokens_seen": 5990552,
      "step": 10320
    },
    {
      "epoch": 1.5378313970807267,
      "grad_norm": 7.15625,
      "learning_rate": 3.844206136431337e-05,
      "loss": 1.2573,
      "num_input_tokens_seen": 5993592,
      "step": 10325
    },
    {
      "epoch": 1.538576109621686,
      "grad_norm": 8.0625,
      "learning_rate": 3.846067917783736e-05,
      "loss": 0.9173,
      "num_input_tokens_seen": 5996536,
      "step": 10330
    },
    {
      "epoch": 1.539320822162645,
      "grad_norm": 10.5,
      "learning_rate": 3.8479296991361336e-05,
      "loss": 1.3319,
      "num_input_tokens_seen": 5999256,
      "step": 10335
    },
    {
      "epoch": 1.5400655347036043,
      "grad_norm": 10.0625,
      "learning_rate": 3.8497914804885314e-05,
      "loss": 1.0771,
      "num_input_tokens_seen": 6002008,
      "step": 10340
    },
    {
      "epoch": 1.5408102472445635,
      "grad_norm": 8.25,
      "learning_rate": 3.851653261840929e-05,
      "loss": 1.2232,
      "num_input_tokens_seen": 6004696,
      "step": 10345
    },
    {
      "epoch": 1.5415549597855227,
      "grad_norm": 14.0625,
      "learning_rate": 3.853515043193328e-05,
      "loss": 1.2466,
      "num_input_tokens_seen": 6007736,
      "step": 10350
    },
    {
      "epoch": 1.542299672326482,
      "grad_norm": 14.8125,
      "learning_rate": 3.8553768245457255e-05,
      "loss": 0.9668,
      "num_input_tokens_seen": 6010712,
      "step": 10355
    },
    {
      "epoch": 1.543044384867441,
      "grad_norm": 5.71875,
      "learning_rate": 3.8572386058981234e-05,
      "loss": 1.0126,
      "num_input_tokens_seen": 6013560,
      "step": 10360
    },
    {
      "epoch": 1.5437890974084003,
      "grad_norm": 11.0,
      "learning_rate": 3.859100387250522e-05,
      "loss": 0.9309,
      "num_input_tokens_seen": 6016312,
      "step": 10365
    },
    {
      "epoch": 1.5445338099493595,
      "grad_norm": 9.125,
      "learning_rate": 3.86096216860292e-05,
      "loss": 0.9595,
      "num_input_tokens_seen": 6019480,
      "step": 10370
    },
    {
      "epoch": 1.5452785224903187,
      "grad_norm": 9.875,
      "learning_rate": 3.8628239499553175e-05,
      "loss": 1.0025,
      "num_input_tokens_seen": 6022296,
      "step": 10375
    },
    {
      "epoch": 1.546023235031278,
      "grad_norm": 13.5625,
      "learning_rate": 3.8646857313077153e-05,
      "loss": 1.26,
      "num_input_tokens_seen": 6025208,
      "step": 10380
    },
    {
      "epoch": 1.5467679475722371,
      "grad_norm": 7.28125,
      "learning_rate": 3.866547512660114e-05,
      "loss": 1.0629,
      "num_input_tokens_seen": 6028248,
      "step": 10385
    },
    {
      "epoch": 1.5475126601131963,
      "grad_norm": 9.75,
      "learning_rate": 3.868409294012512e-05,
      "loss": 0.7582,
      "num_input_tokens_seen": 6031256,
      "step": 10390
    },
    {
      "epoch": 1.5482573726541555,
      "grad_norm": 5.9375,
      "learning_rate": 3.870271075364909e-05,
      "loss": 1.3813,
      "num_input_tokens_seen": 6033944,
      "step": 10395
    },
    {
      "epoch": 1.5490020851951147,
      "grad_norm": 10.8125,
      "learning_rate": 3.872132856717307e-05,
      "loss": 1.3407,
      "num_input_tokens_seen": 6036888,
      "step": 10400
    },
    {
      "epoch": 1.549746797736074,
      "grad_norm": 12.0,
      "learning_rate": 3.873994638069705e-05,
      "loss": 1.0432,
      "num_input_tokens_seen": 6039832,
      "step": 10405
    },
    {
      "epoch": 1.5504915102770331,
      "grad_norm": 14.9375,
      "learning_rate": 3.875856419422103e-05,
      "loss": 1.3548,
      "num_input_tokens_seen": 6042776,
      "step": 10410
    },
    {
      "epoch": 1.5512362228179923,
      "grad_norm": 10.625,
      "learning_rate": 3.877718200774501e-05,
      "loss": 1.3953,
      "num_input_tokens_seen": 6045880,
      "step": 10415
    },
    {
      "epoch": 1.5519809353589515,
      "grad_norm": 14.375,
      "learning_rate": 3.879579982126899e-05,
      "loss": 1.3173,
      "num_input_tokens_seen": 6048920,
      "step": 10420
    },
    {
      "epoch": 1.5527256478999107,
      "grad_norm": 5.71875,
      "learning_rate": 3.881441763479297e-05,
      "loss": 1.2655,
      "num_input_tokens_seen": 6051832,
      "step": 10425
    },
    {
      "epoch": 1.55347036044087,
      "grad_norm": 14.6875,
      "learning_rate": 3.883303544831695e-05,
      "loss": 1.6372,
      "num_input_tokens_seen": 6054744,
      "step": 10430
    },
    {
      "epoch": 1.5542150729818291,
      "grad_norm": 7.65625,
      "learning_rate": 3.8851653261840934e-05,
      "loss": 0.8894,
      "num_input_tokens_seen": 6058072,
      "step": 10435
    },
    {
      "epoch": 1.5549597855227884,
      "grad_norm": 9.5625,
      "learning_rate": 3.887027107536491e-05,
      "loss": 0.9295,
      "num_input_tokens_seen": 6060984,
      "step": 10440
    },
    {
      "epoch": 1.5557044980637476,
      "grad_norm": 11.9375,
      "learning_rate": 3.888888888888889e-05,
      "loss": 1.197,
      "num_input_tokens_seen": 6063768,
      "step": 10445
    },
    {
      "epoch": 1.5564492106047065,
      "grad_norm": 11.8125,
      "learning_rate": 3.890750670241287e-05,
      "loss": 1.3144,
      "num_input_tokens_seen": 6066584,
      "step": 10450
    },
    {
      "epoch": 1.5571939231456657,
      "grad_norm": 9.1875,
      "learning_rate": 3.8926124515936854e-05,
      "loss": 1.1748,
      "num_input_tokens_seen": 6069592,
      "step": 10455
    },
    {
      "epoch": 1.557938635686625,
      "grad_norm": 7.1875,
      "learning_rate": 3.894474232946083e-05,
      "loss": 0.9449,
      "num_input_tokens_seen": 6072344,
      "step": 10460
    },
    {
      "epoch": 1.5586833482275841,
      "grad_norm": 10.0625,
      "learning_rate": 3.896336014298481e-05,
      "loss": 1.0438,
      "num_input_tokens_seen": 6075256,
      "step": 10465
    },
    {
      "epoch": 1.5594280607685433,
      "grad_norm": 13.4375,
      "learning_rate": 3.898197795650879e-05,
      "loss": 1.4599,
      "num_input_tokens_seen": 6078168,
      "step": 10470
    },
    {
      "epoch": 1.5601727733095025,
      "grad_norm": 11.0625,
      "learning_rate": 3.9000595770032774e-05,
      "loss": 1.1988,
      "num_input_tokens_seen": 6080824,
      "step": 10475
    },
    {
      "epoch": 1.5609174858504618,
      "grad_norm": 16.125,
      "learning_rate": 3.9019213583556745e-05,
      "loss": 1.2057,
      "num_input_tokens_seen": 6083480,
      "step": 10480
    },
    {
      "epoch": 1.561662198391421,
      "grad_norm": 6.5,
      "learning_rate": 3.903783139708072e-05,
      "loss": 1.016,
      "num_input_tokens_seen": 6086360,
      "step": 10485
    },
    {
      "epoch": 1.5624069109323802,
      "grad_norm": 8.0,
      "learning_rate": 3.905644921060471e-05,
      "loss": 1.0407,
      "num_input_tokens_seen": 6089240,
      "step": 10490
    },
    {
      "epoch": 1.5631516234733391,
      "grad_norm": 12.125,
      "learning_rate": 3.907506702412869e-05,
      "loss": 1.1067,
      "num_input_tokens_seen": 6092280,
      "step": 10495
    },
    {
      "epoch": 1.5638963360142983,
      "grad_norm": 11.875,
      "learning_rate": 3.9093684837652665e-05,
      "loss": 0.7829,
      "num_input_tokens_seen": 6095064,
      "step": 10500
    },
    {
      "epoch": 1.5646410485552575,
      "grad_norm": 13.9375,
      "learning_rate": 3.911230265117665e-05,
      "loss": 0.9593,
      "num_input_tokens_seen": 6097880,
      "step": 10505
    },
    {
      "epoch": 1.5653857610962167,
      "grad_norm": 9.125,
      "learning_rate": 3.913092046470063e-05,
      "loss": 1.2893,
      "num_input_tokens_seen": 6100984,
      "step": 10510
    },
    {
      "epoch": 1.566130473637176,
      "grad_norm": 8.25,
      "learning_rate": 3.9149538278224606e-05,
      "loss": 1.2156,
      "num_input_tokens_seen": 6103928,
      "step": 10515
    },
    {
      "epoch": 1.5668751861781351,
      "grad_norm": 8.5625,
      "learning_rate": 3.9168156091748585e-05,
      "loss": 1.6827,
      "num_input_tokens_seen": 6107000,
      "step": 10520
    },
    {
      "epoch": 1.5676198987190944,
      "grad_norm": 12.1875,
      "learning_rate": 3.918677390527257e-05,
      "loss": 1.2967,
      "num_input_tokens_seen": 6110744,
      "step": 10525
    },
    {
      "epoch": 1.5683646112600536,
      "grad_norm": 7.125,
      "learning_rate": 3.920539171879655e-05,
      "loss": 1.2138,
      "num_input_tokens_seen": 6113848,
      "step": 10530
    },
    {
      "epoch": 1.5691093238010128,
      "grad_norm": 12.75,
      "learning_rate": 3.9224009532320526e-05,
      "loss": 0.8966,
      "num_input_tokens_seen": 6116856,
      "step": 10535
    },
    {
      "epoch": 1.569854036341972,
      "grad_norm": 11.75,
      "learning_rate": 3.9242627345844504e-05,
      "loss": 1.0661,
      "num_input_tokens_seen": 6119416,
      "step": 10540
    },
    {
      "epoch": 1.5705987488829312,
      "grad_norm": 6.03125,
      "learning_rate": 3.926124515936849e-05,
      "loss": 0.9842,
      "num_input_tokens_seen": 6122168,
      "step": 10545
    },
    {
      "epoch": 1.5713434614238904,
      "grad_norm": 9.8125,
      "learning_rate": 3.927986297289247e-05,
      "loss": 1.0913,
      "num_input_tokens_seen": 6125144,
      "step": 10550
    },
    {
      "epoch": 1.5720881739648496,
      "grad_norm": 8.625,
      "learning_rate": 3.9298480786416446e-05,
      "loss": 1.2373,
      "num_input_tokens_seen": 6128344,
      "step": 10555
    },
    {
      "epoch": 1.5728328865058088,
      "grad_norm": 14.0,
      "learning_rate": 3.9317098599940424e-05,
      "loss": 1.4256,
      "num_input_tokens_seen": 6131384,
      "step": 10560
    },
    {
      "epoch": 1.573577599046768,
      "grad_norm": 12.6875,
      "learning_rate": 3.933571641346441e-05,
      "loss": 0.9289,
      "num_input_tokens_seen": 6134648,
      "step": 10565
    },
    {
      "epoch": 1.5743223115877272,
      "grad_norm": 9.75,
      "learning_rate": 3.935433422698838e-05,
      "loss": 1.0721,
      "num_input_tokens_seen": 6137464,
      "step": 10570
    },
    {
      "epoch": 1.5750670241286864,
      "grad_norm": 8.5625,
      "learning_rate": 3.937295204051236e-05,
      "loss": 1.053,
      "num_input_tokens_seen": 6140408,
      "step": 10575
    },
    {
      "epoch": 1.5758117366696456,
      "grad_norm": 9.8125,
      "learning_rate": 3.9391569854036344e-05,
      "loss": 1.0896,
      "num_input_tokens_seen": 6143160,
      "step": 10580
    },
    {
      "epoch": 1.5765564492106048,
      "grad_norm": 21.25,
      "learning_rate": 3.941018766756032e-05,
      "loss": 1.3133,
      "num_input_tokens_seen": 6145944,
      "step": 10585
    },
    {
      "epoch": 1.577301161751564,
      "grad_norm": 8.5625,
      "learning_rate": 3.94288054810843e-05,
      "loss": 1.2048,
      "num_input_tokens_seen": 6148888,
      "step": 10590
    },
    {
      "epoch": 1.5780458742925232,
      "grad_norm": 16.0,
      "learning_rate": 3.9447423294608285e-05,
      "loss": 1.0416,
      "num_input_tokens_seen": 6151672,
      "step": 10595
    },
    {
      "epoch": 1.5787905868334824,
      "grad_norm": 17.625,
      "learning_rate": 3.946604110813226e-05,
      "loss": 1.527,
      "num_input_tokens_seen": 6154520,
      "step": 10600
    },
    {
      "epoch": 1.5795352993744416,
      "grad_norm": 11.875,
      "learning_rate": 3.948465892165624e-05,
      "loss": 1.1313,
      "num_input_tokens_seen": 6157400,
      "step": 10605
    },
    {
      "epoch": 1.5802800119154008,
      "grad_norm": 10.125,
      "learning_rate": 3.950327673518022e-05,
      "loss": 1.477,
      "num_input_tokens_seen": 6160120,
      "step": 10610
    },
    {
      "epoch": 1.58102472445636,
      "grad_norm": 7.53125,
      "learning_rate": 3.9521894548704205e-05,
      "loss": 0.9169,
      "num_input_tokens_seen": 6163032,
      "step": 10615
    },
    {
      "epoch": 1.5817694369973192,
      "grad_norm": 5.375,
      "learning_rate": 3.954051236222818e-05,
      "loss": 1.2325,
      "num_input_tokens_seen": 6165816,
      "step": 10620
    },
    {
      "epoch": 1.5825141495382782,
      "grad_norm": 7.6875,
      "learning_rate": 3.955913017575216e-05,
      "loss": 1.2522,
      "num_input_tokens_seen": 6168664,
      "step": 10625
    },
    {
      "epoch": 1.5832588620792374,
      "grad_norm": 9.75,
      "learning_rate": 3.957774798927614e-05,
      "loss": 1.4988,
      "num_input_tokens_seen": 6171704,
      "step": 10630
    },
    {
      "epoch": 1.5840035746201966,
      "grad_norm": 11.5,
      "learning_rate": 3.9596365802800125e-05,
      "loss": 1.0161,
      "num_input_tokens_seen": 6174392,
      "step": 10635
    },
    {
      "epoch": 1.5847482871611558,
      "grad_norm": 12.25,
      "learning_rate": 3.96149836163241e-05,
      "loss": 1.161,
      "num_input_tokens_seen": 6177176,
      "step": 10640
    },
    {
      "epoch": 1.585492999702115,
      "grad_norm": 8.0625,
      "learning_rate": 3.963360142984808e-05,
      "loss": 1.2137,
      "num_input_tokens_seen": 6180280,
      "step": 10645
    },
    {
      "epoch": 1.5862377122430742,
      "grad_norm": 15.875,
      "learning_rate": 3.9652219243372066e-05,
      "loss": 1.3467,
      "num_input_tokens_seen": 6183576,
      "step": 10650
    },
    {
      "epoch": 1.5869824247840334,
      "grad_norm": 15.5625,
      "learning_rate": 3.967083705689604e-05,
      "loss": 1.1352,
      "num_input_tokens_seen": 6186904,
      "step": 10655
    },
    {
      "epoch": 1.5877271373249926,
      "grad_norm": 14.5,
      "learning_rate": 3.9689454870420016e-05,
      "loss": 0.9371,
      "num_input_tokens_seen": 6189720,
      "step": 10660
    },
    {
      "epoch": 1.5884718498659516,
      "grad_norm": 6.59375,
      "learning_rate": 3.9708072683944e-05,
      "loss": 1.1314,
      "num_input_tokens_seen": 6192792,
      "step": 10665
    },
    {
      "epoch": 1.5892165624069108,
      "grad_norm": 18.75,
      "learning_rate": 3.972669049746798e-05,
      "loss": 1.3481,
      "num_input_tokens_seen": 6196184,
      "step": 10670
    },
    {
      "epoch": 1.58996127494787,
      "grad_norm": 19.375,
      "learning_rate": 3.974530831099196e-05,
      "loss": 1.3803,
      "num_input_tokens_seen": 6198968,
      "step": 10675
    },
    {
      "epoch": 1.5907059874888292,
      "grad_norm": 6.28125,
      "learning_rate": 3.9763926124515935e-05,
      "loss": 1.0772,
      "num_input_tokens_seen": 6202040,
      "step": 10680
    },
    {
      "epoch": 1.5914507000297884,
      "grad_norm": 8.75,
      "learning_rate": 3.978254393803992e-05,
      "loss": 1.2793,
      "num_input_tokens_seen": 6204952,
      "step": 10685
    },
    {
      "epoch": 1.5921954125707476,
      "grad_norm": 7.84375,
      "learning_rate": 3.98011617515639e-05,
      "loss": 0.7859,
      "num_input_tokens_seen": 6207800,
      "step": 10690
    },
    {
      "epoch": 1.5929401251117068,
      "grad_norm": 8.4375,
      "learning_rate": 3.981977956508788e-05,
      "loss": 1.2886,
      "num_input_tokens_seen": 6210424,
      "step": 10695
    },
    {
      "epoch": 1.593684837652666,
      "grad_norm": 18.25,
      "learning_rate": 3.9838397378611855e-05,
      "loss": 1.0969,
      "num_input_tokens_seen": 6213080,
      "step": 10700
    },
    {
      "epoch": 1.5944295501936252,
      "grad_norm": 7.65625,
      "learning_rate": 3.985701519213584e-05,
      "loss": 1.128,
      "num_input_tokens_seen": 6216088,
      "step": 10705
    },
    {
      "epoch": 1.5951742627345844,
      "grad_norm": 13.875,
      "learning_rate": 3.987563300565982e-05,
      "loss": 1.1481,
      "num_input_tokens_seen": 6219160,
      "step": 10710
    },
    {
      "epoch": 1.5959189752755436,
      "grad_norm": 5.34375,
      "learning_rate": 3.98942508191838e-05,
      "loss": 1.2593,
      "num_input_tokens_seen": 6222424,
      "step": 10715
    },
    {
      "epoch": 1.5966636878165028,
      "grad_norm": 22.375,
      "learning_rate": 3.991286863270778e-05,
      "loss": 1.239,
      "num_input_tokens_seen": 6225400,
      "step": 10720
    },
    {
      "epoch": 1.597408400357462,
      "grad_norm": 15.3125,
      "learning_rate": 3.993148644623176e-05,
      "loss": 1.2497,
      "num_input_tokens_seen": 6227768,
      "step": 10725
    },
    {
      "epoch": 1.5981531128984212,
      "grad_norm": 12.3125,
      "learning_rate": 3.995010425975574e-05,
      "loss": 0.991,
      "num_input_tokens_seen": 6231000,
      "step": 10730
    },
    {
      "epoch": 1.5988978254393804,
      "grad_norm": 7.75,
      "learning_rate": 3.9968722073279716e-05,
      "loss": 0.7573,
      "num_input_tokens_seen": 6233656,
      "step": 10735
    },
    {
      "epoch": 1.5996425379803396,
      "grad_norm": 7.9375,
      "learning_rate": 3.9987339886803695e-05,
      "loss": 1.1099,
      "num_input_tokens_seen": 6236344,
      "step": 10740
    },
    {
      "epoch": 1.6003872505212988,
      "grad_norm": 23.75,
      "learning_rate": 4.000595770032767e-05,
      "loss": 1.1329,
      "num_input_tokens_seen": 6239416,
      "step": 10745
    },
    {
      "epoch": 1.601131963062258,
      "grad_norm": 9.625,
      "learning_rate": 4.002457551385165e-05,
      "loss": 1.137,
      "num_input_tokens_seen": 6242520,
      "step": 10750
    },
    {
      "epoch": 1.6018766756032172,
      "grad_norm": 10.625,
      "learning_rate": 4.0043193327375636e-05,
      "loss": 0.983,
      "num_input_tokens_seen": 6245336,
      "step": 10755
    },
    {
      "epoch": 1.6026213881441764,
      "grad_norm": 9.6875,
      "learning_rate": 4.0061811140899614e-05,
      "loss": 1.195,
      "num_input_tokens_seen": 6247928,
      "step": 10760
    },
    {
      "epoch": 1.6033661006851356,
      "grad_norm": 16.875,
      "learning_rate": 4.008042895442359e-05,
      "loss": 1.4039,
      "num_input_tokens_seen": 6250648,
      "step": 10765
    },
    {
      "epoch": 1.6041108132260948,
      "grad_norm": 11.5625,
      "learning_rate": 4.009904676794757e-05,
      "loss": 1.1194,
      "num_input_tokens_seen": 6253368,
      "step": 10770
    },
    {
      "epoch": 1.604855525767054,
      "grad_norm": 12.25,
      "learning_rate": 4.0117664581471556e-05,
      "loss": 1.4175,
      "num_input_tokens_seen": 6256024,
      "step": 10775
    },
    {
      "epoch": 1.6056002383080132,
      "grad_norm": 17.75,
      "learning_rate": 4.0136282394995534e-05,
      "loss": 0.8883,
      "num_input_tokens_seen": 6258680,
      "step": 10780
    },
    {
      "epoch": 1.6063449508489724,
      "grad_norm": 5.40625,
      "learning_rate": 4.015490020851951e-05,
      "loss": 1.1326,
      "num_input_tokens_seen": 6261432,
      "step": 10785
    },
    {
      "epoch": 1.6070896633899316,
      "grad_norm": 11.3125,
      "learning_rate": 4.017351802204349e-05,
      "loss": 1.1742,
      "num_input_tokens_seen": 6264440,
      "step": 10790
    },
    {
      "epoch": 1.6078343759308906,
      "grad_norm": 5.5,
      "learning_rate": 4.0192135835567475e-05,
      "loss": 1.2166,
      "num_input_tokens_seen": 6267000,
      "step": 10795
    },
    {
      "epoch": 1.6085790884718498,
      "grad_norm": 5.59375,
      "learning_rate": 4.0210753649091454e-05,
      "loss": 1.1928,
      "num_input_tokens_seen": 6269432,
      "step": 10800
    },
    {
      "epoch": 1.609323801012809,
      "grad_norm": 7.40625,
      "learning_rate": 4.022937146261543e-05,
      "loss": 1.264,
      "num_input_tokens_seen": 6272376,
      "step": 10805
    },
    {
      "epoch": 1.6100685135537682,
      "grad_norm": 7.46875,
      "learning_rate": 4.024798927613942e-05,
      "loss": 1.5325,
      "num_input_tokens_seen": 6275480,
      "step": 10810
    },
    {
      "epoch": 1.6108132260947274,
      "grad_norm": 19.125,
      "learning_rate": 4.0266607089663395e-05,
      "loss": 1.2895,
      "num_input_tokens_seen": 6278392,
      "step": 10815
    },
    {
      "epoch": 1.6115579386356866,
      "grad_norm": 13.0625,
      "learning_rate": 4.028522490318737e-05,
      "loss": 1.1691,
      "num_input_tokens_seen": 6281400,
      "step": 10820
    },
    {
      "epoch": 1.6123026511766458,
      "grad_norm": 11.3125,
      "learning_rate": 4.030384271671135e-05,
      "loss": 1.1254,
      "num_input_tokens_seen": 6284248,
      "step": 10825
    },
    {
      "epoch": 1.613047363717605,
      "grad_norm": 12.125,
      "learning_rate": 4.032246053023533e-05,
      "loss": 1.2583,
      "num_input_tokens_seen": 6287288,
      "step": 10830
    },
    {
      "epoch": 1.6137920762585642,
      "grad_norm": 15.1875,
      "learning_rate": 4.034107834375931e-05,
      "loss": 1.0179,
      "num_input_tokens_seen": 6289912,
      "step": 10835
    },
    {
      "epoch": 1.6145367887995232,
      "grad_norm": 23.625,
      "learning_rate": 4.0359696157283286e-05,
      "loss": 1.3324,
      "num_input_tokens_seen": 6292728,
      "step": 10840
    },
    {
      "epoch": 1.6152815013404824,
      "grad_norm": 19.875,
      "learning_rate": 4.037831397080727e-05,
      "loss": 1.1049,
      "num_input_tokens_seen": 6295768,
      "step": 10845
    },
    {
      "epoch": 1.6160262138814416,
      "grad_norm": 8.625,
      "learning_rate": 4.039693178433125e-05,
      "loss": 1.1734,
      "num_input_tokens_seen": 6298520,
      "step": 10850
    },
    {
      "epoch": 1.6167709264224008,
      "grad_norm": 5.28125,
      "learning_rate": 4.041554959785523e-05,
      "loss": 1.1776,
      "num_input_tokens_seen": 6301496,
      "step": 10855
    },
    {
      "epoch": 1.61751563896336,
      "grad_norm": 15.9375,
      "learning_rate": 4.0434167411379206e-05,
      "loss": 0.9146,
      "num_input_tokens_seen": 6304248,
      "step": 10860
    },
    {
      "epoch": 1.6182603515043192,
      "grad_norm": 6.6875,
      "learning_rate": 4.045278522490319e-05,
      "loss": 1.421,
      "num_input_tokens_seen": 6307000,
      "step": 10865
    },
    {
      "epoch": 1.6190050640452784,
      "grad_norm": 14.8125,
      "learning_rate": 4.047140303842717e-05,
      "loss": 1.3346,
      "num_input_tokens_seen": 6309912,
      "step": 10870
    },
    {
      "epoch": 1.6197497765862376,
      "grad_norm": 13.4375,
      "learning_rate": 4.049002085195115e-05,
      "loss": 0.8255,
      "num_input_tokens_seen": 6312632,
      "step": 10875
    },
    {
      "epoch": 1.6204944891271968,
      "grad_norm": 15.8125,
      "learning_rate": 4.050863866547513e-05,
      "loss": 1.4723,
      "num_input_tokens_seen": 6315352,
      "step": 10880
    },
    {
      "epoch": 1.621239201668156,
      "grad_norm": 15.0,
      "learning_rate": 4.052725647899911e-05,
      "loss": 0.9064,
      "num_input_tokens_seen": 6317944,
      "step": 10885
    },
    {
      "epoch": 1.6219839142091153,
      "grad_norm": 8.125,
      "learning_rate": 4.054587429252309e-05,
      "loss": 1.0473,
      "num_input_tokens_seen": 6321016,
      "step": 10890
    },
    {
      "epoch": 1.6227286267500745,
      "grad_norm": 17.125,
      "learning_rate": 4.056449210604707e-05,
      "loss": 1.467,
      "num_input_tokens_seen": 6323768,
      "step": 10895
    },
    {
      "epoch": 1.6234733392910337,
      "grad_norm": 12.5625,
      "learning_rate": 4.058310991957105e-05,
      "loss": 1.1709,
      "num_input_tokens_seen": 6326584,
      "step": 10900
    },
    {
      "epoch": 1.6242180518319929,
      "grad_norm": 5.90625,
      "learning_rate": 4.060172773309503e-05,
      "loss": 1.0295,
      "num_input_tokens_seen": 6329496,
      "step": 10905
    },
    {
      "epoch": 1.624962764372952,
      "grad_norm": 8.4375,
      "learning_rate": 4.062034554661901e-05,
      "loss": 1.1682,
      "num_input_tokens_seen": 6332248,
      "step": 10910
    },
    {
      "epoch": 1.6257074769139113,
      "grad_norm": 21.75,
      "learning_rate": 4.063896336014299e-05,
      "loss": 1.1688,
      "num_input_tokens_seen": 6334904,
      "step": 10915
    },
    {
      "epoch": 1.6264521894548705,
      "grad_norm": 15.3125,
      "learning_rate": 4.0657581173666965e-05,
      "loss": 0.9988,
      "num_input_tokens_seen": 6337592,
      "step": 10920
    },
    {
      "epoch": 1.6271969019958297,
      "grad_norm": 9.875,
      "learning_rate": 4.067619898719094e-05,
      "loss": 1.1033,
      "num_input_tokens_seen": 6340504,
      "step": 10925
    },
    {
      "epoch": 1.6279416145367889,
      "grad_norm": 15.3125,
      "learning_rate": 4.069481680071492e-05,
      "loss": 1.2337,
      "num_input_tokens_seen": 6343352,
      "step": 10930
    },
    {
      "epoch": 1.628686327077748,
      "grad_norm": 7.25,
      "learning_rate": 4.0713434614238907e-05,
      "loss": 0.9678,
      "num_input_tokens_seen": 6346840,
      "step": 10935
    },
    {
      "epoch": 1.6294310396187073,
      "grad_norm": 12.1875,
      "learning_rate": 4.0732052427762885e-05,
      "loss": 1.444,
      "num_input_tokens_seen": 6349752,
      "step": 10940
    },
    {
      "epoch": 1.6301757521596665,
      "grad_norm": 17.375,
      "learning_rate": 4.075067024128686e-05,
      "loss": 1.0466,
      "num_input_tokens_seen": 6352728,
      "step": 10945
    },
    {
      "epoch": 1.6309204647006257,
      "grad_norm": 16.25,
      "learning_rate": 4.076928805481084e-05,
      "loss": 1.0032,
      "num_input_tokens_seen": 6355896,
      "step": 10950
    },
    {
      "epoch": 1.6316651772415849,
      "grad_norm": 16.125,
      "learning_rate": 4.0787905868334826e-05,
      "loss": 1.1108,
      "num_input_tokens_seen": 6358552,
      "step": 10955
    },
    {
      "epoch": 1.632409889782544,
      "grad_norm": 8.625,
      "learning_rate": 4.0806523681858805e-05,
      "loss": 1.1452,
      "num_input_tokens_seen": 6361560,
      "step": 10960
    },
    {
      "epoch": 1.6331546023235033,
      "grad_norm": 17.0,
      "learning_rate": 4.082514149538278e-05,
      "loss": 1.0619,
      "num_input_tokens_seen": 6364696,
      "step": 10965
    },
    {
      "epoch": 1.6338993148644623,
      "grad_norm": 12.625,
      "learning_rate": 4.084375930890677e-05,
      "loss": 1.1099,
      "num_input_tokens_seen": 6367736,
      "step": 10970
    },
    {
      "epoch": 1.6346440274054215,
      "grad_norm": 11.3125,
      "learning_rate": 4.0862377122430746e-05,
      "loss": 1.2975,
      "num_input_tokens_seen": 6370648,
      "step": 10975
    },
    {
      "epoch": 1.6353887399463807,
      "grad_norm": 9.1875,
      "learning_rate": 4.0880994935954724e-05,
      "loss": 1.4353,
      "num_input_tokens_seen": 6373592,
      "step": 10980
    },
    {
      "epoch": 1.6361334524873399,
      "grad_norm": 10.4375,
      "learning_rate": 4.08996127494787e-05,
      "loss": 1.0819,
      "num_input_tokens_seen": 6376408,
      "step": 10985
    },
    {
      "epoch": 1.636878165028299,
      "grad_norm": 12.875,
      "learning_rate": 4.091823056300269e-05,
      "loss": 1.1061,
      "num_input_tokens_seen": 6379384,
      "step": 10990
    },
    {
      "epoch": 1.6376228775692583,
      "grad_norm": 5.34375,
      "learning_rate": 4.0936848376526666e-05,
      "loss": 1.3262,
      "num_input_tokens_seen": 6382392,
      "step": 10995
    },
    {
      "epoch": 1.6383675901102175,
      "grad_norm": 10.5,
      "learning_rate": 4.095546619005064e-05,
      "loss": 1.2297,
      "num_input_tokens_seen": 6385272,
      "step": 11000
    },
    {
      "epoch": 1.6391123026511767,
      "grad_norm": 13.0,
      "learning_rate": 4.097408400357462e-05,
      "loss": 1.3438,
      "num_input_tokens_seen": 6388184,
      "step": 11005
    },
    {
      "epoch": 1.6398570151921357,
      "grad_norm": 10.9375,
      "learning_rate": 4.09927018170986e-05,
      "loss": 0.9925,
      "num_input_tokens_seen": 6391256,
      "step": 11010
    },
    {
      "epoch": 1.6406017277330949,
      "grad_norm": 9.1875,
      "learning_rate": 4.101131963062258e-05,
      "loss": 1.0097,
      "num_input_tokens_seen": 6393688,
      "step": 11015
    },
    {
      "epoch": 1.641346440274054,
      "grad_norm": 18.125,
      "learning_rate": 4.102993744414656e-05,
      "loss": 1.3472,
      "num_input_tokens_seen": 6396728,
      "step": 11020
    },
    {
      "epoch": 1.6420911528150133,
      "grad_norm": 7.6875,
      "learning_rate": 4.104855525767054e-05,
      "loss": 0.9877,
      "num_input_tokens_seen": 6399576,
      "step": 11025
    },
    {
      "epoch": 1.6428358653559725,
      "grad_norm": 10.5,
      "learning_rate": 4.106717307119452e-05,
      "loss": 1.2738,
      "num_input_tokens_seen": 6402680,
      "step": 11030
    },
    {
      "epoch": 1.6435805778969317,
      "grad_norm": 10.625,
      "learning_rate": 4.10857908847185e-05,
      "loss": 1.6186,
      "num_input_tokens_seen": 6405688,
      "step": 11035
    },
    {
      "epoch": 1.6443252904378909,
      "grad_norm": 10.5625,
      "learning_rate": 4.110440869824248e-05,
      "loss": 0.8549,
      "num_input_tokens_seen": 6408408,
      "step": 11040
    },
    {
      "epoch": 1.64507000297885,
      "grad_norm": 10.0625,
      "learning_rate": 4.112302651176646e-05,
      "loss": 1.4855,
      "num_input_tokens_seen": 6411032,
      "step": 11045
    },
    {
      "epoch": 1.6458147155198093,
      "grad_norm": 16.125,
      "learning_rate": 4.114164432529044e-05,
      "loss": 1.2691,
      "num_input_tokens_seen": 6413944,
      "step": 11050
    },
    {
      "epoch": 1.6465594280607685,
      "grad_norm": 8.25,
      "learning_rate": 4.116026213881442e-05,
      "loss": 0.8082,
      "num_input_tokens_seen": 6417048,
      "step": 11055
    },
    {
      "epoch": 1.6473041406017277,
      "grad_norm": 10.75,
      "learning_rate": 4.11788799523384e-05,
      "loss": 1.0426,
      "num_input_tokens_seen": 6419960,
      "step": 11060
    },
    {
      "epoch": 1.648048853142687,
      "grad_norm": 10.9375,
      "learning_rate": 4.119749776586238e-05,
      "loss": 1.1659,
      "num_input_tokens_seen": 6422584,
      "step": 11065
    },
    {
      "epoch": 1.648793565683646,
      "grad_norm": 8.8125,
      "learning_rate": 4.121611557938636e-05,
      "loss": 1.1082,
      "num_input_tokens_seen": 6425624,
      "step": 11070
    },
    {
      "epoch": 1.6495382782246053,
      "grad_norm": 2.96875,
      "learning_rate": 4.123473339291034e-05,
      "loss": 1.1282,
      "num_input_tokens_seen": 6428312,
      "step": 11075
    },
    {
      "epoch": 1.6502829907655645,
      "grad_norm": 10.6875,
      "learning_rate": 4.125335120643432e-05,
      "loss": 0.6746,
      "num_input_tokens_seen": 6431064,
      "step": 11080
    },
    {
      "epoch": 1.6510277033065237,
      "grad_norm": 9.75,
      "learning_rate": 4.1271969019958294e-05,
      "loss": 1.1649,
      "num_input_tokens_seen": 6433976,
      "step": 11085
    },
    {
      "epoch": 1.651772415847483,
      "grad_norm": 9.4375,
      "learning_rate": 4.129058683348227e-05,
      "loss": 1.4467,
      "num_input_tokens_seen": 6437112,
      "step": 11090
    },
    {
      "epoch": 1.6525171283884421,
      "grad_norm": 6.59375,
      "learning_rate": 4.130920464700626e-05,
      "loss": 0.9861,
      "num_input_tokens_seen": 6439928,
      "step": 11095
    },
    {
      "epoch": 1.6532618409294013,
      "grad_norm": 10.75,
      "learning_rate": 4.1327822460530236e-05,
      "loss": 0.9222,
      "num_input_tokens_seen": 6442872,
      "step": 11100
    },
    {
      "epoch": 1.6540065534703605,
      "grad_norm": 5.25,
      "learning_rate": 4.1346440274054214e-05,
      "loss": 1.1081,
      "num_input_tokens_seen": 6445464,
      "step": 11105
    },
    {
      "epoch": 1.6547512660113197,
      "grad_norm": 12.0625,
      "learning_rate": 4.13650580875782e-05,
      "loss": 0.8998,
      "num_input_tokens_seen": 6448312,
      "step": 11110
    },
    {
      "epoch": 1.655495978552279,
      "grad_norm": 12.0625,
      "learning_rate": 4.138367590110218e-05,
      "loss": 1.1367,
      "num_input_tokens_seen": 6450936,
      "step": 11115
    },
    {
      "epoch": 1.6562406910932381,
      "grad_norm": 10.6875,
      "learning_rate": 4.1402293714626155e-05,
      "loss": 1.0773,
      "num_input_tokens_seen": 6453848,
      "step": 11120
    },
    {
      "epoch": 1.6569854036341973,
      "grad_norm": 14.0625,
      "learning_rate": 4.1420911528150134e-05,
      "loss": 1.0954,
      "num_input_tokens_seen": 6456920,
      "step": 11125
    },
    {
      "epoch": 1.6577301161751565,
      "grad_norm": 11.9375,
      "learning_rate": 4.143952934167412e-05,
      "loss": 1.0372,
      "num_input_tokens_seen": 6459768,
      "step": 11130
    },
    {
      "epoch": 1.6584748287161157,
      "grad_norm": 6.84375,
      "learning_rate": 4.14581471551981e-05,
      "loss": 1.0226,
      "num_input_tokens_seen": 6462616,
      "step": 11135
    },
    {
      "epoch": 1.6592195412570747,
      "grad_norm": 9.6875,
      "learning_rate": 4.1476764968722075e-05,
      "loss": 1.089,
      "num_input_tokens_seen": 6465592,
      "step": 11140
    },
    {
      "epoch": 1.659964253798034,
      "grad_norm": 11.0,
      "learning_rate": 4.149538278224605e-05,
      "loss": 1.0925,
      "num_input_tokens_seen": 6468504,
      "step": 11145
    },
    {
      "epoch": 1.6607089663389931,
      "grad_norm": 16.75,
      "learning_rate": 4.151400059577004e-05,
      "loss": 1.1992,
      "num_input_tokens_seen": 6471320,
      "step": 11150
    },
    {
      "epoch": 1.6614536788799523,
      "grad_norm": 20.5,
      "learning_rate": 4.1532618409294017e-05,
      "loss": 1.097,
      "num_input_tokens_seen": 6474136,
      "step": 11155
    },
    {
      "epoch": 1.6621983914209115,
      "grad_norm": 17.875,
      "learning_rate": 4.1551236222817995e-05,
      "loss": 1.2257,
      "num_input_tokens_seen": 6476984,
      "step": 11160
    },
    {
      "epoch": 1.6629431039618707,
      "grad_norm": 11.5625,
      "learning_rate": 4.156985403634197e-05,
      "loss": 1.1597,
      "num_input_tokens_seen": 6479864,
      "step": 11165
    },
    {
      "epoch": 1.66368781650283,
      "grad_norm": 11.0625,
      "learning_rate": 4.158847184986596e-05,
      "loss": 1.1423,
      "num_input_tokens_seen": 6482680,
      "step": 11170
    },
    {
      "epoch": 1.6644325290437891,
      "grad_norm": 16.375,
      "learning_rate": 4.160708966338993e-05,
      "loss": 1.1143,
      "num_input_tokens_seen": 6485336,
      "step": 11175
    },
    {
      "epoch": 1.6651772415847483,
      "grad_norm": 15.4375,
      "learning_rate": 4.162570747691391e-05,
      "loss": 1.0892,
      "num_input_tokens_seen": 6488280,
      "step": 11180
    },
    {
      "epoch": 1.6659219541257073,
      "grad_norm": 9.0,
      "learning_rate": 4.164432529043789e-05,
      "loss": 0.9883,
      "num_input_tokens_seen": 6491128,
      "step": 11185
    },
    {
      "epoch": 1.6666666666666665,
      "grad_norm": 12.6875,
      "learning_rate": 4.166294310396187e-05,
      "loss": 1.0608,
      "num_input_tokens_seen": 6494136,
      "step": 11190
    },
    {
      "epoch": 1.6674113792076257,
      "grad_norm": 16.5,
      "learning_rate": 4.168156091748585e-05,
      "loss": 1.1509,
      "num_input_tokens_seen": 6496824,
      "step": 11195
    },
    {
      "epoch": 1.668156091748585,
      "grad_norm": 5.78125,
      "learning_rate": 4.1700178731009834e-05,
      "loss": 1.079,
      "num_input_tokens_seen": 6499768,
      "step": 11200
    },
    {
      "epoch": 1.6689008042895441,
      "grad_norm": 10.5625,
      "learning_rate": 4.171879654453381e-05,
      "loss": 1.5038,
      "num_input_tokens_seen": 6502712,
      "step": 11205
    },
    {
      "epoch": 1.6696455168305033,
      "grad_norm": 17.5,
      "learning_rate": 4.173741435805779e-05,
      "loss": 0.9323,
      "num_input_tokens_seen": 6505464,
      "step": 11210
    },
    {
      "epoch": 1.6703902293714625,
      "grad_norm": 13.125,
      "learning_rate": 4.175603217158177e-05,
      "loss": 1.1703,
      "num_input_tokens_seen": 6508504,
      "step": 11215
    },
    {
      "epoch": 1.6711349419124217,
      "grad_norm": 17.5,
      "learning_rate": 4.1774649985105754e-05,
      "loss": 1.2228,
      "num_input_tokens_seen": 6511320,
      "step": 11220
    },
    {
      "epoch": 1.671879654453381,
      "grad_norm": 17.0,
      "learning_rate": 4.179326779862973e-05,
      "loss": 1.4853,
      "num_input_tokens_seen": 6514392,
      "step": 11225
    },
    {
      "epoch": 1.6726243669943401,
      "grad_norm": 8.8125,
      "learning_rate": 4.181188561215371e-05,
      "loss": 1.0643,
      "num_input_tokens_seen": 6517752,
      "step": 11230
    },
    {
      "epoch": 1.6733690795352993,
      "grad_norm": 8.1875,
      "learning_rate": 4.183050342567769e-05,
      "loss": 0.9018,
      "num_input_tokens_seen": 6520568,
      "step": 11235
    },
    {
      "epoch": 1.6741137920762585,
      "grad_norm": 10.9375,
      "learning_rate": 4.1849121239201674e-05,
      "loss": 1.0642,
      "num_input_tokens_seen": 6523128,
      "step": 11240
    },
    {
      "epoch": 1.6748585046172177,
      "grad_norm": 8.3125,
      "learning_rate": 4.186773905272565e-05,
      "loss": 1.1197,
      "num_input_tokens_seen": 6526232,
      "step": 11245
    },
    {
      "epoch": 1.675603217158177,
      "grad_norm": 9.375,
      "learning_rate": 4.188635686624963e-05,
      "loss": 1.0953,
      "num_input_tokens_seen": 6528984,
      "step": 11250
    },
    {
      "epoch": 1.6763479296991362,
      "grad_norm": 9.5,
      "learning_rate": 4.1904974679773615e-05,
      "loss": 1.1766,
      "num_input_tokens_seen": 6531896,
      "step": 11255
    },
    {
      "epoch": 1.6770926422400954,
      "grad_norm": 13.4375,
      "learning_rate": 4.1923592493297587e-05,
      "loss": 1.2485,
      "num_input_tokens_seen": 6534552,
      "step": 11260
    },
    {
      "epoch": 1.6778373547810546,
      "grad_norm": 10.5,
      "learning_rate": 4.1942210306821565e-05,
      "loss": 1.053,
      "num_input_tokens_seen": 6537400,
      "step": 11265
    },
    {
      "epoch": 1.6785820673220138,
      "grad_norm": 9.5,
      "learning_rate": 4.196082812034555e-05,
      "loss": 1.218,
      "num_input_tokens_seen": 6540216,
      "step": 11270
    },
    {
      "epoch": 1.679326779862973,
      "grad_norm": 8.625,
      "learning_rate": 4.197944593386953e-05,
      "loss": 1.0275,
      "num_input_tokens_seen": 6542968,
      "step": 11275
    },
    {
      "epoch": 1.6800714924039322,
      "grad_norm": 10.9375,
      "learning_rate": 4.1998063747393506e-05,
      "loss": 1.1394,
      "num_input_tokens_seen": 6545848,
      "step": 11280
    },
    {
      "epoch": 1.6808162049448914,
      "grad_norm": 15.0625,
      "learning_rate": 4.2016681560917485e-05,
      "loss": 1.5218,
      "num_input_tokens_seen": 6548664,
      "step": 11285
    },
    {
      "epoch": 1.6815609174858506,
      "grad_norm": 8.3125,
      "learning_rate": 4.203529937444147e-05,
      "loss": 0.8877,
      "num_input_tokens_seen": 6551416,
      "step": 11290
    },
    {
      "epoch": 1.6823056300268098,
      "grad_norm": 10.5625,
      "learning_rate": 4.205391718796545e-05,
      "loss": 1.396,
      "num_input_tokens_seen": 6554040,
      "step": 11295
    },
    {
      "epoch": 1.683050342567769,
      "grad_norm": 16.125,
      "learning_rate": 4.2072535001489426e-05,
      "loss": 0.9315,
      "num_input_tokens_seen": 6557048,
      "step": 11300
    },
    {
      "epoch": 1.6837950551087282,
      "grad_norm": 10.4375,
      "learning_rate": 4.2091152815013404e-05,
      "loss": 1.2923,
      "num_input_tokens_seen": 6559992,
      "step": 11305
    },
    {
      "epoch": 1.6845397676496874,
      "grad_norm": 3.5625,
      "learning_rate": 4.210977062853739e-05,
      "loss": 0.9469,
      "num_input_tokens_seen": 6562712,
      "step": 11310
    },
    {
      "epoch": 1.6852844801906464,
      "grad_norm": 6.3125,
      "learning_rate": 4.212838844206137e-05,
      "loss": 0.7028,
      "num_input_tokens_seen": 6565688,
      "step": 11315
    },
    {
      "epoch": 1.6860291927316056,
      "grad_norm": 15.375,
      "learning_rate": 4.2147006255585346e-05,
      "loss": 1.2372,
      "num_input_tokens_seen": 6568536,
      "step": 11320
    },
    {
      "epoch": 1.6867739052725648,
      "grad_norm": 22.875,
      "learning_rate": 4.216562406910933e-05,
      "loss": 1.0361,
      "num_input_tokens_seen": 6571448,
      "step": 11325
    },
    {
      "epoch": 1.687518617813524,
      "grad_norm": 11.125,
      "learning_rate": 4.218424188263331e-05,
      "loss": 1.6182,
      "num_input_tokens_seen": 6574616,
      "step": 11330
    },
    {
      "epoch": 1.6882633303544832,
      "grad_norm": 7.84375,
      "learning_rate": 4.220285969615729e-05,
      "loss": 1.2141,
      "num_input_tokens_seen": 6577432,
      "step": 11335
    },
    {
      "epoch": 1.6890080428954424,
      "grad_norm": 26.0,
      "learning_rate": 4.2221477509681265e-05,
      "loss": 0.9032,
      "num_input_tokens_seen": 6580088,
      "step": 11340
    },
    {
      "epoch": 1.6897527554364016,
      "grad_norm": 9.0625,
      "learning_rate": 4.2240095323205244e-05,
      "loss": 1.0824,
      "num_input_tokens_seen": 6582648,
      "step": 11345
    },
    {
      "epoch": 1.6904974679773608,
      "grad_norm": 10.5,
      "learning_rate": 4.225871313672922e-05,
      "loss": 1.1138,
      "num_input_tokens_seen": 6585400,
      "step": 11350
    },
    {
      "epoch": 1.69124218051832,
      "grad_norm": 12.0625,
      "learning_rate": 4.22773309502532e-05,
      "loss": 1.0345,
      "num_input_tokens_seen": 6588472,
      "step": 11355
    },
    {
      "epoch": 1.691986893059279,
      "grad_norm": 10.8125,
      "learning_rate": 4.2295948763777185e-05,
      "loss": 0.9478,
      "num_input_tokens_seen": 6591192,
      "step": 11360
    },
    {
      "epoch": 1.6927316056002382,
      "grad_norm": 11.5625,
      "learning_rate": 4.231456657730116e-05,
      "loss": 0.9276,
      "num_input_tokens_seen": 6593880,
      "step": 11365
    },
    {
      "epoch": 1.6934763181411974,
      "grad_norm": 10.375,
      "learning_rate": 4.233318439082514e-05,
      "loss": 1.1686,
      "num_input_tokens_seen": 6597016,
      "step": 11370
    },
    {
      "epoch": 1.6942210306821566,
      "grad_norm": 12.6875,
      "learning_rate": 4.235180220434912e-05,
      "loss": 1.0729,
      "num_input_tokens_seen": 6599896,
      "step": 11375
    },
    {
      "epoch": 1.6949657432231158,
      "grad_norm": 11.875,
      "learning_rate": 4.2370420017873105e-05,
      "loss": 1.3291,
      "num_input_tokens_seen": 6602808,
      "step": 11380
    },
    {
      "epoch": 1.695710455764075,
      "grad_norm": 6.53125,
      "learning_rate": 4.238903783139708e-05,
      "loss": 0.9051,
      "num_input_tokens_seen": 6605624,
      "step": 11385
    },
    {
      "epoch": 1.6964551683050342,
      "grad_norm": 11.4375,
      "learning_rate": 4.240765564492106e-05,
      "loss": 1.1577,
      "num_input_tokens_seen": 6608504,
      "step": 11390
    },
    {
      "epoch": 1.6971998808459934,
      "grad_norm": 6.625,
      "learning_rate": 4.242627345844504e-05,
      "loss": 0.8608,
      "num_input_tokens_seen": 6611128,
      "step": 11395
    },
    {
      "epoch": 1.6979445933869526,
      "grad_norm": 16.25,
      "learning_rate": 4.2444891271969025e-05,
      "loss": 1.1415,
      "num_input_tokens_seen": 6614168,
      "step": 11400
    },
    {
      "epoch": 1.6986893059279118,
      "grad_norm": 17.625,
      "learning_rate": 4.2463509085493e-05,
      "loss": 1.1771,
      "num_input_tokens_seen": 6617304,
      "step": 11405
    },
    {
      "epoch": 1.699434018468871,
      "grad_norm": 10.875,
      "learning_rate": 4.248212689901698e-05,
      "loss": 0.9856,
      "num_input_tokens_seen": 6620088,
      "step": 11410
    },
    {
      "epoch": 1.7001787310098302,
      "grad_norm": 10.4375,
      "learning_rate": 4.2500744712540966e-05,
      "loss": 1.1803,
      "num_input_tokens_seen": 6622872,
      "step": 11415
    },
    {
      "epoch": 1.7009234435507894,
      "grad_norm": 10.875,
      "learning_rate": 4.2519362526064944e-05,
      "loss": 1.0417,
      "num_input_tokens_seen": 6625560,
      "step": 11420
    },
    {
      "epoch": 1.7016681560917486,
      "grad_norm": 13.8125,
      "learning_rate": 4.253798033958892e-05,
      "loss": 0.9912,
      "num_input_tokens_seen": 6628376,
      "step": 11425
    },
    {
      "epoch": 1.7024128686327078,
      "grad_norm": 12.125,
      "learning_rate": 4.25565981531129e-05,
      "loss": 0.9775,
      "num_input_tokens_seen": 6631096,
      "step": 11430
    },
    {
      "epoch": 1.703157581173667,
      "grad_norm": 12.0625,
      "learning_rate": 4.257521596663688e-05,
      "loss": 1.3462,
      "num_input_tokens_seen": 6634040,
      "step": 11435
    },
    {
      "epoch": 1.7039022937146262,
      "grad_norm": 8.5,
      "learning_rate": 4.259383378016086e-05,
      "loss": 0.8605,
      "num_input_tokens_seen": 6636920,
      "step": 11440
    },
    {
      "epoch": 1.7046470062555854,
      "grad_norm": 7.03125,
      "learning_rate": 4.2612451593684835e-05,
      "loss": 1.4365,
      "num_input_tokens_seen": 6639800,
      "step": 11445
    },
    {
      "epoch": 1.7053917187965446,
      "grad_norm": 11.1875,
      "learning_rate": 4.263106940720882e-05,
      "loss": 1.0331,
      "num_input_tokens_seen": 6642680,
      "step": 11450
    },
    {
      "epoch": 1.7061364313375038,
      "grad_norm": 7.875,
      "learning_rate": 4.26496872207328e-05,
      "loss": 1.0894,
      "num_input_tokens_seen": 6645784,
      "step": 11455
    },
    {
      "epoch": 1.706881143878463,
      "grad_norm": 9.75,
      "learning_rate": 4.266830503425678e-05,
      "loss": 1.1736,
      "num_input_tokens_seen": 6649048,
      "step": 11460
    },
    {
      "epoch": 1.7076258564194222,
      "grad_norm": 10.375,
      "learning_rate": 4.2686922847780755e-05,
      "loss": 0.8163,
      "num_input_tokens_seen": 6651960,
      "step": 11465
    },
    {
      "epoch": 1.7083705689603814,
      "grad_norm": 11.9375,
      "learning_rate": 4.270554066130474e-05,
      "loss": 1.0543,
      "num_input_tokens_seen": 6654968,
      "step": 11470
    },
    {
      "epoch": 1.7091152815013406,
      "grad_norm": 14.4375,
      "learning_rate": 4.272415847482872e-05,
      "loss": 1.4073,
      "num_input_tokens_seen": 6657944,
      "step": 11475
    },
    {
      "epoch": 1.7098599940422998,
      "grad_norm": 7.65625,
      "learning_rate": 4.2742776288352697e-05,
      "loss": 1.4262,
      "num_input_tokens_seen": 6660824,
      "step": 11480
    },
    {
      "epoch": 1.710604706583259,
      "grad_norm": 9.25,
      "learning_rate": 4.276139410187668e-05,
      "loss": 1.1404,
      "num_input_tokens_seen": 6663800,
      "step": 11485
    },
    {
      "epoch": 1.711349419124218,
      "grad_norm": 9.375,
      "learning_rate": 4.278001191540066e-05,
      "loss": 0.9624,
      "num_input_tokens_seen": 6666872,
      "step": 11490
    },
    {
      "epoch": 1.7120941316651772,
      "grad_norm": 7.625,
      "learning_rate": 4.279862972892464e-05,
      "loss": 1.0421,
      "num_input_tokens_seen": 6669624,
      "step": 11495
    },
    {
      "epoch": 1.7128388442061364,
      "grad_norm": 15.0,
      "learning_rate": 4.2817247542448616e-05,
      "loss": 0.9614,
      "num_input_tokens_seen": 6672248,
      "step": 11500
    },
    {
      "epoch": 1.7135835567470956,
      "grad_norm": 7.15625,
      "learning_rate": 4.28358653559726e-05,
      "loss": 1.0722,
      "num_input_tokens_seen": 6675288,
      "step": 11505
    },
    {
      "epoch": 1.7143282692880548,
      "grad_norm": 14.75,
      "learning_rate": 4.285448316949658e-05,
      "loss": 1.1822,
      "num_input_tokens_seen": 6678168,
      "step": 11510
    },
    {
      "epoch": 1.715072981829014,
      "grad_norm": 15.9375,
      "learning_rate": 4.287310098302056e-05,
      "loss": 1.0395,
      "num_input_tokens_seen": 6680792,
      "step": 11515
    },
    {
      "epoch": 1.7158176943699732,
      "grad_norm": 8.9375,
      "learning_rate": 4.2891718796544536e-05,
      "loss": 0.7989,
      "num_input_tokens_seen": 6683672,
      "step": 11520
    },
    {
      "epoch": 1.7165624069109324,
      "grad_norm": 10.3125,
      "learning_rate": 4.2910336610068514e-05,
      "loss": 1.0374,
      "num_input_tokens_seen": 6686584,
      "step": 11525
    },
    {
      "epoch": 1.7173071194518914,
      "grad_norm": 7.59375,
      "learning_rate": 4.292895442359249e-05,
      "loss": 1.3342,
      "num_input_tokens_seen": 6689752,
      "step": 11530
    },
    {
      "epoch": 1.7180518319928506,
      "grad_norm": 8.3125,
      "learning_rate": 4.294757223711647e-05,
      "loss": 1.062,
      "num_input_tokens_seen": 6692568,
      "step": 11535
    },
    {
      "epoch": 1.7187965445338098,
      "grad_norm": 6.15625,
      "learning_rate": 4.2966190050640456e-05,
      "loss": 1.1567,
      "num_input_tokens_seen": 6695544,
      "step": 11540
    },
    {
      "epoch": 1.719541257074769,
      "grad_norm": 11.375,
      "learning_rate": 4.2984807864164434e-05,
      "loss": 0.9498,
      "num_input_tokens_seen": 6698424,
      "step": 11545
    },
    {
      "epoch": 1.7202859696157282,
      "grad_norm": 11.6875,
      "learning_rate": 4.300342567768841e-05,
      "loss": 1.3496,
      "num_input_tokens_seen": 6701272,
      "step": 11550
    },
    {
      "epoch": 1.7210306821566874,
      "grad_norm": 13.3125,
      "learning_rate": 4.302204349121239e-05,
      "loss": 0.8805,
      "num_input_tokens_seen": 6704088,
      "step": 11555
    },
    {
      "epoch": 1.7217753946976466,
      "grad_norm": 9.5625,
      "learning_rate": 4.3040661304736375e-05,
      "loss": 1.1368,
      "num_input_tokens_seen": 6706744,
      "step": 11560
    },
    {
      "epoch": 1.7225201072386058,
      "grad_norm": 8.4375,
      "learning_rate": 4.3059279118260354e-05,
      "loss": 1.1116,
      "num_input_tokens_seen": 6709496,
      "step": 11565
    },
    {
      "epoch": 1.723264819779565,
      "grad_norm": 14.25,
      "learning_rate": 4.307789693178433e-05,
      "loss": 0.9843,
      "num_input_tokens_seen": 6712504,
      "step": 11570
    },
    {
      "epoch": 1.7240095323205242,
      "grad_norm": 11.8125,
      "learning_rate": 4.309651474530832e-05,
      "loss": 1.5991,
      "num_input_tokens_seen": 6715768,
      "step": 11575
    },
    {
      "epoch": 1.7247542448614834,
      "grad_norm": 10.0625,
      "learning_rate": 4.3115132558832295e-05,
      "loss": 0.9318,
      "num_input_tokens_seen": 6718520,
      "step": 11580
    },
    {
      "epoch": 1.7254989574024426,
      "grad_norm": 10.5,
      "learning_rate": 4.313375037235627e-05,
      "loss": 0.9902,
      "num_input_tokens_seen": 6721304,
      "step": 11585
    },
    {
      "epoch": 1.7262436699434018,
      "grad_norm": 5.09375,
      "learning_rate": 4.315236818588025e-05,
      "loss": 0.9076,
      "num_input_tokens_seen": 6724024,
      "step": 11590
    },
    {
      "epoch": 1.726988382484361,
      "grad_norm": 7.3125,
      "learning_rate": 4.3170985999404237e-05,
      "loss": 1.1016,
      "num_input_tokens_seen": 6726904,
      "step": 11595
    },
    {
      "epoch": 1.7277330950253202,
      "grad_norm": 8.875,
      "learning_rate": 4.3189603812928215e-05,
      "loss": 1.0002,
      "num_input_tokens_seen": 6729784,
      "step": 11600
    },
    {
      "epoch": 1.7284778075662794,
      "grad_norm": 6.28125,
      "learning_rate": 4.3208221626452186e-05,
      "loss": 0.8324,
      "num_input_tokens_seen": 6732504,
      "step": 11605
    },
    {
      "epoch": 1.7292225201072386,
      "grad_norm": 10.6875,
      "learning_rate": 4.322683943997617e-05,
      "loss": 1.4383,
      "num_input_tokens_seen": 6735320,
      "step": 11610
    },
    {
      "epoch": 1.7299672326481979,
      "grad_norm": 14.3125,
      "learning_rate": 4.324545725350015e-05,
      "loss": 1.2307,
      "num_input_tokens_seen": 6738200,
      "step": 11615
    },
    {
      "epoch": 1.730711945189157,
      "grad_norm": 15.0,
      "learning_rate": 4.326407506702413e-05,
      "loss": 1.4083,
      "num_input_tokens_seen": 6741304,
      "step": 11620
    },
    {
      "epoch": 1.7314566577301163,
      "grad_norm": 13.5,
      "learning_rate": 4.3282692880548106e-05,
      "loss": 1.4944,
      "num_input_tokens_seen": 6744184,
      "step": 11625
    },
    {
      "epoch": 1.7322013702710755,
      "grad_norm": 26.625,
      "learning_rate": 4.330131069407209e-05,
      "loss": 1.2352,
      "num_input_tokens_seen": 6747032,
      "step": 11630
    },
    {
      "epoch": 1.7329460828120347,
      "grad_norm": 14.875,
      "learning_rate": 4.331992850759607e-05,
      "loss": 1.2258,
      "num_input_tokens_seen": 6749880,
      "step": 11635
    },
    {
      "epoch": 1.7336907953529939,
      "grad_norm": 10.0625,
      "learning_rate": 4.333854632112005e-05,
      "loss": 1.5673,
      "num_input_tokens_seen": 6752728,
      "step": 11640
    },
    {
      "epoch": 1.734435507893953,
      "grad_norm": 8.9375,
      "learning_rate": 4.335716413464403e-05,
      "loss": 1.3544,
      "num_input_tokens_seen": 6755672,
      "step": 11645
    },
    {
      "epoch": 1.7351802204349123,
      "grad_norm": 8.6875,
      "learning_rate": 4.337578194816801e-05,
      "loss": 0.919,
      "num_input_tokens_seen": 6758200,
      "step": 11650
    },
    {
      "epoch": 1.7359249329758715,
      "grad_norm": 10.125,
      "learning_rate": 4.339439976169199e-05,
      "loss": 1.001,
      "num_input_tokens_seen": 6760888,
      "step": 11655
    },
    {
      "epoch": 1.7366696455168305,
      "grad_norm": 11.75,
      "learning_rate": 4.341301757521597e-05,
      "loss": 1.2343,
      "num_input_tokens_seen": 6763800,
      "step": 11660
    },
    {
      "epoch": 1.7374143580577897,
      "grad_norm": 8.625,
      "learning_rate": 4.343163538873995e-05,
      "loss": 1.2254,
      "num_input_tokens_seen": 6766520,
      "step": 11665
    },
    {
      "epoch": 1.7381590705987489,
      "grad_norm": 20.0,
      "learning_rate": 4.345025320226393e-05,
      "loss": 1.083,
      "num_input_tokens_seen": 6769560,
      "step": 11670
    },
    {
      "epoch": 1.738903783139708,
      "grad_norm": 12.6875,
      "learning_rate": 4.346887101578791e-05,
      "loss": 1.2854,
      "num_input_tokens_seen": 6772344,
      "step": 11675
    },
    {
      "epoch": 1.7396484956806673,
      "grad_norm": 6.34375,
      "learning_rate": 4.348748882931189e-05,
      "loss": 1.3467,
      "num_input_tokens_seen": 6774872,
      "step": 11680
    },
    {
      "epoch": 1.7403932082216265,
      "grad_norm": 13.4375,
      "learning_rate": 4.350610664283587e-05,
      "loss": 0.9955,
      "num_input_tokens_seen": 6777880,
      "step": 11685
    },
    {
      "epoch": 1.7411379207625857,
      "grad_norm": 11.5625,
      "learning_rate": 4.352472445635984e-05,
      "loss": 0.9975,
      "num_input_tokens_seen": 6780696,
      "step": 11690
    },
    {
      "epoch": 1.7418826333035449,
      "grad_norm": 11.625,
      "learning_rate": 4.354334226988382e-05,
      "loss": 1.2761,
      "num_input_tokens_seen": 6783544,
      "step": 11695
    },
    {
      "epoch": 1.742627345844504,
      "grad_norm": 8.125,
      "learning_rate": 4.3561960083407807e-05,
      "loss": 1.0734,
      "num_input_tokens_seen": 6786680,
      "step": 11700
    },
    {
      "epoch": 1.743372058385463,
      "grad_norm": 11.4375,
      "learning_rate": 4.3580577896931785e-05,
      "loss": 0.9804,
      "num_input_tokens_seen": 6789688,
      "step": 11705
    },
    {
      "epoch": 1.7441167709264223,
      "grad_norm": 16.625,
      "learning_rate": 4.359919571045576e-05,
      "loss": 1.2709,
      "num_input_tokens_seen": 6792952,
      "step": 11710
    },
    {
      "epoch": 1.7448614834673815,
      "grad_norm": 14.875,
      "learning_rate": 4.361781352397975e-05,
      "loss": 1.2376,
      "num_input_tokens_seen": 6795928,
      "step": 11715
    },
    {
      "epoch": 1.7456061960083407,
      "grad_norm": 5.6875,
      "learning_rate": 4.3636431337503726e-05,
      "loss": 1.0606,
      "num_input_tokens_seen": 6799160,
      "step": 11720
    },
    {
      "epoch": 1.7463509085492999,
      "grad_norm": 10.625,
      "learning_rate": 4.3655049151027704e-05,
      "loss": 1.0274,
      "num_input_tokens_seen": 6802232,
      "step": 11725
    },
    {
      "epoch": 1.747095621090259,
      "grad_norm": 11.9375,
      "learning_rate": 4.367366696455168e-05,
      "loss": 1.1888,
      "num_input_tokens_seen": 6805208,
      "step": 11730
    },
    {
      "epoch": 1.7478403336312183,
      "grad_norm": 10.4375,
      "learning_rate": 4.369228477807567e-05,
      "loss": 0.9675,
      "num_input_tokens_seen": 6808184,
      "step": 11735
    },
    {
      "epoch": 1.7485850461721775,
      "grad_norm": 22.75,
      "learning_rate": 4.3710902591599646e-05,
      "loss": 1.6261,
      "num_input_tokens_seen": 6811288,
      "step": 11740
    },
    {
      "epoch": 1.7493297587131367,
      "grad_norm": 8.125,
      "learning_rate": 4.3729520405123624e-05,
      "loss": 1.234,
      "num_input_tokens_seen": 6814328,
      "step": 11745
    },
    {
      "epoch": 1.7500744712540959,
      "grad_norm": 12.4375,
      "learning_rate": 4.37481382186476e-05,
      "loss": 1.2599,
      "num_input_tokens_seen": 6817240,
      "step": 11750
    },
    {
      "epoch": 1.750819183795055,
      "grad_norm": 13.8125,
      "learning_rate": 4.376675603217159e-05,
      "loss": 0.8884,
      "num_input_tokens_seen": 6819896,
      "step": 11755
    },
    {
      "epoch": 1.7515638963360143,
      "grad_norm": 9.9375,
      "learning_rate": 4.3785373845695566e-05,
      "loss": 0.9751,
      "num_input_tokens_seen": 6822840,
      "step": 11760
    },
    {
      "epoch": 1.7523086088769735,
      "grad_norm": 11.9375,
      "learning_rate": 4.3803991659219544e-05,
      "loss": 1.2536,
      "num_input_tokens_seen": 6825624,
      "step": 11765
    },
    {
      "epoch": 1.7530533214179327,
      "grad_norm": 9.9375,
      "learning_rate": 4.382260947274352e-05,
      "loss": 1.1791,
      "num_input_tokens_seen": 6828344,
      "step": 11770
    },
    {
      "epoch": 1.7537980339588919,
      "grad_norm": 15.5625,
      "learning_rate": 4.384122728626751e-05,
      "loss": 0.8482,
      "num_input_tokens_seen": 6831736,
      "step": 11775
    },
    {
      "epoch": 1.754542746499851,
      "grad_norm": 8.5625,
      "learning_rate": 4.385984509979148e-05,
      "loss": 1.152,
      "num_input_tokens_seen": 6834552,
      "step": 11780
    },
    {
      "epoch": 1.7552874590408103,
      "grad_norm": 12.75,
      "learning_rate": 4.387846291331546e-05,
      "loss": 0.8668,
      "num_input_tokens_seen": 6837560,
      "step": 11785
    },
    {
      "epoch": 1.7560321715817695,
      "grad_norm": 19.125,
      "learning_rate": 4.389708072683944e-05,
      "loss": 1.4533,
      "num_input_tokens_seen": 6840280,
      "step": 11790
    },
    {
      "epoch": 1.7567768841227287,
      "grad_norm": 12.9375,
      "learning_rate": 4.391569854036342e-05,
      "loss": 1.1727,
      "num_input_tokens_seen": 6843128,
      "step": 11795
    },
    {
      "epoch": 1.757521596663688,
      "grad_norm": 9.9375,
      "learning_rate": 4.39343163538874e-05,
      "loss": 1.0242,
      "num_input_tokens_seen": 6846328,
      "step": 11800
    },
    {
      "epoch": 1.758266309204647,
      "grad_norm": 14.0625,
      "learning_rate": 4.395293416741138e-05,
      "loss": 1.2978,
      "num_input_tokens_seen": 6849592,
      "step": 11805
    },
    {
      "epoch": 1.7590110217456063,
      "grad_norm": 7.46875,
      "learning_rate": 4.397155198093536e-05,
      "loss": 1.2133,
      "num_input_tokens_seen": 6852408,
      "step": 11810
    },
    {
      "epoch": 1.7597557342865655,
      "grad_norm": 11.1875,
      "learning_rate": 4.399016979445934e-05,
      "loss": 1.0979,
      "num_input_tokens_seen": 6855160,
      "step": 11815
    },
    {
      "epoch": 1.7605004468275247,
      "grad_norm": 11.0,
      "learning_rate": 4.400878760798332e-05,
      "loss": 1.031,
      "num_input_tokens_seen": 6857912,
      "step": 11820
    },
    {
      "epoch": 1.761245159368484,
      "grad_norm": 19.5,
      "learning_rate": 4.40274054215073e-05,
      "loss": 1.3905,
      "num_input_tokens_seen": 6860792,
      "step": 11825
    },
    {
      "epoch": 1.7619898719094431,
      "grad_norm": 19.25,
      "learning_rate": 4.404602323503128e-05,
      "loss": 1.4873,
      "num_input_tokens_seen": 6863640,
      "step": 11830
    },
    {
      "epoch": 1.762734584450402,
      "grad_norm": 12.875,
      "learning_rate": 4.406464104855526e-05,
      "loss": 1.2055,
      "num_input_tokens_seen": 6866296,
      "step": 11835
    },
    {
      "epoch": 1.7634792969913613,
      "grad_norm": 7.25,
      "learning_rate": 4.408325886207924e-05,
      "loss": 0.825,
      "num_input_tokens_seen": 6869592,
      "step": 11840
    },
    {
      "epoch": 1.7642240095323205,
      "grad_norm": 8.0625,
      "learning_rate": 4.410187667560322e-05,
      "loss": 0.9955,
      "num_input_tokens_seen": 6872056,
      "step": 11845
    },
    {
      "epoch": 1.7649687220732797,
      "grad_norm": 8.9375,
      "learning_rate": 4.41204944891272e-05,
      "loss": 1.0757,
      "num_input_tokens_seen": 6874968,
      "step": 11850
    },
    {
      "epoch": 1.765713434614239,
      "grad_norm": 14.375,
      "learning_rate": 4.413911230265118e-05,
      "loss": 0.7705,
      "num_input_tokens_seen": 6877976,
      "step": 11855
    },
    {
      "epoch": 1.766458147155198,
      "grad_norm": 12.1875,
      "learning_rate": 4.4157730116175164e-05,
      "loss": 0.7936,
      "num_input_tokens_seen": 6881016,
      "step": 11860
    },
    {
      "epoch": 1.7672028596961573,
      "grad_norm": 8.4375,
      "learning_rate": 4.4176347929699136e-05,
      "loss": 1.1228,
      "num_input_tokens_seen": 6883832,
      "step": 11865
    },
    {
      "epoch": 1.7679475722371165,
      "grad_norm": 8.875,
      "learning_rate": 4.4194965743223114e-05,
      "loss": 1.0044,
      "num_input_tokens_seen": 6886680,
      "step": 11870
    },
    {
      "epoch": 1.7686922847780755,
      "grad_norm": 8.25,
      "learning_rate": 4.42135835567471e-05,
      "loss": 1.1756,
      "num_input_tokens_seen": 6889656,
      "step": 11875
    },
    {
      "epoch": 1.7694369973190347,
      "grad_norm": 7.875,
      "learning_rate": 4.423220137027108e-05,
      "loss": 0.8816,
      "num_input_tokens_seen": 6892920,
      "step": 11880
    },
    {
      "epoch": 1.770181709859994,
      "grad_norm": 16.875,
      "learning_rate": 4.4250819183795055e-05,
      "loss": 1.129,
      "num_input_tokens_seen": 6895928,
      "step": 11885
    },
    {
      "epoch": 1.770926422400953,
      "grad_norm": 9.75,
      "learning_rate": 4.4269436997319034e-05,
      "loss": 1.2624,
      "num_input_tokens_seen": 6898680,
      "step": 11890
    },
    {
      "epoch": 1.7716711349419123,
      "grad_norm": 6.34375,
      "learning_rate": 4.428805481084302e-05,
      "loss": 1.016,
      "num_input_tokens_seen": 6901752,
      "step": 11895
    },
    {
      "epoch": 1.7724158474828715,
      "grad_norm": 13.5625,
      "learning_rate": 4.4306672624367e-05,
      "loss": 1.2275,
      "num_input_tokens_seen": 6904472,
      "step": 11900
    },
    {
      "epoch": 1.7731605600238307,
      "grad_norm": 11.5625,
      "learning_rate": 4.4325290437890975e-05,
      "loss": 0.901,
      "num_input_tokens_seen": 6907384,
      "step": 11905
    },
    {
      "epoch": 1.77390527256479,
      "grad_norm": 16.875,
      "learning_rate": 4.434390825141495e-05,
      "loss": 1.0932,
      "num_input_tokens_seen": 6910168,
      "step": 11910
    },
    {
      "epoch": 1.7746499851057491,
      "grad_norm": 8.0625,
      "learning_rate": 4.436252606493894e-05,
      "loss": 1.2485,
      "num_input_tokens_seen": 6912888,
      "step": 11915
    },
    {
      "epoch": 1.7753946976467083,
      "grad_norm": 16.625,
      "learning_rate": 4.4381143878462917e-05,
      "loss": 1.1383,
      "num_input_tokens_seen": 6915544,
      "step": 11920
    },
    {
      "epoch": 1.7761394101876675,
      "grad_norm": 11.6875,
      "learning_rate": 4.4399761691986895e-05,
      "loss": 1.7622,
      "num_input_tokens_seen": 6918456,
      "step": 11925
    },
    {
      "epoch": 1.7768841227286267,
      "grad_norm": 10.75,
      "learning_rate": 4.441837950551088e-05,
      "loss": 0.7957,
      "num_input_tokens_seen": 6921464,
      "step": 11930
    },
    {
      "epoch": 1.777628835269586,
      "grad_norm": 13.125,
      "learning_rate": 4.443699731903486e-05,
      "loss": 1.3254,
      "num_input_tokens_seen": 6924216,
      "step": 11935
    },
    {
      "epoch": 1.7783735478105451,
      "grad_norm": 8.0,
      "learning_rate": 4.4455615132558836e-05,
      "loss": 1.1092,
      "num_input_tokens_seen": 6927032,
      "step": 11940
    },
    {
      "epoch": 1.7791182603515043,
      "grad_norm": 6.0625,
      "learning_rate": 4.4474232946082814e-05,
      "loss": 1.1089,
      "num_input_tokens_seen": 6929912,
      "step": 11945
    },
    {
      "epoch": 1.7798629728924635,
      "grad_norm": 10.25,
      "learning_rate": 4.449285075960679e-05,
      "loss": 0.9616,
      "num_input_tokens_seen": 6933240,
      "step": 11950
    },
    {
      "epoch": 1.7806076854334227,
      "grad_norm": 13.25,
      "learning_rate": 4.451146857313077e-05,
      "loss": 1.111,
      "num_input_tokens_seen": 6936056,
      "step": 11955
    },
    {
      "epoch": 1.781352397974382,
      "grad_norm": 19.0,
      "learning_rate": 4.453008638665475e-05,
      "loss": 0.9463,
      "num_input_tokens_seen": 6938744,
      "step": 11960
    },
    {
      "epoch": 1.7820971105153411,
      "grad_norm": 16.125,
      "learning_rate": 4.4548704200178734e-05,
      "loss": 1.1942,
      "num_input_tokens_seen": 6941752,
      "step": 11965
    },
    {
      "epoch": 1.7828418230563003,
      "grad_norm": 19.75,
      "learning_rate": 4.456732201370271e-05,
      "loss": 1.1351,
      "num_input_tokens_seen": 6944856,
      "step": 11970
    },
    {
      "epoch": 1.7835865355972595,
      "grad_norm": 5.1875,
      "learning_rate": 4.458593982722669e-05,
      "loss": 0.965,
      "num_input_tokens_seen": 6947800,
      "step": 11975
    },
    {
      "epoch": 1.7843312481382188,
      "grad_norm": 8.875,
      "learning_rate": 4.460455764075067e-05,
      "loss": 0.7821,
      "num_input_tokens_seen": 6950520,
      "step": 11980
    },
    {
      "epoch": 1.785075960679178,
      "grad_norm": 15.625,
      "learning_rate": 4.4623175454274654e-05,
      "loss": 1.7408,
      "num_input_tokens_seen": 6953368,
      "step": 11985
    },
    {
      "epoch": 1.7858206732201372,
      "grad_norm": 15.0625,
      "learning_rate": 4.464179326779863e-05,
      "loss": 1.1463,
      "num_input_tokens_seen": 6956376,
      "step": 11990
    },
    {
      "epoch": 1.7865653857610964,
      "grad_norm": 11.0,
      "learning_rate": 4.466041108132261e-05,
      "loss": 0.9291,
      "num_input_tokens_seen": 6959224,
      "step": 11995
    },
    {
      "epoch": 1.7873100983020556,
      "grad_norm": 6.6875,
      "learning_rate": 4.467902889484659e-05,
      "loss": 1.0634,
      "num_input_tokens_seen": 6962072,
      "step": 12000
    },
    {
      "epoch": 1.7880548108430145,
      "grad_norm": 7.34375,
      "learning_rate": 4.4697646708370574e-05,
      "loss": 0.985,
      "num_input_tokens_seen": 6965208,
      "step": 12005
    },
    {
      "epoch": 1.7887995233839737,
      "grad_norm": 16.875,
      "learning_rate": 4.471626452189455e-05,
      "loss": 0.9683,
      "num_input_tokens_seen": 6968280,
      "step": 12010
    },
    {
      "epoch": 1.789544235924933,
      "grad_norm": 16.875,
      "learning_rate": 4.473488233541853e-05,
      "loss": 1.069,
      "num_input_tokens_seen": 6970936,
      "step": 12015
    },
    {
      "epoch": 1.7902889484658921,
      "grad_norm": 7.96875,
      "learning_rate": 4.4753500148942515e-05,
      "loss": 0.9058,
      "num_input_tokens_seen": 6973816,
      "step": 12020
    },
    {
      "epoch": 1.7910336610068514,
      "grad_norm": 18.375,
      "learning_rate": 4.477211796246649e-05,
      "loss": 1.0349,
      "num_input_tokens_seen": 6976664,
      "step": 12025
    },
    {
      "epoch": 1.7917783735478106,
      "grad_norm": 8.6875,
      "learning_rate": 4.479073577599047e-05,
      "loss": 0.7928,
      "num_input_tokens_seen": 6979704,
      "step": 12030
    },
    {
      "epoch": 1.7925230860887698,
      "grad_norm": 11.125,
      "learning_rate": 4.480935358951445e-05,
      "loss": 1.2175,
      "num_input_tokens_seen": 6982456,
      "step": 12035
    },
    {
      "epoch": 1.793267798629729,
      "grad_norm": 10.4375,
      "learning_rate": 4.482797140303843e-05,
      "loss": 0.904,
      "num_input_tokens_seen": 6985176,
      "step": 12040
    },
    {
      "epoch": 1.7940125111706882,
      "grad_norm": 5.25,
      "learning_rate": 4.4846589216562406e-05,
      "loss": 1.372,
      "num_input_tokens_seen": 6987960,
      "step": 12045
    },
    {
      "epoch": 1.7947572237116471,
      "grad_norm": 14.9375,
      "learning_rate": 4.4865207030086384e-05,
      "loss": 1.2007,
      "num_input_tokens_seen": 6991064,
      "step": 12050
    },
    {
      "epoch": 1.7955019362526063,
      "grad_norm": 11.8125,
      "learning_rate": 4.488382484361037e-05,
      "loss": 0.7677,
      "num_input_tokens_seen": 6994008,
      "step": 12055
    },
    {
      "epoch": 1.7962466487935655,
      "grad_norm": 6.625,
      "learning_rate": 4.490244265713435e-05,
      "loss": 1.2801,
      "num_input_tokens_seen": 6997208,
      "step": 12060
    },
    {
      "epoch": 1.7969913613345248,
      "grad_norm": 11.8125,
      "learning_rate": 4.4921060470658326e-05,
      "loss": 1.354,
      "num_input_tokens_seen": 7000120,
      "step": 12065
    },
    {
      "epoch": 1.797736073875484,
      "grad_norm": 9.8125,
      "learning_rate": 4.4939678284182304e-05,
      "loss": 0.815,
      "num_input_tokens_seen": 7003160,
      "step": 12070
    },
    {
      "epoch": 1.7984807864164432,
      "grad_norm": 13.5,
      "learning_rate": 4.495829609770629e-05,
      "loss": 0.8087,
      "num_input_tokens_seen": 7006008,
      "step": 12075
    },
    {
      "epoch": 1.7992254989574024,
      "grad_norm": 11.0,
      "learning_rate": 4.497691391123027e-05,
      "loss": 1.078,
      "num_input_tokens_seen": 7008792,
      "step": 12080
    },
    {
      "epoch": 1.7999702114983616,
      "grad_norm": 16.375,
      "learning_rate": 4.4995531724754246e-05,
      "loss": 1.011,
      "num_input_tokens_seen": 7011736,
      "step": 12085
    },
    {
      "epoch": 1.8007149240393208,
      "grad_norm": 21.25,
      "learning_rate": 4.501414953827823e-05,
      "loss": 1.363,
      "num_input_tokens_seen": 7014296,
      "step": 12090
    },
    {
      "epoch": 1.80145963658028,
      "grad_norm": 13.3125,
      "learning_rate": 4.503276735180221e-05,
      "loss": 1.3046,
      "num_input_tokens_seen": 7017656,
      "step": 12095
    },
    {
      "epoch": 1.8022043491212392,
      "grad_norm": 8.25,
      "learning_rate": 4.505138516532619e-05,
      "loss": 1.058,
      "num_input_tokens_seen": 7020536,
      "step": 12100
    },
    {
      "epoch": 1.8029490616621984,
      "grad_norm": 17.625,
      "learning_rate": 4.5070002978850165e-05,
      "loss": 1.0642,
      "num_input_tokens_seen": 7023672,
      "step": 12105
    },
    {
      "epoch": 1.8036937742031576,
      "grad_norm": 15.25,
      "learning_rate": 4.508862079237415e-05,
      "loss": 1.2025,
      "num_input_tokens_seen": 7026232,
      "step": 12110
    },
    {
      "epoch": 1.8044384867441168,
      "grad_norm": 14.9375,
      "learning_rate": 4.510723860589813e-05,
      "loss": 1.2689,
      "num_input_tokens_seen": 7029496,
      "step": 12115
    },
    {
      "epoch": 1.805183199285076,
      "grad_norm": 15.5625,
      "learning_rate": 4.512585641942211e-05,
      "loss": 1.1384,
      "num_input_tokens_seen": 7032312,
      "step": 12120
    },
    {
      "epoch": 1.8059279118260352,
      "grad_norm": 23.25,
      "learning_rate": 4.5144474232946085e-05,
      "loss": 1.3252,
      "num_input_tokens_seen": 7035256,
      "step": 12125
    },
    {
      "epoch": 1.8066726243669944,
      "grad_norm": 15.1875,
      "learning_rate": 4.516309204647006e-05,
      "loss": 1.0359,
      "num_input_tokens_seen": 7038008,
      "step": 12130
    },
    {
      "epoch": 1.8074173369079536,
      "grad_norm": 15.25,
      "learning_rate": 4.518170985999404e-05,
      "loss": 1.3087,
      "num_input_tokens_seen": 7040824,
      "step": 12135
    },
    {
      "epoch": 1.8081620494489128,
      "grad_norm": 8.8125,
      "learning_rate": 4.520032767351802e-05,
      "loss": 1.0642,
      "num_input_tokens_seen": 7043960,
      "step": 12140
    },
    {
      "epoch": 1.808906761989872,
      "grad_norm": 12.625,
      "learning_rate": 4.5218945487042005e-05,
      "loss": 1.1775,
      "num_input_tokens_seen": 7046840,
      "step": 12145
    },
    {
      "epoch": 1.8096514745308312,
      "grad_norm": 7.875,
      "learning_rate": 4.523756330056598e-05,
      "loss": 1.1381,
      "num_input_tokens_seen": 7050104,
      "step": 12150
    },
    {
      "epoch": 1.8103961870717904,
      "grad_norm": 16.5,
      "learning_rate": 4.525618111408996e-05,
      "loss": 1.092,
      "num_input_tokens_seen": 7052728,
      "step": 12155
    },
    {
      "epoch": 1.8111408996127496,
      "grad_norm": 10.3125,
      "learning_rate": 4.527479892761394e-05,
      "loss": 1.1828,
      "num_input_tokens_seen": 7055800,
      "step": 12160
    },
    {
      "epoch": 1.8118856121537088,
      "grad_norm": 7.46875,
      "learning_rate": 4.5293416741137924e-05,
      "loss": 1.2968,
      "num_input_tokens_seen": 7058904,
      "step": 12165
    },
    {
      "epoch": 1.812630324694668,
      "grad_norm": 10.4375,
      "learning_rate": 4.53120345546619e-05,
      "loss": 1.1059,
      "num_input_tokens_seen": 7061688,
      "step": 12170
    },
    {
      "epoch": 1.8133750372356272,
      "grad_norm": 10.75,
      "learning_rate": 4.533065236818588e-05,
      "loss": 1.0268,
      "num_input_tokens_seen": 7064504,
      "step": 12175
    },
    {
      "epoch": 1.8141197497765862,
      "grad_norm": 3.484375,
      "learning_rate": 4.5349270181709866e-05,
      "loss": 0.8907,
      "num_input_tokens_seen": 7067512,
      "step": 12180
    },
    {
      "epoch": 1.8148644623175454,
      "grad_norm": 7.0,
      "learning_rate": 4.5367887995233844e-05,
      "loss": 1.1242,
      "num_input_tokens_seen": 7070648,
      "step": 12185
    },
    {
      "epoch": 1.8156091748585046,
      "grad_norm": 16.75,
      "learning_rate": 4.538650580875782e-05,
      "loss": 1.1532,
      "num_input_tokens_seen": 7073336,
      "step": 12190
    },
    {
      "epoch": 1.8163538873994638,
      "grad_norm": 8.4375,
      "learning_rate": 4.54051236222818e-05,
      "loss": 0.9241,
      "num_input_tokens_seen": 7076152,
      "step": 12195
    },
    {
      "epoch": 1.817098599940423,
      "grad_norm": 13.3125,
      "learning_rate": 4.5423741435805786e-05,
      "loss": 0.9868,
      "num_input_tokens_seen": 7078968,
      "step": 12200
    },
    {
      "epoch": 1.8178433124813822,
      "grad_norm": 6.15625,
      "learning_rate": 4.5442359249329764e-05,
      "loss": 1.2634,
      "num_input_tokens_seen": 7082168,
      "step": 12205
    },
    {
      "epoch": 1.8185880250223414,
      "grad_norm": 7.90625,
      "learning_rate": 4.5460977062853735e-05,
      "loss": 1.0089,
      "num_input_tokens_seen": 7085144,
      "step": 12210
    },
    {
      "epoch": 1.8193327375633006,
      "grad_norm": 9.0625,
      "learning_rate": 4.547959487637772e-05,
      "loss": 1.1643,
      "num_input_tokens_seen": 7087896,
      "step": 12215
    },
    {
      "epoch": 1.8200774501042598,
      "grad_norm": 4.1875,
      "learning_rate": 4.54982126899017e-05,
      "loss": 1.0471,
      "num_input_tokens_seen": 7090808,
      "step": 12220
    },
    {
      "epoch": 1.8208221626452188,
      "grad_norm": 15.4375,
      "learning_rate": 4.551683050342568e-05,
      "loss": 1.3766,
      "num_input_tokens_seen": 7093624,
      "step": 12225
    },
    {
      "epoch": 1.821566875186178,
      "grad_norm": 10.0625,
      "learning_rate": 4.5535448316949655e-05,
      "loss": 1.1809,
      "num_input_tokens_seen": 7096408,
      "step": 12230
    },
    {
      "epoch": 1.8223115877271372,
      "grad_norm": 7.90625,
      "learning_rate": 4.555406613047364e-05,
      "loss": 0.9865,
      "num_input_tokens_seen": 7099192,
      "step": 12235
    },
    {
      "epoch": 1.8230563002680964,
      "grad_norm": 8.375,
      "learning_rate": 4.557268394399762e-05,
      "loss": 1.1568,
      "num_input_tokens_seen": 7102008,
      "step": 12240
    },
    {
      "epoch": 1.8238010128090556,
      "grad_norm": 5.59375,
      "learning_rate": 4.5591301757521596e-05,
      "loss": 1.0802,
      "num_input_tokens_seen": 7105016,
      "step": 12245
    },
    {
      "epoch": 1.8245457253500148,
      "grad_norm": 24.25,
      "learning_rate": 4.560991957104558e-05,
      "loss": 1.2795,
      "num_input_tokens_seen": 7107640,
      "step": 12250
    },
    {
      "epoch": 1.825290437890974,
      "grad_norm": 10.625,
      "learning_rate": 4.562853738456956e-05,
      "loss": 1.1755,
      "num_input_tokens_seen": 7110328,
      "step": 12255
    },
    {
      "epoch": 1.8260351504319332,
      "grad_norm": 11.1875,
      "learning_rate": 4.564715519809354e-05,
      "loss": 0.7838,
      "num_input_tokens_seen": 7113112,
      "step": 12260
    },
    {
      "epoch": 1.8267798629728924,
      "grad_norm": 9.9375,
      "learning_rate": 4.5665773011617516e-05,
      "loss": 1.189,
      "num_input_tokens_seen": 7116216,
      "step": 12265
    },
    {
      "epoch": 1.8275245755138516,
      "grad_norm": 16.375,
      "learning_rate": 4.56843908251415e-05,
      "loss": 1.1576,
      "num_input_tokens_seen": 7119096,
      "step": 12270
    },
    {
      "epoch": 1.8282692880548108,
      "grad_norm": 8.625,
      "learning_rate": 4.570300863866548e-05,
      "loss": 1.0926,
      "num_input_tokens_seen": 7122104,
      "step": 12275
    },
    {
      "epoch": 1.82901400059577,
      "grad_norm": 12.8125,
      "learning_rate": 4.572162645218946e-05,
      "loss": 1.3316,
      "num_input_tokens_seen": 7125048,
      "step": 12280
    },
    {
      "epoch": 1.8297587131367292,
      "grad_norm": 14.625,
      "learning_rate": 4.5740244265713436e-05,
      "loss": 1.1078,
      "num_input_tokens_seen": 7127992,
      "step": 12285
    },
    {
      "epoch": 1.8305034256776884,
      "grad_norm": 12.4375,
      "learning_rate": 4.575886207923742e-05,
      "loss": 0.7511,
      "num_input_tokens_seen": 7130744,
      "step": 12290
    },
    {
      "epoch": 1.8312481382186476,
      "grad_norm": 14.3125,
      "learning_rate": 4.57774798927614e-05,
      "loss": 1.2694,
      "num_input_tokens_seen": 7133496,
      "step": 12295
    },
    {
      "epoch": 1.8319928507596068,
      "grad_norm": 10.8125,
      "learning_rate": 4.579609770628537e-05,
      "loss": 0.8611,
      "num_input_tokens_seen": 7136504,
      "step": 12300
    },
    {
      "epoch": 1.832737563300566,
      "grad_norm": 10.0,
      "learning_rate": 4.5814715519809356e-05,
      "loss": 1.2243,
      "num_input_tokens_seen": 7139896,
      "step": 12305
    },
    {
      "epoch": 1.8334822758415252,
      "grad_norm": 13.8125,
      "learning_rate": 4.5833333333333334e-05,
      "loss": 0.9805,
      "num_input_tokens_seen": 7142520,
      "step": 12310
    },
    {
      "epoch": 1.8342269883824844,
      "grad_norm": 11.3125,
      "learning_rate": 4.585195114685731e-05,
      "loss": 1.3165,
      "num_input_tokens_seen": 7145432,
      "step": 12315
    },
    {
      "epoch": 1.8349717009234436,
      "grad_norm": 9.1875,
      "learning_rate": 4.58705689603813e-05,
      "loss": 1.0759,
      "num_input_tokens_seen": 7148408,
      "step": 12320
    },
    {
      "epoch": 1.8357164134644028,
      "grad_norm": 7.15625,
      "learning_rate": 4.5889186773905275e-05,
      "loss": 0.9793,
      "num_input_tokens_seen": 7151160,
      "step": 12325
    },
    {
      "epoch": 1.836461126005362,
      "grad_norm": 7.9375,
      "learning_rate": 4.5907804587429254e-05,
      "loss": 1.215,
      "num_input_tokens_seen": 7153976,
      "step": 12330
    },
    {
      "epoch": 1.8372058385463212,
      "grad_norm": 7.0625,
      "learning_rate": 4.592642240095323e-05,
      "loss": 1.1141,
      "num_input_tokens_seen": 7156888,
      "step": 12335
    },
    {
      "epoch": 1.8379505510872804,
      "grad_norm": 10.5625,
      "learning_rate": 4.594504021447722e-05,
      "loss": 0.9729,
      "num_input_tokens_seen": 7159800,
      "step": 12340
    },
    {
      "epoch": 1.8386952636282397,
      "grad_norm": 13.3125,
      "learning_rate": 4.5963658028001195e-05,
      "loss": 1.2642,
      "num_input_tokens_seen": 7162936,
      "step": 12345
    },
    {
      "epoch": 1.8394399761691989,
      "grad_norm": 17.5,
      "learning_rate": 4.598227584152517e-05,
      "loss": 1.2284,
      "num_input_tokens_seen": 7165688,
      "step": 12350
    },
    {
      "epoch": 1.8401846887101578,
      "grad_norm": 14.6875,
      "learning_rate": 4.600089365504915e-05,
      "loss": 1.1347,
      "num_input_tokens_seen": 7168248,
      "step": 12355
    },
    {
      "epoch": 1.840929401251117,
      "grad_norm": 9.8125,
      "learning_rate": 4.6019511468573136e-05,
      "loss": 1.3651,
      "num_input_tokens_seen": 7170872,
      "step": 12360
    },
    {
      "epoch": 1.8416741137920762,
      "grad_norm": 11.125,
      "learning_rate": 4.6038129282097115e-05,
      "loss": 0.7939,
      "num_input_tokens_seen": 7173656,
      "step": 12365
    },
    {
      "epoch": 1.8424188263330354,
      "grad_norm": 12.5,
      "learning_rate": 4.605674709562109e-05,
      "loss": 1.1986,
      "num_input_tokens_seen": 7176728,
      "step": 12370
    },
    {
      "epoch": 1.8431635388739946,
      "grad_norm": 9.5,
      "learning_rate": 4.607536490914507e-05,
      "loss": 1.1633,
      "num_input_tokens_seen": 7179640,
      "step": 12375
    },
    {
      "epoch": 1.8439082514149538,
      "grad_norm": 17.25,
      "learning_rate": 4.6093982722669056e-05,
      "loss": 1.1809,
      "num_input_tokens_seen": 7182424,
      "step": 12380
    },
    {
      "epoch": 1.844652963955913,
      "grad_norm": 8.625,
      "learning_rate": 4.611260053619303e-05,
      "loss": 1.403,
      "num_input_tokens_seen": 7185304,
      "step": 12385
    },
    {
      "epoch": 1.8453976764968723,
      "grad_norm": 18.625,
      "learning_rate": 4.6131218349717006e-05,
      "loss": 1.0721,
      "num_input_tokens_seen": 7188248,
      "step": 12390
    },
    {
      "epoch": 1.8461423890378312,
      "grad_norm": 11.8125,
      "learning_rate": 4.614983616324099e-05,
      "loss": 1.7234,
      "num_input_tokens_seen": 7191192,
      "step": 12395
    },
    {
      "epoch": 1.8468871015787904,
      "grad_norm": 18.25,
      "learning_rate": 4.616845397676497e-05,
      "loss": 1.093,
      "num_input_tokens_seen": 7193976,
      "step": 12400
    },
    {
      "epoch": 1.8476318141197496,
      "grad_norm": 9.0,
      "learning_rate": 4.618707179028895e-05,
      "loss": 0.8507,
      "num_input_tokens_seen": 7196824,
      "step": 12405
    },
    {
      "epoch": 1.8483765266607088,
      "grad_norm": 9.0,
      "learning_rate": 4.620568960381293e-05,
      "loss": 0.9252,
      "num_input_tokens_seen": 7199672,
      "step": 12410
    },
    {
      "epoch": 1.849121239201668,
      "grad_norm": 8.1875,
      "learning_rate": 4.622430741733691e-05,
      "loss": 1.1168,
      "num_input_tokens_seen": 7202648,
      "step": 12415
    },
    {
      "epoch": 1.8498659517426272,
      "grad_norm": 10.8125,
      "learning_rate": 4.624292523086089e-05,
      "loss": 1.0072,
      "num_input_tokens_seen": 7205688,
      "step": 12420
    },
    {
      "epoch": 1.8506106642835864,
      "grad_norm": 10.125,
      "learning_rate": 4.626154304438487e-05,
      "loss": 1.0126,
      "num_input_tokens_seen": 7208824,
      "step": 12425
    },
    {
      "epoch": 1.8513553768245457,
      "grad_norm": 9.1875,
      "learning_rate": 4.628016085790885e-05,
      "loss": 1.0513,
      "num_input_tokens_seen": 7211960,
      "step": 12430
    },
    {
      "epoch": 1.8521000893655049,
      "grad_norm": 16.625,
      "learning_rate": 4.629877867143283e-05,
      "loss": 1.0496,
      "num_input_tokens_seen": 7214776,
      "step": 12435
    },
    {
      "epoch": 1.852844801906464,
      "grad_norm": 11.9375,
      "learning_rate": 4.631739648495681e-05,
      "loss": 1.1432,
      "num_input_tokens_seen": 7218072,
      "step": 12440
    },
    {
      "epoch": 1.8535895144474233,
      "grad_norm": 11.0,
      "learning_rate": 4.633601429848079e-05,
      "loss": 1.2413,
      "num_input_tokens_seen": 7221272,
      "step": 12445
    },
    {
      "epoch": 1.8543342269883825,
      "grad_norm": 17.875,
      "learning_rate": 4.635463211200477e-05,
      "loss": 1.156,
      "num_input_tokens_seen": 7224120,
      "step": 12450
    },
    {
      "epoch": 1.8550789395293417,
      "grad_norm": 10.25,
      "learning_rate": 4.637324992552875e-05,
      "loss": 1.2875,
      "num_input_tokens_seen": 7227256,
      "step": 12455
    },
    {
      "epoch": 1.8558236520703009,
      "grad_norm": 7.96875,
      "learning_rate": 4.639186773905273e-05,
      "loss": 1.0006,
      "num_input_tokens_seen": 7230520,
      "step": 12460
    },
    {
      "epoch": 1.85656836461126,
      "grad_norm": 10.625,
      "learning_rate": 4.641048555257671e-05,
      "loss": 0.9327,
      "num_input_tokens_seen": 7233272,
      "step": 12465
    },
    {
      "epoch": 1.8573130771522193,
      "grad_norm": 8.375,
      "learning_rate": 4.6429103366100685e-05,
      "loss": 1.1957,
      "num_input_tokens_seen": 7236024,
      "step": 12470
    },
    {
      "epoch": 1.8580577896931785,
      "grad_norm": 9.375,
      "learning_rate": 4.644772117962466e-05,
      "loss": 1.1361,
      "num_input_tokens_seen": 7238840,
      "step": 12475
    },
    {
      "epoch": 1.8588025022341377,
      "grad_norm": 8.5625,
      "learning_rate": 4.646633899314865e-05,
      "loss": 0.8064,
      "num_input_tokens_seen": 7241784,
      "step": 12480
    },
    {
      "epoch": 1.8595472147750969,
      "grad_norm": 4.28125,
      "learning_rate": 4.6484956806672626e-05,
      "loss": 1.089,
      "num_input_tokens_seen": 7244440,
      "step": 12485
    },
    {
      "epoch": 1.860291927316056,
      "grad_norm": 7.4375,
      "learning_rate": 4.6503574620196604e-05,
      "loss": 0.9703,
      "num_input_tokens_seen": 7247320,
      "step": 12490
    },
    {
      "epoch": 1.8610366398570153,
      "grad_norm": 13.3125,
      "learning_rate": 4.652219243372058e-05,
      "loss": 1.3872,
      "num_input_tokens_seen": 7249944,
      "step": 12495
    },
    {
      "epoch": 1.8617813523979745,
      "grad_norm": 10.5625,
      "learning_rate": 4.654081024724457e-05,
      "loss": 1.4435,
      "num_input_tokens_seen": 7253112,
      "step": 12500
    },
    {
      "epoch": 1.8625260649389337,
      "grad_norm": 7.6875,
      "learning_rate": 4.6559428060768546e-05,
      "loss": 0.8527,
      "num_input_tokens_seen": 7256472,
      "step": 12505
    },
    {
      "epoch": 1.863270777479893,
      "grad_norm": 8.0,
      "learning_rate": 4.6578045874292524e-05,
      "loss": 0.8526,
      "num_input_tokens_seen": 7259224,
      "step": 12510
    },
    {
      "epoch": 1.864015490020852,
      "grad_norm": 13.875,
      "learning_rate": 4.65966636878165e-05,
      "loss": 1.1167,
      "num_input_tokens_seen": 7261912,
      "step": 12515
    },
    {
      "epoch": 1.8647602025618113,
      "grad_norm": 10.125,
      "learning_rate": 4.661528150134049e-05,
      "loss": 1.0339,
      "num_input_tokens_seen": 7264728,
      "step": 12520
    },
    {
      "epoch": 1.8655049151027703,
      "grad_norm": 5.84375,
      "learning_rate": 4.6633899314864466e-05,
      "loss": 0.9818,
      "num_input_tokens_seen": 7267672,
      "step": 12525
    },
    {
      "epoch": 1.8662496276437295,
      "grad_norm": 17.375,
      "learning_rate": 4.6652517128388444e-05,
      "loss": 1.2419,
      "num_input_tokens_seen": 7270552,
      "step": 12530
    },
    {
      "epoch": 1.8669943401846887,
      "grad_norm": 7.8125,
      "learning_rate": 4.667113494191243e-05,
      "loss": 0.9462,
      "num_input_tokens_seen": 7273240,
      "step": 12535
    },
    {
      "epoch": 1.8677390527256479,
      "grad_norm": 6.84375,
      "learning_rate": 4.668975275543641e-05,
      "loss": 1.0538,
      "num_input_tokens_seen": 7276216,
      "step": 12540
    },
    {
      "epoch": 1.868483765266607,
      "grad_norm": 12.75,
      "learning_rate": 4.6708370568960385e-05,
      "loss": 1.0676,
      "num_input_tokens_seen": 7278904,
      "step": 12545
    },
    {
      "epoch": 1.8692284778075663,
      "grad_norm": 11.6875,
      "learning_rate": 4.6726988382484364e-05,
      "loss": 1.2239,
      "num_input_tokens_seen": 7281720,
      "step": 12550
    },
    {
      "epoch": 1.8699731903485255,
      "grad_norm": 17.5,
      "learning_rate": 4.674560619600835e-05,
      "loss": 1.2463,
      "num_input_tokens_seen": 7284504,
      "step": 12555
    },
    {
      "epoch": 1.8707179028894847,
      "grad_norm": 7.59375,
      "learning_rate": 4.676422400953232e-05,
      "loss": 1.2725,
      "num_input_tokens_seen": 7287160,
      "step": 12560
    },
    {
      "epoch": 1.871462615430444,
      "grad_norm": 25.125,
      "learning_rate": 4.67828418230563e-05,
      "loss": 1.3635,
      "num_input_tokens_seen": 7290008,
      "step": 12565
    },
    {
      "epoch": 1.8722073279714029,
      "grad_norm": 3.53125,
      "learning_rate": 4.680145963658028e-05,
      "loss": 0.6521,
      "num_input_tokens_seen": 7292920,
      "step": 12570
    },
    {
      "epoch": 1.872952040512362,
      "grad_norm": 6.90625,
      "learning_rate": 4.682007745010426e-05,
      "loss": 0.8572,
      "num_input_tokens_seen": 7295928,
      "step": 12575
    },
    {
      "epoch": 1.8736967530533213,
      "grad_norm": 13.5,
      "learning_rate": 4.683869526362824e-05,
      "loss": 1.1599,
      "num_input_tokens_seen": 7298968,
      "step": 12580
    },
    {
      "epoch": 1.8744414655942805,
      "grad_norm": 6.75,
      "learning_rate": 4.685731307715222e-05,
      "loss": 1.2418,
      "num_input_tokens_seen": 7302104,
      "step": 12585
    },
    {
      "epoch": 1.8751861781352397,
      "grad_norm": 3.5625,
      "learning_rate": 4.68759308906762e-05,
      "loss": 0.9288,
      "num_input_tokens_seen": 7305080,
      "step": 12590
    },
    {
      "epoch": 1.875930890676199,
      "grad_norm": 9.125,
      "learning_rate": 4.689454870420018e-05,
      "loss": 1.2507,
      "num_input_tokens_seen": 7308152,
      "step": 12595
    },
    {
      "epoch": 1.876675603217158,
      "grad_norm": 10.1875,
      "learning_rate": 4.691316651772416e-05,
      "loss": 1.0123,
      "num_input_tokens_seen": 7310840,
      "step": 12600
    },
    {
      "epoch": 1.8774203157581173,
      "grad_norm": 14.25,
      "learning_rate": 4.693178433124814e-05,
      "loss": 0.9607,
      "num_input_tokens_seen": 7313368,
      "step": 12605
    },
    {
      "epoch": 1.8781650282990765,
      "grad_norm": 13.5,
      "learning_rate": 4.695040214477212e-05,
      "loss": 1.0751,
      "num_input_tokens_seen": 7316728,
      "step": 12610
    },
    {
      "epoch": 1.8789097408400357,
      "grad_norm": 15.3125,
      "learning_rate": 4.69690199582961e-05,
      "loss": 1.3943,
      "num_input_tokens_seen": 7319704,
      "step": 12615
    },
    {
      "epoch": 1.879654453380995,
      "grad_norm": 11.4375,
      "learning_rate": 4.698763777182008e-05,
      "loss": 1.067,
      "num_input_tokens_seen": 7322392,
      "step": 12620
    },
    {
      "epoch": 1.880399165921954,
      "grad_norm": 15.0,
      "learning_rate": 4.7006255585344064e-05,
      "loss": 1.07,
      "num_input_tokens_seen": 7325208,
      "step": 12625
    },
    {
      "epoch": 1.8811438784629133,
      "grad_norm": 11.4375,
      "learning_rate": 4.702487339886804e-05,
      "loss": 1.3317,
      "num_input_tokens_seen": 7328120,
      "step": 12630
    },
    {
      "epoch": 1.8818885910038725,
      "grad_norm": 21.5,
      "learning_rate": 4.704349121239202e-05,
      "loss": 1.0526,
      "num_input_tokens_seen": 7331096,
      "step": 12635
    },
    {
      "epoch": 1.8826333035448317,
      "grad_norm": 13.9375,
      "learning_rate": 4.7062109025916e-05,
      "loss": 1.0678,
      "num_input_tokens_seen": 7333944,
      "step": 12640
    },
    {
      "epoch": 1.883378016085791,
      "grad_norm": 16.125,
      "learning_rate": 4.708072683943998e-05,
      "loss": 1.4477,
      "num_input_tokens_seen": 7336824,
      "step": 12645
    },
    {
      "epoch": 1.8841227286267501,
      "grad_norm": 8.5,
      "learning_rate": 4.7099344652963955e-05,
      "loss": 0.8903,
      "num_input_tokens_seen": 7339736,
      "step": 12650
    },
    {
      "epoch": 1.8848674411677093,
      "grad_norm": 9.0,
      "learning_rate": 4.7117962466487934e-05,
      "loss": 1.1249,
      "num_input_tokens_seen": 7342936,
      "step": 12655
    },
    {
      "epoch": 1.8856121537086685,
      "grad_norm": 7.125,
      "learning_rate": 4.713658028001192e-05,
      "loss": 1.069,
      "num_input_tokens_seen": 7345752,
      "step": 12660
    },
    {
      "epoch": 1.8863568662496277,
      "grad_norm": 5.8125,
      "learning_rate": 4.71551980935359e-05,
      "loss": 0.8193,
      "num_input_tokens_seen": 7348408,
      "step": 12665
    },
    {
      "epoch": 1.887101578790587,
      "grad_norm": 14.5,
      "learning_rate": 4.7173815907059875e-05,
      "loss": 0.929,
      "num_input_tokens_seen": 7351064,
      "step": 12670
    },
    {
      "epoch": 1.8878462913315461,
      "grad_norm": 17.125,
      "learning_rate": 4.719243372058385e-05,
      "loss": 1.3993,
      "num_input_tokens_seen": 7353624,
      "step": 12675
    },
    {
      "epoch": 1.8885910038725053,
      "grad_norm": 8.125,
      "learning_rate": 4.721105153410784e-05,
      "loss": 1.0203,
      "num_input_tokens_seen": 7356376,
      "step": 12680
    },
    {
      "epoch": 1.8893357164134645,
      "grad_norm": 12.0,
      "learning_rate": 4.7229669347631816e-05,
      "loss": 1.0574,
      "num_input_tokens_seen": 7359256,
      "step": 12685
    },
    {
      "epoch": 1.8900804289544237,
      "grad_norm": 13.4375,
      "learning_rate": 4.7248287161155795e-05,
      "loss": 1.3919,
      "num_input_tokens_seen": 7362392,
      "step": 12690
    },
    {
      "epoch": 1.890825141495383,
      "grad_norm": 9.0,
      "learning_rate": 4.726690497467978e-05,
      "loss": 0.9958,
      "num_input_tokens_seen": 7365176,
      "step": 12695
    },
    {
      "epoch": 1.891569854036342,
      "grad_norm": 20.875,
      "learning_rate": 4.728552278820376e-05,
      "loss": 1.3807,
      "num_input_tokens_seen": 7368056,
      "step": 12700
    },
    {
      "epoch": 1.8923145665773011,
      "grad_norm": 14.125,
      "learning_rate": 4.7304140601727736e-05,
      "loss": 1.1366,
      "num_input_tokens_seen": 7371064,
      "step": 12705
    },
    {
      "epoch": 1.8930592791182603,
      "grad_norm": 7.3125,
      "learning_rate": 4.7322758415251714e-05,
      "loss": 1.0218,
      "num_input_tokens_seen": 7373912,
      "step": 12710
    },
    {
      "epoch": 1.8938039916592195,
      "grad_norm": 4.71875,
      "learning_rate": 4.73413762287757e-05,
      "loss": 1.0702,
      "num_input_tokens_seen": 7376888,
      "step": 12715
    },
    {
      "epoch": 1.8945487042001787,
      "grad_norm": 6.96875,
      "learning_rate": 4.735999404229968e-05,
      "loss": 0.945,
      "num_input_tokens_seen": 7379736,
      "step": 12720
    },
    {
      "epoch": 1.895293416741138,
      "grad_norm": 9.125,
      "learning_rate": 4.7378611855823656e-05,
      "loss": 0.8728,
      "num_input_tokens_seen": 7382776,
      "step": 12725
    },
    {
      "epoch": 1.8960381292820971,
      "grad_norm": 10.0,
      "learning_rate": 4.7397229669347634e-05,
      "loss": 1.0191,
      "num_input_tokens_seen": 7385912,
      "step": 12730
    },
    {
      "epoch": 1.8967828418230563,
      "grad_norm": 8.625,
      "learning_rate": 4.741584748287161e-05,
      "loss": 1.0479,
      "num_input_tokens_seen": 7388728,
      "step": 12735
    },
    {
      "epoch": 1.8975275543640155,
      "grad_norm": 11.125,
      "learning_rate": 4.743446529639559e-05,
      "loss": 1.0521,
      "num_input_tokens_seen": 7391384,
      "step": 12740
    },
    {
      "epoch": 1.8982722669049745,
      "grad_norm": 10.125,
      "learning_rate": 4.745308310991957e-05,
      "loss": 1.1041,
      "num_input_tokens_seen": 7394328,
      "step": 12745
    },
    {
      "epoch": 1.8990169794459337,
      "grad_norm": 14.1875,
      "learning_rate": 4.7471700923443554e-05,
      "loss": 0.8959,
      "num_input_tokens_seen": 7397112,
      "step": 12750
    },
    {
      "epoch": 1.899761691986893,
      "grad_norm": 19.5,
      "learning_rate": 4.749031873696753e-05,
      "loss": 1.2438,
      "num_input_tokens_seen": 7400152,
      "step": 12755
    },
    {
      "epoch": 1.9005064045278521,
      "grad_norm": 12.875,
      "learning_rate": 4.750893655049151e-05,
      "loss": 1.2746,
      "num_input_tokens_seen": 7403064,
      "step": 12760
    },
    {
      "epoch": 1.9012511170688113,
      "grad_norm": 15.0,
      "learning_rate": 4.752755436401549e-05,
      "loss": 1.0224,
      "num_input_tokens_seen": 7405720,
      "step": 12765
    },
    {
      "epoch": 1.9019958296097705,
      "grad_norm": 7.53125,
      "learning_rate": 4.7546172177539474e-05,
      "loss": 0.9573,
      "num_input_tokens_seen": 7408536,
      "step": 12770
    },
    {
      "epoch": 1.9027405421507297,
      "grad_norm": 19.25,
      "learning_rate": 4.756478999106345e-05,
      "loss": 1.3226,
      "num_input_tokens_seen": 7411736,
      "step": 12775
    },
    {
      "epoch": 1.903485254691689,
      "grad_norm": 16.875,
      "learning_rate": 4.758340780458743e-05,
      "loss": 1.2892,
      "num_input_tokens_seen": 7414552,
      "step": 12780
    },
    {
      "epoch": 1.9042299672326481,
      "grad_norm": 5.15625,
      "learning_rate": 4.7602025618111415e-05,
      "loss": 1.1857,
      "num_input_tokens_seen": 7417368,
      "step": 12785
    },
    {
      "epoch": 1.9049746797736073,
      "grad_norm": 7.625,
      "learning_rate": 4.762064343163539e-05,
      "loss": 0.9877,
      "num_input_tokens_seen": 7420504,
      "step": 12790
    },
    {
      "epoch": 1.9057193923145666,
      "grad_norm": 14.0625,
      "learning_rate": 4.763926124515937e-05,
      "loss": 1.0501,
      "num_input_tokens_seen": 7423448,
      "step": 12795
    },
    {
      "epoch": 1.9064641048555258,
      "grad_norm": 12.0,
      "learning_rate": 4.765787905868335e-05,
      "loss": 1.2576,
      "num_input_tokens_seen": 7426328,
      "step": 12800
    },
    {
      "epoch": 1.907208817396485,
      "grad_norm": 12.1875,
      "learning_rate": 4.7676496872207335e-05,
      "loss": 1.6554,
      "num_input_tokens_seen": 7428824,
      "step": 12805
    },
    {
      "epoch": 1.9079535299374442,
      "grad_norm": 7.21875,
      "learning_rate": 4.769511468573131e-05,
      "loss": 1.0232,
      "num_input_tokens_seen": 7432184,
      "step": 12810
    },
    {
      "epoch": 1.9086982424784034,
      "grad_norm": 5.90625,
      "learning_rate": 4.7713732499255284e-05,
      "loss": 0.8809,
      "num_input_tokens_seen": 7435160,
      "step": 12815
    },
    {
      "epoch": 1.9094429550193626,
      "grad_norm": 11.9375,
      "learning_rate": 4.773235031277927e-05,
      "loss": 1.1673,
      "num_input_tokens_seen": 7437944,
      "step": 12820
    },
    {
      "epoch": 1.9101876675603218,
      "grad_norm": 18.25,
      "learning_rate": 4.775096812630325e-05,
      "loss": 1.236,
      "num_input_tokens_seen": 7440952,
      "step": 12825
    },
    {
      "epoch": 1.910932380101281,
      "grad_norm": 14.0625,
      "learning_rate": 4.7769585939827226e-05,
      "loss": 0.7807,
      "num_input_tokens_seen": 7444216,
      "step": 12830
    },
    {
      "epoch": 1.9116770926422402,
      "grad_norm": 12.375,
      "learning_rate": 4.7788203753351204e-05,
      "loss": 0.6974,
      "num_input_tokens_seen": 7447160,
      "step": 12835
    },
    {
      "epoch": 1.9124218051831994,
      "grad_norm": 9.3125,
      "learning_rate": 4.780682156687519e-05,
      "loss": 1.1669,
      "num_input_tokens_seen": 7450008,
      "step": 12840
    },
    {
      "epoch": 1.9131665177241586,
      "grad_norm": 15.8125,
      "learning_rate": 4.782543938039917e-05,
      "loss": 0.9607,
      "num_input_tokens_seen": 7453048,
      "step": 12845
    },
    {
      "epoch": 1.9139112302651178,
      "grad_norm": 20.75,
      "learning_rate": 4.7844057193923146e-05,
      "loss": 1.3554,
      "num_input_tokens_seen": 7456152,
      "step": 12850
    },
    {
      "epoch": 1.914655942806077,
      "grad_norm": 7.25,
      "learning_rate": 4.786267500744713e-05,
      "loss": 0.777,
      "num_input_tokens_seen": 7459288,
      "step": 12855
    },
    {
      "epoch": 1.9154006553470362,
      "grad_norm": 7.8125,
      "learning_rate": 4.788129282097111e-05,
      "loss": 1.1933,
      "num_input_tokens_seen": 7462168,
      "step": 12860
    },
    {
      "epoch": 1.9161453678879954,
      "grad_norm": 17.625,
      "learning_rate": 4.789991063449509e-05,
      "loss": 1.0283,
      "num_input_tokens_seen": 7464888,
      "step": 12865
    },
    {
      "epoch": 1.9168900804289544,
      "grad_norm": 6.84375,
      "learning_rate": 4.7918528448019065e-05,
      "loss": 1.032,
      "num_input_tokens_seen": 7468120,
      "step": 12870
    },
    {
      "epoch": 1.9176347929699136,
      "grad_norm": 6.28125,
      "learning_rate": 4.793714626154305e-05,
      "loss": 0.9571,
      "num_input_tokens_seen": 7471288,
      "step": 12875
    },
    {
      "epoch": 1.9183795055108728,
      "grad_norm": 7.3125,
      "learning_rate": 4.795576407506703e-05,
      "loss": 1.0542,
      "num_input_tokens_seen": 7474520,
      "step": 12880
    },
    {
      "epoch": 1.919124218051832,
      "grad_norm": 15.6875,
      "learning_rate": 4.797438188859101e-05,
      "loss": 1.1354,
      "num_input_tokens_seen": 7477240,
      "step": 12885
    },
    {
      "epoch": 1.9198689305927912,
      "grad_norm": 29.875,
      "learning_rate": 4.7992999702114985e-05,
      "loss": 1.2084,
      "num_input_tokens_seen": 7480184,
      "step": 12890
    },
    {
      "epoch": 1.9206136431337504,
      "grad_norm": 10.9375,
      "learning_rate": 4.801161751563897e-05,
      "loss": 1.0342,
      "num_input_tokens_seen": 7483128,
      "step": 12895
    },
    {
      "epoch": 1.9213583556747096,
      "grad_norm": 9.8125,
      "learning_rate": 4.803023532916295e-05,
      "loss": 1.0567,
      "num_input_tokens_seen": 7486264,
      "step": 12900
    },
    {
      "epoch": 1.9221030682156688,
      "grad_norm": 5.78125,
      "learning_rate": 4.804885314268692e-05,
      "loss": 1.4167,
      "num_input_tokens_seen": 7488984,
      "step": 12905
    },
    {
      "epoch": 1.922847780756628,
      "grad_norm": 16.25,
      "learning_rate": 4.8067470956210905e-05,
      "loss": 1.157,
      "num_input_tokens_seen": 7491576,
      "step": 12910
    },
    {
      "epoch": 1.923592493297587,
      "grad_norm": 5.6875,
      "learning_rate": 4.808608876973488e-05,
      "loss": 0.8588,
      "num_input_tokens_seen": 7494616,
      "step": 12915
    },
    {
      "epoch": 1.9243372058385462,
      "grad_norm": 24.125,
      "learning_rate": 4.810470658325886e-05,
      "loss": 1.3275,
      "num_input_tokens_seen": 7497144,
      "step": 12920
    },
    {
      "epoch": 1.9250819183795054,
      "grad_norm": 10.0,
      "learning_rate": 4.8123324396782846e-05,
      "loss": 0.9101,
      "num_input_tokens_seen": 7499896,
      "step": 12925
    },
    {
      "epoch": 1.9258266309204646,
      "grad_norm": 16.625,
      "learning_rate": 4.8141942210306824e-05,
      "loss": 0.9033,
      "num_input_tokens_seen": 7502680,
      "step": 12930
    },
    {
      "epoch": 1.9265713434614238,
      "grad_norm": 10.0,
      "learning_rate": 4.81605600238308e-05,
      "loss": 0.9043,
      "num_input_tokens_seen": 7505432,
      "step": 12935
    },
    {
      "epoch": 1.927316056002383,
      "grad_norm": 14.6875,
      "learning_rate": 4.817917783735478e-05,
      "loss": 1.0721,
      "num_input_tokens_seen": 7508376,
      "step": 12940
    },
    {
      "epoch": 1.9280607685433422,
      "grad_norm": 21.5,
      "learning_rate": 4.8197795650878766e-05,
      "loss": 1.3673,
      "num_input_tokens_seen": 7511384,
      "step": 12945
    },
    {
      "epoch": 1.9288054810843014,
      "grad_norm": 12.375,
      "learning_rate": 4.8216413464402744e-05,
      "loss": 1.2585,
      "num_input_tokens_seen": 7514360,
      "step": 12950
    },
    {
      "epoch": 1.9295501936252606,
      "grad_norm": 12.8125,
      "learning_rate": 4.823503127792672e-05,
      "loss": 0.9891,
      "num_input_tokens_seen": 7517176,
      "step": 12955
    },
    {
      "epoch": 1.9302949061662198,
      "grad_norm": 7.625,
      "learning_rate": 4.82536490914507e-05,
      "loss": 0.8973,
      "num_input_tokens_seen": 7520120,
      "step": 12960
    },
    {
      "epoch": 1.931039618707179,
      "grad_norm": 7.96875,
      "learning_rate": 4.8272266904974686e-05,
      "loss": 0.9687,
      "num_input_tokens_seen": 7523224,
      "step": 12965
    },
    {
      "epoch": 1.9317843312481382,
      "grad_norm": 6.75,
      "learning_rate": 4.8290884718498664e-05,
      "loss": 1.0107,
      "num_input_tokens_seen": 7525752,
      "step": 12970
    },
    {
      "epoch": 1.9325290437890974,
      "grad_norm": 11.3125,
      "learning_rate": 4.830950253202264e-05,
      "loss": 1.2807,
      "num_input_tokens_seen": 7528504,
      "step": 12975
    },
    {
      "epoch": 1.9332737563300566,
      "grad_norm": 13.375,
      "learning_rate": 4.832812034554662e-05,
      "loss": 1.4186,
      "num_input_tokens_seen": 7531416,
      "step": 12980
    },
    {
      "epoch": 1.9340184688710158,
      "grad_norm": 5.125,
      "learning_rate": 4.8346738159070605e-05,
      "loss": 1.2215,
      "num_input_tokens_seen": 7534200,
      "step": 12985
    },
    {
      "epoch": 1.934763181411975,
      "grad_norm": 6.90625,
      "learning_rate": 4.836535597259458e-05,
      "loss": 1.0541,
      "num_input_tokens_seen": 7537112,
      "step": 12990
    },
    {
      "epoch": 1.9355078939529342,
      "grad_norm": 10.3125,
      "learning_rate": 4.8383973786118555e-05,
      "loss": 1.1252,
      "num_input_tokens_seen": 7539704,
      "step": 12995
    },
    {
      "epoch": 1.9362526064938934,
      "grad_norm": 9.625,
      "learning_rate": 4.840259159964254e-05,
      "loss": 1.0944,
      "num_input_tokens_seen": 7542456,
      "step": 13000
    },
    {
      "epoch": 1.9369973190348526,
      "grad_norm": 11.5625,
      "learning_rate": 4.842120941316652e-05,
      "loss": 1.0385,
      "num_input_tokens_seen": 7545336,
      "step": 13005
    },
    {
      "epoch": 1.9377420315758118,
      "grad_norm": 4.71875,
      "learning_rate": 4.8439827226690496e-05,
      "loss": 1.0616,
      "num_input_tokens_seen": 7548248,
      "step": 13010
    },
    {
      "epoch": 1.938486744116771,
      "grad_norm": 7.28125,
      "learning_rate": 4.845844504021448e-05,
      "loss": 0.8651,
      "num_input_tokens_seen": 7551192,
      "step": 13015
    },
    {
      "epoch": 1.9392314566577302,
      "grad_norm": 7.875,
      "learning_rate": 4.847706285373846e-05,
      "loss": 0.8418,
      "num_input_tokens_seen": 7554328,
      "step": 13020
    },
    {
      "epoch": 1.9399761691986894,
      "grad_norm": 11.0,
      "learning_rate": 4.849568066726244e-05,
      "loss": 1.1895,
      "num_input_tokens_seen": 7557432,
      "step": 13025
    },
    {
      "epoch": 1.9407208817396486,
      "grad_norm": 10.4375,
      "learning_rate": 4.8514298480786416e-05,
      "loss": 1.2644,
      "num_input_tokens_seen": 7560504,
      "step": 13030
    },
    {
      "epoch": 1.9414655942806078,
      "grad_norm": 7.3125,
      "learning_rate": 4.85329162943104e-05,
      "loss": 0.902,
      "num_input_tokens_seen": 7563128,
      "step": 13035
    },
    {
      "epoch": 1.942210306821567,
      "grad_norm": 8.6875,
      "learning_rate": 4.855153410783438e-05,
      "loss": 1.3484,
      "num_input_tokens_seen": 7566232,
      "step": 13040
    },
    {
      "epoch": 1.942955019362526,
      "grad_norm": 15.0625,
      "learning_rate": 4.857015192135836e-05,
      "loss": 1.2825,
      "num_input_tokens_seen": 7569400,
      "step": 13045
    },
    {
      "epoch": 1.9436997319034852,
      "grad_norm": 11.875,
      "learning_rate": 4.8588769734882336e-05,
      "loss": 1.1836,
      "num_input_tokens_seen": 7572376,
      "step": 13050
    },
    {
      "epoch": 1.9444444444444444,
      "grad_norm": 11.0,
      "learning_rate": 4.860738754840632e-05,
      "loss": 1.0819,
      "num_input_tokens_seen": 7575160,
      "step": 13055
    },
    {
      "epoch": 1.9451891569854036,
      "grad_norm": 14.875,
      "learning_rate": 4.86260053619303e-05,
      "loss": 0.8462,
      "num_input_tokens_seen": 7578072,
      "step": 13060
    },
    {
      "epoch": 1.9459338695263628,
      "grad_norm": 9.125,
      "learning_rate": 4.864462317545428e-05,
      "loss": 1.0915,
      "num_input_tokens_seen": 7580696,
      "step": 13065
    },
    {
      "epoch": 1.946678582067322,
      "grad_norm": 8.9375,
      "learning_rate": 4.866324098897826e-05,
      "loss": 1.243,
      "num_input_tokens_seen": 7583640,
      "step": 13070
    },
    {
      "epoch": 1.9474232946082812,
      "grad_norm": 12.0,
      "learning_rate": 4.8681858802502234e-05,
      "loss": 0.9462,
      "num_input_tokens_seen": 7586616,
      "step": 13075
    },
    {
      "epoch": 1.9481680071492404,
      "grad_norm": 14.125,
      "learning_rate": 4.870047661602621e-05,
      "loss": 0.8535,
      "num_input_tokens_seen": 7589304,
      "step": 13080
    },
    {
      "epoch": 1.9489127196901996,
      "grad_norm": 18.875,
      "learning_rate": 4.87190944295502e-05,
      "loss": 1.2912,
      "num_input_tokens_seen": 7592280,
      "step": 13085
    },
    {
      "epoch": 1.9496574322311586,
      "grad_norm": 11.8125,
      "learning_rate": 4.8737712243074175e-05,
      "loss": 1.1646,
      "num_input_tokens_seen": 7595352,
      "step": 13090
    },
    {
      "epoch": 1.9504021447721178,
      "grad_norm": 6.15625,
      "learning_rate": 4.8756330056598153e-05,
      "loss": 0.9879,
      "num_input_tokens_seen": 7598040,
      "step": 13095
    },
    {
      "epoch": 1.951146857313077,
      "grad_norm": 10.875,
      "learning_rate": 4.877494787012213e-05,
      "loss": 0.8726,
      "num_input_tokens_seen": 7600760,
      "step": 13100
    },
    {
      "epoch": 1.9518915698540362,
      "grad_norm": 16.375,
      "learning_rate": 4.879356568364612e-05,
      "loss": 1.1914,
      "num_input_tokens_seen": 7603736,
      "step": 13105
    },
    {
      "epoch": 1.9526362823949954,
      "grad_norm": 10.9375,
      "learning_rate": 4.8812183497170095e-05,
      "loss": 0.9951,
      "num_input_tokens_seen": 7607000,
      "step": 13110
    },
    {
      "epoch": 1.9533809949359546,
      "grad_norm": 6.03125,
      "learning_rate": 4.883080131069407e-05,
      "loss": 1.3205,
      "num_input_tokens_seen": 7609752,
      "step": 13115
    },
    {
      "epoch": 1.9541257074769138,
      "grad_norm": 5.0,
      "learning_rate": 4.884941912421805e-05,
      "loss": 0.9779,
      "num_input_tokens_seen": 7612600,
      "step": 13120
    },
    {
      "epoch": 1.954870420017873,
      "grad_norm": 14.6875,
      "learning_rate": 4.8868036937742036e-05,
      "loss": 1.2102,
      "num_input_tokens_seen": 7615480,
      "step": 13125
    },
    {
      "epoch": 1.9556151325588322,
      "grad_norm": 13.8125,
      "learning_rate": 4.8886654751266015e-05,
      "loss": 0.8978,
      "num_input_tokens_seen": 7618488,
      "step": 13130
    },
    {
      "epoch": 1.9563598450997914,
      "grad_norm": 9.375,
      "learning_rate": 4.890527256478999e-05,
      "loss": 1.0422,
      "num_input_tokens_seen": 7621304,
      "step": 13135
    },
    {
      "epoch": 1.9571045576407506,
      "grad_norm": 7.34375,
      "learning_rate": 4.892389037831398e-05,
      "loss": 1.0225,
      "num_input_tokens_seen": 7624184,
      "step": 13140
    },
    {
      "epoch": 1.9578492701817098,
      "grad_norm": 11.4375,
      "learning_rate": 4.8942508191837956e-05,
      "loss": 1.2366,
      "num_input_tokens_seen": 7626936,
      "step": 13145
    },
    {
      "epoch": 1.958593982722669,
      "grad_norm": 5.5,
      "learning_rate": 4.8961126005361934e-05,
      "loss": 0.8843,
      "num_input_tokens_seen": 7629752,
      "step": 13150
    },
    {
      "epoch": 1.9593386952636282,
      "grad_norm": 14.625,
      "learning_rate": 4.897974381888591e-05,
      "loss": 0.958,
      "num_input_tokens_seen": 7632696,
      "step": 13155
    },
    {
      "epoch": 1.9600834078045875,
      "grad_norm": 15.4375,
      "learning_rate": 4.89983616324099e-05,
      "loss": 1.1106,
      "num_input_tokens_seen": 7635480,
      "step": 13160
    },
    {
      "epoch": 1.9608281203455467,
      "grad_norm": 18.5,
      "learning_rate": 4.901697944593387e-05,
      "loss": 0.914,
      "num_input_tokens_seen": 7638520,
      "step": 13165
    },
    {
      "epoch": 1.9615728328865059,
      "grad_norm": 17.875,
      "learning_rate": 4.903559725945785e-05,
      "loss": 1.0989,
      "num_input_tokens_seen": 7641464,
      "step": 13170
    },
    {
      "epoch": 1.962317545427465,
      "grad_norm": 6.9375,
      "learning_rate": 4.905421507298183e-05,
      "loss": 0.9779,
      "num_input_tokens_seen": 7644280,
      "step": 13175
    },
    {
      "epoch": 1.9630622579684243,
      "grad_norm": 5.125,
      "learning_rate": 4.907283288650581e-05,
      "loss": 1.0001,
      "num_input_tokens_seen": 7646968,
      "step": 13180
    },
    {
      "epoch": 1.9638069705093835,
      "grad_norm": 13.5,
      "learning_rate": 4.909145070002979e-05,
      "loss": 1.1526,
      "num_input_tokens_seen": 7649816,
      "step": 13185
    },
    {
      "epoch": 1.9645516830503427,
      "grad_norm": 12.5625,
      "learning_rate": 4.911006851355377e-05,
      "loss": 1.0587,
      "num_input_tokens_seen": 7653048,
      "step": 13190
    },
    {
      "epoch": 1.9652963955913019,
      "grad_norm": 15.1875,
      "learning_rate": 4.912868632707775e-05,
      "loss": 0.9145,
      "num_input_tokens_seen": 7655992,
      "step": 13195
    },
    {
      "epoch": 1.966041108132261,
      "grad_norm": 22.375,
      "learning_rate": 4.914730414060173e-05,
      "loss": 1.2473,
      "num_input_tokens_seen": 7658840,
      "step": 13200
    },
    {
      "epoch": 1.9667858206732203,
      "grad_norm": 13.375,
      "learning_rate": 4.916592195412571e-05,
      "loss": 1.1258,
      "num_input_tokens_seen": 7661912,
      "step": 13205
    },
    {
      "epoch": 1.9675305332141795,
      "grad_norm": 17.375,
      "learning_rate": 4.918453976764969e-05,
      "loss": 1.0191,
      "num_input_tokens_seen": 7664792,
      "step": 13210
    },
    {
      "epoch": 1.9682752457551387,
      "grad_norm": 7.40625,
      "learning_rate": 4.920315758117367e-05,
      "loss": 0.6086,
      "num_input_tokens_seen": 7667448,
      "step": 13215
    },
    {
      "epoch": 1.9690199582960977,
      "grad_norm": 12.1875,
      "learning_rate": 4.922177539469765e-05,
      "loss": 1.0879,
      "num_input_tokens_seen": 7670232,
      "step": 13220
    },
    {
      "epoch": 1.9697646708370569,
      "grad_norm": 12.4375,
      "learning_rate": 4.924039320822163e-05,
      "loss": 1.146,
      "num_input_tokens_seen": 7672984,
      "step": 13225
    },
    {
      "epoch": 1.970509383378016,
      "grad_norm": 11.625,
      "learning_rate": 4.925901102174561e-05,
      "loss": 0.9519,
      "num_input_tokens_seen": 7675704,
      "step": 13230
    },
    {
      "epoch": 1.9712540959189753,
      "grad_norm": 11.25,
      "learning_rate": 4.927762883526959e-05,
      "loss": 1.0419,
      "num_input_tokens_seen": 7679000,
      "step": 13235
    },
    {
      "epoch": 1.9719988084599345,
      "grad_norm": 7.65625,
      "learning_rate": 4.929624664879357e-05,
      "loss": 0.8323,
      "num_input_tokens_seen": 7681880,
      "step": 13240
    },
    {
      "epoch": 1.9727435210008937,
      "grad_norm": 10.3125,
      "learning_rate": 4.931486446231755e-05,
      "loss": 1.1329,
      "num_input_tokens_seen": 7684952,
      "step": 13245
    },
    {
      "epoch": 1.9734882335418529,
      "grad_norm": 9.875,
      "learning_rate": 4.9333482275841526e-05,
      "loss": 0.8103,
      "num_input_tokens_seen": 7687768,
      "step": 13250
    },
    {
      "epoch": 1.974232946082812,
      "grad_norm": 13.5,
      "learning_rate": 4.9352100089365504e-05,
      "loss": 1.1796,
      "num_input_tokens_seen": 7690680,
      "step": 13255
    },
    {
      "epoch": 1.974977658623771,
      "grad_norm": 17.75,
      "learning_rate": 4.937071790288948e-05,
      "loss": 1.0722,
      "num_input_tokens_seen": 7693560,
      "step": 13260
    },
    {
      "epoch": 1.9757223711647303,
      "grad_norm": 7.53125,
      "learning_rate": 4.938933571641347e-05,
      "loss": 0.9972,
      "num_input_tokens_seen": 7696536,
      "step": 13265
    },
    {
      "epoch": 1.9764670837056895,
      "grad_norm": 8.125,
      "learning_rate": 4.9407953529937446e-05,
      "loss": 1.2945,
      "num_input_tokens_seen": 7699224,
      "step": 13270
    },
    {
      "epoch": 1.9772117962466487,
      "grad_norm": 7.6875,
      "learning_rate": 4.9426571343461424e-05,
      "loss": 1.0666,
      "num_input_tokens_seen": 7702648,
      "step": 13275
    },
    {
      "epoch": 1.9779565087876079,
      "grad_norm": 6.90625,
      "learning_rate": 4.94451891569854e-05,
      "loss": 0.889,
      "num_input_tokens_seen": 7705528,
      "step": 13280
    },
    {
      "epoch": 1.978701221328567,
      "grad_norm": 9.1875,
      "learning_rate": 4.946380697050939e-05,
      "loss": 0.9434,
      "num_input_tokens_seen": 7708376,
      "step": 13285
    },
    {
      "epoch": 1.9794459338695263,
      "grad_norm": 12.25,
      "learning_rate": 4.9482424784033366e-05,
      "loss": 1.192,
      "num_input_tokens_seen": 7711256,
      "step": 13290
    },
    {
      "epoch": 1.9801906464104855,
      "grad_norm": 12.6875,
      "learning_rate": 4.9501042597557344e-05,
      "loss": 1.2967,
      "num_input_tokens_seen": 7714360,
      "step": 13295
    },
    {
      "epoch": 1.9809353589514447,
      "grad_norm": 11.875,
      "learning_rate": 4.951966041108133e-05,
      "loss": 1.3538,
      "num_input_tokens_seen": 7717016,
      "step": 13300
    },
    {
      "epoch": 1.9816800714924039,
      "grad_norm": 11.5,
      "learning_rate": 4.953827822460531e-05,
      "loss": 1.0497,
      "num_input_tokens_seen": 7720440,
      "step": 13305
    },
    {
      "epoch": 1.982424784033363,
      "grad_norm": 13.375,
      "learning_rate": 4.9556896038129285e-05,
      "loss": 1.2425,
      "num_input_tokens_seen": 7723672,
      "step": 13310
    },
    {
      "epoch": 1.9831694965743223,
      "grad_norm": 13.75,
      "learning_rate": 4.9575513851653263e-05,
      "loss": 1.2736,
      "num_input_tokens_seen": 7726360,
      "step": 13315
    },
    {
      "epoch": 1.9839142091152815,
      "grad_norm": 14.8125,
      "learning_rate": 4.959413166517725e-05,
      "loss": 0.9472,
      "num_input_tokens_seen": 7729016,
      "step": 13320
    },
    {
      "epoch": 1.9846589216562407,
      "grad_norm": 11.1875,
      "learning_rate": 4.961274947870123e-05,
      "loss": 0.9906,
      "num_input_tokens_seen": 7731864,
      "step": 13325
    },
    {
      "epoch": 1.9854036341972,
      "grad_norm": 26.25,
      "learning_rate": 4.9631367292225205e-05,
      "loss": 1.3823,
      "num_input_tokens_seen": 7734552,
      "step": 13330
    },
    {
      "epoch": 1.986148346738159,
      "grad_norm": 14.0625,
      "learning_rate": 4.964998510574918e-05,
      "loss": 0.9001,
      "num_input_tokens_seen": 7737528,
      "step": 13335
    },
    {
      "epoch": 1.9868930592791183,
      "grad_norm": 7.0625,
      "learning_rate": 4.966860291927316e-05,
      "loss": 1.1088,
      "num_input_tokens_seen": 7740568,
      "step": 13340
    },
    {
      "epoch": 1.9876377718200775,
      "grad_norm": 14.1875,
      "learning_rate": 4.968722073279714e-05,
      "loss": 0.8542,
      "num_input_tokens_seen": 7743704,
      "step": 13345
    },
    {
      "epoch": 1.9883824843610367,
      "grad_norm": 11.0625,
      "learning_rate": 4.970583854632112e-05,
      "loss": 0.9944,
      "num_input_tokens_seen": 7746648,
      "step": 13350
    },
    {
      "epoch": 1.989127196901996,
      "grad_norm": 8.75,
      "learning_rate": 4.97244563598451e-05,
      "loss": 1.2554,
      "num_input_tokens_seen": 7749624,
      "step": 13355
    },
    {
      "epoch": 1.9898719094429551,
      "grad_norm": 13.625,
      "learning_rate": 4.974307417336908e-05,
      "loss": 1.6792,
      "num_input_tokens_seen": 7752184,
      "step": 13360
    },
    {
      "epoch": 1.9906166219839143,
      "grad_norm": 11.3125,
      "learning_rate": 4.976169198689306e-05,
      "loss": 1.0361,
      "num_input_tokens_seen": 7754936,
      "step": 13365
    },
    {
      "epoch": 1.9913613345248735,
      "grad_norm": 7.65625,
      "learning_rate": 4.978030980041704e-05,
      "loss": 0.8824,
      "num_input_tokens_seen": 7757688,
      "step": 13370
    },
    {
      "epoch": 1.9921060470658327,
      "grad_norm": 21.375,
      "learning_rate": 4.979892761394102e-05,
      "loss": 1.1252,
      "num_input_tokens_seen": 7760472,
      "step": 13375
    },
    {
      "epoch": 1.992850759606792,
      "grad_norm": 10.5625,
      "learning_rate": 4.9817545427465e-05,
      "loss": 1.2022,
      "num_input_tokens_seen": 7763576,
      "step": 13380
    },
    {
      "epoch": 1.9935954721477511,
      "grad_norm": 8.8125,
      "learning_rate": 4.983616324098898e-05,
      "loss": 1.1339,
      "num_input_tokens_seen": 7766488,
      "step": 13385
    },
    {
      "epoch": 1.99434018468871,
      "grad_norm": 6.28125,
      "learning_rate": 4.9854781054512964e-05,
      "loss": 1.0005,
      "num_input_tokens_seen": 7769432,
      "step": 13390
    },
    {
      "epoch": 1.9950848972296693,
      "grad_norm": 11.0,
      "learning_rate": 4.987339886803694e-05,
      "loss": 1.1674,
      "num_input_tokens_seen": 7771992,
      "step": 13395
    },
    {
      "epoch": 1.9958296097706285,
      "grad_norm": 7.5625,
      "learning_rate": 4.989201668156092e-05,
      "loss": 0.9932,
      "num_input_tokens_seen": 7774808,
      "step": 13400
    },
    {
      "epoch": 1.9965743223115877,
      "grad_norm": 9.0625,
      "learning_rate": 4.99106344950849e-05,
      "loss": 1.0292,
      "num_input_tokens_seen": 7777624,
      "step": 13405
    },
    {
      "epoch": 1.997319034852547,
      "grad_norm": 13.75,
      "learning_rate": 4.9929252308608884e-05,
      "loss": 1.1089,
      "num_input_tokens_seen": 7780696,
      "step": 13410
    },
    {
      "epoch": 1.9980637473935061,
      "grad_norm": 19.0,
      "learning_rate": 4.994787012213286e-05,
      "loss": 0.9749,
      "num_input_tokens_seen": 7783736,
      "step": 13415
    },
    {
      "epoch": 1.9988084599344653,
      "grad_norm": 12.5625,
      "learning_rate": 4.996648793565684e-05,
      "loss": 1.0491,
      "num_input_tokens_seen": 7786520,
      "step": 13420
    },
    {
      "epoch": 1.9995531724754245,
      "grad_norm": 5.34375,
      "learning_rate": 4.998510574918082e-05,
      "loss": 1.2032,
      "num_input_tokens_seen": 7789368,
      "step": 13425
    },
    {
      "epoch": 2.0,
      "eval_loss": 1.101166844367981,
      "eval_runtime": 43.7693,
      "eval_samples_per_second": 68.176,
      "eval_steps_per_second": 17.044,
      "num_input_tokens_seen": 7790784,
      "step": 13428
    },
    {
      "epoch": 2.0002978850163835,
      "grad_norm": 6.6875,
      "learning_rate": 4.999999999155301e-05,
      "loss": 1.3319,
      "num_input_tokens_seen": 7792000,
      "step": 13430
    },
    {
      "epoch": 2.0010425975573427,
      "grad_norm": 5.875,
      "learning_rate": 4.9999999695908296e-05,
      "loss": 1.3345,
      "num_input_tokens_seen": 7794848,
      "step": 13435
    },
    {
      "epoch": 2.001787310098302,
      "grad_norm": 13.0625,
      "learning_rate": 4.9999998977913995e-05,
      "loss": 0.89,
      "num_input_tokens_seen": 7797760,
      "step": 13440
    },
    {
      "epoch": 2.002532022639261,
      "grad_norm": 6.65625,
      "learning_rate": 4.999999783757012e-05,
      "loss": 0.9123,
      "num_input_tokens_seen": 7800896,
      "step": 13445
    },
    {
      "epoch": 2.0032767351802203,
      "grad_norm": 15.0625,
      "learning_rate": 4.999999627487669e-05,
      "loss": 1.1466,
      "num_input_tokens_seen": 7803936,
      "step": 13450
    },
    {
      "epoch": 2.0040214477211795,
      "grad_norm": 17.375,
      "learning_rate": 4.999999428983374e-05,
      "loss": 1.178,
      "num_input_tokens_seen": 7806784,
      "step": 13455
    },
    {
      "epoch": 2.0047661602621387,
      "grad_norm": 5.59375,
      "learning_rate": 4.999999188244129e-05,
      "loss": 0.8866,
      "num_input_tokens_seen": 7809664,
      "step": 13460
    },
    {
      "epoch": 2.005510872803098,
      "grad_norm": 13.375,
      "learning_rate": 4.999998905269938e-05,
      "loss": 1.1343,
      "num_input_tokens_seen": 7812448,
      "step": 13465
    },
    {
      "epoch": 2.006255585344057,
      "grad_norm": 8.3125,
      "learning_rate": 4.9999985800608076e-05,
      "loss": 0.8758,
      "num_input_tokens_seen": 7815456,
      "step": 13470
    },
    {
      "epoch": 2.0070002978850163,
      "grad_norm": 17.25,
      "learning_rate": 4.9999982126167414e-05,
      "loss": 1.3629,
      "num_input_tokens_seen": 7818432,
      "step": 13475
    },
    {
      "epoch": 2.0077450104259755,
      "grad_norm": 13.3125,
      "learning_rate": 4.9999978029377456e-05,
      "loss": 1.1093,
      "num_input_tokens_seen": 7821152,
      "step": 13480
    },
    {
      "epoch": 2.0084897229669347,
      "grad_norm": 12.0625,
      "learning_rate": 4.9999973510238284e-05,
      "loss": 1.1267,
      "num_input_tokens_seen": 7824288,
      "step": 13485
    },
    {
      "epoch": 2.009234435507894,
      "grad_norm": 13.3125,
      "learning_rate": 4.999996856874997e-05,
      "loss": 1.0271,
      "num_input_tokens_seen": 7827136,
      "step": 13490
    },
    {
      "epoch": 2.009979148048853,
      "grad_norm": 7.59375,
      "learning_rate": 4.99999632049126e-05,
      "loss": 1.2617,
      "num_input_tokens_seen": 7829984,
      "step": 13495
    },
    {
      "epoch": 2.0107238605898123,
      "grad_norm": 7.375,
      "learning_rate": 4.999995741872625e-05,
      "loss": 0.9157,
      "num_input_tokens_seen": 7832928,
      "step": 13500
    },
    {
      "epoch": 2.0114685731307715,
      "grad_norm": 11.75,
      "learning_rate": 4.999995121019103e-05,
      "loss": 0.9851,
      "num_input_tokens_seen": 7835872,
      "step": 13505
    },
    {
      "epoch": 2.0122132856717307,
      "grad_norm": 11.875,
      "learning_rate": 4.999994457930705e-05,
      "loss": 1.4314,
      "num_input_tokens_seen": 7838784,
      "step": 13510
    },
    {
      "epoch": 2.01295799821269,
      "grad_norm": 11.625,
      "learning_rate": 4.999993752607441e-05,
      "loss": 1.0095,
      "num_input_tokens_seen": 7841792,
      "step": 13515
    },
    {
      "epoch": 2.013702710753649,
      "grad_norm": 11.75,
      "learning_rate": 4.999993005049324e-05,
      "loss": 1.1439,
      "num_input_tokens_seen": 7844768,
      "step": 13520
    },
    {
      "epoch": 2.0144474232946084,
      "grad_norm": 10.9375,
      "learning_rate": 4.999992215256365e-05,
      "loss": 1.083,
      "num_input_tokens_seen": 7848224,
      "step": 13525
    },
    {
      "epoch": 2.0151921358355676,
      "grad_norm": 7.75,
      "learning_rate": 4.9999913832285796e-05,
      "loss": 0.8196,
      "num_input_tokens_seen": 7851072,
      "step": 13530
    },
    {
      "epoch": 2.0159368483765268,
      "grad_norm": 11.9375,
      "learning_rate": 4.99999050896598e-05,
      "loss": 0.8168,
      "num_input_tokens_seen": 7853984,
      "step": 13535
    },
    {
      "epoch": 2.016681560917486,
      "grad_norm": 17.625,
      "learning_rate": 4.999989592468582e-05,
      "loss": 1.0963,
      "num_input_tokens_seen": 7856640,
      "step": 13540
    },
    {
      "epoch": 2.017426273458445,
      "grad_norm": 7.15625,
      "learning_rate": 4.9999886337364004e-05,
      "loss": 0.9139,
      "num_input_tokens_seen": 7859392,
      "step": 13545
    },
    {
      "epoch": 2.0181709859994044,
      "grad_norm": 5.375,
      "learning_rate": 4.999987632769452e-05,
      "loss": 0.7641,
      "num_input_tokens_seen": 7862016,
      "step": 13550
    },
    {
      "epoch": 2.0189156985403636,
      "grad_norm": 10.0,
      "learning_rate": 4.9999865895677534e-05,
      "loss": 0.7673,
      "num_input_tokens_seen": 7864640,
      "step": 13555
    },
    {
      "epoch": 2.0196604110813228,
      "grad_norm": 8.0625,
      "learning_rate": 4.999985504131322e-05,
      "loss": 0.8318,
      "num_input_tokens_seen": 7867488,
      "step": 13560
    },
    {
      "epoch": 2.020405123622282,
      "grad_norm": 7.46875,
      "learning_rate": 4.999984376460176e-05,
      "loss": 1.1349,
      "num_input_tokens_seen": 7870496,
      "step": 13565
    },
    {
      "epoch": 2.021149836163241,
      "grad_norm": 13.125,
      "learning_rate": 4.999983206554335e-05,
      "loss": 1.145,
      "num_input_tokens_seen": 7873344,
      "step": 13570
    },
    {
      "epoch": 2.0218945487042004,
      "grad_norm": 15.0,
      "learning_rate": 4.9999819944138194e-05,
      "loss": 1.4455,
      "num_input_tokens_seen": 7876512,
      "step": 13575
    },
    {
      "epoch": 2.0226392612451596,
      "grad_norm": 15.4375,
      "learning_rate": 4.999980740038648e-05,
      "loss": 1.2001,
      "num_input_tokens_seen": 7879360,
      "step": 13580
    },
    {
      "epoch": 2.0233839737861183,
      "grad_norm": 15.375,
      "learning_rate": 4.9999794434288434e-05,
      "loss": 0.9122,
      "num_input_tokens_seen": 7881792,
      "step": 13585
    },
    {
      "epoch": 2.0241286863270775,
      "grad_norm": 12.25,
      "learning_rate": 4.9999781045844266e-05,
      "loss": 1.0415,
      "num_input_tokens_seen": 7884960,
      "step": 13590
    },
    {
      "epoch": 2.0248733988680367,
      "grad_norm": 15.125,
      "learning_rate": 4.999976723505421e-05,
      "loss": 1.3175,
      "num_input_tokens_seen": 7887776,
      "step": 13595
    },
    {
      "epoch": 2.025618111408996,
      "grad_norm": 9.125,
      "learning_rate": 4.999975300191849e-05,
      "loss": 0.9942,
      "num_input_tokens_seen": 7890880,
      "step": 13600
    },
    {
      "epoch": 2.026362823949955,
      "grad_norm": 6.65625,
      "learning_rate": 4.9999738346437355e-05,
      "loss": 1.0929,
      "num_input_tokens_seen": 7894336,
      "step": 13605
    },
    {
      "epoch": 2.0271075364909144,
      "grad_norm": 14.25,
      "learning_rate": 4.9999723268611046e-05,
      "loss": 1.1415,
      "num_input_tokens_seen": 7897664,
      "step": 13610
    },
    {
      "epoch": 2.0278522490318736,
      "grad_norm": 11.875,
      "learning_rate": 4.9999707768439824e-05,
      "loss": 1.1008,
      "num_input_tokens_seen": 7900544,
      "step": 13615
    },
    {
      "epoch": 2.0285969615728328,
      "grad_norm": 8.6875,
      "learning_rate": 4.999969184592395e-05,
      "loss": 1.4377,
      "num_input_tokens_seen": 7903392,
      "step": 13620
    },
    {
      "epoch": 2.029341674113792,
      "grad_norm": 9.125,
      "learning_rate": 4.999967550106368e-05,
      "loss": 0.7423,
      "num_input_tokens_seen": 7906336,
      "step": 13625
    },
    {
      "epoch": 2.030086386654751,
      "grad_norm": 10.75,
      "learning_rate": 4.999965873385931e-05,
      "loss": 1.0707,
      "num_input_tokens_seen": 7909088,
      "step": 13630
    },
    {
      "epoch": 2.0308310991957104,
      "grad_norm": 13.375,
      "learning_rate": 4.999964154431112e-05,
      "loss": 1.0865,
      "num_input_tokens_seen": 7912000,
      "step": 13635
    },
    {
      "epoch": 2.0315758117366696,
      "grad_norm": 10.375,
      "learning_rate": 4.999962393241938e-05,
      "loss": 1.4291,
      "num_input_tokens_seen": 7915072,
      "step": 13640
    },
    {
      "epoch": 2.0323205242776288,
      "grad_norm": 10.8125,
      "learning_rate": 4.999960589818441e-05,
      "loss": 1.2963,
      "num_input_tokens_seen": 7917952,
      "step": 13645
    },
    {
      "epoch": 2.033065236818588,
      "grad_norm": 8.4375,
      "learning_rate": 4.999958744160651e-05,
      "loss": 1.1897,
      "num_input_tokens_seen": 7920704,
      "step": 13650
    },
    {
      "epoch": 2.033809949359547,
      "grad_norm": 11.8125,
      "learning_rate": 4.9999568562685986e-05,
      "loss": 0.9579,
      "num_input_tokens_seen": 7923392,
      "step": 13655
    },
    {
      "epoch": 2.0345546619005064,
      "grad_norm": 19.5,
      "learning_rate": 4.999954926142316e-05,
      "loss": 1.4009,
      "num_input_tokens_seen": 7926112,
      "step": 13660
    },
    {
      "epoch": 2.0352993744414656,
      "grad_norm": 16.75,
      "learning_rate": 4.999952953781836e-05,
      "loss": 1.1753,
      "num_input_tokens_seen": 7929472,
      "step": 13665
    },
    {
      "epoch": 2.036044086982425,
      "grad_norm": 19.5,
      "learning_rate": 4.9999509391871905e-05,
      "loss": 1.0989,
      "num_input_tokens_seen": 7932224,
      "step": 13670
    },
    {
      "epoch": 2.036788799523384,
      "grad_norm": 8.0625,
      "learning_rate": 4.999948882358416e-05,
      "loss": 1.0973,
      "num_input_tokens_seen": 7934944,
      "step": 13675
    },
    {
      "epoch": 2.037533512064343,
      "grad_norm": 7.96875,
      "learning_rate": 4.9999467832955454e-05,
      "loss": 1.028,
      "num_input_tokens_seen": 7937920,
      "step": 13680
    },
    {
      "epoch": 2.0382782246053024,
      "grad_norm": 17.625,
      "learning_rate": 4.999944641998615e-05,
      "loss": 1.1869,
      "num_input_tokens_seen": 7940832,
      "step": 13685
    },
    {
      "epoch": 2.0390229371462616,
      "grad_norm": 8.0625,
      "learning_rate": 4.99994245846766e-05,
      "loss": 0.9802,
      "num_input_tokens_seen": 7944192,
      "step": 13690
    },
    {
      "epoch": 2.039767649687221,
      "grad_norm": 14.0625,
      "learning_rate": 4.999940232702719e-05,
      "loss": 1.4084,
      "num_input_tokens_seen": 7946944,
      "step": 13695
    },
    {
      "epoch": 2.04051236222818,
      "grad_norm": 6.90625,
      "learning_rate": 4.999937964703828e-05,
      "loss": 1.3644,
      "num_input_tokens_seen": 7949920,
      "step": 13700
    },
    {
      "epoch": 2.041257074769139,
      "grad_norm": 7.375,
      "learning_rate": 4.999935654471026e-05,
      "loss": 1.1059,
      "num_input_tokens_seen": 7952736,
      "step": 13705
    },
    {
      "epoch": 2.0420017873100984,
      "grad_norm": 13.1875,
      "learning_rate": 4.999933302004352e-05,
      "loss": 0.9744,
      "num_input_tokens_seen": 7955392,
      "step": 13710
    },
    {
      "epoch": 2.0427464998510576,
      "grad_norm": 8.3125,
      "learning_rate": 4.999930907303846e-05,
      "loss": 0.8443,
      "num_input_tokens_seen": 7958080,
      "step": 13715
    },
    {
      "epoch": 2.043491212392017,
      "grad_norm": 10.8125,
      "learning_rate": 4.9999284703695474e-05,
      "loss": 1.1493,
      "num_input_tokens_seen": 7960960,
      "step": 13720
    },
    {
      "epoch": 2.044235924932976,
      "grad_norm": 12.875,
      "learning_rate": 4.9999259912014986e-05,
      "loss": 1.0714,
      "num_input_tokens_seen": 7963744,
      "step": 13725
    },
    {
      "epoch": 2.044980637473935,
      "grad_norm": 9.0625,
      "learning_rate": 4.999923469799741e-05,
      "loss": 1.2932,
      "num_input_tokens_seen": 7966336,
      "step": 13730
    },
    {
      "epoch": 2.0457253500148944,
      "grad_norm": 6.96875,
      "learning_rate": 4.9999209061643174e-05,
      "loss": 1.2566,
      "num_input_tokens_seen": 7969088,
      "step": 13735
    },
    {
      "epoch": 2.0464700625558536,
      "grad_norm": 19.625,
      "learning_rate": 4.99991830029527e-05,
      "loss": 1.0675,
      "num_input_tokens_seen": 7972160,
      "step": 13740
    },
    {
      "epoch": 2.047214775096813,
      "grad_norm": 8.8125,
      "learning_rate": 4.999915652192645e-05,
      "loss": 0.62,
      "num_input_tokens_seen": 7975072,
      "step": 13745
    },
    {
      "epoch": 2.047959487637772,
      "grad_norm": 8.8125,
      "learning_rate": 4.9999129618564844e-05,
      "loss": 1.0199,
      "num_input_tokens_seen": 7977792,
      "step": 13750
    },
    {
      "epoch": 2.0487042001787312,
      "grad_norm": 11.375,
      "learning_rate": 4.999910229286836e-05,
      "loss": 1.2209,
      "num_input_tokens_seen": 7980672,
      "step": 13755
    },
    {
      "epoch": 2.04944891271969,
      "grad_norm": 10.1875,
      "learning_rate": 4.999907454483745e-05,
      "loss": 1.1191,
      "num_input_tokens_seen": 7983456,
      "step": 13760
    },
    {
      "epoch": 2.050193625260649,
      "grad_norm": 7.6875,
      "learning_rate": 4.999904637447258e-05,
      "loss": 1.1345,
      "num_input_tokens_seen": 7986496,
      "step": 13765
    },
    {
      "epoch": 2.0509383378016084,
      "grad_norm": 11.6875,
      "learning_rate": 4.9999017781774236e-05,
      "loss": 0.9551,
      "num_input_tokens_seen": 7989504,
      "step": 13770
    },
    {
      "epoch": 2.0516830503425676,
      "grad_norm": 15.6875,
      "learning_rate": 4.999898876674289e-05,
      "loss": 1.1992,
      "num_input_tokens_seen": 7992064,
      "step": 13775
    },
    {
      "epoch": 2.052427762883527,
      "grad_norm": 7.96875,
      "learning_rate": 4.9998959329379036e-05,
      "loss": 1.0411,
      "num_input_tokens_seen": 7994592,
      "step": 13780
    },
    {
      "epoch": 2.053172475424486,
      "grad_norm": 6.03125,
      "learning_rate": 4.999892946968318e-05,
      "loss": 0.9424,
      "num_input_tokens_seen": 7997664,
      "step": 13785
    },
    {
      "epoch": 2.053917187965445,
      "grad_norm": 9.125,
      "learning_rate": 4.999889918765581e-05,
      "loss": 1.1472,
      "num_input_tokens_seen": 8000736,
      "step": 13790
    },
    {
      "epoch": 2.0546619005064044,
      "grad_norm": 4.84375,
      "learning_rate": 4.999886848329744e-05,
      "loss": 1.0974,
      "num_input_tokens_seen": 8003488,
      "step": 13795
    },
    {
      "epoch": 2.0554066130473636,
      "grad_norm": 6.46875,
      "learning_rate": 4.999883735660861e-05,
      "loss": 1.0698,
      "num_input_tokens_seen": 8006400,
      "step": 13800
    },
    {
      "epoch": 2.056151325588323,
      "grad_norm": 6.21875,
      "learning_rate": 4.999880580758982e-05,
      "loss": 0.7589,
      "num_input_tokens_seen": 8009312,
      "step": 13805
    },
    {
      "epoch": 2.056896038129282,
      "grad_norm": 12.75,
      "learning_rate": 4.999877383624162e-05,
      "loss": 0.888,
      "num_input_tokens_seen": 8012128,
      "step": 13810
    },
    {
      "epoch": 2.057640750670241,
      "grad_norm": 17.0,
      "learning_rate": 4.9998741442564535e-05,
      "loss": 1.1194,
      "num_input_tokens_seen": 8014944,
      "step": 13815
    },
    {
      "epoch": 2.0583854632112004,
      "grad_norm": 7.21875,
      "learning_rate": 4.999870862655913e-05,
      "loss": 0.8995,
      "num_input_tokens_seen": 8017568,
      "step": 13820
    },
    {
      "epoch": 2.0591301757521596,
      "grad_norm": 13.9375,
      "learning_rate": 4.999867538822595e-05,
      "loss": 1.3479,
      "num_input_tokens_seen": 8020672,
      "step": 13825
    },
    {
      "epoch": 2.059874888293119,
      "grad_norm": 18.25,
      "learning_rate": 4.999864172756554e-05,
      "loss": 1.1089,
      "num_input_tokens_seen": 8023520,
      "step": 13830
    },
    {
      "epoch": 2.060619600834078,
      "grad_norm": 10.75,
      "learning_rate": 4.9998607644578505e-05,
      "loss": 0.8861,
      "num_input_tokens_seen": 8026528,
      "step": 13835
    },
    {
      "epoch": 2.0613643133750372,
      "grad_norm": 9.0625,
      "learning_rate": 4.9998573139265395e-05,
      "loss": 1.1938,
      "num_input_tokens_seen": 8029440,
      "step": 13840
    },
    {
      "epoch": 2.0621090259159964,
      "grad_norm": 5.21875,
      "learning_rate": 4.99985382116268e-05,
      "loss": 1.1439,
      "num_input_tokens_seen": 8032160,
      "step": 13845
    },
    {
      "epoch": 2.0628537384569556,
      "grad_norm": 8.875,
      "learning_rate": 4.999850286166331e-05,
      "loss": 1.039,
      "num_input_tokens_seen": 8034912,
      "step": 13850
    },
    {
      "epoch": 2.063598450997915,
      "grad_norm": 6.59375,
      "learning_rate": 4.999846708937552e-05,
      "loss": 1.1181,
      "num_input_tokens_seen": 8037920,
      "step": 13855
    },
    {
      "epoch": 2.064343163538874,
      "grad_norm": 11.5625,
      "learning_rate": 4.9998430894764034e-05,
      "loss": 1.4127,
      "num_input_tokens_seen": 8040800,
      "step": 13860
    },
    {
      "epoch": 2.0650878760798332,
      "grad_norm": 10.125,
      "learning_rate": 4.9998394277829466e-05,
      "loss": 1.2383,
      "num_input_tokens_seen": 8044000,
      "step": 13865
    },
    {
      "epoch": 2.0658325886207924,
      "grad_norm": 10.3125,
      "learning_rate": 4.9998357238572435e-05,
      "loss": 0.9677,
      "num_input_tokens_seen": 8046784,
      "step": 13870
    },
    {
      "epoch": 2.0665773011617516,
      "grad_norm": 11.375,
      "learning_rate": 4.9998319776993566e-05,
      "loss": 1.3271,
      "num_input_tokens_seen": 8049376,
      "step": 13875
    },
    {
      "epoch": 2.067322013702711,
      "grad_norm": 18.25,
      "learning_rate": 4.999828189309349e-05,
      "loss": 1.2142,
      "num_input_tokens_seen": 8052160,
      "step": 13880
    },
    {
      "epoch": 2.06806672624367,
      "grad_norm": 24.75,
      "learning_rate": 4.999824358687285e-05,
      "loss": 0.9895,
      "num_input_tokens_seen": 8055072,
      "step": 13885
    },
    {
      "epoch": 2.0688114387846293,
      "grad_norm": 7.125,
      "learning_rate": 4.9998204858332295e-05,
      "loss": 1.0674,
      "num_input_tokens_seen": 8057984,
      "step": 13890
    },
    {
      "epoch": 2.0695561513255885,
      "grad_norm": 8.9375,
      "learning_rate": 4.999816570747247e-05,
      "loss": 1.0928,
      "num_input_tokens_seen": 8060640,
      "step": 13895
    },
    {
      "epoch": 2.0703008638665477,
      "grad_norm": 12.5,
      "learning_rate": 4.999812613429404e-05,
      "loss": 1.2101,
      "num_input_tokens_seen": 8063648,
      "step": 13900
    },
    {
      "epoch": 2.071045576407507,
      "grad_norm": 16.0,
      "learning_rate": 4.9998086138797685e-05,
      "loss": 1.5044,
      "num_input_tokens_seen": 8066624,
      "step": 13905
    },
    {
      "epoch": 2.071790288948466,
      "grad_norm": 29.125,
      "learning_rate": 4.9998045720984065e-05,
      "loss": 1.0977,
      "num_input_tokens_seen": 8069664,
      "step": 13910
    },
    {
      "epoch": 2.0725350014894253,
      "grad_norm": 9.875,
      "learning_rate": 4.999800488085388e-05,
      "loss": 1.2855,
      "num_input_tokens_seen": 8072672,
      "step": 13915
    },
    {
      "epoch": 2.0732797140303845,
      "grad_norm": 13.0625,
      "learning_rate": 4.9997963618407794e-05,
      "loss": 1.1065,
      "num_input_tokens_seen": 8075744,
      "step": 13920
    },
    {
      "epoch": 2.0740244265713437,
      "grad_norm": 13.75,
      "learning_rate": 4.999792193364653e-05,
      "loss": 1.1883,
      "num_input_tokens_seen": 8079040,
      "step": 13925
    },
    {
      "epoch": 2.074769139112303,
      "grad_norm": 10.1875,
      "learning_rate": 4.999787982657077e-05,
      "loss": 0.8987,
      "num_input_tokens_seen": 8081984,
      "step": 13930
    },
    {
      "epoch": 2.0755138516532616,
      "grad_norm": 10.3125,
      "learning_rate": 4.999783729718125e-05,
      "loss": 1.0379,
      "num_input_tokens_seen": 8084864,
      "step": 13935
    },
    {
      "epoch": 2.076258564194221,
      "grad_norm": 15.125,
      "learning_rate": 4.999779434547867e-05,
      "loss": 1.3043,
      "num_input_tokens_seen": 8087776,
      "step": 13940
    },
    {
      "epoch": 2.07700327673518,
      "grad_norm": 10.6875,
      "learning_rate": 4.999775097146376e-05,
      "loss": 0.9869,
      "num_input_tokens_seen": 8090560,
      "step": 13945
    },
    {
      "epoch": 2.0777479892761392,
      "grad_norm": 8.0,
      "learning_rate": 4.999770717513726e-05,
      "loss": 0.8028,
      "num_input_tokens_seen": 8093408,
      "step": 13950
    },
    {
      "epoch": 2.0784927018170984,
      "grad_norm": 6.53125,
      "learning_rate": 4.99976629564999e-05,
      "loss": 1.3579,
      "num_input_tokens_seen": 8096320,
      "step": 13955
    },
    {
      "epoch": 2.0792374143580576,
      "grad_norm": 18.125,
      "learning_rate": 4.999761831555243e-05,
      "loss": 1.2219,
      "num_input_tokens_seen": 8099136,
      "step": 13960
    },
    {
      "epoch": 2.079982126899017,
      "grad_norm": 11.0625,
      "learning_rate": 4.9997573252295604e-05,
      "loss": 1.0434,
      "num_input_tokens_seen": 8102080,
      "step": 13965
    },
    {
      "epoch": 2.080726839439976,
      "grad_norm": 11.1875,
      "learning_rate": 4.999752776673018e-05,
      "loss": 1.2418,
      "num_input_tokens_seen": 8104800,
      "step": 13970
    },
    {
      "epoch": 2.0814715519809353,
      "grad_norm": 11.8125,
      "learning_rate": 4.999748185885694e-05,
      "loss": 1.3723,
      "num_input_tokens_seen": 8108064,
      "step": 13975
    },
    {
      "epoch": 2.0822162645218945,
      "grad_norm": 15.3125,
      "learning_rate": 4.999743552867665e-05,
      "loss": 0.96,
      "num_input_tokens_seen": 8110976,
      "step": 13980
    },
    {
      "epoch": 2.0829609770628537,
      "grad_norm": 5.25,
      "learning_rate": 4.999738877619009e-05,
      "loss": 0.8989,
      "num_input_tokens_seen": 8113824,
      "step": 13985
    },
    {
      "epoch": 2.083705689603813,
      "grad_norm": 10.0625,
      "learning_rate": 4.999734160139805e-05,
      "loss": 1.2811,
      "num_input_tokens_seen": 8116608,
      "step": 13990
    },
    {
      "epoch": 2.084450402144772,
      "grad_norm": 7.65625,
      "learning_rate": 4.999729400430133e-05,
      "loss": 1.0885,
      "num_input_tokens_seen": 8119520,
      "step": 13995
    },
    {
      "epoch": 2.0851951146857313,
      "grad_norm": 8.4375,
      "learning_rate": 4.9997245984900745e-05,
      "loss": 1.1528,
      "num_input_tokens_seen": 8122208,
      "step": 14000
    },
    {
      "epoch": 2.0859398272266905,
      "grad_norm": 7.9375,
      "learning_rate": 4.999719754319708e-05,
      "loss": 0.8082,
      "num_input_tokens_seen": 8125440,
      "step": 14005
    },
    {
      "epoch": 2.0866845397676497,
      "grad_norm": 13.125,
      "learning_rate": 4.9997148679191174e-05,
      "loss": 1.2972,
      "num_input_tokens_seen": 8128128,
      "step": 14010
    },
    {
      "epoch": 2.087429252308609,
      "grad_norm": 9.75,
      "learning_rate": 4.999709939288385e-05,
      "loss": 0.9918,
      "num_input_tokens_seen": 8131104,
      "step": 14015
    },
    {
      "epoch": 2.088173964849568,
      "grad_norm": 9.8125,
      "learning_rate": 4.9997049684275936e-05,
      "loss": 0.9122,
      "num_input_tokens_seen": 8133952,
      "step": 14020
    },
    {
      "epoch": 2.0889186773905273,
      "grad_norm": 14.0625,
      "learning_rate": 4.999699955336827e-05,
      "loss": 1.2547,
      "num_input_tokens_seen": 8136992,
      "step": 14025
    },
    {
      "epoch": 2.0896633899314865,
      "grad_norm": 18.125,
      "learning_rate": 4.9996949000161705e-05,
      "loss": 1.4493,
      "num_input_tokens_seen": 8139744,
      "step": 14030
    },
    {
      "epoch": 2.0904081024724457,
      "grad_norm": 9.25,
      "learning_rate": 4.99968980246571e-05,
      "loss": 1.0526,
      "num_input_tokens_seen": 8142688,
      "step": 14035
    },
    {
      "epoch": 2.091152815013405,
      "grad_norm": 14.1875,
      "learning_rate": 4.99968466268553e-05,
      "loss": 1.2237,
      "num_input_tokens_seen": 8145632,
      "step": 14040
    },
    {
      "epoch": 2.091897527554364,
      "grad_norm": 16.25,
      "learning_rate": 4.999679480675719e-05,
      "loss": 1.5079,
      "num_input_tokens_seen": 8148736,
      "step": 14045
    },
    {
      "epoch": 2.0926422400953233,
      "grad_norm": 8.875,
      "learning_rate": 4.9996742564363616e-05,
      "loss": 0.9836,
      "num_input_tokens_seen": 8151648,
      "step": 14050
    },
    {
      "epoch": 2.0933869526362825,
      "grad_norm": 14.3125,
      "learning_rate": 4.99966898996755e-05,
      "loss": 1.3144,
      "num_input_tokens_seen": 8154560,
      "step": 14055
    },
    {
      "epoch": 2.0941316651772417,
      "grad_norm": 8.3125,
      "learning_rate": 4.999663681269372e-05,
      "loss": 1.0216,
      "num_input_tokens_seen": 8157568,
      "step": 14060
    },
    {
      "epoch": 2.094876377718201,
      "grad_norm": 6.6875,
      "learning_rate": 4.999658330341915e-05,
      "loss": 1.1483,
      "num_input_tokens_seen": 8160384,
      "step": 14065
    },
    {
      "epoch": 2.09562109025916,
      "grad_norm": 5.25,
      "learning_rate": 4.9996529371852716e-05,
      "loss": 0.9209,
      "num_input_tokens_seen": 8163296,
      "step": 14070
    },
    {
      "epoch": 2.0963658028001193,
      "grad_norm": 12.5,
      "learning_rate": 4.999647501799532e-05,
      "loss": 0.978,
      "num_input_tokens_seen": 8166304,
      "step": 14075
    },
    {
      "epoch": 2.0971105153410785,
      "grad_norm": 7.125,
      "learning_rate": 4.99964202418479e-05,
      "loss": 1.0349,
      "num_input_tokens_seen": 8169600,
      "step": 14080
    },
    {
      "epoch": 2.0978552278820377,
      "grad_norm": 5.46875,
      "learning_rate": 4.999636504341135e-05,
      "loss": 1.1623,
      "num_input_tokens_seen": 8172160,
      "step": 14085
    },
    {
      "epoch": 2.098599940422997,
      "grad_norm": 6.9375,
      "learning_rate": 4.9996309422686624e-05,
      "loss": 0.8464,
      "num_input_tokens_seen": 8175264,
      "step": 14090
    },
    {
      "epoch": 2.099344652963956,
      "grad_norm": 8.5,
      "learning_rate": 4.999625337967465e-05,
      "loss": 0.8014,
      "num_input_tokens_seen": 8178144,
      "step": 14095
    },
    {
      "epoch": 2.1000893655049153,
      "grad_norm": 7.65625,
      "learning_rate": 4.999619691437638e-05,
      "loss": 0.9741,
      "num_input_tokens_seen": 8181088,
      "step": 14100
    },
    {
      "epoch": 2.1008340780458745,
      "grad_norm": 10.5,
      "learning_rate": 4.9996140026792774e-05,
      "loss": 1.1539,
      "num_input_tokens_seen": 8183840,
      "step": 14105
    },
    {
      "epoch": 2.1015787905868333,
      "grad_norm": 6.78125,
      "learning_rate": 4.999608271692479e-05,
      "loss": 0.9404,
      "num_input_tokens_seen": 8186944,
      "step": 14110
    },
    {
      "epoch": 2.1023235031277925,
      "grad_norm": 11.5,
      "learning_rate": 4.999602498477338e-05,
      "loss": 1.2361,
      "num_input_tokens_seen": 8190144,
      "step": 14115
    },
    {
      "epoch": 2.1030682156687517,
      "grad_norm": 13.0625,
      "learning_rate": 4.999596683033955e-05,
      "loss": 1.179,
      "num_input_tokens_seen": 8193056,
      "step": 14120
    },
    {
      "epoch": 2.103812928209711,
      "grad_norm": 7.75,
      "learning_rate": 4.999590825362425e-05,
      "loss": 1.3069,
      "num_input_tokens_seen": 8196032,
      "step": 14125
    },
    {
      "epoch": 2.10455764075067,
      "grad_norm": 6.125,
      "learning_rate": 4.999584925462849e-05,
      "loss": 1.5362,
      "num_input_tokens_seen": 8198880,
      "step": 14130
    },
    {
      "epoch": 2.1053023532916293,
      "grad_norm": 9.8125,
      "learning_rate": 4.999578983335327e-05,
      "loss": 1.1233,
      "num_input_tokens_seen": 8201728,
      "step": 14135
    },
    {
      "epoch": 2.1060470658325885,
      "grad_norm": 5.4375,
      "learning_rate": 4.999572998979957e-05,
      "loss": 0.7181,
      "num_input_tokens_seen": 8204512,
      "step": 14140
    },
    {
      "epoch": 2.1067917783735477,
      "grad_norm": 13.8125,
      "learning_rate": 4.9995669723968426e-05,
      "loss": 0.9299,
      "num_input_tokens_seen": 8207392,
      "step": 14145
    },
    {
      "epoch": 2.107536490914507,
      "grad_norm": 14.875,
      "learning_rate": 4.9995609035860845e-05,
      "loss": 1.1679,
      "num_input_tokens_seen": 8210304,
      "step": 14150
    },
    {
      "epoch": 2.108281203455466,
      "grad_norm": 6.625,
      "learning_rate": 4.9995547925477856e-05,
      "loss": 0.9974,
      "num_input_tokens_seen": 8213248,
      "step": 14155
    },
    {
      "epoch": 2.1090259159964253,
      "grad_norm": 12.6875,
      "learning_rate": 4.999548639282048e-05,
      "loss": 1.0152,
      "num_input_tokens_seen": 8216000,
      "step": 14160
    },
    {
      "epoch": 2.1097706285373845,
      "grad_norm": 5.90625,
      "learning_rate": 4.9995424437889774e-05,
      "loss": 0.8451,
      "num_input_tokens_seen": 8218752,
      "step": 14165
    },
    {
      "epoch": 2.1105153410783437,
      "grad_norm": 9.8125,
      "learning_rate": 4.999536206068678e-05,
      "loss": 1.1748,
      "num_input_tokens_seen": 8221824,
      "step": 14170
    },
    {
      "epoch": 2.111260053619303,
      "grad_norm": 11.0,
      "learning_rate": 4.9995299261212536e-05,
      "loss": 0.9658,
      "num_input_tokens_seen": 8224512,
      "step": 14175
    },
    {
      "epoch": 2.112004766160262,
      "grad_norm": 12.625,
      "learning_rate": 4.999523603946812e-05,
      "loss": 1.0341,
      "num_input_tokens_seen": 8227520,
      "step": 14180
    },
    {
      "epoch": 2.1127494787012213,
      "grad_norm": 7.375,
      "learning_rate": 4.9995172395454606e-05,
      "loss": 0.7786,
      "num_input_tokens_seen": 8230304,
      "step": 14185
    },
    {
      "epoch": 2.1134941912421805,
      "grad_norm": 10.0625,
      "learning_rate": 4.999510832917304e-05,
      "loss": 0.7926,
      "num_input_tokens_seen": 8233024,
      "step": 14190
    },
    {
      "epoch": 2.1142389037831397,
      "grad_norm": 10.0625,
      "learning_rate": 4.9995043840624536e-05,
      "loss": 1.3574,
      "num_input_tokens_seen": 8236032,
      "step": 14195
    },
    {
      "epoch": 2.114983616324099,
      "grad_norm": 12.75,
      "learning_rate": 4.999497892981017e-05,
      "loss": 1.1649,
      "num_input_tokens_seen": 8238656,
      "step": 14200
    },
    {
      "epoch": 2.115728328865058,
      "grad_norm": 9.0,
      "learning_rate": 4.999491359673103e-05,
      "loss": 1.4109,
      "num_input_tokens_seen": 8241440,
      "step": 14205
    },
    {
      "epoch": 2.1164730414060173,
      "grad_norm": 8.9375,
      "learning_rate": 4.999484784138823e-05,
      "loss": 0.7182,
      "num_input_tokens_seen": 8244416,
      "step": 14210
    },
    {
      "epoch": 2.1172177539469765,
      "grad_norm": 11.375,
      "learning_rate": 4.9994781663782884e-05,
      "loss": 1.05,
      "num_input_tokens_seen": 8247296,
      "step": 14215
    },
    {
      "epoch": 2.1179624664879357,
      "grad_norm": 7.40625,
      "learning_rate": 4.99947150639161e-05,
      "loss": 0.8955,
      "num_input_tokens_seen": 8250304,
      "step": 14220
    },
    {
      "epoch": 2.118707179028895,
      "grad_norm": 9.3125,
      "learning_rate": 4.9994648041789016e-05,
      "loss": 1.4188,
      "num_input_tokens_seen": 8253472,
      "step": 14225
    },
    {
      "epoch": 2.119451891569854,
      "grad_norm": 6.5625,
      "learning_rate": 4.999458059740275e-05,
      "loss": 0.8507,
      "num_input_tokens_seen": 8256288,
      "step": 14230
    },
    {
      "epoch": 2.1201966041108133,
      "grad_norm": 8.25,
      "learning_rate": 4.9994512730758454e-05,
      "loss": 1.1337,
      "num_input_tokens_seen": 8259104,
      "step": 14235
    },
    {
      "epoch": 2.1209413166517725,
      "grad_norm": 12.625,
      "learning_rate": 4.999444444185727e-05,
      "loss": 1.0595,
      "num_input_tokens_seen": 8261984,
      "step": 14240
    },
    {
      "epoch": 2.1216860291927317,
      "grad_norm": 10.5625,
      "learning_rate": 4.999437573070034e-05,
      "loss": 0.7705,
      "num_input_tokens_seen": 8264768,
      "step": 14245
    },
    {
      "epoch": 2.122430741733691,
      "grad_norm": 5.125,
      "learning_rate": 4.999430659728884e-05,
      "loss": 0.8694,
      "num_input_tokens_seen": 8267840,
      "step": 14250
    },
    {
      "epoch": 2.12317545427465,
      "grad_norm": 11.4375,
      "learning_rate": 4.9994237041623935e-05,
      "loss": 1.1313,
      "num_input_tokens_seen": 8270784,
      "step": 14255
    },
    {
      "epoch": 2.1239201668156094,
      "grad_norm": 7.875,
      "learning_rate": 4.99941670637068e-05,
      "loss": 1.0879,
      "num_input_tokens_seen": 8273952,
      "step": 14260
    },
    {
      "epoch": 2.1246648793565686,
      "grad_norm": 10.9375,
      "learning_rate": 4.999409666353861e-05,
      "loss": 1.3327,
      "num_input_tokens_seen": 8277088,
      "step": 14265
    },
    {
      "epoch": 2.1254095918975278,
      "grad_norm": 11.9375,
      "learning_rate": 4.999402584112057e-05,
      "loss": 1.0276,
      "num_input_tokens_seen": 8280000,
      "step": 14270
    },
    {
      "epoch": 2.1261543044384865,
      "grad_norm": 9.0625,
      "learning_rate": 4.999395459645385e-05,
      "loss": 0.8536,
      "num_input_tokens_seen": 8282784,
      "step": 14275
    },
    {
      "epoch": 2.126899016979446,
      "grad_norm": 13.0625,
      "learning_rate": 4.999388292953968e-05,
      "loss": 1.1355,
      "num_input_tokens_seen": 8285792,
      "step": 14280
    },
    {
      "epoch": 2.127643729520405,
      "grad_norm": 10.4375,
      "learning_rate": 4.999381084037926e-05,
      "loss": 1.3253,
      "num_input_tokens_seen": 8288544,
      "step": 14285
    },
    {
      "epoch": 2.128388442061364,
      "grad_norm": 8.25,
      "learning_rate": 4.999373832897381e-05,
      "loss": 0.9938,
      "num_input_tokens_seen": 8291200,
      "step": 14290
    },
    {
      "epoch": 2.1291331546023233,
      "grad_norm": 6.9375,
      "learning_rate": 4.9993665395324554e-05,
      "loss": 1.1294,
      "num_input_tokens_seen": 8294208,
      "step": 14295
    },
    {
      "epoch": 2.1298778671432825,
      "grad_norm": 12.5625,
      "learning_rate": 4.999359203943272e-05,
      "loss": 1.2398,
      "num_input_tokens_seen": 8297248,
      "step": 14300
    },
    {
      "epoch": 2.1306225796842417,
      "grad_norm": 9.5,
      "learning_rate": 4.999351826129955e-05,
      "loss": 1.1736,
      "num_input_tokens_seen": 8300064,
      "step": 14305
    },
    {
      "epoch": 2.131367292225201,
      "grad_norm": 14.0,
      "learning_rate": 4.9993444060926296e-05,
      "loss": 1.1815,
      "num_input_tokens_seen": 8302912,
      "step": 14310
    },
    {
      "epoch": 2.13211200476616,
      "grad_norm": 12.5625,
      "learning_rate": 4.9993369438314204e-05,
      "loss": 1.273,
      "num_input_tokens_seen": 8305888,
      "step": 14315
    },
    {
      "epoch": 2.1328567173071193,
      "grad_norm": 9.9375,
      "learning_rate": 4.9993294393464536e-05,
      "loss": 0.9116,
      "num_input_tokens_seen": 8308448,
      "step": 14320
    },
    {
      "epoch": 2.1336014298480785,
      "grad_norm": 6.28125,
      "learning_rate": 4.999321892637856e-05,
      "loss": 0.7636,
      "num_input_tokens_seen": 8311424,
      "step": 14325
    },
    {
      "epoch": 2.1343461423890377,
      "grad_norm": 10.4375,
      "learning_rate": 4.9993143037057554e-05,
      "loss": 1.2643,
      "num_input_tokens_seen": 8314208,
      "step": 14330
    },
    {
      "epoch": 2.135090854929997,
      "grad_norm": 14.375,
      "learning_rate": 4.99930667255028e-05,
      "loss": 1.4576,
      "num_input_tokens_seen": 8317216,
      "step": 14335
    },
    {
      "epoch": 2.135835567470956,
      "grad_norm": 9.125,
      "learning_rate": 4.999298999171559e-05,
      "loss": 1.3939,
      "num_input_tokens_seen": 8319936,
      "step": 14340
    },
    {
      "epoch": 2.1365802800119154,
      "grad_norm": 21.125,
      "learning_rate": 4.99929128356972e-05,
      "loss": 1.1929,
      "num_input_tokens_seen": 8322912,
      "step": 14345
    },
    {
      "epoch": 2.1373249925528746,
      "grad_norm": 13.875,
      "learning_rate": 4.9992835257448965e-05,
      "loss": 1.1022,
      "num_input_tokens_seen": 8326432,
      "step": 14350
    },
    {
      "epoch": 2.1380697050938338,
      "grad_norm": 11.625,
      "learning_rate": 4.999275725697218e-05,
      "loss": 0.86,
      "num_input_tokens_seen": 8329408,
      "step": 14355
    },
    {
      "epoch": 2.138814417634793,
      "grad_norm": 4.8125,
      "learning_rate": 4.9992678834268154e-05,
      "loss": 0.683,
      "num_input_tokens_seen": 8332256,
      "step": 14360
    },
    {
      "epoch": 2.139559130175752,
      "grad_norm": 15.8125,
      "learning_rate": 4.999259998933822e-05,
      "loss": 1.3898,
      "num_input_tokens_seen": 8335232,
      "step": 14365
    },
    {
      "epoch": 2.1403038427167114,
      "grad_norm": 6.4375,
      "learning_rate": 4.9992520722183714e-05,
      "loss": 1.3715,
      "num_input_tokens_seen": 8337920,
      "step": 14370
    },
    {
      "epoch": 2.1410485552576706,
      "grad_norm": 13.375,
      "learning_rate": 4.999244103280597e-05,
      "loss": 1.1415,
      "num_input_tokens_seen": 8340480,
      "step": 14375
    },
    {
      "epoch": 2.1417932677986298,
      "grad_norm": 8.25,
      "learning_rate": 4.999236092120634e-05,
      "loss": 0.9956,
      "num_input_tokens_seen": 8343520,
      "step": 14380
    },
    {
      "epoch": 2.142537980339589,
      "grad_norm": 8.3125,
      "learning_rate": 4.999228038738617e-05,
      "loss": 0.8205,
      "num_input_tokens_seen": 8346592,
      "step": 14385
    },
    {
      "epoch": 2.143282692880548,
      "grad_norm": 6.90625,
      "learning_rate": 4.999219943134683e-05,
      "loss": 0.761,
      "num_input_tokens_seen": 8349504,
      "step": 14390
    },
    {
      "epoch": 2.1440274054215074,
      "grad_norm": 8.125,
      "learning_rate": 4.9992118053089675e-05,
      "loss": 0.8233,
      "num_input_tokens_seen": 8352384,
      "step": 14395
    },
    {
      "epoch": 2.1447721179624666,
      "grad_norm": 11.625,
      "learning_rate": 4.999203625261609e-05,
      "loss": 0.8531,
      "num_input_tokens_seen": 8355584,
      "step": 14400
    },
    {
      "epoch": 2.145516830503426,
      "grad_norm": 15.4375,
      "learning_rate": 4.999195402992745e-05,
      "loss": 1.2453,
      "num_input_tokens_seen": 8358624,
      "step": 14405
    },
    {
      "epoch": 2.146261543044385,
      "grad_norm": 12.0,
      "learning_rate": 4.999187138502515e-05,
      "loss": 0.981,
      "num_input_tokens_seen": 8361472,
      "step": 14410
    },
    {
      "epoch": 2.147006255585344,
      "grad_norm": 16.125,
      "learning_rate": 4.999178831791058e-05,
      "loss": 1.2288,
      "num_input_tokens_seen": 8364384,
      "step": 14415
    },
    {
      "epoch": 2.1477509681263034,
      "grad_norm": 4.34375,
      "learning_rate": 4.999170482858515e-05,
      "loss": 0.8912,
      "num_input_tokens_seen": 8367136,
      "step": 14420
    },
    {
      "epoch": 2.1484956806672626,
      "grad_norm": 11.25,
      "learning_rate": 4.999162091705026e-05,
      "loss": 1.0237,
      "num_input_tokens_seen": 8370112,
      "step": 14425
    },
    {
      "epoch": 2.149240393208222,
      "grad_norm": 8.1875,
      "learning_rate": 4.9991536583307344e-05,
      "loss": 1.2578,
      "num_input_tokens_seen": 8373024,
      "step": 14430
    },
    {
      "epoch": 2.149985105749181,
      "grad_norm": 7.15625,
      "learning_rate": 4.999145182735782e-05,
      "loss": 0.8923,
      "num_input_tokens_seen": 8375648,
      "step": 14435
    },
    {
      "epoch": 2.15072981829014,
      "grad_norm": 10.0625,
      "learning_rate": 4.999136664920311e-05,
      "loss": 0.9747,
      "num_input_tokens_seen": 8378656,
      "step": 14440
    },
    {
      "epoch": 2.1514745308310994,
      "grad_norm": 7.5,
      "learning_rate": 4.999128104884466e-05,
      "loss": 1.162,
      "num_input_tokens_seen": 8381408,
      "step": 14445
    },
    {
      "epoch": 2.152219243372058,
      "grad_norm": 16.75,
      "learning_rate": 4.999119502628392e-05,
      "loss": 1.1071,
      "num_input_tokens_seen": 8384416,
      "step": 14450
    },
    {
      "epoch": 2.1529639559130174,
      "grad_norm": 11.5,
      "learning_rate": 4.999110858152234e-05,
      "loss": 1.099,
      "num_input_tokens_seen": 8387712,
      "step": 14455
    },
    {
      "epoch": 2.1537086684539766,
      "grad_norm": 11.0625,
      "learning_rate": 4.999102171456138e-05,
      "loss": 0.8502,
      "num_input_tokens_seen": 8390688,
      "step": 14460
    },
    {
      "epoch": 2.1544533809949358,
      "grad_norm": 9.875,
      "learning_rate": 4.999093442540251e-05,
      "loss": 1.5198,
      "num_input_tokens_seen": 8393664,
      "step": 14465
    },
    {
      "epoch": 2.155198093535895,
      "grad_norm": 6.0,
      "learning_rate": 4.9990846714047204e-05,
      "loss": 0.8299,
      "num_input_tokens_seen": 8396480,
      "step": 14470
    },
    {
      "epoch": 2.155942806076854,
      "grad_norm": 11.5625,
      "learning_rate": 4.9990758580496935e-05,
      "loss": 1.2164,
      "num_input_tokens_seen": 8399264,
      "step": 14475
    },
    {
      "epoch": 2.1566875186178134,
      "grad_norm": 5.78125,
      "learning_rate": 4.99906700247532e-05,
      "loss": 1.1903,
      "num_input_tokens_seen": 8402016,
      "step": 14480
    },
    {
      "epoch": 2.1574322311587726,
      "grad_norm": 10.1875,
      "learning_rate": 4.99905810468175e-05,
      "loss": 1.034,
      "num_input_tokens_seen": 8404672,
      "step": 14485
    },
    {
      "epoch": 2.158176943699732,
      "grad_norm": 7.0625,
      "learning_rate": 4.999049164669133e-05,
      "loss": 0.9011,
      "num_input_tokens_seen": 8407328,
      "step": 14490
    },
    {
      "epoch": 2.158921656240691,
      "grad_norm": 12.75,
      "learning_rate": 4.9990401824376196e-05,
      "loss": 0.876,
      "num_input_tokens_seen": 8410240,
      "step": 14495
    },
    {
      "epoch": 2.15966636878165,
      "grad_norm": 23.25,
      "learning_rate": 4.999031157987364e-05,
      "loss": 1.1117,
      "num_input_tokens_seen": 8413152,
      "step": 14500
    },
    {
      "epoch": 2.1604110813226094,
      "grad_norm": 12.4375,
      "learning_rate": 4.9990220913185146e-05,
      "loss": 0.8861,
      "num_input_tokens_seen": 8416160,
      "step": 14505
    },
    {
      "epoch": 2.1611557938635686,
      "grad_norm": 8.6875,
      "learning_rate": 4.9990129824312285e-05,
      "loss": 1.1851,
      "num_input_tokens_seen": 8418976,
      "step": 14510
    },
    {
      "epoch": 2.161900506404528,
      "grad_norm": 9.75,
      "learning_rate": 4.9990038313256573e-05,
      "loss": 1.0576,
      "num_input_tokens_seen": 8421856,
      "step": 14515
    },
    {
      "epoch": 2.162645218945487,
      "grad_norm": 8.0,
      "learning_rate": 4.998994638001957e-05,
      "loss": 0.8923,
      "num_input_tokens_seen": 8424448,
      "step": 14520
    },
    {
      "epoch": 2.163389931486446,
      "grad_norm": 6.03125,
      "learning_rate": 4.998985402460281e-05,
      "loss": 0.9423,
      "num_input_tokens_seen": 8427392,
      "step": 14525
    },
    {
      "epoch": 2.1641346440274054,
      "grad_norm": 8.625,
      "learning_rate": 4.998976124700787e-05,
      "loss": 1.3639,
      "num_input_tokens_seen": 8430528,
      "step": 14530
    },
    {
      "epoch": 2.1648793565683646,
      "grad_norm": 9.6875,
      "learning_rate": 4.9989668047236316e-05,
      "loss": 0.8812,
      "num_input_tokens_seen": 8433408,
      "step": 14535
    },
    {
      "epoch": 2.165624069109324,
      "grad_norm": 7.21875,
      "learning_rate": 4.998957442528972e-05,
      "loss": 1.3089,
      "num_input_tokens_seen": 8436480,
      "step": 14540
    },
    {
      "epoch": 2.166368781650283,
      "grad_norm": 8.625,
      "learning_rate": 4.998948038116965e-05,
      "loss": 1.1544,
      "num_input_tokens_seen": 8439200,
      "step": 14545
    },
    {
      "epoch": 2.167113494191242,
      "grad_norm": 6.34375,
      "learning_rate": 4.9989385914877717e-05,
      "loss": 0.7478,
      "num_input_tokens_seen": 8442048,
      "step": 14550
    },
    {
      "epoch": 2.1678582067322014,
      "grad_norm": 8.9375,
      "learning_rate": 4.998929102641551e-05,
      "loss": 1.1137,
      "num_input_tokens_seen": 8444832,
      "step": 14555
    },
    {
      "epoch": 2.1686029192731606,
      "grad_norm": 14.6875,
      "learning_rate": 4.998919571578462e-05,
      "loss": 1.0496,
      "num_input_tokens_seen": 8447616,
      "step": 14560
    },
    {
      "epoch": 2.16934763181412,
      "grad_norm": 6.6875,
      "learning_rate": 4.998909998298668e-05,
      "loss": 0.8987,
      "num_input_tokens_seen": 8450656,
      "step": 14565
    },
    {
      "epoch": 2.170092344355079,
      "grad_norm": 13.3125,
      "learning_rate": 4.998900382802327e-05,
      "loss": 1.5305,
      "num_input_tokens_seen": 8453536,
      "step": 14570
    },
    {
      "epoch": 2.1708370568960382,
      "grad_norm": 8.75,
      "learning_rate": 4.9988907250896056e-05,
      "loss": 1.1379,
      "num_input_tokens_seen": 8456576,
      "step": 14575
    },
    {
      "epoch": 2.1715817694369974,
      "grad_norm": 5.8125,
      "learning_rate": 4.998881025160665e-05,
      "loss": 1.0005,
      "num_input_tokens_seen": 8459296,
      "step": 14580
    },
    {
      "epoch": 2.1723264819779566,
      "grad_norm": 4.5,
      "learning_rate": 4.9988712830156694e-05,
      "loss": 0.8047,
      "num_input_tokens_seen": 8462272,
      "step": 14585
    },
    {
      "epoch": 2.173071194518916,
      "grad_norm": 6.3125,
      "learning_rate": 4.998861498654782e-05,
      "loss": 1.0096,
      "num_input_tokens_seen": 8465152,
      "step": 14590
    },
    {
      "epoch": 2.173815907059875,
      "grad_norm": 6.46875,
      "learning_rate": 4.9988516720781705e-05,
      "loss": 0.859,
      "num_input_tokens_seen": 8468032,
      "step": 14595
    },
    {
      "epoch": 2.1745606196008342,
      "grad_norm": 9.4375,
      "learning_rate": 4.998841803286e-05,
      "loss": 0.9008,
      "num_input_tokens_seen": 8470784,
      "step": 14600
    },
    {
      "epoch": 2.1753053321417934,
      "grad_norm": 11.75,
      "learning_rate": 4.9988318922784364e-05,
      "loss": 1.1246,
      "num_input_tokens_seen": 8473568,
      "step": 14605
    },
    {
      "epoch": 2.1760500446827526,
      "grad_norm": 17.0,
      "learning_rate": 4.9988219390556466e-05,
      "loss": 0.957,
      "num_input_tokens_seen": 8476512,
      "step": 14610
    },
    {
      "epoch": 2.176794757223712,
      "grad_norm": 11.1875,
      "learning_rate": 4.998811943617801e-05,
      "loss": 1.2953,
      "num_input_tokens_seen": 8479392,
      "step": 14615
    },
    {
      "epoch": 2.177539469764671,
      "grad_norm": 13.1875,
      "learning_rate": 4.998801905965067e-05,
      "loss": 1.1981,
      "num_input_tokens_seen": 8482240,
      "step": 14620
    },
    {
      "epoch": 2.17828418230563,
      "grad_norm": 9.5625,
      "learning_rate": 4.998791826097615e-05,
      "loss": 1.0438,
      "num_input_tokens_seen": 8484960,
      "step": 14625
    },
    {
      "epoch": 2.179028894846589,
      "grad_norm": 14.3125,
      "learning_rate": 4.998781704015614e-05,
      "loss": 0.9735,
      "num_input_tokens_seen": 8487616,
      "step": 14630
    },
    {
      "epoch": 2.179773607387548,
      "grad_norm": 5.25,
      "learning_rate": 4.998771539719236e-05,
      "loss": 0.9937,
      "num_input_tokens_seen": 8490368,
      "step": 14635
    },
    {
      "epoch": 2.1805183199285074,
      "grad_norm": 17.375,
      "learning_rate": 4.998761333208652e-05,
      "loss": 1.3712,
      "num_input_tokens_seen": 8493216,
      "step": 14640
    },
    {
      "epoch": 2.1812630324694666,
      "grad_norm": 12.1875,
      "learning_rate": 4.9987510844840354e-05,
      "loss": 1.0841,
      "num_input_tokens_seen": 8496128,
      "step": 14645
    },
    {
      "epoch": 2.182007745010426,
      "grad_norm": 10.875,
      "learning_rate": 4.998740793545559e-05,
      "loss": 1.068,
      "num_input_tokens_seen": 8498848,
      "step": 14650
    },
    {
      "epoch": 2.182752457551385,
      "grad_norm": 5.75,
      "learning_rate": 4.998730460393397e-05,
      "loss": 0.8812,
      "num_input_tokens_seen": 8501856,
      "step": 14655
    },
    {
      "epoch": 2.1834971700923442,
      "grad_norm": 19.0,
      "learning_rate": 4.998720085027723e-05,
      "loss": 1.3009,
      "num_input_tokens_seen": 8504704,
      "step": 14660
    },
    {
      "epoch": 2.1842418826333034,
      "grad_norm": 12.3125,
      "learning_rate": 4.998709667448712e-05,
      "loss": 0.9856,
      "num_input_tokens_seen": 8507360,
      "step": 14665
    },
    {
      "epoch": 2.1849865951742626,
      "grad_norm": 6.0625,
      "learning_rate": 4.998699207656542e-05,
      "loss": 1.133,
      "num_input_tokens_seen": 8510400,
      "step": 14670
    },
    {
      "epoch": 2.185731307715222,
      "grad_norm": 7.65625,
      "learning_rate": 4.9986887056513874e-05,
      "loss": 0.8329,
      "num_input_tokens_seen": 8513184,
      "step": 14675
    },
    {
      "epoch": 2.186476020256181,
      "grad_norm": 7.75,
      "learning_rate": 4.998678161433427e-05,
      "loss": 0.9666,
      "num_input_tokens_seen": 8516704,
      "step": 14680
    },
    {
      "epoch": 2.1872207327971402,
      "grad_norm": 8.25,
      "learning_rate": 4.998667575002839e-05,
      "loss": 1.0932,
      "num_input_tokens_seen": 8519584,
      "step": 14685
    },
    {
      "epoch": 2.1879654453380994,
      "grad_norm": 11.5625,
      "learning_rate": 4.998656946359801e-05,
      "loss": 1.1663,
      "num_input_tokens_seen": 8522144,
      "step": 14690
    },
    {
      "epoch": 2.1887101578790586,
      "grad_norm": 10.5,
      "learning_rate": 4.998646275504494e-05,
      "loss": 1.3783,
      "num_input_tokens_seen": 8525248,
      "step": 14695
    },
    {
      "epoch": 2.189454870420018,
      "grad_norm": 8.6875,
      "learning_rate": 4.998635562437098e-05,
      "loss": 1.1082,
      "num_input_tokens_seen": 8527904,
      "step": 14700
    },
    {
      "epoch": 2.190199582960977,
      "grad_norm": 3.296875,
      "learning_rate": 4.9986248071577934e-05,
      "loss": 0.919,
      "num_input_tokens_seen": 8530720,
      "step": 14705
    },
    {
      "epoch": 2.1909442955019363,
      "grad_norm": 11.125,
      "learning_rate": 4.998614009666762e-05,
      "loss": 1.0719,
      "num_input_tokens_seen": 8533344,
      "step": 14710
    },
    {
      "epoch": 2.1916890080428955,
      "grad_norm": 8.4375,
      "learning_rate": 4.9986031699641866e-05,
      "loss": 1.0427,
      "num_input_tokens_seen": 8536032,
      "step": 14715
    },
    {
      "epoch": 2.1924337205838547,
      "grad_norm": 8.625,
      "learning_rate": 4.99859228805025e-05,
      "loss": 1.0975,
      "num_input_tokens_seen": 8539040,
      "step": 14720
    },
    {
      "epoch": 2.193178433124814,
      "grad_norm": 5.5625,
      "learning_rate": 4.9985813639251355e-05,
      "loss": 1.4399,
      "num_input_tokens_seen": 8542048,
      "step": 14725
    },
    {
      "epoch": 2.193923145665773,
      "grad_norm": 13.4375,
      "learning_rate": 4.9985703975890294e-05,
      "loss": 1.2099,
      "num_input_tokens_seen": 8544672,
      "step": 14730
    },
    {
      "epoch": 2.1946678582067323,
      "grad_norm": 10.125,
      "learning_rate": 4.998559389042115e-05,
      "loss": 1.3236,
      "num_input_tokens_seen": 8547456,
      "step": 14735
    },
    {
      "epoch": 2.1954125707476915,
      "grad_norm": 22.0,
      "learning_rate": 4.99854833828458e-05,
      "loss": 1.2383,
      "num_input_tokens_seen": 8550304,
      "step": 14740
    },
    {
      "epoch": 2.1961572832886507,
      "grad_norm": 7.375,
      "learning_rate": 4.998537245316609e-05,
      "loss": 1.3345,
      "num_input_tokens_seen": 8553408,
      "step": 14745
    },
    {
      "epoch": 2.19690199582961,
      "grad_norm": 17.0,
      "learning_rate": 4.998526110138392e-05,
      "loss": 1.3244,
      "num_input_tokens_seen": 8556128,
      "step": 14750
    },
    {
      "epoch": 2.197646708370569,
      "grad_norm": 9.25,
      "learning_rate": 4.9985149327501146e-05,
      "loss": 1.2306,
      "num_input_tokens_seen": 8558976,
      "step": 14755
    },
    {
      "epoch": 2.1983914209115283,
      "grad_norm": 14.0,
      "learning_rate": 4.998503713151967e-05,
      "loss": 0.7918,
      "num_input_tokens_seen": 8561888,
      "step": 14760
    },
    {
      "epoch": 2.1991361334524875,
      "grad_norm": 12.4375,
      "learning_rate": 4.9984924513441397e-05,
      "loss": 1.4048,
      "num_input_tokens_seen": 8564640,
      "step": 14765
    },
    {
      "epoch": 2.1998808459934467,
      "grad_norm": 15.5625,
      "learning_rate": 4.9984811473268214e-05,
      "loss": 1.313,
      "num_input_tokens_seen": 8567456,
      "step": 14770
    },
    {
      "epoch": 2.200625558534406,
      "grad_norm": 6.84375,
      "learning_rate": 4.998469801100203e-05,
      "loss": 1.1103,
      "num_input_tokens_seen": 8570528,
      "step": 14775
    },
    {
      "epoch": 2.201370271075365,
      "grad_norm": 15.75,
      "learning_rate": 4.998458412664476e-05,
      "loss": 1.1309,
      "num_input_tokens_seen": 8573664,
      "step": 14780
    },
    {
      "epoch": 2.2021149836163243,
      "grad_norm": 6.90625,
      "learning_rate": 4.9984469820198345e-05,
      "loss": 1.0072,
      "num_input_tokens_seen": 8576704,
      "step": 14785
    },
    {
      "epoch": 2.202859696157283,
      "grad_norm": 9.0625,
      "learning_rate": 4.9984355091664705e-05,
      "loss": 0.9232,
      "num_input_tokens_seen": 8579552,
      "step": 14790
    },
    {
      "epoch": 2.2036044086982427,
      "grad_norm": 13.3125,
      "learning_rate": 4.9984239941045766e-05,
      "loss": 1.1337,
      "num_input_tokens_seen": 8582400,
      "step": 14795
    },
    {
      "epoch": 2.2043491212392015,
      "grad_norm": 10.9375,
      "learning_rate": 4.99841243683435e-05,
      "loss": 0.9191,
      "num_input_tokens_seen": 8585344,
      "step": 14800
    },
    {
      "epoch": 2.2050938337801607,
      "grad_norm": 15.9375,
      "learning_rate": 4.998400837355984e-05,
      "loss": 1.1776,
      "num_input_tokens_seen": 8588320,
      "step": 14805
    },
    {
      "epoch": 2.20583854632112,
      "grad_norm": 13.5625,
      "learning_rate": 4.998389195669675e-05,
      "loss": 1.2323,
      "num_input_tokens_seen": 8591264,
      "step": 14810
    },
    {
      "epoch": 2.206583258862079,
      "grad_norm": 10.0,
      "learning_rate": 4.998377511775621e-05,
      "loss": 1.3,
      "num_input_tokens_seen": 8594176,
      "step": 14815
    },
    {
      "epoch": 2.2073279714030383,
      "grad_norm": 4.5625,
      "learning_rate": 4.9983657856740165e-05,
      "loss": 1.0137,
      "num_input_tokens_seen": 8597280,
      "step": 14820
    },
    {
      "epoch": 2.2080726839439975,
      "grad_norm": 8.5,
      "learning_rate": 4.9983540173650614e-05,
      "loss": 1.0032,
      "num_input_tokens_seen": 8600192,
      "step": 14825
    },
    {
      "epoch": 2.2088173964849567,
      "grad_norm": 9.75,
      "learning_rate": 4.9983422068489546e-05,
      "loss": 0.9824,
      "num_input_tokens_seen": 8603232,
      "step": 14830
    },
    {
      "epoch": 2.209562109025916,
      "grad_norm": 14.6875,
      "learning_rate": 4.998330354125896e-05,
      "loss": 1.7561,
      "num_input_tokens_seen": 8606176,
      "step": 14835
    },
    {
      "epoch": 2.210306821566875,
      "grad_norm": 9.125,
      "learning_rate": 4.998318459196085e-05,
      "loss": 0.874,
      "num_input_tokens_seen": 8609120,
      "step": 14840
    },
    {
      "epoch": 2.2110515341078343,
      "grad_norm": 9.0625,
      "learning_rate": 4.998306522059723e-05,
      "loss": 0.9429,
      "num_input_tokens_seen": 8611936,
      "step": 14845
    },
    {
      "epoch": 2.2117962466487935,
      "grad_norm": 13.6875,
      "learning_rate": 4.9982945427170115e-05,
      "loss": 1.2539,
      "num_input_tokens_seen": 8614784,
      "step": 14850
    },
    {
      "epoch": 2.2125409591897527,
      "grad_norm": 8.25,
      "learning_rate": 4.998282521168153e-05,
      "loss": 1.1755,
      "num_input_tokens_seen": 8617888,
      "step": 14855
    },
    {
      "epoch": 2.213285671730712,
      "grad_norm": 7.0625,
      "learning_rate": 4.9982704574133497e-05,
      "loss": 1.443,
      "num_input_tokens_seen": 8620896,
      "step": 14860
    },
    {
      "epoch": 2.214030384271671,
      "grad_norm": 17.5,
      "learning_rate": 4.998258351452806e-05,
      "loss": 1.1343,
      "num_input_tokens_seen": 8623808,
      "step": 14865
    },
    {
      "epoch": 2.2147750968126303,
      "grad_norm": 10.8125,
      "learning_rate": 4.998246203286727e-05,
      "loss": 1.2614,
      "num_input_tokens_seen": 8626688,
      "step": 14870
    },
    {
      "epoch": 2.2155198093535895,
      "grad_norm": 15.8125,
      "learning_rate": 4.9982340129153185e-05,
      "loss": 1.3141,
      "num_input_tokens_seen": 8629536,
      "step": 14875
    },
    {
      "epoch": 2.2162645218945487,
      "grad_norm": 10.75,
      "learning_rate": 4.9982217803387844e-05,
      "loss": 1.2341,
      "num_input_tokens_seen": 8632480,
      "step": 14880
    },
    {
      "epoch": 2.217009234435508,
      "grad_norm": 6.71875,
      "learning_rate": 4.998209505557333e-05,
      "loss": 1.4118,
      "num_input_tokens_seen": 8635520,
      "step": 14885
    },
    {
      "epoch": 2.217753946976467,
      "grad_norm": 10.875,
      "learning_rate": 4.99819718857117e-05,
      "loss": 1.124,
      "num_input_tokens_seen": 8638752,
      "step": 14890
    },
    {
      "epoch": 2.2184986595174263,
      "grad_norm": 19.875,
      "learning_rate": 4.998184829380505e-05,
      "loss": 0.9778,
      "num_input_tokens_seen": 8641824,
      "step": 14895
    },
    {
      "epoch": 2.2192433720583855,
      "grad_norm": 10.4375,
      "learning_rate": 4.9981724279855466e-05,
      "loss": 0.961,
      "num_input_tokens_seen": 8644928,
      "step": 14900
    },
    {
      "epoch": 2.2199880845993447,
      "grad_norm": 5.625,
      "learning_rate": 4.998159984386504e-05,
      "loss": 0.9778,
      "num_input_tokens_seen": 8648000,
      "step": 14905
    },
    {
      "epoch": 2.220732797140304,
      "grad_norm": 8.0,
      "learning_rate": 4.9981474985835875e-05,
      "loss": 1.3095,
      "num_input_tokens_seen": 8651008,
      "step": 14910
    },
    {
      "epoch": 2.221477509681263,
      "grad_norm": 4.40625,
      "learning_rate": 4.9981349705770074e-05,
      "loss": 0.98,
      "num_input_tokens_seen": 8653696,
      "step": 14915
    },
    {
      "epoch": 2.2222222222222223,
      "grad_norm": 12.9375,
      "learning_rate": 4.998122400366977e-05,
      "loss": 1.1534,
      "num_input_tokens_seen": 8656576,
      "step": 14920
    },
    {
      "epoch": 2.2229669347631815,
      "grad_norm": 8.1875,
      "learning_rate": 4.998109787953708e-05,
      "loss": 1.1079,
      "num_input_tokens_seen": 8659488,
      "step": 14925
    },
    {
      "epoch": 2.2237116473041407,
      "grad_norm": 14.875,
      "learning_rate": 4.998097133337412e-05,
      "loss": 0.9247,
      "num_input_tokens_seen": 8662208,
      "step": 14930
    },
    {
      "epoch": 2.2244563598451,
      "grad_norm": 8.125,
      "learning_rate": 4.998084436518303e-05,
      "loss": 1.0598,
      "num_input_tokens_seen": 8664896,
      "step": 14935
    },
    {
      "epoch": 2.225201072386059,
      "grad_norm": 10.75,
      "learning_rate": 4.998071697496598e-05,
      "loss": 1.0248,
      "num_input_tokens_seen": 8667808,
      "step": 14940
    },
    {
      "epoch": 2.2259457849270183,
      "grad_norm": 10.8125,
      "learning_rate": 4.99805891627251e-05,
      "loss": 1.1192,
      "num_input_tokens_seen": 8670880,
      "step": 14945
    },
    {
      "epoch": 2.2266904974679775,
      "grad_norm": 9.9375,
      "learning_rate": 4.998046092846256e-05,
      "loss": 1.0548,
      "num_input_tokens_seen": 8673856,
      "step": 14950
    },
    {
      "epoch": 2.2274352100089367,
      "grad_norm": 9.9375,
      "learning_rate": 4.998033227218052e-05,
      "loss": 1.1779,
      "num_input_tokens_seen": 8676736,
      "step": 14955
    },
    {
      "epoch": 2.228179922549896,
      "grad_norm": 13.5625,
      "learning_rate": 4.998020319388115e-05,
      "loss": 1.0152,
      "num_input_tokens_seen": 8679872,
      "step": 14960
    },
    {
      "epoch": 2.2289246350908547,
      "grad_norm": 5.5625,
      "learning_rate": 4.998007369356664e-05,
      "loss": 1.2413,
      "num_input_tokens_seen": 8682496,
      "step": 14965
    },
    {
      "epoch": 2.2296693476318143,
      "grad_norm": 6.65625,
      "learning_rate": 4.997994377123917e-05,
      "loss": 1.149,
      "num_input_tokens_seen": 8685280,
      "step": 14970
    },
    {
      "epoch": 2.230414060172773,
      "grad_norm": 13.9375,
      "learning_rate": 4.997981342690095e-05,
      "loss": 1.0051,
      "num_input_tokens_seen": 8688128,
      "step": 14975
    },
    {
      "epoch": 2.2311587727137323,
      "grad_norm": 13.875,
      "learning_rate": 4.9979682660554154e-05,
      "loss": 1.0236,
      "num_input_tokens_seen": 8691200,
      "step": 14980
    },
    {
      "epoch": 2.2319034852546915,
      "grad_norm": 19.125,
      "learning_rate": 4.997955147220101e-05,
      "loss": 1.2794,
      "num_input_tokens_seen": 8693888,
      "step": 14985
    },
    {
      "epoch": 2.2326481977956507,
      "grad_norm": 7.6875,
      "learning_rate": 4.997941986184375e-05,
      "loss": 0.9884,
      "num_input_tokens_seen": 8696448,
      "step": 14990
    },
    {
      "epoch": 2.23339291033661,
      "grad_norm": 13.375,
      "learning_rate": 4.9979287829484555e-05,
      "loss": 0.7165,
      "num_input_tokens_seen": 8699360,
      "step": 14995
    },
    {
      "epoch": 2.234137622877569,
      "grad_norm": 9.875,
      "learning_rate": 4.99791553751257e-05,
      "loss": 1.3495,
      "num_input_tokens_seen": 8702272,
      "step": 15000
    },
    {
      "epoch": 2.2348823354185283,
      "grad_norm": 13.125,
      "learning_rate": 4.997902249876939e-05,
      "loss": 0.8412,
      "num_input_tokens_seen": 8705472,
      "step": 15005
    },
    {
      "epoch": 2.2356270479594875,
      "grad_norm": 8.625,
      "learning_rate": 4.997888920041789e-05,
      "loss": 1.1819,
      "num_input_tokens_seen": 8708192,
      "step": 15010
    },
    {
      "epoch": 2.2363717605004467,
      "grad_norm": 17.875,
      "learning_rate": 4.997875548007343e-05,
      "loss": 1.1222,
      "num_input_tokens_seen": 8711072,
      "step": 15015
    },
    {
      "epoch": 2.237116473041406,
      "grad_norm": 6.375,
      "learning_rate": 4.99786213377383e-05,
      "loss": 0.8829,
      "num_input_tokens_seen": 8713792,
      "step": 15020
    },
    {
      "epoch": 2.237861185582365,
      "grad_norm": 14.375,
      "learning_rate": 4.997848677341474e-05,
      "loss": 0.9319,
      "num_input_tokens_seen": 8716480,
      "step": 15025
    },
    {
      "epoch": 2.2386058981233243,
      "grad_norm": 15.75,
      "learning_rate": 4.997835178710504e-05,
      "loss": 1.1757,
      "num_input_tokens_seen": 8719200,
      "step": 15030
    },
    {
      "epoch": 2.2393506106642835,
      "grad_norm": 8.1875,
      "learning_rate": 4.997821637881147e-05,
      "loss": 0.9812,
      "num_input_tokens_seen": 8722112,
      "step": 15035
    },
    {
      "epoch": 2.2400953232052427,
      "grad_norm": 4.875,
      "learning_rate": 4.997808054853632e-05,
      "loss": 0.9095,
      "num_input_tokens_seen": 8724800,
      "step": 15040
    },
    {
      "epoch": 2.240840035746202,
      "grad_norm": 13.0,
      "learning_rate": 4.9977944296281895e-05,
      "loss": 1.137,
      "num_input_tokens_seen": 8727968,
      "step": 15045
    },
    {
      "epoch": 2.241584748287161,
      "grad_norm": 9.9375,
      "learning_rate": 4.997780762205047e-05,
      "loss": 1.218,
      "num_input_tokens_seen": 8730848,
      "step": 15050
    },
    {
      "epoch": 2.2423294608281203,
      "grad_norm": 6.03125,
      "learning_rate": 4.997767052584439e-05,
      "loss": 1.2982,
      "num_input_tokens_seen": 8733728,
      "step": 15055
    },
    {
      "epoch": 2.2430741733690795,
      "grad_norm": 14.375,
      "learning_rate": 4.9977533007665944e-05,
      "loss": 1.0647,
      "num_input_tokens_seen": 8736704,
      "step": 15060
    },
    {
      "epoch": 2.2438188859100388,
      "grad_norm": 11.25,
      "learning_rate": 4.9977395067517464e-05,
      "loss": 0.9614,
      "num_input_tokens_seen": 8739392,
      "step": 15065
    },
    {
      "epoch": 2.244563598450998,
      "grad_norm": 8.25,
      "learning_rate": 4.997725670540128e-05,
      "loss": 1.0869,
      "num_input_tokens_seen": 8742528,
      "step": 15070
    },
    {
      "epoch": 2.245308310991957,
      "grad_norm": 8.25,
      "learning_rate": 4.997711792131973e-05,
      "loss": 1.2944,
      "num_input_tokens_seen": 8745504,
      "step": 15075
    },
    {
      "epoch": 2.2460530235329164,
      "grad_norm": 13.5,
      "learning_rate": 4.9976978715275155e-05,
      "loss": 1.3217,
      "num_input_tokens_seen": 8748128,
      "step": 15080
    },
    {
      "epoch": 2.2467977360738756,
      "grad_norm": 6.78125,
      "learning_rate": 4.997683908726991e-05,
      "loss": 0.7599,
      "num_input_tokens_seen": 8750880,
      "step": 15085
    },
    {
      "epoch": 2.2475424486148348,
      "grad_norm": 9.25,
      "learning_rate": 4.9976699037306356e-05,
      "loss": 0.9894,
      "num_input_tokens_seen": 8753792,
      "step": 15090
    },
    {
      "epoch": 2.248287161155794,
      "grad_norm": 10.75,
      "learning_rate": 4.997655856538686e-05,
      "loss": 0.9065,
      "num_input_tokens_seen": 8756608,
      "step": 15095
    },
    {
      "epoch": 2.249031873696753,
      "grad_norm": 5.03125,
      "learning_rate": 4.9976417671513787e-05,
      "loss": 1.1211,
      "num_input_tokens_seen": 8759456,
      "step": 15100
    },
    {
      "epoch": 2.2497765862377124,
      "grad_norm": 12.1875,
      "learning_rate": 4.997627635568953e-05,
      "loss": 1.2529,
      "num_input_tokens_seen": 8762208,
      "step": 15105
    },
    {
      "epoch": 2.2505212987786716,
      "grad_norm": 10.8125,
      "learning_rate": 4.997613461791646e-05,
      "loss": 0.9582,
      "num_input_tokens_seen": 8765152,
      "step": 15110
    },
    {
      "epoch": 2.2512660113196308,
      "grad_norm": 9.3125,
      "learning_rate": 4.9975992458196986e-05,
      "loss": 0.982,
      "num_input_tokens_seen": 8768032,
      "step": 15115
    },
    {
      "epoch": 2.25201072386059,
      "grad_norm": 9.6875,
      "learning_rate": 4.99758498765335e-05,
      "loss": 1.1154,
      "num_input_tokens_seen": 8771104,
      "step": 15120
    },
    {
      "epoch": 2.252755436401549,
      "grad_norm": 9.1875,
      "learning_rate": 4.997570687292842e-05,
      "loss": 1.4598,
      "num_input_tokens_seen": 8773856,
      "step": 15125
    },
    {
      "epoch": 2.2535001489425084,
      "grad_norm": 22.0,
      "learning_rate": 4.9975563447384156e-05,
      "loss": 1.17,
      "num_input_tokens_seen": 8776512,
      "step": 15130
    },
    {
      "epoch": 2.2542448614834676,
      "grad_norm": 10.5625,
      "learning_rate": 4.997541959990313e-05,
      "loss": 0.8745,
      "num_input_tokens_seen": 8779328,
      "step": 15135
    },
    {
      "epoch": 2.2549895740244263,
      "grad_norm": 12.875,
      "learning_rate": 4.997527533048777e-05,
      "loss": 1.2317,
      "num_input_tokens_seen": 8782400,
      "step": 15140
    },
    {
      "epoch": 2.255734286565386,
      "grad_norm": 8.9375,
      "learning_rate": 4.997513063914052e-05,
      "loss": 0.9372,
      "num_input_tokens_seen": 8785504,
      "step": 15145
    },
    {
      "epoch": 2.2564789991063448,
      "grad_norm": 10.1875,
      "learning_rate": 4.997498552586382e-05,
      "loss": 1.323,
      "num_input_tokens_seen": 8788480,
      "step": 15150
    },
    {
      "epoch": 2.257223711647304,
      "grad_norm": 13.4375,
      "learning_rate": 4.9974839990660124e-05,
      "loss": 1.2281,
      "num_input_tokens_seen": 8791072,
      "step": 15155
    },
    {
      "epoch": 2.257968424188263,
      "grad_norm": 8.625,
      "learning_rate": 4.997469403353189e-05,
      "loss": 1.3427,
      "num_input_tokens_seen": 8793920,
      "step": 15160
    },
    {
      "epoch": 2.2587131367292224,
      "grad_norm": 10.9375,
      "learning_rate": 4.9974547654481585e-05,
      "loss": 1.4443,
      "num_input_tokens_seen": 8796704,
      "step": 15165
    },
    {
      "epoch": 2.2594578492701816,
      "grad_norm": 14.4375,
      "learning_rate": 4.997440085351168e-05,
      "loss": 1.1884,
      "num_input_tokens_seen": 8799360,
      "step": 15170
    },
    {
      "epoch": 2.2602025618111408,
      "grad_norm": 6.34375,
      "learning_rate": 4.9974253630624654e-05,
      "loss": 1.0375,
      "num_input_tokens_seen": 8802016,
      "step": 15175
    },
    {
      "epoch": 2.2609472743521,
      "grad_norm": 7.5625,
      "learning_rate": 4.9974105985822996e-05,
      "loss": 1.109,
      "num_input_tokens_seen": 8804800,
      "step": 15180
    },
    {
      "epoch": 2.261691986893059,
      "grad_norm": 11.3125,
      "learning_rate": 4.997395791910919e-05,
      "loss": 1.4975,
      "num_input_tokens_seen": 8807968,
      "step": 15185
    },
    {
      "epoch": 2.2624366994340184,
      "grad_norm": 5.71875,
      "learning_rate": 4.997380943048576e-05,
      "loss": 1.1182,
      "num_input_tokens_seen": 8811168,
      "step": 15190
    },
    {
      "epoch": 2.2631814119749776,
      "grad_norm": 13.8125,
      "learning_rate": 4.99736605199552e-05,
      "loss": 1.541,
      "num_input_tokens_seen": 8813824,
      "step": 15195
    },
    {
      "epoch": 2.2639261245159368,
      "grad_norm": 10.375,
      "learning_rate": 4.9973511187520025e-05,
      "loss": 1.1083,
      "num_input_tokens_seen": 8816640,
      "step": 15200
    },
    {
      "epoch": 2.264670837056896,
      "grad_norm": 10.25,
      "learning_rate": 4.9973361433182764e-05,
      "loss": 1.3686,
      "num_input_tokens_seen": 8819616,
      "step": 15205
    },
    {
      "epoch": 2.265415549597855,
      "grad_norm": 25.875,
      "learning_rate": 4.997321125694594e-05,
      "loss": 1.2953,
      "num_input_tokens_seen": 8822560,
      "step": 15210
    },
    {
      "epoch": 2.2661602621388144,
      "grad_norm": 19.625,
      "learning_rate": 4.99730606588121e-05,
      "loss": 1.5025,
      "num_input_tokens_seen": 8825280,
      "step": 15215
    },
    {
      "epoch": 2.2669049746797736,
      "grad_norm": 11.4375,
      "learning_rate": 4.997290963878377e-05,
      "loss": 1.2326,
      "num_input_tokens_seen": 8828160,
      "step": 15220
    },
    {
      "epoch": 2.267649687220733,
      "grad_norm": 13.5625,
      "learning_rate": 4.9972758196863524e-05,
      "loss": 1.3705,
      "num_input_tokens_seen": 8830848,
      "step": 15225
    },
    {
      "epoch": 2.268394399761692,
      "grad_norm": 8.9375,
      "learning_rate": 4.9972606333053903e-05,
      "loss": 0.9617,
      "num_input_tokens_seen": 8833920,
      "step": 15230
    },
    {
      "epoch": 2.269139112302651,
      "grad_norm": 9.4375,
      "learning_rate": 4.997245404735748e-05,
      "loss": 0.8863,
      "num_input_tokens_seen": 8836800,
      "step": 15235
    },
    {
      "epoch": 2.2698838248436104,
      "grad_norm": 17.125,
      "learning_rate": 4.997230133977683e-05,
      "loss": 1.2041,
      "num_input_tokens_seen": 8840032,
      "step": 15240
    },
    {
      "epoch": 2.2706285373845696,
      "grad_norm": 9.625,
      "learning_rate": 4.997214821031453e-05,
      "loss": 1.4369,
      "num_input_tokens_seen": 8842848,
      "step": 15245
    },
    {
      "epoch": 2.271373249925529,
      "grad_norm": 6.0625,
      "learning_rate": 4.997199465897316e-05,
      "loss": 0.8942,
      "num_input_tokens_seen": 8845632,
      "step": 15250
    },
    {
      "epoch": 2.272117962466488,
      "grad_norm": 7.09375,
      "learning_rate": 4.9971840685755324e-05,
      "loss": 0.7889,
      "num_input_tokens_seen": 8848128,
      "step": 15255
    },
    {
      "epoch": 2.272862675007447,
      "grad_norm": 7.90625,
      "learning_rate": 4.997168629066362e-05,
      "loss": 1.2471,
      "num_input_tokens_seen": 8851328,
      "step": 15260
    },
    {
      "epoch": 2.2736073875484064,
      "grad_norm": 15.1875,
      "learning_rate": 4.9971531473700654e-05,
      "loss": 1.3637,
      "num_input_tokens_seen": 8854368,
      "step": 15265
    },
    {
      "epoch": 2.2743521000893656,
      "grad_norm": 13.5625,
      "learning_rate": 4.997137623486905e-05,
      "loss": 0.9556,
      "num_input_tokens_seen": 8857056,
      "step": 15270
    },
    {
      "epoch": 2.275096812630325,
      "grad_norm": 17.0,
      "learning_rate": 4.9971220574171415e-05,
      "loss": 1.0358,
      "num_input_tokens_seen": 8859872,
      "step": 15275
    },
    {
      "epoch": 2.275841525171284,
      "grad_norm": 11.8125,
      "learning_rate": 4.9971064491610396e-05,
      "loss": 1.1671,
      "num_input_tokens_seen": 8862848,
      "step": 15280
    },
    {
      "epoch": 2.276586237712243,
      "grad_norm": 12.375,
      "learning_rate": 4.997090798718862e-05,
      "loss": 1.13,
      "num_input_tokens_seen": 8866080,
      "step": 15285
    },
    {
      "epoch": 2.2773309502532024,
      "grad_norm": 10.0625,
      "learning_rate": 4.9970751060908735e-05,
      "loss": 1.0146,
      "num_input_tokens_seen": 8869056,
      "step": 15290
    },
    {
      "epoch": 2.2780756627941616,
      "grad_norm": 11.6875,
      "learning_rate": 4.997059371277339e-05,
      "loss": 1.0064,
      "num_input_tokens_seen": 8871808,
      "step": 15295
    },
    {
      "epoch": 2.278820375335121,
      "grad_norm": 8.8125,
      "learning_rate": 4.997043594278523e-05,
      "loss": 1.1272,
      "num_input_tokens_seen": 8874752,
      "step": 15300
    },
    {
      "epoch": 2.2795650878760796,
      "grad_norm": 9.0,
      "learning_rate": 4.997027775094695e-05,
      "loss": 1.5847,
      "num_input_tokens_seen": 8877440,
      "step": 15305
    },
    {
      "epoch": 2.2803098004170392,
      "grad_norm": 4.4375,
      "learning_rate": 4.99701191372612e-05,
      "loss": 0.777,
      "num_input_tokens_seen": 8880192,
      "step": 15310
    },
    {
      "epoch": 2.281054512957998,
      "grad_norm": 13.25,
      "learning_rate": 4.9969960101730664e-05,
      "loss": 0.971,
      "num_input_tokens_seen": 8882624,
      "step": 15315
    },
    {
      "epoch": 2.2817992254989576,
      "grad_norm": 25.625,
      "learning_rate": 4.996980064435803e-05,
      "loss": 1.1163,
      "num_input_tokens_seen": 8885472,
      "step": 15320
    },
    {
      "epoch": 2.2825439380399164,
      "grad_norm": 6.0625,
      "learning_rate": 4.9969640765145996e-05,
      "loss": 1.2753,
      "num_input_tokens_seen": 8888448,
      "step": 15325
    },
    {
      "epoch": 2.2832886505808756,
      "grad_norm": 7.8125,
      "learning_rate": 4.9969480464097255e-05,
      "loss": 0.9157,
      "num_input_tokens_seen": 8891200,
      "step": 15330
    },
    {
      "epoch": 2.284033363121835,
      "grad_norm": 11.0,
      "learning_rate": 4.9969319741214525e-05,
      "loss": 1.1655,
      "num_input_tokens_seen": 8894144,
      "step": 15335
    },
    {
      "epoch": 2.284778075662794,
      "grad_norm": 17.375,
      "learning_rate": 4.996915859650051e-05,
      "loss": 1.6436,
      "num_input_tokens_seen": 8896864,
      "step": 15340
    },
    {
      "epoch": 2.285522788203753,
      "grad_norm": 20.875,
      "learning_rate": 4.996899702995794e-05,
      "loss": 1.1883,
      "num_input_tokens_seen": 8899488,
      "step": 15345
    },
    {
      "epoch": 2.2862675007447124,
      "grad_norm": 10.75,
      "learning_rate": 4.9968835041589546e-05,
      "loss": 1.0607,
      "num_input_tokens_seen": 8902368,
      "step": 15350
    },
    {
      "epoch": 2.2870122132856716,
      "grad_norm": 6.1875,
      "learning_rate": 4.996867263139806e-05,
      "loss": 0.8732,
      "num_input_tokens_seen": 8905312,
      "step": 15355
    },
    {
      "epoch": 2.287756925826631,
      "grad_norm": 12.625,
      "learning_rate": 4.996850979938622e-05,
      "loss": 1.0781,
      "num_input_tokens_seen": 8908192,
      "step": 15360
    },
    {
      "epoch": 2.28850163836759,
      "grad_norm": 11.1875,
      "learning_rate": 4.996834654555679e-05,
      "loss": 1.2048,
      "num_input_tokens_seen": 8911040,
      "step": 15365
    },
    {
      "epoch": 2.289246350908549,
      "grad_norm": 8.75,
      "learning_rate": 4.9968182869912525e-05,
      "loss": 0.9603,
      "num_input_tokens_seen": 8913664,
      "step": 15370
    },
    {
      "epoch": 2.2899910634495084,
      "grad_norm": 14.4375,
      "learning_rate": 4.9968018772456185e-05,
      "loss": 1.3106,
      "num_input_tokens_seen": 8916736,
      "step": 15375
    },
    {
      "epoch": 2.2907357759904676,
      "grad_norm": 16.25,
      "learning_rate": 4.9967854253190536e-05,
      "loss": 1.1596,
      "num_input_tokens_seen": 8919264,
      "step": 15380
    },
    {
      "epoch": 2.291480488531427,
      "grad_norm": 11.0,
      "learning_rate": 4.996768931211837e-05,
      "loss": 1.0174,
      "num_input_tokens_seen": 8922368,
      "step": 15385
    },
    {
      "epoch": 2.292225201072386,
      "grad_norm": 11.8125,
      "learning_rate": 4.996752394924247e-05,
      "loss": 1.0157,
      "num_input_tokens_seen": 8925248,
      "step": 15390
    },
    {
      "epoch": 2.2929699136133452,
      "grad_norm": 3.71875,
      "learning_rate": 4.996735816456564e-05,
      "loss": 0.969,
      "num_input_tokens_seen": 8928160,
      "step": 15395
    },
    {
      "epoch": 2.2937146261543044,
      "grad_norm": 6.53125,
      "learning_rate": 4.9967191958090656e-05,
      "loss": 0.8736,
      "num_input_tokens_seen": 8931104,
      "step": 15400
    },
    {
      "epoch": 2.2944593386952636,
      "grad_norm": 9.375,
      "learning_rate": 4.996702532982034e-05,
      "loss": 0.9896,
      "num_input_tokens_seen": 8933984,
      "step": 15405
    },
    {
      "epoch": 2.295204051236223,
      "grad_norm": 14.9375,
      "learning_rate": 4.99668582797575e-05,
      "loss": 1.0376,
      "num_input_tokens_seen": 8936832,
      "step": 15410
    },
    {
      "epoch": 2.295948763777182,
      "grad_norm": 5.65625,
      "learning_rate": 4.996669080790498e-05,
      "loss": 0.8919,
      "num_input_tokens_seen": 8939616,
      "step": 15415
    },
    {
      "epoch": 2.2966934763181412,
      "grad_norm": 6.375,
      "learning_rate": 4.996652291426559e-05,
      "loss": 1.0544,
      "num_input_tokens_seen": 8942720,
      "step": 15420
    },
    {
      "epoch": 2.2974381888591004,
      "grad_norm": 8.875,
      "learning_rate": 4.996635459884216e-05,
      "loss": 1.152,
      "num_input_tokens_seen": 8945504,
      "step": 15425
    },
    {
      "epoch": 2.2981829014000597,
      "grad_norm": 18.0,
      "learning_rate": 4.996618586163755e-05,
      "loss": 1.2021,
      "num_input_tokens_seen": 8948160,
      "step": 15430
    },
    {
      "epoch": 2.298927613941019,
      "grad_norm": 12.75,
      "learning_rate": 4.996601670265461e-05,
      "loss": 1.265,
      "num_input_tokens_seen": 8950912,
      "step": 15435
    },
    {
      "epoch": 2.299672326481978,
      "grad_norm": 12.375,
      "learning_rate": 4.996584712189618e-05,
      "loss": 0.9586,
      "num_input_tokens_seen": 8954048,
      "step": 15440
    },
    {
      "epoch": 2.3004170390229373,
      "grad_norm": 13.25,
      "learning_rate": 4.996567711936515e-05,
      "loss": 1.0646,
      "num_input_tokens_seen": 8956832,
      "step": 15445
    },
    {
      "epoch": 2.3011617515638965,
      "grad_norm": 5.1875,
      "learning_rate": 4.996550669506438e-05,
      "loss": 0.9341,
      "num_input_tokens_seen": 8959808,
      "step": 15450
    },
    {
      "epoch": 2.3019064641048557,
      "grad_norm": 14.875,
      "learning_rate": 4.996533584899674e-05,
      "loss": 1.1794,
      "num_input_tokens_seen": 8962752,
      "step": 15455
    },
    {
      "epoch": 2.302651176645815,
      "grad_norm": 14.125,
      "learning_rate": 4.996516458116512e-05,
      "loss": 1.2664,
      "num_input_tokens_seen": 8965824,
      "step": 15460
    },
    {
      "epoch": 2.303395889186774,
      "grad_norm": 9.0625,
      "learning_rate": 4.9964992891572425e-05,
      "loss": 0.929,
      "num_input_tokens_seen": 8968608,
      "step": 15465
    },
    {
      "epoch": 2.3041406017277333,
      "grad_norm": 8.1875,
      "learning_rate": 4.996482078022155e-05,
      "loss": 1.2259,
      "num_input_tokens_seen": 8971488,
      "step": 15470
    },
    {
      "epoch": 2.3048853142686925,
      "grad_norm": 13.9375,
      "learning_rate": 4.9964648247115395e-05,
      "loss": 0.9089,
      "num_input_tokens_seen": 8974368,
      "step": 15475
    },
    {
      "epoch": 2.3056300268096512,
      "grad_norm": 10.6875,
      "learning_rate": 4.9964475292256884e-05,
      "loss": 1.0451,
      "num_input_tokens_seen": 8977280,
      "step": 15480
    },
    {
      "epoch": 2.306374739350611,
      "grad_norm": 17.0,
      "learning_rate": 4.996430191564894e-05,
      "loss": 1.0091,
      "num_input_tokens_seen": 8980256,
      "step": 15485
    },
    {
      "epoch": 2.3071194518915696,
      "grad_norm": 12.75,
      "learning_rate": 4.996412811729448e-05,
      "loss": 1.1773,
      "num_input_tokens_seen": 8982912,
      "step": 15490
    },
    {
      "epoch": 2.3078641644325293,
      "grad_norm": 9.5625,
      "learning_rate": 4.996395389719646e-05,
      "loss": 1.0079,
      "num_input_tokens_seen": 8985824,
      "step": 15495
    },
    {
      "epoch": 2.308608876973488,
      "grad_norm": 8.6875,
      "learning_rate": 4.99637792553578e-05,
      "loss": 1.5381,
      "num_input_tokens_seen": 8988672,
      "step": 15500
    },
    {
      "epoch": 2.3093535895144472,
      "grad_norm": 18.125,
      "learning_rate": 4.996360419178147e-05,
      "loss": 1.3176,
      "num_input_tokens_seen": 8991424,
      "step": 15505
    },
    {
      "epoch": 2.3100983020554064,
      "grad_norm": 4.0625,
      "learning_rate": 4.9963428706470405e-05,
      "loss": 1.0127,
      "num_input_tokens_seen": 8994400,
      "step": 15510
    },
    {
      "epoch": 2.3108430145963657,
      "grad_norm": 8.8125,
      "learning_rate": 4.9963252799427594e-05,
      "loss": 0.9753,
      "num_input_tokens_seen": 8997152,
      "step": 15515
    },
    {
      "epoch": 2.311587727137325,
      "grad_norm": 6.3125,
      "learning_rate": 4.9963076470655995e-05,
      "loss": 1.4641,
      "num_input_tokens_seen": 8999840,
      "step": 15520
    },
    {
      "epoch": 2.312332439678284,
      "grad_norm": 12.8125,
      "learning_rate": 4.996289972015859e-05,
      "loss": 1.0442,
      "num_input_tokens_seen": 9002720,
      "step": 15525
    },
    {
      "epoch": 2.3130771522192433,
      "grad_norm": 15.0,
      "learning_rate": 4.9962722547938365e-05,
      "loss": 1.2296,
      "num_input_tokens_seen": 9005856,
      "step": 15530
    },
    {
      "epoch": 2.3138218647602025,
      "grad_norm": 10.6875,
      "learning_rate": 4.9962544953998316e-05,
      "loss": 1.3369,
      "num_input_tokens_seen": 9008928,
      "step": 15535
    },
    {
      "epoch": 2.3145665773011617,
      "grad_norm": 8.5625,
      "learning_rate": 4.996236693834144e-05,
      "loss": 1.0273,
      "num_input_tokens_seen": 9011680,
      "step": 15540
    },
    {
      "epoch": 2.315311289842121,
      "grad_norm": 12.75,
      "learning_rate": 4.996218850097075e-05,
      "loss": 0.9439,
      "num_input_tokens_seen": 9014752,
      "step": 15545
    },
    {
      "epoch": 2.31605600238308,
      "grad_norm": 4.5,
      "learning_rate": 4.996200964188925e-05,
      "loss": 0.8033,
      "num_input_tokens_seen": 9017984,
      "step": 15550
    },
    {
      "epoch": 2.3168007149240393,
      "grad_norm": 7.0,
      "learning_rate": 4.996183036109997e-05,
      "loss": 1.1056,
      "num_input_tokens_seen": 9020992,
      "step": 15555
    },
    {
      "epoch": 2.3175454274649985,
      "grad_norm": 7.15625,
      "learning_rate": 4.996165065860594e-05,
      "loss": 1.4473,
      "num_input_tokens_seen": 9023936,
      "step": 15560
    },
    {
      "epoch": 2.3182901400059577,
      "grad_norm": 11.9375,
      "learning_rate": 4.996147053441018e-05,
      "loss": 1.087,
      "num_input_tokens_seen": 9026784,
      "step": 15565
    },
    {
      "epoch": 2.319034852546917,
      "grad_norm": 9.1875,
      "learning_rate": 4.996128998851575e-05,
      "loss": 0.8899,
      "num_input_tokens_seen": 9029696,
      "step": 15570
    },
    {
      "epoch": 2.319779565087876,
      "grad_norm": 20.25,
      "learning_rate": 4.99611090209257e-05,
      "loss": 1.1355,
      "num_input_tokens_seen": 9032736,
      "step": 15575
    },
    {
      "epoch": 2.3205242776288353,
      "grad_norm": 17.0,
      "learning_rate": 4.9960927631643086e-05,
      "loss": 1.0747,
      "num_input_tokens_seen": 9035680,
      "step": 15580
    },
    {
      "epoch": 2.3212689901697945,
      "grad_norm": 7.625,
      "learning_rate": 4.996074582067096e-05,
      "loss": 1.0105,
      "num_input_tokens_seen": 9038464,
      "step": 15585
    },
    {
      "epoch": 2.3220137027107537,
      "grad_norm": 11.625,
      "learning_rate": 4.9960563588012396e-05,
      "loss": 1.0284,
      "num_input_tokens_seen": 9041312,
      "step": 15590
    },
    {
      "epoch": 2.322758415251713,
      "grad_norm": 9.75,
      "learning_rate": 4.9960380933670495e-05,
      "loss": 0.8627,
      "num_input_tokens_seen": 9043968,
      "step": 15595
    },
    {
      "epoch": 2.323503127792672,
      "grad_norm": 10.4375,
      "learning_rate": 4.996019785764832e-05,
      "loss": 1.3822,
      "num_input_tokens_seen": 9046784,
      "step": 15600
    },
    {
      "epoch": 2.3242478403336313,
      "grad_norm": 12.3125,
      "learning_rate": 4.996001435994897e-05,
      "loss": 0.9204,
      "num_input_tokens_seen": 9049504,
      "step": 15605
    },
    {
      "epoch": 2.3249925528745905,
      "grad_norm": 23.25,
      "learning_rate": 4.995983044057554e-05,
      "loss": 0.9467,
      "num_input_tokens_seen": 9052544,
      "step": 15610
    },
    {
      "epoch": 2.3257372654155497,
      "grad_norm": 12.125,
      "learning_rate": 4.9959646099531156e-05,
      "loss": 1.3112,
      "num_input_tokens_seen": 9055424,
      "step": 15615
    },
    {
      "epoch": 2.326481977956509,
      "grad_norm": 7.625,
      "learning_rate": 4.99594613368189e-05,
      "loss": 1.2808,
      "num_input_tokens_seen": 9058528,
      "step": 15620
    },
    {
      "epoch": 2.327226690497468,
      "grad_norm": 13.0,
      "learning_rate": 4.995927615244193e-05,
      "loss": 0.9975,
      "num_input_tokens_seen": 9061280,
      "step": 15625
    },
    {
      "epoch": 2.3279714030384273,
      "grad_norm": 14.3125,
      "learning_rate": 4.9959090546403356e-05,
      "loss": 1.1683,
      "num_input_tokens_seen": 9064160,
      "step": 15630
    },
    {
      "epoch": 2.3287161155793865,
      "grad_norm": 12.875,
      "learning_rate": 4.9958904518706305e-05,
      "loss": 1.276,
      "num_input_tokens_seen": 9067328,
      "step": 15635
    },
    {
      "epoch": 2.3294608281203457,
      "grad_norm": 10.375,
      "learning_rate": 4.9958718069353935e-05,
      "loss": 0.8913,
      "num_input_tokens_seen": 9070112,
      "step": 15640
    },
    {
      "epoch": 2.330205540661305,
      "grad_norm": 5.0,
      "learning_rate": 4.9958531198349384e-05,
      "loss": 0.9709,
      "num_input_tokens_seen": 9072800,
      "step": 15645
    },
    {
      "epoch": 2.330950253202264,
      "grad_norm": 6.96875,
      "learning_rate": 4.9958343905695823e-05,
      "loss": 1.0676,
      "num_input_tokens_seen": 9075776,
      "step": 15650
    },
    {
      "epoch": 2.331694965743223,
      "grad_norm": 8.875,
      "learning_rate": 4.99581561913964e-05,
      "loss": 1.1703,
      "num_input_tokens_seen": 9078944,
      "step": 15655
    },
    {
      "epoch": 2.3324396782841825,
      "grad_norm": 9.5,
      "learning_rate": 4.99579680554543e-05,
      "loss": 1.3644,
      "num_input_tokens_seen": 9081984,
      "step": 15660
    },
    {
      "epoch": 2.3331843908251413,
      "grad_norm": 9.6875,
      "learning_rate": 4.99577794978727e-05,
      "loss": 1.0773,
      "num_input_tokens_seen": 9084992,
      "step": 15665
    },
    {
      "epoch": 2.333929103366101,
      "grad_norm": 8.5,
      "learning_rate": 4.995759051865477e-05,
      "loss": 0.9799,
      "num_input_tokens_seen": 9088032,
      "step": 15670
    },
    {
      "epoch": 2.3346738159070597,
      "grad_norm": 6.6875,
      "learning_rate": 4.995740111780372e-05,
      "loss": 0.9571,
      "num_input_tokens_seen": 9090912,
      "step": 15675
    },
    {
      "epoch": 2.335418528448019,
      "grad_norm": 17.375,
      "learning_rate": 4.995721129532275e-05,
      "loss": 1.5103,
      "num_input_tokens_seen": 9093760,
      "step": 15680
    },
    {
      "epoch": 2.336163240988978,
      "grad_norm": 7.84375,
      "learning_rate": 4.9957021051215055e-05,
      "loss": 1.0454,
      "num_input_tokens_seen": 9096640,
      "step": 15685
    },
    {
      "epoch": 2.3369079535299373,
      "grad_norm": 8.0625,
      "learning_rate": 4.995683038548385e-05,
      "loss": 1.0085,
      "num_input_tokens_seen": 9099648,
      "step": 15690
    },
    {
      "epoch": 2.3376526660708965,
      "grad_norm": 16.875,
      "learning_rate": 4.995663929813237e-05,
      "loss": 1.0001,
      "num_input_tokens_seen": 9102432,
      "step": 15695
    },
    {
      "epoch": 2.3383973786118557,
      "grad_norm": 12.125,
      "learning_rate": 4.995644778916383e-05,
      "loss": 0.9671,
      "num_input_tokens_seen": 9105088,
      "step": 15700
    },
    {
      "epoch": 2.339142091152815,
      "grad_norm": 5.46875,
      "learning_rate": 4.995625585858146e-05,
      "loss": 0.8758,
      "num_input_tokens_seen": 9107776,
      "step": 15705
    },
    {
      "epoch": 2.339886803693774,
      "grad_norm": 13.6875,
      "learning_rate": 4.9956063506388524e-05,
      "loss": 1.0611,
      "num_input_tokens_seen": 9110656,
      "step": 15710
    },
    {
      "epoch": 2.3406315162347333,
      "grad_norm": 11.125,
      "learning_rate": 4.995587073258825e-05,
      "loss": 1.3616,
      "num_input_tokens_seen": 9113536,
      "step": 15715
    },
    {
      "epoch": 2.3413762287756925,
      "grad_norm": 12.625,
      "learning_rate": 4.995567753718391e-05,
      "loss": 1.2419,
      "num_input_tokens_seen": 9116576,
      "step": 15720
    },
    {
      "epoch": 2.3421209413166517,
      "grad_norm": 10.375,
      "learning_rate": 4.995548392017876e-05,
      "loss": 1.0847,
      "num_input_tokens_seen": 9119808,
      "step": 15725
    },
    {
      "epoch": 2.342865653857611,
      "grad_norm": 15.125,
      "learning_rate": 4.995528988157608e-05,
      "loss": 1.1213,
      "num_input_tokens_seen": 9122816,
      "step": 15730
    },
    {
      "epoch": 2.34361036639857,
      "grad_norm": 4.5,
      "learning_rate": 4.995509542137913e-05,
      "loss": 1.006,
      "num_input_tokens_seen": 9125408,
      "step": 15735
    },
    {
      "epoch": 2.3443550789395293,
      "grad_norm": 10.25,
      "learning_rate": 4.995490053959121e-05,
      "loss": 1.3413,
      "num_input_tokens_seen": 9128384,
      "step": 15740
    },
    {
      "epoch": 2.3450997914804885,
      "grad_norm": 8.875,
      "learning_rate": 4.995470523621561e-05,
      "loss": 1.1584,
      "num_input_tokens_seen": 9131104,
      "step": 15745
    },
    {
      "epoch": 2.3458445040214477,
      "grad_norm": 9.25,
      "learning_rate": 4.9954509511255625e-05,
      "loss": 1.2376,
      "num_input_tokens_seen": 9133792,
      "step": 15750
    },
    {
      "epoch": 2.346589216562407,
      "grad_norm": 9.875,
      "learning_rate": 4.9954313364714565e-05,
      "loss": 0.915,
      "num_input_tokens_seen": 9136800,
      "step": 15755
    },
    {
      "epoch": 2.347333929103366,
      "grad_norm": 12.9375,
      "learning_rate": 4.9954116796595754e-05,
      "loss": 1.0171,
      "num_input_tokens_seen": 9139840,
      "step": 15760
    },
    {
      "epoch": 2.3480786416443253,
      "grad_norm": 9.5,
      "learning_rate": 4.9953919806902486e-05,
      "loss": 0.9686,
      "num_input_tokens_seen": 9142656,
      "step": 15765
    },
    {
      "epoch": 2.3488233541852845,
      "grad_norm": 7.65625,
      "learning_rate": 4.9953722395638115e-05,
      "loss": 1.1476,
      "num_input_tokens_seen": 9145600,
      "step": 15770
    },
    {
      "epoch": 2.3495680667262437,
      "grad_norm": 9.8125,
      "learning_rate": 4.995352456280596e-05,
      "loss": 1.2368,
      "num_input_tokens_seen": 9148672,
      "step": 15775
    },
    {
      "epoch": 2.350312779267203,
      "grad_norm": 10.0625,
      "learning_rate": 4.9953326308409364e-05,
      "loss": 0.9541,
      "num_input_tokens_seen": 9151680,
      "step": 15780
    },
    {
      "epoch": 2.351057491808162,
      "grad_norm": 10.6875,
      "learning_rate": 4.9953127632451694e-05,
      "loss": 0.9121,
      "num_input_tokens_seen": 9154848,
      "step": 15785
    },
    {
      "epoch": 2.3518022043491214,
      "grad_norm": 12.3125,
      "learning_rate": 4.995292853493629e-05,
      "loss": 1.0724,
      "num_input_tokens_seen": 9157536,
      "step": 15790
    },
    {
      "epoch": 2.3525469168900806,
      "grad_norm": 5.84375,
      "learning_rate": 4.995272901586652e-05,
      "loss": 1.247,
      "num_input_tokens_seen": 9160576,
      "step": 15795
    },
    {
      "epoch": 2.3532916294310398,
      "grad_norm": 9.5625,
      "learning_rate": 4.9952529075245744e-05,
      "loss": 0.9522,
      "num_input_tokens_seen": 9163680,
      "step": 15800
    },
    {
      "epoch": 2.354036341971999,
      "grad_norm": 10.9375,
      "learning_rate": 4.995232871307736e-05,
      "loss": 1.0806,
      "num_input_tokens_seen": 9166912,
      "step": 15805
    },
    {
      "epoch": 2.354781054512958,
      "grad_norm": 8.8125,
      "learning_rate": 4.9952127929364746e-05,
      "loss": 1.0,
      "num_input_tokens_seen": 9170048,
      "step": 15810
    },
    {
      "epoch": 2.3555257670539174,
      "grad_norm": 11.25,
      "learning_rate": 4.995192672411128e-05,
      "loss": 1.0446,
      "num_input_tokens_seen": 9173152,
      "step": 15815
    },
    {
      "epoch": 2.3562704795948766,
      "grad_norm": 5.09375,
      "learning_rate": 4.995172509732038e-05,
      "loss": 1.2427,
      "num_input_tokens_seen": 9176064,
      "step": 15820
    },
    {
      "epoch": 2.3570151921358358,
      "grad_norm": 9.75,
      "learning_rate": 4.995152304899544e-05,
      "loss": 1.0589,
      "num_input_tokens_seen": 9178656,
      "step": 15825
    },
    {
      "epoch": 2.3577599046767945,
      "grad_norm": 6.71875,
      "learning_rate": 4.9951320579139884e-05,
      "loss": 1.0602,
      "num_input_tokens_seen": 9181536,
      "step": 15830
    },
    {
      "epoch": 2.358504617217754,
      "grad_norm": 9.6875,
      "learning_rate": 4.995111768775712e-05,
      "loss": 1.1645,
      "num_input_tokens_seen": 9184768,
      "step": 15835
    },
    {
      "epoch": 2.359249329758713,
      "grad_norm": 10.5625,
      "learning_rate": 4.995091437485058e-05,
      "loss": 1.1017,
      "num_input_tokens_seen": 9187616,
      "step": 15840
    },
    {
      "epoch": 2.359994042299672,
      "grad_norm": 13.0625,
      "learning_rate": 4.9950710640423705e-05,
      "loss": 0.9067,
      "num_input_tokens_seen": 9190560,
      "step": 15845
    },
    {
      "epoch": 2.3607387548406313,
      "grad_norm": 15.625,
      "learning_rate": 4.9950506484479934e-05,
      "loss": 1.1436,
      "num_input_tokens_seen": 9193248,
      "step": 15850
    },
    {
      "epoch": 2.3614834673815905,
      "grad_norm": 10.625,
      "learning_rate": 4.995030190702271e-05,
      "loss": 1.2057,
      "num_input_tokens_seen": 9195936,
      "step": 15855
    },
    {
      "epoch": 2.3622281799225497,
      "grad_norm": 6.875,
      "learning_rate": 4.99500969080555e-05,
      "loss": 1.1063,
      "num_input_tokens_seen": 9198560,
      "step": 15860
    },
    {
      "epoch": 2.362972892463509,
      "grad_norm": 20.125,
      "learning_rate": 4.994989148758176e-05,
      "loss": 1.2079,
      "num_input_tokens_seen": 9201344,
      "step": 15865
    },
    {
      "epoch": 2.363717605004468,
      "grad_norm": 11.125,
      "learning_rate": 4.994968564560495e-05,
      "loss": 1.0313,
      "num_input_tokens_seen": 9204448,
      "step": 15870
    },
    {
      "epoch": 2.3644623175454274,
      "grad_norm": 12.3125,
      "learning_rate": 4.994947938212857e-05,
      "loss": 1.196,
      "num_input_tokens_seen": 9207360,
      "step": 15875
    },
    {
      "epoch": 2.3652070300863866,
      "grad_norm": 16.0,
      "learning_rate": 4.994927269715609e-05,
      "loss": 1.0744,
      "num_input_tokens_seen": 9210176,
      "step": 15880
    },
    {
      "epoch": 2.3659517426273458,
      "grad_norm": 4.6875,
      "learning_rate": 4.9949065590691e-05,
      "loss": 0.8841,
      "num_input_tokens_seen": 9213376,
      "step": 15885
    },
    {
      "epoch": 2.366696455168305,
      "grad_norm": 11.625,
      "learning_rate": 4.9948858062736814e-05,
      "loss": 1.2373,
      "num_input_tokens_seen": 9216416,
      "step": 15890
    },
    {
      "epoch": 2.367441167709264,
      "grad_norm": 13.1875,
      "learning_rate": 4.994865011329702e-05,
      "loss": 1.2065,
      "num_input_tokens_seen": 9219424,
      "step": 15895
    },
    {
      "epoch": 2.3681858802502234,
      "grad_norm": 10.5625,
      "learning_rate": 4.994844174237514e-05,
      "loss": 0.98,
      "num_input_tokens_seen": 9222208,
      "step": 15900
    },
    {
      "epoch": 2.3689305927911826,
      "grad_norm": 11.5625,
      "learning_rate": 4.99482329499747e-05,
      "loss": 1.3642,
      "num_input_tokens_seen": 9224960,
      "step": 15905
    },
    {
      "epoch": 2.3696753053321418,
      "grad_norm": 14.1875,
      "learning_rate": 4.994802373609922e-05,
      "loss": 1.344,
      "num_input_tokens_seen": 9227776,
      "step": 15910
    },
    {
      "epoch": 2.370420017873101,
      "grad_norm": 13.6875,
      "learning_rate": 4.9947814100752226e-05,
      "loss": 1.0649,
      "num_input_tokens_seen": 9230432,
      "step": 15915
    },
    {
      "epoch": 2.37116473041406,
      "grad_norm": 9.8125,
      "learning_rate": 4.994760404393727e-05,
      "loss": 1.1933,
      "num_input_tokens_seen": 9233440,
      "step": 15920
    },
    {
      "epoch": 2.3719094429550194,
      "grad_norm": 11.1875,
      "learning_rate": 4.994739356565791e-05,
      "loss": 0.7894,
      "num_input_tokens_seen": 9236288,
      "step": 15925
    },
    {
      "epoch": 2.3726541554959786,
      "grad_norm": 18.5,
      "learning_rate": 4.994718266591768e-05,
      "loss": 1.376,
      "num_input_tokens_seen": 9238976,
      "step": 15930
    },
    {
      "epoch": 2.373398868036938,
      "grad_norm": 8.75,
      "learning_rate": 4.994697134472016e-05,
      "loss": 0.9244,
      "num_input_tokens_seen": 9241888,
      "step": 15935
    },
    {
      "epoch": 2.374143580577897,
      "grad_norm": 20.0,
      "learning_rate": 4.994675960206891e-05,
      "loss": 0.9013,
      "num_input_tokens_seen": 9244736,
      "step": 15940
    },
    {
      "epoch": 2.374888293118856,
      "grad_norm": 20.625,
      "learning_rate": 4.9946547437967515e-05,
      "loss": 1.5907,
      "num_input_tokens_seen": 9247712,
      "step": 15945
    },
    {
      "epoch": 2.3756330056598154,
      "grad_norm": 12.875,
      "learning_rate": 4.9946334852419555e-05,
      "loss": 0.9683,
      "num_input_tokens_seen": 9250816,
      "step": 15950
    },
    {
      "epoch": 2.3763777182007746,
      "grad_norm": 10.8125,
      "learning_rate": 4.9946121845428616e-05,
      "loss": 1.3643,
      "num_input_tokens_seen": 9253536,
      "step": 15955
    },
    {
      "epoch": 2.377122430741734,
      "grad_norm": 11.25,
      "learning_rate": 4.994590841699831e-05,
      "loss": 1.0987,
      "num_input_tokens_seen": 9256544,
      "step": 15960
    },
    {
      "epoch": 2.377867143282693,
      "grad_norm": 8.1875,
      "learning_rate": 4.9945694567132227e-05,
      "loss": 1.0324,
      "num_input_tokens_seen": 9259360,
      "step": 15965
    },
    {
      "epoch": 2.378611855823652,
      "grad_norm": 15.0625,
      "learning_rate": 4.9945480295834e-05,
      "loss": 1.3104,
      "num_input_tokens_seen": 9262208,
      "step": 15970
    },
    {
      "epoch": 2.3793565683646114,
      "grad_norm": 5.53125,
      "learning_rate": 4.994526560310723e-05,
      "loss": 0.9437,
      "num_input_tokens_seen": 9265280,
      "step": 15975
    },
    {
      "epoch": 2.3801012809055706,
      "grad_norm": 16.125,
      "learning_rate": 4.994505048895555e-05,
      "loss": 0.9978,
      "num_input_tokens_seen": 9268000,
      "step": 15980
    },
    {
      "epoch": 2.38084599344653,
      "grad_norm": 17.5,
      "learning_rate": 4.99448349533826e-05,
      "loss": 0.8726,
      "num_input_tokens_seen": 9271008,
      "step": 15985
    },
    {
      "epoch": 2.381590705987489,
      "grad_norm": 8.125,
      "learning_rate": 4.9944618996392014e-05,
      "loss": 1.5122,
      "num_input_tokens_seen": 9274368,
      "step": 15990
    },
    {
      "epoch": 2.382335418528448,
      "grad_norm": 16.5,
      "learning_rate": 4.994440261798743e-05,
      "loss": 1.1801,
      "num_input_tokens_seen": 9277344,
      "step": 15995
    },
    {
      "epoch": 2.3830801310694074,
      "grad_norm": 12.875,
      "learning_rate": 4.994418581817254e-05,
      "loss": 1.4275,
      "num_input_tokens_seen": 9280160,
      "step": 16000
    },
    {
      "epoch": 2.383824843610366,
      "grad_norm": 12.6875,
      "learning_rate": 4.994396859695096e-05,
      "loss": 1.2792,
      "num_input_tokens_seen": 9282912,
      "step": 16005
    },
    {
      "epoch": 2.384569556151326,
      "grad_norm": 21.75,
      "learning_rate": 4.99437509543264e-05,
      "loss": 1.0841,
      "num_input_tokens_seen": 9285600,
      "step": 16010
    },
    {
      "epoch": 2.3853142686922846,
      "grad_norm": 8.0,
      "learning_rate": 4.994353289030251e-05,
      "loss": 1.3087,
      "num_input_tokens_seen": 9288576,
      "step": 16015
    },
    {
      "epoch": 2.386058981233244,
      "grad_norm": 9.625,
      "learning_rate": 4.994331440488298e-05,
      "loss": 1.0398,
      "num_input_tokens_seen": 9291552,
      "step": 16020
    },
    {
      "epoch": 2.386803693774203,
      "grad_norm": 6.3125,
      "learning_rate": 4.994309549807151e-05,
      "loss": 1.0766,
      "num_input_tokens_seen": 9294528,
      "step": 16025
    },
    {
      "epoch": 2.387548406315162,
      "grad_norm": 21.75,
      "learning_rate": 4.9942876169871794e-05,
      "loss": 1.0611,
      "num_input_tokens_seen": 9297440,
      "step": 16030
    },
    {
      "epoch": 2.3882931188561214,
      "grad_norm": 9.9375,
      "learning_rate": 4.9942656420287535e-05,
      "loss": 1.2354,
      "num_input_tokens_seen": 9300256,
      "step": 16035
    },
    {
      "epoch": 2.3890378313970806,
      "grad_norm": 13.5,
      "learning_rate": 4.9942436249322444e-05,
      "loss": 1.3175,
      "num_input_tokens_seen": 9303392,
      "step": 16040
    },
    {
      "epoch": 2.38978254393804,
      "grad_norm": 9.6875,
      "learning_rate": 4.994221565698025e-05,
      "loss": 1.0797,
      "num_input_tokens_seen": 9306432,
      "step": 16045
    },
    {
      "epoch": 2.390527256478999,
      "grad_norm": 3.65625,
      "learning_rate": 4.9941994643264665e-05,
      "loss": 0.8783,
      "num_input_tokens_seen": 9309312,
      "step": 16050
    },
    {
      "epoch": 2.391271969019958,
      "grad_norm": 14.3125,
      "learning_rate": 4.994177320817943e-05,
      "loss": 1.0064,
      "num_input_tokens_seen": 9312544,
      "step": 16055
    },
    {
      "epoch": 2.3920166815609174,
      "grad_norm": 12.3125,
      "learning_rate": 4.9941551351728286e-05,
      "loss": 1.0263,
      "num_input_tokens_seen": 9315520,
      "step": 16060
    },
    {
      "epoch": 2.3927613941018766,
      "grad_norm": 8.1875,
      "learning_rate": 4.994132907391499e-05,
      "loss": 1.1296,
      "num_input_tokens_seen": 9318336,
      "step": 16065
    },
    {
      "epoch": 2.393506106642836,
      "grad_norm": 8.3125,
      "learning_rate": 4.994110637474327e-05,
      "loss": 1.1354,
      "num_input_tokens_seen": 9321600,
      "step": 16070
    },
    {
      "epoch": 2.394250819183795,
      "grad_norm": 8.1875,
      "learning_rate": 4.994088325421693e-05,
      "loss": 0.9115,
      "num_input_tokens_seen": 9324384,
      "step": 16075
    },
    {
      "epoch": 2.394995531724754,
      "grad_norm": 5.25,
      "learning_rate": 4.99406597123397e-05,
      "loss": 0.9319,
      "num_input_tokens_seen": 9327744,
      "step": 16080
    },
    {
      "epoch": 2.3957402442657134,
      "grad_norm": 8.75,
      "learning_rate": 4.994043574911538e-05,
      "loss": 0.892,
      "num_input_tokens_seen": 9330560,
      "step": 16085
    },
    {
      "epoch": 2.3964849568066726,
      "grad_norm": 15.0625,
      "learning_rate": 4.9940211364547744e-05,
      "loss": 1.1997,
      "num_input_tokens_seen": 9333952,
      "step": 16090
    },
    {
      "epoch": 2.397229669347632,
      "grad_norm": 6.28125,
      "learning_rate": 4.9939986558640585e-05,
      "loss": 0.933,
      "num_input_tokens_seen": 9336992,
      "step": 16095
    },
    {
      "epoch": 2.397974381888591,
      "grad_norm": 5.0,
      "learning_rate": 4.99397613313977e-05,
      "loss": 1.5015,
      "num_input_tokens_seen": 9340096,
      "step": 16100
    },
    {
      "epoch": 2.3987190944295502,
      "grad_norm": 17.0,
      "learning_rate": 4.99395356828229e-05,
      "loss": 1.2906,
      "num_input_tokens_seen": 9343104,
      "step": 16105
    },
    {
      "epoch": 2.3994638069705094,
      "grad_norm": 8.25,
      "learning_rate": 4.993930961291999e-05,
      "loss": 1.1555,
      "num_input_tokens_seen": 9346240,
      "step": 16110
    },
    {
      "epoch": 2.4002085195114686,
      "grad_norm": 15.0,
      "learning_rate": 4.993908312169279e-05,
      "loss": 1.1697,
      "num_input_tokens_seen": 9349056,
      "step": 16115
    },
    {
      "epoch": 2.400953232052428,
      "grad_norm": 7.71875,
      "learning_rate": 4.9938856209145123e-05,
      "loss": 1.2494,
      "num_input_tokens_seen": 9352000,
      "step": 16120
    },
    {
      "epoch": 2.401697944593387,
      "grad_norm": 8.0,
      "learning_rate": 4.993862887528083e-05,
      "loss": 0.906,
      "num_input_tokens_seen": 9354880,
      "step": 16125
    },
    {
      "epoch": 2.4024426571343462,
      "grad_norm": 13.125,
      "learning_rate": 4.9938401120103755e-05,
      "loss": 0.8687,
      "num_input_tokens_seen": 9357792,
      "step": 16130
    },
    {
      "epoch": 2.4031873696753054,
      "grad_norm": 8.8125,
      "learning_rate": 4.9938172943617735e-05,
      "loss": 0.7273,
      "num_input_tokens_seen": 9360608,
      "step": 16135
    },
    {
      "epoch": 2.4039320822162646,
      "grad_norm": 24.25,
      "learning_rate": 4.993794434582663e-05,
      "loss": 0.8679,
      "num_input_tokens_seen": 9363744,
      "step": 16140
    },
    {
      "epoch": 2.404676794757224,
      "grad_norm": 6.125,
      "learning_rate": 4.99377153267343e-05,
      "loss": 0.9164,
      "num_input_tokens_seen": 9366912,
      "step": 16145
    },
    {
      "epoch": 2.405421507298183,
      "grad_norm": 11.9375,
      "learning_rate": 4.9937485886344614e-05,
      "loss": 1.1195,
      "num_input_tokens_seen": 9369856,
      "step": 16150
    },
    {
      "epoch": 2.4061662198391423,
      "grad_norm": 12.9375,
      "learning_rate": 4.993725602466145e-05,
      "loss": 1.3924,
      "num_input_tokens_seen": 9372640,
      "step": 16155
    },
    {
      "epoch": 2.4069109323801015,
      "grad_norm": 9.6875,
      "learning_rate": 4.9937025741688694e-05,
      "loss": 1.1698,
      "num_input_tokens_seen": 9375648,
      "step": 16160
    },
    {
      "epoch": 2.4076556449210607,
      "grad_norm": 6.3125,
      "learning_rate": 4.993679503743023e-05,
      "loss": 0.9864,
      "num_input_tokens_seen": 9378720,
      "step": 16165
    },
    {
      "epoch": 2.4084003574620194,
      "grad_norm": 12.125,
      "learning_rate": 4.993656391188995e-05,
      "loss": 0.7491,
      "num_input_tokens_seen": 9381632,
      "step": 16170
    },
    {
      "epoch": 2.409145070002979,
      "grad_norm": 12.75,
      "learning_rate": 4.993633236507178e-05,
      "loss": 1.2186,
      "num_input_tokens_seen": 9384192,
      "step": 16175
    },
    {
      "epoch": 2.409889782543938,
      "grad_norm": 17.5,
      "learning_rate": 4.9936100396979614e-05,
      "loss": 1.0595,
      "num_input_tokens_seen": 9387200,
      "step": 16180
    },
    {
      "epoch": 2.4106344950848975,
      "grad_norm": 7.625,
      "learning_rate": 4.993586800761738e-05,
      "loss": 1.0643,
      "num_input_tokens_seen": 9389888,
      "step": 16185
    },
    {
      "epoch": 2.4113792076258562,
      "grad_norm": 6.78125,
      "learning_rate": 4.9935635196989e-05,
      "loss": 0.8381,
      "num_input_tokens_seen": 9392992,
      "step": 16190
    },
    {
      "epoch": 2.4121239201668154,
      "grad_norm": 9.125,
      "learning_rate": 4.9935401965098395e-05,
      "loss": 0.9402,
      "num_input_tokens_seen": 9395744,
      "step": 16195
    },
    {
      "epoch": 2.4128686327077746,
      "grad_norm": 7.375,
      "learning_rate": 4.9935168311949524e-05,
      "loss": 0.8944,
      "num_input_tokens_seen": 9398560,
      "step": 16200
    },
    {
      "epoch": 2.413613345248734,
      "grad_norm": 4.0,
      "learning_rate": 4.9934934237546326e-05,
      "loss": 1.0245,
      "num_input_tokens_seen": 9401568,
      "step": 16205
    },
    {
      "epoch": 2.414358057789693,
      "grad_norm": 11.5625,
      "learning_rate": 4.993469974189275e-05,
      "loss": 1.2824,
      "num_input_tokens_seen": 9404640,
      "step": 16210
    },
    {
      "epoch": 2.4151027703306522,
      "grad_norm": 7.96875,
      "learning_rate": 4.993446482499278e-05,
      "loss": 0.8987,
      "num_input_tokens_seen": 9407776,
      "step": 16215
    },
    {
      "epoch": 2.4158474828716114,
      "grad_norm": 12.75,
      "learning_rate": 4.993422948685036e-05,
      "loss": 1.3167,
      "num_input_tokens_seen": 9410432,
      "step": 16220
    },
    {
      "epoch": 2.4165921954125706,
      "grad_norm": 8.6875,
      "learning_rate": 4.993399372746948e-05,
      "loss": 1.2772,
      "num_input_tokens_seen": 9413280,
      "step": 16225
    },
    {
      "epoch": 2.41733690795353,
      "grad_norm": 10.1875,
      "learning_rate": 4.9933757546854115e-05,
      "loss": 0.7407,
      "num_input_tokens_seen": 9416032,
      "step": 16230
    },
    {
      "epoch": 2.418081620494489,
      "grad_norm": 4.3125,
      "learning_rate": 4.993352094500825e-05,
      "loss": 0.5683,
      "num_input_tokens_seen": 9419392,
      "step": 16235
    },
    {
      "epoch": 2.4188263330354483,
      "grad_norm": 9.3125,
      "learning_rate": 4.99332839219359e-05,
      "loss": 0.8205,
      "num_input_tokens_seen": 9422080,
      "step": 16240
    },
    {
      "epoch": 2.4195710455764075,
      "grad_norm": 9.3125,
      "learning_rate": 4.993304647764106e-05,
      "loss": 1.4257,
      "num_input_tokens_seen": 9424896,
      "step": 16245
    },
    {
      "epoch": 2.4203157581173667,
      "grad_norm": 10.5625,
      "learning_rate": 4.993280861212773e-05,
      "loss": 1.0787,
      "num_input_tokens_seen": 9427904,
      "step": 16250
    },
    {
      "epoch": 2.421060470658326,
      "grad_norm": 11.0,
      "learning_rate": 4.993257032539995e-05,
      "loss": 1.1627,
      "num_input_tokens_seen": 9430944,
      "step": 16255
    },
    {
      "epoch": 2.421805183199285,
      "grad_norm": 9.5,
      "learning_rate": 4.993233161746174e-05,
      "loss": 1.0048,
      "num_input_tokens_seen": 9433952,
      "step": 16260
    },
    {
      "epoch": 2.4225498957402443,
      "grad_norm": 5.28125,
      "learning_rate": 4.993209248831711e-05,
      "loss": 1.0861,
      "num_input_tokens_seen": 9436960,
      "step": 16265
    },
    {
      "epoch": 2.4232946082812035,
      "grad_norm": 8.3125,
      "learning_rate": 4.9931852937970124e-05,
      "loss": 1.0254,
      "num_input_tokens_seen": 9439712,
      "step": 16270
    },
    {
      "epoch": 2.4240393208221627,
      "grad_norm": 6.8125,
      "learning_rate": 4.9931612966424824e-05,
      "loss": 1.0991,
      "num_input_tokens_seen": 9442656,
      "step": 16275
    },
    {
      "epoch": 2.424784033363122,
      "grad_norm": 8.625,
      "learning_rate": 4.993137257368526e-05,
      "loss": 1.0423,
      "num_input_tokens_seen": 9445632,
      "step": 16280
    },
    {
      "epoch": 2.425528745904081,
      "grad_norm": 6.3125,
      "learning_rate": 4.99311317597555e-05,
      "loss": 1.158,
      "num_input_tokens_seen": 9448672,
      "step": 16285
    },
    {
      "epoch": 2.4262734584450403,
      "grad_norm": 12.0,
      "learning_rate": 4.993089052463961e-05,
      "loss": 1.3244,
      "num_input_tokens_seen": 9451520,
      "step": 16290
    },
    {
      "epoch": 2.4270181709859995,
      "grad_norm": 7.71875,
      "learning_rate": 4.993064886834166e-05,
      "loss": 1.0434,
      "num_input_tokens_seen": 9454240,
      "step": 16295
    },
    {
      "epoch": 2.4277628835269587,
      "grad_norm": 19.625,
      "learning_rate": 4.993040679086573e-05,
      "loss": 1.1138,
      "num_input_tokens_seen": 9457216,
      "step": 16300
    },
    {
      "epoch": 2.428507596067918,
      "grad_norm": 6.875,
      "learning_rate": 4.993016429221592e-05,
      "loss": 0.829,
      "num_input_tokens_seen": 9460096,
      "step": 16305
    },
    {
      "epoch": 2.429252308608877,
      "grad_norm": 12.8125,
      "learning_rate": 4.992992137239632e-05,
      "loss": 1.31,
      "num_input_tokens_seen": 9462976,
      "step": 16310
    },
    {
      "epoch": 2.4299970211498363,
      "grad_norm": 13.5,
      "learning_rate": 4.992967803141104e-05,
      "loss": 1.1099,
      "num_input_tokens_seen": 9466016,
      "step": 16315
    },
    {
      "epoch": 2.4307417336907955,
      "grad_norm": 9.9375,
      "learning_rate": 4.992943426926419e-05,
      "loss": 0.9853,
      "num_input_tokens_seen": 9469248,
      "step": 16320
    },
    {
      "epoch": 2.4314864462317547,
      "grad_norm": 5.03125,
      "learning_rate": 4.9929190085959874e-05,
      "loss": 1.1218,
      "num_input_tokens_seen": 9472224,
      "step": 16325
    },
    {
      "epoch": 2.432231158772714,
      "grad_norm": 10.625,
      "learning_rate": 4.9928945481502234e-05,
      "loss": 1.1408,
      "num_input_tokens_seen": 9474816,
      "step": 16330
    },
    {
      "epoch": 2.432975871313673,
      "grad_norm": 12.1875,
      "learning_rate": 4.9928700455895394e-05,
      "loss": 1.2627,
      "num_input_tokens_seen": 9477504,
      "step": 16335
    },
    {
      "epoch": 2.4337205838546323,
      "grad_norm": 19.5,
      "learning_rate": 4.99284550091435e-05,
      "loss": 1.2827,
      "num_input_tokens_seen": 9480288,
      "step": 16340
    },
    {
      "epoch": 2.434465296395591,
      "grad_norm": 8.25,
      "learning_rate": 4.992820914125069e-05,
      "loss": 0.978,
      "num_input_tokens_seen": 9483392,
      "step": 16345
    },
    {
      "epoch": 2.4352100089365507,
      "grad_norm": 7.875,
      "learning_rate": 4.9927962852221136e-05,
      "loss": 1.0926,
      "num_input_tokens_seen": 9486208,
      "step": 16350
    },
    {
      "epoch": 2.4359547214775095,
      "grad_norm": 6.4375,
      "learning_rate": 4.9927716142058976e-05,
      "loss": 1.0156,
      "num_input_tokens_seen": 9488960,
      "step": 16355
    },
    {
      "epoch": 2.436699434018469,
      "grad_norm": 16.875,
      "learning_rate": 4.992746901076838e-05,
      "loss": 1.2187,
      "num_input_tokens_seen": 9491936,
      "step": 16360
    },
    {
      "epoch": 2.437444146559428,
      "grad_norm": 11.375,
      "learning_rate": 4.992722145835354e-05,
      "loss": 0.8973,
      "num_input_tokens_seen": 9494944,
      "step": 16365
    },
    {
      "epoch": 2.438188859100387,
      "grad_norm": 6.21875,
      "learning_rate": 4.992697348481863e-05,
      "loss": 1.272,
      "num_input_tokens_seen": 9497824,
      "step": 16370
    },
    {
      "epoch": 2.4389335716413463,
      "grad_norm": 13.4375,
      "learning_rate": 4.992672509016782e-05,
      "loss": 0.9616,
      "num_input_tokens_seen": 9500672,
      "step": 16375
    },
    {
      "epoch": 2.4396782841823055,
      "grad_norm": 13.9375,
      "learning_rate": 4.992647627440534e-05,
      "loss": 1.1014,
      "num_input_tokens_seen": 9503936,
      "step": 16380
    },
    {
      "epoch": 2.4404229967232647,
      "grad_norm": 7.28125,
      "learning_rate": 4.992622703753538e-05,
      "loss": 1.0775,
      "num_input_tokens_seen": 9506720,
      "step": 16385
    },
    {
      "epoch": 2.441167709264224,
      "grad_norm": 11.6875,
      "learning_rate": 4.9925977379562145e-05,
      "loss": 0.8484,
      "num_input_tokens_seen": 9509568,
      "step": 16390
    },
    {
      "epoch": 2.441912421805183,
      "grad_norm": 8.6875,
      "learning_rate": 4.9925727300489853e-05,
      "loss": 1.5385,
      "num_input_tokens_seen": 9512512,
      "step": 16395
    },
    {
      "epoch": 2.4426571343461423,
      "grad_norm": 16.5,
      "learning_rate": 4.9925476800322735e-05,
      "loss": 1.1466,
      "num_input_tokens_seen": 9515680,
      "step": 16400
    },
    {
      "epoch": 2.4434018468871015,
      "grad_norm": 12.25,
      "learning_rate": 4.992522587906501e-05,
      "loss": 1.236,
      "num_input_tokens_seen": 9518816,
      "step": 16405
    },
    {
      "epoch": 2.4441465594280607,
      "grad_norm": 10.25,
      "learning_rate": 4.992497453672094e-05,
      "loss": 1.3857,
      "num_input_tokens_seen": 9521632,
      "step": 16410
    },
    {
      "epoch": 2.44489127196902,
      "grad_norm": 16.125,
      "learning_rate": 4.9924722773294745e-05,
      "loss": 1.245,
      "num_input_tokens_seen": 9524512,
      "step": 16415
    },
    {
      "epoch": 2.445635984509979,
      "grad_norm": 4.8125,
      "learning_rate": 4.99244705887907e-05,
      "loss": 0.7561,
      "num_input_tokens_seen": 9527200,
      "step": 16420
    },
    {
      "epoch": 2.4463806970509383,
      "grad_norm": 12.0625,
      "learning_rate": 4.992421798321305e-05,
      "loss": 1.5127,
      "num_input_tokens_seen": 9529984,
      "step": 16425
    },
    {
      "epoch": 2.4471254095918975,
      "grad_norm": 10.5625,
      "learning_rate": 4.992396495656608e-05,
      "loss": 1.0578,
      "num_input_tokens_seen": 9532864,
      "step": 16430
    },
    {
      "epoch": 2.4478701221328567,
      "grad_norm": 11.1875,
      "learning_rate": 4.992371150885404e-05,
      "loss": 0.9043,
      "num_input_tokens_seen": 9535680,
      "step": 16435
    },
    {
      "epoch": 2.448614834673816,
      "grad_norm": 10.9375,
      "learning_rate": 4.9923457640081236e-05,
      "loss": 0.9557,
      "num_input_tokens_seen": 9538752,
      "step": 16440
    },
    {
      "epoch": 2.449359547214775,
      "grad_norm": 4.8125,
      "learning_rate": 4.992320335025194e-05,
      "loss": 1.0105,
      "num_input_tokens_seen": 9541856,
      "step": 16445
    },
    {
      "epoch": 2.4501042597557343,
      "grad_norm": 20.5,
      "learning_rate": 4.992294863937046e-05,
      "loss": 1.2022,
      "num_input_tokens_seen": 9544672,
      "step": 16450
    },
    {
      "epoch": 2.4508489722966935,
      "grad_norm": 8.9375,
      "learning_rate": 4.9922693507441084e-05,
      "loss": 1.0298,
      "num_input_tokens_seen": 9547328,
      "step": 16455
    },
    {
      "epoch": 2.4515936848376527,
      "grad_norm": 8.75,
      "learning_rate": 4.9922437954468136e-05,
      "loss": 1.172,
      "num_input_tokens_seen": 9550528,
      "step": 16460
    },
    {
      "epoch": 2.452338397378612,
      "grad_norm": 10.875,
      "learning_rate": 4.992218198045593e-05,
      "loss": 1.5561,
      "num_input_tokens_seen": 9553440,
      "step": 16465
    },
    {
      "epoch": 2.453083109919571,
      "grad_norm": 11.1875,
      "learning_rate": 4.992192558540879e-05,
      "loss": 1.354,
      "num_input_tokens_seen": 9556288,
      "step": 16470
    },
    {
      "epoch": 2.4538278224605303,
      "grad_norm": 13.0,
      "learning_rate": 4.992166876933105e-05,
      "loss": 1.1653,
      "num_input_tokens_seen": 9559232,
      "step": 16475
    },
    {
      "epoch": 2.4545725350014895,
      "grad_norm": 8.3125,
      "learning_rate": 4.9921411532227036e-05,
      "loss": 0.8697,
      "num_input_tokens_seen": 9562464,
      "step": 16480
    },
    {
      "epoch": 2.4553172475424487,
      "grad_norm": 9.5625,
      "learning_rate": 4.992115387410111e-05,
      "loss": 0.8621,
      "num_input_tokens_seen": 9565472,
      "step": 16485
    },
    {
      "epoch": 2.456061960083408,
      "grad_norm": 9.8125,
      "learning_rate": 4.992089579495762e-05,
      "loss": 1.0722,
      "num_input_tokens_seen": 9568448,
      "step": 16490
    },
    {
      "epoch": 2.456806672624367,
      "grad_norm": 11.625,
      "learning_rate": 4.992063729480092e-05,
      "loss": 1.5426,
      "num_input_tokens_seen": 9571360,
      "step": 16495
    },
    {
      "epoch": 2.4575513851653263,
      "grad_norm": 9.8125,
      "learning_rate": 4.992037837363538e-05,
      "loss": 1.3651,
      "num_input_tokens_seen": 9574176,
      "step": 16500
    },
    {
      "epoch": 2.4582960977062855,
      "grad_norm": 12.6875,
      "learning_rate": 4.992011903146537e-05,
      "loss": 1.0909,
      "num_input_tokens_seen": 9577056,
      "step": 16505
    },
    {
      "epoch": 2.4590408102472447,
      "grad_norm": 8.875,
      "learning_rate": 4.991985926829529e-05,
      "loss": 1.4246,
      "num_input_tokens_seen": 9579968,
      "step": 16510
    },
    {
      "epoch": 2.459785522788204,
      "grad_norm": 6.59375,
      "learning_rate": 4.991959908412951e-05,
      "loss": 1.0155,
      "num_input_tokens_seen": 9582720,
      "step": 16515
    },
    {
      "epoch": 2.4605302353291627,
      "grad_norm": 5.96875,
      "learning_rate": 4.9919338478972424e-05,
      "loss": 1.1137,
      "num_input_tokens_seen": 9585600,
      "step": 16520
    },
    {
      "epoch": 2.4612749478701224,
      "grad_norm": 9.5,
      "learning_rate": 4.991907745282845e-05,
      "loss": 0.9938,
      "num_input_tokens_seen": 9588416,
      "step": 16525
    },
    {
      "epoch": 2.462019660411081,
      "grad_norm": 9.75,
      "learning_rate": 4.9918816005701986e-05,
      "loss": 1.0675,
      "num_input_tokens_seen": 9591424,
      "step": 16530
    },
    {
      "epoch": 2.4627643729520408,
      "grad_norm": 8.6875,
      "learning_rate": 4.9918554137597454e-05,
      "loss": 1.0316,
      "num_input_tokens_seen": 9594560,
      "step": 16535
    },
    {
      "epoch": 2.4635090854929995,
      "grad_norm": 8.1875,
      "learning_rate": 4.9918291848519275e-05,
      "loss": 1.0634,
      "num_input_tokens_seen": 9597504,
      "step": 16540
    },
    {
      "epoch": 2.4642537980339587,
      "grad_norm": 9.0625,
      "learning_rate": 4.991802913847188e-05,
      "loss": 0.7658,
      "num_input_tokens_seen": 9600480,
      "step": 16545
    },
    {
      "epoch": 2.464998510574918,
      "grad_norm": 10.6875,
      "learning_rate": 4.9917766007459696e-05,
      "loss": 0.8911,
      "num_input_tokens_seen": 9603232,
      "step": 16550
    },
    {
      "epoch": 2.465743223115877,
      "grad_norm": 18.0,
      "learning_rate": 4.99175024554872e-05,
      "loss": 1.2758,
      "num_input_tokens_seen": 9606464,
      "step": 16555
    },
    {
      "epoch": 2.4664879356568363,
      "grad_norm": 6.03125,
      "learning_rate": 4.991723848255881e-05,
      "loss": 1.0028,
      "num_input_tokens_seen": 9609504,
      "step": 16560
    },
    {
      "epoch": 2.4672326481977955,
      "grad_norm": 12.0625,
      "learning_rate": 4.9916974088679015e-05,
      "loss": 1.0824,
      "num_input_tokens_seen": 9612256,
      "step": 16565
    },
    {
      "epoch": 2.4679773607387547,
      "grad_norm": 8.75,
      "learning_rate": 4.991670927385226e-05,
      "loss": 0.8404,
      "num_input_tokens_seen": 9615264,
      "step": 16570
    },
    {
      "epoch": 2.468722073279714,
      "grad_norm": 16.375,
      "learning_rate": 4.9916444038083024e-05,
      "loss": 1.2644,
      "num_input_tokens_seen": 9618112,
      "step": 16575
    },
    {
      "epoch": 2.469466785820673,
      "grad_norm": 15.8125,
      "learning_rate": 4.991617838137579e-05,
      "loss": 1.8519,
      "num_input_tokens_seen": 9621152,
      "step": 16580
    },
    {
      "epoch": 2.4702114983616323,
      "grad_norm": 8.875,
      "learning_rate": 4.991591230373505e-05,
      "loss": 1.4461,
      "num_input_tokens_seen": 9623968,
      "step": 16585
    },
    {
      "epoch": 2.4709562109025915,
      "grad_norm": 5.8125,
      "learning_rate": 4.99156458051653e-05,
      "loss": 0.7216,
      "num_input_tokens_seen": 9626720,
      "step": 16590
    },
    {
      "epoch": 2.4717009234435507,
      "grad_norm": 21.5,
      "learning_rate": 4.9915378885671026e-05,
      "loss": 1.2463,
      "num_input_tokens_seen": 9629664,
      "step": 16595
    },
    {
      "epoch": 2.47244563598451,
      "grad_norm": 9.375,
      "learning_rate": 4.9915111545256754e-05,
      "loss": 1.4902,
      "num_input_tokens_seen": 9632320,
      "step": 16600
    },
    {
      "epoch": 2.473190348525469,
      "grad_norm": 12.875,
      "learning_rate": 4.991484378392699e-05,
      "loss": 1.276,
      "num_input_tokens_seen": 9635168,
      "step": 16605
    },
    {
      "epoch": 2.4739350610664284,
      "grad_norm": 12.0,
      "learning_rate": 4.9914575601686266e-05,
      "loss": 1.0035,
      "num_input_tokens_seen": 9638432,
      "step": 16610
    },
    {
      "epoch": 2.4746797736073876,
      "grad_norm": 15.5625,
      "learning_rate": 4.9914306998539115e-05,
      "loss": 1.0306,
      "num_input_tokens_seen": 9641632,
      "step": 16615
    },
    {
      "epoch": 2.4754244861483468,
      "grad_norm": 8.5,
      "learning_rate": 4.991403797449006e-05,
      "loss": 1.2368,
      "num_input_tokens_seen": 9644672,
      "step": 16620
    },
    {
      "epoch": 2.476169198689306,
      "grad_norm": 16.375,
      "learning_rate": 4.9913768529543656e-05,
      "loss": 1.1291,
      "num_input_tokens_seen": 9647360,
      "step": 16625
    },
    {
      "epoch": 2.476913911230265,
      "grad_norm": 11.5,
      "learning_rate": 4.991349866370446e-05,
      "loss": 1.1556,
      "num_input_tokens_seen": 9650464,
      "step": 16630
    },
    {
      "epoch": 2.4776586237712244,
      "grad_norm": 8.9375,
      "learning_rate": 4.9913228376977026e-05,
      "loss": 1.0824,
      "num_input_tokens_seen": 9653056,
      "step": 16635
    },
    {
      "epoch": 2.4784033363121836,
      "grad_norm": 8.25,
      "learning_rate": 4.991295766936591e-05,
      "loss": 1.1165,
      "num_input_tokens_seen": 9655808,
      "step": 16640
    },
    {
      "epoch": 2.4791480488531428,
      "grad_norm": 25.5,
      "learning_rate": 4.9912686540875696e-05,
      "loss": 0.9669,
      "num_input_tokens_seen": 9658592,
      "step": 16645
    },
    {
      "epoch": 2.479892761394102,
      "grad_norm": 15.9375,
      "learning_rate": 4.991241499151097e-05,
      "loss": 1.0769,
      "num_input_tokens_seen": 9661600,
      "step": 16650
    },
    {
      "epoch": 2.480637473935061,
      "grad_norm": 9.5,
      "learning_rate": 4.9912143021276306e-05,
      "loss": 1.0193,
      "num_input_tokens_seen": 9664576,
      "step": 16655
    },
    {
      "epoch": 2.4813821864760204,
      "grad_norm": 14.4375,
      "learning_rate": 4.991187063017631e-05,
      "loss": 0.9464,
      "num_input_tokens_seen": 9667776,
      "step": 16660
    },
    {
      "epoch": 2.4821268990169796,
      "grad_norm": 10.25,
      "learning_rate": 4.9911597818215575e-05,
      "loss": 1.2112,
      "num_input_tokens_seen": 9670720,
      "step": 16665
    },
    {
      "epoch": 2.482871611557939,
      "grad_norm": 6.625,
      "learning_rate": 4.9911324585398724e-05,
      "loss": 1.0425,
      "num_input_tokens_seen": 9673856,
      "step": 16670
    },
    {
      "epoch": 2.483616324098898,
      "grad_norm": 6.15625,
      "learning_rate": 4.9911050931730354e-05,
      "loss": 1.1267,
      "num_input_tokens_seen": 9676576,
      "step": 16675
    },
    {
      "epoch": 2.484361036639857,
      "grad_norm": 6.90625,
      "learning_rate": 4.9910776857215094e-05,
      "loss": 1.0932,
      "num_input_tokens_seen": 9679872,
      "step": 16680
    },
    {
      "epoch": 2.4851057491808164,
      "grad_norm": 7.34375,
      "learning_rate": 4.991050236185758e-05,
      "loss": 0.7573,
      "num_input_tokens_seen": 9682624,
      "step": 16685
    },
    {
      "epoch": 2.4858504617217756,
      "grad_norm": 9.8125,
      "learning_rate": 4.991022744566245e-05,
      "loss": 1.1415,
      "num_input_tokens_seen": 9685504,
      "step": 16690
    },
    {
      "epoch": 2.4865951742627344,
      "grad_norm": 14.5,
      "learning_rate": 4.990995210863434e-05,
      "loss": 1.076,
      "num_input_tokens_seen": 9688416,
      "step": 16695
    },
    {
      "epoch": 2.487339886803694,
      "grad_norm": 7.59375,
      "learning_rate": 4.9909676350777914e-05,
      "loss": 0.8529,
      "num_input_tokens_seen": 9691360,
      "step": 16700
    },
    {
      "epoch": 2.4880845993446528,
      "grad_norm": 10.0625,
      "learning_rate": 4.990940017209782e-05,
      "loss": 1.427,
      "num_input_tokens_seen": 9694400,
      "step": 16705
    },
    {
      "epoch": 2.488829311885612,
      "grad_norm": 7.0,
      "learning_rate": 4.990912357259872e-05,
      "loss": 1.3197,
      "num_input_tokens_seen": 9697504,
      "step": 16710
    },
    {
      "epoch": 2.489574024426571,
      "grad_norm": 7.09375,
      "learning_rate": 4.99088465522853e-05,
      "loss": 0.9722,
      "num_input_tokens_seen": 9700480,
      "step": 16715
    },
    {
      "epoch": 2.4903187369675304,
      "grad_norm": 16.125,
      "learning_rate": 4.9908569111162226e-05,
      "loss": 1.1015,
      "num_input_tokens_seen": 9703296,
      "step": 16720
    },
    {
      "epoch": 2.4910634495084896,
      "grad_norm": 17.25,
      "learning_rate": 4.9908291249234207e-05,
      "loss": 0.9376,
      "num_input_tokens_seen": 9706464,
      "step": 16725
    },
    {
      "epoch": 2.4918081620494488,
      "grad_norm": 8.5,
      "learning_rate": 4.990801296650591e-05,
      "loss": 0.8722,
      "num_input_tokens_seen": 9709120,
      "step": 16730
    },
    {
      "epoch": 2.492552874590408,
      "grad_norm": 13.9375,
      "learning_rate": 4.9907734262982056e-05,
      "loss": 1.1975,
      "num_input_tokens_seen": 9712192,
      "step": 16735
    },
    {
      "epoch": 2.493297587131367,
      "grad_norm": 9.0,
      "learning_rate": 4.990745513866735e-05,
      "loss": 1.0176,
      "num_input_tokens_seen": 9714848,
      "step": 16740
    },
    {
      "epoch": 2.4940422996723264,
      "grad_norm": 14.8125,
      "learning_rate": 4.99071755935665e-05,
      "loss": 1.1491,
      "num_input_tokens_seen": 9717664,
      "step": 16745
    },
    {
      "epoch": 2.4947870122132856,
      "grad_norm": 6.875,
      "learning_rate": 4.990689562768423e-05,
      "loss": 1.0053,
      "num_input_tokens_seen": 9720960,
      "step": 16750
    },
    {
      "epoch": 2.495531724754245,
      "grad_norm": 10.3125,
      "learning_rate": 4.990661524102528e-05,
      "loss": 1.4286,
      "num_input_tokens_seen": 9723744,
      "step": 16755
    },
    {
      "epoch": 2.496276437295204,
      "grad_norm": 9.0625,
      "learning_rate": 4.990633443359437e-05,
      "loss": 1.1305,
      "num_input_tokens_seen": 9726624,
      "step": 16760
    },
    {
      "epoch": 2.497021149836163,
      "grad_norm": 8.0625,
      "learning_rate": 4.990605320539626e-05,
      "loss": 1.0855,
      "num_input_tokens_seen": 9729504,
      "step": 16765
    },
    {
      "epoch": 2.4977658623771224,
      "grad_norm": 6.9375,
      "learning_rate": 4.990577155643569e-05,
      "loss": 0.9623,
      "num_input_tokens_seen": 9732384,
      "step": 16770
    },
    {
      "epoch": 2.4985105749180816,
      "grad_norm": 10.6875,
      "learning_rate": 4.9905489486717425e-05,
      "loss": 1.1852,
      "num_input_tokens_seen": 9735296,
      "step": 16775
    },
    {
      "epoch": 2.499255287459041,
      "grad_norm": 9.0,
      "learning_rate": 4.990520699624623e-05,
      "loss": 1.0177,
      "num_input_tokens_seen": 9738144,
      "step": 16780
    },
    {
      "epoch": 2.5,
      "grad_norm": 8.5625,
      "learning_rate": 4.990492408502687e-05,
      "loss": 1.0549,
      "num_input_tokens_seen": 9740864,
      "step": 16785
    },
    {
      "epoch": 2.500744712540959,
      "grad_norm": 7.09375,
      "learning_rate": 4.9904640753064135e-05,
      "loss": 1.216,
      "num_input_tokens_seen": 9743584,
      "step": 16790
    },
    {
      "epoch": 2.5014894250819184,
      "grad_norm": 5.5,
      "learning_rate": 4.99043570003628e-05,
      "loss": 0.984,
      "num_input_tokens_seen": 9746304,
      "step": 16795
    },
    {
      "epoch": 2.5022341376228776,
      "grad_norm": 7.875,
      "learning_rate": 4.990407282692767e-05,
      "loss": 1.0989,
      "num_input_tokens_seen": 9749248,
      "step": 16800
    },
    {
      "epoch": 2.502978850163837,
      "grad_norm": 7.4375,
      "learning_rate": 4.9903788232763545e-05,
      "loss": 0.9722,
      "num_input_tokens_seen": 9752096,
      "step": 16805
    },
    {
      "epoch": 2.503723562704796,
      "grad_norm": 5.78125,
      "learning_rate": 4.9903503217875227e-05,
      "loss": 1.3824,
      "num_input_tokens_seen": 9755456,
      "step": 16810
    },
    {
      "epoch": 2.504468275245755,
      "grad_norm": 13.3125,
      "learning_rate": 4.990321778226753e-05,
      "loss": 1.2185,
      "num_input_tokens_seen": 9758944,
      "step": 16815
    },
    {
      "epoch": 2.5052129877867144,
      "grad_norm": 7.875,
      "learning_rate": 4.990293192594527e-05,
      "loss": 1.3264,
      "num_input_tokens_seen": 9761664,
      "step": 16820
    },
    {
      "epoch": 2.5059577003276736,
      "grad_norm": 11.9375,
      "learning_rate": 4.9902645648913305e-05,
      "loss": 0.8146,
      "num_input_tokens_seen": 9764384,
      "step": 16825
    },
    {
      "epoch": 2.506702412868633,
      "grad_norm": 5.75,
      "learning_rate": 4.990235895117644e-05,
      "loss": 0.8531,
      "num_input_tokens_seen": 9767488,
      "step": 16830
    },
    {
      "epoch": 2.507447125409592,
      "grad_norm": 7.09375,
      "learning_rate": 4.990207183273953e-05,
      "loss": 0.9446,
      "num_input_tokens_seen": 9770432,
      "step": 16835
    },
    {
      "epoch": 2.5081918379505512,
      "grad_norm": 6.5625,
      "learning_rate": 4.9901784293607424e-05,
      "loss": 1.0491,
      "num_input_tokens_seen": 9773760,
      "step": 16840
    },
    {
      "epoch": 2.5089365504915104,
      "grad_norm": 8.375,
      "learning_rate": 4.990149633378499e-05,
      "loss": 1.2533,
      "num_input_tokens_seen": 9776416,
      "step": 16845
    },
    {
      "epoch": 2.509681263032469,
      "grad_norm": 10.625,
      "learning_rate": 4.990120795327707e-05,
      "loss": 0.986,
      "num_input_tokens_seen": 9779328,
      "step": 16850
    },
    {
      "epoch": 2.510425975573429,
      "grad_norm": 9.4375,
      "learning_rate": 4.990091915208857e-05,
      "loss": 0.4976,
      "num_input_tokens_seen": 9782080,
      "step": 16855
    },
    {
      "epoch": 2.5111706881143876,
      "grad_norm": 14.125,
      "learning_rate": 4.990062993022433e-05,
      "loss": 1.2037,
      "num_input_tokens_seen": 9784992,
      "step": 16860
    },
    {
      "epoch": 2.5119154006553472,
      "grad_norm": 10.4375,
      "learning_rate": 4.990034028768927e-05,
      "loss": 1.2823,
      "num_input_tokens_seen": 9788160,
      "step": 16865
    },
    {
      "epoch": 2.512660113196306,
      "grad_norm": 7.25,
      "learning_rate": 4.990005022448826e-05,
      "loss": 1.3508,
      "num_input_tokens_seen": 9790944,
      "step": 16870
    },
    {
      "epoch": 2.5134048257372656,
      "grad_norm": 9.1875,
      "learning_rate": 4.989975974062621e-05,
      "loss": 0.7253,
      "num_input_tokens_seen": 9793952,
      "step": 16875
    },
    {
      "epoch": 2.5141495382782244,
      "grad_norm": 11.3125,
      "learning_rate": 4.989946883610803e-05,
      "loss": 0.8068,
      "num_input_tokens_seen": 9796928,
      "step": 16880
    },
    {
      "epoch": 2.514894250819184,
      "grad_norm": 14.5625,
      "learning_rate": 4.9899177510938624e-05,
      "loss": 1.0774,
      "num_input_tokens_seen": 9799872,
      "step": 16885
    },
    {
      "epoch": 2.515638963360143,
      "grad_norm": 14.0625,
      "learning_rate": 4.9898885765122927e-05,
      "loss": 1.0523,
      "num_input_tokens_seen": 9802976,
      "step": 16890
    },
    {
      "epoch": 2.516383675901102,
      "grad_norm": 13.5,
      "learning_rate": 4.9898593598665856e-05,
      "loss": 1.2098,
      "num_input_tokens_seen": 9805952,
      "step": 16895
    },
    {
      "epoch": 2.517128388442061,
      "grad_norm": 9.1875,
      "learning_rate": 4.989830101157235e-05,
      "loss": 0.9647,
      "num_input_tokens_seen": 9808992,
      "step": 16900
    },
    {
      "epoch": 2.5178731009830204,
      "grad_norm": 11.25,
      "learning_rate": 4.9898008003847363e-05,
      "loss": 1.3339,
      "num_input_tokens_seen": 9811904,
      "step": 16905
    },
    {
      "epoch": 2.5186178135239796,
      "grad_norm": 12.0,
      "learning_rate": 4.989771457549582e-05,
      "loss": 1.0762,
      "num_input_tokens_seen": 9814688,
      "step": 16910
    },
    {
      "epoch": 2.519362526064939,
      "grad_norm": 18.0,
      "learning_rate": 4.989742072652271e-05,
      "loss": 1.2092,
      "num_input_tokens_seen": 9817824,
      "step": 16915
    },
    {
      "epoch": 2.520107238605898,
      "grad_norm": 6.25,
      "learning_rate": 4.989712645693297e-05,
      "loss": 1.005,
      "num_input_tokens_seen": 9821152,
      "step": 16920
    },
    {
      "epoch": 2.5208519511468572,
      "grad_norm": 6.03125,
      "learning_rate": 4.98968317667316e-05,
      "loss": 0.7686,
      "num_input_tokens_seen": 9823904,
      "step": 16925
    },
    {
      "epoch": 2.5215966636878164,
      "grad_norm": 18.125,
      "learning_rate": 4.989653665592355e-05,
      "loss": 1.3021,
      "num_input_tokens_seen": 9826656,
      "step": 16930
    },
    {
      "epoch": 2.5223413762287756,
      "grad_norm": 14.0625,
      "learning_rate": 4.989624112451381e-05,
      "loss": 1.1178,
      "num_input_tokens_seen": 9829376,
      "step": 16935
    },
    {
      "epoch": 2.523086088769735,
      "grad_norm": 13.8125,
      "learning_rate": 4.989594517250739e-05,
      "loss": 1.1754,
      "num_input_tokens_seen": 9832448,
      "step": 16940
    },
    {
      "epoch": 2.523830801310694,
      "grad_norm": 11.4375,
      "learning_rate": 4.989564879990928e-05,
      "loss": 1.605,
      "num_input_tokens_seen": 9835392,
      "step": 16945
    },
    {
      "epoch": 2.5245755138516532,
      "grad_norm": 20.125,
      "learning_rate": 4.9895352006724485e-05,
      "loss": 1.1123,
      "num_input_tokens_seen": 9838176,
      "step": 16950
    },
    {
      "epoch": 2.5253202263926124,
      "grad_norm": 7.28125,
      "learning_rate": 4.9895054792958015e-05,
      "loss": 0.7573,
      "num_input_tokens_seen": 9841024,
      "step": 16955
    },
    {
      "epoch": 2.5260649389335716,
      "grad_norm": 5.8125,
      "learning_rate": 4.98947571586149e-05,
      "loss": 0.9329,
      "num_input_tokens_seen": 9843744,
      "step": 16960
    },
    {
      "epoch": 2.526809651474531,
      "grad_norm": 7.40625,
      "learning_rate": 4.9894459103700166e-05,
      "loss": 0.9216,
      "num_input_tokens_seen": 9846496,
      "step": 16965
    },
    {
      "epoch": 2.52755436401549,
      "grad_norm": 8.375,
      "learning_rate": 4.989416062821884e-05,
      "loss": 0.896,
      "num_input_tokens_seen": 9849120,
      "step": 16970
    },
    {
      "epoch": 2.5282990765564493,
      "grad_norm": 12.5625,
      "learning_rate": 4.989386173217598e-05,
      "loss": 0.9459,
      "num_input_tokens_seen": 9852320,
      "step": 16975
    },
    {
      "epoch": 2.5290437890974085,
      "grad_norm": 6.53125,
      "learning_rate": 4.989356241557662e-05,
      "loss": 1.0394,
      "num_input_tokens_seen": 9855136,
      "step": 16980
    },
    {
      "epoch": 2.5297885016383677,
      "grad_norm": 11.3125,
      "learning_rate": 4.989326267842583e-05,
      "loss": 1.0079,
      "num_input_tokens_seen": 9857856,
      "step": 16985
    },
    {
      "epoch": 2.530533214179327,
      "grad_norm": 9.0,
      "learning_rate": 4.9892962520728664e-05,
      "loss": 1.1278,
      "num_input_tokens_seen": 9860768,
      "step": 16990
    },
    {
      "epoch": 2.531277926720286,
      "grad_norm": 12.375,
      "learning_rate": 4.989266194249019e-05,
      "loss": 0.8717,
      "num_input_tokens_seen": 9863584,
      "step": 16995
    },
    {
      "epoch": 2.5320226392612453,
      "grad_norm": 10.875,
      "learning_rate": 4.989236094371551e-05,
      "loss": 0.9329,
      "num_input_tokens_seen": 9866752,
      "step": 17000
    },
    {
      "epoch": 2.5327673518022045,
      "grad_norm": 4.4375,
      "learning_rate": 4.9892059524409676e-05,
      "loss": 0.8395,
      "num_input_tokens_seen": 9869760,
      "step": 17005
    },
    {
      "epoch": 2.5335120643431637,
      "grad_norm": 18.5,
      "learning_rate": 4.98917576845778e-05,
      "loss": 1.183,
      "num_input_tokens_seen": 9872800,
      "step": 17010
    },
    {
      "epoch": 2.534256776884123,
      "grad_norm": 5.3125,
      "learning_rate": 4.989145542422498e-05,
      "loss": 0.9001,
      "num_input_tokens_seen": 9875840,
      "step": 17015
    },
    {
      "epoch": 2.535001489425082,
      "grad_norm": 9.5,
      "learning_rate": 4.989115274335632e-05,
      "loss": 1.298,
      "num_input_tokens_seen": 9878816,
      "step": 17020
    },
    {
      "epoch": 2.535746201966041,
      "grad_norm": 7.53125,
      "learning_rate": 4.9890849641976924e-05,
      "loss": 1.0614,
      "num_input_tokens_seen": 9881504,
      "step": 17025
    },
    {
      "epoch": 2.5364909145070005,
      "grad_norm": 19.875,
      "learning_rate": 4.989054612009192e-05,
      "loss": 1.5957,
      "num_input_tokens_seen": 9884992,
      "step": 17030
    },
    {
      "epoch": 2.5372356270479592,
      "grad_norm": 8.375,
      "learning_rate": 4.989024217770645e-05,
      "loss": 0.8625,
      "num_input_tokens_seen": 9887616,
      "step": 17035
    },
    {
      "epoch": 2.537980339588919,
      "grad_norm": 11.625,
      "learning_rate": 4.988993781482563e-05,
      "loss": 1.0229,
      "num_input_tokens_seen": 9890240,
      "step": 17040
    },
    {
      "epoch": 2.5387250521298776,
      "grad_norm": 6.46875,
      "learning_rate": 4.9889633031454604e-05,
      "loss": 0.7553,
      "num_input_tokens_seen": 9893184,
      "step": 17045
    },
    {
      "epoch": 2.5394697646708373,
      "grad_norm": 10.9375,
      "learning_rate": 4.9889327827598526e-05,
      "loss": 0.8622,
      "num_input_tokens_seen": 9896064,
      "step": 17050
    },
    {
      "epoch": 2.540214477211796,
      "grad_norm": 8.875,
      "learning_rate": 4.988902220326255e-05,
      "loss": 0.9569,
      "num_input_tokens_seen": 9898752,
      "step": 17055
    },
    {
      "epoch": 2.5409591897527557,
      "grad_norm": 3.953125,
      "learning_rate": 4.9888716158451844e-05,
      "loss": 0.828,
      "num_input_tokens_seen": 9901632,
      "step": 17060
    },
    {
      "epoch": 2.5417039022937145,
      "grad_norm": 9.625,
      "learning_rate": 4.988840969317157e-05,
      "loss": 0.9124,
      "num_input_tokens_seen": 9904608,
      "step": 17065
    },
    {
      "epoch": 2.5424486148346737,
      "grad_norm": 12.0625,
      "learning_rate": 4.988810280742691e-05,
      "loss": 1.2068,
      "num_input_tokens_seen": 9907296,
      "step": 17070
    },
    {
      "epoch": 2.543193327375633,
      "grad_norm": 11.6875,
      "learning_rate": 4.988779550122305e-05,
      "loss": 0.9577,
      "num_input_tokens_seen": 9910240,
      "step": 17075
    },
    {
      "epoch": 2.543938039916592,
      "grad_norm": 15.4375,
      "learning_rate": 4.9887487774565176e-05,
      "loss": 1.3485,
      "num_input_tokens_seen": 9913280,
      "step": 17080
    },
    {
      "epoch": 2.5446827524575513,
      "grad_norm": 8.0625,
      "learning_rate": 4.98871796274585e-05,
      "loss": 1.1092,
      "num_input_tokens_seen": 9916160,
      "step": 17085
    },
    {
      "epoch": 2.5454274649985105,
      "grad_norm": 10.375,
      "learning_rate": 4.9886871059908213e-05,
      "loss": 1.3408,
      "num_input_tokens_seen": 9919072,
      "step": 17090
    },
    {
      "epoch": 2.5461721775394697,
      "grad_norm": 20.375,
      "learning_rate": 4.988656207191953e-05,
      "loss": 1.1493,
      "num_input_tokens_seen": 9922272,
      "step": 17095
    },
    {
      "epoch": 2.546916890080429,
      "grad_norm": 7.78125,
      "learning_rate": 4.988625266349768e-05,
      "loss": 1.0544,
      "num_input_tokens_seen": 9925024,
      "step": 17100
    },
    {
      "epoch": 2.547661602621388,
      "grad_norm": 11.3125,
      "learning_rate": 4.988594283464788e-05,
      "loss": 1.4644,
      "num_input_tokens_seen": 9928160,
      "step": 17105
    },
    {
      "epoch": 2.5484063151623473,
      "grad_norm": 8.625,
      "learning_rate": 4.988563258537537e-05,
      "loss": 1.1145,
      "num_input_tokens_seen": 9930688,
      "step": 17110
    },
    {
      "epoch": 2.5491510277033065,
      "grad_norm": 8.6875,
      "learning_rate": 4.988532191568539e-05,
      "loss": 1.1381,
      "num_input_tokens_seen": 9933568,
      "step": 17115
    },
    {
      "epoch": 2.5498957402442657,
      "grad_norm": 8.375,
      "learning_rate": 4.988501082558319e-05,
      "loss": 0.8956,
      "num_input_tokens_seen": 9936128,
      "step": 17120
    },
    {
      "epoch": 2.550640452785225,
      "grad_norm": 9.5625,
      "learning_rate": 4.988469931507402e-05,
      "loss": 1.0187,
      "num_input_tokens_seen": 9939296,
      "step": 17125
    },
    {
      "epoch": 2.551385165326184,
      "grad_norm": 7.53125,
      "learning_rate": 4.988438738416316e-05,
      "loss": 0.8756,
      "num_input_tokens_seen": 9942400,
      "step": 17130
    },
    {
      "epoch": 2.5521298778671433,
      "grad_norm": 11.75,
      "learning_rate": 4.988407503285585e-05,
      "loss": 1.0118,
      "num_input_tokens_seen": 9945056,
      "step": 17135
    },
    {
      "epoch": 2.5528745904081025,
      "grad_norm": 6.625,
      "learning_rate": 4.988376226115739e-05,
      "loss": 0.839,
      "num_input_tokens_seen": 9947744,
      "step": 17140
    },
    {
      "epoch": 2.5536193029490617,
      "grad_norm": 11.3125,
      "learning_rate": 4.9883449069073055e-05,
      "loss": 1.0669,
      "num_input_tokens_seen": 9950464,
      "step": 17145
    },
    {
      "epoch": 2.554364015490021,
      "grad_norm": 9.9375,
      "learning_rate": 4.9883135456608146e-05,
      "loss": 1.1595,
      "num_input_tokens_seen": 9953632,
      "step": 17150
    },
    {
      "epoch": 2.55510872803098,
      "grad_norm": 15.375,
      "learning_rate": 4.988282142376795e-05,
      "loss": 1.1702,
      "num_input_tokens_seen": 9956736,
      "step": 17155
    },
    {
      "epoch": 2.5558534405719393,
      "grad_norm": 15.3125,
      "learning_rate": 4.9882506970557766e-05,
      "loss": 1.2583,
      "num_input_tokens_seen": 9959424,
      "step": 17160
    },
    {
      "epoch": 2.5565981531128985,
      "grad_norm": 9.5,
      "learning_rate": 4.988219209698293e-05,
      "loss": 0.9395,
      "num_input_tokens_seen": 9962112,
      "step": 17165
    },
    {
      "epoch": 2.5573428656538577,
      "grad_norm": 9.8125,
      "learning_rate": 4.988187680304874e-05,
      "loss": 0.8719,
      "num_input_tokens_seen": 9964768,
      "step": 17170
    },
    {
      "epoch": 2.558087578194817,
      "grad_norm": 12.25,
      "learning_rate": 4.988156108876053e-05,
      "loss": 1.0415,
      "num_input_tokens_seen": 9967680,
      "step": 17175
    },
    {
      "epoch": 2.558832290735776,
      "grad_norm": 7.21875,
      "learning_rate": 4.988124495412364e-05,
      "loss": 1.0374,
      "num_input_tokens_seen": 9970496,
      "step": 17180
    },
    {
      "epoch": 2.5595770032767353,
      "grad_norm": 9.625,
      "learning_rate": 4.98809283991434e-05,
      "loss": 1.1511,
      "num_input_tokens_seen": 9973152,
      "step": 17185
    },
    {
      "epoch": 2.5603217158176945,
      "grad_norm": 15.75,
      "learning_rate": 4.988061142382516e-05,
      "loss": 1.2338,
      "num_input_tokens_seen": 9976000,
      "step": 17190
    },
    {
      "epoch": 2.5610664283586537,
      "grad_norm": 9.1875,
      "learning_rate": 4.988029402817428e-05,
      "loss": 0.8157,
      "num_input_tokens_seen": 9978656,
      "step": 17195
    },
    {
      "epoch": 2.5618111408996125,
      "grad_norm": 8.1875,
      "learning_rate": 4.9879976212196124e-05,
      "loss": 0.8955,
      "num_input_tokens_seen": 9981696,
      "step": 17200
    },
    {
      "epoch": 2.562555853440572,
      "grad_norm": 14.8125,
      "learning_rate": 4.987965797589605e-05,
      "loss": 1.0481,
      "num_input_tokens_seen": 9984608,
      "step": 17205
    },
    {
      "epoch": 2.563300565981531,
      "grad_norm": 11.1875,
      "learning_rate": 4.987933931927944e-05,
      "loss": 1.0943,
      "num_input_tokens_seen": 9987616,
      "step": 17210
    },
    {
      "epoch": 2.5640452785224905,
      "grad_norm": 12.3125,
      "learning_rate": 4.987902024235169e-05,
      "loss": 1.138,
      "num_input_tokens_seen": 9990784,
      "step": 17215
    },
    {
      "epoch": 2.5647899910634493,
      "grad_norm": 14.25,
      "learning_rate": 4.987870074511817e-05,
      "loss": 1.7264,
      "num_input_tokens_seen": 9993696,
      "step": 17220
    },
    {
      "epoch": 2.565534703604409,
      "grad_norm": 14.1875,
      "learning_rate": 4.987838082758428e-05,
      "loss": 1.0309,
      "num_input_tokens_seen": 9996416,
      "step": 17225
    },
    {
      "epoch": 2.5662794161453677,
      "grad_norm": 11.1875,
      "learning_rate": 4.987806048975544e-05,
      "loss": 1.2359,
      "num_input_tokens_seen": 9999424,
      "step": 17230
    },
    {
      "epoch": 2.5670241286863273,
      "grad_norm": 12.0625,
      "learning_rate": 4.987773973163706e-05,
      "loss": 0.9775,
      "num_input_tokens_seen": 10002720,
      "step": 17235
    },
    {
      "epoch": 2.567768841227286,
      "grad_norm": 12.375,
      "learning_rate": 4.987741855323454e-05,
      "loss": 0.7962,
      "num_input_tokens_seen": 10005632,
      "step": 17240
    },
    {
      "epoch": 2.5685135537682453,
      "grad_norm": 11.6875,
      "learning_rate": 4.9877096954553324e-05,
      "loss": 0.8899,
      "num_input_tokens_seen": 10008576,
      "step": 17245
    },
    {
      "epoch": 2.5692582663092045,
      "grad_norm": 9.0,
      "learning_rate": 4.987677493559883e-05,
      "loss": 1.3857,
      "num_input_tokens_seen": 10011456,
      "step": 17250
    },
    {
      "epoch": 2.5700029788501637,
      "grad_norm": 13.4375,
      "learning_rate": 4.987645249637652e-05,
      "loss": 1.1235,
      "num_input_tokens_seen": 10014496,
      "step": 17255
    },
    {
      "epoch": 2.570747691391123,
      "grad_norm": 14.75,
      "learning_rate": 4.987612963689182e-05,
      "loss": 1.1654,
      "num_input_tokens_seen": 10017664,
      "step": 17260
    },
    {
      "epoch": 2.571492403932082,
      "grad_norm": 7.21875,
      "learning_rate": 4.987580635715019e-05,
      "loss": 1.0614,
      "num_input_tokens_seen": 10020384,
      "step": 17265
    },
    {
      "epoch": 2.5722371164730413,
      "grad_norm": 10.125,
      "learning_rate": 4.98754826571571e-05,
      "loss": 1.2333,
      "num_input_tokens_seen": 10023360,
      "step": 17270
    },
    {
      "epoch": 2.5729818290140005,
      "grad_norm": 8.5625,
      "learning_rate": 4.9875158536918015e-05,
      "loss": 0.9547,
      "num_input_tokens_seen": 10026208,
      "step": 17275
    },
    {
      "epoch": 2.5737265415549597,
      "grad_norm": 13.3125,
      "learning_rate": 4.987483399643841e-05,
      "loss": 0.7291,
      "num_input_tokens_seen": 10029056,
      "step": 17280
    },
    {
      "epoch": 2.574471254095919,
      "grad_norm": 4.0,
      "learning_rate": 4.987450903572376e-05,
      "loss": 1.0213,
      "num_input_tokens_seen": 10031936,
      "step": 17285
    },
    {
      "epoch": 2.575215966636878,
      "grad_norm": 10.5,
      "learning_rate": 4.987418365477956e-05,
      "loss": 0.6944,
      "num_input_tokens_seen": 10034816,
      "step": 17290
    },
    {
      "epoch": 2.5759606791778373,
      "grad_norm": 4.4375,
      "learning_rate": 4.987385785361131e-05,
      "loss": 1.286,
      "num_input_tokens_seen": 10037856,
      "step": 17295
    },
    {
      "epoch": 2.5767053917187965,
      "grad_norm": 9.9375,
      "learning_rate": 4.987353163222451e-05,
      "loss": 0.8079,
      "num_input_tokens_seen": 10040576,
      "step": 17300
    },
    {
      "epoch": 2.5774501042597557,
      "grad_norm": 15.0625,
      "learning_rate": 4.9873204990624676e-05,
      "loss": 1.0964,
      "num_input_tokens_seen": 10043392,
      "step": 17305
    },
    {
      "epoch": 2.578194816800715,
      "grad_norm": 7.625,
      "learning_rate": 4.987287792881733e-05,
      "loss": 1.651,
      "num_input_tokens_seen": 10045888,
      "step": 17310
    },
    {
      "epoch": 2.578939529341674,
      "grad_norm": 6.09375,
      "learning_rate": 4.9872550446807986e-05,
      "loss": 1.2949,
      "num_input_tokens_seen": 10049056,
      "step": 17315
    },
    {
      "epoch": 2.5796842418826333,
      "grad_norm": 14.9375,
      "learning_rate": 4.987222254460218e-05,
      "loss": 0.914,
      "num_input_tokens_seen": 10052032,
      "step": 17320
    },
    {
      "epoch": 2.5804289544235925,
      "grad_norm": 5.65625,
      "learning_rate": 4.987189422220545e-05,
      "loss": 0.9032,
      "num_input_tokens_seen": 10054880,
      "step": 17325
    },
    {
      "epoch": 2.5811736669645517,
      "grad_norm": 6.25,
      "learning_rate": 4.987156547962335e-05,
      "loss": 1.0446,
      "num_input_tokens_seen": 10057280,
      "step": 17330
    },
    {
      "epoch": 2.581918379505511,
      "grad_norm": 16.0,
      "learning_rate": 4.987123631686143e-05,
      "loss": 0.8119,
      "num_input_tokens_seen": 10060288,
      "step": 17335
    },
    {
      "epoch": 2.58266309204647,
      "grad_norm": 7.59375,
      "learning_rate": 4.987090673392525e-05,
      "loss": 1.1154,
      "num_input_tokens_seen": 10063072,
      "step": 17340
    },
    {
      "epoch": 2.5834078045874294,
      "grad_norm": 8.375,
      "learning_rate": 4.987057673082038e-05,
      "loss": 1.0852,
      "num_input_tokens_seen": 10066208,
      "step": 17345
    },
    {
      "epoch": 2.5841525171283886,
      "grad_norm": 9.8125,
      "learning_rate": 4.987024630755239e-05,
      "loss": 1.0302,
      "num_input_tokens_seen": 10069120,
      "step": 17350
    },
    {
      "epoch": 2.5848972296693478,
      "grad_norm": 18.75,
      "learning_rate": 4.986991546412687e-05,
      "loss": 1.0552,
      "num_input_tokens_seen": 10072000,
      "step": 17355
    },
    {
      "epoch": 2.585641942210307,
      "grad_norm": 10.6875,
      "learning_rate": 4.98695842005494e-05,
      "loss": 1.0907,
      "num_input_tokens_seen": 10074720,
      "step": 17360
    },
    {
      "epoch": 2.586386654751266,
      "grad_norm": 7.75,
      "learning_rate": 4.9869252516825585e-05,
      "loss": 0.8638,
      "num_input_tokens_seen": 10077568,
      "step": 17365
    },
    {
      "epoch": 2.5871313672922254,
      "grad_norm": 10.0625,
      "learning_rate": 4.986892041296102e-05,
      "loss": 0.9335,
      "num_input_tokens_seen": 10080288,
      "step": 17370
    },
    {
      "epoch": 2.587876079833184,
      "grad_norm": 8.375,
      "learning_rate": 4.9868587888961325e-05,
      "loss": 0.9185,
      "num_input_tokens_seen": 10083104,
      "step": 17375
    },
    {
      "epoch": 2.5886207923741438,
      "grad_norm": 7.28125,
      "learning_rate": 4.986825494483211e-05,
      "loss": 1.0151,
      "num_input_tokens_seen": 10086080,
      "step": 17380
    },
    {
      "epoch": 2.5893655049151025,
      "grad_norm": 10.9375,
      "learning_rate": 4.9867921580579e-05,
      "loss": 1.1719,
      "num_input_tokens_seen": 10089056,
      "step": 17385
    },
    {
      "epoch": 2.590110217456062,
      "grad_norm": 5.0625,
      "learning_rate": 4.986758779620764e-05,
      "loss": 1.4206,
      "num_input_tokens_seen": 10091904,
      "step": 17390
    },
    {
      "epoch": 2.590854929997021,
      "grad_norm": 12.5625,
      "learning_rate": 4.986725359172365e-05,
      "loss": 1.0909,
      "num_input_tokens_seen": 10094656,
      "step": 17395
    },
    {
      "epoch": 2.5915996425379806,
      "grad_norm": 10.3125,
      "learning_rate": 4.986691896713269e-05,
      "loss": 0.8421,
      "num_input_tokens_seen": 10097408,
      "step": 17400
    },
    {
      "epoch": 2.5923443550789393,
      "grad_norm": 16.25,
      "learning_rate": 4.98665839224404e-05,
      "loss": 1.0418,
      "num_input_tokens_seen": 10100640,
      "step": 17405
    },
    {
      "epoch": 2.593089067619899,
      "grad_norm": 10.25,
      "learning_rate": 4.9866248457652455e-05,
      "loss": 0.7574,
      "num_input_tokens_seen": 10103328,
      "step": 17410
    },
    {
      "epoch": 2.5938337801608577,
      "grad_norm": 9.875,
      "learning_rate": 4.986591257277451e-05,
      "loss": 1.0655,
      "num_input_tokens_seen": 10106144,
      "step": 17415
    },
    {
      "epoch": 2.594578492701817,
      "grad_norm": 7.46875,
      "learning_rate": 4.986557626781224e-05,
      "loss": 1.4423,
      "num_input_tokens_seen": 10109216,
      "step": 17420
    },
    {
      "epoch": 2.595323205242776,
      "grad_norm": 7.75,
      "learning_rate": 4.9865239542771345e-05,
      "loss": 1.032,
      "num_input_tokens_seen": 10112096,
      "step": 17425
    },
    {
      "epoch": 2.5960679177837354,
      "grad_norm": 12.75,
      "learning_rate": 4.986490239765749e-05,
      "loss": 1.0429,
      "num_input_tokens_seen": 10114880,
      "step": 17430
    },
    {
      "epoch": 2.5968126303246946,
      "grad_norm": 16.375,
      "learning_rate": 4.9864564832476386e-05,
      "loss": 0.6941,
      "num_input_tokens_seen": 10117824,
      "step": 17435
    },
    {
      "epoch": 2.5975573428656538,
      "grad_norm": 13.625,
      "learning_rate": 4.986422684723373e-05,
      "loss": 1.3461,
      "num_input_tokens_seen": 10120736,
      "step": 17440
    },
    {
      "epoch": 2.598302055406613,
      "grad_norm": 6.53125,
      "learning_rate": 4.986388844193523e-05,
      "loss": 0.8798,
      "num_input_tokens_seen": 10123456,
      "step": 17445
    },
    {
      "epoch": 2.599046767947572,
      "grad_norm": 9.5,
      "learning_rate": 4.9863549616586604e-05,
      "loss": 1.0752,
      "num_input_tokens_seen": 10126464,
      "step": 17450
    },
    {
      "epoch": 2.5997914804885314,
      "grad_norm": 13.6875,
      "learning_rate": 4.986321037119358e-05,
      "loss": 1.2931,
      "num_input_tokens_seen": 10129408,
      "step": 17455
    },
    {
      "epoch": 2.6005361930294906,
      "grad_norm": 7.625,
      "learning_rate": 4.986287070576188e-05,
      "loss": 1.3276,
      "num_input_tokens_seen": 10132160,
      "step": 17460
    },
    {
      "epoch": 2.6012809055704498,
      "grad_norm": 16.375,
      "learning_rate": 4.986253062029725e-05,
      "loss": 1.4028,
      "num_input_tokens_seen": 10135200,
      "step": 17465
    },
    {
      "epoch": 2.602025618111409,
      "grad_norm": 6.875,
      "learning_rate": 4.986219011480544e-05,
      "loss": 1.3235,
      "num_input_tokens_seen": 10138112,
      "step": 17470
    },
    {
      "epoch": 2.602770330652368,
      "grad_norm": 13.625,
      "learning_rate": 4.98618491892922e-05,
      "loss": 1.1252,
      "num_input_tokens_seen": 10141184,
      "step": 17475
    },
    {
      "epoch": 2.6035150431933274,
      "grad_norm": 9.75,
      "learning_rate": 4.986150784376328e-05,
      "loss": 1.2277,
      "num_input_tokens_seen": 10143840,
      "step": 17480
    },
    {
      "epoch": 2.6042597557342866,
      "grad_norm": 11.4375,
      "learning_rate": 4.986116607822445e-05,
      "loss": 0.8264,
      "num_input_tokens_seen": 10146848,
      "step": 17485
    },
    {
      "epoch": 2.605004468275246,
      "grad_norm": 19.5,
      "learning_rate": 4.9860823892681496e-05,
      "loss": 1.3248,
      "num_input_tokens_seen": 10149952,
      "step": 17490
    },
    {
      "epoch": 2.605749180816205,
      "grad_norm": 3.6875,
      "learning_rate": 4.986048128714019e-05,
      "loss": 1.1781,
      "num_input_tokens_seen": 10152608,
      "step": 17495
    },
    {
      "epoch": 2.606493893357164,
      "grad_norm": 9.625,
      "learning_rate": 4.986013826160631e-05,
      "loss": 1.3545,
      "num_input_tokens_seen": 10156384,
      "step": 17500
    },
    {
      "epoch": 2.6072386058981234,
      "grad_norm": 7.625,
      "learning_rate": 4.985979481608567e-05,
      "loss": 1.033,
      "num_input_tokens_seen": 10159552,
      "step": 17505
    },
    {
      "epoch": 2.6079833184390826,
      "grad_norm": 11.875,
      "learning_rate": 4.9859450950584056e-05,
      "loss": 1.2628,
      "num_input_tokens_seen": 10162336,
      "step": 17510
    },
    {
      "epoch": 2.608728030980042,
      "grad_norm": 15.9375,
      "learning_rate": 4.9859106665107294e-05,
      "loss": 1.1231,
      "num_input_tokens_seen": 10165504,
      "step": 17515
    },
    {
      "epoch": 2.609472743521001,
      "grad_norm": 16.5,
      "learning_rate": 4.985876195966118e-05,
      "loss": 1.2544,
      "num_input_tokens_seen": 10168288,
      "step": 17520
    },
    {
      "epoch": 2.61021745606196,
      "grad_norm": 6.375,
      "learning_rate": 4.985841683425155e-05,
      "loss": 0.8186,
      "num_input_tokens_seen": 10170880,
      "step": 17525
    },
    {
      "epoch": 2.6109621686029194,
      "grad_norm": 8.375,
      "learning_rate": 4.9858071288884236e-05,
      "loss": 1.0541,
      "num_input_tokens_seen": 10173984,
      "step": 17530
    },
    {
      "epoch": 2.6117068811438786,
      "grad_norm": 12.625,
      "learning_rate": 4.985772532356507e-05,
      "loss": 0.9747,
      "num_input_tokens_seen": 10176800,
      "step": 17535
    },
    {
      "epoch": 2.612451593684838,
      "grad_norm": 9.625,
      "learning_rate": 4.9857378938299895e-05,
      "loss": 1.0657,
      "num_input_tokens_seen": 10179872,
      "step": 17540
    },
    {
      "epoch": 2.613196306225797,
      "grad_norm": 7.21875,
      "learning_rate": 4.985703213309457e-05,
      "loss": 1.3146,
      "num_input_tokens_seen": 10182816,
      "step": 17545
    },
    {
      "epoch": 2.6139410187667558,
      "grad_norm": 9.5625,
      "learning_rate": 4.9856684907954955e-05,
      "loss": 1.2523,
      "num_input_tokens_seen": 10185600,
      "step": 17550
    },
    {
      "epoch": 2.6146857313077154,
      "grad_norm": 11.9375,
      "learning_rate": 4.985633726288691e-05,
      "loss": 1.0302,
      "num_input_tokens_seen": 10188384,
      "step": 17555
    },
    {
      "epoch": 2.615430443848674,
      "grad_norm": 12.625,
      "learning_rate": 4.985598919789631e-05,
      "loss": 1.3514,
      "num_input_tokens_seen": 10191488,
      "step": 17560
    },
    {
      "epoch": 2.616175156389634,
      "grad_norm": 5.0625,
      "learning_rate": 4.9855640712989035e-05,
      "loss": 1.3376,
      "num_input_tokens_seen": 10194304,
      "step": 17565
    },
    {
      "epoch": 2.6169198689305926,
      "grad_norm": 14.4375,
      "learning_rate": 4.9855291808170966e-05,
      "loss": 1.1397,
      "num_input_tokens_seen": 10196960,
      "step": 17570
    },
    {
      "epoch": 2.6176645814715522,
      "grad_norm": 13.875,
      "learning_rate": 4.985494248344801e-05,
      "loss": 0.8192,
      "num_input_tokens_seen": 10199936,
      "step": 17575
    },
    {
      "epoch": 2.618409294012511,
      "grad_norm": 12.25,
      "learning_rate": 4.9854592738826054e-05,
      "loss": 1.1976,
      "num_input_tokens_seen": 10202912,
      "step": 17580
    },
    {
      "epoch": 2.6191540065534706,
      "grad_norm": 7.125,
      "learning_rate": 4.985424257431103e-05,
      "loss": 1.0573,
      "num_input_tokens_seen": 10205888,
      "step": 17585
    },
    {
      "epoch": 2.6198987190944294,
      "grad_norm": 10.9375,
      "learning_rate": 4.985389198990883e-05,
      "loss": 0.9909,
      "num_input_tokens_seen": 10209888,
      "step": 17590
    },
    {
      "epoch": 2.6206434316353886,
      "grad_norm": 12.5625,
      "learning_rate": 4.985354098562538e-05,
      "loss": 0.8789,
      "num_input_tokens_seen": 10212800,
      "step": 17595
    },
    {
      "epoch": 2.621388144176348,
      "grad_norm": 5.875,
      "learning_rate": 4.985318956146662e-05,
      "loss": 0.8157,
      "num_input_tokens_seen": 10215584,
      "step": 17600
    },
    {
      "epoch": 2.622132856717307,
      "grad_norm": 8.6875,
      "learning_rate": 4.9852837717438485e-05,
      "loss": 0.9551,
      "num_input_tokens_seen": 10218656,
      "step": 17605
    },
    {
      "epoch": 2.622877569258266,
      "grad_norm": 11.875,
      "learning_rate": 4.985248545354692e-05,
      "loss": 0.9289,
      "num_input_tokens_seen": 10221632,
      "step": 17610
    },
    {
      "epoch": 2.6236222817992254,
      "grad_norm": 11.1875,
      "learning_rate": 4.985213276979785e-05,
      "loss": 1.2183,
      "num_input_tokens_seen": 10224384,
      "step": 17615
    },
    {
      "epoch": 2.6243669943401846,
      "grad_norm": 15.875,
      "learning_rate": 4.985177966619727e-05,
      "loss": 0.9789,
      "num_input_tokens_seen": 10227392,
      "step": 17620
    },
    {
      "epoch": 2.625111706881144,
      "grad_norm": 19.375,
      "learning_rate": 4.985142614275114e-05,
      "loss": 1.0507,
      "num_input_tokens_seen": 10230080,
      "step": 17625
    },
    {
      "epoch": 2.625856419422103,
      "grad_norm": 14.25,
      "learning_rate": 4.985107219946541e-05,
      "loss": 0.9309,
      "num_input_tokens_seen": 10232928,
      "step": 17630
    },
    {
      "epoch": 2.626601131963062,
      "grad_norm": 7.90625,
      "learning_rate": 4.985071783634608e-05,
      "loss": 1.4134,
      "num_input_tokens_seen": 10236000,
      "step": 17635
    },
    {
      "epoch": 2.6273458445040214,
      "grad_norm": 8.5,
      "learning_rate": 4.985036305339913e-05,
      "loss": 1.0635,
      "num_input_tokens_seen": 10239008,
      "step": 17640
    },
    {
      "epoch": 2.6280905570449806,
      "grad_norm": 13.0,
      "learning_rate": 4.9850007850630545e-05,
      "loss": 1.1151,
      "num_input_tokens_seen": 10241888,
      "step": 17645
    },
    {
      "epoch": 2.62883526958594,
      "grad_norm": 9.6875,
      "learning_rate": 4.984965222804634e-05,
      "loss": 1.0763,
      "num_input_tokens_seen": 10244736,
      "step": 17650
    },
    {
      "epoch": 2.629579982126899,
      "grad_norm": 8.0625,
      "learning_rate": 4.984929618565252e-05,
      "loss": 1.1672,
      "num_input_tokens_seen": 10247424,
      "step": 17655
    },
    {
      "epoch": 2.6303246946678582,
      "grad_norm": 12.3125,
      "learning_rate": 4.9848939723455085e-05,
      "loss": 1.1464,
      "num_input_tokens_seen": 10250144,
      "step": 17660
    },
    {
      "epoch": 2.6310694072088174,
      "grad_norm": 8.8125,
      "learning_rate": 4.984858284146008e-05,
      "loss": 1.1731,
      "num_input_tokens_seen": 10253184,
      "step": 17665
    },
    {
      "epoch": 2.6318141197497766,
      "grad_norm": 11.625,
      "learning_rate": 4.9848225539673513e-05,
      "loss": 1.3604,
      "num_input_tokens_seen": 10256160,
      "step": 17670
    },
    {
      "epoch": 2.632558832290736,
      "grad_norm": 17.75,
      "learning_rate": 4.9847867818101436e-05,
      "loss": 1.0287,
      "num_input_tokens_seen": 10259072,
      "step": 17675
    },
    {
      "epoch": 2.633303544831695,
      "grad_norm": 9.4375,
      "learning_rate": 4.984750967674989e-05,
      "loss": 0.8821,
      "num_input_tokens_seen": 10262144,
      "step": 17680
    },
    {
      "epoch": 2.6340482573726542,
      "grad_norm": 18.125,
      "learning_rate": 4.9847151115624916e-05,
      "loss": 1.1633,
      "num_input_tokens_seen": 10265024,
      "step": 17685
    },
    {
      "epoch": 2.6347929699136134,
      "grad_norm": 11.0625,
      "learning_rate": 4.984679213473258e-05,
      "loss": 1.2085,
      "num_input_tokens_seen": 10267968,
      "step": 17690
    },
    {
      "epoch": 2.6355376824545726,
      "grad_norm": 12.0625,
      "learning_rate": 4.984643273407894e-05,
      "loss": 1.0705,
      "num_input_tokens_seen": 10270688,
      "step": 17695
    },
    {
      "epoch": 2.636282394995532,
      "grad_norm": 8.0,
      "learning_rate": 4.984607291367007e-05,
      "loss": 1.0736,
      "num_input_tokens_seen": 10273568,
      "step": 17700
    },
    {
      "epoch": 2.637027107536491,
      "grad_norm": 12.875,
      "learning_rate": 4.984571267351206e-05,
      "loss": 1.4194,
      "num_input_tokens_seen": 10276416,
      "step": 17705
    },
    {
      "epoch": 2.6377718200774503,
      "grad_norm": 9.4375,
      "learning_rate": 4.984535201361098e-05,
      "loss": 0.9473,
      "num_input_tokens_seen": 10279104,
      "step": 17710
    },
    {
      "epoch": 2.6385165326184095,
      "grad_norm": 9.1875,
      "learning_rate": 4.984499093397294e-05,
      "loss": 1.0385,
      "num_input_tokens_seen": 10282112,
      "step": 17715
    },
    {
      "epoch": 2.6392612451593687,
      "grad_norm": 11.625,
      "learning_rate": 4.984462943460402e-05,
      "loss": 1.0809,
      "num_input_tokens_seen": 10285216,
      "step": 17720
    },
    {
      "epoch": 2.6400059577003274,
      "grad_norm": 6.25,
      "learning_rate": 4.984426751551033e-05,
      "loss": 1.1585,
      "num_input_tokens_seen": 10287904,
      "step": 17725
    },
    {
      "epoch": 2.640750670241287,
      "grad_norm": 9.9375,
      "learning_rate": 4.9843905176698004e-05,
      "loss": 0.7135,
      "num_input_tokens_seen": 10290912,
      "step": 17730
    },
    {
      "epoch": 2.641495382782246,
      "grad_norm": 13.5,
      "learning_rate": 4.984354241817314e-05,
      "loss": 1.5575,
      "num_input_tokens_seen": 10294048,
      "step": 17735
    },
    {
      "epoch": 2.6422400953232055,
      "grad_norm": 8.25,
      "learning_rate": 4.984317923994188e-05,
      "loss": 0.9773,
      "num_input_tokens_seen": 10297056,
      "step": 17740
    },
    {
      "epoch": 2.6429848078641642,
      "grad_norm": 16.0,
      "learning_rate": 4.984281564201036e-05,
      "loss": 1.0748,
      "num_input_tokens_seen": 10299776,
      "step": 17745
    },
    {
      "epoch": 2.643729520405124,
      "grad_norm": 8.4375,
      "learning_rate": 4.9842451624384715e-05,
      "loss": 1.2266,
      "num_input_tokens_seen": 10302720,
      "step": 17750
    },
    {
      "epoch": 2.6444742329460826,
      "grad_norm": 12.25,
      "learning_rate": 4.98420871870711e-05,
      "loss": 1.2395,
      "num_input_tokens_seen": 10305472,
      "step": 17755
    },
    {
      "epoch": 2.645218945487042,
      "grad_norm": 16.25,
      "learning_rate": 4.984172233007567e-05,
      "loss": 1.3534,
      "num_input_tokens_seen": 10308192,
      "step": 17760
    },
    {
      "epoch": 2.645963658028001,
      "grad_norm": 7.9375,
      "learning_rate": 4.984135705340459e-05,
      "loss": 1.1833,
      "num_input_tokens_seen": 10311296,
      "step": 17765
    },
    {
      "epoch": 2.6467083705689602,
      "grad_norm": 15.9375,
      "learning_rate": 4.984099135706402e-05,
      "loss": 1.4578,
      "num_input_tokens_seen": 10313984,
      "step": 17770
    },
    {
      "epoch": 2.6474530831099194,
      "grad_norm": 6.3125,
      "learning_rate": 4.984062524106017e-05,
      "loss": 0.8014,
      "num_input_tokens_seen": 10316768,
      "step": 17775
    },
    {
      "epoch": 2.6481977956508786,
      "grad_norm": 13.625,
      "learning_rate": 4.984025870539919e-05,
      "loss": 1.3812,
      "num_input_tokens_seen": 10319776,
      "step": 17780
    },
    {
      "epoch": 2.648942508191838,
      "grad_norm": 13.0,
      "learning_rate": 4.983989175008729e-05,
      "loss": 1.1044,
      "num_input_tokens_seen": 10322368,
      "step": 17785
    },
    {
      "epoch": 2.649687220732797,
      "grad_norm": 9.0625,
      "learning_rate": 4.983952437513066e-05,
      "loss": 1.3277,
      "num_input_tokens_seen": 10325600,
      "step": 17790
    },
    {
      "epoch": 2.6504319332737563,
      "grad_norm": 8.5,
      "learning_rate": 4.983915658053551e-05,
      "loss": 0.9881,
      "num_input_tokens_seen": 10328608,
      "step": 17795
    },
    {
      "epoch": 2.6511766458147155,
      "grad_norm": 11.9375,
      "learning_rate": 4.983878836630806e-05,
      "loss": 1.0767,
      "num_input_tokens_seen": 10331456,
      "step": 17800
    },
    {
      "epoch": 2.6519213583556747,
      "grad_norm": 12.0625,
      "learning_rate": 4.983841973245452e-05,
      "loss": 1.2743,
      "num_input_tokens_seen": 10334336,
      "step": 17805
    },
    {
      "epoch": 2.652666070896634,
      "grad_norm": 7.625,
      "learning_rate": 4.983805067898113e-05,
      "loss": 0.9133,
      "num_input_tokens_seen": 10337152,
      "step": 17810
    },
    {
      "epoch": 2.653410783437593,
      "grad_norm": 5.125,
      "learning_rate": 4.983768120589411e-05,
      "loss": 1.0471,
      "num_input_tokens_seen": 10340000,
      "step": 17815
    },
    {
      "epoch": 2.6541554959785523,
      "grad_norm": 10.0,
      "learning_rate": 4.983731131319972e-05,
      "loss": 1.621,
      "num_input_tokens_seen": 10342880,
      "step": 17820
    },
    {
      "epoch": 2.6549002085195115,
      "grad_norm": 6.6875,
      "learning_rate": 4.98369410009042e-05,
      "loss": 0.592,
      "num_input_tokens_seen": 10345824,
      "step": 17825
    },
    {
      "epoch": 2.6556449210604707,
      "grad_norm": 13.0,
      "learning_rate": 4.9836570269013796e-05,
      "loss": 0.7643,
      "num_input_tokens_seen": 10348608,
      "step": 17830
    },
    {
      "epoch": 2.65638963360143,
      "grad_norm": 10.625,
      "learning_rate": 4.983619911753478e-05,
      "loss": 0.9205,
      "num_input_tokens_seen": 10351648,
      "step": 17835
    },
    {
      "epoch": 2.657134346142389,
      "grad_norm": 16.125,
      "learning_rate": 4.983582754647343e-05,
      "loss": 1.1103,
      "num_input_tokens_seen": 10354752,
      "step": 17840
    },
    {
      "epoch": 2.6578790586833483,
      "grad_norm": 6.65625,
      "learning_rate": 4.983545555583601e-05,
      "loss": 1.298,
      "num_input_tokens_seen": 10357632,
      "step": 17845
    },
    {
      "epoch": 2.6586237712243075,
      "grad_norm": 15.875,
      "learning_rate": 4.9835083145628816e-05,
      "loss": 1.2839,
      "num_input_tokens_seen": 10360736,
      "step": 17850
    },
    {
      "epoch": 2.6593684837652667,
      "grad_norm": 7.59375,
      "learning_rate": 4.9834710315858125e-05,
      "loss": 1.2456,
      "num_input_tokens_seen": 10363520,
      "step": 17855
    },
    {
      "epoch": 2.660113196306226,
      "grad_norm": 12.9375,
      "learning_rate": 4.983433706653024e-05,
      "loss": 1.1492,
      "num_input_tokens_seen": 10366400,
      "step": 17860
    },
    {
      "epoch": 2.660857908847185,
      "grad_norm": 6.21875,
      "learning_rate": 4.9833963397651485e-05,
      "loss": 1.0047,
      "num_input_tokens_seen": 10369248,
      "step": 17865
    },
    {
      "epoch": 2.6616026213881443,
      "grad_norm": 15.0,
      "learning_rate": 4.9833589309228154e-05,
      "loss": 1.4143,
      "num_input_tokens_seen": 10372160,
      "step": 17870
    },
    {
      "epoch": 2.6623473339291035,
      "grad_norm": 13.25,
      "learning_rate": 4.9833214801266565e-05,
      "loss": 0.9142,
      "num_input_tokens_seen": 10374848,
      "step": 17875
    },
    {
      "epoch": 2.6630920464700627,
      "grad_norm": 17.5,
      "learning_rate": 4.9832839873773054e-05,
      "loss": 1.2779,
      "num_input_tokens_seen": 10377600,
      "step": 17880
    },
    {
      "epoch": 2.663836759011022,
      "grad_norm": 12.0625,
      "learning_rate": 4.983246452675395e-05,
      "loss": 1.0683,
      "num_input_tokens_seen": 10380736,
      "step": 17885
    },
    {
      "epoch": 2.6645814715519807,
      "grad_norm": 7.34375,
      "learning_rate": 4.983208876021561e-05,
      "loss": 1.0711,
      "num_input_tokens_seen": 10383808,
      "step": 17890
    },
    {
      "epoch": 2.6653261840929403,
      "grad_norm": 12.25,
      "learning_rate": 4.983171257416436e-05,
      "loss": 1.0756,
      "num_input_tokens_seen": 10386720,
      "step": 17895
    },
    {
      "epoch": 2.666070896633899,
      "grad_norm": 15.0,
      "learning_rate": 4.983133596860656e-05,
      "loss": 1.5392,
      "num_input_tokens_seen": 10389696,
      "step": 17900
    },
    {
      "epoch": 2.6668156091748587,
      "grad_norm": 8.875,
      "learning_rate": 4.983095894354858e-05,
      "loss": 1.6628,
      "num_input_tokens_seen": 10392544,
      "step": 17905
    },
    {
      "epoch": 2.6675603217158175,
      "grad_norm": 11.75,
      "learning_rate": 4.9830581498996784e-05,
      "loss": 1.4958,
      "num_input_tokens_seen": 10395296,
      "step": 17910
    },
    {
      "epoch": 2.668305034256777,
      "grad_norm": 12.1875,
      "learning_rate": 4.983020363495755e-05,
      "loss": 0.9797,
      "num_input_tokens_seen": 10398304,
      "step": 17915
    },
    {
      "epoch": 2.669049746797736,
      "grad_norm": 13.0625,
      "learning_rate": 4.982982535143727e-05,
      "loss": 1.4829,
      "num_input_tokens_seen": 10401088,
      "step": 17920
    },
    {
      "epoch": 2.6697944593386955,
      "grad_norm": 7.4375,
      "learning_rate": 4.982944664844231e-05,
      "loss": 0.98,
      "num_input_tokens_seen": 10403808,
      "step": 17925
    },
    {
      "epoch": 2.6705391718796543,
      "grad_norm": 15.5625,
      "learning_rate": 4.98290675259791e-05,
      "loss": 1.3291,
      "num_input_tokens_seen": 10406752,
      "step": 17930
    },
    {
      "epoch": 2.6712838844206135,
      "grad_norm": 14.375,
      "learning_rate": 4.9828687984054015e-05,
      "loss": 1.1477,
      "num_input_tokens_seen": 10409440,
      "step": 17935
    },
    {
      "epoch": 2.6720285969615727,
      "grad_norm": 5.46875,
      "learning_rate": 4.9828308022673494e-05,
      "loss": 1.2114,
      "num_input_tokens_seen": 10412128,
      "step": 17940
    },
    {
      "epoch": 2.672773309502532,
      "grad_norm": 18.125,
      "learning_rate": 4.9827927641843944e-05,
      "loss": 0.9685,
      "num_input_tokens_seen": 10414752,
      "step": 17945
    },
    {
      "epoch": 2.673518022043491,
      "grad_norm": 10.0625,
      "learning_rate": 4.982754684157178e-05,
      "loss": 0.9261,
      "num_input_tokens_seen": 10417312,
      "step": 17950
    },
    {
      "epoch": 2.6742627345844503,
      "grad_norm": 11.3125,
      "learning_rate": 4.982716562186345e-05,
      "loss": 1.6758,
      "num_input_tokens_seen": 10420288,
      "step": 17955
    },
    {
      "epoch": 2.6750074471254095,
      "grad_norm": 14.0,
      "learning_rate": 4.982678398272539e-05,
      "loss": 1.0474,
      "num_input_tokens_seen": 10423264,
      "step": 17960
    },
    {
      "epoch": 2.6757521596663687,
      "grad_norm": 13.75,
      "learning_rate": 4.982640192416404e-05,
      "loss": 1.3268,
      "num_input_tokens_seen": 10426336,
      "step": 17965
    },
    {
      "epoch": 2.676496872207328,
      "grad_norm": 9.4375,
      "learning_rate": 4.982601944618588e-05,
      "loss": 0.9542,
      "num_input_tokens_seen": 10429312,
      "step": 17970
    },
    {
      "epoch": 2.677241584748287,
      "grad_norm": 16.125,
      "learning_rate": 4.982563654879734e-05,
      "loss": 0.951,
      "num_input_tokens_seen": 10432256,
      "step": 17975
    },
    {
      "epoch": 2.6779862972892463,
      "grad_norm": 12.375,
      "learning_rate": 4.982525323200491e-05,
      "loss": 0.882,
      "num_input_tokens_seen": 10434880,
      "step": 17980
    },
    {
      "epoch": 2.6787310098302055,
      "grad_norm": 12.25,
      "learning_rate": 4.982486949581505e-05,
      "loss": 1.2351,
      "num_input_tokens_seen": 10437920,
      "step": 17985
    },
    {
      "epoch": 2.6794757223711647,
      "grad_norm": 21.375,
      "learning_rate": 4.982448534023426e-05,
      "loss": 1.2568,
      "num_input_tokens_seen": 10440928,
      "step": 17990
    },
    {
      "epoch": 2.680220434912124,
      "grad_norm": 10.8125,
      "learning_rate": 4.982410076526901e-05,
      "loss": 1.1039,
      "num_input_tokens_seen": 10443520,
      "step": 17995
    },
    {
      "epoch": 2.680965147453083,
      "grad_norm": 9.0,
      "learning_rate": 4.9823715770925814e-05,
      "loss": 1.2639,
      "num_input_tokens_seen": 10446464,
      "step": 18000
    },
    {
      "epoch": 2.6817098599940423,
      "grad_norm": 11.3125,
      "learning_rate": 4.982333035721117e-05,
      "loss": 0.8294,
      "num_input_tokens_seen": 10449152,
      "step": 18005
    },
    {
      "epoch": 2.6824545725350015,
      "grad_norm": 13.625,
      "learning_rate": 4.982294452413159e-05,
      "loss": 1.0819,
      "num_input_tokens_seen": 10451776,
      "step": 18010
    },
    {
      "epoch": 2.6831992850759607,
      "grad_norm": 7.96875,
      "learning_rate": 4.982255827169359e-05,
      "loss": 0.9446,
      "num_input_tokens_seen": 10454432,
      "step": 18015
    },
    {
      "epoch": 2.68394399761692,
      "grad_norm": 5.28125,
      "learning_rate": 4.982217159990369e-05,
      "loss": 1.1091,
      "num_input_tokens_seen": 10457184,
      "step": 18020
    },
    {
      "epoch": 2.684688710157879,
      "grad_norm": 9.25,
      "learning_rate": 4.982178450876843e-05,
      "loss": 1.4583,
      "num_input_tokens_seen": 10460064,
      "step": 18025
    },
    {
      "epoch": 2.6854334226988383,
      "grad_norm": 3.515625,
      "learning_rate": 4.9821396998294356e-05,
      "loss": 1.0576,
      "num_input_tokens_seen": 10462848,
      "step": 18030
    },
    {
      "epoch": 2.6861781352397975,
      "grad_norm": 9.6875,
      "learning_rate": 4.982100906848801e-05,
      "loss": 0.7155,
      "num_input_tokens_seen": 10465472,
      "step": 18035
    },
    {
      "epoch": 2.6869228477807567,
      "grad_norm": 9.5,
      "learning_rate": 4.9820620719355934e-05,
      "loss": 1.1407,
      "num_input_tokens_seen": 10468384,
      "step": 18040
    },
    {
      "epoch": 2.687667560321716,
      "grad_norm": 8.6875,
      "learning_rate": 4.982023195090469e-05,
      "loss": 1.0674,
      "num_input_tokens_seen": 10471424,
      "step": 18045
    },
    {
      "epoch": 2.688412272862675,
      "grad_norm": 21.0,
      "learning_rate": 4.981984276314087e-05,
      "loss": 1.0715,
      "num_input_tokens_seen": 10474272,
      "step": 18050
    },
    {
      "epoch": 2.6891569854036343,
      "grad_norm": 5.5625,
      "learning_rate": 4.981945315607103e-05,
      "loss": 0.9403,
      "num_input_tokens_seen": 10477312,
      "step": 18055
    },
    {
      "epoch": 2.6899016979445936,
      "grad_norm": 11.8125,
      "learning_rate": 4.981906312970175e-05,
      "loss": 1.1543,
      "num_input_tokens_seen": 10480128,
      "step": 18060
    },
    {
      "epoch": 2.6906464104855523,
      "grad_norm": 4.71875,
      "learning_rate": 4.981867268403962e-05,
      "loss": 0.6724,
      "num_input_tokens_seen": 10483264,
      "step": 18065
    },
    {
      "epoch": 2.691391123026512,
      "grad_norm": 13.0625,
      "learning_rate": 4.981828181909124e-05,
      "loss": 1.2558,
      "num_input_tokens_seen": 10486272,
      "step": 18070
    },
    {
      "epoch": 2.6921358355674707,
      "grad_norm": 8.25,
      "learning_rate": 4.981789053486322e-05,
      "loss": 1.0039,
      "num_input_tokens_seen": 10489344,
      "step": 18075
    },
    {
      "epoch": 2.6928805481084304,
      "grad_norm": 9.4375,
      "learning_rate": 4.981749883136215e-05,
      "loss": 0.9668,
      "num_input_tokens_seen": 10492128,
      "step": 18080
    },
    {
      "epoch": 2.693625260649389,
      "grad_norm": 6.96875,
      "learning_rate": 4.981710670859467e-05,
      "loss": 1.1675,
      "num_input_tokens_seen": 10495104,
      "step": 18085
    },
    {
      "epoch": 2.6943699731903488,
      "grad_norm": 19.375,
      "learning_rate": 4.98167141665674e-05,
      "loss": 1.4656,
      "num_input_tokens_seen": 10497888,
      "step": 18090
    },
    {
      "epoch": 2.6951146857313075,
      "grad_norm": 10.3125,
      "learning_rate": 4.981632120528696e-05,
      "loss": 1.3094,
      "num_input_tokens_seen": 10500864,
      "step": 18095
    },
    {
      "epoch": 2.695859398272267,
      "grad_norm": 9.375,
      "learning_rate": 4.981592782476e-05,
      "loss": 1.1253,
      "num_input_tokens_seen": 10503456,
      "step": 18100
    },
    {
      "epoch": 2.696604110813226,
      "grad_norm": 11.8125,
      "learning_rate": 4.981553402499316e-05,
      "loss": 0.7484,
      "num_input_tokens_seen": 10506784,
      "step": 18105
    },
    {
      "epoch": 2.697348823354185,
      "grad_norm": 8.0625,
      "learning_rate": 4.9815139805993086e-05,
      "loss": 1.1911,
      "num_input_tokens_seen": 10509728,
      "step": 18110
    },
    {
      "epoch": 2.6980935358951443,
      "grad_norm": 6.25,
      "learning_rate": 4.9814745167766455e-05,
      "loss": 0.9596,
      "num_input_tokens_seen": 10512512,
      "step": 18115
    },
    {
      "epoch": 2.6988382484361035,
      "grad_norm": 19.0,
      "learning_rate": 4.981435011031992e-05,
      "loss": 1.5762,
      "num_input_tokens_seen": 10515488,
      "step": 18120
    },
    {
      "epoch": 2.6995829609770627,
      "grad_norm": 18.0,
      "learning_rate": 4.9813954633660166e-05,
      "loss": 1.29,
      "num_input_tokens_seen": 10518336,
      "step": 18125
    },
    {
      "epoch": 2.700327673518022,
      "grad_norm": 8.4375,
      "learning_rate": 4.9813558737793865e-05,
      "loss": 1.1062,
      "num_input_tokens_seen": 10521248,
      "step": 18130
    },
    {
      "epoch": 2.701072386058981,
      "grad_norm": 11.4375,
      "learning_rate": 4.9813162422727705e-05,
      "loss": 1.2107,
      "num_input_tokens_seen": 10523968,
      "step": 18135
    },
    {
      "epoch": 2.7018170985999403,
      "grad_norm": 5.03125,
      "learning_rate": 4.981276568846839e-05,
      "loss": 1.0526,
      "num_input_tokens_seen": 10526624,
      "step": 18140
    },
    {
      "epoch": 2.7025618111408996,
      "grad_norm": 8.9375,
      "learning_rate": 4.981236853502261e-05,
      "loss": 1.4042,
      "num_input_tokens_seen": 10529344,
      "step": 18145
    },
    {
      "epoch": 2.7033065236818588,
      "grad_norm": 6.5,
      "learning_rate": 4.9811970962397095e-05,
      "loss": 1.1841,
      "num_input_tokens_seen": 10532224,
      "step": 18150
    },
    {
      "epoch": 2.704051236222818,
      "grad_norm": 10.75,
      "learning_rate": 4.981157297059853e-05,
      "loss": 0.8586,
      "num_input_tokens_seen": 10535040,
      "step": 18155
    },
    {
      "epoch": 2.704795948763777,
      "grad_norm": 14.25,
      "learning_rate": 4.981117455963367e-05,
      "loss": 0.9624,
      "num_input_tokens_seen": 10537888,
      "step": 18160
    },
    {
      "epoch": 2.7055406613047364,
      "grad_norm": 7.625,
      "learning_rate": 4.981077572950923e-05,
      "loss": 1.2956,
      "num_input_tokens_seen": 10540640,
      "step": 18165
    },
    {
      "epoch": 2.7062853738456956,
      "grad_norm": 7.53125,
      "learning_rate": 4.9810376480231944e-05,
      "loss": 1.0712,
      "num_input_tokens_seen": 10543552,
      "step": 18170
    },
    {
      "epoch": 2.7070300863866548,
      "grad_norm": 7.90625,
      "learning_rate": 4.980997681180858e-05,
      "loss": 1.1145,
      "num_input_tokens_seen": 10546432,
      "step": 18175
    },
    {
      "epoch": 2.707774798927614,
      "grad_norm": 8.75,
      "learning_rate": 4.980957672424586e-05,
      "loss": 1.0457,
      "num_input_tokens_seen": 10549408,
      "step": 18180
    },
    {
      "epoch": 2.708519511468573,
      "grad_norm": 13.375,
      "learning_rate": 4.980917621755056e-05,
      "loss": 1.2812,
      "num_input_tokens_seen": 10552224,
      "step": 18185
    },
    {
      "epoch": 2.7092642240095324,
      "grad_norm": 9.6875,
      "learning_rate": 4.9808775291729445e-05,
      "loss": 0.8561,
      "num_input_tokens_seen": 10554848,
      "step": 18190
    },
    {
      "epoch": 2.7100089365504916,
      "grad_norm": 11.5,
      "learning_rate": 4.980837394678928e-05,
      "loss": 1.2115,
      "num_input_tokens_seen": 10557760,
      "step": 18195
    },
    {
      "epoch": 2.710753649091451,
      "grad_norm": 6.4375,
      "learning_rate": 4.980797218273685e-05,
      "loss": 1.0071,
      "num_input_tokens_seen": 10560992,
      "step": 18200
    },
    {
      "epoch": 2.71149836163241,
      "grad_norm": 7.03125,
      "learning_rate": 4.980756999957895e-05,
      "loss": 0.8751,
      "num_input_tokens_seen": 10563488,
      "step": 18205
    },
    {
      "epoch": 2.712243074173369,
      "grad_norm": 11.4375,
      "learning_rate": 4.9807167397322376e-05,
      "loss": 1.3043,
      "num_input_tokens_seen": 10566304,
      "step": 18210
    },
    {
      "epoch": 2.7129877867143284,
      "grad_norm": 10.875,
      "learning_rate": 4.980676437597391e-05,
      "loss": 0.9929,
      "num_input_tokens_seen": 10568832,
      "step": 18215
    },
    {
      "epoch": 2.7137324992552876,
      "grad_norm": 7.6875,
      "learning_rate": 4.980636093554038e-05,
      "loss": 0.987,
      "num_input_tokens_seen": 10571808,
      "step": 18220
    },
    {
      "epoch": 2.714477211796247,
      "grad_norm": 15.6875,
      "learning_rate": 4.980595707602858e-05,
      "loss": 0.9858,
      "num_input_tokens_seen": 10574624,
      "step": 18225
    },
    {
      "epoch": 2.715221924337206,
      "grad_norm": 6.75,
      "learning_rate": 4.980555279744535e-05,
      "loss": 1.0497,
      "num_input_tokens_seen": 10577632,
      "step": 18230
    },
    {
      "epoch": 2.715966636878165,
      "grad_norm": 8.5625,
      "learning_rate": 4.980514809979753e-05,
      "loss": 0.9534,
      "num_input_tokens_seen": 10580352,
      "step": 18235
    },
    {
      "epoch": 2.716711349419124,
      "grad_norm": 5.0625,
      "learning_rate": 4.9804742983091934e-05,
      "loss": 1.0024,
      "num_input_tokens_seen": 10583328,
      "step": 18240
    },
    {
      "epoch": 2.7174560619600836,
      "grad_norm": 9.1875,
      "learning_rate": 4.9804337447335414e-05,
      "loss": 1.0167,
      "num_input_tokens_seen": 10586144,
      "step": 18245
    },
    {
      "epoch": 2.7182007745010424,
      "grad_norm": 28.75,
      "learning_rate": 4.980393149253483e-05,
      "loss": 1.3582,
      "num_input_tokens_seen": 10589408,
      "step": 18250
    },
    {
      "epoch": 2.718945487042002,
      "grad_norm": 19.625,
      "learning_rate": 4.980352511869703e-05,
      "loss": 1.028,
      "num_input_tokens_seen": 10592480,
      "step": 18255
    },
    {
      "epoch": 2.7196901995829608,
      "grad_norm": 15.6875,
      "learning_rate": 4.980311832582888e-05,
      "loss": 1.4085,
      "num_input_tokens_seen": 10595488,
      "step": 18260
    },
    {
      "epoch": 2.7204349121239204,
      "grad_norm": 7.75,
      "learning_rate": 4.980271111393726e-05,
      "loss": 0.9784,
      "num_input_tokens_seen": 10598880,
      "step": 18265
    },
    {
      "epoch": 2.721179624664879,
      "grad_norm": 11.125,
      "learning_rate": 4.980230348302904e-05,
      "loss": 1.0444,
      "num_input_tokens_seen": 10601632,
      "step": 18270
    },
    {
      "epoch": 2.721924337205839,
      "grad_norm": 7.40625,
      "learning_rate": 4.9801895433111115e-05,
      "loss": 1.2578,
      "num_input_tokens_seen": 10604544,
      "step": 18275
    },
    {
      "epoch": 2.7226690497467976,
      "grad_norm": 15.4375,
      "learning_rate": 4.9801486964190366e-05,
      "loss": 1.2478,
      "num_input_tokens_seen": 10607904,
      "step": 18280
    },
    {
      "epoch": 2.723413762287757,
      "grad_norm": 10.875,
      "learning_rate": 4.9801078076273704e-05,
      "loss": 0.8532,
      "num_input_tokens_seen": 10610752,
      "step": 18285
    },
    {
      "epoch": 2.724158474828716,
      "grad_norm": 14.3125,
      "learning_rate": 4.980066876936804e-05,
      "loss": 0.9032,
      "num_input_tokens_seen": 10613376,
      "step": 18290
    },
    {
      "epoch": 2.724903187369675,
      "grad_norm": 10.75,
      "learning_rate": 4.980025904348028e-05,
      "loss": 1.2846,
      "num_input_tokens_seen": 10616064,
      "step": 18295
    },
    {
      "epoch": 2.7256478999106344,
      "grad_norm": 8.0625,
      "learning_rate": 4.979984889861735e-05,
      "loss": 0.9438,
      "num_input_tokens_seen": 10618848,
      "step": 18300
    },
    {
      "epoch": 2.7263926124515936,
      "grad_norm": 11.8125,
      "learning_rate": 4.9799438334786174e-05,
      "loss": 1.239,
      "num_input_tokens_seen": 10621568,
      "step": 18305
    },
    {
      "epoch": 2.727137324992553,
      "grad_norm": 13.4375,
      "learning_rate": 4.97990273519937e-05,
      "loss": 1.0805,
      "num_input_tokens_seen": 10624448,
      "step": 18310
    },
    {
      "epoch": 2.727882037533512,
      "grad_norm": 10.0625,
      "learning_rate": 4.9798615950246855e-05,
      "loss": 1.1527,
      "num_input_tokens_seen": 10627648,
      "step": 18315
    },
    {
      "epoch": 2.728626750074471,
      "grad_norm": 11.875,
      "learning_rate": 4.97982041295526e-05,
      "loss": 1.3283,
      "num_input_tokens_seen": 10630400,
      "step": 18320
    },
    {
      "epoch": 2.7293714626154304,
      "grad_norm": 22.375,
      "learning_rate": 4.97977918899179e-05,
      "loss": 1.1472,
      "num_input_tokens_seen": 10633248,
      "step": 18325
    },
    {
      "epoch": 2.7301161751563896,
      "grad_norm": 17.5,
      "learning_rate": 4.97973792313497e-05,
      "loss": 0.859,
      "num_input_tokens_seen": 10636448,
      "step": 18330
    },
    {
      "epoch": 2.730860887697349,
      "grad_norm": 12.5625,
      "learning_rate": 4.979696615385499e-05,
      "loss": 1.0396,
      "num_input_tokens_seen": 10639136,
      "step": 18335
    },
    {
      "epoch": 2.731605600238308,
      "grad_norm": 9.75,
      "learning_rate": 4.979655265744072e-05,
      "loss": 1.0977,
      "num_input_tokens_seen": 10641952,
      "step": 18340
    },
    {
      "epoch": 2.732350312779267,
      "grad_norm": 11.3125,
      "learning_rate": 4.979613874211391e-05,
      "loss": 1.2279,
      "num_input_tokens_seen": 10644480,
      "step": 18345
    },
    {
      "epoch": 2.7330950253202264,
      "grad_norm": 12.5,
      "learning_rate": 4.979572440788154e-05,
      "loss": 0.9441,
      "num_input_tokens_seen": 10647488,
      "step": 18350
    },
    {
      "epoch": 2.7338397378611856,
      "grad_norm": 6.6875,
      "learning_rate": 4.97953096547506e-05,
      "loss": 1.0487,
      "num_input_tokens_seen": 10650368,
      "step": 18355
    },
    {
      "epoch": 2.734584450402145,
      "grad_norm": 5.9375,
      "learning_rate": 4.9794894482728105e-05,
      "loss": 0.9909,
      "num_input_tokens_seen": 10653344,
      "step": 18360
    },
    {
      "epoch": 2.735329162943104,
      "grad_norm": 18.25,
      "learning_rate": 4.979447889182107e-05,
      "loss": 1.1446,
      "num_input_tokens_seen": 10656352,
      "step": 18365
    },
    {
      "epoch": 2.7360738754840632,
      "grad_norm": 16.625,
      "learning_rate": 4.979406288203651e-05,
      "loss": 0.9749,
      "num_input_tokens_seen": 10659264,
      "step": 18370
    },
    {
      "epoch": 2.7368185880250224,
      "grad_norm": 10.375,
      "learning_rate": 4.979364645338146e-05,
      "loss": 1.1619,
      "num_input_tokens_seen": 10662176,
      "step": 18375
    },
    {
      "epoch": 2.7375633005659816,
      "grad_norm": 11.0625,
      "learning_rate": 4.979322960586296e-05,
      "loss": 0.9704,
      "num_input_tokens_seen": 10665024,
      "step": 18380
    },
    {
      "epoch": 2.738308013106941,
      "grad_norm": 16.125,
      "learning_rate": 4.979281233948803e-05,
      "loss": 1.0188,
      "num_input_tokens_seen": 10667904,
      "step": 18385
    },
    {
      "epoch": 2.7390527256479,
      "grad_norm": 3.9375,
      "learning_rate": 4.9792394654263744e-05,
      "loss": 0.9085,
      "num_input_tokens_seen": 10670656,
      "step": 18390
    },
    {
      "epoch": 2.7397974381888592,
      "grad_norm": 7.9375,
      "learning_rate": 4.9791976550197144e-05,
      "loss": 1.2148,
      "num_input_tokens_seen": 10673408,
      "step": 18395
    },
    {
      "epoch": 2.7405421507298184,
      "grad_norm": 11.625,
      "learning_rate": 4.9791558027295296e-05,
      "loss": 1.1543,
      "num_input_tokens_seen": 10676128,
      "step": 18400
    },
    {
      "epoch": 2.7412868632707776,
      "grad_norm": 10.5,
      "learning_rate": 4.9791139085565274e-05,
      "loss": 1.2091,
      "num_input_tokens_seen": 10679072,
      "step": 18405
    },
    {
      "epoch": 2.742031575811737,
      "grad_norm": 13.875,
      "learning_rate": 4.9790719725014154e-05,
      "loss": 1.2701,
      "num_input_tokens_seen": 10682080,
      "step": 18410
    },
    {
      "epoch": 2.7427762883526956,
      "grad_norm": 6.78125,
      "learning_rate": 4.979029994564902e-05,
      "loss": 0.9581,
      "num_input_tokens_seen": 10684864,
      "step": 18415
    },
    {
      "epoch": 2.7435210008936552,
      "grad_norm": 8.4375,
      "learning_rate": 4.978987974747697e-05,
      "loss": 1.3366,
      "num_input_tokens_seen": 10687488,
      "step": 18420
    },
    {
      "epoch": 2.744265713434614,
      "grad_norm": 6.09375,
      "learning_rate": 4.9789459130505086e-05,
      "loss": 0.9427,
      "num_input_tokens_seen": 10690560,
      "step": 18425
    },
    {
      "epoch": 2.7450104259755737,
      "grad_norm": 10.4375,
      "learning_rate": 4.97890380947405e-05,
      "loss": 1.0524,
      "num_input_tokens_seen": 10693632,
      "step": 18430
    },
    {
      "epoch": 2.7457551385165324,
      "grad_norm": 19.25,
      "learning_rate": 4.97886166401903e-05,
      "loss": 1.2297,
      "num_input_tokens_seen": 10696640,
      "step": 18435
    },
    {
      "epoch": 2.746499851057492,
      "grad_norm": 10.8125,
      "learning_rate": 4.978819476686162e-05,
      "loss": 1.6409,
      "num_input_tokens_seen": 10699456,
      "step": 18440
    },
    {
      "epoch": 2.747244563598451,
      "grad_norm": 15.125,
      "learning_rate": 4.9787772474761575e-05,
      "loss": 1.2215,
      "num_input_tokens_seen": 10702176,
      "step": 18445
    },
    {
      "epoch": 2.7479892761394105,
      "grad_norm": 9.0625,
      "learning_rate": 4.978734976389732e-05,
      "loss": 0.8816,
      "num_input_tokens_seen": 10705184,
      "step": 18450
    },
    {
      "epoch": 2.7487339886803692,
      "grad_norm": 11.375,
      "learning_rate": 4.9786926634275964e-05,
      "loss": 1.0373,
      "num_input_tokens_seen": 10708352,
      "step": 18455
    },
    {
      "epoch": 2.7494787012213284,
      "grad_norm": 6.625,
      "learning_rate": 4.978650308590469e-05,
      "loss": 1.2032,
      "num_input_tokens_seen": 10711136,
      "step": 18460
    },
    {
      "epoch": 2.7502234137622876,
      "grad_norm": 7.09375,
      "learning_rate": 4.9786079118790635e-05,
      "loss": 1.1172,
      "num_input_tokens_seen": 10714016,
      "step": 18465
    },
    {
      "epoch": 2.750968126303247,
      "grad_norm": 12.4375,
      "learning_rate": 4.9785654732940964e-05,
      "loss": 1.3861,
      "num_input_tokens_seen": 10716896,
      "step": 18470
    },
    {
      "epoch": 2.751712838844206,
      "grad_norm": 10.5,
      "learning_rate": 4.9785229928362854e-05,
      "loss": 1.4106,
      "num_input_tokens_seen": 10719744,
      "step": 18475
    },
    {
      "epoch": 2.7524575513851652,
      "grad_norm": 7.0,
      "learning_rate": 4.9784804705063465e-05,
      "loss": 0.8659,
      "num_input_tokens_seen": 10722752,
      "step": 18480
    },
    {
      "epoch": 2.7532022639261244,
      "grad_norm": 14.625,
      "learning_rate": 4.978437906304999e-05,
      "loss": 0.7031,
      "num_input_tokens_seen": 10725696,
      "step": 18485
    },
    {
      "epoch": 2.7539469764670836,
      "grad_norm": 6.46875,
      "learning_rate": 4.978395300232963e-05,
      "loss": 0.9893,
      "num_input_tokens_seen": 10728384,
      "step": 18490
    },
    {
      "epoch": 2.754691689008043,
      "grad_norm": 13.375,
      "learning_rate": 4.978352652290956e-05,
      "loss": 0.8778,
      "num_input_tokens_seen": 10731520,
      "step": 18495
    },
    {
      "epoch": 2.755436401549002,
      "grad_norm": 13.875,
      "learning_rate": 4.978309962479701e-05,
      "loss": 1.1286,
      "num_input_tokens_seen": 10734688,
      "step": 18500
    },
    {
      "epoch": 2.7561811140899612,
      "grad_norm": 10.0625,
      "learning_rate": 4.978267230799918e-05,
      "loss": 1.3226,
      "num_input_tokens_seen": 10737728,
      "step": 18505
    },
    {
      "epoch": 2.7569258266309205,
      "grad_norm": 14.9375,
      "learning_rate": 4.9782244572523284e-05,
      "loss": 0.9476,
      "num_input_tokens_seen": 10740416,
      "step": 18510
    },
    {
      "epoch": 2.7576705391718797,
      "grad_norm": 12.4375,
      "learning_rate": 4.978181641837656e-05,
      "loss": 1.2145,
      "num_input_tokens_seen": 10743168,
      "step": 18515
    },
    {
      "epoch": 2.758415251712839,
      "grad_norm": 8.5,
      "learning_rate": 4.978138784556623e-05,
      "loss": 1.2305,
      "num_input_tokens_seen": 10746240,
      "step": 18520
    },
    {
      "epoch": 2.759159964253798,
      "grad_norm": 9.8125,
      "learning_rate": 4.9780958854099535e-05,
      "loss": 1.1045,
      "num_input_tokens_seen": 10749152,
      "step": 18525
    },
    {
      "epoch": 2.7599046767947573,
      "grad_norm": 8.4375,
      "learning_rate": 4.978052944398373e-05,
      "loss": 1.1855,
      "num_input_tokens_seen": 10752288,
      "step": 18530
    },
    {
      "epoch": 2.7606493893357165,
      "grad_norm": 20.0,
      "learning_rate": 4.978009961522607e-05,
      "loss": 1.6989,
      "num_input_tokens_seen": 10755168,
      "step": 18535
    },
    {
      "epoch": 2.7613941018766757,
      "grad_norm": 8.25,
      "learning_rate": 4.9779669367833804e-05,
      "loss": 0.6897,
      "num_input_tokens_seen": 10758080,
      "step": 18540
    },
    {
      "epoch": 2.762138814417635,
      "grad_norm": 9.5,
      "learning_rate": 4.9779238701814214e-05,
      "loss": 0.8756,
      "num_input_tokens_seen": 10761120,
      "step": 18545
    },
    {
      "epoch": 2.762883526958594,
      "grad_norm": 9.9375,
      "learning_rate": 4.977880761717457e-05,
      "loss": 0.8536,
      "num_input_tokens_seen": 10764064,
      "step": 18550
    },
    {
      "epoch": 2.7636282394995533,
      "grad_norm": 14.5625,
      "learning_rate": 4.977837611392216e-05,
      "loss": 0.9859,
      "num_input_tokens_seen": 10766784,
      "step": 18555
    },
    {
      "epoch": 2.7643729520405125,
      "grad_norm": 18.75,
      "learning_rate": 4.9777944192064264e-05,
      "loss": 1.2707,
      "num_input_tokens_seen": 10769664,
      "step": 18560
    },
    {
      "epoch": 2.7651176645814717,
      "grad_norm": 5.59375,
      "learning_rate": 4.9777511851608185e-05,
      "loss": 1.2971,
      "num_input_tokens_seen": 10772544,
      "step": 18565
    },
    {
      "epoch": 2.765862377122431,
      "grad_norm": 7.375,
      "learning_rate": 4.9777079092561224e-05,
      "loss": 1.2674,
      "num_input_tokens_seen": 10775648,
      "step": 18570
    },
    {
      "epoch": 2.76660708966339,
      "grad_norm": 7.90625,
      "learning_rate": 4.97766459149307e-05,
      "loss": 1.0498,
      "num_input_tokens_seen": 10778368,
      "step": 18575
    },
    {
      "epoch": 2.7673518022043493,
      "grad_norm": 11.8125,
      "learning_rate": 4.977621231872392e-05,
      "loss": 1.1958,
      "num_input_tokens_seen": 10781216,
      "step": 18580
    },
    {
      "epoch": 2.7680965147453085,
      "grad_norm": 7.4375,
      "learning_rate": 4.977577830394822e-05,
      "loss": 1.0164,
      "num_input_tokens_seen": 10783904,
      "step": 18585
    },
    {
      "epoch": 2.7688412272862672,
      "grad_norm": 16.125,
      "learning_rate": 4.977534387061091e-05,
      "loss": 1.3929,
      "num_input_tokens_seen": 10786752,
      "step": 18590
    },
    {
      "epoch": 2.769585939827227,
      "grad_norm": 8.8125,
      "learning_rate": 4.977490901871936e-05,
      "loss": 1.3109,
      "num_input_tokens_seen": 10789760,
      "step": 18595
    },
    {
      "epoch": 2.7703306523681857,
      "grad_norm": 9.8125,
      "learning_rate": 4.97744737482809e-05,
      "loss": 0.8078,
      "num_input_tokens_seen": 10792640,
      "step": 18600
    },
    {
      "epoch": 2.7710753649091453,
      "grad_norm": 15.8125,
      "learning_rate": 4.977403805930288e-05,
      "loss": 1.3063,
      "num_input_tokens_seen": 10795392,
      "step": 18605
    },
    {
      "epoch": 2.771820077450104,
      "grad_norm": 8.4375,
      "learning_rate": 4.977360195179268e-05,
      "loss": 1.0781,
      "num_input_tokens_seen": 10798368,
      "step": 18610
    },
    {
      "epoch": 2.7725647899910637,
      "grad_norm": 11.375,
      "learning_rate": 4.9773165425757646e-05,
      "loss": 1.0465,
      "num_input_tokens_seen": 10801248,
      "step": 18615
    },
    {
      "epoch": 2.7733095025320225,
      "grad_norm": 14.75,
      "learning_rate": 4.977272848120516e-05,
      "loss": 0.9547,
      "num_input_tokens_seen": 10803936,
      "step": 18620
    },
    {
      "epoch": 2.7740542150729817,
      "grad_norm": 9.6875,
      "learning_rate": 4.9772291118142604e-05,
      "loss": 1.069,
      "num_input_tokens_seen": 10807008,
      "step": 18625
    },
    {
      "epoch": 2.774798927613941,
      "grad_norm": 13.6875,
      "learning_rate": 4.9771853336577366e-05,
      "loss": 1.191,
      "num_input_tokens_seen": 10809952,
      "step": 18630
    },
    {
      "epoch": 2.7755436401549,
      "grad_norm": 14.125,
      "learning_rate": 4.9771415136516846e-05,
      "loss": 1.3259,
      "num_input_tokens_seen": 10813120,
      "step": 18635
    },
    {
      "epoch": 2.7762883526958593,
      "grad_norm": 15.6875,
      "learning_rate": 4.977097651796844e-05,
      "loss": 1.2256,
      "num_input_tokens_seen": 10816000,
      "step": 18640
    },
    {
      "epoch": 2.7770330652368185,
      "grad_norm": 8.0625,
      "learning_rate": 4.977053748093956e-05,
      "loss": 1.0674,
      "num_input_tokens_seen": 10818592,
      "step": 18645
    },
    {
      "epoch": 2.7777777777777777,
      "grad_norm": 15.25,
      "learning_rate": 4.9770098025437634e-05,
      "loss": 1.2088,
      "num_input_tokens_seen": 10821312,
      "step": 18650
    },
    {
      "epoch": 2.778522490318737,
      "grad_norm": 15.0625,
      "learning_rate": 4.9769658151470075e-05,
      "loss": 1.1597,
      "num_input_tokens_seen": 10824128,
      "step": 18655
    },
    {
      "epoch": 2.779267202859696,
      "grad_norm": 13.375,
      "learning_rate": 4.976921785904431e-05,
      "loss": 1.0039,
      "num_input_tokens_seen": 10827104,
      "step": 18660
    },
    {
      "epoch": 2.7800119154006553,
      "grad_norm": 12.25,
      "learning_rate": 4.976877714816779e-05,
      "loss": 1.0971,
      "num_input_tokens_seen": 10829888,
      "step": 18665
    },
    {
      "epoch": 2.7807566279416145,
      "grad_norm": 12.1875,
      "learning_rate": 4.976833601884795e-05,
      "loss": 0.8997,
      "num_input_tokens_seen": 10832896,
      "step": 18670
    },
    {
      "epoch": 2.7815013404825737,
      "grad_norm": 13.0,
      "learning_rate": 4.9767894471092246e-05,
      "loss": 1.0344,
      "num_input_tokens_seen": 10835776,
      "step": 18675
    },
    {
      "epoch": 2.782246053023533,
      "grad_norm": 21.375,
      "learning_rate": 4.9767452504908143e-05,
      "loss": 0.9433,
      "num_input_tokens_seen": 10838784,
      "step": 18680
    },
    {
      "epoch": 2.782990765564492,
      "grad_norm": 8.0625,
      "learning_rate": 4.9767010120303094e-05,
      "loss": 0.7431,
      "num_input_tokens_seen": 10841440,
      "step": 18685
    },
    {
      "epoch": 2.7837354781054513,
      "grad_norm": 16.5,
      "learning_rate": 4.9766567317284585e-05,
      "loss": 1.1939,
      "num_input_tokens_seen": 10844480,
      "step": 18690
    },
    {
      "epoch": 2.7844801906464105,
      "grad_norm": 13.8125,
      "learning_rate": 4.976612409586009e-05,
      "loss": 1.1957,
      "num_input_tokens_seen": 10847488,
      "step": 18695
    },
    {
      "epoch": 2.7852249031873697,
      "grad_norm": 15.875,
      "learning_rate": 4.9765680456037106e-05,
      "loss": 0.9634,
      "num_input_tokens_seen": 10850304,
      "step": 18700
    },
    {
      "epoch": 2.785969615728329,
      "grad_norm": 21.875,
      "learning_rate": 4.976523639782312e-05,
      "loss": 1.4427,
      "num_input_tokens_seen": 10852896,
      "step": 18705
    },
    {
      "epoch": 2.786714328269288,
      "grad_norm": 11.25,
      "learning_rate": 4.976479192122563e-05,
      "loss": 0.9172,
      "num_input_tokens_seen": 10855584,
      "step": 18710
    },
    {
      "epoch": 2.7874590408102473,
      "grad_norm": 7.0625,
      "learning_rate": 4.9764347026252156e-05,
      "loss": 0.7861,
      "num_input_tokens_seen": 10858688,
      "step": 18715
    },
    {
      "epoch": 2.7882037533512065,
      "grad_norm": 12.625,
      "learning_rate": 4.97639017129102e-05,
      "loss": 1.1139,
      "num_input_tokens_seen": 10861728,
      "step": 18720
    },
    {
      "epoch": 2.7889484658921657,
      "grad_norm": 13.3125,
      "learning_rate": 4.9763455981207305e-05,
      "loss": 0.9323,
      "num_input_tokens_seen": 10864576,
      "step": 18725
    },
    {
      "epoch": 2.789693178433125,
      "grad_norm": 15.9375,
      "learning_rate": 4.976300983115099e-05,
      "loss": 0.7488,
      "num_input_tokens_seen": 10867296,
      "step": 18730
    },
    {
      "epoch": 2.790437890974084,
      "grad_norm": 7.28125,
      "learning_rate": 4.976256326274878e-05,
      "loss": 0.8286,
      "num_input_tokens_seen": 10870336,
      "step": 18735
    },
    {
      "epoch": 2.7911826035150433,
      "grad_norm": 7.46875,
      "learning_rate": 4.976211627600823e-05,
      "loss": 0.968,
      "num_input_tokens_seen": 10873184,
      "step": 18740
    },
    {
      "epoch": 2.7919273160560025,
      "grad_norm": 16.75,
      "learning_rate": 4.976166887093691e-05,
      "loss": 1.3027,
      "num_input_tokens_seen": 10876128,
      "step": 18745
    },
    {
      "epoch": 2.7926720285969617,
      "grad_norm": 9.375,
      "learning_rate": 4.976122104754235e-05,
      "loss": 1.2342,
      "num_input_tokens_seen": 10879104,
      "step": 18750
    },
    {
      "epoch": 2.7934167411379205,
      "grad_norm": 8.375,
      "learning_rate": 4.976077280583212e-05,
      "loss": 0.9768,
      "num_input_tokens_seen": 10881920,
      "step": 18755
    },
    {
      "epoch": 2.79416145367888,
      "grad_norm": 16.625,
      "learning_rate": 4.9760324145813806e-05,
      "loss": 1.4537,
      "num_input_tokens_seen": 10884960,
      "step": 18760
    },
    {
      "epoch": 2.794906166219839,
      "grad_norm": 4.90625,
      "learning_rate": 4.975987506749499e-05,
      "loss": 0.911,
      "num_input_tokens_seen": 10887936,
      "step": 18765
    },
    {
      "epoch": 2.7956508787607985,
      "grad_norm": 11.5,
      "learning_rate": 4.975942557088324e-05,
      "loss": 1.3192,
      "num_input_tokens_seen": 10890976,
      "step": 18770
    },
    {
      "epoch": 2.7963955913017573,
      "grad_norm": 11.75,
      "learning_rate": 4.9758975655986164e-05,
      "loss": 1.2683,
      "num_input_tokens_seen": 10893920,
      "step": 18775
    },
    {
      "epoch": 2.797140303842717,
      "grad_norm": 6.21875,
      "learning_rate": 4.975852532281135e-05,
      "loss": 0.8045,
      "num_input_tokens_seen": 10896672,
      "step": 18780
    },
    {
      "epoch": 2.7978850163836757,
      "grad_norm": 27.125,
      "learning_rate": 4.975807457136642e-05,
      "loss": 1.1551,
      "num_input_tokens_seen": 10900160,
      "step": 18785
    },
    {
      "epoch": 2.7986297289246354,
      "grad_norm": 5.15625,
      "learning_rate": 4.975762340165898e-05,
      "loss": 1.0587,
      "num_input_tokens_seen": 10903008,
      "step": 18790
    },
    {
      "epoch": 2.799374441465594,
      "grad_norm": 10.125,
      "learning_rate": 4.975717181369666e-05,
      "loss": 1.2406,
      "num_input_tokens_seen": 10905760,
      "step": 18795
    },
    {
      "epoch": 2.8001191540065533,
      "grad_norm": 10.6875,
      "learning_rate": 4.9756719807487076e-05,
      "loss": 1.2365,
      "num_input_tokens_seen": 10908608,
      "step": 18800
    },
    {
      "epoch": 2.8008638665475125,
      "grad_norm": 4.8125,
      "learning_rate": 4.975626738303788e-05,
      "loss": 0.8481,
      "num_input_tokens_seen": 10911584,
      "step": 18805
    },
    {
      "epoch": 2.8016085790884717,
      "grad_norm": 12.3125,
      "learning_rate": 4.975581454035671e-05,
      "loss": 1.0738,
      "num_input_tokens_seen": 10914432,
      "step": 18810
    },
    {
      "epoch": 2.802353291629431,
      "grad_norm": 5.46875,
      "learning_rate": 4.975536127945121e-05,
      "loss": 0.9585,
      "num_input_tokens_seen": 10917344,
      "step": 18815
    },
    {
      "epoch": 2.80309800417039,
      "grad_norm": 5.96875,
      "learning_rate": 4.975490760032904e-05,
      "loss": 1.2423,
      "num_input_tokens_seen": 10920256,
      "step": 18820
    },
    {
      "epoch": 2.8038427167113493,
      "grad_norm": 12.3125,
      "learning_rate": 4.975445350299787e-05,
      "loss": 0.869,
      "num_input_tokens_seen": 10923040,
      "step": 18825
    },
    {
      "epoch": 2.8045874292523085,
      "grad_norm": 6.71875,
      "learning_rate": 4.975399898746536e-05,
      "loss": 1.0725,
      "num_input_tokens_seen": 10926144,
      "step": 18830
    },
    {
      "epoch": 2.8053321417932677,
      "grad_norm": 15.0625,
      "learning_rate": 4.9753544053739197e-05,
      "loss": 0.9579,
      "num_input_tokens_seen": 10928992,
      "step": 18835
    },
    {
      "epoch": 2.806076854334227,
      "grad_norm": 9.8125,
      "learning_rate": 4.975308870182707e-05,
      "loss": 0.8158,
      "num_input_tokens_seen": 10931872,
      "step": 18840
    },
    {
      "epoch": 2.806821566875186,
      "grad_norm": 7.5625,
      "learning_rate": 4.9752632931736665e-05,
      "loss": 0.8767,
      "num_input_tokens_seen": 10934848,
      "step": 18845
    },
    {
      "epoch": 2.8075662794161453,
      "grad_norm": 8.375,
      "learning_rate": 4.9752176743475684e-05,
      "loss": 0.9372,
      "num_input_tokens_seen": 10937792,
      "step": 18850
    },
    {
      "epoch": 2.8083109919571045,
      "grad_norm": 6.53125,
      "learning_rate": 4.9751720137051836e-05,
      "loss": 0.947,
      "num_input_tokens_seen": 10940960,
      "step": 18855
    },
    {
      "epoch": 2.8090557044980637,
      "grad_norm": 10.3125,
      "learning_rate": 4.9751263112472834e-05,
      "loss": 1.0674,
      "num_input_tokens_seen": 10943616,
      "step": 18860
    },
    {
      "epoch": 2.809800417039023,
      "grad_norm": 7.5625,
      "learning_rate": 4.9750805669746395e-05,
      "loss": 1.2001,
      "num_input_tokens_seen": 10946720,
      "step": 18865
    },
    {
      "epoch": 2.810545129579982,
      "grad_norm": 4.8125,
      "learning_rate": 4.975034780888025e-05,
      "loss": 1.059,
      "num_input_tokens_seen": 10949376,
      "step": 18870
    },
    {
      "epoch": 2.8112898421209414,
      "grad_norm": 14.0625,
      "learning_rate": 4.9749889529882134e-05,
      "loss": 1.1814,
      "num_input_tokens_seen": 10952128,
      "step": 18875
    },
    {
      "epoch": 2.8120345546619006,
      "grad_norm": 5.5625,
      "learning_rate": 4.974943083275979e-05,
      "loss": 0.9823,
      "num_input_tokens_seen": 10954880,
      "step": 18880
    },
    {
      "epoch": 2.8127792672028598,
      "grad_norm": 10.8125,
      "learning_rate": 4.974897171752097e-05,
      "loss": 1.0401,
      "num_input_tokens_seen": 10957696,
      "step": 18885
    },
    {
      "epoch": 2.813523979743819,
      "grad_norm": 11.1875,
      "learning_rate": 4.9748512184173416e-05,
      "loss": 1.1934,
      "num_input_tokens_seen": 10960736,
      "step": 18890
    },
    {
      "epoch": 2.814268692284778,
      "grad_norm": 6.21875,
      "learning_rate": 4.9748052232724905e-05,
      "loss": 0.9638,
      "num_input_tokens_seen": 10963424,
      "step": 18895
    },
    {
      "epoch": 2.8150134048257374,
      "grad_norm": 5.625,
      "learning_rate": 4.974759186318321e-05,
      "loss": 0.9488,
      "num_input_tokens_seen": 10966272,
      "step": 18900
    },
    {
      "epoch": 2.8157581173666966,
      "grad_norm": 8.5625,
      "learning_rate": 4.97471310755561e-05,
      "loss": 1.0505,
      "num_input_tokens_seen": 10969504,
      "step": 18905
    },
    {
      "epoch": 2.8165028299076558,
      "grad_norm": 7.46875,
      "learning_rate": 4.974666986985136e-05,
      "loss": 0.7455,
      "num_input_tokens_seen": 10972704,
      "step": 18910
    },
    {
      "epoch": 2.817247542448615,
      "grad_norm": 12.8125,
      "learning_rate": 4.974620824607679e-05,
      "loss": 1.624,
      "num_input_tokens_seen": 10975776,
      "step": 18915
    },
    {
      "epoch": 2.817992254989574,
      "grad_norm": 6.6875,
      "learning_rate": 4.9745746204240175e-05,
      "loss": 0.9172,
      "num_input_tokens_seen": 10978432,
      "step": 18920
    },
    {
      "epoch": 2.8187369675305334,
      "grad_norm": 10.0,
      "learning_rate": 4.974528374434934e-05,
      "loss": 1.3829,
      "num_input_tokens_seen": 10981568,
      "step": 18925
    },
    {
      "epoch": 2.819481680071492,
      "grad_norm": 12.8125,
      "learning_rate": 4.974482086641207e-05,
      "loss": 1.0746,
      "num_input_tokens_seen": 10984352,
      "step": 18930
    },
    {
      "epoch": 2.820226392612452,
      "grad_norm": 8.9375,
      "learning_rate": 4.974435757043621e-05,
      "loss": 0.9932,
      "num_input_tokens_seen": 10987168,
      "step": 18935
    },
    {
      "epoch": 2.8209711051534105,
      "grad_norm": 7.375,
      "learning_rate": 4.974389385642958e-05,
      "loss": 0.7832,
      "num_input_tokens_seen": 10990144,
      "step": 18940
    },
    {
      "epoch": 2.82171581769437,
      "grad_norm": 12.25,
      "learning_rate": 4.9743429724400007e-05,
      "loss": 1.168,
      "num_input_tokens_seen": 10993056,
      "step": 18945
    },
    {
      "epoch": 2.822460530235329,
      "grad_norm": 6.6875,
      "learning_rate": 4.974296517435534e-05,
      "loss": 0.9364,
      "num_input_tokens_seen": 10996064,
      "step": 18950
    },
    {
      "epoch": 2.8232052427762886,
      "grad_norm": 12.0,
      "learning_rate": 4.974250020630342e-05,
      "loss": 1.5215,
      "num_input_tokens_seen": 10999008,
      "step": 18955
    },
    {
      "epoch": 2.8239499553172474,
      "grad_norm": 12.0625,
      "learning_rate": 4.9742034820252116e-05,
      "loss": 1.0449,
      "num_input_tokens_seen": 11001824,
      "step": 18960
    },
    {
      "epoch": 2.824694667858207,
      "grad_norm": 13.25,
      "learning_rate": 4.974156901620927e-05,
      "loss": 0.8294,
      "num_input_tokens_seen": 11004896,
      "step": 18965
    },
    {
      "epoch": 2.8254393803991658,
      "grad_norm": 23.5,
      "learning_rate": 4.974110279418277e-05,
      "loss": 1.3766,
      "num_input_tokens_seen": 11007712,
      "step": 18970
    },
    {
      "epoch": 2.826184092940125,
      "grad_norm": 8.0,
      "learning_rate": 4.9740636154180476e-05,
      "loss": 0.9932,
      "num_input_tokens_seen": 11010656,
      "step": 18975
    },
    {
      "epoch": 2.826928805481084,
      "grad_norm": 12.25,
      "learning_rate": 4.974016909621029e-05,
      "loss": 0.9993,
      "num_input_tokens_seen": 11013536,
      "step": 18980
    },
    {
      "epoch": 2.8276735180220434,
      "grad_norm": 11.1875,
      "learning_rate": 4.9739701620280076e-05,
      "loss": 0.8174,
      "num_input_tokens_seen": 11016224,
      "step": 18985
    },
    {
      "epoch": 2.8284182305630026,
      "grad_norm": 9.875,
      "learning_rate": 4.973923372639776e-05,
      "loss": 1.2627,
      "num_input_tokens_seen": 11019328,
      "step": 18990
    },
    {
      "epoch": 2.8291629431039618,
      "grad_norm": 5.78125,
      "learning_rate": 4.973876541457123e-05,
      "loss": 1.0074,
      "num_input_tokens_seen": 11022208,
      "step": 18995
    },
    {
      "epoch": 2.829907655644921,
      "grad_norm": 3.75,
      "learning_rate": 4.97382966848084e-05,
      "loss": 0.717,
      "num_input_tokens_seen": 11025568,
      "step": 19000
    },
    {
      "epoch": 2.83065236818588,
      "grad_norm": 15.1875,
      "learning_rate": 4.9737827537117196e-05,
      "loss": 1.107,
      "num_input_tokens_seen": 11028384,
      "step": 19005
    },
    {
      "epoch": 2.8313970807268394,
      "grad_norm": 16.25,
      "learning_rate": 4.973735797150553e-05,
      "loss": 1.251,
      "num_input_tokens_seen": 11031328,
      "step": 19010
    },
    {
      "epoch": 2.8321417932677986,
      "grad_norm": 10.0625,
      "learning_rate": 4.973688798798135e-05,
      "loss": 1.0833,
      "num_input_tokens_seen": 11034464,
      "step": 19015
    },
    {
      "epoch": 2.832886505808758,
      "grad_norm": 7.5,
      "learning_rate": 4.973641758655259e-05,
      "loss": 1.0331,
      "num_input_tokens_seen": 11037216,
      "step": 19020
    },
    {
      "epoch": 2.833631218349717,
      "grad_norm": 14.375,
      "learning_rate": 4.973594676722719e-05,
      "loss": 1.0522,
      "num_input_tokens_seen": 11040384,
      "step": 19025
    },
    {
      "epoch": 2.834375930890676,
      "grad_norm": 11.1875,
      "learning_rate": 4.973547553001311e-05,
      "loss": 1.1074,
      "num_input_tokens_seen": 11043424,
      "step": 19030
    },
    {
      "epoch": 2.8351206434316354,
      "grad_norm": 6.75,
      "learning_rate": 4.9735003874918314e-05,
      "loss": 1.1582,
      "num_input_tokens_seen": 11046336,
      "step": 19035
    },
    {
      "epoch": 2.8358653559725946,
      "grad_norm": 11.25,
      "learning_rate": 4.9734531801950765e-05,
      "loss": 1.1316,
      "num_input_tokens_seen": 11049216,
      "step": 19040
    },
    {
      "epoch": 2.836610068513554,
      "grad_norm": 15.25,
      "learning_rate": 4.9734059311118444e-05,
      "loss": 1.2904,
      "num_input_tokens_seen": 11052064,
      "step": 19045
    },
    {
      "epoch": 2.837354781054513,
      "grad_norm": 10.375,
      "learning_rate": 4.973358640242932e-05,
      "loss": 1.0414,
      "num_input_tokens_seen": 11054976,
      "step": 19050
    },
    {
      "epoch": 2.838099493595472,
      "grad_norm": 7.1875,
      "learning_rate": 4.97331130758914e-05,
      "loss": 0.8992,
      "num_input_tokens_seen": 11057984,
      "step": 19055
    },
    {
      "epoch": 2.8388442061364314,
      "grad_norm": 7.5,
      "learning_rate": 4.9732639331512675e-05,
      "loss": 0.9448,
      "num_input_tokens_seen": 11060768,
      "step": 19060
    },
    {
      "epoch": 2.8395889186773906,
      "grad_norm": 13.4375,
      "learning_rate": 4.973216516930114e-05,
      "loss": 1.3529,
      "num_input_tokens_seen": 11063744,
      "step": 19065
    },
    {
      "epoch": 2.84033363121835,
      "grad_norm": 13.3125,
      "learning_rate": 4.973169058926481e-05,
      "loss": 0.9109,
      "num_input_tokens_seen": 11066592,
      "step": 19070
    },
    {
      "epoch": 2.841078343759309,
      "grad_norm": 6.15625,
      "learning_rate": 4.973121559141171e-05,
      "loss": 1.0008,
      "num_input_tokens_seen": 11069280,
      "step": 19075
    },
    {
      "epoch": 2.841823056300268,
      "grad_norm": 16.375,
      "learning_rate": 4.9730740175749854e-05,
      "loss": 0.8299,
      "num_input_tokens_seen": 11072384,
      "step": 19080
    },
    {
      "epoch": 2.8425677688412274,
      "grad_norm": 15.5,
      "learning_rate": 4.973026434228728e-05,
      "loss": 1.2069,
      "num_input_tokens_seen": 11075072,
      "step": 19085
    },
    {
      "epoch": 2.8433124813821866,
      "grad_norm": 15.3125,
      "learning_rate": 4.972978809103202e-05,
      "loss": 1.1703,
      "num_input_tokens_seen": 11077856,
      "step": 19090
    },
    {
      "epoch": 2.844057193923146,
      "grad_norm": 13.9375,
      "learning_rate": 4.972931142199213e-05,
      "loss": 1.1893,
      "num_input_tokens_seen": 11080672,
      "step": 19095
    },
    {
      "epoch": 2.844801906464105,
      "grad_norm": 14.75,
      "learning_rate": 4.972883433517566e-05,
      "loss": 1.1536,
      "num_input_tokens_seen": 11083360,
      "step": 19100
    },
    {
      "epoch": 2.845546619005064,
      "grad_norm": 11.125,
      "learning_rate": 4.972835683059065e-05,
      "loss": 1.1904,
      "num_input_tokens_seen": 11086464,
      "step": 19105
    },
    {
      "epoch": 2.8462913315460234,
      "grad_norm": 14.5625,
      "learning_rate": 4.97278789082452e-05,
      "loss": 1.0642,
      "num_input_tokens_seen": 11089504,
      "step": 19110
    },
    {
      "epoch": 2.847036044086982,
      "grad_norm": 18.125,
      "learning_rate": 4.9727400568147364e-05,
      "loss": 1.3196,
      "num_input_tokens_seen": 11092288,
      "step": 19115
    },
    {
      "epoch": 2.847780756627942,
      "grad_norm": 5.375,
      "learning_rate": 4.972692181030523e-05,
      "loss": 1.0454,
      "num_input_tokens_seen": 11095136,
      "step": 19120
    },
    {
      "epoch": 2.8485254691689006,
      "grad_norm": 7.34375,
      "learning_rate": 4.972644263472688e-05,
      "loss": 1.0355,
      "num_input_tokens_seen": 11098112,
      "step": 19125
    },
    {
      "epoch": 2.8492701817098602,
      "grad_norm": 11.875,
      "learning_rate": 4.972596304142041e-05,
      "loss": 0.8964,
      "num_input_tokens_seen": 11100992,
      "step": 19130
    },
    {
      "epoch": 2.850014894250819,
      "grad_norm": 7.8125,
      "learning_rate": 4.9725483030393924e-05,
      "loss": 0.6845,
      "num_input_tokens_seen": 11104000,
      "step": 19135
    },
    {
      "epoch": 2.8507596067917786,
      "grad_norm": 8.0625,
      "learning_rate": 4.972500260165555e-05,
      "loss": 1.0782,
      "num_input_tokens_seen": 11106976,
      "step": 19140
    },
    {
      "epoch": 2.8515043193327374,
      "grad_norm": 13.1875,
      "learning_rate": 4.972452175521337e-05,
      "loss": 1.1018,
      "num_input_tokens_seen": 11109728,
      "step": 19145
    },
    {
      "epoch": 2.8522490318736966,
      "grad_norm": 8.875,
      "learning_rate": 4.972404049107552e-05,
      "loss": 0.9478,
      "num_input_tokens_seen": 11112576,
      "step": 19150
    },
    {
      "epoch": 2.852993744414656,
      "grad_norm": 10.875,
      "learning_rate": 4.972355880925014e-05,
      "loss": 1.0693,
      "num_input_tokens_seen": 11115328,
      "step": 19155
    },
    {
      "epoch": 2.853738456955615,
      "grad_norm": 11.8125,
      "learning_rate": 4.9723076709745365e-05,
      "loss": 1.3238,
      "num_input_tokens_seen": 11118528,
      "step": 19160
    },
    {
      "epoch": 2.854483169496574,
      "grad_norm": 10.1875,
      "learning_rate": 4.972259419256933e-05,
      "loss": 1.0409,
      "num_input_tokens_seen": 11121216,
      "step": 19165
    },
    {
      "epoch": 2.8552278820375334,
      "grad_norm": 11.8125,
      "learning_rate": 4.97221112577302e-05,
      "loss": 1.1433,
      "num_input_tokens_seen": 11123904,
      "step": 19170
    },
    {
      "epoch": 2.8559725945784926,
      "grad_norm": 11.0625,
      "learning_rate": 4.972162790523612e-05,
      "loss": 1.0267,
      "num_input_tokens_seen": 11126752,
      "step": 19175
    },
    {
      "epoch": 2.856717307119452,
      "grad_norm": 8.1875,
      "learning_rate": 4.9721144135095265e-05,
      "loss": 0.9349,
      "num_input_tokens_seen": 11129888,
      "step": 19180
    },
    {
      "epoch": 2.857462019660411,
      "grad_norm": 11.4375,
      "learning_rate": 4.9720659947315815e-05,
      "loss": 1.2895,
      "num_input_tokens_seen": 11132608,
      "step": 19185
    },
    {
      "epoch": 2.8582067322013702,
      "grad_norm": 14.5625,
      "learning_rate": 4.972017534190593e-05,
      "loss": 1.0376,
      "num_input_tokens_seen": 11135392,
      "step": 19190
    },
    {
      "epoch": 2.8589514447423294,
      "grad_norm": 12.5,
      "learning_rate": 4.971969031887381e-05,
      "loss": 0.9009,
      "num_input_tokens_seen": 11138720,
      "step": 19195
    },
    {
      "epoch": 2.8596961572832886,
      "grad_norm": 15.3125,
      "learning_rate": 4.971920487822764e-05,
      "loss": 1.1893,
      "num_input_tokens_seen": 11141504,
      "step": 19200
    },
    {
      "epoch": 2.860440869824248,
      "grad_norm": 8.0,
      "learning_rate": 4.971871901997563e-05,
      "loss": 1.2711,
      "num_input_tokens_seen": 11144352,
      "step": 19205
    },
    {
      "epoch": 2.861185582365207,
      "grad_norm": 14.875,
      "learning_rate": 4.9718232744125995e-05,
      "loss": 1.2469,
      "num_input_tokens_seen": 11147424,
      "step": 19210
    },
    {
      "epoch": 2.8619302949061662,
      "grad_norm": 10.625,
      "learning_rate": 4.9717746050686925e-05,
      "loss": 1.2022,
      "num_input_tokens_seen": 11150400,
      "step": 19215
    },
    {
      "epoch": 2.8626750074471254,
      "grad_norm": 3.421875,
      "learning_rate": 4.9717258939666663e-05,
      "loss": 0.8701,
      "num_input_tokens_seen": 11153344,
      "step": 19220
    },
    {
      "epoch": 2.8634197199880846,
      "grad_norm": 7.625,
      "learning_rate": 4.9716771411073436e-05,
      "loss": 1.0303,
      "num_input_tokens_seen": 11156032,
      "step": 19225
    },
    {
      "epoch": 2.864164432529044,
      "grad_norm": 20.5,
      "learning_rate": 4.9716283464915484e-05,
      "loss": 1.001,
      "num_input_tokens_seen": 11158720,
      "step": 19230
    },
    {
      "epoch": 2.864909145070003,
      "grad_norm": 13.4375,
      "learning_rate": 4.9715795101201025e-05,
      "loss": 1.4105,
      "num_input_tokens_seen": 11161440,
      "step": 19235
    },
    {
      "epoch": 2.8656538576109623,
      "grad_norm": 7.875,
      "learning_rate": 4.9715306319938335e-05,
      "loss": 1.0183,
      "num_input_tokens_seen": 11164256,
      "step": 19240
    },
    {
      "epoch": 2.8663985701519215,
      "grad_norm": 8.5625,
      "learning_rate": 4.971481712113567e-05,
      "loss": 1.0139,
      "num_input_tokens_seen": 11167200,
      "step": 19245
    },
    {
      "epoch": 2.8671432826928807,
      "grad_norm": 10.3125,
      "learning_rate": 4.9714327504801286e-05,
      "loss": 1.106,
      "num_input_tokens_seen": 11170208,
      "step": 19250
    },
    {
      "epoch": 2.86788799523384,
      "grad_norm": 14.4375,
      "learning_rate": 4.971383747094346e-05,
      "loss": 1.5059,
      "num_input_tokens_seen": 11173056,
      "step": 19255
    },
    {
      "epoch": 2.868632707774799,
      "grad_norm": 7.5625,
      "learning_rate": 4.9713347019570465e-05,
      "loss": 0.8521,
      "num_input_tokens_seen": 11175968,
      "step": 19260
    },
    {
      "epoch": 2.8693774203157583,
      "grad_norm": 8.1875,
      "learning_rate": 4.971285615069059e-05,
      "loss": 0.9514,
      "num_input_tokens_seen": 11178880,
      "step": 19265
    },
    {
      "epoch": 2.8701221328567175,
      "grad_norm": 12.8125,
      "learning_rate": 4.9712364864312125e-05,
      "loss": 1.0714,
      "num_input_tokens_seen": 11181696,
      "step": 19270
    },
    {
      "epoch": 2.8708668453976767,
      "grad_norm": 12.0625,
      "learning_rate": 4.9711873160443375e-05,
      "loss": 1.5187,
      "num_input_tokens_seen": 11184448,
      "step": 19275
    },
    {
      "epoch": 2.8716115579386354,
      "grad_norm": 19.0,
      "learning_rate": 4.971138103909264e-05,
      "loss": 1.3512,
      "num_input_tokens_seen": 11187776,
      "step": 19280
    },
    {
      "epoch": 2.872356270479595,
      "grad_norm": 6.8125,
      "learning_rate": 4.9710888500268236e-05,
      "loss": 0.9291,
      "num_input_tokens_seen": 11190752,
      "step": 19285
    },
    {
      "epoch": 2.873100983020554,
      "grad_norm": 7.03125,
      "learning_rate": 4.9710395543978495e-05,
      "loss": 0.9955,
      "num_input_tokens_seen": 11193536,
      "step": 19290
    },
    {
      "epoch": 2.8738456955615135,
      "grad_norm": 6.78125,
      "learning_rate": 4.970990217023173e-05,
      "loss": 0.9174,
      "num_input_tokens_seen": 11196352,
      "step": 19295
    },
    {
      "epoch": 2.8745904081024722,
      "grad_norm": 9.6875,
      "learning_rate": 4.9709408379036284e-05,
      "loss": 1.1384,
      "num_input_tokens_seen": 11199328,
      "step": 19300
    },
    {
      "epoch": 2.875335120643432,
      "grad_norm": 5.4375,
      "learning_rate": 4.97089141704005e-05,
      "loss": 0.6378,
      "num_input_tokens_seen": 11202176,
      "step": 19305
    },
    {
      "epoch": 2.8760798331843906,
      "grad_norm": 11.1875,
      "learning_rate": 4.970841954433272e-05,
      "loss": 1.2926,
      "num_input_tokens_seen": 11204992,
      "step": 19310
    },
    {
      "epoch": 2.8768245457253503,
      "grad_norm": 15.3125,
      "learning_rate": 4.97079245008413e-05,
      "loss": 0.7988,
      "num_input_tokens_seen": 11207616,
      "step": 19315
    },
    {
      "epoch": 2.877569258266309,
      "grad_norm": 9.0625,
      "learning_rate": 4.970742903993462e-05,
      "loss": 1.2456,
      "num_input_tokens_seen": 11210720,
      "step": 19320
    },
    {
      "epoch": 2.8783139708072683,
      "grad_norm": 8.6875,
      "learning_rate": 4.970693316162103e-05,
      "loss": 1.2059,
      "num_input_tokens_seen": 11213600,
      "step": 19325
    },
    {
      "epoch": 2.8790586833482275,
      "grad_norm": 10.1875,
      "learning_rate": 4.9706436865908915e-05,
      "loss": 1.1707,
      "num_input_tokens_seen": 11216320,
      "step": 19330
    },
    {
      "epoch": 2.8798033958891867,
      "grad_norm": 17.125,
      "learning_rate": 4.970594015280665e-05,
      "loss": 1.1038,
      "num_input_tokens_seen": 11219424,
      "step": 19335
    },
    {
      "epoch": 2.880548108430146,
      "grad_norm": 8.875,
      "learning_rate": 4.970544302232265e-05,
      "loss": 1.0579,
      "num_input_tokens_seen": 11222336,
      "step": 19340
    },
    {
      "epoch": 2.881292820971105,
      "grad_norm": 11.0625,
      "learning_rate": 4.97049454744653e-05,
      "loss": 1.1125,
      "num_input_tokens_seen": 11225376,
      "step": 19345
    },
    {
      "epoch": 2.8820375335120643,
      "grad_norm": 9.75,
      "learning_rate": 4.9704447509243e-05,
      "loss": 1.4205,
      "num_input_tokens_seen": 11228448,
      "step": 19350
    },
    {
      "epoch": 2.8827822460530235,
      "grad_norm": 7.4375,
      "learning_rate": 4.970394912666416e-05,
      "loss": 0.9783,
      "num_input_tokens_seen": 11231072,
      "step": 19355
    },
    {
      "epoch": 2.8835269585939827,
      "grad_norm": 11.75,
      "learning_rate": 4.970345032673722e-05,
      "loss": 0.9695,
      "num_input_tokens_seen": 11234048,
      "step": 19360
    },
    {
      "epoch": 2.884271671134942,
      "grad_norm": 12.1875,
      "learning_rate": 4.97029511094706e-05,
      "loss": 1.2761,
      "num_input_tokens_seen": 11236928,
      "step": 19365
    },
    {
      "epoch": 2.885016383675901,
      "grad_norm": 5.4375,
      "learning_rate": 4.970245147487271e-05,
      "loss": 0.9991,
      "num_input_tokens_seen": 11240064,
      "step": 19370
    },
    {
      "epoch": 2.8857610962168603,
      "grad_norm": 7.375,
      "learning_rate": 4.970195142295202e-05,
      "loss": 1.4764,
      "num_input_tokens_seen": 11242656,
      "step": 19375
    },
    {
      "epoch": 2.8865058087578195,
      "grad_norm": 14.1875,
      "learning_rate": 4.9701450953716965e-05,
      "loss": 1.5598,
      "num_input_tokens_seen": 11245792,
      "step": 19380
    },
    {
      "epoch": 2.8872505212987787,
      "grad_norm": 9.5,
      "learning_rate": 4.9700950067176e-05,
      "loss": 1.1305,
      "num_input_tokens_seen": 11248800,
      "step": 19385
    },
    {
      "epoch": 2.887995233839738,
      "grad_norm": 12.3125,
      "learning_rate": 4.970044876333759e-05,
      "loss": 0.9832,
      "num_input_tokens_seen": 11251904,
      "step": 19390
    },
    {
      "epoch": 2.888739946380697,
      "grad_norm": 13.75,
      "learning_rate": 4.9699947042210196e-05,
      "loss": 1.3011,
      "num_input_tokens_seen": 11254944,
      "step": 19395
    },
    {
      "epoch": 2.8894846589216563,
      "grad_norm": 10.875,
      "learning_rate": 4.96994449038023e-05,
      "loss": 1.1763,
      "num_input_tokens_seen": 11257728,
      "step": 19400
    },
    {
      "epoch": 2.8902293714626155,
      "grad_norm": 15.6875,
      "learning_rate": 4.9698942348122404e-05,
      "loss": 1.2717,
      "num_input_tokens_seen": 11260832,
      "step": 19405
    },
    {
      "epoch": 2.8909740840035747,
      "grad_norm": 8.0625,
      "learning_rate": 4.9698439375178965e-05,
      "loss": 0.8775,
      "num_input_tokens_seen": 11263712,
      "step": 19410
    },
    {
      "epoch": 2.891718796544534,
      "grad_norm": 11.875,
      "learning_rate": 4.9697935984980496e-05,
      "loss": 1.2087,
      "num_input_tokens_seen": 11266848,
      "step": 19415
    },
    {
      "epoch": 2.892463509085493,
      "grad_norm": 10.4375,
      "learning_rate": 4.96974321775355e-05,
      "loss": 1.1041,
      "num_input_tokens_seen": 11269440,
      "step": 19420
    },
    {
      "epoch": 2.8932082216264523,
      "grad_norm": 9.875,
      "learning_rate": 4.969692795285249e-05,
      "loss": 1.1174,
      "num_input_tokens_seen": 11272288,
      "step": 19425
    },
    {
      "epoch": 2.8939529341674115,
      "grad_norm": 8.5,
      "learning_rate": 4.9696423310939985e-05,
      "loss": 1.5486,
      "num_input_tokens_seen": 11275008,
      "step": 19430
    },
    {
      "epoch": 2.8946976467083707,
      "grad_norm": 18.75,
      "learning_rate": 4.9695918251806506e-05,
      "loss": 1.2323,
      "num_input_tokens_seen": 11277856,
      "step": 19435
    },
    {
      "epoch": 2.89544235924933,
      "grad_norm": 11.0,
      "learning_rate": 4.969541277546059e-05,
      "loss": 1.0159,
      "num_input_tokens_seen": 11280480,
      "step": 19440
    },
    {
      "epoch": 2.896187071790289,
      "grad_norm": 6.78125,
      "learning_rate": 4.9694906881910776e-05,
      "loss": 0.8862,
      "num_input_tokens_seen": 11283360,
      "step": 19445
    },
    {
      "epoch": 2.8969317843312483,
      "grad_norm": 7.375,
      "learning_rate": 4.969440057116561e-05,
      "loss": 0.9614,
      "num_input_tokens_seen": 11286400,
      "step": 19450
    },
    {
      "epoch": 2.897676496872207,
      "grad_norm": 16.25,
      "learning_rate": 4.969389384323364e-05,
      "loss": 1.0288,
      "num_input_tokens_seen": 11289184,
      "step": 19455
    },
    {
      "epoch": 2.8984212094131667,
      "grad_norm": 16.375,
      "learning_rate": 4.969338669812343e-05,
      "loss": 1.0341,
      "num_input_tokens_seen": 11291712,
      "step": 19460
    },
    {
      "epoch": 2.8991659219541255,
      "grad_norm": 30.875,
      "learning_rate": 4.969287913584355e-05,
      "loss": 1.299,
      "num_input_tokens_seen": 11294688,
      "step": 19465
    },
    {
      "epoch": 2.899910634495085,
      "grad_norm": 10.9375,
      "learning_rate": 4.969237115640258e-05,
      "loss": 1.0262,
      "num_input_tokens_seen": 11297440,
      "step": 19470
    },
    {
      "epoch": 2.900655347036044,
      "grad_norm": 9.125,
      "learning_rate": 4.969186275980909e-05,
      "loss": 1.0743,
      "num_input_tokens_seen": 11300544,
      "step": 19475
    },
    {
      "epoch": 2.9014000595770035,
      "grad_norm": 11.0625,
      "learning_rate": 4.969135394607167e-05,
      "loss": 0.9244,
      "num_input_tokens_seen": 11303552,
      "step": 19480
    },
    {
      "epoch": 2.9021447721179623,
      "grad_norm": 15.5625,
      "learning_rate": 4.969084471519893e-05,
      "loss": 1.4413,
      "num_input_tokens_seen": 11306304,
      "step": 19485
    },
    {
      "epoch": 2.9028894846589215,
      "grad_norm": 5.125,
      "learning_rate": 4.9690335067199464e-05,
      "loss": 0.8644,
      "num_input_tokens_seen": 11309056,
      "step": 19490
    },
    {
      "epoch": 2.9036341971998807,
      "grad_norm": 17.625,
      "learning_rate": 4.9689825002081866e-05,
      "loss": 1.2702,
      "num_input_tokens_seen": 11312608,
      "step": 19495
    },
    {
      "epoch": 2.90437890974084,
      "grad_norm": 7.1875,
      "learning_rate": 4.9689314519854786e-05,
      "loss": 0.8674,
      "num_input_tokens_seen": 11315328,
      "step": 19500
    },
    {
      "epoch": 2.905123622281799,
      "grad_norm": 8.625,
      "learning_rate": 4.968880362052682e-05,
      "loss": 0.9724,
      "num_input_tokens_seen": 11318208,
      "step": 19505
    },
    {
      "epoch": 2.9058683348227583,
      "grad_norm": 10.875,
      "learning_rate": 4.968829230410661e-05,
      "loss": 1.2725,
      "num_input_tokens_seen": 11320928,
      "step": 19510
    },
    {
      "epoch": 2.9066130473637175,
      "grad_norm": 8.3125,
      "learning_rate": 4.96877805706028e-05,
      "loss": 0.9562,
      "num_input_tokens_seen": 11324160,
      "step": 19515
    },
    {
      "epoch": 2.9073577599046767,
      "grad_norm": 9.5625,
      "learning_rate": 4.968726842002402e-05,
      "loss": 1.2482,
      "num_input_tokens_seen": 11326912,
      "step": 19520
    },
    {
      "epoch": 2.908102472445636,
      "grad_norm": 11.8125,
      "learning_rate": 4.968675585237894e-05,
      "loss": 0.9362,
      "num_input_tokens_seen": 11329824,
      "step": 19525
    },
    {
      "epoch": 2.908847184986595,
      "grad_norm": 6.375,
      "learning_rate": 4.9686242867676204e-05,
      "loss": 0.9713,
      "num_input_tokens_seen": 11332512,
      "step": 19530
    },
    {
      "epoch": 2.9095918975275543,
      "grad_norm": 15.125,
      "learning_rate": 4.968572946592448e-05,
      "loss": 1.0552,
      "num_input_tokens_seen": 11335360,
      "step": 19535
    },
    {
      "epoch": 2.9103366100685135,
      "grad_norm": 14.1875,
      "learning_rate": 4.968521564713246e-05,
      "loss": 1.0681,
      "num_input_tokens_seen": 11338080,
      "step": 19540
    },
    {
      "epoch": 2.9110813226094727,
      "grad_norm": 6.46875,
      "learning_rate": 4.9684701411308796e-05,
      "loss": 1.2461,
      "num_input_tokens_seen": 11340832,
      "step": 19545
    },
    {
      "epoch": 2.911826035150432,
      "grad_norm": 8.25,
      "learning_rate": 4.9684186758462205e-05,
      "loss": 0.9111,
      "num_input_tokens_seen": 11343456,
      "step": 19550
    },
    {
      "epoch": 2.912570747691391,
      "grad_norm": 8.1875,
      "learning_rate": 4.968367168860136e-05,
      "loss": 0.9971,
      "num_input_tokens_seen": 11346176,
      "step": 19555
    },
    {
      "epoch": 2.9133154602323503,
      "grad_norm": 11.1875,
      "learning_rate": 4.968315620173496e-05,
      "loss": 1.0297,
      "num_input_tokens_seen": 11348928,
      "step": 19560
    },
    {
      "epoch": 2.9140601727733095,
      "grad_norm": 14.8125,
      "learning_rate": 4.968264029787173e-05,
      "loss": 1.1266,
      "num_input_tokens_seen": 11351520,
      "step": 19565
    },
    {
      "epoch": 2.9148048853142687,
      "grad_norm": 11.25,
      "learning_rate": 4.9682123977020385e-05,
      "loss": 1.1049,
      "num_input_tokens_seen": 11354464,
      "step": 19570
    },
    {
      "epoch": 2.915549597855228,
      "grad_norm": 18.375,
      "learning_rate": 4.968160723918963e-05,
      "loss": 1.5109,
      "num_input_tokens_seen": 11357632,
      "step": 19575
    },
    {
      "epoch": 2.916294310396187,
      "grad_norm": 6.53125,
      "learning_rate": 4.968109008438821e-05,
      "loss": 1.1685,
      "num_input_tokens_seen": 11360320,
      "step": 19580
    },
    {
      "epoch": 2.9170390229371463,
      "grad_norm": 11.75,
      "learning_rate": 4.9680572512624865e-05,
      "loss": 1.1181,
      "num_input_tokens_seen": 11363136,
      "step": 19585
    },
    {
      "epoch": 2.9177837354781055,
      "grad_norm": 10.0,
      "learning_rate": 4.968005452390832e-05,
      "loss": 0.8603,
      "num_input_tokens_seen": 11365888,
      "step": 19590
    },
    {
      "epoch": 2.9185284480190647,
      "grad_norm": 13.1875,
      "learning_rate": 4.967953611824735e-05,
      "loss": 1.1579,
      "num_input_tokens_seen": 11368832,
      "step": 19595
    },
    {
      "epoch": 2.919273160560024,
      "grad_norm": 9.5,
      "learning_rate": 4.9679017295650694e-05,
      "loss": 0.8712,
      "num_input_tokens_seen": 11371648,
      "step": 19600
    },
    {
      "epoch": 2.920017873100983,
      "grad_norm": 15.25,
      "learning_rate": 4.9678498056127124e-05,
      "loss": 1.1831,
      "num_input_tokens_seen": 11374560,
      "step": 19605
    },
    {
      "epoch": 2.9207625856419424,
      "grad_norm": 10.3125,
      "learning_rate": 4.967797839968541e-05,
      "loss": 1.1193,
      "num_input_tokens_seen": 11377600,
      "step": 19610
    },
    {
      "epoch": 2.9215072981829016,
      "grad_norm": 8.75,
      "learning_rate": 4.9677458326334336e-05,
      "loss": 0.8161,
      "num_input_tokens_seen": 11380544,
      "step": 19615
    },
    {
      "epoch": 2.9222520107238603,
      "grad_norm": 18.5,
      "learning_rate": 4.967693783608268e-05,
      "loss": 1.3193,
      "num_input_tokens_seen": 11383520,
      "step": 19620
    },
    {
      "epoch": 2.92299672326482,
      "grad_norm": 10.0625,
      "learning_rate": 4.967641692893924e-05,
      "loss": 1.0077,
      "num_input_tokens_seen": 11386336,
      "step": 19625
    },
    {
      "epoch": 2.9237414358057787,
      "grad_norm": 10.1875,
      "learning_rate": 4.967589560491282e-05,
      "loss": 1.1199,
      "num_input_tokens_seen": 11388960,
      "step": 19630
    },
    {
      "epoch": 2.9244861483467384,
      "grad_norm": 8.6875,
      "learning_rate": 4.967537386401222e-05,
      "loss": 0.8413,
      "num_input_tokens_seen": 11391520,
      "step": 19635
    },
    {
      "epoch": 2.925230860887697,
      "grad_norm": 14.0,
      "learning_rate": 4.967485170624625e-05,
      "loss": 1.1429,
      "num_input_tokens_seen": 11394624,
      "step": 19640
    },
    {
      "epoch": 2.9259755734286568,
      "grad_norm": 6.625,
      "learning_rate": 4.9674329131623756e-05,
      "loss": 1.1752,
      "num_input_tokens_seen": 11397408,
      "step": 19645
    },
    {
      "epoch": 2.9267202859696155,
      "grad_norm": 12.8125,
      "learning_rate": 4.967380614015354e-05,
      "loss": 1.2417,
      "num_input_tokens_seen": 11400256,
      "step": 19650
    },
    {
      "epoch": 2.927464998510575,
      "grad_norm": 7.0625,
      "learning_rate": 4.9673282731844444e-05,
      "loss": 1.0017,
      "num_input_tokens_seen": 11403136,
      "step": 19655
    },
    {
      "epoch": 2.928209711051534,
      "grad_norm": 8.3125,
      "learning_rate": 4.967275890670532e-05,
      "loss": 1.1591,
      "num_input_tokens_seen": 11406016,
      "step": 19660
    },
    {
      "epoch": 2.928954423592493,
      "grad_norm": 9.625,
      "learning_rate": 4.967223466474501e-05,
      "loss": 1.0587,
      "num_input_tokens_seen": 11408736,
      "step": 19665
    },
    {
      "epoch": 2.9296991361334523,
      "grad_norm": 6.59375,
      "learning_rate": 4.967171000597236e-05,
      "loss": 1.3274,
      "num_input_tokens_seen": 11411968,
      "step": 19670
    },
    {
      "epoch": 2.9304438486744115,
      "grad_norm": 10.375,
      "learning_rate": 4.967118493039625e-05,
      "loss": 1.2789,
      "num_input_tokens_seen": 11414880,
      "step": 19675
    },
    {
      "epoch": 2.9311885612153707,
      "grad_norm": 12.1875,
      "learning_rate": 4.9670659438025545e-05,
      "loss": 0.9178,
      "num_input_tokens_seen": 11417632,
      "step": 19680
    },
    {
      "epoch": 2.93193327375633,
      "grad_norm": 9.5625,
      "learning_rate": 4.967013352886913e-05,
      "loss": 1.0468,
      "num_input_tokens_seen": 11420416,
      "step": 19685
    },
    {
      "epoch": 2.932677986297289,
      "grad_norm": 13.0,
      "learning_rate": 4.9669607202935876e-05,
      "loss": 1.0102,
      "num_input_tokens_seen": 11423328,
      "step": 19690
    },
    {
      "epoch": 2.9334226988382484,
      "grad_norm": 12.875,
      "learning_rate": 4.966908046023468e-05,
      "loss": 0.9427,
      "num_input_tokens_seen": 11426048,
      "step": 19695
    },
    {
      "epoch": 2.9341674113792076,
      "grad_norm": 17.0,
      "learning_rate": 4.966855330077445e-05,
      "loss": 1.0921,
      "num_input_tokens_seen": 11428800,
      "step": 19700
    },
    {
      "epoch": 2.9349121239201668,
      "grad_norm": 15.75,
      "learning_rate": 4.966802572456408e-05,
      "loss": 1.3286,
      "num_input_tokens_seen": 11431872,
      "step": 19705
    },
    {
      "epoch": 2.935656836461126,
      "grad_norm": 11.1875,
      "learning_rate": 4.966749773161249e-05,
      "loss": 1.1157,
      "num_input_tokens_seen": 11434976,
      "step": 19710
    },
    {
      "epoch": 2.936401549002085,
      "grad_norm": 9.75,
      "learning_rate": 4.966696932192859e-05,
      "loss": 1.4911,
      "num_input_tokens_seen": 11437792,
      "step": 19715
    },
    {
      "epoch": 2.9371462615430444,
      "grad_norm": 9.25,
      "learning_rate": 4.9666440495521313e-05,
      "loss": 1.2092,
      "num_input_tokens_seen": 11440544,
      "step": 19720
    },
    {
      "epoch": 2.9378909740840036,
      "grad_norm": 19.125,
      "learning_rate": 4.96659112523996e-05,
      "loss": 1.172,
      "num_input_tokens_seen": 11443360,
      "step": 19725
    },
    {
      "epoch": 2.9386356866249628,
      "grad_norm": 10.8125,
      "learning_rate": 4.9665381592572387e-05,
      "loss": 1.3444,
      "num_input_tokens_seen": 11446272,
      "step": 19730
    },
    {
      "epoch": 2.939380399165922,
      "grad_norm": 13.5625,
      "learning_rate": 4.9664851516048615e-05,
      "loss": 0.9315,
      "num_input_tokens_seen": 11449568,
      "step": 19735
    },
    {
      "epoch": 2.940125111706881,
      "grad_norm": 7.4375,
      "learning_rate": 4.9664321022837244e-05,
      "loss": 1.053,
      "num_input_tokens_seen": 11452512,
      "step": 19740
    },
    {
      "epoch": 2.9408698242478404,
      "grad_norm": 6.03125,
      "learning_rate": 4.966379011294724e-05,
      "loss": 1.1086,
      "num_input_tokens_seen": 11455168,
      "step": 19745
    },
    {
      "epoch": 2.9416145367887996,
      "grad_norm": 10.0,
      "learning_rate": 4.966325878638757e-05,
      "loss": 1.2114,
      "num_input_tokens_seen": 11458592,
      "step": 19750
    },
    {
      "epoch": 2.942359249329759,
      "grad_norm": 8.8125,
      "learning_rate": 4.966272704316721e-05,
      "loss": 0.7639,
      "num_input_tokens_seen": 11461408,
      "step": 19755
    },
    {
      "epoch": 2.943103961870718,
      "grad_norm": 9.8125,
      "learning_rate": 4.966219488329514e-05,
      "loss": 0.8962,
      "num_input_tokens_seen": 11464256,
      "step": 19760
    },
    {
      "epoch": 2.943848674411677,
      "grad_norm": 12.3125,
      "learning_rate": 4.966166230678035e-05,
      "loss": 1.1634,
      "num_input_tokens_seen": 11467072,
      "step": 19765
    },
    {
      "epoch": 2.9445933869526364,
      "grad_norm": 12.0625,
      "learning_rate": 4.966112931363185e-05,
      "loss": 0.7162,
      "num_input_tokens_seen": 11469856,
      "step": 19770
    },
    {
      "epoch": 2.9453380994935956,
      "grad_norm": 10.4375,
      "learning_rate": 4.966059590385863e-05,
      "loss": 0.7132,
      "num_input_tokens_seen": 11472736,
      "step": 19775
    },
    {
      "epoch": 2.946082812034555,
      "grad_norm": 9.5625,
      "learning_rate": 4.9660062077469706e-05,
      "loss": 1.0917,
      "num_input_tokens_seen": 11476000,
      "step": 19780
    },
    {
      "epoch": 2.946827524575514,
      "grad_norm": 8.8125,
      "learning_rate": 4.965952783447409e-05,
      "loss": 1.1348,
      "num_input_tokens_seen": 11479104,
      "step": 19785
    },
    {
      "epoch": 2.947572237116473,
      "grad_norm": 3.078125,
      "learning_rate": 4.965899317488082e-05,
      "loss": 0.8351,
      "num_input_tokens_seen": 11481888,
      "step": 19790
    },
    {
      "epoch": 2.948316949657432,
      "grad_norm": 9.1875,
      "learning_rate": 4.9658458098698926e-05,
      "loss": 0.7284,
      "num_input_tokens_seen": 11484768,
      "step": 19795
    },
    {
      "epoch": 2.9490616621983916,
      "grad_norm": 11.875,
      "learning_rate": 4.965792260593744e-05,
      "loss": 1.0094,
      "num_input_tokens_seen": 11488064,
      "step": 19800
    },
    {
      "epoch": 2.9498063747393504,
      "grad_norm": 14.6875,
      "learning_rate": 4.965738669660541e-05,
      "loss": 1.0408,
      "num_input_tokens_seen": 11490944,
      "step": 19805
    },
    {
      "epoch": 2.95055108728031,
      "grad_norm": 7.90625,
      "learning_rate": 4.96568503707119e-05,
      "loss": 1.145,
      "num_input_tokens_seen": 11493920,
      "step": 19810
    },
    {
      "epoch": 2.9512957998212688,
      "grad_norm": 10.9375,
      "learning_rate": 4.965631362826596e-05,
      "loss": 1.0432,
      "num_input_tokens_seen": 11496800,
      "step": 19815
    },
    {
      "epoch": 2.9520405123622284,
      "grad_norm": 8.375,
      "learning_rate": 4.965577646927666e-05,
      "loss": 1.0185,
      "num_input_tokens_seen": 11499456,
      "step": 19820
    },
    {
      "epoch": 2.952785224903187,
      "grad_norm": 5.46875,
      "learning_rate": 4.965523889375308e-05,
      "loss": 1.141,
      "num_input_tokens_seen": 11502368,
      "step": 19825
    },
    {
      "epoch": 2.953529937444147,
      "grad_norm": 15.625,
      "learning_rate": 4.9654700901704286e-05,
      "loss": 1.1742,
      "num_input_tokens_seen": 11504896,
      "step": 19830
    },
    {
      "epoch": 2.9542746499851056,
      "grad_norm": 7.96875,
      "learning_rate": 4.965416249313939e-05,
      "loss": 1.0585,
      "num_input_tokens_seen": 11507488,
      "step": 19835
    },
    {
      "epoch": 2.955019362526065,
      "grad_norm": 7.84375,
      "learning_rate": 4.965362366806747e-05,
      "loss": 1.0701,
      "num_input_tokens_seen": 11510368,
      "step": 19840
    },
    {
      "epoch": 2.955764075067024,
      "grad_norm": 11.5,
      "learning_rate": 4.9653084426497633e-05,
      "loss": 0.9299,
      "num_input_tokens_seen": 11513088,
      "step": 19845
    },
    {
      "epoch": 2.956508787607983,
      "grad_norm": 7.0,
      "learning_rate": 4.965254476843899e-05,
      "loss": 0.7069,
      "num_input_tokens_seen": 11515872,
      "step": 19850
    },
    {
      "epoch": 2.9572535001489424,
      "grad_norm": 9.125,
      "learning_rate": 4.965200469390067e-05,
      "loss": 1.1191,
      "num_input_tokens_seen": 11518528,
      "step": 19855
    },
    {
      "epoch": 2.9579982126899016,
      "grad_norm": 11.0625,
      "learning_rate": 4.965146420289177e-05,
      "loss": 1.2145,
      "num_input_tokens_seen": 11521568,
      "step": 19860
    },
    {
      "epoch": 2.958742925230861,
      "grad_norm": 4.0625,
      "learning_rate": 4.965092329542145e-05,
      "loss": 1.0121,
      "num_input_tokens_seen": 11524224,
      "step": 19865
    },
    {
      "epoch": 2.95948763777182,
      "grad_norm": 12.3125,
      "learning_rate": 4.9650381971498824e-05,
      "loss": 0.87,
      "num_input_tokens_seen": 11527136,
      "step": 19870
    },
    {
      "epoch": 2.960232350312779,
      "grad_norm": 10.4375,
      "learning_rate": 4.964984023113306e-05,
      "loss": 1.1336,
      "num_input_tokens_seen": 11529760,
      "step": 19875
    },
    {
      "epoch": 2.9609770628537384,
      "grad_norm": 13.3125,
      "learning_rate": 4.9649298074333294e-05,
      "loss": 1.0775,
      "num_input_tokens_seen": 11532480,
      "step": 19880
    },
    {
      "epoch": 2.9617217753946976,
      "grad_norm": 16.375,
      "learning_rate": 4.964875550110869e-05,
      "loss": 1.1856,
      "num_input_tokens_seen": 11535552,
      "step": 19885
    },
    {
      "epoch": 2.962466487935657,
      "grad_norm": 7.59375,
      "learning_rate": 4.964821251146841e-05,
      "loss": 0.8977,
      "num_input_tokens_seen": 11538368,
      "step": 19890
    },
    {
      "epoch": 2.963211200476616,
      "grad_norm": 15.125,
      "learning_rate": 4.964766910542164e-05,
      "loss": 1.2328,
      "num_input_tokens_seen": 11540768,
      "step": 19895
    },
    {
      "epoch": 2.963955913017575,
      "grad_norm": 8.75,
      "learning_rate": 4.9647125282977536e-05,
      "loss": 0.8934,
      "num_input_tokens_seen": 11543808,
      "step": 19900
    },
    {
      "epoch": 2.9647006255585344,
      "grad_norm": 13.75,
      "learning_rate": 4.964658104414531e-05,
      "loss": 1.0636,
      "num_input_tokens_seen": 11547232,
      "step": 19905
    },
    {
      "epoch": 2.9654453380994936,
      "grad_norm": 11.1875,
      "learning_rate": 4.964603638893415e-05,
      "loss": 1.0413,
      "num_input_tokens_seen": 11550400,
      "step": 19910
    },
    {
      "epoch": 2.966190050640453,
      "grad_norm": 8.75,
      "learning_rate": 4.9645491317353246e-05,
      "loss": 1.1333,
      "num_input_tokens_seen": 11553120,
      "step": 19915
    },
    {
      "epoch": 2.966934763181412,
      "grad_norm": 5.46875,
      "learning_rate": 4.9644945829411815e-05,
      "loss": 0.9867,
      "num_input_tokens_seen": 11555968,
      "step": 19920
    },
    {
      "epoch": 2.9676794757223712,
      "grad_norm": 9.1875,
      "learning_rate": 4.964439992511908e-05,
      "loss": 1.3202,
      "num_input_tokens_seen": 11558784,
      "step": 19925
    },
    {
      "epoch": 2.9684241882633304,
      "grad_norm": 5.90625,
      "learning_rate": 4.964385360448425e-05,
      "loss": 1.0325,
      "num_input_tokens_seen": 11561664,
      "step": 19930
    },
    {
      "epoch": 2.9691689008042896,
      "grad_norm": 13.4375,
      "learning_rate": 4.964330686751656e-05,
      "loss": 0.8093,
      "num_input_tokens_seen": 11564512,
      "step": 19935
    },
    {
      "epoch": 2.969913613345249,
      "grad_norm": 15.4375,
      "learning_rate": 4.964275971422525e-05,
      "loss": 1.174,
      "num_input_tokens_seen": 11567136,
      "step": 19940
    },
    {
      "epoch": 2.970658325886208,
      "grad_norm": 4.34375,
      "learning_rate": 4.964221214461956e-05,
      "loss": 1.1467,
      "num_input_tokens_seen": 11569856,
      "step": 19945
    },
    {
      "epoch": 2.9714030384271672,
      "grad_norm": 12.4375,
      "learning_rate": 4.964166415870874e-05,
      "loss": 1.2491,
      "num_input_tokens_seen": 11573024,
      "step": 19950
    },
    {
      "epoch": 2.9721477509681264,
      "grad_norm": 6.875,
      "learning_rate": 4.964111575650205e-05,
      "loss": 0.7912,
      "num_input_tokens_seen": 11575616,
      "step": 19955
    },
    {
      "epoch": 2.9728924635090856,
      "grad_norm": 8.125,
      "learning_rate": 4.9640566938008745e-05,
      "loss": 1.0685,
      "num_input_tokens_seen": 11578976,
      "step": 19960
    },
    {
      "epoch": 2.973637176050045,
      "grad_norm": 4.6875,
      "learning_rate": 4.964001770323812e-05,
      "loss": 0.6298,
      "num_input_tokens_seen": 11582016,
      "step": 19965
    },
    {
      "epoch": 2.9743818885910036,
      "grad_norm": 4.28125,
      "learning_rate": 4.9639468052199426e-05,
      "loss": 0.8942,
      "num_input_tokens_seen": 11585376,
      "step": 19970
    },
    {
      "epoch": 2.9751266011319633,
      "grad_norm": 11.625,
      "learning_rate": 4.963891798490197e-05,
      "loss": 0.8813,
      "num_input_tokens_seen": 11588224,
      "step": 19975
    },
    {
      "epoch": 2.975871313672922,
      "grad_norm": 12.4375,
      "learning_rate": 4.963836750135503e-05,
      "loss": 1.2448,
      "num_input_tokens_seen": 11591072,
      "step": 19980
    },
    {
      "epoch": 2.9766160262138817,
      "grad_norm": 9.375,
      "learning_rate": 4.963781660156792e-05,
      "loss": 1.2982,
      "num_input_tokens_seen": 11593728,
      "step": 19985
    },
    {
      "epoch": 2.9773607387548404,
      "grad_norm": 12.25,
      "learning_rate": 4.9637265285549935e-05,
      "loss": 0.9883,
      "num_input_tokens_seen": 11596384,
      "step": 19990
    },
    {
      "epoch": 2.9781054512958,
      "grad_norm": 9.6875,
      "learning_rate": 4.9636713553310396e-05,
      "loss": 1.2303,
      "num_input_tokens_seen": 11599456,
      "step": 19995
    },
    {
      "epoch": 2.978850163836759,
      "grad_norm": 15.3125,
      "learning_rate": 4.963616140485862e-05,
      "loss": 1.0129,
      "num_input_tokens_seen": 11602208,
      "step": 20000
    },
    {
      "epoch": 2.9795948763777185,
      "grad_norm": 14.3125,
      "learning_rate": 4.963560884020393e-05,
      "loss": 1.3647,
      "num_input_tokens_seen": 11605184,
      "step": 20005
    },
    {
      "epoch": 2.9803395889186772,
      "grad_norm": 15.8125,
      "learning_rate": 4.963505585935567e-05,
      "loss": 1.0575,
      "num_input_tokens_seen": 11608000,
      "step": 20010
    },
    {
      "epoch": 2.9810843014596364,
      "grad_norm": 11.9375,
      "learning_rate": 4.9634502462323186e-05,
      "loss": 1.0779,
      "num_input_tokens_seen": 11610848,
      "step": 20015
    },
    {
      "epoch": 2.9818290140005956,
      "grad_norm": 12.9375,
      "learning_rate": 4.9633948649115816e-05,
      "loss": 0.9983,
      "num_input_tokens_seen": 11613696,
      "step": 20020
    },
    {
      "epoch": 2.982573726541555,
      "grad_norm": 6.84375,
      "learning_rate": 4.9633394419742917e-05,
      "loss": 1.2594,
      "num_input_tokens_seen": 11616864,
      "step": 20025
    },
    {
      "epoch": 2.983318439082514,
      "grad_norm": 5.53125,
      "learning_rate": 4.963283977421386e-05,
      "loss": 0.9793,
      "num_input_tokens_seen": 11619904,
      "step": 20030
    },
    {
      "epoch": 2.9840631516234732,
      "grad_norm": 9.125,
      "learning_rate": 4.9632284712538005e-05,
      "loss": 1.3096,
      "num_input_tokens_seen": 11622816,
      "step": 20035
    },
    {
      "epoch": 2.9848078641644324,
      "grad_norm": 17.25,
      "learning_rate": 4.9631729234724736e-05,
      "loss": 1.0562,
      "num_input_tokens_seen": 11625888,
      "step": 20040
    },
    {
      "epoch": 2.9855525767053916,
      "grad_norm": 11.875,
      "learning_rate": 4.9631173340783445e-05,
      "loss": 0.9758,
      "num_input_tokens_seen": 11629088,
      "step": 20045
    },
    {
      "epoch": 2.986297289246351,
      "grad_norm": 12.9375,
      "learning_rate": 4.96306170307235e-05,
      "loss": 1.39,
      "num_input_tokens_seen": 11632128,
      "step": 20050
    },
    {
      "epoch": 2.98704200178731,
      "grad_norm": 11.875,
      "learning_rate": 4.963006030455433e-05,
      "loss": 0.9554,
      "num_input_tokens_seen": 11634912,
      "step": 20055
    },
    {
      "epoch": 2.9877867143282693,
      "grad_norm": 15.5625,
      "learning_rate": 4.962950316228532e-05,
      "loss": 1.2868,
      "num_input_tokens_seen": 11637664,
      "step": 20060
    },
    {
      "epoch": 2.9885314268692285,
      "grad_norm": 12.0,
      "learning_rate": 4.9628945603925884e-05,
      "loss": 1.2949,
      "num_input_tokens_seen": 11640416,
      "step": 20065
    },
    {
      "epoch": 2.9892761394101877,
      "grad_norm": 6.875,
      "learning_rate": 4.9628387629485435e-05,
      "loss": 1.0952,
      "num_input_tokens_seen": 11643424,
      "step": 20070
    },
    {
      "epoch": 2.990020851951147,
      "grad_norm": 9.25,
      "learning_rate": 4.962782923897342e-05,
      "loss": 0.9656,
      "num_input_tokens_seen": 11646400,
      "step": 20075
    },
    {
      "epoch": 2.990765564492106,
      "grad_norm": 12.25,
      "learning_rate": 4.962727043239925e-05,
      "loss": 0.9651,
      "num_input_tokens_seen": 11649664,
      "step": 20080
    },
    {
      "epoch": 2.9915102770330653,
      "grad_norm": 12.8125,
      "learning_rate": 4.962671120977238e-05,
      "loss": 1.2234,
      "num_input_tokens_seen": 11652768,
      "step": 20085
    },
    {
      "epoch": 2.9922549895740245,
      "grad_norm": 10.5625,
      "learning_rate": 4.962615157110226e-05,
      "loss": 0.8727,
      "num_input_tokens_seen": 11655552,
      "step": 20090
    },
    {
      "epoch": 2.9929997021149837,
      "grad_norm": 14.8125,
      "learning_rate": 4.9625591516398336e-05,
      "loss": 0.6815,
      "num_input_tokens_seen": 11658272,
      "step": 20095
    },
    {
      "epoch": 2.993744414655943,
      "grad_norm": 11.5,
      "learning_rate": 4.962503104567007e-05,
      "loss": 1.1181,
      "num_input_tokens_seen": 11661248,
      "step": 20100
    },
    {
      "epoch": 2.994489127196902,
      "grad_norm": 14.6875,
      "learning_rate": 4.9624470158926925e-05,
      "loss": 1.083,
      "num_input_tokens_seen": 11664000,
      "step": 20105
    },
    {
      "epoch": 2.9952338397378613,
      "grad_norm": 7.875,
      "learning_rate": 4.962390885617839e-05,
      "loss": 0.9426,
      "num_input_tokens_seen": 11666848,
      "step": 20110
    },
    {
      "epoch": 2.9959785522788205,
      "grad_norm": 9.75,
      "learning_rate": 4.9623347137433954e-05,
      "loss": 0.9013,
      "num_input_tokens_seen": 11669600,
      "step": 20115
    },
    {
      "epoch": 2.9967232648197797,
      "grad_norm": 12.5625,
      "learning_rate": 4.962278500270307e-05,
      "loss": 1.0397,
      "num_input_tokens_seen": 11672352,
      "step": 20120
    },
    {
      "epoch": 2.997467977360739,
      "grad_norm": 3.765625,
      "learning_rate": 4.9622222451995274e-05,
      "loss": 0.8998,
      "num_input_tokens_seen": 11675104,
      "step": 20125
    },
    {
      "epoch": 2.998212689901698,
      "grad_norm": 18.125,
      "learning_rate": 4.962165948532006e-05,
      "loss": 1.1491,
      "num_input_tokens_seen": 11677696,
      "step": 20130
    },
    {
      "epoch": 2.9989574024426573,
      "grad_norm": 12.625,
      "learning_rate": 4.962109610268692e-05,
      "loss": 0.6968,
      "num_input_tokens_seen": 11680640,
      "step": 20135
    },
    {
      "epoch": 2.9997021149836165,
      "grad_norm": 4.625,
      "learning_rate": 4.9620532304105385e-05,
      "loss": 0.9582,
      "num_input_tokens_seen": 11683648,
      "step": 20140
    },
    {
      "epoch": 3.0,
      "eval_loss": 1.0888701677322388,
      "eval_runtime": 43.7416,
      "eval_samples_per_second": 68.219,
      "eval_steps_per_second": 17.055,
      "num_input_tokens_seen": 11684296,
      "step": 20142
    },
    {
      "epoch": 3.0004468275245757,
      "grad_norm": 7.90625,
      "learning_rate": 4.961996808958499e-05,
      "loss": 1.4373,
      "num_input_tokens_seen": 11685992,
      "step": 20145
    },
    {
      "epoch": 3.001191540065535,
      "grad_norm": 8.0625,
      "learning_rate": 4.961940345913525e-05,
      "loss": 0.8752,
      "num_input_tokens_seen": 11688648,
      "step": 20150
    },
    {
      "epoch": 3.001936252606494,
      "grad_norm": 10.875,
      "learning_rate": 4.961883841276571e-05,
      "loss": 0.9999,
      "num_input_tokens_seen": 11691816,
      "step": 20155
    },
    {
      "epoch": 3.002680965147453,
      "grad_norm": 7.59375,
      "learning_rate": 4.961827295048592e-05,
      "loss": 0.8102,
      "num_input_tokens_seen": 11694664,
      "step": 20160
    },
    {
      "epoch": 3.003425677688412,
      "grad_norm": 6.75,
      "learning_rate": 4.961770707230543e-05,
      "loss": 1.1432,
      "num_input_tokens_seen": 11697512,
      "step": 20165
    },
    {
      "epoch": 3.0041703902293713,
      "grad_norm": 10.25,
      "learning_rate": 4.961714077823379e-05,
      "loss": 1.2767,
      "num_input_tokens_seen": 11700392,
      "step": 20170
    },
    {
      "epoch": 3.0049151027703305,
      "grad_norm": 22.5,
      "learning_rate": 4.961657406828059e-05,
      "loss": 0.9132,
      "num_input_tokens_seen": 11703688,
      "step": 20175
    },
    {
      "epoch": 3.0056598153112897,
      "grad_norm": 20.125,
      "learning_rate": 4.961600694245539e-05,
      "loss": 1.3944,
      "num_input_tokens_seen": 11706312,
      "step": 20180
    },
    {
      "epoch": 3.006404527852249,
      "grad_norm": 11.625,
      "learning_rate": 4.961543940076776e-05,
      "loss": 1.2746,
      "num_input_tokens_seen": 11709064,
      "step": 20185
    },
    {
      "epoch": 3.007149240393208,
      "grad_norm": 8.5625,
      "learning_rate": 4.961487144322731e-05,
      "loss": 1.0201,
      "num_input_tokens_seen": 11711816,
      "step": 20190
    },
    {
      "epoch": 3.0078939529341673,
      "grad_norm": 10.3125,
      "learning_rate": 4.961430306984362e-05,
      "loss": 1.0728,
      "num_input_tokens_seen": 11714760,
      "step": 20195
    },
    {
      "epoch": 3.0086386654751265,
      "grad_norm": 6.25,
      "learning_rate": 4.9613734280626287e-05,
      "loss": 0.7642,
      "num_input_tokens_seen": 11717416,
      "step": 20200
    },
    {
      "epoch": 3.0093833780160857,
      "grad_norm": 16.25,
      "learning_rate": 4.961316507558494e-05,
      "loss": 1.4872,
      "num_input_tokens_seen": 11720168,
      "step": 20205
    },
    {
      "epoch": 3.010128090557045,
      "grad_norm": 10.8125,
      "learning_rate": 4.961259545472918e-05,
      "loss": 1.141,
      "num_input_tokens_seen": 11723144,
      "step": 20210
    },
    {
      "epoch": 3.010872803098004,
      "grad_norm": 9.5,
      "learning_rate": 4.961202541806864e-05,
      "loss": 1.3997,
      "num_input_tokens_seen": 11725896,
      "step": 20215
    },
    {
      "epoch": 3.0116175156389633,
      "grad_norm": 8.6875,
      "learning_rate": 4.9611454965612944e-05,
      "loss": 0.9988,
      "num_input_tokens_seen": 11728840,
      "step": 20220
    },
    {
      "epoch": 3.0123622281799225,
      "grad_norm": 9.75,
      "learning_rate": 4.9610884097371736e-05,
      "loss": 0.9928,
      "num_input_tokens_seen": 11731464,
      "step": 20225
    },
    {
      "epoch": 3.0131069407208817,
      "grad_norm": 5.09375,
      "learning_rate": 4.961031281335464e-05,
      "loss": 0.8491,
      "num_input_tokens_seen": 11734408,
      "step": 20230
    },
    {
      "epoch": 3.013851653261841,
      "grad_norm": 14.0,
      "learning_rate": 4.9609741113571336e-05,
      "loss": 1.1741,
      "num_input_tokens_seen": 11737064,
      "step": 20235
    },
    {
      "epoch": 3.0145963658028,
      "grad_norm": 11.5625,
      "learning_rate": 4.960916899803146e-05,
      "loss": 1.46,
      "num_input_tokens_seen": 11739848,
      "step": 20240
    },
    {
      "epoch": 3.0153410783437593,
      "grad_norm": 23.125,
      "learning_rate": 4.960859646674469e-05,
      "loss": 0.9969,
      "num_input_tokens_seen": 11742760,
      "step": 20245
    },
    {
      "epoch": 3.0160857908847185,
      "grad_norm": 14.5,
      "learning_rate": 4.960802351972069e-05,
      "loss": 1.1441,
      "num_input_tokens_seen": 11745768,
      "step": 20250
    },
    {
      "epoch": 3.0168305034256777,
      "grad_norm": 10.6875,
      "learning_rate": 4.960745015696914e-05,
      "loss": 0.9147,
      "num_input_tokens_seen": 11748808,
      "step": 20255
    },
    {
      "epoch": 3.017575215966637,
      "grad_norm": 6.53125,
      "learning_rate": 4.960687637849974e-05,
      "loss": 0.9779,
      "num_input_tokens_seen": 11751368,
      "step": 20260
    },
    {
      "epoch": 3.018319928507596,
      "grad_norm": 16.375,
      "learning_rate": 4.960630218432216e-05,
      "loss": 1.0987,
      "num_input_tokens_seen": 11754376,
      "step": 20265
    },
    {
      "epoch": 3.0190646410485553,
      "grad_norm": 9.5625,
      "learning_rate": 4.960572757444612e-05,
      "loss": 1.0941,
      "num_input_tokens_seen": 11757768,
      "step": 20270
    },
    {
      "epoch": 3.0198093535895145,
      "grad_norm": 6.34375,
      "learning_rate": 4.960515254888133e-05,
      "loss": 1.3188,
      "num_input_tokens_seen": 11760488,
      "step": 20275
    },
    {
      "epoch": 3.0205540661304737,
      "grad_norm": 13.0,
      "learning_rate": 4.9604577107637484e-05,
      "loss": 1.3248,
      "num_input_tokens_seen": 11763304,
      "step": 20280
    },
    {
      "epoch": 3.021298778671433,
      "grad_norm": 6.96875,
      "learning_rate": 4.960400125072431e-05,
      "loss": 1.083,
      "num_input_tokens_seen": 11766152,
      "step": 20285
    },
    {
      "epoch": 3.022043491212392,
      "grad_norm": 8.875,
      "learning_rate": 4.960342497815155e-05,
      "loss": 1.3739,
      "num_input_tokens_seen": 11769032,
      "step": 20290
    },
    {
      "epoch": 3.0227882037533513,
      "grad_norm": 10.25,
      "learning_rate": 4.9602848289928926e-05,
      "loss": 1.0713,
      "num_input_tokens_seen": 11771880,
      "step": 20295
    },
    {
      "epoch": 3.0235329162943105,
      "grad_norm": 9.875,
      "learning_rate": 4.9602271186066194e-05,
      "loss": 1.4513,
      "num_input_tokens_seen": 11774856,
      "step": 20300
    },
    {
      "epoch": 3.0242776288352697,
      "grad_norm": 11.125,
      "learning_rate": 4.960169366657309e-05,
      "loss": 0.9831,
      "num_input_tokens_seen": 11777640,
      "step": 20305
    },
    {
      "epoch": 3.025022341376229,
      "grad_norm": 7.0625,
      "learning_rate": 4.960111573145937e-05,
      "loss": 0.965,
      "num_input_tokens_seen": 11780552,
      "step": 20310
    },
    {
      "epoch": 3.025767053917188,
      "grad_norm": 15.6875,
      "learning_rate": 4.960053738073481e-05,
      "loss": 0.9318,
      "num_input_tokens_seen": 11783496,
      "step": 20315
    },
    {
      "epoch": 3.0265117664581473,
      "grad_norm": 7.78125,
      "learning_rate": 4.959995861440917e-05,
      "loss": 0.9615,
      "num_input_tokens_seen": 11786184,
      "step": 20320
    },
    {
      "epoch": 3.0272564789991065,
      "grad_norm": 19.125,
      "learning_rate": 4.959937943249223e-05,
      "loss": 1.1938,
      "num_input_tokens_seen": 11788904,
      "step": 20325
    },
    {
      "epoch": 3.0280011915400658,
      "grad_norm": 12.8125,
      "learning_rate": 4.9598799834993784e-05,
      "loss": 1.2966,
      "num_input_tokens_seen": 11791848,
      "step": 20330
    },
    {
      "epoch": 3.0287459040810245,
      "grad_norm": 9.5,
      "learning_rate": 4.9598219821923605e-05,
      "loss": 1.169,
      "num_input_tokens_seen": 11794856,
      "step": 20335
    },
    {
      "epoch": 3.0294906166219837,
      "grad_norm": 11.5,
      "learning_rate": 4.959763939329152e-05,
      "loss": 1.0836,
      "num_input_tokens_seen": 11797608,
      "step": 20340
    },
    {
      "epoch": 3.030235329162943,
      "grad_norm": 9.25,
      "learning_rate": 4.95970585491073e-05,
      "loss": 0.9295,
      "num_input_tokens_seen": 11800456,
      "step": 20345
    },
    {
      "epoch": 3.030980041703902,
      "grad_norm": 5.03125,
      "learning_rate": 4.9596477289380786e-05,
      "loss": 0.675,
      "num_input_tokens_seen": 11803048,
      "step": 20350
    },
    {
      "epoch": 3.0317247542448613,
      "grad_norm": 6.625,
      "learning_rate": 4.959589561412178e-05,
      "loss": 1.0941,
      "num_input_tokens_seen": 11805960,
      "step": 20355
    },
    {
      "epoch": 3.0324694667858205,
      "grad_norm": 23.125,
      "learning_rate": 4.959531352334012e-05,
      "loss": 1.3088,
      "num_input_tokens_seen": 11808776,
      "step": 20360
    },
    {
      "epoch": 3.0332141793267797,
      "grad_norm": 18.375,
      "learning_rate": 4.959473101704563e-05,
      "loss": 1.514,
      "num_input_tokens_seen": 11811656,
      "step": 20365
    },
    {
      "epoch": 3.033958891867739,
      "grad_norm": 4.875,
      "learning_rate": 4.959414809524816e-05,
      "loss": 1.1588,
      "num_input_tokens_seen": 11814600,
      "step": 20370
    },
    {
      "epoch": 3.034703604408698,
      "grad_norm": 12.5625,
      "learning_rate": 4.9593564757957554e-05,
      "loss": 0.9693,
      "num_input_tokens_seen": 11817352,
      "step": 20375
    },
    {
      "epoch": 3.0354483169496573,
      "grad_norm": 13.9375,
      "learning_rate": 4.959298100518367e-05,
      "loss": 0.6614,
      "num_input_tokens_seen": 11820360,
      "step": 20380
    },
    {
      "epoch": 3.0361930294906165,
      "grad_norm": 4.8125,
      "learning_rate": 4.959239683693636e-05,
      "loss": 0.925,
      "num_input_tokens_seen": 11823432,
      "step": 20385
    },
    {
      "epoch": 3.0369377420315757,
      "grad_norm": 23.125,
      "learning_rate": 4.959181225322551e-05,
      "loss": 1.2771,
      "num_input_tokens_seen": 11826440,
      "step": 20390
    },
    {
      "epoch": 3.037682454572535,
      "grad_norm": 16.375,
      "learning_rate": 4.959122725406098e-05,
      "loss": 1.1499,
      "num_input_tokens_seen": 11829192,
      "step": 20395
    },
    {
      "epoch": 3.038427167113494,
      "grad_norm": 5.96875,
      "learning_rate": 4.959064183945266e-05,
      "loss": 1.121,
      "num_input_tokens_seen": 11832104,
      "step": 20400
    },
    {
      "epoch": 3.0391718796544533,
      "grad_norm": 8.9375,
      "learning_rate": 4.959005600941043e-05,
      "loss": 0.9175,
      "num_input_tokens_seen": 11835240,
      "step": 20405
    },
    {
      "epoch": 3.0399165921954125,
      "grad_norm": 11.0,
      "learning_rate": 4.958946976394421e-05,
      "loss": 1.0484,
      "num_input_tokens_seen": 11837960,
      "step": 20410
    },
    {
      "epoch": 3.0406613047363718,
      "grad_norm": 7.125,
      "learning_rate": 4.958888310306389e-05,
      "loss": 1.0426,
      "num_input_tokens_seen": 11840584,
      "step": 20415
    },
    {
      "epoch": 3.041406017277331,
      "grad_norm": 16.625,
      "learning_rate": 4.958829602677937e-05,
      "loss": 1.1798,
      "num_input_tokens_seen": 11843496,
      "step": 20420
    },
    {
      "epoch": 3.04215072981829,
      "grad_norm": 11.125,
      "learning_rate": 4.9587708535100584e-05,
      "loss": 1.3027,
      "num_input_tokens_seen": 11846184,
      "step": 20425
    },
    {
      "epoch": 3.0428954423592494,
      "grad_norm": 6.4375,
      "learning_rate": 4.958712062803745e-05,
      "loss": 1.0733,
      "num_input_tokens_seen": 11848872,
      "step": 20430
    },
    {
      "epoch": 3.0436401549002086,
      "grad_norm": 17.125,
      "learning_rate": 4.958653230559991e-05,
      "loss": 0.9755,
      "num_input_tokens_seen": 11851848,
      "step": 20435
    },
    {
      "epoch": 3.0443848674411678,
      "grad_norm": 8.0625,
      "learning_rate": 4.958594356779789e-05,
      "loss": 1.2241,
      "num_input_tokens_seen": 11854888,
      "step": 20440
    },
    {
      "epoch": 3.045129579982127,
      "grad_norm": 12.8125,
      "learning_rate": 4.958535441464134e-05,
      "loss": 1.0116,
      "num_input_tokens_seen": 11857800,
      "step": 20445
    },
    {
      "epoch": 3.045874292523086,
      "grad_norm": 6.625,
      "learning_rate": 4.958476484614022e-05,
      "loss": 1.0663,
      "num_input_tokens_seen": 11860968,
      "step": 20450
    },
    {
      "epoch": 3.0466190050640454,
      "grad_norm": 16.125,
      "learning_rate": 4.958417486230448e-05,
      "loss": 1.395,
      "num_input_tokens_seen": 11863944,
      "step": 20455
    },
    {
      "epoch": 3.0473637176050046,
      "grad_norm": 11.125,
      "learning_rate": 4.95835844631441e-05,
      "loss": 1.1828,
      "num_input_tokens_seen": 11867496,
      "step": 20460
    },
    {
      "epoch": 3.0481084301459638,
      "grad_norm": 20.125,
      "learning_rate": 4.958299364866903e-05,
      "loss": 1.2905,
      "num_input_tokens_seen": 11870568,
      "step": 20465
    },
    {
      "epoch": 3.048853142686923,
      "grad_norm": 10.0,
      "learning_rate": 4.958240241888928e-05,
      "loss": 1.0541,
      "num_input_tokens_seen": 11873512,
      "step": 20470
    },
    {
      "epoch": 3.049597855227882,
      "grad_norm": 18.875,
      "learning_rate": 4.958181077381482e-05,
      "loss": 1.0337,
      "num_input_tokens_seen": 11876360,
      "step": 20475
    },
    {
      "epoch": 3.0503425677688414,
      "grad_norm": 11.125,
      "learning_rate": 4.958121871345565e-05,
      "loss": 1.455,
      "num_input_tokens_seen": 11879048,
      "step": 20480
    },
    {
      "epoch": 3.0510872803098006,
      "grad_norm": 10.5,
      "learning_rate": 4.958062623782178e-05,
      "loss": 1.0771,
      "num_input_tokens_seen": 11882024,
      "step": 20485
    },
    {
      "epoch": 3.05183199285076,
      "grad_norm": 14.625,
      "learning_rate": 4.958003334692321e-05,
      "loss": 1.0908,
      "num_input_tokens_seen": 11884808,
      "step": 20490
    },
    {
      "epoch": 3.052576705391719,
      "grad_norm": 8.0,
      "learning_rate": 4.957944004076995e-05,
      "loss": 1.1916,
      "num_input_tokens_seen": 11888040,
      "step": 20495
    },
    {
      "epoch": 3.053321417932678,
      "grad_norm": 12.4375,
      "learning_rate": 4.957884631937204e-05,
      "loss": 1.1758,
      "num_input_tokens_seen": 11890696,
      "step": 20500
    },
    {
      "epoch": 3.054066130473637,
      "grad_norm": 17.375,
      "learning_rate": 4.9578252182739506e-05,
      "loss": 1.3444,
      "num_input_tokens_seen": 11894248,
      "step": 20505
    },
    {
      "epoch": 3.054810843014596,
      "grad_norm": 7.8125,
      "learning_rate": 4.957765763088237e-05,
      "loss": 0.9682,
      "num_input_tokens_seen": 11896936,
      "step": 20510
    },
    {
      "epoch": 3.0555555555555554,
      "grad_norm": 2.875,
      "learning_rate": 4.95770626638107e-05,
      "loss": 0.8252,
      "num_input_tokens_seen": 11899432,
      "step": 20515
    },
    {
      "epoch": 3.0563002680965146,
      "grad_norm": 11.875,
      "learning_rate": 4.9576467281534526e-05,
      "loss": 0.8775,
      "num_input_tokens_seen": 11902568,
      "step": 20520
    },
    {
      "epoch": 3.0570449806374738,
      "grad_norm": 26.75,
      "learning_rate": 4.9575871484063915e-05,
      "loss": 1.3879,
      "num_input_tokens_seen": 11905384,
      "step": 20525
    },
    {
      "epoch": 3.057789693178433,
      "grad_norm": 13.375,
      "learning_rate": 4.9575275271408944e-05,
      "loss": 1.434,
      "num_input_tokens_seen": 11908200,
      "step": 20530
    },
    {
      "epoch": 3.058534405719392,
      "grad_norm": 6.8125,
      "learning_rate": 4.957467864357967e-05,
      "loss": 0.665,
      "num_input_tokens_seen": 11911016,
      "step": 20535
    },
    {
      "epoch": 3.0592791182603514,
      "grad_norm": 13.1875,
      "learning_rate": 4.9574081600586175e-05,
      "loss": 1.5807,
      "num_input_tokens_seen": 11914088,
      "step": 20540
    },
    {
      "epoch": 3.0600238308013106,
      "grad_norm": 7.1875,
      "learning_rate": 4.957348414243855e-05,
      "loss": 0.9964,
      "num_input_tokens_seen": 11917128,
      "step": 20545
    },
    {
      "epoch": 3.0607685433422698,
      "grad_norm": 9.125,
      "learning_rate": 4.9572886269146877e-05,
      "loss": 1.2255,
      "num_input_tokens_seen": 11920104,
      "step": 20550
    },
    {
      "epoch": 3.061513255883229,
      "grad_norm": 11.625,
      "learning_rate": 4.957228798072128e-05,
      "loss": 1.0904,
      "num_input_tokens_seen": 11923048,
      "step": 20555
    },
    {
      "epoch": 3.062257968424188,
      "grad_norm": 9.0625,
      "learning_rate": 4.957168927717184e-05,
      "loss": 1.0121,
      "num_input_tokens_seen": 11925800,
      "step": 20560
    },
    {
      "epoch": 3.0630026809651474,
      "grad_norm": 14.4375,
      "learning_rate": 4.957109015850868e-05,
      "loss": 0.8794,
      "num_input_tokens_seen": 11928680,
      "step": 20565
    },
    {
      "epoch": 3.0637473935061066,
      "grad_norm": 6.0625,
      "learning_rate": 4.957049062474194e-05,
      "loss": 1.0195,
      "num_input_tokens_seen": 11931976,
      "step": 20570
    },
    {
      "epoch": 3.064492106047066,
      "grad_norm": 6.125,
      "learning_rate": 4.956989067588172e-05,
      "loss": 0.9812,
      "num_input_tokens_seen": 11935048,
      "step": 20575
    },
    {
      "epoch": 3.065236818588025,
      "grad_norm": 10.3125,
      "learning_rate": 4.956929031193817e-05,
      "loss": 1.0992,
      "num_input_tokens_seen": 11937928,
      "step": 20580
    },
    {
      "epoch": 3.065981531128984,
      "grad_norm": 5.96875,
      "learning_rate": 4.956868953292143e-05,
      "loss": 1.0756,
      "num_input_tokens_seen": 11940872,
      "step": 20585
    },
    {
      "epoch": 3.0667262436699434,
      "grad_norm": 11.125,
      "learning_rate": 4.9568088338841664e-05,
      "loss": 1.0473,
      "num_input_tokens_seen": 11943656,
      "step": 20590
    },
    {
      "epoch": 3.0674709562109026,
      "grad_norm": 9.6875,
      "learning_rate": 4.9567486729709e-05,
      "loss": 1.0025,
      "num_input_tokens_seen": 11946440,
      "step": 20595
    },
    {
      "epoch": 3.068215668751862,
      "grad_norm": 3.921875,
      "learning_rate": 4.956688470553363e-05,
      "loss": 1.1092,
      "num_input_tokens_seen": 11949320,
      "step": 20600
    },
    {
      "epoch": 3.068960381292821,
      "grad_norm": 14.5625,
      "learning_rate": 4.95662822663257e-05,
      "loss": 0.939,
      "num_input_tokens_seen": 11952104,
      "step": 20605
    },
    {
      "epoch": 3.06970509383378,
      "grad_norm": 14.25,
      "learning_rate": 4.9565679412095415e-05,
      "loss": 1.0772,
      "num_input_tokens_seen": 11954920,
      "step": 20610
    },
    {
      "epoch": 3.0704498063747394,
      "grad_norm": 5.1875,
      "learning_rate": 4.956507614285293e-05,
      "loss": 0.6516,
      "num_input_tokens_seen": 11957608,
      "step": 20615
    },
    {
      "epoch": 3.0711945189156986,
      "grad_norm": 10.125,
      "learning_rate": 4.9564472458608445e-05,
      "loss": 1.0942,
      "num_input_tokens_seen": 11960392,
      "step": 20620
    },
    {
      "epoch": 3.071939231456658,
      "grad_norm": 8.375,
      "learning_rate": 4.956386835937218e-05,
      "loss": 1.2655,
      "num_input_tokens_seen": 11963496,
      "step": 20625
    },
    {
      "epoch": 3.072683943997617,
      "grad_norm": 9.0625,
      "learning_rate": 4.9563263845154315e-05,
      "loss": 1.0755,
      "num_input_tokens_seen": 11966408,
      "step": 20630
    },
    {
      "epoch": 3.073428656538576,
      "grad_norm": 12.0625,
      "learning_rate": 4.9562658915965075e-05,
      "loss": 1.1818,
      "num_input_tokens_seen": 11969192,
      "step": 20635
    },
    {
      "epoch": 3.0741733690795354,
      "grad_norm": 13.125,
      "learning_rate": 4.956205357181467e-05,
      "loss": 1.1043,
      "num_input_tokens_seen": 11971912,
      "step": 20640
    },
    {
      "epoch": 3.0749180816204946,
      "grad_norm": 18.125,
      "learning_rate": 4.9561447812713345e-05,
      "loss": 1.1944,
      "num_input_tokens_seen": 11974760,
      "step": 20645
    },
    {
      "epoch": 3.075662794161454,
      "grad_norm": 10.0,
      "learning_rate": 4.956084163867132e-05,
      "loss": 0.9007,
      "num_input_tokens_seen": 11977384,
      "step": 20650
    },
    {
      "epoch": 3.076407506702413,
      "grad_norm": 10.75,
      "learning_rate": 4.9560235049698834e-05,
      "loss": 1.3957,
      "num_input_tokens_seen": 11981128,
      "step": 20655
    },
    {
      "epoch": 3.0771522192433722,
      "grad_norm": 9.0,
      "learning_rate": 4.955962804580614e-05,
      "loss": 1.1754,
      "num_input_tokens_seen": 11984072,
      "step": 20660
    },
    {
      "epoch": 3.0778969317843314,
      "grad_norm": 7.125,
      "learning_rate": 4.9559020627003494e-05,
      "loss": 1.66,
      "num_input_tokens_seen": 11986952,
      "step": 20665
    },
    {
      "epoch": 3.0786416443252906,
      "grad_norm": 18.25,
      "learning_rate": 4.955841279330115e-05,
      "loss": 0.8807,
      "num_input_tokens_seen": 11989960,
      "step": 20670
    },
    {
      "epoch": 3.07938635686625,
      "grad_norm": 8.3125,
      "learning_rate": 4.9557804544709385e-05,
      "loss": 1.1235,
      "num_input_tokens_seen": 11992840,
      "step": 20675
    },
    {
      "epoch": 3.0801310694072086,
      "grad_norm": 12.4375,
      "learning_rate": 4.955719588123847e-05,
      "loss": 1.3226,
      "num_input_tokens_seen": 11995784,
      "step": 20680
    },
    {
      "epoch": 3.080875781948168,
      "grad_norm": 12.375,
      "learning_rate": 4.955658680289869e-05,
      "loss": 1.0964,
      "num_input_tokens_seen": 11998888,
      "step": 20685
    },
    {
      "epoch": 3.081620494489127,
      "grad_norm": 10.125,
      "learning_rate": 4.955597730970034e-05,
      "loss": 1.1143,
      "num_input_tokens_seen": 12001832,
      "step": 20690
    },
    {
      "epoch": 3.082365207030086,
      "grad_norm": 9.3125,
      "learning_rate": 4.95553674016537e-05,
      "loss": 1.1042,
      "num_input_tokens_seen": 12004584,
      "step": 20695
    },
    {
      "epoch": 3.0831099195710454,
      "grad_norm": 8.9375,
      "learning_rate": 4.9554757078769095e-05,
      "loss": 0.9167,
      "num_input_tokens_seen": 12007624,
      "step": 20700
    },
    {
      "epoch": 3.0838546321120046,
      "grad_norm": 12.625,
      "learning_rate": 4.955414634105682e-05,
      "loss": 1.0711,
      "num_input_tokens_seen": 12010536,
      "step": 20705
    },
    {
      "epoch": 3.084599344652964,
      "grad_norm": 6.3125,
      "learning_rate": 4.95535351885272e-05,
      "loss": 1.1687,
      "num_input_tokens_seen": 12013672,
      "step": 20710
    },
    {
      "epoch": 3.085344057193923,
      "grad_norm": 9.4375,
      "learning_rate": 4.955292362119055e-05,
      "loss": 1.3144,
      "num_input_tokens_seen": 12016232,
      "step": 20715
    },
    {
      "epoch": 3.086088769734882,
      "grad_norm": 16.75,
      "learning_rate": 4.955231163905723e-05,
      "loss": 1.0319,
      "num_input_tokens_seen": 12019496,
      "step": 20720
    },
    {
      "epoch": 3.0868334822758414,
      "grad_norm": 12.75,
      "learning_rate": 4.955169924213754e-05,
      "loss": 1.403,
      "num_input_tokens_seen": 12022440,
      "step": 20725
    },
    {
      "epoch": 3.0875781948168006,
      "grad_norm": 11.875,
      "learning_rate": 4.955108643044185e-05,
      "loss": 1.0266,
      "num_input_tokens_seen": 12025448,
      "step": 20730
    },
    {
      "epoch": 3.08832290735776,
      "grad_norm": 14.4375,
      "learning_rate": 4.955047320398051e-05,
      "loss": 1.0549,
      "num_input_tokens_seen": 12028168,
      "step": 20735
    },
    {
      "epoch": 3.089067619898719,
      "grad_norm": 7.28125,
      "learning_rate": 4.954985956276388e-05,
      "loss": 0.8595,
      "num_input_tokens_seen": 12030952,
      "step": 20740
    },
    {
      "epoch": 3.0898123324396782,
      "grad_norm": 8.875,
      "learning_rate": 4.954924550680231e-05,
      "loss": 0.878,
      "num_input_tokens_seen": 12033768,
      "step": 20745
    },
    {
      "epoch": 3.0905570449806374,
      "grad_norm": 4.21875,
      "learning_rate": 4.95486310361062e-05,
      "loss": 1.0241,
      "num_input_tokens_seen": 12036488,
      "step": 20750
    },
    {
      "epoch": 3.0913017575215966,
      "grad_norm": 5.1875,
      "learning_rate": 4.954801615068592e-05,
      "loss": 1.0534,
      "num_input_tokens_seen": 12039496,
      "step": 20755
    },
    {
      "epoch": 3.092046470062556,
      "grad_norm": 19.25,
      "learning_rate": 4.9547400850551853e-05,
      "loss": 1.0948,
      "num_input_tokens_seen": 12042280,
      "step": 20760
    },
    {
      "epoch": 3.092791182603515,
      "grad_norm": 8.125,
      "learning_rate": 4.9546785135714394e-05,
      "loss": 0.9848,
      "num_input_tokens_seen": 12045256,
      "step": 20765
    },
    {
      "epoch": 3.0935358951444742,
      "grad_norm": 14.0625,
      "learning_rate": 4.954616900618395e-05,
      "loss": 1.0682,
      "num_input_tokens_seen": 12047944,
      "step": 20770
    },
    {
      "epoch": 3.0942806076854334,
      "grad_norm": 6.65625,
      "learning_rate": 4.954555246197093e-05,
      "loss": 1.017,
      "num_input_tokens_seen": 12050728,
      "step": 20775
    },
    {
      "epoch": 3.0950253202263927,
      "grad_norm": 13.75,
      "learning_rate": 4.954493550308575e-05,
      "loss": 0.8774,
      "num_input_tokens_seen": 12053448,
      "step": 20780
    },
    {
      "epoch": 3.095770032767352,
      "grad_norm": 9.8125,
      "learning_rate": 4.9544318129538824e-05,
      "loss": 1.5424,
      "num_input_tokens_seen": 12056296,
      "step": 20785
    },
    {
      "epoch": 3.096514745308311,
      "grad_norm": 10.5625,
      "learning_rate": 4.95437003413406e-05,
      "loss": 0.9362,
      "num_input_tokens_seen": 12058952,
      "step": 20790
    },
    {
      "epoch": 3.0972594578492703,
      "grad_norm": 8.5,
      "learning_rate": 4.9543082138501495e-05,
      "loss": 0.8733,
      "num_input_tokens_seen": 12061768,
      "step": 20795
    },
    {
      "epoch": 3.0980041703902295,
      "grad_norm": 8.1875,
      "learning_rate": 4.954246352103197e-05,
      "loss": 0.8728,
      "num_input_tokens_seen": 12064392,
      "step": 20800
    },
    {
      "epoch": 3.0987488829311887,
      "grad_norm": 14.8125,
      "learning_rate": 4.954184448894246e-05,
      "loss": 1.4414,
      "num_input_tokens_seen": 12067208,
      "step": 20805
    },
    {
      "epoch": 3.099493595472148,
      "grad_norm": 15.75,
      "learning_rate": 4.954122504224343e-05,
      "loss": 1.286,
      "num_input_tokens_seen": 12070216,
      "step": 20810
    },
    {
      "epoch": 3.100238308013107,
      "grad_norm": 15.5625,
      "learning_rate": 4.954060518094535e-05,
      "loss": 1.0911,
      "num_input_tokens_seen": 12072936,
      "step": 20815
    },
    {
      "epoch": 3.1009830205540663,
      "grad_norm": 10.375,
      "learning_rate": 4.953998490505868e-05,
      "loss": 1.0019,
      "num_input_tokens_seen": 12075784,
      "step": 20820
    },
    {
      "epoch": 3.1017277330950255,
      "grad_norm": 7.71875,
      "learning_rate": 4.953936421459392e-05,
      "loss": 1.1917,
      "num_input_tokens_seen": 12078760,
      "step": 20825
    },
    {
      "epoch": 3.1024724456359847,
      "grad_norm": 16.75,
      "learning_rate": 4.953874310956153e-05,
      "loss": 1.0565,
      "num_input_tokens_seen": 12081416,
      "step": 20830
    },
    {
      "epoch": 3.103217158176944,
      "grad_norm": 11.1875,
      "learning_rate": 4.953812158997202e-05,
      "loss": 1.1621,
      "num_input_tokens_seen": 12084104,
      "step": 20835
    },
    {
      "epoch": 3.103961870717903,
      "grad_norm": 10.125,
      "learning_rate": 4.953749965583588e-05,
      "loss": 0.9436,
      "num_input_tokens_seen": 12087080,
      "step": 20840
    },
    {
      "epoch": 3.1047065832588623,
      "grad_norm": 17.875,
      "learning_rate": 4.953687730716363e-05,
      "loss": 1.3954,
      "num_input_tokens_seen": 12090344,
      "step": 20845
    },
    {
      "epoch": 3.1054512957998215,
      "grad_norm": 9.5,
      "learning_rate": 4.9536254543965775e-05,
      "loss": 1.1207,
      "num_input_tokens_seen": 12093416,
      "step": 20850
    },
    {
      "epoch": 3.1061960083407802,
      "grad_norm": 10.5625,
      "learning_rate": 4.953563136625283e-05,
      "loss": 1.0086,
      "num_input_tokens_seen": 12096040,
      "step": 20855
    },
    {
      "epoch": 3.1069407208817394,
      "grad_norm": 9.3125,
      "learning_rate": 4.9535007774035335e-05,
      "loss": 0.8583,
      "num_input_tokens_seen": 12099176,
      "step": 20860
    },
    {
      "epoch": 3.1076854334226987,
      "grad_norm": 10.3125,
      "learning_rate": 4.9534383767323825e-05,
      "loss": 1.0093,
      "num_input_tokens_seen": 12102408,
      "step": 20865
    },
    {
      "epoch": 3.108430145963658,
      "grad_norm": 11.8125,
      "learning_rate": 4.9533759346128824e-05,
      "loss": 1.1618,
      "num_input_tokens_seen": 12105160,
      "step": 20870
    },
    {
      "epoch": 3.109174858504617,
      "grad_norm": 16.0,
      "learning_rate": 4.953313451046091e-05,
      "loss": 1.2983,
      "num_input_tokens_seen": 12108360,
      "step": 20875
    },
    {
      "epoch": 3.1099195710455763,
      "grad_norm": 16.375,
      "learning_rate": 4.9532509260330615e-05,
      "loss": 1.2305,
      "num_input_tokens_seen": 12111528,
      "step": 20880
    },
    {
      "epoch": 3.1106642835865355,
      "grad_norm": 5.25,
      "learning_rate": 4.953188359574851e-05,
      "loss": 1.3222,
      "num_input_tokens_seen": 12114184,
      "step": 20885
    },
    {
      "epoch": 3.1114089961274947,
      "grad_norm": 9.625,
      "learning_rate": 4.953125751672516e-05,
      "loss": 0.9059,
      "num_input_tokens_seen": 12117160,
      "step": 20890
    },
    {
      "epoch": 3.112153708668454,
      "grad_norm": 16.875,
      "learning_rate": 4.953063102327115e-05,
      "loss": 0.9014,
      "num_input_tokens_seen": 12120488,
      "step": 20895
    },
    {
      "epoch": 3.112898421209413,
      "grad_norm": 8.875,
      "learning_rate": 4.953000411539706e-05,
      "loss": 1.3161,
      "num_input_tokens_seen": 12123240,
      "step": 20900
    },
    {
      "epoch": 3.1136431337503723,
      "grad_norm": 15.375,
      "learning_rate": 4.952937679311348e-05,
      "loss": 1.5275,
      "num_input_tokens_seen": 12125960,
      "step": 20905
    },
    {
      "epoch": 3.1143878462913315,
      "grad_norm": 7.59375,
      "learning_rate": 4.9528749056431015e-05,
      "loss": 1.081,
      "num_input_tokens_seen": 12128680,
      "step": 20910
    },
    {
      "epoch": 3.1151325588322907,
      "grad_norm": 10.75,
      "learning_rate": 4.9528120905360265e-05,
      "loss": 1.1131,
      "num_input_tokens_seen": 12131304,
      "step": 20915
    },
    {
      "epoch": 3.11587727137325,
      "grad_norm": 14.0625,
      "learning_rate": 4.9527492339911836e-05,
      "loss": 0.9769,
      "num_input_tokens_seen": 12133992,
      "step": 20920
    },
    {
      "epoch": 3.116621983914209,
      "grad_norm": 6.625,
      "learning_rate": 4.952686336009635e-05,
      "loss": 0.8481,
      "num_input_tokens_seen": 12136616,
      "step": 20925
    },
    {
      "epoch": 3.1173666964551683,
      "grad_norm": 7.125,
      "learning_rate": 4.952623396592445e-05,
      "loss": 0.6317,
      "num_input_tokens_seen": 12139368,
      "step": 20930
    },
    {
      "epoch": 3.1181114089961275,
      "grad_norm": 11.625,
      "learning_rate": 4.952560415740674e-05,
      "loss": 1.0561,
      "num_input_tokens_seen": 12142248,
      "step": 20935
    },
    {
      "epoch": 3.1188561215370867,
      "grad_norm": 8.4375,
      "learning_rate": 4.9524973934553884e-05,
      "loss": 1.2709,
      "num_input_tokens_seen": 12145416,
      "step": 20940
    },
    {
      "epoch": 3.119600834078046,
      "grad_norm": 6.34375,
      "learning_rate": 4.952434329737651e-05,
      "loss": 1.022,
      "num_input_tokens_seen": 12148648,
      "step": 20945
    },
    {
      "epoch": 3.120345546619005,
      "grad_norm": 7.09375,
      "learning_rate": 4.952371224588529e-05,
      "loss": 1.2112,
      "num_input_tokens_seen": 12151464,
      "step": 20950
    },
    {
      "epoch": 3.1210902591599643,
      "grad_norm": 15.0,
      "learning_rate": 4.952308078009087e-05,
      "loss": 1.0568,
      "num_input_tokens_seen": 12154216,
      "step": 20955
    },
    {
      "epoch": 3.1218349717009235,
      "grad_norm": 10.75,
      "learning_rate": 4.9522448900003925e-05,
      "loss": 1.0302,
      "num_input_tokens_seen": 12156936,
      "step": 20960
    },
    {
      "epoch": 3.1225796842418827,
      "grad_norm": 11.5625,
      "learning_rate": 4.952181660563514e-05,
      "loss": 1.1846,
      "num_input_tokens_seen": 12159560,
      "step": 20965
    },
    {
      "epoch": 3.123324396782842,
      "grad_norm": 14.375,
      "learning_rate": 4.952118389699517e-05,
      "loss": 1.3416,
      "num_input_tokens_seen": 12162280,
      "step": 20970
    },
    {
      "epoch": 3.124069109323801,
      "grad_norm": 6.96875,
      "learning_rate": 4.9520550774094735e-05,
      "loss": 1.0417,
      "num_input_tokens_seen": 12165448,
      "step": 20975
    },
    {
      "epoch": 3.1248138218647603,
      "grad_norm": 12.0625,
      "learning_rate": 4.9519917236944504e-05,
      "loss": 0.8634,
      "num_input_tokens_seen": 12168520,
      "step": 20980
    },
    {
      "epoch": 3.1255585344057195,
      "grad_norm": 6.125,
      "learning_rate": 4.9519283285555195e-05,
      "loss": 1.0509,
      "num_input_tokens_seen": 12171656,
      "step": 20985
    },
    {
      "epoch": 3.1263032469466787,
      "grad_norm": 15.5625,
      "learning_rate": 4.951864891993752e-05,
      "loss": 1.1008,
      "num_input_tokens_seen": 12174728,
      "step": 20990
    },
    {
      "epoch": 3.127047959487638,
      "grad_norm": 7.4375,
      "learning_rate": 4.951801414010219e-05,
      "loss": 1.0778,
      "num_input_tokens_seen": 12177640,
      "step": 20995
    },
    {
      "epoch": 3.127792672028597,
      "grad_norm": 6.0,
      "learning_rate": 4.9517378946059936e-05,
      "loss": 1.2502,
      "num_input_tokens_seen": 12180392,
      "step": 21000
    },
    {
      "epoch": 3.1285373845695563,
      "grad_norm": 13.0,
      "learning_rate": 4.951674333782147e-05,
      "loss": 0.8213,
      "num_input_tokens_seen": 12183016,
      "step": 21005
    },
    {
      "epoch": 3.1292820971105155,
      "grad_norm": 8.25,
      "learning_rate": 4.9516107315397554e-05,
      "loss": 1.0611,
      "num_input_tokens_seen": 12185992,
      "step": 21010
    },
    {
      "epoch": 3.1300268096514747,
      "grad_norm": 7.28125,
      "learning_rate": 4.951547087879891e-05,
      "loss": 1.8405,
      "num_input_tokens_seen": 12188776,
      "step": 21015
    },
    {
      "epoch": 3.1307715221924335,
      "grad_norm": 13.0625,
      "learning_rate": 4.951483402803631e-05,
      "loss": 0.9799,
      "num_input_tokens_seen": 12191336,
      "step": 21020
    },
    {
      "epoch": 3.131516234733393,
      "grad_norm": 10.375,
      "learning_rate": 4.95141967631205e-05,
      "loss": 1.2963,
      "num_input_tokens_seen": 12194312,
      "step": 21025
    },
    {
      "epoch": 3.132260947274352,
      "grad_norm": 10.375,
      "learning_rate": 4.951355908406226e-05,
      "loss": 1.1141,
      "num_input_tokens_seen": 12197288,
      "step": 21030
    },
    {
      "epoch": 3.133005659815311,
      "grad_norm": 7.53125,
      "learning_rate": 4.951292099087235e-05,
      "loss": 1.1594,
      "num_input_tokens_seen": 12200072,
      "step": 21035
    },
    {
      "epoch": 3.1337503723562703,
      "grad_norm": 12.25,
      "learning_rate": 4.951228248356155e-05,
      "loss": 1.1443,
      "num_input_tokens_seen": 12202856,
      "step": 21040
    },
    {
      "epoch": 3.1344950848972295,
      "grad_norm": 10.0,
      "learning_rate": 4.951164356214065e-05,
      "loss": 1.1215,
      "num_input_tokens_seen": 12205864,
      "step": 21045
    },
    {
      "epoch": 3.1352397974381887,
      "grad_norm": 10.5,
      "learning_rate": 4.951100422662045e-05,
      "loss": 0.9858,
      "num_input_tokens_seen": 12209064,
      "step": 21050
    },
    {
      "epoch": 3.135984509979148,
      "grad_norm": 7.46875,
      "learning_rate": 4.951036447701174e-05,
      "loss": 1.027,
      "num_input_tokens_seen": 12211752,
      "step": 21055
    },
    {
      "epoch": 3.136729222520107,
      "grad_norm": 14.4375,
      "learning_rate": 4.950972431332534e-05,
      "loss": 0.9652,
      "num_input_tokens_seen": 12214632,
      "step": 21060
    },
    {
      "epoch": 3.1374739350610663,
      "grad_norm": 12.625,
      "learning_rate": 4.9509083735572055e-05,
      "loss": 1.161,
      "num_input_tokens_seen": 12217224,
      "step": 21065
    },
    {
      "epoch": 3.1382186476020255,
      "grad_norm": 11.875,
      "learning_rate": 4.950844274376271e-05,
      "loss": 1.0009,
      "num_input_tokens_seen": 12220424,
      "step": 21070
    },
    {
      "epoch": 3.1389633601429847,
      "grad_norm": 16.625,
      "learning_rate": 4.950780133790813e-05,
      "loss": 0.9358,
      "num_input_tokens_seen": 12223560,
      "step": 21075
    },
    {
      "epoch": 3.139708072683944,
      "grad_norm": 14.375,
      "learning_rate": 4.950715951801916e-05,
      "loss": 0.9998,
      "num_input_tokens_seen": 12226440,
      "step": 21080
    },
    {
      "epoch": 3.140452785224903,
      "grad_norm": 10.75,
      "learning_rate": 4.950651728410663e-05,
      "loss": 1.3718,
      "num_input_tokens_seen": 12229256,
      "step": 21085
    },
    {
      "epoch": 3.1411974977658623,
      "grad_norm": 8.3125,
      "learning_rate": 4.9505874636181414e-05,
      "loss": 1.1286,
      "num_input_tokens_seen": 12232104,
      "step": 21090
    },
    {
      "epoch": 3.1419422103068215,
      "grad_norm": 12.75,
      "learning_rate": 4.950523157425434e-05,
      "loss": 1.4912,
      "num_input_tokens_seen": 12234984,
      "step": 21095
    },
    {
      "epoch": 3.1426869228477807,
      "grad_norm": 10.25,
      "learning_rate": 4.950458809833629e-05,
      "loss": 0.9238,
      "num_input_tokens_seen": 12237832,
      "step": 21100
    },
    {
      "epoch": 3.14343163538874,
      "grad_norm": 10.75,
      "learning_rate": 4.9503944208438124e-05,
      "loss": 0.7698,
      "num_input_tokens_seen": 12240776,
      "step": 21105
    },
    {
      "epoch": 3.144176347929699,
      "grad_norm": 5.78125,
      "learning_rate": 4.950329990457073e-05,
      "loss": 0.5299,
      "num_input_tokens_seen": 12243720,
      "step": 21110
    },
    {
      "epoch": 3.1449210604706583,
      "grad_norm": 18.75,
      "learning_rate": 4.950265518674498e-05,
      "loss": 1.3015,
      "num_input_tokens_seen": 12246888,
      "step": 21115
    },
    {
      "epoch": 3.1456657730116175,
      "grad_norm": 8.75,
      "learning_rate": 4.950201005497179e-05,
      "loss": 1.0589,
      "num_input_tokens_seen": 12249640,
      "step": 21120
    },
    {
      "epoch": 3.1464104855525767,
      "grad_norm": 10.4375,
      "learning_rate": 4.950136450926203e-05,
      "loss": 1.2089,
      "num_input_tokens_seen": 12252712,
      "step": 21125
    },
    {
      "epoch": 3.147155198093536,
      "grad_norm": 6.25,
      "learning_rate": 4.950071854962662e-05,
      "loss": 0.994,
      "num_input_tokens_seen": 12255464,
      "step": 21130
    },
    {
      "epoch": 3.147899910634495,
      "grad_norm": 10.4375,
      "learning_rate": 4.950007217607647e-05,
      "loss": 0.9199,
      "num_input_tokens_seen": 12258280,
      "step": 21135
    },
    {
      "epoch": 3.1486446231754543,
      "grad_norm": 8.875,
      "learning_rate": 4.949942538862251e-05,
      "loss": 0.8512,
      "num_input_tokens_seen": 12261128,
      "step": 21140
    },
    {
      "epoch": 3.1493893357164136,
      "grad_norm": 15.8125,
      "learning_rate": 4.949877818727565e-05,
      "loss": 1.2727,
      "num_input_tokens_seen": 12264104,
      "step": 21145
    },
    {
      "epoch": 3.1501340482573728,
      "grad_norm": 20.375,
      "learning_rate": 4.949813057204684e-05,
      "loss": 1.157,
      "num_input_tokens_seen": 12266760,
      "step": 21150
    },
    {
      "epoch": 3.150878760798332,
      "grad_norm": 10.375,
      "learning_rate": 4.9497482542947004e-05,
      "loss": 1.0595,
      "num_input_tokens_seen": 12269448,
      "step": 21155
    },
    {
      "epoch": 3.151623473339291,
      "grad_norm": 13.0,
      "learning_rate": 4.9496834099987106e-05,
      "loss": 1.1422,
      "num_input_tokens_seen": 12272264,
      "step": 21160
    },
    {
      "epoch": 3.1523681858802504,
      "grad_norm": 8.75,
      "learning_rate": 4.949618524317809e-05,
      "loss": 0.7743,
      "num_input_tokens_seen": 12275016,
      "step": 21165
    },
    {
      "epoch": 3.1531128984212096,
      "grad_norm": 9.25,
      "learning_rate": 4.9495535972530924e-05,
      "loss": 0.6867,
      "num_input_tokens_seen": 12278088,
      "step": 21170
    },
    {
      "epoch": 3.1538576109621688,
      "grad_norm": 12.25,
      "learning_rate": 4.949488628805657e-05,
      "loss": 1.0301,
      "num_input_tokens_seen": 12280808,
      "step": 21175
    },
    {
      "epoch": 3.154602323503128,
      "grad_norm": 8.75,
      "learning_rate": 4.9494236189766005e-05,
      "loss": 1.275,
      "num_input_tokens_seen": 12283976,
      "step": 21180
    },
    {
      "epoch": 3.155347036044087,
      "grad_norm": 8.625,
      "learning_rate": 4.9493585677670216e-05,
      "loss": 0.9089,
      "num_input_tokens_seen": 12287112,
      "step": 21185
    },
    {
      "epoch": 3.1560917485850464,
      "grad_norm": 6.21875,
      "learning_rate": 4.94929347517802e-05,
      "loss": 0.7298,
      "num_input_tokens_seen": 12290088,
      "step": 21190
    },
    {
      "epoch": 3.156836461126005,
      "grad_norm": 8.875,
      "learning_rate": 4.9492283412106934e-05,
      "loss": 1.0272,
      "num_input_tokens_seen": 12292904,
      "step": 21195
    },
    {
      "epoch": 3.157581173666965,
      "grad_norm": 5.4375,
      "learning_rate": 4.9491631658661436e-05,
      "loss": 0.8951,
      "num_input_tokens_seen": 12295560,
      "step": 21200
    },
    {
      "epoch": 3.1583258862079235,
      "grad_norm": 16.75,
      "learning_rate": 4.9490979491454716e-05,
      "loss": 1.4068,
      "num_input_tokens_seen": 12298728,
      "step": 21205
    },
    {
      "epoch": 3.1590705987488827,
      "grad_norm": 10.4375,
      "learning_rate": 4.9490326910497786e-05,
      "loss": 1.3593,
      "num_input_tokens_seen": 12301640,
      "step": 21210
    },
    {
      "epoch": 3.159815311289842,
      "grad_norm": 11.75,
      "learning_rate": 4.948967391580167e-05,
      "loss": 0.9999,
      "num_input_tokens_seen": 12304488,
      "step": 21215
    },
    {
      "epoch": 3.160560023830801,
      "grad_norm": 16.75,
      "learning_rate": 4.948902050737741e-05,
      "loss": 0.9842,
      "num_input_tokens_seen": 12307432,
      "step": 21220
    },
    {
      "epoch": 3.1613047363717603,
      "grad_norm": 9.3125,
      "learning_rate": 4.948836668523604e-05,
      "loss": 0.7834,
      "num_input_tokens_seen": 12310408,
      "step": 21225
    },
    {
      "epoch": 3.1620494489127196,
      "grad_norm": 9.875,
      "learning_rate": 4.9487712449388604e-05,
      "loss": 1.2592,
      "num_input_tokens_seen": 12313448,
      "step": 21230
    },
    {
      "epoch": 3.1627941614536788,
      "grad_norm": 8.5625,
      "learning_rate": 4.948705779984614e-05,
      "loss": 1.1104,
      "num_input_tokens_seen": 12316264,
      "step": 21235
    },
    {
      "epoch": 3.163538873994638,
      "grad_norm": 7.625,
      "learning_rate": 4.9486402736619736e-05,
      "loss": 1.2313,
      "num_input_tokens_seen": 12319112,
      "step": 21240
    },
    {
      "epoch": 3.164283586535597,
      "grad_norm": 10.25,
      "learning_rate": 4.9485747259720435e-05,
      "loss": 0.9054,
      "num_input_tokens_seen": 12322152,
      "step": 21245
    },
    {
      "epoch": 3.1650282990765564,
      "grad_norm": 12.5,
      "learning_rate": 4.9485091369159334e-05,
      "loss": 1.1355,
      "num_input_tokens_seen": 12325032,
      "step": 21250
    },
    {
      "epoch": 3.1657730116175156,
      "grad_norm": 7.15625,
      "learning_rate": 4.948443506494749e-05,
      "loss": 1.442,
      "num_input_tokens_seen": 12328040,
      "step": 21255
    },
    {
      "epoch": 3.1665177241584748,
      "grad_norm": 16.375,
      "learning_rate": 4.9483778347096e-05,
      "loss": 0.843,
      "num_input_tokens_seen": 12330920,
      "step": 21260
    },
    {
      "epoch": 3.167262436699434,
      "grad_norm": 9.0625,
      "learning_rate": 4.948312121561596e-05,
      "loss": 1.3942,
      "num_input_tokens_seen": 12333832,
      "step": 21265
    },
    {
      "epoch": 3.168007149240393,
      "grad_norm": 7.96875,
      "learning_rate": 4.9482463670518476e-05,
      "loss": 0.9554,
      "num_input_tokens_seen": 12336872,
      "step": 21270
    },
    {
      "epoch": 3.1687518617813524,
      "grad_norm": 8.375,
      "learning_rate": 4.9481805711814645e-05,
      "loss": 1.088,
      "num_input_tokens_seen": 12339848,
      "step": 21275
    },
    {
      "epoch": 3.1694965743223116,
      "grad_norm": 12.0,
      "learning_rate": 4.948114733951559e-05,
      "loss": 0.9123,
      "num_input_tokens_seen": 12342824,
      "step": 21280
    },
    {
      "epoch": 3.170241286863271,
      "grad_norm": 13.375,
      "learning_rate": 4.948048855363243e-05,
      "loss": 0.9263,
      "num_input_tokens_seen": 12345800,
      "step": 21285
    },
    {
      "epoch": 3.17098599940423,
      "grad_norm": 15.1875,
      "learning_rate": 4.94798293541763e-05,
      "loss": 1.1922,
      "num_input_tokens_seen": 12349128,
      "step": 21290
    },
    {
      "epoch": 3.171730711945189,
      "grad_norm": 11.8125,
      "learning_rate": 4.9479169741158336e-05,
      "loss": 1.0529,
      "num_input_tokens_seen": 12352232,
      "step": 21295
    },
    {
      "epoch": 3.1724754244861484,
      "grad_norm": 8.9375,
      "learning_rate": 4.947850971458968e-05,
      "loss": 1.1641,
      "num_input_tokens_seen": 12355368,
      "step": 21300
    },
    {
      "epoch": 3.1732201370271076,
      "grad_norm": 7.15625,
      "learning_rate": 4.947784927448147e-05,
      "loss": 1.0608,
      "num_input_tokens_seen": 12357992,
      "step": 21305
    },
    {
      "epoch": 3.173964849568067,
      "grad_norm": 14.4375,
      "learning_rate": 4.9477188420844886e-05,
      "loss": 1.1342,
      "num_input_tokens_seen": 12360616,
      "step": 21310
    },
    {
      "epoch": 3.174709562109026,
      "grad_norm": 12.4375,
      "learning_rate": 4.947652715369108e-05,
      "loss": 1.2647,
      "num_input_tokens_seen": 12363432,
      "step": 21315
    },
    {
      "epoch": 3.175454274649985,
      "grad_norm": 9.75,
      "learning_rate": 4.947586547303121e-05,
      "loss": 0.8343,
      "num_input_tokens_seen": 12366248,
      "step": 21320
    },
    {
      "epoch": 3.1761989871909444,
      "grad_norm": 5.5625,
      "learning_rate": 4.947520337887649e-05,
      "loss": 0.847,
      "num_input_tokens_seen": 12369064,
      "step": 21325
    },
    {
      "epoch": 3.1769436997319036,
      "grad_norm": 10.9375,
      "learning_rate": 4.947454087123807e-05,
      "loss": 1.1935,
      "num_input_tokens_seen": 12372168,
      "step": 21330
    },
    {
      "epoch": 3.177688412272863,
      "grad_norm": 9.5,
      "learning_rate": 4.947387795012716e-05,
      "loss": 0.9886,
      "num_input_tokens_seen": 12375176,
      "step": 21335
    },
    {
      "epoch": 3.178433124813822,
      "grad_norm": 5.03125,
      "learning_rate": 4.947321461555496e-05,
      "loss": 0.7512,
      "num_input_tokens_seen": 12378088,
      "step": 21340
    },
    {
      "epoch": 3.179177837354781,
      "grad_norm": 10.25,
      "learning_rate": 4.947255086753268e-05,
      "loss": 1.0044,
      "num_input_tokens_seen": 12381064,
      "step": 21345
    },
    {
      "epoch": 3.1799225498957404,
      "grad_norm": 14.0,
      "learning_rate": 4.9471886706071504e-05,
      "loss": 1.4652,
      "num_input_tokens_seen": 12384104,
      "step": 21350
    },
    {
      "epoch": 3.1806672624366996,
      "grad_norm": 12.5625,
      "learning_rate": 4.9471222131182685e-05,
      "loss": 1.3543,
      "num_input_tokens_seen": 12386792,
      "step": 21355
    },
    {
      "epoch": 3.181411974977659,
      "grad_norm": 5.53125,
      "learning_rate": 4.9470557142877446e-05,
      "loss": 0.9358,
      "num_input_tokens_seen": 12389864,
      "step": 21360
    },
    {
      "epoch": 3.182156687518618,
      "grad_norm": 9.4375,
      "learning_rate": 4.946989174116701e-05,
      "loss": 1.1173,
      "num_input_tokens_seen": 12392776,
      "step": 21365
    },
    {
      "epoch": 3.182901400059577,
      "grad_norm": 4.84375,
      "learning_rate": 4.9469225926062625e-05,
      "loss": 1.0048,
      "num_input_tokens_seen": 12395752,
      "step": 21370
    },
    {
      "epoch": 3.1836461126005364,
      "grad_norm": 11.8125,
      "learning_rate": 4.946855969757553e-05,
      "loss": 0.9502,
      "num_input_tokens_seen": 12398280,
      "step": 21375
    },
    {
      "epoch": 3.184390825141495,
      "grad_norm": 13.375,
      "learning_rate": 4.9467893055716996e-05,
      "loss": 1.2103,
      "num_input_tokens_seen": 12401448,
      "step": 21380
    },
    {
      "epoch": 3.1851355376824544,
      "grad_norm": 6.34375,
      "learning_rate": 4.946722600049827e-05,
      "loss": 1.1467,
      "num_input_tokens_seen": 12404264,
      "step": 21385
    },
    {
      "epoch": 3.1858802502234136,
      "grad_norm": 17.375,
      "learning_rate": 4.946655853193063e-05,
      "loss": 1.3284,
      "num_input_tokens_seen": 12407112,
      "step": 21390
    },
    {
      "epoch": 3.186624962764373,
      "grad_norm": 9.0,
      "learning_rate": 4.946589065002535e-05,
      "loss": 0.89,
      "num_input_tokens_seen": 12409864,
      "step": 21395
    },
    {
      "epoch": 3.187369675305332,
      "grad_norm": 8.0625,
      "learning_rate": 4.946522235479372e-05,
      "loss": 1.1667,
      "num_input_tokens_seen": 12412648,
      "step": 21400
    },
    {
      "epoch": 3.188114387846291,
      "grad_norm": 8.625,
      "learning_rate": 4.946455364624702e-05,
      "loss": 0.9584,
      "num_input_tokens_seen": 12415528,
      "step": 21405
    },
    {
      "epoch": 3.1888591003872504,
      "grad_norm": 7.40625,
      "learning_rate": 4.9463884524396555e-05,
      "loss": 1.071,
      "num_input_tokens_seen": 12418504,
      "step": 21410
    },
    {
      "epoch": 3.1896038129282096,
      "grad_norm": 22.875,
      "learning_rate": 4.946321498925362e-05,
      "loss": 1.0495,
      "num_input_tokens_seen": 12421192,
      "step": 21415
    },
    {
      "epoch": 3.190348525469169,
      "grad_norm": 6.34375,
      "learning_rate": 4.946254504082952e-05,
      "loss": 0.9432,
      "num_input_tokens_seen": 12424104,
      "step": 21420
    },
    {
      "epoch": 3.191093238010128,
      "grad_norm": 5.875,
      "learning_rate": 4.94618746791356e-05,
      "loss": 1.0843,
      "num_input_tokens_seen": 12426760,
      "step": 21425
    },
    {
      "epoch": 3.191837950551087,
      "grad_norm": 10.6875,
      "learning_rate": 4.946120390418316e-05,
      "loss": 0.7242,
      "num_input_tokens_seen": 12429544,
      "step": 21430
    },
    {
      "epoch": 3.1925826630920464,
      "grad_norm": 17.0,
      "learning_rate": 4.946053271598355e-05,
      "loss": 0.9443,
      "num_input_tokens_seen": 12432552,
      "step": 21435
    },
    {
      "epoch": 3.1933273756330056,
      "grad_norm": 10.75,
      "learning_rate": 4.94598611145481e-05,
      "loss": 1.0917,
      "num_input_tokens_seen": 12435272,
      "step": 21440
    },
    {
      "epoch": 3.194072088173965,
      "grad_norm": 7.5,
      "learning_rate": 4.945918909988815e-05,
      "loss": 1.1563,
      "num_input_tokens_seen": 12437992,
      "step": 21445
    },
    {
      "epoch": 3.194816800714924,
      "grad_norm": 9.25,
      "learning_rate": 4.945851667201507e-05,
      "loss": 0.946,
      "num_input_tokens_seen": 12440840,
      "step": 21450
    },
    {
      "epoch": 3.1955615132558832,
      "grad_norm": 10.875,
      "learning_rate": 4.945784383094019e-05,
      "loss": 0.6106,
      "num_input_tokens_seen": 12443880,
      "step": 21455
    },
    {
      "epoch": 3.1963062257968424,
      "grad_norm": 16.625,
      "learning_rate": 4.9457170576674914e-05,
      "loss": 1.1092,
      "num_input_tokens_seen": 12446888,
      "step": 21460
    },
    {
      "epoch": 3.1970509383378016,
      "grad_norm": 7.34375,
      "learning_rate": 4.945649690923059e-05,
      "loss": 0.8897,
      "num_input_tokens_seen": 12449544,
      "step": 21465
    },
    {
      "epoch": 3.197795650878761,
      "grad_norm": 9.8125,
      "learning_rate": 4.94558228286186e-05,
      "loss": 0.9752,
      "num_input_tokens_seen": 12452552,
      "step": 21470
    },
    {
      "epoch": 3.19854036341972,
      "grad_norm": 7.6875,
      "learning_rate": 4.945514833485036e-05,
      "loss": 1.1031,
      "num_input_tokens_seen": 12455336,
      "step": 21475
    },
    {
      "epoch": 3.1992850759606792,
      "grad_norm": 9.1875,
      "learning_rate": 4.9454473427937225e-05,
      "loss": 1.1184,
      "num_input_tokens_seen": 12458376,
      "step": 21480
    },
    {
      "epoch": 3.2000297885016384,
      "grad_norm": 10.1875,
      "learning_rate": 4.9453798107890624e-05,
      "loss": 0.8914,
      "num_input_tokens_seen": 12461512,
      "step": 21485
    },
    {
      "epoch": 3.2007745010425976,
      "grad_norm": 9.625,
      "learning_rate": 4.945312237472196e-05,
      "loss": 0.8667,
      "num_input_tokens_seen": 12464456,
      "step": 21490
    },
    {
      "epoch": 3.201519213583557,
      "grad_norm": 6.96875,
      "learning_rate": 4.945244622844264e-05,
      "loss": 0.8749,
      "num_input_tokens_seen": 12467048,
      "step": 21495
    },
    {
      "epoch": 3.202263926124516,
      "grad_norm": 5.9375,
      "learning_rate": 4.9451769669064096e-05,
      "loss": 0.7307,
      "num_input_tokens_seen": 12469960,
      "step": 21500
    },
    {
      "epoch": 3.2030086386654752,
      "grad_norm": 8.375,
      "learning_rate": 4.945109269659776e-05,
      "loss": 0.9523,
      "num_input_tokens_seen": 12472808,
      "step": 21505
    },
    {
      "epoch": 3.2037533512064345,
      "grad_norm": 9.9375,
      "learning_rate": 4.945041531105505e-05,
      "loss": 1.1495,
      "num_input_tokens_seen": 12475880,
      "step": 21510
    },
    {
      "epoch": 3.2044980637473937,
      "grad_norm": 13.375,
      "learning_rate": 4.9449737512447435e-05,
      "loss": 1.3432,
      "num_input_tokens_seen": 12478696,
      "step": 21515
    },
    {
      "epoch": 3.205242776288353,
      "grad_norm": 6.5625,
      "learning_rate": 4.9449059300786355e-05,
      "loss": 0.8777,
      "num_input_tokens_seen": 12481672,
      "step": 21520
    },
    {
      "epoch": 3.205987488829312,
      "grad_norm": 5.40625,
      "learning_rate": 4.944838067608326e-05,
      "loss": 1.0545,
      "num_input_tokens_seen": 12484584,
      "step": 21525
    },
    {
      "epoch": 3.2067322013702713,
      "grad_norm": 10.375,
      "learning_rate": 4.944770163834963e-05,
      "loss": 1.0657,
      "num_input_tokens_seen": 12487080,
      "step": 21530
    },
    {
      "epoch": 3.2074769139112305,
      "grad_norm": 16.0,
      "learning_rate": 4.944702218759692e-05,
      "loss": 1.1607,
      "num_input_tokens_seen": 12489864,
      "step": 21535
    },
    {
      "epoch": 3.2082216264521897,
      "grad_norm": 14.4375,
      "learning_rate": 4.944634232383662e-05,
      "loss": 1.1024,
      "num_input_tokens_seen": 12493192,
      "step": 21540
    },
    {
      "epoch": 3.2089663389931484,
      "grad_norm": 10.0625,
      "learning_rate": 4.944566204708022e-05,
      "loss": 1.3703,
      "num_input_tokens_seen": 12495816,
      "step": 21545
    },
    {
      "epoch": 3.2097110515341076,
      "grad_norm": 20.125,
      "learning_rate": 4.94449813573392e-05,
      "loss": 1.0036,
      "num_input_tokens_seen": 12498824,
      "step": 21550
    },
    {
      "epoch": 3.210455764075067,
      "grad_norm": 15.1875,
      "learning_rate": 4.944430025462507e-05,
      "loss": 0.9295,
      "num_input_tokens_seen": 12501576,
      "step": 21555
    },
    {
      "epoch": 3.211200476616026,
      "grad_norm": 5.84375,
      "learning_rate": 4.944361873894932e-05,
      "loss": 0.6704,
      "num_input_tokens_seen": 12504488,
      "step": 21560
    },
    {
      "epoch": 3.2119451891569852,
      "grad_norm": 7.96875,
      "learning_rate": 4.944293681032348e-05,
      "loss": 1.1675,
      "num_input_tokens_seen": 12507272,
      "step": 21565
    },
    {
      "epoch": 3.2126899016979444,
      "grad_norm": 9.375,
      "learning_rate": 4.9442254468759065e-05,
      "loss": 1.0315,
      "num_input_tokens_seen": 12509928,
      "step": 21570
    },
    {
      "epoch": 3.2134346142389036,
      "grad_norm": 8.6875,
      "learning_rate": 4.94415717142676e-05,
      "loss": 1.3533,
      "num_input_tokens_seen": 12512712,
      "step": 21575
    },
    {
      "epoch": 3.214179326779863,
      "grad_norm": 17.625,
      "learning_rate": 4.944088854686062e-05,
      "loss": 1.0637,
      "num_input_tokens_seen": 12515656,
      "step": 21580
    },
    {
      "epoch": 3.214924039320822,
      "grad_norm": 10.0625,
      "learning_rate": 4.944020496654968e-05,
      "loss": 1.0889,
      "num_input_tokens_seen": 12518632,
      "step": 21585
    },
    {
      "epoch": 3.2156687518617812,
      "grad_norm": 9.6875,
      "learning_rate": 4.943952097334631e-05,
      "loss": 0.8833,
      "num_input_tokens_seen": 12521640,
      "step": 21590
    },
    {
      "epoch": 3.2164134644027405,
      "grad_norm": 8.5625,
      "learning_rate": 4.943883656726207e-05,
      "loss": 1.0472,
      "num_input_tokens_seen": 12524520,
      "step": 21595
    },
    {
      "epoch": 3.2171581769436997,
      "grad_norm": 7.75,
      "learning_rate": 4.943815174830853e-05,
      "loss": 1.0263,
      "num_input_tokens_seen": 12527272,
      "step": 21600
    },
    {
      "epoch": 3.217902889484659,
      "grad_norm": 14.3125,
      "learning_rate": 4.9437466516497255e-05,
      "loss": 1.136,
      "num_input_tokens_seen": 12530248,
      "step": 21605
    },
    {
      "epoch": 3.218647602025618,
      "grad_norm": 10.875,
      "learning_rate": 4.943678087183982e-05,
      "loss": 1.0245,
      "num_input_tokens_seen": 12532968,
      "step": 21610
    },
    {
      "epoch": 3.2193923145665773,
      "grad_norm": 12.3125,
      "learning_rate": 4.94360948143478e-05,
      "loss": 0.7976,
      "num_input_tokens_seen": 12535624,
      "step": 21615
    },
    {
      "epoch": 3.2201370271075365,
      "grad_norm": 11.3125,
      "learning_rate": 4.94354083440328e-05,
      "loss": 1.3831,
      "num_input_tokens_seen": 12538632,
      "step": 21620
    },
    {
      "epoch": 3.2208817396484957,
      "grad_norm": 7.46875,
      "learning_rate": 4.9434721460906406e-05,
      "loss": 0.8733,
      "num_input_tokens_seen": 12541576,
      "step": 21625
    },
    {
      "epoch": 3.221626452189455,
      "grad_norm": 10.9375,
      "learning_rate": 4.9434034164980233e-05,
      "loss": 0.9303,
      "num_input_tokens_seen": 12544584,
      "step": 21630
    },
    {
      "epoch": 3.222371164730414,
      "grad_norm": 8.5625,
      "learning_rate": 4.94333464562659e-05,
      "loss": 0.8129,
      "num_input_tokens_seen": 12547400,
      "step": 21635
    },
    {
      "epoch": 3.2231158772713733,
      "grad_norm": 7.71875,
      "learning_rate": 4.9432658334774984e-05,
      "loss": 1.0469,
      "num_input_tokens_seen": 12550152,
      "step": 21640
    },
    {
      "epoch": 3.2238605898123325,
      "grad_norm": 11.6875,
      "learning_rate": 4.943196980051915e-05,
      "loss": 0.9751,
      "num_input_tokens_seen": 12553128,
      "step": 21645
    },
    {
      "epoch": 3.2246053023532917,
      "grad_norm": 5.53125,
      "learning_rate": 4.943128085351002e-05,
      "loss": 1.0558,
      "num_input_tokens_seen": 12556168,
      "step": 21650
    },
    {
      "epoch": 3.225350014894251,
      "grad_norm": 6.46875,
      "learning_rate": 4.943059149375923e-05,
      "loss": 1.1582,
      "num_input_tokens_seen": 12559208,
      "step": 21655
    },
    {
      "epoch": 3.22609472743521,
      "grad_norm": 6.3125,
      "learning_rate": 4.9429901721278426e-05,
      "loss": 0.8908,
      "num_input_tokens_seen": 12562440,
      "step": 21660
    },
    {
      "epoch": 3.2268394399761693,
      "grad_norm": 14.375,
      "learning_rate": 4.9429211536079266e-05,
      "loss": 1.1612,
      "num_input_tokens_seen": 12565608,
      "step": 21665
    },
    {
      "epoch": 3.2275841525171285,
      "grad_norm": 13.375,
      "learning_rate": 4.94285209381734e-05,
      "loss": 0.9516,
      "num_input_tokens_seen": 12568584,
      "step": 21670
    },
    {
      "epoch": 3.2283288650580877,
      "grad_norm": 21.5,
      "learning_rate": 4.94278299275725e-05,
      "loss": 0.991,
      "num_input_tokens_seen": 12571464,
      "step": 21675
    },
    {
      "epoch": 3.229073577599047,
      "grad_norm": 16.75,
      "learning_rate": 4.9427138504288245e-05,
      "loss": 1.1621,
      "num_input_tokens_seen": 12574696,
      "step": 21680
    },
    {
      "epoch": 3.229818290140006,
      "grad_norm": 6.84375,
      "learning_rate": 4.942644666833231e-05,
      "loss": 1.1168,
      "num_input_tokens_seen": 12577608,
      "step": 21685
    },
    {
      "epoch": 3.2305630026809653,
      "grad_norm": 7.125,
      "learning_rate": 4.9425754419716383e-05,
      "loss": 0.9074,
      "num_input_tokens_seen": 12580680,
      "step": 21690
    },
    {
      "epoch": 3.2313077152219245,
      "grad_norm": 9.5,
      "learning_rate": 4.942506175845216e-05,
      "loss": 1.1881,
      "num_input_tokens_seen": 12583464,
      "step": 21695
    },
    {
      "epoch": 3.2320524277628837,
      "grad_norm": 5.9375,
      "learning_rate": 4.9424368684551347e-05,
      "loss": 0.8427,
      "num_input_tokens_seen": 12586408,
      "step": 21700
    },
    {
      "epoch": 3.232797140303843,
      "grad_norm": 7.78125,
      "learning_rate": 4.942367519802565e-05,
      "loss": 1.1408,
      "num_input_tokens_seen": 12589224,
      "step": 21705
    },
    {
      "epoch": 3.233541852844802,
      "grad_norm": 15.5625,
      "learning_rate": 4.9422981298886776e-05,
      "loss": 1.1846,
      "num_input_tokens_seen": 12592232,
      "step": 21710
    },
    {
      "epoch": 3.2342865653857613,
      "grad_norm": 12.0625,
      "learning_rate": 4.942228698714646e-05,
      "loss": 0.8433,
      "num_input_tokens_seen": 12594888,
      "step": 21715
    },
    {
      "epoch": 3.23503127792672,
      "grad_norm": 9.5625,
      "learning_rate": 4.942159226281643e-05,
      "loss": 1.2318,
      "num_input_tokens_seen": 12597800,
      "step": 21720
    },
    {
      "epoch": 3.2357759904676793,
      "grad_norm": 11.9375,
      "learning_rate": 4.942089712590842e-05,
      "loss": 1.1371,
      "num_input_tokens_seen": 12600936,
      "step": 21725
    },
    {
      "epoch": 3.2365207030086385,
      "grad_norm": 4.9375,
      "learning_rate": 4.9420201576434165e-05,
      "loss": 1.0067,
      "num_input_tokens_seen": 12603912,
      "step": 21730
    },
    {
      "epoch": 3.2372654155495977,
      "grad_norm": 7.03125,
      "learning_rate": 4.941950561440543e-05,
      "loss": 0.9506,
      "num_input_tokens_seen": 12606984,
      "step": 21735
    },
    {
      "epoch": 3.238010128090557,
      "grad_norm": 12.3125,
      "learning_rate": 4.9418809239833964e-05,
      "loss": 1.3378,
      "num_input_tokens_seen": 12610024,
      "step": 21740
    },
    {
      "epoch": 3.238754840631516,
      "grad_norm": 8.4375,
      "learning_rate": 4.9418112452731534e-05,
      "loss": 1.0182,
      "num_input_tokens_seen": 12612840,
      "step": 21745
    },
    {
      "epoch": 3.2394995531724753,
      "grad_norm": 7.9375,
      "learning_rate": 4.941741525310991e-05,
      "loss": 1.0557,
      "num_input_tokens_seen": 12615592,
      "step": 21750
    },
    {
      "epoch": 3.2402442657134345,
      "grad_norm": 5.75,
      "learning_rate": 4.9416717640980884e-05,
      "loss": 0.9673,
      "num_input_tokens_seen": 12618696,
      "step": 21755
    },
    {
      "epoch": 3.2409889782543937,
      "grad_norm": 9.125,
      "learning_rate": 4.941601961635621e-05,
      "loss": 1.0601,
      "num_input_tokens_seen": 12621448,
      "step": 21760
    },
    {
      "epoch": 3.241733690795353,
      "grad_norm": 12.0625,
      "learning_rate": 4.941532117924772e-05,
      "loss": 1.1779,
      "num_input_tokens_seen": 12624168,
      "step": 21765
    },
    {
      "epoch": 3.242478403336312,
      "grad_norm": 7.6875,
      "learning_rate": 4.941462232966718e-05,
      "loss": 1.232,
      "num_input_tokens_seen": 12626952,
      "step": 21770
    },
    {
      "epoch": 3.2432231158772713,
      "grad_norm": 8.875,
      "learning_rate": 4.9413923067626413e-05,
      "loss": 1.1342,
      "num_input_tokens_seen": 12630056,
      "step": 21775
    },
    {
      "epoch": 3.2439678284182305,
      "grad_norm": 10.4375,
      "learning_rate": 4.941322339313723e-05,
      "loss": 0.8357,
      "num_input_tokens_seen": 12633512,
      "step": 21780
    },
    {
      "epoch": 3.2447125409591897,
      "grad_norm": 16.625,
      "learning_rate": 4.941252330621145e-05,
      "loss": 1.3474,
      "num_input_tokens_seen": 12636520,
      "step": 21785
    },
    {
      "epoch": 3.245457253500149,
      "grad_norm": 6.0625,
      "learning_rate": 4.94118228068609e-05,
      "loss": 1.1672,
      "num_input_tokens_seen": 12639528,
      "step": 21790
    },
    {
      "epoch": 3.246201966041108,
      "grad_norm": 6.4375,
      "learning_rate": 4.9411121895097414e-05,
      "loss": 1.0725,
      "num_input_tokens_seen": 12642408,
      "step": 21795
    },
    {
      "epoch": 3.2469466785820673,
      "grad_norm": 7.9375,
      "learning_rate": 4.941042057093284e-05,
      "loss": 0.8017,
      "num_input_tokens_seen": 12645352,
      "step": 21800
    },
    {
      "epoch": 3.2476913911230265,
      "grad_norm": 7.96875,
      "learning_rate": 4.940971883437901e-05,
      "loss": 0.7418,
      "num_input_tokens_seen": 12648392,
      "step": 21805
    },
    {
      "epoch": 3.2484361036639857,
      "grad_norm": 7.96875,
      "learning_rate": 4.94090166854478e-05,
      "loss": 1.3367,
      "num_input_tokens_seen": 12651432,
      "step": 21810
    },
    {
      "epoch": 3.249180816204945,
      "grad_norm": 5.1875,
      "learning_rate": 4.940831412415105e-05,
      "loss": 1.1129,
      "num_input_tokens_seen": 12654184,
      "step": 21815
    },
    {
      "epoch": 3.249925528745904,
      "grad_norm": 10.5,
      "learning_rate": 4.9407611150500646e-05,
      "loss": 1.3675,
      "num_input_tokens_seen": 12656872,
      "step": 21820
    },
    {
      "epoch": 3.2506702412868633,
      "grad_norm": 8.1875,
      "learning_rate": 4.940690776450846e-05,
      "loss": 1.0907,
      "num_input_tokens_seen": 12659912,
      "step": 21825
    },
    {
      "epoch": 3.2514149538278225,
      "grad_norm": 14.125,
      "learning_rate": 4.940620396618637e-05,
      "loss": 1.387,
      "num_input_tokens_seen": 12662600,
      "step": 21830
    },
    {
      "epoch": 3.2521596663687817,
      "grad_norm": 12.4375,
      "learning_rate": 4.940549975554627e-05,
      "loss": 1.0065,
      "num_input_tokens_seen": 12665736,
      "step": 21835
    },
    {
      "epoch": 3.252904378909741,
      "grad_norm": 12.4375,
      "learning_rate": 4.940479513260006e-05,
      "loss": 1.0217,
      "num_input_tokens_seen": 12668616,
      "step": 21840
    },
    {
      "epoch": 3.2536490914507,
      "grad_norm": 4.28125,
      "learning_rate": 4.940409009735964e-05,
      "loss": 1.1744,
      "num_input_tokens_seen": 12671656,
      "step": 21845
    },
    {
      "epoch": 3.2543938039916593,
      "grad_norm": 10.5,
      "learning_rate": 4.940338464983691e-05,
      "loss": 1.0387,
      "num_input_tokens_seen": 12674504,
      "step": 21850
    },
    {
      "epoch": 3.2551385165326185,
      "grad_norm": 5.3125,
      "learning_rate": 4.940267879004381e-05,
      "loss": 1.0591,
      "num_input_tokens_seen": 12677480,
      "step": 21855
    },
    {
      "epoch": 3.2558832290735777,
      "grad_norm": 10.125,
      "learning_rate": 4.9401972517992254e-05,
      "loss": 0.9099,
      "num_input_tokens_seen": 12680072,
      "step": 21860
    },
    {
      "epoch": 3.256627941614537,
      "grad_norm": 7.40625,
      "learning_rate": 4.9401265833694166e-05,
      "loss": 0.8537,
      "num_input_tokens_seen": 12683176,
      "step": 21865
    },
    {
      "epoch": 3.257372654155496,
      "grad_norm": 10.75,
      "learning_rate": 4.940055873716149e-05,
      "loss": 1.1864,
      "num_input_tokens_seen": 12686216,
      "step": 21870
    },
    {
      "epoch": 3.2581173666964554,
      "grad_norm": 10.9375,
      "learning_rate": 4.939985122840619e-05,
      "loss": 1.008,
      "num_input_tokens_seen": 12689064,
      "step": 21875
    },
    {
      "epoch": 3.2588620792374146,
      "grad_norm": 7.03125,
      "learning_rate": 4.939914330744019e-05,
      "loss": 0.9618,
      "num_input_tokens_seen": 12692328,
      "step": 21880
    },
    {
      "epoch": 3.2596067917783733,
      "grad_norm": 15.5625,
      "learning_rate": 4.939843497427547e-05,
      "loss": 1.1144,
      "num_input_tokens_seen": 12695528,
      "step": 21885
    },
    {
      "epoch": 3.260351504319333,
      "grad_norm": 11.875,
      "learning_rate": 4.939772622892398e-05,
      "loss": 0.9455,
      "num_input_tokens_seen": 12698600,
      "step": 21890
    },
    {
      "epoch": 3.2610962168602917,
      "grad_norm": 13.5,
      "learning_rate": 4.93970170713977e-05,
      "loss": 1.3017,
      "num_input_tokens_seen": 12701448,
      "step": 21895
    },
    {
      "epoch": 3.2618409294012514,
      "grad_norm": 11.75,
      "learning_rate": 4.9396307501708625e-05,
      "loss": 1.2489,
      "num_input_tokens_seen": 12704200,
      "step": 21900
    },
    {
      "epoch": 3.26258564194221,
      "grad_norm": 3.28125,
      "learning_rate": 4.939559751986872e-05,
      "loss": 1.3578,
      "num_input_tokens_seen": 12706984,
      "step": 21905
    },
    {
      "epoch": 3.2633303544831693,
      "grad_norm": 7.0625,
      "learning_rate": 4.939488712588999e-05,
      "loss": 0.9158,
      "num_input_tokens_seen": 12709960,
      "step": 21910
    },
    {
      "epoch": 3.2640750670241285,
      "grad_norm": 9.125,
      "learning_rate": 4.939417631978444e-05,
      "loss": 1.0017,
      "num_input_tokens_seen": 12712744,
      "step": 21915
    },
    {
      "epoch": 3.2648197795650877,
      "grad_norm": 13.9375,
      "learning_rate": 4.939346510156407e-05,
      "loss": 1.2003,
      "num_input_tokens_seen": 12715656,
      "step": 21920
    },
    {
      "epoch": 3.265564492106047,
      "grad_norm": 8.4375,
      "learning_rate": 4.93927534712409e-05,
      "loss": 0.9923,
      "num_input_tokens_seen": 12718504,
      "step": 21925
    },
    {
      "epoch": 3.266309204647006,
      "grad_norm": 9.625,
      "learning_rate": 4.939204142882696e-05,
      "loss": 1.3899,
      "num_input_tokens_seen": 12721384,
      "step": 21930
    },
    {
      "epoch": 3.2670539171879653,
      "grad_norm": 9.25,
      "learning_rate": 4.939132897433426e-05,
      "loss": 1.1427,
      "num_input_tokens_seen": 12724424,
      "step": 21935
    },
    {
      "epoch": 3.2677986297289245,
      "grad_norm": 5.1875,
      "learning_rate": 4.939061610777486e-05,
      "loss": 0.9195,
      "num_input_tokens_seen": 12727624,
      "step": 21940
    },
    {
      "epoch": 3.2685433422698837,
      "grad_norm": 16.375,
      "learning_rate": 4.938990282916078e-05,
      "loss": 1.1094,
      "num_input_tokens_seen": 12730376,
      "step": 21945
    },
    {
      "epoch": 3.269288054810843,
      "grad_norm": 17.125,
      "learning_rate": 4.938918913850408e-05,
      "loss": 1.2643,
      "num_input_tokens_seen": 12733128,
      "step": 21950
    },
    {
      "epoch": 3.270032767351802,
      "grad_norm": 30.25,
      "learning_rate": 4.938847503581682e-05,
      "loss": 1.2309,
      "num_input_tokens_seen": 12735912,
      "step": 21955
    },
    {
      "epoch": 3.2707774798927614,
      "grad_norm": 18.875,
      "learning_rate": 4.938776052111106e-05,
      "loss": 1.3738,
      "num_input_tokens_seen": 12738664,
      "step": 21960
    },
    {
      "epoch": 3.2715221924337206,
      "grad_norm": 9.4375,
      "learning_rate": 4.9387045594398875e-05,
      "loss": 1.1718,
      "num_input_tokens_seen": 12741384,
      "step": 21965
    },
    {
      "epoch": 3.2722669049746798,
      "grad_norm": 8.3125,
      "learning_rate": 4.9386330255692346e-05,
      "loss": 1.0666,
      "num_input_tokens_seen": 12743976,
      "step": 21970
    },
    {
      "epoch": 3.273011617515639,
      "grad_norm": 22.75,
      "learning_rate": 4.938561450500354e-05,
      "loss": 1.2428,
      "num_input_tokens_seen": 12747016,
      "step": 21975
    },
    {
      "epoch": 3.273756330056598,
      "grad_norm": 8.375,
      "learning_rate": 4.938489834234457e-05,
      "loss": 1.0017,
      "num_input_tokens_seen": 12749960,
      "step": 21980
    },
    {
      "epoch": 3.2745010425975574,
      "grad_norm": 4.875,
      "learning_rate": 4.9384181767727524e-05,
      "loss": 0.7149,
      "num_input_tokens_seen": 12752808,
      "step": 21985
    },
    {
      "epoch": 3.2752457551385166,
      "grad_norm": 15.375,
      "learning_rate": 4.9383464781164515e-05,
      "loss": 0.9499,
      "num_input_tokens_seen": 12755656,
      "step": 21990
    },
    {
      "epoch": 3.2759904676794758,
      "grad_norm": 6.1875,
      "learning_rate": 4.938274738266764e-05,
      "loss": 1.113,
      "num_input_tokens_seen": 12758952,
      "step": 21995
    },
    {
      "epoch": 3.276735180220435,
      "grad_norm": 8.5,
      "learning_rate": 4.938202957224903e-05,
      "loss": 0.8638,
      "num_input_tokens_seen": 12761640,
      "step": 22000
    },
    {
      "epoch": 3.277479892761394,
      "grad_norm": 18.375,
      "learning_rate": 4.938131134992082e-05,
      "loss": 1.4735,
      "num_input_tokens_seen": 12764328,
      "step": 22005
    },
    {
      "epoch": 3.2782246053023534,
      "grad_norm": 11.5625,
      "learning_rate": 4.938059271569513e-05,
      "loss": 0.9408,
      "num_input_tokens_seen": 12767400,
      "step": 22010
    },
    {
      "epoch": 3.2789693178433126,
      "grad_norm": 5.46875,
      "learning_rate": 4.937987366958411e-05,
      "loss": 1.0356,
      "num_input_tokens_seen": 12769960,
      "step": 22015
    },
    {
      "epoch": 3.279714030384272,
      "grad_norm": 16.875,
      "learning_rate": 4.93791542115999e-05,
      "loss": 1.1879,
      "num_input_tokens_seen": 12773288,
      "step": 22020
    },
    {
      "epoch": 3.280458742925231,
      "grad_norm": 16.25,
      "learning_rate": 4.937843434175466e-05,
      "loss": 1.3191,
      "num_input_tokens_seen": 12776040,
      "step": 22025
    },
    {
      "epoch": 3.28120345546619,
      "grad_norm": 7.375,
      "learning_rate": 4.937771406006054e-05,
      "loss": 1.3686,
      "num_input_tokens_seen": 12779016,
      "step": 22030
    },
    {
      "epoch": 3.2819481680071494,
      "grad_norm": 9.375,
      "learning_rate": 4.937699336652973e-05,
      "loss": 1.0974,
      "num_input_tokens_seen": 12782216,
      "step": 22035
    },
    {
      "epoch": 3.2826928805481086,
      "grad_norm": 6.0,
      "learning_rate": 4.937627226117438e-05,
      "loss": 1.0221,
      "num_input_tokens_seen": 12785032,
      "step": 22040
    },
    {
      "epoch": 3.283437593089068,
      "grad_norm": 7.71875,
      "learning_rate": 4.9375550744006695e-05,
      "loss": 1.0337,
      "num_input_tokens_seen": 12787784,
      "step": 22045
    },
    {
      "epoch": 3.284182305630027,
      "grad_norm": 7.375,
      "learning_rate": 4.9374828815038856e-05,
      "loss": 1.0307,
      "num_input_tokens_seen": 12790728,
      "step": 22050
    },
    {
      "epoch": 3.284927018170986,
      "grad_norm": 9.6875,
      "learning_rate": 4.937410647428304e-05,
      "loss": 0.8147,
      "num_input_tokens_seen": 12793544,
      "step": 22055
    },
    {
      "epoch": 3.285671730711945,
      "grad_norm": 10.4375,
      "learning_rate": 4.9373383721751486e-05,
      "loss": 1.1473,
      "num_input_tokens_seen": 12796520,
      "step": 22060
    },
    {
      "epoch": 3.2864164432529046,
      "grad_norm": 6.84375,
      "learning_rate": 4.9372660557456384e-05,
      "loss": 1.0865,
      "num_input_tokens_seen": 12799432,
      "step": 22065
    },
    {
      "epoch": 3.2871611557938634,
      "grad_norm": 8.6875,
      "learning_rate": 4.937193698140995e-05,
      "loss": 1.4237,
      "num_input_tokens_seen": 12802696,
      "step": 22070
    },
    {
      "epoch": 3.2879058683348226,
      "grad_norm": 14.0,
      "learning_rate": 4.9371212993624405e-05,
      "loss": 1.1791,
      "num_input_tokens_seen": 12805800,
      "step": 22075
    },
    {
      "epoch": 3.2886505808757818,
      "grad_norm": 6.875,
      "learning_rate": 4.9370488594112e-05,
      "loss": 1.1572,
      "num_input_tokens_seen": 12808648,
      "step": 22080
    },
    {
      "epoch": 3.289395293416741,
      "grad_norm": 7.625,
      "learning_rate": 4.936976378288495e-05,
      "loss": 0.9525,
      "num_input_tokens_seen": 12811784,
      "step": 22085
    },
    {
      "epoch": 3.2901400059577,
      "grad_norm": 12.8125,
      "learning_rate": 4.93690385599555e-05,
      "loss": 1.2013,
      "num_input_tokens_seen": 12814792,
      "step": 22090
    },
    {
      "epoch": 3.2908847184986594,
      "grad_norm": 8.4375,
      "learning_rate": 4.9368312925335925e-05,
      "loss": 1.0466,
      "num_input_tokens_seen": 12817448,
      "step": 22095
    },
    {
      "epoch": 3.2916294310396186,
      "grad_norm": 10.875,
      "learning_rate": 4.9367586879038466e-05,
      "loss": 0.9135,
      "num_input_tokens_seen": 12820232,
      "step": 22100
    },
    {
      "epoch": 3.292374143580578,
      "grad_norm": 9.25,
      "learning_rate": 4.93668604210754e-05,
      "loss": 1.2014,
      "num_input_tokens_seen": 12823272,
      "step": 22105
    },
    {
      "epoch": 3.293118856121537,
      "grad_norm": 8.9375,
      "learning_rate": 4.936613355145898e-05,
      "loss": 0.9881,
      "num_input_tokens_seen": 12826280,
      "step": 22110
    },
    {
      "epoch": 3.293863568662496,
      "grad_norm": 7.78125,
      "learning_rate": 4.936540627020151e-05,
      "loss": 1.0788,
      "num_input_tokens_seen": 12829096,
      "step": 22115
    },
    {
      "epoch": 3.2946082812034554,
      "grad_norm": 7.25,
      "learning_rate": 4.936467857731526e-05,
      "loss": 1.0383,
      "num_input_tokens_seen": 12831912,
      "step": 22120
    },
    {
      "epoch": 3.2953529937444146,
      "grad_norm": 10.0625,
      "learning_rate": 4.9363950472812524e-05,
      "loss": 0.8682,
      "num_input_tokens_seen": 12834536,
      "step": 22125
    },
    {
      "epoch": 3.296097706285374,
      "grad_norm": 11.9375,
      "learning_rate": 4.936322195670561e-05,
      "loss": 1.0217,
      "num_input_tokens_seen": 12837288,
      "step": 22130
    },
    {
      "epoch": 3.296842418826333,
      "grad_norm": 9.5625,
      "learning_rate": 4.936249302900682e-05,
      "loss": 1.4588,
      "num_input_tokens_seen": 12840392,
      "step": 22135
    },
    {
      "epoch": 3.297587131367292,
      "grad_norm": 9.75,
      "learning_rate": 4.936176368972848e-05,
      "loss": 1.1207,
      "num_input_tokens_seen": 12843304,
      "step": 22140
    },
    {
      "epoch": 3.2983318439082514,
      "grad_norm": 10.0625,
      "learning_rate": 4.93610339388829e-05,
      "loss": 1.3829,
      "num_input_tokens_seen": 12846184,
      "step": 22145
    },
    {
      "epoch": 3.2990765564492106,
      "grad_norm": 15.1875,
      "learning_rate": 4.936030377648241e-05,
      "loss": 1.3571,
      "num_input_tokens_seen": 12848968,
      "step": 22150
    },
    {
      "epoch": 3.29982126899017,
      "grad_norm": 4.75,
      "learning_rate": 4.935957320253934e-05,
      "loss": 0.7946,
      "num_input_tokens_seen": 12852264,
      "step": 22155
    },
    {
      "epoch": 3.300565981531129,
      "grad_norm": 6.53125,
      "learning_rate": 4.9358842217066044e-05,
      "loss": 1.0231,
      "num_input_tokens_seen": 12855048,
      "step": 22160
    },
    {
      "epoch": 3.301310694072088,
      "grad_norm": 16.625,
      "learning_rate": 4.935811082007487e-05,
      "loss": 1.223,
      "num_input_tokens_seen": 12857864,
      "step": 22165
    },
    {
      "epoch": 3.3020554066130474,
      "grad_norm": 4.90625,
      "learning_rate": 4.935737901157816e-05,
      "loss": 1.0495,
      "num_input_tokens_seen": 12860936,
      "step": 22170
    },
    {
      "epoch": 3.3028001191540066,
      "grad_norm": 12.6875,
      "learning_rate": 4.935664679158829e-05,
      "loss": 1.163,
      "num_input_tokens_seen": 12864008,
      "step": 22175
    },
    {
      "epoch": 3.303544831694966,
      "grad_norm": 15.875,
      "learning_rate": 4.935591416011763e-05,
      "loss": 1.1561,
      "num_input_tokens_seen": 12866760,
      "step": 22180
    },
    {
      "epoch": 3.304289544235925,
      "grad_norm": 8.9375,
      "learning_rate": 4.9355181117178564e-05,
      "loss": 1.1298,
      "num_input_tokens_seen": 12869576,
      "step": 22185
    },
    {
      "epoch": 3.3050342567768842,
      "grad_norm": 12.1875,
      "learning_rate": 4.935444766278345e-05,
      "loss": 1.318,
      "num_input_tokens_seen": 12872648,
      "step": 22190
    },
    {
      "epoch": 3.3057789693178434,
      "grad_norm": 5.21875,
      "learning_rate": 4.93537137969447e-05,
      "loss": 1.1636,
      "num_input_tokens_seen": 12875336,
      "step": 22195
    },
    {
      "epoch": 3.3065236818588026,
      "grad_norm": 20.875,
      "learning_rate": 4.935297951967471e-05,
      "loss": 1.1655,
      "num_input_tokens_seen": 12878088,
      "step": 22200
    },
    {
      "epoch": 3.307268394399762,
      "grad_norm": 11.6875,
      "learning_rate": 4.9352244830985886e-05,
      "loss": 0.8791,
      "num_input_tokens_seen": 12880936,
      "step": 22205
    },
    {
      "epoch": 3.308013106940721,
      "grad_norm": 11.0625,
      "learning_rate": 4.935150973089063e-05,
      "loss": 1.0217,
      "num_input_tokens_seen": 12883656,
      "step": 22210
    },
    {
      "epoch": 3.3087578194816802,
      "grad_norm": 16.75,
      "learning_rate": 4.935077421940137e-05,
      "loss": 0.9508,
      "num_input_tokens_seen": 12886568,
      "step": 22215
    },
    {
      "epoch": 3.3095025320226394,
      "grad_norm": 4.53125,
      "learning_rate": 4.935003829653053e-05,
      "loss": 1.1566,
      "num_input_tokens_seen": 12889896,
      "step": 22220
    },
    {
      "epoch": 3.310247244563598,
      "grad_norm": 12.875,
      "learning_rate": 4.934930196229054e-05,
      "loss": 1.0889,
      "num_input_tokens_seen": 12892680,
      "step": 22225
    },
    {
      "epoch": 3.310991957104558,
      "grad_norm": 7.375,
      "learning_rate": 4.9348565216693845e-05,
      "loss": 0.9637,
      "num_input_tokens_seen": 12895688,
      "step": 22230
    },
    {
      "epoch": 3.3117366696455166,
      "grad_norm": 7.96875,
      "learning_rate": 4.9347828059752874e-05,
      "loss": 1.1458,
      "num_input_tokens_seen": 12898376,
      "step": 22235
    },
    {
      "epoch": 3.3124813821864763,
      "grad_norm": 7.40625,
      "learning_rate": 4.934709049148011e-05,
      "loss": 1.0571,
      "num_input_tokens_seen": 12901384,
      "step": 22240
    },
    {
      "epoch": 3.313226094727435,
      "grad_norm": 14.1875,
      "learning_rate": 4.934635251188799e-05,
      "loss": 1.0176,
      "num_input_tokens_seen": 12904200,
      "step": 22245
    },
    {
      "epoch": 3.313970807268394,
      "grad_norm": 13.5,
      "learning_rate": 4.934561412098899e-05,
      "loss": 0.9075,
      "num_input_tokens_seen": 12907304,
      "step": 22250
    },
    {
      "epoch": 3.3147155198093534,
      "grad_norm": 13.625,
      "learning_rate": 4.934487531879558e-05,
      "loss": 1.4628,
      "num_input_tokens_seen": 12910056,
      "step": 22255
    },
    {
      "epoch": 3.3154602323503126,
      "grad_norm": 12.25,
      "learning_rate": 4.934413610532025e-05,
      "loss": 0.9452,
      "num_input_tokens_seen": 12912936,
      "step": 22260
    },
    {
      "epoch": 3.316204944891272,
      "grad_norm": 11.6875,
      "learning_rate": 4.9343396480575474e-05,
      "loss": 1.3679,
      "num_input_tokens_seen": 12915880,
      "step": 22265
    },
    {
      "epoch": 3.316949657432231,
      "grad_norm": 13.875,
      "learning_rate": 4.9342656444573764e-05,
      "loss": 0.9263,
      "num_input_tokens_seen": 12918760,
      "step": 22270
    },
    {
      "epoch": 3.3176943699731902,
      "grad_norm": 11.25,
      "learning_rate": 4.934191599732762e-05,
      "loss": 1.1795,
      "num_input_tokens_seen": 12921672,
      "step": 22275
    },
    {
      "epoch": 3.3184390825141494,
      "grad_norm": 17.125,
      "learning_rate": 4.934117513884953e-05,
      "loss": 1.2555,
      "num_input_tokens_seen": 12924648,
      "step": 22280
    },
    {
      "epoch": 3.3191837950551086,
      "grad_norm": 15.0625,
      "learning_rate": 4.934043386915203e-05,
      "loss": 1.2654,
      "num_input_tokens_seen": 12927560,
      "step": 22285
    },
    {
      "epoch": 3.319928507596068,
      "grad_norm": 8.375,
      "learning_rate": 4.933969218824764e-05,
      "loss": 1.1975,
      "num_input_tokens_seen": 12930760,
      "step": 22290
    },
    {
      "epoch": 3.320673220137027,
      "grad_norm": 11.6875,
      "learning_rate": 4.933895009614889e-05,
      "loss": 1.0639,
      "num_input_tokens_seen": 12933576,
      "step": 22295
    },
    {
      "epoch": 3.3214179326779862,
      "grad_norm": 7.09375,
      "learning_rate": 4.933820759286831e-05,
      "loss": 0.9613,
      "num_input_tokens_seen": 12936808,
      "step": 22300
    },
    {
      "epoch": 3.3221626452189454,
      "grad_norm": 5.59375,
      "learning_rate": 4.933746467841846e-05,
      "loss": 0.9041,
      "num_input_tokens_seen": 12939464,
      "step": 22305
    },
    {
      "epoch": 3.3229073577599046,
      "grad_norm": 4.4375,
      "learning_rate": 4.9336721352811864e-05,
      "loss": 0.9743,
      "num_input_tokens_seen": 12942472,
      "step": 22310
    },
    {
      "epoch": 3.323652070300864,
      "grad_norm": 4.09375,
      "learning_rate": 4.933597761606111e-05,
      "loss": 0.8853,
      "num_input_tokens_seen": 12945512,
      "step": 22315
    },
    {
      "epoch": 3.324396782841823,
      "grad_norm": 13.75,
      "learning_rate": 4.9335233468178744e-05,
      "loss": 1.003,
      "num_input_tokens_seen": 12948424,
      "step": 22320
    },
    {
      "epoch": 3.3251414953827823,
      "grad_norm": 12.25,
      "learning_rate": 4.9334488909177336e-05,
      "loss": 1.2228,
      "num_input_tokens_seen": 12951144,
      "step": 22325
    },
    {
      "epoch": 3.3258862079237415,
      "grad_norm": 7.84375,
      "learning_rate": 4.9333743939069476e-05,
      "loss": 0.9415,
      "num_input_tokens_seen": 12954280,
      "step": 22330
    },
    {
      "epoch": 3.3266309204647007,
      "grad_norm": 5.78125,
      "learning_rate": 4.9332998557867735e-05,
      "loss": 1.1016,
      "num_input_tokens_seen": 12957320,
      "step": 22335
    },
    {
      "epoch": 3.32737563300566,
      "grad_norm": 14.1875,
      "learning_rate": 4.933225276558473e-05,
      "loss": 1.315,
      "num_input_tokens_seen": 12960136,
      "step": 22340
    },
    {
      "epoch": 3.328120345546619,
      "grad_norm": 7.875,
      "learning_rate": 4.933150656223303e-05,
      "loss": 0.8829,
      "num_input_tokens_seen": 12962888,
      "step": 22345
    },
    {
      "epoch": 3.3288650580875783,
      "grad_norm": 12.8125,
      "learning_rate": 4.933075994782527e-05,
      "loss": 1.0409,
      "num_input_tokens_seen": 12965608,
      "step": 22350
    },
    {
      "epoch": 3.3296097706285375,
      "grad_norm": 11.1875,
      "learning_rate": 4.933001292237404e-05,
      "loss": 0.8182,
      "num_input_tokens_seen": 12968840,
      "step": 22355
    },
    {
      "epoch": 3.3303544831694967,
      "grad_norm": 10.9375,
      "learning_rate": 4.9329265485891966e-05,
      "loss": 1.1612,
      "num_input_tokens_seen": 12971496,
      "step": 22360
    },
    {
      "epoch": 3.331099195710456,
      "grad_norm": 9.0,
      "learning_rate": 4.9328517638391684e-05,
      "loss": 0.8852,
      "num_input_tokens_seen": 12974280,
      "step": 22365
    },
    {
      "epoch": 3.331843908251415,
      "grad_norm": 8.375,
      "learning_rate": 4.932776937988582e-05,
      "loss": 1.2675,
      "num_input_tokens_seen": 12977352,
      "step": 22370
    },
    {
      "epoch": 3.3325886207923743,
      "grad_norm": 15.25,
      "learning_rate": 4.932702071038703e-05,
      "loss": 1.1558,
      "num_input_tokens_seen": 12980072,
      "step": 22375
    },
    {
      "epoch": 3.3333333333333335,
      "grad_norm": 6.25,
      "learning_rate": 4.932627162990794e-05,
      "loss": 1.0639,
      "num_input_tokens_seen": 12982984,
      "step": 22380
    },
    {
      "epoch": 3.3340780458742927,
      "grad_norm": 13.9375,
      "learning_rate": 4.932552213846121e-05,
      "loss": 0.9581,
      "num_input_tokens_seen": 12985800,
      "step": 22385
    },
    {
      "epoch": 3.334822758415252,
      "grad_norm": 5.9375,
      "learning_rate": 4.932477223605951e-05,
      "loss": 0.8061,
      "num_input_tokens_seen": 12988456,
      "step": 22390
    },
    {
      "epoch": 3.335567470956211,
      "grad_norm": 4.40625,
      "learning_rate": 4.932402192271551e-05,
      "loss": 0.9184,
      "num_input_tokens_seen": 12991240,
      "step": 22395
    },
    {
      "epoch": 3.33631218349717,
      "grad_norm": 10.625,
      "learning_rate": 4.9323271198441886e-05,
      "loss": 1.1123,
      "num_input_tokens_seen": 12994120,
      "step": 22400
    },
    {
      "epoch": 3.3370568960381295,
      "grad_norm": 5.4375,
      "learning_rate": 4.932252006325131e-05,
      "loss": 0.8705,
      "num_input_tokens_seen": 12997096,
      "step": 22405
    },
    {
      "epoch": 3.3378016085790883,
      "grad_norm": 6.8125,
      "learning_rate": 4.932176851715647e-05,
      "loss": 0.5685,
      "num_input_tokens_seen": 12999720,
      "step": 22410
    },
    {
      "epoch": 3.338546321120048,
      "grad_norm": 6.59375,
      "learning_rate": 4.932101656017008e-05,
      "loss": 1.0985,
      "num_input_tokens_seen": 13003016,
      "step": 22415
    },
    {
      "epoch": 3.3392910336610067,
      "grad_norm": 15.75,
      "learning_rate": 4.9320264192304835e-05,
      "loss": 0.9232,
      "num_input_tokens_seen": 13005864,
      "step": 22420
    },
    {
      "epoch": 3.340035746201966,
      "grad_norm": 11.625,
      "learning_rate": 4.931951141357344e-05,
      "loss": 1.4996,
      "num_input_tokens_seen": 13008936,
      "step": 22425
    },
    {
      "epoch": 3.340780458742925,
      "grad_norm": 18.125,
      "learning_rate": 4.931875822398862e-05,
      "loss": 0.9446,
      "num_input_tokens_seen": 13011816,
      "step": 22430
    },
    {
      "epoch": 3.3415251712838843,
      "grad_norm": 17.875,
      "learning_rate": 4.93180046235631e-05,
      "loss": 1.2572,
      "num_input_tokens_seen": 13014568,
      "step": 22435
    },
    {
      "epoch": 3.3422698838248435,
      "grad_norm": 7.1875,
      "learning_rate": 4.9317250612309594e-05,
      "loss": 1.115,
      "num_input_tokens_seen": 13017512,
      "step": 22440
    },
    {
      "epoch": 3.3430145963658027,
      "grad_norm": 12.3125,
      "learning_rate": 4.9316496190240866e-05,
      "loss": 1.1404,
      "num_input_tokens_seen": 13020424,
      "step": 22445
    },
    {
      "epoch": 3.343759308906762,
      "grad_norm": 11.0625,
      "learning_rate": 4.931574135736965e-05,
      "loss": 0.9969,
      "num_input_tokens_seen": 13023368,
      "step": 22450
    },
    {
      "epoch": 3.344504021447721,
      "grad_norm": 6.4375,
      "learning_rate": 4.931498611370869e-05,
      "loss": 1.236,
      "num_input_tokens_seen": 13026056,
      "step": 22455
    },
    {
      "epoch": 3.3452487339886803,
      "grad_norm": 12.1875,
      "learning_rate": 4.9314230459270756e-05,
      "loss": 1.236,
      "num_input_tokens_seen": 13028808,
      "step": 22460
    },
    {
      "epoch": 3.3459934465296395,
      "grad_norm": 3.890625,
      "learning_rate": 4.9313474394068604e-05,
      "loss": 0.8086,
      "num_input_tokens_seen": 13031624,
      "step": 22465
    },
    {
      "epoch": 3.3467381590705987,
      "grad_norm": 4.34375,
      "learning_rate": 4.931271791811502e-05,
      "loss": 1.3054,
      "num_input_tokens_seen": 13034568,
      "step": 22470
    },
    {
      "epoch": 3.347482871611558,
      "grad_norm": 20.125,
      "learning_rate": 4.931196103142278e-05,
      "loss": 1.1342,
      "num_input_tokens_seen": 13037512,
      "step": 22475
    },
    {
      "epoch": 3.348227584152517,
      "grad_norm": 7.46875,
      "learning_rate": 4.9311203734004665e-05,
      "loss": 0.839,
      "num_input_tokens_seen": 13040520,
      "step": 22480
    },
    {
      "epoch": 3.3489722966934763,
      "grad_norm": 9.5,
      "learning_rate": 4.931044602587346e-05,
      "loss": 1.2828,
      "num_input_tokens_seen": 13043272,
      "step": 22485
    },
    {
      "epoch": 3.3497170092344355,
      "grad_norm": 7.78125,
      "learning_rate": 4.930968790704199e-05,
      "loss": 0.7664,
      "num_input_tokens_seen": 13046472,
      "step": 22490
    },
    {
      "epoch": 3.3504617217753947,
      "grad_norm": 9.375,
      "learning_rate": 4.930892937752305e-05,
      "loss": 1.1673,
      "num_input_tokens_seen": 13049416,
      "step": 22495
    },
    {
      "epoch": 3.351206434316354,
      "grad_norm": 13.4375,
      "learning_rate": 4.930817043732945e-05,
      "loss": 1.1519,
      "num_input_tokens_seen": 13052360,
      "step": 22500
    },
    {
      "epoch": 3.351951146857313,
      "grad_norm": 15.0,
      "learning_rate": 4.930741108647402e-05,
      "loss": 1.577,
      "num_input_tokens_seen": 13055240,
      "step": 22505
    },
    {
      "epoch": 3.3526958593982723,
      "grad_norm": 9.9375,
      "learning_rate": 4.9306651324969583e-05,
      "loss": 0.7868,
      "num_input_tokens_seen": 13058152,
      "step": 22510
    },
    {
      "epoch": 3.3534405719392315,
      "grad_norm": 10.3125,
      "learning_rate": 4.9305891152828976e-05,
      "loss": 1.2081,
      "num_input_tokens_seen": 13060936,
      "step": 22515
    },
    {
      "epoch": 3.3541852844801907,
      "grad_norm": 8.25,
      "learning_rate": 4.930513057006504e-05,
      "loss": 1.2583,
      "num_input_tokens_seen": 13063784,
      "step": 22520
    },
    {
      "epoch": 3.35492999702115,
      "grad_norm": 12.0,
      "learning_rate": 4.930436957669063e-05,
      "loss": 1.135,
      "num_input_tokens_seen": 13066536,
      "step": 22525
    },
    {
      "epoch": 3.355674709562109,
      "grad_norm": 13.375,
      "learning_rate": 4.93036081727186e-05,
      "loss": 1.5033,
      "num_input_tokens_seen": 13069224,
      "step": 22530
    },
    {
      "epoch": 3.3564194221030683,
      "grad_norm": 6.5625,
      "learning_rate": 4.93028463581618e-05,
      "loss": 0.8692,
      "num_input_tokens_seen": 13072072,
      "step": 22535
    },
    {
      "epoch": 3.3571641346440275,
      "grad_norm": 10.0,
      "learning_rate": 4.930208413303312e-05,
      "loss": 0.9636,
      "num_input_tokens_seen": 13074920,
      "step": 22540
    },
    {
      "epoch": 3.3579088471849867,
      "grad_norm": 10.6875,
      "learning_rate": 4.930132149734542e-05,
      "loss": 1.076,
      "num_input_tokens_seen": 13077800,
      "step": 22545
    },
    {
      "epoch": 3.358653559725946,
      "grad_norm": 13.375,
      "learning_rate": 4.93005584511116e-05,
      "loss": 1.3488,
      "num_input_tokens_seen": 13080456,
      "step": 22550
    },
    {
      "epoch": 3.359398272266905,
      "grad_norm": 9.0625,
      "learning_rate": 4.929979499434454e-05,
      "loss": 1.2345,
      "num_input_tokens_seen": 13083496,
      "step": 22555
    },
    {
      "epoch": 3.3601429848078643,
      "grad_norm": 6.34375,
      "learning_rate": 4.929903112705714e-05,
      "loss": 0.8712,
      "num_input_tokens_seen": 13086952,
      "step": 22560
    },
    {
      "epoch": 3.3608876973488235,
      "grad_norm": 15.375,
      "learning_rate": 4.9298266849262306e-05,
      "loss": 1.2334,
      "num_input_tokens_seen": 13089480,
      "step": 22565
    },
    {
      "epoch": 3.3616324098897827,
      "grad_norm": 5.53125,
      "learning_rate": 4.929750216097295e-05,
      "loss": 0.9507,
      "num_input_tokens_seen": 13092488,
      "step": 22570
    },
    {
      "epoch": 3.3623771224307415,
      "grad_norm": 5.9375,
      "learning_rate": 4.929673706220199e-05,
      "loss": 0.7167,
      "num_input_tokens_seen": 13095400,
      "step": 22575
    },
    {
      "epoch": 3.363121834971701,
      "grad_norm": 8.6875,
      "learning_rate": 4.929597155296235e-05,
      "loss": 1.1887,
      "num_input_tokens_seen": 13098600,
      "step": 22580
    },
    {
      "epoch": 3.36386654751266,
      "grad_norm": 7.90625,
      "learning_rate": 4.929520563326697e-05,
      "loss": 1.1572,
      "num_input_tokens_seen": 13101192,
      "step": 22585
    },
    {
      "epoch": 3.3646112600536195,
      "grad_norm": 10.4375,
      "learning_rate": 4.929443930312878e-05,
      "loss": 1.4253,
      "num_input_tokens_seen": 13104008,
      "step": 22590
    },
    {
      "epoch": 3.3653559725945783,
      "grad_norm": 14.875,
      "learning_rate": 4.929367256256072e-05,
      "loss": 1.1497,
      "num_input_tokens_seen": 13106952,
      "step": 22595
    },
    {
      "epoch": 3.3661006851355375,
      "grad_norm": 10.1875,
      "learning_rate": 4.929290541157576e-05,
      "loss": 1.0964,
      "num_input_tokens_seen": 13109832,
      "step": 22600
    },
    {
      "epoch": 3.3668453976764967,
      "grad_norm": 9.4375,
      "learning_rate": 4.929213785018686e-05,
      "loss": 0.8703,
      "num_input_tokens_seen": 13112776,
      "step": 22605
    },
    {
      "epoch": 3.367590110217456,
      "grad_norm": 17.125,
      "learning_rate": 4.9291369878406975e-05,
      "loss": 0.7758,
      "num_input_tokens_seen": 13115560,
      "step": 22610
    },
    {
      "epoch": 3.368334822758415,
      "grad_norm": 10.6875,
      "learning_rate": 4.929060149624909e-05,
      "loss": 1.2399,
      "num_input_tokens_seen": 13118568,
      "step": 22615
    },
    {
      "epoch": 3.3690795352993743,
      "grad_norm": 6.03125,
      "learning_rate": 4.928983270372617e-05,
      "loss": 0.9495,
      "num_input_tokens_seen": 13121320,
      "step": 22620
    },
    {
      "epoch": 3.3698242478403335,
      "grad_norm": 12.0625,
      "learning_rate": 4.928906350085122e-05,
      "loss": 1.0821,
      "num_input_tokens_seen": 13124200,
      "step": 22625
    },
    {
      "epoch": 3.3705689603812927,
      "grad_norm": 7.65625,
      "learning_rate": 4.928829388763723e-05,
      "loss": 0.9966,
      "num_input_tokens_seen": 13126984,
      "step": 22630
    },
    {
      "epoch": 3.371313672922252,
      "grad_norm": 7.78125,
      "learning_rate": 4.928752386409719e-05,
      "loss": 1.2786,
      "num_input_tokens_seen": 13129832,
      "step": 22635
    },
    {
      "epoch": 3.372058385463211,
      "grad_norm": 16.75,
      "learning_rate": 4.9286753430244126e-05,
      "loss": 1.2555,
      "num_input_tokens_seen": 13132904,
      "step": 22640
    },
    {
      "epoch": 3.3728030980041703,
      "grad_norm": 8.5,
      "learning_rate": 4.928598258609105e-05,
      "loss": 1.0217,
      "num_input_tokens_seen": 13135656,
      "step": 22645
    },
    {
      "epoch": 3.3735478105451295,
      "grad_norm": 7.71875,
      "learning_rate": 4.928521133165098e-05,
      "loss": 0.7481,
      "num_input_tokens_seen": 13138472,
      "step": 22650
    },
    {
      "epoch": 3.3742925230860887,
      "grad_norm": 8.75,
      "learning_rate": 4.928443966693694e-05,
      "loss": 0.7982,
      "num_input_tokens_seen": 13141384,
      "step": 22655
    },
    {
      "epoch": 3.375037235627048,
      "grad_norm": 11.25,
      "learning_rate": 4.928366759196198e-05,
      "loss": 1.217,
      "num_input_tokens_seen": 13144264,
      "step": 22660
    },
    {
      "epoch": 3.375781948168007,
      "grad_norm": 8.375,
      "learning_rate": 4.9282895106739136e-05,
      "loss": 0.9548,
      "num_input_tokens_seen": 13146888,
      "step": 22665
    },
    {
      "epoch": 3.3765266607089663,
      "grad_norm": 10.1875,
      "learning_rate": 4.928212221128146e-05,
      "loss": 0.9705,
      "num_input_tokens_seen": 13149640,
      "step": 22670
    },
    {
      "epoch": 3.3772713732499255,
      "grad_norm": 15.5625,
      "learning_rate": 4.928134890560201e-05,
      "loss": 1.067,
      "num_input_tokens_seen": 13152392,
      "step": 22675
    },
    {
      "epoch": 3.3780160857908847,
      "grad_norm": 11.5625,
      "learning_rate": 4.928057518971384e-05,
      "loss": 1.0909,
      "num_input_tokens_seen": 13155176,
      "step": 22680
    },
    {
      "epoch": 3.378760798331844,
      "grad_norm": 6.0625,
      "learning_rate": 4.9279801063630035e-05,
      "loss": 1.2016,
      "num_input_tokens_seen": 13158120,
      "step": 22685
    },
    {
      "epoch": 3.379505510872803,
      "grad_norm": 12.6875,
      "learning_rate": 4.9279026527363666e-05,
      "loss": 0.9533,
      "num_input_tokens_seen": 13160776,
      "step": 22690
    },
    {
      "epoch": 3.3802502234137624,
      "grad_norm": 17.5,
      "learning_rate": 4.927825158092783e-05,
      "loss": 1.2683,
      "num_input_tokens_seen": 13163912,
      "step": 22695
    },
    {
      "epoch": 3.3809949359547216,
      "grad_norm": 12.125,
      "learning_rate": 4.9277476224335603e-05,
      "loss": 0.7508,
      "num_input_tokens_seen": 13166824,
      "step": 22700
    },
    {
      "epoch": 3.3817396484956808,
      "grad_norm": 8.125,
      "learning_rate": 4.927670045760009e-05,
      "loss": 1.1084,
      "num_input_tokens_seen": 13169640,
      "step": 22705
    },
    {
      "epoch": 3.38248436103664,
      "grad_norm": 12.5,
      "learning_rate": 4.927592428073439e-05,
      "loss": 1.264,
      "num_input_tokens_seen": 13172360,
      "step": 22710
    },
    {
      "epoch": 3.383229073577599,
      "grad_norm": 10.25,
      "learning_rate": 4.927514769375163e-05,
      "loss": 0.8921,
      "num_input_tokens_seen": 13175144,
      "step": 22715
    },
    {
      "epoch": 3.3839737861185584,
      "grad_norm": 16.5,
      "learning_rate": 4.9274370696664916e-05,
      "loss": 1.5143,
      "num_input_tokens_seen": 13177768,
      "step": 22720
    },
    {
      "epoch": 3.3847184986595176,
      "grad_norm": 4.53125,
      "learning_rate": 4.9273593289487384e-05,
      "loss": 0.8482,
      "num_input_tokens_seen": 13180520,
      "step": 22725
    },
    {
      "epoch": 3.3854632112004768,
      "grad_norm": 7.125,
      "learning_rate": 4.9272815472232165e-05,
      "loss": 1.3003,
      "num_input_tokens_seen": 13183368,
      "step": 22730
    },
    {
      "epoch": 3.386207923741436,
      "grad_norm": 11.0,
      "learning_rate": 4.9272037244912394e-05,
      "loss": 1.1379,
      "num_input_tokens_seen": 13185896,
      "step": 22735
    },
    {
      "epoch": 3.386952636282395,
      "grad_norm": 9.875,
      "learning_rate": 4.927125860754123e-05,
      "loss": 1.0185,
      "num_input_tokens_seen": 13188712,
      "step": 22740
    },
    {
      "epoch": 3.3876973488233544,
      "grad_norm": 11.375,
      "learning_rate": 4.9270479560131813e-05,
      "loss": 0.9933,
      "num_input_tokens_seen": 13191688,
      "step": 22745
    },
    {
      "epoch": 3.388442061364313,
      "grad_norm": 6.21875,
      "learning_rate": 4.926970010269731e-05,
      "loss": 0.949,
      "num_input_tokens_seen": 13194440,
      "step": 22750
    },
    {
      "epoch": 3.389186773905273,
      "grad_norm": 9.25,
      "learning_rate": 4.92689202352509e-05,
      "loss": 0.8517,
      "num_input_tokens_seen": 13197512,
      "step": 22755
    },
    {
      "epoch": 3.3899314864462315,
      "grad_norm": 19.875,
      "learning_rate": 4.926813995780574e-05,
      "loss": 0.9716,
      "num_input_tokens_seen": 13200360,
      "step": 22760
    },
    {
      "epoch": 3.390676198987191,
      "grad_norm": 5.03125,
      "learning_rate": 4.926735927037503e-05,
      "loss": 1.095,
      "num_input_tokens_seen": 13203336,
      "step": 22765
    },
    {
      "epoch": 3.39142091152815,
      "grad_norm": 10.875,
      "learning_rate": 4.9266578172971934e-05,
      "loss": 1.0577,
      "num_input_tokens_seen": 13206184,
      "step": 22770
    },
    {
      "epoch": 3.392165624069109,
      "grad_norm": 10.0,
      "learning_rate": 4.926579666560968e-05,
      "loss": 1.4397,
      "num_input_tokens_seen": 13209256,
      "step": 22775
    },
    {
      "epoch": 3.3929103366100684,
      "grad_norm": 10.125,
      "learning_rate": 4.926501474830144e-05,
      "loss": 0.9071,
      "num_input_tokens_seen": 13212808,
      "step": 22780
    },
    {
      "epoch": 3.3936550491510276,
      "grad_norm": 10.0,
      "learning_rate": 4.926423242106044e-05,
      "loss": 1.064,
      "num_input_tokens_seen": 13215496,
      "step": 22785
    },
    {
      "epoch": 3.3943997616919868,
      "grad_norm": 15.3125,
      "learning_rate": 4.92634496838999e-05,
      "loss": 0.9452,
      "num_input_tokens_seen": 13218216,
      "step": 22790
    },
    {
      "epoch": 3.395144474232946,
      "grad_norm": 13.9375,
      "learning_rate": 4.9262666536833035e-05,
      "loss": 1.3332,
      "num_input_tokens_seen": 13220936,
      "step": 22795
    },
    {
      "epoch": 3.395889186773905,
      "grad_norm": 10.8125,
      "learning_rate": 4.926188297987308e-05,
      "loss": 1.541,
      "num_input_tokens_seen": 13224360,
      "step": 22800
    },
    {
      "epoch": 3.3966338993148644,
      "grad_norm": 18.5,
      "learning_rate": 4.926109901303327e-05,
      "loss": 1.1027,
      "num_input_tokens_seen": 13227080,
      "step": 22805
    },
    {
      "epoch": 3.3973786118558236,
      "grad_norm": 18.0,
      "learning_rate": 4.9260314636326846e-05,
      "loss": 1.1513,
      "num_input_tokens_seen": 13230056,
      "step": 22810
    },
    {
      "epoch": 3.3981233243967828,
      "grad_norm": 10.8125,
      "learning_rate": 4.925952984976707e-05,
      "loss": 1.2968,
      "num_input_tokens_seen": 13232968,
      "step": 22815
    },
    {
      "epoch": 3.398868036937742,
      "grad_norm": 10.1875,
      "learning_rate": 4.925874465336719e-05,
      "loss": 1.0861,
      "num_input_tokens_seen": 13235688,
      "step": 22820
    },
    {
      "epoch": 3.399612749478701,
      "grad_norm": 17.5,
      "learning_rate": 4.9257959047140476e-05,
      "loss": 1.1536,
      "num_input_tokens_seen": 13238408,
      "step": 22825
    },
    {
      "epoch": 3.4003574620196604,
      "grad_norm": 6.0,
      "learning_rate": 4.9257173031100196e-05,
      "loss": 0.8197,
      "num_input_tokens_seen": 13241256,
      "step": 22830
    },
    {
      "epoch": 3.4011021745606196,
      "grad_norm": 12.875,
      "learning_rate": 4.925638660525963e-05,
      "loss": 1.0918,
      "num_input_tokens_seen": 13244168,
      "step": 22835
    },
    {
      "epoch": 3.401846887101579,
      "grad_norm": 4.90625,
      "learning_rate": 4.925559976963207e-05,
      "loss": 1.1821,
      "num_input_tokens_seen": 13246920,
      "step": 22840
    },
    {
      "epoch": 3.402591599642538,
      "grad_norm": 12.4375,
      "learning_rate": 4.9254812524230806e-05,
      "loss": 1.2806,
      "num_input_tokens_seen": 13249992,
      "step": 22845
    },
    {
      "epoch": 3.403336312183497,
      "grad_norm": 10.875,
      "learning_rate": 4.925402486906913e-05,
      "loss": 1.4581,
      "num_input_tokens_seen": 13252776,
      "step": 22850
    },
    {
      "epoch": 3.4040810247244564,
      "grad_norm": 13.75,
      "learning_rate": 4.925323680416036e-05,
      "loss": 0.8995,
      "num_input_tokens_seen": 13255848,
      "step": 22855
    },
    {
      "epoch": 3.4048257372654156,
      "grad_norm": 8.25,
      "learning_rate": 4.92524483295178e-05,
      "loss": 1.1585,
      "num_input_tokens_seen": 13258696,
      "step": 22860
    },
    {
      "epoch": 3.405570449806375,
      "grad_norm": 8.75,
      "learning_rate": 4.925165944515477e-05,
      "loss": 1.1684,
      "num_input_tokens_seen": 13261320,
      "step": 22865
    },
    {
      "epoch": 3.406315162347334,
      "grad_norm": 9.6875,
      "learning_rate": 4.9250870151084614e-05,
      "loss": 1.101,
      "num_input_tokens_seen": 13264296,
      "step": 22870
    },
    {
      "epoch": 3.407059874888293,
      "grad_norm": 19.375,
      "learning_rate": 4.9250080447320644e-05,
      "loss": 1.2753,
      "num_input_tokens_seen": 13266888,
      "step": 22875
    },
    {
      "epoch": 3.4078045874292524,
      "grad_norm": 8.875,
      "learning_rate": 4.924929033387622e-05,
      "loss": 1.1736,
      "num_input_tokens_seen": 13269800,
      "step": 22880
    },
    {
      "epoch": 3.4085492999702116,
      "grad_norm": 13.0625,
      "learning_rate": 4.9248499810764675e-05,
      "loss": 1.3163,
      "num_input_tokens_seen": 13272808,
      "step": 22885
    },
    {
      "epoch": 3.409294012511171,
      "grad_norm": 12.75,
      "learning_rate": 4.9247708877999375e-05,
      "loss": 1.249,
      "num_input_tokens_seen": 13275752,
      "step": 22890
    },
    {
      "epoch": 3.41003872505213,
      "grad_norm": 8.9375,
      "learning_rate": 4.9246917535593675e-05,
      "loss": 1.149,
      "num_input_tokens_seen": 13278472,
      "step": 22895
    },
    {
      "epoch": 3.410783437593089,
      "grad_norm": 13.5,
      "learning_rate": 4.924612578356095e-05,
      "loss": 1.0032,
      "num_input_tokens_seen": 13281160,
      "step": 22900
    },
    {
      "epoch": 3.4115281501340484,
      "grad_norm": 5.71875,
      "learning_rate": 4.9245333621914566e-05,
      "loss": 0.7953,
      "num_input_tokens_seen": 13284072,
      "step": 22905
    },
    {
      "epoch": 3.4122728626750076,
      "grad_norm": 9.6875,
      "learning_rate": 4.9244541050667916e-05,
      "loss": 0.946,
      "num_input_tokens_seen": 13287048,
      "step": 22910
    },
    {
      "epoch": 3.413017575215967,
      "grad_norm": 7.03125,
      "learning_rate": 4.9243748069834386e-05,
      "loss": 1.1751,
      "num_input_tokens_seen": 13290280,
      "step": 22915
    },
    {
      "epoch": 3.413762287756926,
      "grad_norm": 9.8125,
      "learning_rate": 4.924295467942737e-05,
      "loss": 0.8268,
      "num_input_tokens_seen": 13293288,
      "step": 22920
    },
    {
      "epoch": 3.414507000297885,
      "grad_norm": 9.625,
      "learning_rate": 4.924216087946028e-05,
      "loss": 1.1823,
      "num_input_tokens_seen": 13295912,
      "step": 22925
    },
    {
      "epoch": 3.4152517128388444,
      "grad_norm": 16.125,
      "learning_rate": 4.924136666994652e-05,
      "loss": 0.9571,
      "num_input_tokens_seen": 13298760,
      "step": 22930
    },
    {
      "epoch": 3.415996425379803,
      "grad_norm": 7.5625,
      "learning_rate": 4.9240572050899505e-05,
      "loss": 1.0821,
      "num_input_tokens_seen": 13301608,
      "step": 22935
    },
    {
      "epoch": 3.4167411379207624,
      "grad_norm": 12.4375,
      "learning_rate": 4.923977702233266e-05,
      "loss": 0.9469,
      "num_input_tokens_seen": 13304488,
      "step": 22940
    },
    {
      "epoch": 3.4174858504617216,
      "grad_norm": 6.5,
      "learning_rate": 4.923898158425942e-05,
      "loss": 1.1091,
      "num_input_tokens_seen": 13307496,
      "step": 22945
    },
    {
      "epoch": 3.418230563002681,
      "grad_norm": 9.1875,
      "learning_rate": 4.923818573669322e-05,
      "loss": 1.0814,
      "num_input_tokens_seen": 13310344,
      "step": 22950
    },
    {
      "epoch": 3.41897527554364,
      "grad_norm": 8.5625,
      "learning_rate": 4.923738947964751e-05,
      "loss": 0.9814,
      "num_input_tokens_seen": 13313448,
      "step": 22955
    },
    {
      "epoch": 3.419719988084599,
      "grad_norm": 13.1875,
      "learning_rate": 4.923659281313574e-05,
      "loss": 1.2842,
      "num_input_tokens_seen": 13316360,
      "step": 22960
    },
    {
      "epoch": 3.4204647006255584,
      "grad_norm": 24.25,
      "learning_rate": 4.9235795737171365e-05,
      "loss": 1.2853,
      "num_input_tokens_seen": 13319336,
      "step": 22965
    },
    {
      "epoch": 3.4212094131665176,
      "grad_norm": 12.5625,
      "learning_rate": 4.923499825176786e-05,
      "loss": 0.9277,
      "num_input_tokens_seen": 13322120,
      "step": 22970
    },
    {
      "epoch": 3.421954125707477,
      "grad_norm": 20.25,
      "learning_rate": 4.923420035693868e-05,
      "loss": 1.2606,
      "num_input_tokens_seen": 13325128,
      "step": 22975
    },
    {
      "epoch": 3.422698838248436,
      "grad_norm": 3.609375,
      "learning_rate": 4.923340205269732e-05,
      "loss": 1.0538,
      "num_input_tokens_seen": 13327976,
      "step": 22980
    },
    {
      "epoch": 3.423443550789395,
      "grad_norm": 8.3125,
      "learning_rate": 4.923260333905726e-05,
      "loss": 1.149,
      "num_input_tokens_seen": 13330568,
      "step": 22985
    },
    {
      "epoch": 3.4241882633303544,
      "grad_norm": 6.09375,
      "learning_rate": 4.9231804216031995e-05,
      "loss": 0.8874,
      "num_input_tokens_seen": 13333192,
      "step": 22990
    },
    {
      "epoch": 3.4249329758713136,
      "grad_norm": 14.3125,
      "learning_rate": 4.923100468363503e-05,
      "loss": 1.0982,
      "num_input_tokens_seen": 13336296,
      "step": 22995
    },
    {
      "epoch": 3.425677688412273,
      "grad_norm": 15.25,
      "learning_rate": 4.923020474187987e-05,
      "loss": 1.4418,
      "num_input_tokens_seen": 13339432,
      "step": 23000
    },
    {
      "epoch": 3.426422400953232,
      "grad_norm": 9.75,
      "learning_rate": 4.922940439078002e-05,
      "loss": 0.9152,
      "num_input_tokens_seen": 13342248,
      "step": 23005
    },
    {
      "epoch": 3.4271671134941912,
      "grad_norm": 7.625,
      "learning_rate": 4.922860363034901e-05,
      "loss": 1.0672,
      "num_input_tokens_seen": 13345000,
      "step": 23010
    },
    {
      "epoch": 3.4279118260351504,
      "grad_norm": 11.5,
      "learning_rate": 4.922780246060037e-05,
      "loss": 0.8175,
      "num_input_tokens_seen": 13347720,
      "step": 23015
    },
    {
      "epoch": 3.4286565385761096,
      "grad_norm": 9.25,
      "learning_rate": 4.922700088154764e-05,
      "loss": 0.8924,
      "num_input_tokens_seen": 13350696,
      "step": 23020
    },
    {
      "epoch": 3.429401251117069,
      "grad_norm": 16.125,
      "learning_rate": 4.9226198893204335e-05,
      "loss": 1.0144,
      "num_input_tokens_seen": 13353224,
      "step": 23025
    },
    {
      "epoch": 3.430145963658028,
      "grad_norm": 19.875,
      "learning_rate": 4.922539649558403e-05,
      "loss": 1.0832,
      "num_input_tokens_seen": 13355976,
      "step": 23030
    },
    {
      "epoch": 3.4308906761989872,
      "grad_norm": 4.46875,
      "learning_rate": 4.9224593688700274e-05,
      "loss": 0.7955,
      "num_input_tokens_seen": 13358664,
      "step": 23035
    },
    {
      "epoch": 3.4316353887399464,
      "grad_norm": 9.3125,
      "learning_rate": 4.922379047256663e-05,
      "loss": 1.2683,
      "num_input_tokens_seen": 13361352,
      "step": 23040
    },
    {
      "epoch": 3.4323801012809056,
      "grad_norm": 20.375,
      "learning_rate": 4.922298684719666e-05,
      "loss": 1.1448,
      "num_input_tokens_seen": 13364296,
      "step": 23045
    },
    {
      "epoch": 3.433124813821865,
      "grad_norm": 8.8125,
      "learning_rate": 4.922218281260395e-05,
      "loss": 0.8845,
      "num_input_tokens_seen": 13367272,
      "step": 23050
    },
    {
      "epoch": 3.433869526362824,
      "grad_norm": 3.6875,
      "learning_rate": 4.9221378368802085e-05,
      "loss": 0.8236,
      "num_input_tokens_seen": 13369928,
      "step": 23055
    },
    {
      "epoch": 3.4346142389037833,
      "grad_norm": 5.625,
      "learning_rate": 4.9220573515804644e-05,
      "loss": 1.1539,
      "num_input_tokens_seen": 13372808,
      "step": 23060
    },
    {
      "epoch": 3.4353589514447425,
      "grad_norm": 21.875,
      "learning_rate": 4.921976825362523e-05,
      "loss": 1.3101,
      "num_input_tokens_seen": 13375528,
      "step": 23065
    },
    {
      "epoch": 3.4361036639857017,
      "grad_norm": 8.5625,
      "learning_rate": 4.921896258227745e-05,
      "loss": 0.8724,
      "num_input_tokens_seen": 13378088,
      "step": 23070
    },
    {
      "epoch": 3.436848376526661,
      "grad_norm": 15.5625,
      "learning_rate": 4.921815650177491e-05,
      "loss": 1.4801,
      "num_input_tokens_seen": 13381288,
      "step": 23075
    },
    {
      "epoch": 3.43759308906762,
      "grad_norm": 8.375,
      "learning_rate": 4.9217350012131223e-05,
      "loss": 1.0689,
      "num_input_tokens_seen": 13384424,
      "step": 23080
    },
    {
      "epoch": 3.4383378016085793,
      "grad_norm": 8.9375,
      "learning_rate": 4.9216543113360035e-05,
      "loss": 0.9733,
      "num_input_tokens_seen": 13387016,
      "step": 23085
    },
    {
      "epoch": 3.4390825141495385,
      "grad_norm": 13.0625,
      "learning_rate": 4.9215735805474956e-05,
      "loss": 0.8714,
      "num_input_tokens_seen": 13389832,
      "step": 23090
    },
    {
      "epoch": 3.4398272266904977,
      "grad_norm": 6.21875,
      "learning_rate": 4.921492808848963e-05,
      "loss": 1.2249,
      "num_input_tokens_seen": 13392712,
      "step": 23095
    },
    {
      "epoch": 3.4405719392314564,
      "grad_norm": 6.625,
      "learning_rate": 4.921411996241771e-05,
      "loss": 1.1219,
      "num_input_tokens_seen": 13395496,
      "step": 23100
    },
    {
      "epoch": 3.441316651772416,
      "grad_norm": 15.8125,
      "learning_rate": 4.921331142727284e-05,
      "loss": 0.9665,
      "num_input_tokens_seen": 13398216,
      "step": 23105
    },
    {
      "epoch": 3.442061364313375,
      "grad_norm": 18.75,
      "learning_rate": 4.921250248306869e-05,
      "loss": 0.9463,
      "num_input_tokens_seen": 13401000,
      "step": 23110
    },
    {
      "epoch": 3.442806076854334,
      "grad_norm": 4.34375,
      "learning_rate": 4.9211693129818915e-05,
      "loss": 0.6668,
      "num_input_tokens_seen": 13404072,
      "step": 23115
    },
    {
      "epoch": 3.4435507893952932,
      "grad_norm": 8.5,
      "learning_rate": 4.9210883367537184e-05,
      "loss": 0.897,
      "num_input_tokens_seen": 13406952,
      "step": 23120
    },
    {
      "epoch": 3.4442955019362524,
      "grad_norm": 12.1875,
      "learning_rate": 4.9210073196237196e-05,
      "loss": 1.2285,
      "num_input_tokens_seen": 13409512,
      "step": 23125
    },
    {
      "epoch": 3.4450402144772116,
      "grad_norm": 6.1875,
      "learning_rate": 4.9209262615932624e-05,
      "loss": 0.9251,
      "num_input_tokens_seen": 13412360,
      "step": 23130
    },
    {
      "epoch": 3.445784927018171,
      "grad_norm": 17.75,
      "learning_rate": 4.9208451626637164e-05,
      "loss": 1.0656,
      "num_input_tokens_seen": 13415496,
      "step": 23135
    },
    {
      "epoch": 3.44652963955913,
      "grad_norm": 8.5,
      "learning_rate": 4.920764022836452e-05,
      "loss": 0.9408,
      "num_input_tokens_seen": 13418248,
      "step": 23140
    },
    {
      "epoch": 3.4472743521000893,
      "grad_norm": 10.25,
      "learning_rate": 4.920682842112839e-05,
      "loss": 1.1688,
      "num_input_tokens_seen": 13421352,
      "step": 23145
    },
    {
      "epoch": 3.4480190646410485,
      "grad_norm": 12.8125,
      "learning_rate": 4.920601620494251e-05,
      "loss": 1.0071,
      "num_input_tokens_seen": 13424264,
      "step": 23150
    },
    {
      "epoch": 3.4487637771820077,
      "grad_norm": 8.1875,
      "learning_rate": 4.920520357982058e-05,
      "loss": 0.8362,
      "num_input_tokens_seen": 13427048,
      "step": 23155
    },
    {
      "epoch": 3.449508489722967,
      "grad_norm": 8.75,
      "learning_rate": 4.9204390545776334e-05,
      "loss": 1.0607,
      "num_input_tokens_seen": 13430280,
      "step": 23160
    },
    {
      "epoch": 3.450253202263926,
      "grad_norm": 5.75,
      "learning_rate": 4.920357710282352e-05,
      "loss": 1.2855,
      "num_input_tokens_seen": 13433192,
      "step": 23165
    },
    {
      "epoch": 3.4509979148048853,
      "grad_norm": 9.875,
      "learning_rate": 4.9202763250975864e-05,
      "loss": 0.9739,
      "num_input_tokens_seen": 13435784,
      "step": 23170
    },
    {
      "epoch": 3.4517426273458445,
      "grad_norm": 14.0625,
      "learning_rate": 4.920194899024712e-05,
      "loss": 0.9098,
      "num_input_tokens_seen": 13438888,
      "step": 23175
    },
    {
      "epoch": 3.4524873398868037,
      "grad_norm": 13.75,
      "learning_rate": 4.920113432065105e-05,
      "loss": 1.1118,
      "num_input_tokens_seen": 13442024,
      "step": 23180
    },
    {
      "epoch": 3.453232052427763,
      "grad_norm": 7.25,
      "learning_rate": 4.920031924220141e-05,
      "loss": 1.1637,
      "num_input_tokens_seen": 13445416,
      "step": 23185
    },
    {
      "epoch": 3.453976764968722,
      "grad_norm": 14.6875,
      "learning_rate": 4.919950375491197e-05,
      "loss": 1.0131,
      "num_input_tokens_seen": 13448520,
      "step": 23190
    },
    {
      "epoch": 3.4547214775096813,
      "grad_norm": 17.0,
      "learning_rate": 4.919868785879651e-05,
      "loss": 1.2385,
      "num_input_tokens_seen": 13451304,
      "step": 23195
    },
    {
      "epoch": 3.4554661900506405,
      "grad_norm": 10.25,
      "learning_rate": 4.919787155386882e-05,
      "loss": 1.1753,
      "num_input_tokens_seen": 13454024,
      "step": 23200
    },
    {
      "epoch": 3.4562109025915997,
      "grad_norm": 3.5625,
      "learning_rate": 4.919705484014268e-05,
      "loss": 0.9765,
      "num_input_tokens_seen": 13456936,
      "step": 23205
    },
    {
      "epoch": 3.456955615132559,
      "grad_norm": 9.625,
      "learning_rate": 4.919623771763189e-05,
      "loss": 1.154,
      "num_input_tokens_seen": 13460072,
      "step": 23210
    },
    {
      "epoch": 3.457700327673518,
      "grad_norm": 13.1875,
      "learning_rate": 4.919542018635025e-05,
      "loss": 0.9946,
      "num_input_tokens_seen": 13463080,
      "step": 23215
    },
    {
      "epoch": 3.4584450402144773,
      "grad_norm": 7.75,
      "learning_rate": 4.919460224631158e-05,
      "loss": 0.8849,
      "num_input_tokens_seen": 13465992,
      "step": 23220
    },
    {
      "epoch": 3.4591897527554365,
      "grad_norm": 8.375,
      "learning_rate": 4.91937838975297e-05,
      "loss": 0.7968,
      "num_input_tokens_seen": 13468776,
      "step": 23225
    },
    {
      "epoch": 3.4599344652963957,
      "grad_norm": 20.75,
      "learning_rate": 4.9192965140018435e-05,
      "loss": 1.5884,
      "num_input_tokens_seen": 13471656,
      "step": 23230
    },
    {
      "epoch": 3.460679177837355,
      "grad_norm": 6.28125,
      "learning_rate": 4.919214597379161e-05,
      "loss": 1.1641,
      "num_input_tokens_seen": 13474440,
      "step": 23235
    },
    {
      "epoch": 3.461423890378314,
      "grad_norm": 14.75,
      "learning_rate": 4.919132639886306e-05,
      "loss": 1.1459,
      "num_input_tokens_seen": 13476968,
      "step": 23240
    },
    {
      "epoch": 3.4621686029192733,
      "grad_norm": 10.4375,
      "learning_rate": 4.919050641524663e-05,
      "loss": 1.2679,
      "num_input_tokens_seen": 13479976,
      "step": 23245
    },
    {
      "epoch": 3.4629133154602325,
      "grad_norm": 9.625,
      "learning_rate": 4.9189686022956195e-05,
      "loss": 1.4268,
      "num_input_tokens_seen": 13482888,
      "step": 23250
    },
    {
      "epoch": 3.4636580280011917,
      "grad_norm": 8.375,
      "learning_rate": 4.91888652220056e-05,
      "loss": 0.9121,
      "num_input_tokens_seen": 13485704,
      "step": 23255
    },
    {
      "epoch": 3.464402740542151,
      "grad_norm": 10.9375,
      "learning_rate": 4.91880440124087e-05,
      "loss": 0.9857,
      "num_input_tokens_seen": 13488520,
      "step": 23260
    },
    {
      "epoch": 3.4651474530831097,
      "grad_norm": 11.125,
      "learning_rate": 4.918722239417939e-05,
      "loss": 0.7136,
      "num_input_tokens_seen": 13491496,
      "step": 23265
    },
    {
      "epoch": 3.4658921656240693,
      "grad_norm": 12.375,
      "learning_rate": 4.918640036733154e-05,
      "loss": 1.6547,
      "num_input_tokens_seen": 13494376,
      "step": 23270
    },
    {
      "epoch": 3.466636878165028,
      "grad_norm": 12.8125,
      "learning_rate": 4.9185577931879034e-05,
      "loss": 1.0496,
      "num_input_tokens_seen": 13497288,
      "step": 23275
    },
    {
      "epoch": 3.4673815907059877,
      "grad_norm": 9.1875,
      "learning_rate": 4.9184755087835766e-05,
      "loss": 1.2524,
      "num_input_tokens_seen": 13500168,
      "step": 23280
    },
    {
      "epoch": 3.4681263032469465,
      "grad_norm": 11.5625,
      "learning_rate": 4.9183931835215645e-05,
      "loss": 1.1849,
      "num_input_tokens_seen": 13502952,
      "step": 23285
    },
    {
      "epoch": 3.4688710157879057,
      "grad_norm": 8.0,
      "learning_rate": 4.918310817403258e-05,
      "loss": 1.0255,
      "num_input_tokens_seen": 13506216,
      "step": 23290
    },
    {
      "epoch": 3.469615728328865,
      "grad_norm": 8.875,
      "learning_rate": 4.918228410430048e-05,
      "loss": 1.0992,
      "num_input_tokens_seen": 13509672,
      "step": 23295
    },
    {
      "epoch": 3.470360440869824,
      "grad_norm": 11.6875,
      "learning_rate": 4.918145962603326e-05,
      "loss": 1.1749,
      "num_input_tokens_seen": 13512840,
      "step": 23300
    },
    {
      "epoch": 3.4711051534107833,
      "grad_norm": 13.0625,
      "learning_rate": 4.918063473924486e-05,
      "loss": 1.2399,
      "num_input_tokens_seen": 13515848,
      "step": 23305
    },
    {
      "epoch": 3.4718498659517425,
      "grad_norm": 11.625,
      "learning_rate": 4.917980944394922e-05,
      "loss": 1.0234,
      "num_input_tokens_seen": 13518696,
      "step": 23310
    },
    {
      "epoch": 3.4725945784927017,
      "grad_norm": 12.5,
      "learning_rate": 4.9178983740160264e-05,
      "loss": 1.1046,
      "num_input_tokens_seen": 13521768,
      "step": 23315
    },
    {
      "epoch": 3.473339291033661,
      "grad_norm": 13.0,
      "learning_rate": 4.9178157627891956e-05,
      "loss": 0.9937,
      "num_input_tokens_seen": 13524456,
      "step": 23320
    },
    {
      "epoch": 3.47408400357462,
      "grad_norm": 10.8125,
      "learning_rate": 4.917733110715825e-05,
      "loss": 1.0255,
      "num_input_tokens_seen": 13527272,
      "step": 23325
    },
    {
      "epoch": 3.4748287161155793,
      "grad_norm": 14.4375,
      "learning_rate": 4.9176504177973105e-05,
      "loss": 1.2099,
      "num_input_tokens_seen": 13530440,
      "step": 23330
    },
    {
      "epoch": 3.4755734286565385,
      "grad_norm": 14.25,
      "learning_rate": 4.91756768403505e-05,
      "loss": 1.2122,
      "num_input_tokens_seen": 13533256,
      "step": 23335
    },
    {
      "epoch": 3.4763181411974977,
      "grad_norm": 7.84375,
      "learning_rate": 4.9174849094304396e-05,
      "loss": 1.4961,
      "num_input_tokens_seen": 13536136,
      "step": 23340
    },
    {
      "epoch": 3.477062853738457,
      "grad_norm": 10.25,
      "learning_rate": 4.91740209398488e-05,
      "loss": 1.0034,
      "num_input_tokens_seen": 13538920,
      "step": 23345
    },
    {
      "epoch": 3.477807566279416,
      "grad_norm": 11.1875,
      "learning_rate": 4.917319237699768e-05,
      "loss": 1.2944,
      "num_input_tokens_seen": 13541896,
      "step": 23350
    },
    {
      "epoch": 3.4785522788203753,
      "grad_norm": 11.375,
      "learning_rate": 4.9172363405765044e-05,
      "loss": 1.4694,
      "num_input_tokens_seen": 13544552,
      "step": 23355
    },
    {
      "epoch": 3.4792969913613345,
      "grad_norm": 14.5625,
      "learning_rate": 4.91715340261649e-05,
      "loss": 1.1855,
      "num_input_tokens_seen": 13547464,
      "step": 23360
    },
    {
      "epoch": 3.4800417039022937,
      "grad_norm": 11.0,
      "learning_rate": 4.917070423821125e-05,
      "loss": 0.9783,
      "num_input_tokens_seen": 13550216,
      "step": 23365
    },
    {
      "epoch": 3.480786416443253,
      "grad_norm": 5.28125,
      "learning_rate": 4.9169874041918116e-05,
      "loss": 1.267,
      "num_input_tokens_seen": 13553192,
      "step": 23370
    },
    {
      "epoch": 3.481531128984212,
      "grad_norm": 8.0,
      "learning_rate": 4.916904343729954e-05,
      "loss": 1.1501,
      "num_input_tokens_seen": 13555912,
      "step": 23375
    },
    {
      "epoch": 3.4822758415251713,
      "grad_norm": 9.9375,
      "learning_rate": 4.916821242436952e-05,
      "loss": 0.8135,
      "num_input_tokens_seen": 13558664,
      "step": 23380
    },
    {
      "epoch": 3.4830205540661305,
      "grad_norm": 10.3125,
      "learning_rate": 4.916738100314213e-05,
      "loss": 1.2579,
      "num_input_tokens_seen": 13561512,
      "step": 23385
    },
    {
      "epoch": 3.4837652666070897,
      "grad_norm": 9.5,
      "learning_rate": 4.916654917363139e-05,
      "loss": 1.1446,
      "num_input_tokens_seen": 13564296,
      "step": 23390
    },
    {
      "epoch": 3.484509979148049,
      "grad_norm": 5.25,
      "learning_rate": 4.916571693585137e-05,
      "loss": 1.0766,
      "num_input_tokens_seen": 13567144,
      "step": 23395
    },
    {
      "epoch": 3.485254691689008,
      "grad_norm": 11.6875,
      "learning_rate": 4.9164884289816115e-05,
      "loss": 0.8813,
      "num_input_tokens_seen": 13569896,
      "step": 23400
    },
    {
      "epoch": 3.4859994042299673,
      "grad_norm": 14.125,
      "learning_rate": 4.916405123553971e-05,
      "loss": 1.0741,
      "num_input_tokens_seen": 13572840,
      "step": 23405
    },
    {
      "epoch": 3.4867441167709265,
      "grad_norm": 16.25,
      "learning_rate": 4.9163217773036214e-05,
      "loss": 1.2607,
      "num_input_tokens_seen": 13575752,
      "step": 23410
    },
    {
      "epoch": 3.4874888293118858,
      "grad_norm": 15.0,
      "learning_rate": 4.916238390231971e-05,
      "loss": 0.9304,
      "num_input_tokens_seen": 13578600,
      "step": 23415
    },
    {
      "epoch": 3.488233541852845,
      "grad_norm": 12.5625,
      "learning_rate": 4.916154962340429e-05,
      "loss": 1.0109,
      "num_input_tokens_seen": 13581288,
      "step": 23420
    },
    {
      "epoch": 3.488978254393804,
      "grad_norm": 5.125,
      "learning_rate": 4.916071493630405e-05,
      "loss": 0.7706,
      "num_input_tokens_seen": 13584072,
      "step": 23425
    },
    {
      "epoch": 3.4897229669347634,
      "grad_norm": 9.125,
      "learning_rate": 4.915987984103309e-05,
      "loss": 1.1234,
      "num_input_tokens_seen": 13587080,
      "step": 23430
    },
    {
      "epoch": 3.4904676794757226,
      "grad_norm": 9.3125,
      "learning_rate": 4.9159044337605495e-05,
      "loss": 1.1441,
      "num_input_tokens_seen": 13589896,
      "step": 23435
    },
    {
      "epoch": 3.4912123920166813,
      "grad_norm": 13.8125,
      "learning_rate": 4.915820842603542e-05,
      "loss": 1.0812,
      "num_input_tokens_seen": 13592712,
      "step": 23440
    },
    {
      "epoch": 3.491957104557641,
      "grad_norm": 10.375,
      "learning_rate": 4.9157372106336965e-05,
      "loss": 1.0266,
      "num_input_tokens_seen": 13595752,
      "step": 23445
    },
    {
      "epoch": 3.4927018170985997,
      "grad_norm": 10.5625,
      "learning_rate": 4.915653537852425e-05,
      "loss": 1.2327,
      "num_input_tokens_seen": 13598312,
      "step": 23450
    },
    {
      "epoch": 3.4934465296395594,
      "grad_norm": 8.0,
      "learning_rate": 4.915569824261143e-05,
      "loss": 1.0803,
      "num_input_tokens_seen": 13601128,
      "step": 23455
    },
    {
      "epoch": 3.494191242180518,
      "grad_norm": 12.875,
      "learning_rate": 4.915486069861264e-05,
      "loss": 1.3793,
      "num_input_tokens_seen": 13603944,
      "step": 23460
    },
    {
      "epoch": 3.4949359547214773,
      "grad_norm": 8.375,
      "learning_rate": 4.915402274654202e-05,
      "loss": 0.867,
      "num_input_tokens_seen": 13606568,
      "step": 23465
    },
    {
      "epoch": 3.4956806672624365,
      "grad_norm": 9.5,
      "learning_rate": 4.915318438641374e-05,
      "loss": 1.4123,
      "num_input_tokens_seen": 13609608,
      "step": 23470
    },
    {
      "epoch": 3.4964253798033957,
      "grad_norm": 13.75,
      "learning_rate": 4.915234561824196e-05,
      "loss": 0.8482,
      "num_input_tokens_seen": 13612552,
      "step": 23475
    },
    {
      "epoch": 3.497170092344355,
      "grad_norm": 12.0,
      "learning_rate": 4.915150644204084e-05,
      "loss": 1.1889,
      "num_input_tokens_seen": 13615560,
      "step": 23480
    },
    {
      "epoch": 3.497914804885314,
      "grad_norm": 5.90625,
      "learning_rate": 4.915066685782457e-05,
      "loss": 1.0083,
      "num_input_tokens_seen": 13618312,
      "step": 23485
    },
    {
      "epoch": 3.4986595174262733,
      "grad_norm": 10.375,
      "learning_rate": 4.914982686560733e-05,
      "loss": 0.9941,
      "num_input_tokens_seen": 13621064,
      "step": 23490
    },
    {
      "epoch": 3.4994042299672325,
      "grad_norm": 10.6875,
      "learning_rate": 4.914898646540331e-05,
      "loss": 0.8468,
      "num_input_tokens_seen": 13624136,
      "step": 23495
    },
    {
      "epoch": 3.5001489425081918,
      "grad_norm": 19.0,
      "learning_rate": 4.914814565722671e-05,
      "loss": 1.1898,
      "num_input_tokens_seen": 13627112,
      "step": 23500
    },
    {
      "epoch": 3.500893655049151,
      "grad_norm": 7.875,
      "learning_rate": 4.914730444109173e-05,
      "loss": 1.5901,
      "num_input_tokens_seen": 13629928,
      "step": 23505
    },
    {
      "epoch": 3.50163836759011,
      "grad_norm": 10.5625,
      "learning_rate": 4.9146462817012586e-05,
      "loss": 1.1876,
      "num_input_tokens_seen": 13632904,
      "step": 23510
    },
    {
      "epoch": 3.5023830801310694,
      "grad_norm": 10.5625,
      "learning_rate": 4.9145620785003485e-05,
      "loss": 1.029,
      "num_input_tokens_seen": 13635656,
      "step": 23515
    },
    {
      "epoch": 3.5031277926720286,
      "grad_norm": 15.5625,
      "learning_rate": 4.9144778345078665e-05,
      "loss": 1.1089,
      "num_input_tokens_seen": 13638440,
      "step": 23520
    },
    {
      "epoch": 3.5038725052129878,
      "grad_norm": 7.21875,
      "learning_rate": 4.914393549725236e-05,
      "loss": 1.0673,
      "num_input_tokens_seen": 13641352,
      "step": 23525
    },
    {
      "epoch": 3.504617217753947,
      "grad_norm": 10.0625,
      "learning_rate": 4.91430922415388e-05,
      "loss": 1.3351,
      "num_input_tokens_seen": 13644264,
      "step": 23530
    },
    {
      "epoch": 3.505361930294906,
      "grad_norm": 10.125,
      "learning_rate": 4.914224857795224e-05,
      "loss": 1.5315,
      "num_input_tokens_seen": 13647080,
      "step": 23535
    },
    {
      "epoch": 3.5061066428358654,
      "grad_norm": 8.4375,
      "learning_rate": 4.914140450650692e-05,
      "loss": 1.1339,
      "num_input_tokens_seen": 13650216,
      "step": 23540
    },
    {
      "epoch": 3.5068513553768246,
      "grad_norm": 7.59375,
      "learning_rate": 4.9140560027217106e-05,
      "loss": 1.0383,
      "num_input_tokens_seen": 13653352,
      "step": 23545
    },
    {
      "epoch": 3.5075960679177838,
      "grad_norm": 7.5,
      "learning_rate": 4.9139715140097075e-05,
      "loss": 0.9886,
      "num_input_tokens_seen": 13656072,
      "step": 23550
    },
    {
      "epoch": 3.508340780458743,
      "grad_norm": 11.0,
      "learning_rate": 4.9138869845161086e-05,
      "loss": 1.0976,
      "num_input_tokens_seen": 13658824,
      "step": 23555
    },
    {
      "epoch": 3.509085492999702,
      "grad_norm": 6.46875,
      "learning_rate": 4.913802414242342e-05,
      "loss": 0.8315,
      "num_input_tokens_seen": 13661992,
      "step": 23560
    },
    {
      "epoch": 3.5098302055406614,
      "grad_norm": 12.0,
      "learning_rate": 4.913717803189838e-05,
      "loss": 1.1921,
      "num_input_tokens_seen": 13664904,
      "step": 23565
    },
    {
      "epoch": 3.5105749180816206,
      "grad_norm": 4.1875,
      "learning_rate": 4.913633151360024e-05,
      "loss": 1.0031,
      "num_input_tokens_seen": 13667688,
      "step": 23570
    },
    {
      "epoch": 3.51131963062258,
      "grad_norm": 8.3125,
      "learning_rate": 4.913548458754331e-05,
      "loss": 0.801,
      "num_input_tokens_seen": 13670440,
      "step": 23575
    },
    {
      "epoch": 3.512064343163539,
      "grad_norm": 8.375,
      "learning_rate": 4.91346372537419e-05,
      "loss": 0.9411,
      "num_input_tokens_seen": 13673192,
      "step": 23580
    },
    {
      "epoch": 3.512809055704498,
      "grad_norm": 10.25,
      "learning_rate": 4.913378951221033e-05,
      "loss": 1.1539,
      "num_input_tokens_seen": 13675976,
      "step": 23585
    },
    {
      "epoch": 3.5135537682454574,
      "grad_norm": 11.375,
      "learning_rate": 4.9132941362962905e-05,
      "loss": 1.2077,
      "num_input_tokens_seen": 13678728,
      "step": 23590
    },
    {
      "epoch": 3.5142984807864166,
      "grad_norm": 6.59375,
      "learning_rate": 4.913209280601396e-05,
      "loss": 0.8766,
      "num_input_tokens_seen": 13681544,
      "step": 23595
    },
    {
      "epoch": 3.515043193327376,
      "grad_norm": 4.75,
      "learning_rate": 4.913124384137784e-05,
      "loss": 1.091,
      "num_input_tokens_seen": 13684392,
      "step": 23600
    },
    {
      "epoch": 3.5157879058683346,
      "grad_norm": 11.3125,
      "learning_rate": 4.9130394469068886e-05,
      "loss": 1.1082,
      "num_input_tokens_seen": 13687144,
      "step": 23605
    },
    {
      "epoch": 3.516532618409294,
      "grad_norm": 19.0,
      "learning_rate": 4.9129544689101437e-05,
      "loss": 1.0389,
      "num_input_tokens_seen": 13690056,
      "step": 23610
    },
    {
      "epoch": 3.517277330950253,
      "grad_norm": 7.875,
      "learning_rate": 4.912869450148986e-05,
      "loss": 1.2795,
      "num_input_tokens_seen": 13692872,
      "step": 23615
    },
    {
      "epoch": 3.5180220434912126,
      "grad_norm": 13.4375,
      "learning_rate": 4.9127843906248504e-05,
      "loss": 1.071,
      "num_input_tokens_seen": 13695720,
      "step": 23620
    },
    {
      "epoch": 3.5187667560321714,
      "grad_norm": 13.3125,
      "learning_rate": 4.912699290339175e-05,
      "loss": 1.069,
      "num_input_tokens_seen": 13698728,
      "step": 23625
    },
    {
      "epoch": 3.519511468573131,
      "grad_norm": 14.0625,
      "learning_rate": 4.912614149293398e-05,
      "loss": 1.3284,
      "num_input_tokens_seen": 13701288,
      "step": 23630
    },
    {
      "epoch": 3.5202561811140898,
      "grad_norm": 6.9375,
      "learning_rate": 4.9125289674889566e-05,
      "loss": 1.0197,
      "num_input_tokens_seen": 13704200,
      "step": 23635
    },
    {
      "epoch": 3.5210008936550494,
      "grad_norm": 6.75,
      "learning_rate": 4.91244374492729e-05,
      "loss": 1.2334,
      "num_input_tokens_seen": 13707208,
      "step": 23640
    },
    {
      "epoch": 3.521745606196008,
      "grad_norm": 4.75,
      "learning_rate": 4.912358481609838e-05,
      "loss": 0.8878,
      "num_input_tokens_seen": 13710248,
      "step": 23645
    },
    {
      "epoch": 3.5224903187369674,
      "grad_norm": 13.125,
      "learning_rate": 4.912273177538041e-05,
      "loss": 1.1051,
      "num_input_tokens_seen": 13713416,
      "step": 23650
    },
    {
      "epoch": 3.5232350312779266,
      "grad_norm": 10.4375,
      "learning_rate": 4.912187832713342e-05,
      "loss": 1.0796,
      "num_input_tokens_seen": 13716104,
      "step": 23655
    },
    {
      "epoch": 3.523979743818886,
      "grad_norm": 8.9375,
      "learning_rate": 4.91210244713718e-05,
      "loss": 0.9221,
      "num_input_tokens_seen": 13718984,
      "step": 23660
    },
    {
      "epoch": 3.524724456359845,
      "grad_norm": 7.46875,
      "learning_rate": 4.912017020810999e-05,
      "loss": 1.1469,
      "num_input_tokens_seen": 13721864,
      "step": 23665
    },
    {
      "epoch": 3.525469168900804,
      "grad_norm": 14.3125,
      "learning_rate": 4.911931553736242e-05,
      "loss": 1.0498,
      "num_input_tokens_seen": 13724648,
      "step": 23670
    },
    {
      "epoch": 3.5262138814417634,
      "grad_norm": 10.1875,
      "learning_rate": 4.9118460459143524e-05,
      "loss": 0.9142,
      "num_input_tokens_seen": 13728104,
      "step": 23675
    },
    {
      "epoch": 3.5269585939827226,
      "grad_norm": 13.6875,
      "learning_rate": 4.9117604973467756e-05,
      "loss": 1.1054,
      "num_input_tokens_seen": 13731080,
      "step": 23680
    },
    {
      "epoch": 3.527703306523682,
      "grad_norm": 10.1875,
      "learning_rate": 4.9116749080349556e-05,
      "loss": 1.257,
      "num_input_tokens_seen": 13733800,
      "step": 23685
    },
    {
      "epoch": 3.528448019064641,
      "grad_norm": 9.875,
      "learning_rate": 4.911589277980339e-05,
      "loss": 1.1189,
      "num_input_tokens_seen": 13736776,
      "step": 23690
    },
    {
      "epoch": 3.5291927316056,
      "grad_norm": 10.6875,
      "learning_rate": 4.911503607184375e-05,
      "loss": 1.0026,
      "num_input_tokens_seen": 13739912,
      "step": 23695
    },
    {
      "epoch": 3.5299374441465594,
      "grad_norm": 8.5,
      "learning_rate": 4.911417895648506e-05,
      "loss": 0.8261,
      "num_input_tokens_seen": 13743048,
      "step": 23700
    },
    {
      "epoch": 3.5306821566875186,
      "grad_norm": 10.8125,
      "learning_rate": 4.9113321433741835e-05,
      "loss": 0.9092,
      "num_input_tokens_seen": 13746120,
      "step": 23705
    },
    {
      "epoch": 3.531426869228478,
      "grad_norm": 8.4375,
      "learning_rate": 4.9112463503628545e-05,
      "loss": 1.084,
      "num_input_tokens_seen": 13748808,
      "step": 23710
    },
    {
      "epoch": 3.532171581769437,
      "grad_norm": 6.96875,
      "learning_rate": 4.91116051661597e-05,
      "loss": 1.2157,
      "num_input_tokens_seen": 13751720,
      "step": 23715
    },
    {
      "epoch": 3.532916294310396,
      "grad_norm": 9.0,
      "learning_rate": 4.911074642134979e-05,
      "loss": 0.9328,
      "num_input_tokens_seen": 13754376,
      "step": 23720
    },
    {
      "epoch": 3.5336610068513554,
      "grad_norm": 5.8125,
      "learning_rate": 4.9109887269213315e-05,
      "loss": 1.0073,
      "num_input_tokens_seen": 13757544,
      "step": 23725
    },
    {
      "epoch": 3.5344057193923146,
      "grad_norm": 9.5,
      "learning_rate": 4.910902770976481e-05,
      "loss": 1.202,
      "num_input_tokens_seen": 13760296,
      "step": 23730
    },
    {
      "epoch": 3.535150431933274,
      "grad_norm": 10.0,
      "learning_rate": 4.910816774301878e-05,
      "loss": 1.0273,
      "num_input_tokens_seen": 13763144,
      "step": 23735
    },
    {
      "epoch": 3.535895144474233,
      "grad_norm": 12.4375,
      "learning_rate": 4.910730736898976e-05,
      "loss": 0.8445,
      "num_input_tokens_seen": 13766024,
      "step": 23740
    },
    {
      "epoch": 3.5366398570151922,
      "grad_norm": 7.03125,
      "learning_rate": 4.9106446587692276e-05,
      "loss": 1.1846,
      "num_input_tokens_seen": 13768872,
      "step": 23745
    },
    {
      "epoch": 3.5373845695561514,
      "grad_norm": 9.75,
      "learning_rate": 4.910558539914088e-05,
      "loss": 0.961,
      "num_input_tokens_seen": 13771784,
      "step": 23750
    },
    {
      "epoch": 3.5381292820971106,
      "grad_norm": 11.125,
      "learning_rate": 4.910472380335013e-05,
      "loss": 1.0785,
      "num_input_tokens_seen": 13774760,
      "step": 23755
    },
    {
      "epoch": 3.53887399463807,
      "grad_norm": 11.5625,
      "learning_rate": 4.9103861800334567e-05,
      "loss": 0.8611,
      "num_input_tokens_seen": 13777832,
      "step": 23760
    },
    {
      "epoch": 3.539618707179029,
      "grad_norm": 9.9375,
      "learning_rate": 4.9102999390108753e-05,
      "loss": 0.9443,
      "num_input_tokens_seen": 13780616,
      "step": 23765
    },
    {
      "epoch": 3.5403634197199882,
      "grad_norm": 8.875,
      "learning_rate": 4.910213657268726e-05,
      "loss": 0.8914,
      "num_input_tokens_seen": 13783272,
      "step": 23770
    },
    {
      "epoch": 3.5411081322609474,
      "grad_norm": 11.5625,
      "learning_rate": 4.910127334808466e-05,
      "loss": 1.2436,
      "num_input_tokens_seen": 13785960,
      "step": 23775
    },
    {
      "epoch": 3.541852844801906,
      "grad_norm": 10.0,
      "learning_rate": 4.9100409716315556e-05,
      "loss": 0.9759,
      "num_input_tokens_seen": 13788680,
      "step": 23780
    },
    {
      "epoch": 3.542597557342866,
      "grad_norm": 6.4375,
      "learning_rate": 4.909954567739452e-05,
      "loss": 0.9406,
      "num_input_tokens_seen": 13791592,
      "step": 23785
    },
    {
      "epoch": 3.5433422698838246,
      "grad_norm": 9.25,
      "learning_rate": 4.909868123133615e-05,
      "loss": 0.9594,
      "num_input_tokens_seen": 13794344,
      "step": 23790
    },
    {
      "epoch": 3.5440869824247843,
      "grad_norm": 9.4375,
      "learning_rate": 4.909781637815506e-05,
      "loss": 0.8783,
      "num_input_tokens_seen": 13797288,
      "step": 23795
    },
    {
      "epoch": 3.544831694965743,
      "grad_norm": 18.5,
      "learning_rate": 4.909695111786584e-05,
      "loss": 1.0134,
      "num_input_tokens_seen": 13800168,
      "step": 23800
    },
    {
      "epoch": 3.5455764075067027,
      "grad_norm": 12.625,
      "learning_rate": 4.9096085450483134e-05,
      "loss": 1.1485,
      "num_input_tokens_seen": 13802984,
      "step": 23805
    },
    {
      "epoch": 3.5463211200476614,
      "grad_norm": 7.3125,
      "learning_rate": 4.909521937602155e-05,
      "loss": 0.8699,
      "num_input_tokens_seen": 13805928,
      "step": 23810
    },
    {
      "epoch": 3.5470658325886206,
      "grad_norm": 9.125,
      "learning_rate": 4.909435289449573e-05,
      "loss": 1.1521,
      "num_input_tokens_seen": 13808936,
      "step": 23815
    },
    {
      "epoch": 3.54781054512958,
      "grad_norm": 12.875,
      "learning_rate": 4.90934860059203e-05,
      "loss": 1.1626,
      "num_input_tokens_seen": 13811976,
      "step": 23820
    },
    {
      "epoch": 3.548555257670539,
      "grad_norm": 14.8125,
      "learning_rate": 4.909261871030991e-05,
      "loss": 1.1219,
      "num_input_tokens_seen": 13814632,
      "step": 23825
    },
    {
      "epoch": 3.5492999702114982,
      "grad_norm": 8.25,
      "learning_rate": 4.9091751007679224e-05,
      "loss": 1.1947,
      "num_input_tokens_seen": 13817704,
      "step": 23830
    },
    {
      "epoch": 3.5500446827524574,
      "grad_norm": 11.0,
      "learning_rate": 4.9090882898042876e-05,
      "loss": 1.1036,
      "num_input_tokens_seen": 13821000,
      "step": 23835
    },
    {
      "epoch": 3.5507893952934166,
      "grad_norm": 7.5,
      "learning_rate": 4.909001438141556e-05,
      "loss": 0.7595,
      "num_input_tokens_seen": 13823592,
      "step": 23840
    },
    {
      "epoch": 3.551534107834376,
      "grad_norm": 17.375,
      "learning_rate": 4.908914545781192e-05,
      "loss": 1.1964,
      "num_input_tokens_seen": 13826280,
      "step": 23845
    },
    {
      "epoch": 3.552278820375335,
      "grad_norm": 16.875,
      "learning_rate": 4.9088276127246666e-05,
      "loss": 1.068,
      "num_input_tokens_seen": 13828840,
      "step": 23850
    },
    {
      "epoch": 3.5530235329162942,
      "grad_norm": 23.875,
      "learning_rate": 4.9087406389734465e-05,
      "loss": 1.1343,
      "num_input_tokens_seen": 13831592,
      "step": 23855
    },
    {
      "epoch": 3.5537682454572534,
      "grad_norm": 14.5,
      "learning_rate": 4.908653624529001e-05,
      "loss": 1.271,
      "num_input_tokens_seen": 13834344,
      "step": 23860
    },
    {
      "epoch": 3.5545129579982127,
      "grad_norm": 6.3125,
      "learning_rate": 4.908566569392801e-05,
      "loss": 1.2483,
      "num_input_tokens_seen": 13837384,
      "step": 23865
    },
    {
      "epoch": 3.555257670539172,
      "grad_norm": 7.0625,
      "learning_rate": 4.908479473566316e-05,
      "loss": 0.7671,
      "num_input_tokens_seen": 13840072,
      "step": 23870
    },
    {
      "epoch": 3.556002383080131,
      "grad_norm": 20.25,
      "learning_rate": 4.9083923370510184e-05,
      "loss": 1.4316,
      "num_input_tokens_seen": 13842728,
      "step": 23875
    },
    {
      "epoch": 3.5567470956210903,
      "grad_norm": 9.1875,
      "learning_rate": 4.908305159848381e-05,
      "loss": 1.1685,
      "num_input_tokens_seen": 13845864,
      "step": 23880
    },
    {
      "epoch": 3.5574918081620495,
      "grad_norm": 9.875,
      "learning_rate": 4.908217941959875e-05,
      "loss": 1.0976,
      "num_input_tokens_seen": 13848648,
      "step": 23885
    },
    {
      "epoch": 3.5582365207030087,
      "grad_norm": 16.75,
      "learning_rate": 4.908130683386974e-05,
      "loss": 1.1416,
      "num_input_tokens_seen": 13851688,
      "step": 23890
    },
    {
      "epoch": 3.558981233243968,
      "grad_norm": 7.125,
      "learning_rate": 4.9080433841311526e-05,
      "loss": 1.1158,
      "num_input_tokens_seen": 13854664,
      "step": 23895
    },
    {
      "epoch": 3.559725945784927,
      "grad_norm": 14.0,
      "learning_rate": 4.9079560441938865e-05,
      "loss": 1.0005,
      "num_input_tokens_seen": 13857512,
      "step": 23900
    },
    {
      "epoch": 3.5604706583258863,
      "grad_norm": 13.125,
      "learning_rate": 4.90786866357665e-05,
      "loss": 1.4393,
      "num_input_tokens_seen": 13860776,
      "step": 23905
    },
    {
      "epoch": 3.5612153708668455,
      "grad_norm": 10.125,
      "learning_rate": 4.90778124228092e-05,
      "loss": 0.9472,
      "num_input_tokens_seen": 13863816,
      "step": 23910
    },
    {
      "epoch": 3.5619600834078047,
      "grad_norm": 11.6875,
      "learning_rate": 4.907693780308172e-05,
      "loss": 1.3562,
      "num_input_tokens_seen": 13866568,
      "step": 23915
    },
    {
      "epoch": 3.562704795948764,
      "grad_norm": 6.375,
      "learning_rate": 4.907606277659885e-05,
      "loss": 0.8591,
      "num_input_tokens_seen": 13869384,
      "step": 23920
    },
    {
      "epoch": 3.563449508489723,
      "grad_norm": 5.875,
      "learning_rate": 4.907518734337538e-05,
      "loss": 1.3406,
      "num_input_tokens_seen": 13871912,
      "step": 23925
    },
    {
      "epoch": 3.5641942210306823,
      "grad_norm": 13.5,
      "learning_rate": 4.907431150342608e-05,
      "loss": 0.9404,
      "num_input_tokens_seen": 13874888,
      "step": 23930
    },
    {
      "epoch": 3.5649389335716415,
      "grad_norm": 6.0,
      "learning_rate": 4.907343525676575e-05,
      "loss": 0.947,
      "num_input_tokens_seen": 13877960,
      "step": 23935
    },
    {
      "epoch": 3.5656836461126007,
      "grad_norm": 12.9375,
      "learning_rate": 4.9072558603409216e-05,
      "loss": 1.3509,
      "num_input_tokens_seen": 13880744,
      "step": 23940
    },
    {
      "epoch": 3.5664283586535594,
      "grad_norm": 9.6875,
      "learning_rate": 4.907168154337125e-05,
      "loss": 0.8382,
      "num_input_tokens_seen": 13883624,
      "step": 23945
    },
    {
      "epoch": 3.567173071194519,
      "grad_norm": 8.5,
      "learning_rate": 4.90708040766667e-05,
      "loss": 1.2026,
      "num_input_tokens_seen": 13887080,
      "step": 23950
    },
    {
      "epoch": 3.567917783735478,
      "grad_norm": 9.875,
      "learning_rate": 4.906992620331038e-05,
      "loss": 1.057,
      "num_input_tokens_seen": 13890184,
      "step": 23955
    },
    {
      "epoch": 3.5686624962764375,
      "grad_norm": 11.0,
      "learning_rate": 4.906904792331712e-05,
      "loss": 1.3202,
      "num_input_tokens_seen": 13893032,
      "step": 23960
    },
    {
      "epoch": 3.5694072088173963,
      "grad_norm": 12.875,
      "learning_rate": 4.906816923670176e-05,
      "loss": 0.9146,
      "num_input_tokens_seen": 13896488,
      "step": 23965
    },
    {
      "epoch": 3.570151921358356,
      "grad_norm": 7.375,
      "learning_rate": 4.906729014347914e-05,
      "loss": 0.9023,
      "num_input_tokens_seen": 13899272,
      "step": 23970
    },
    {
      "epoch": 3.5708966338993147,
      "grad_norm": 12.9375,
      "learning_rate": 4.9066410643664113e-05,
      "loss": 1.1285,
      "num_input_tokens_seen": 13902088,
      "step": 23975
    },
    {
      "epoch": 3.5716413464402743,
      "grad_norm": 5.5,
      "learning_rate": 4.906553073727154e-05,
      "loss": 0.9994,
      "num_input_tokens_seen": 13904744,
      "step": 23980
    },
    {
      "epoch": 3.572386058981233,
      "grad_norm": 10.8125,
      "learning_rate": 4.9064650424316284e-05,
      "loss": 0.7971,
      "num_input_tokens_seen": 13907528,
      "step": 23985
    },
    {
      "epoch": 3.5731307715221923,
      "grad_norm": 7.53125,
      "learning_rate": 4.906376970481321e-05,
      "loss": 0.8707,
      "num_input_tokens_seen": 13910632,
      "step": 23990
    },
    {
      "epoch": 3.5738754840631515,
      "grad_norm": 10.6875,
      "learning_rate": 4.9062888578777214e-05,
      "loss": 1.39,
      "num_input_tokens_seen": 13913480,
      "step": 23995
    },
    {
      "epoch": 3.5746201966041107,
      "grad_norm": 14.8125,
      "learning_rate": 4.906200704622317e-05,
      "loss": 1.1142,
      "num_input_tokens_seen": 13916456,
      "step": 24000
    },
    {
      "epoch": 3.57536490914507,
      "grad_norm": 5.5625,
      "learning_rate": 4.906112510716597e-05,
      "loss": 1.0901,
      "num_input_tokens_seen": 13919240,
      "step": 24005
    },
    {
      "epoch": 3.576109621686029,
      "grad_norm": 7.59375,
      "learning_rate": 4.906024276162052e-05,
      "loss": 0.9633,
      "num_input_tokens_seen": 13922248,
      "step": 24010
    },
    {
      "epoch": 3.5768543342269883,
      "grad_norm": 13.75,
      "learning_rate": 4.905936000960172e-05,
      "loss": 0.9472,
      "num_input_tokens_seen": 13924968,
      "step": 24015
    },
    {
      "epoch": 3.5775990467679475,
      "grad_norm": 16.625,
      "learning_rate": 4.905847685112448e-05,
      "loss": 0.898,
      "num_input_tokens_seen": 13927816,
      "step": 24020
    },
    {
      "epoch": 3.5783437593089067,
      "grad_norm": 11.9375,
      "learning_rate": 4.905759328620373e-05,
      "loss": 1.3852,
      "num_input_tokens_seen": 13930632,
      "step": 24025
    },
    {
      "epoch": 3.579088471849866,
      "grad_norm": 8.8125,
      "learning_rate": 4.90567093148544e-05,
      "loss": 1.3585,
      "num_input_tokens_seen": 13933480,
      "step": 24030
    },
    {
      "epoch": 3.579833184390825,
      "grad_norm": 13.25,
      "learning_rate": 4.9055824937091406e-05,
      "loss": 0.8605,
      "num_input_tokens_seen": 13936232,
      "step": 24035
    },
    {
      "epoch": 3.5805778969317843,
      "grad_norm": 10.25,
      "learning_rate": 4.9054940152929704e-05,
      "loss": 0.8544,
      "num_input_tokens_seen": 13939368,
      "step": 24040
    },
    {
      "epoch": 3.5813226094727435,
      "grad_norm": 7.34375,
      "learning_rate": 4.9054054962384235e-05,
      "loss": 1.114,
      "num_input_tokens_seen": 13942152,
      "step": 24045
    },
    {
      "epoch": 3.5820673220137027,
      "grad_norm": 14.0,
      "learning_rate": 4.905316936546995e-05,
      "loss": 1.3367,
      "num_input_tokens_seen": 13945032,
      "step": 24050
    },
    {
      "epoch": 3.582812034554662,
      "grad_norm": 15.875,
      "learning_rate": 4.9052283362201823e-05,
      "loss": 0.7851,
      "num_input_tokens_seen": 13947816,
      "step": 24055
    },
    {
      "epoch": 3.583556747095621,
      "grad_norm": 11.75,
      "learning_rate": 4.9051396952594806e-05,
      "loss": 1.1855,
      "num_input_tokens_seen": 13950760,
      "step": 24060
    },
    {
      "epoch": 3.5843014596365803,
      "grad_norm": 12.5625,
      "learning_rate": 4.905051013666389e-05,
      "loss": 1.4873,
      "num_input_tokens_seen": 13953608,
      "step": 24065
    },
    {
      "epoch": 3.5850461721775395,
      "grad_norm": 17.25,
      "learning_rate": 4.904962291442404e-05,
      "loss": 1.2012,
      "num_input_tokens_seen": 13956488,
      "step": 24070
    },
    {
      "epoch": 3.5857908847184987,
      "grad_norm": 6.90625,
      "learning_rate": 4.904873528589027e-05,
      "loss": 0.722,
      "num_input_tokens_seen": 13959080,
      "step": 24075
    },
    {
      "epoch": 3.586535597259458,
      "grad_norm": 9.5,
      "learning_rate": 4.9047847251077544e-05,
      "loss": 1.2649,
      "num_input_tokens_seen": 13961800,
      "step": 24080
    },
    {
      "epoch": 3.587280309800417,
      "grad_norm": 13.0625,
      "learning_rate": 4.904695881000089e-05,
      "loss": 1.0891,
      "num_input_tokens_seen": 13964648,
      "step": 24085
    },
    {
      "epoch": 3.5880250223413763,
      "grad_norm": 17.75,
      "learning_rate": 4.90460699626753e-05,
      "loss": 1.0457,
      "num_input_tokens_seen": 13967528,
      "step": 24090
    },
    {
      "epoch": 3.5887697348823355,
      "grad_norm": 5.0625,
      "learning_rate": 4.90451807091158e-05,
      "loss": 1.0219,
      "num_input_tokens_seen": 13970696,
      "step": 24095
    },
    {
      "epoch": 3.5895144474232947,
      "grad_norm": 7.53125,
      "learning_rate": 4.904429104933741e-05,
      "loss": 0.9798,
      "num_input_tokens_seen": 13973736,
      "step": 24100
    },
    {
      "epoch": 3.590259159964254,
      "grad_norm": 11.8125,
      "learning_rate": 4.904340098335516e-05,
      "loss": 1.1215,
      "num_input_tokens_seen": 13976840,
      "step": 24105
    },
    {
      "epoch": 3.591003872505213,
      "grad_norm": 4.65625,
      "learning_rate": 4.904251051118408e-05,
      "loss": 0.8536,
      "num_input_tokens_seen": 13979688,
      "step": 24110
    },
    {
      "epoch": 3.5917485850461723,
      "grad_norm": 11.3125,
      "learning_rate": 4.904161963283923e-05,
      "loss": 1.0158,
      "num_input_tokens_seen": 13982504,
      "step": 24115
    },
    {
      "epoch": 3.592493297587131,
      "grad_norm": 20.75,
      "learning_rate": 4.9040728348335655e-05,
      "loss": 1.611,
      "num_input_tokens_seen": 13985320,
      "step": 24120
    },
    {
      "epoch": 3.5932380101280907,
      "grad_norm": 9.5625,
      "learning_rate": 4.90398366576884e-05,
      "loss": 1.0915,
      "num_input_tokens_seen": 13988040,
      "step": 24125
    },
    {
      "epoch": 3.5939827226690495,
      "grad_norm": 14.0625,
      "learning_rate": 4.903894456091254e-05,
      "loss": 1.177,
      "num_input_tokens_seen": 13991176,
      "step": 24130
    },
    {
      "epoch": 3.594727435210009,
      "grad_norm": 8.6875,
      "learning_rate": 4.903805205802314e-05,
      "loss": 1.0814,
      "num_input_tokens_seen": 13994088,
      "step": 24135
    },
    {
      "epoch": 3.595472147750968,
      "grad_norm": 13.9375,
      "learning_rate": 4.903715914903529e-05,
      "loss": 1.0622,
      "num_input_tokens_seen": 13997096,
      "step": 24140
    },
    {
      "epoch": 3.5962168602919276,
      "grad_norm": 7.75,
      "learning_rate": 4.9036265833964057e-05,
      "loss": 1.2851,
      "num_input_tokens_seen": 14000072,
      "step": 24145
    },
    {
      "epoch": 3.5969615728328863,
      "grad_norm": 16.375,
      "learning_rate": 4.903537211282455e-05,
      "loss": 1.0689,
      "num_input_tokens_seen": 14002728,
      "step": 24150
    },
    {
      "epoch": 3.597706285373846,
      "grad_norm": 5.46875,
      "learning_rate": 4.9034477985631854e-05,
      "loss": 1.0235,
      "num_input_tokens_seen": 14005640,
      "step": 24155
    },
    {
      "epoch": 3.5984509979148047,
      "grad_norm": 12.75,
      "learning_rate": 4.903358345240109e-05,
      "loss": 1.1101,
      "num_input_tokens_seen": 14008488,
      "step": 24160
    },
    {
      "epoch": 3.599195710455764,
      "grad_norm": 5.46875,
      "learning_rate": 4.903268851314735e-05,
      "loss": 1.034,
      "num_input_tokens_seen": 14011432,
      "step": 24165
    },
    {
      "epoch": 3.599940422996723,
      "grad_norm": 16.75,
      "learning_rate": 4.903179316788577e-05,
      "loss": 1.0226,
      "num_input_tokens_seen": 14014280,
      "step": 24170
    },
    {
      "epoch": 3.6006851355376823,
      "grad_norm": 14.6875,
      "learning_rate": 4.903089741663146e-05,
      "loss": 1.5,
      "num_input_tokens_seen": 14017128,
      "step": 24175
    },
    {
      "epoch": 3.6014298480786415,
      "grad_norm": 7.8125,
      "learning_rate": 4.9030001259399563e-05,
      "loss": 1.1125,
      "num_input_tokens_seen": 14020200,
      "step": 24180
    },
    {
      "epoch": 3.6021745606196007,
      "grad_norm": 10.375,
      "learning_rate": 4.9029104696205225e-05,
      "loss": 1.0698,
      "num_input_tokens_seen": 14023304,
      "step": 24185
    },
    {
      "epoch": 3.60291927316056,
      "grad_norm": 9.1875,
      "learning_rate": 4.9028207727063576e-05,
      "loss": 1.1851,
      "num_input_tokens_seen": 14026120,
      "step": 24190
    },
    {
      "epoch": 3.603663985701519,
      "grad_norm": 12.6875,
      "learning_rate": 4.902731035198979e-05,
      "loss": 1.0115,
      "num_input_tokens_seen": 14029256,
      "step": 24195
    },
    {
      "epoch": 3.6044086982424783,
      "grad_norm": 7.5,
      "learning_rate": 4.902641257099901e-05,
      "loss": 0.9484,
      "num_input_tokens_seen": 14032040,
      "step": 24200
    },
    {
      "epoch": 3.6051534107834375,
      "grad_norm": 10.5625,
      "learning_rate": 4.9025514384106414e-05,
      "loss": 0.9578,
      "num_input_tokens_seen": 14034888,
      "step": 24205
    },
    {
      "epoch": 3.6058981233243967,
      "grad_norm": 8.8125,
      "learning_rate": 4.902461579132717e-05,
      "loss": 1.0869,
      "num_input_tokens_seen": 14037768,
      "step": 24210
    },
    {
      "epoch": 3.606642835865356,
      "grad_norm": 12.5,
      "learning_rate": 4.902371679267646e-05,
      "loss": 1.1232,
      "num_input_tokens_seen": 14040392,
      "step": 24215
    },
    {
      "epoch": 3.607387548406315,
      "grad_norm": 6.96875,
      "learning_rate": 4.9022817388169464e-05,
      "loss": 1.0221,
      "num_input_tokens_seen": 14043208,
      "step": 24220
    },
    {
      "epoch": 3.6081322609472744,
      "grad_norm": 12.25,
      "learning_rate": 4.9021917577821386e-05,
      "loss": 0.9075,
      "num_input_tokens_seen": 14046024,
      "step": 24225
    },
    {
      "epoch": 3.6088769734882336,
      "grad_norm": 11.3125,
      "learning_rate": 4.902101736164742e-05,
      "loss": 1.0174,
      "num_input_tokens_seen": 14048904,
      "step": 24230
    },
    {
      "epoch": 3.6096216860291928,
      "grad_norm": 8.8125,
      "learning_rate": 4.902011673966279e-05,
      "loss": 1.1455,
      "num_input_tokens_seen": 14052296,
      "step": 24235
    },
    {
      "epoch": 3.610366398570152,
      "grad_norm": 10.3125,
      "learning_rate": 4.90192157118827e-05,
      "loss": 1.0719,
      "num_input_tokens_seen": 14055016,
      "step": 24240
    },
    {
      "epoch": 3.611111111111111,
      "grad_norm": 6.65625,
      "learning_rate": 4.901831427832237e-05,
      "loss": 1.1399,
      "num_input_tokens_seen": 14057800,
      "step": 24245
    },
    {
      "epoch": 3.6118558236520704,
      "grad_norm": 12.1875,
      "learning_rate": 4.9017412438997026e-05,
      "loss": 1.1527,
      "num_input_tokens_seen": 14060680,
      "step": 24250
    },
    {
      "epoch": 3.6126005361930296,
      "grad_norm": 9.6875,
      "learning_rate": 4.901651019392191e-05,
      "loss": 0.9193,
      "num_input_tokens_seen": 14063304,
      "step": 24255
    },
    {
      "epoch": 3.6133452487339888,
      "grad_norm": 6.5625,
      "learning_rate": 4.901560754311227e-05,
      "loss": 1.2612,
      "num_input_tokens_seen": 14066408,
      "step": 24260
    },
    {
      "epoch": 3.614089961274948,
      "grad_norm": 8.75,
      "learning_rate": 4.901470448658335e-05,
      "loss": 0.911,
      "num_input_tokens_seen": 14069288,
      "step": 24265
    },
    {
      "epoch": 3.614834673815907,
      "grad_norm": 10.375,
      "learning_rate": 4.9013801024350406e-05,
      "loss": 0.6612,
      "num_input_tokens_seen": 14072104,
      "step": 24270
    },
    {
      "epoch": 3.6155793863568664,
      "grad_norm": 10.125,
      "learning_rate": 4.9012897156428694e-05,
      "loss": 0.9261,
      "num_input_tokens_seen": 14074920,
      "step": 24275
    },
    {
      "epoch": 3.6163240988978256,
      "grad_norm": 11.4375,
      "learning_rate": 4.901199288283349e-05,
      "loss": 0.9602,
      "num_input_tokens_seen": 14077800,
      "step": 24280
    },
    {
      "epoch": 3.617068811438785,
      "grad_norm": 8.8125,
      "learning_rate": 4.901108820358008e-05,
      "loss": 0.9405,
      "num_input_tokens_seen": 14080456,
      "step": 24285
    },
    {
      "epoch": 3.617813523979744,
      "grad_norm": 7.8125,
      "learning_rate": 4.901018311868373e-05,
      "loss": 1.0885,
      "num_input_tokens_seen": 14083176,
      "step": 24290
    },
    {
      "epoch": 3.6185582365207027,
      "grad_norm": 9.5,
      "learning_rate": 4.9009277628159744e-05,
      "loss": 1.2667,
      "num_input_tokens_seen": 14085832,
      "step": 24295
    },
    {
      "epoch": 3.6193029490616624,
      "grad_norm": 10.25,
      "learning_rate": 4.900837173202341e-05,
      "loss": 0.9093,
      "num_input_tokens_seen": 14088616,
      "step": 24300
    },
    {
      "epoch": 3.620047661602621,
      "grad_norm": 11.0,
      "learning_rate": 4.900746543029003e-05,
      "loss": 1.3702,
      "num_input_tokens_seen": 14091240,
      "step": 24305
    },
    {
      "epoch": 3.620792374143581,
      "grad_norm": 18.5,
      "learning_rate": 4.900655872297494e-05,
      "loss": 1.3881,
      "num_input_tokens_seen": 14094152,
      "step": 24310
    },
    {
      "epoch": 3.6215370866845396,
      "grad_norm": 10.75,
      "learning_rate": 4.900565161009343e-05,
      "loss": 1.2712,
      "num_input_tokens_seen": 14097192,
      "step": 24315
    },
    {
      "epoch": 3.622281799225499,
      "grad_norm": 12.5625,
      "learning_rate": 4.9004744091660826e-05,
      "loss": 1.1016,
      "num_input_tokens_seen": 14099912,
      "step": 24320
    },
    {
      "epoch": 3.623026511766458,
      "grad_norm": 14.25,
      "learning_rate": 4.900383616769247e-05,
      "loss": 1.1333,
      "num_input_tokens_seen": 14103112,
      "step": 24325
    },
    {
      "epoch": 3.6237712243074176,
      "grad_norm": 11.5625,
      "learning_rate": 4.900292783820371e-05,
      "loss": 1.0799,
      "num_input_tokens_seen": 14106216,
      "step": 24330
    },
    {
      "epoch": 3.6245159368483764,
      "grad_norm": 6.4375,
      "learning_rate": 4.9002019103209875e-05,
      "loss": 0.8108,
      "num_input_tokens_seen": 14109128,
      "step": 24335
    },
    {
      "epoch": 3.6252606493893356,
      "grad_norm": 13.6875,
      "learning_rate": 4.9001109962726323e-05,
      "loss": 1.32,
      "num_input_tokens_seen": 14111592,
      "step": 24340
    },
    {
      "epoch": 3.6260053619302948,
      "grad_norm": 5.75,
      "learning_rate": 4.9000200416768405e-05,
      "loss": 1.014,
      "num_input_tokens_seen": 14113992,
      "step": 24345
    },
    {
      "epoch": 3.626750074471254,
      "grad_norm": 20.375,
      "learning_rate": 4.89992904653515e-05,
      "loss": 1.1488,
      "num_input_tokens_seen": 14116776,
      "step": 24350
    },
    {
      "epoch": 3.627494787012213,
      "grad_norm": 5.375,
      "learning_rate": 4.899838010849097e-05,
      "loss": 0.969,
      "num_input_tokens_seen": 14119592,
      "step": 24355
    },
    {
      "epoch": 3.6282394995531724,
      "grad_norm": 14.5625,
      "learning_rate": 4.89974693462022e-05,
      "loss": 1.0264,
      "num_input_tokens_seen": 14122600,
      "step": 24360
    },
    {
      "epoch": 3.6289842120941316,
      "grad_norm": 9.8125,
      "learning_rate": 4.899655817850058e-05,
      "loss": 0.9071,
      "num_input_tokens_seen": 14125416,
      "step": 24365
    },
    {
      "epoch": 3.629728924635091,
      "grad_norm": 8.4375,
      "learning_rate": 4.899564660540149e-05,
      "loss": 1.2761,
      "num_input_tokens_seen": 14128424,
      "step": 24370
    },
    {
      "epoch": 3.63047363717605,
      "grad_norm": 11.1875,
      "learning_rate": 4.899473462692035e-05,
      "loss": 1.0494,
      "num_input_tokens_seen": 14131368,
      "step": 24375
    },
    {
      "epoch": 3.631218349717009,
      "grad_norm": 8.0625,
      "learning_rate": 4.899382224307255e-05,
      "loss": 1.3577,
      "num_input_tokens_seen": 14134312,
      "step": 24380
    },
    {
      "epoch": 3.6319630622579684,
      "grad_norm": 10.0625,
      "learning_rate": 4.8992909453873505e-05,
      "loss": 1.1648,
      "num_input_tokens_seen": 14137224,
      "step": 24385
    },
    {
      "epoch": 3.6327077747989276,
      "grad_norm": 8.1875,
      "learning_rate": 4.899199625933865e-05,
      "loss": 0.9281,
      "num_input_tokens_seen": 14140264,
      "step": 24390
    },
    {
      "epoch": 3.633452487339887,
      "grad_norm": 12.75,
      "learning_rate": 4.899108265948339e-05,
      "loss": 1.1557,
      "num_input_tokens_seen": 14143048,
      "step": 24395
    },
    {
      "epoch": 3.634197199880846,
      "grad_norm": 6.25,
      "learning_rate": 4.899016865432318e-05,
      "loss": 0.9423,
      "num_input_tokens_seen": 14146088,
      "step": 24400
    },
    {
      "epoch": 3.634941912421805,
      "grad_norm": 15.5,
      "learning_rate": 4.898925424387345e-05,
      "loss": 1.2145,
      "num_input_tokens_seen": 14149000,
      "step": 24405
    },
    {
      "epoch": 3.6356866249627644,
      "grad_norm": 7.0,
      "learning_rate": 4.8988339428149656e-05,
      "loss": 0.8245,
      "num_input_tokens_seen": 14151944,
      "step": 24410
    },
    {
      "epoch": 3.6364313375037236,
      "grad_norm": 9.1875,
      "learning_rate": 4.898742420716724e-05,
      "loss": 0.9519,
      "num_input_tokens_seen": 14155208,
      "step": 24415
    },
    {
      "epoch": 3.637176050044683,
      "grad_norm": 10.625,
      "learning_rate": 4.898650858094168e-05,
      "loss": 1.1838,
      "num_input_tokens_seen": 14158216,
      "step": 24420
    },
    {
      "epoch": 3.637920762585642,
      "grad_norm": 9.5,
      "learning_rate": 4.898559254948843e-05,
      "loss": 1.1729,
      "num_input_tokens_seen": 14161064,
      "step": 24425
    },
    {
      "epoch": 3.638665475126601,
      "grad_norm": 10.125,
      "learning_rate": 4.898467611282297e-05,
      "loss": 0.8263,
      "num_input_tokens_seen": 14163848,
      "step": 24430
    },
    {
      "epoch": 3.6394101876675604,
      "grad_norm": 7.1875,
      "learning_rate": 4.8983759270960796e-05,
      "loss": 0.8831,
      "num_input_tokens_seen": 14166696,
      "step": 24435
    },
    {
      "epoch": 3.6401549002085196,
      "grad_norm": 9.125,
      "learning_rate": 4.8982842023917374e-05,
      "loss": 1.1061,
      "num_input_tokens_seen": 14169512,
      "step": 24440
    },
    {
      "epoch": 3.640899612749479,
      "grad_norm": 8.25,
      "learning_rate": 4.898192437170822e-05,
      "loss": 1.0955,
      "num_input_tokens_seen": 14172296,
      "step": 24445
    },
    {
      "epoch": 3.641644325290438,
      "grad_norm": 11.125,
      "learning_rate": 4.898100631434882e-05,
      "loss": 1.1743,
      "num_input_tokens_seen": 14175016,
      "step": 24450
    },
    {
      "epoch": 3.6423890378313972,
      "grad_norm": 7.28125,
      "learning_rate": 4.898008785185469e-05,
      "loss": 0.9133,
      "num_input_tokens_seen": 14178568,
      "step": 24455
    },
    {
      "epoch": 3.6431337503723564,
      "grad_norm": 4.59375,
      "learning_rate": 4.8979168984241354e-05,
      "loss": 0.9547,
      "num_input_tokens_seen": 14181512,
      "step": 24460
    },
    {
      "epoch": 3.6438784629133156,
      "grad_norm": 6.28125,
      "learning_rate": 4.8978249711524324e-05,
      "loss": 0.9648,
      "num_input_tokens_seen": 14184712,
      "step": 24465
    },
    {
      "epoch": 3.6446231754542744,
      "grad_norm": 12.125,
      "learning_rate": 4.8977330033719147e-05,
      "loss": 1.1381,
      "num_input_tokens_seen": 14187464,
      "step": 24470
    },
    {
      "epoch": 3.645367887995234,
      "grad_norm": 10.5625,
      "learning_rate": 4.897640995084133e-05,
      "loss": 0.7837,
      "num_input_tokens_seen": 14190440,
      "step": 24475
    },
    {
      "epoch": 3.646112600536193,
      "grad_norm": 24.5,
      "learning_rate": 4.8975489462906456e-05,
      "loss": 1.1666,
      "num_input_tokens_seen": 14193384,
      "step": 24480
    },
    {
      "epoch": 3.6468573130771524,
      "grad_norm": 11.4375,
      "learning_rate": 4.897456856993004e-05,
      "loss": 0.9456,
      "num_input_tokens_seen": 14196520,
      "step": 24485
    },
    {
      "epoch": 3.647602025618111,
      "grad_norm": 6.0625,
      "learning_rate": 4.897364727192766e-05,
      "loss": 0.9273,
      "num_input_tokens_seen": 14199464,
      "step": 24490
    },
    {
      "epoch": 3.648346738159071,
      "grad_norm": 3.375,
      "learning_rate": 4.897272556891487e-05,
      "loss": 1.0641,
      "num_input_tokens_seen": 14202280,
      "step": 24495
    },
    {
      "epoch": 3.6490914507000296,
      "grad_norm": 16.375,
      "learning_rate": 4.897180346090726e-05,
      "loss": 1.3599,
      "num_input_tokens_seen": 14205288,
      "step": 24500
    },
    {
      "epoch": 3.6498361632409893,
      "grad_norm": 9.9375,
      "learning_rate": 4.8970880947920386e-05,
      "loss": 1.0969,
      "num_input_tokens_seen": 14208456,
      "step": 24505
    },
    {
      "epoch": 3.650580875781948,
      "grad_norm": 8.0625,
      "learning_rate": 4.8969958029969834e-05,
      "loss": 1.2501,
      "num_input_tokens_seen": 14211432,
      "step": 24510
    },
    {
      "epoch": 3.651325588322907,
      "grad_norm": 12.875,
      "learning_rate": 4.896903470707121e-05,
      "loss": 1.1183,
      "num_input_tokens_seen": 14214408,
      "step": 24515
    },
    {
      "epoch": 3.6520703008638664,
      "grad_norm": 15.4375,
      "learning_rate": 4.89681109792401e-05,
      "loss": 1.341,
      "num_input_tokens_seen": 14217128,
      "step": 24520
    },
    {
      "epoch": 3.6528150134048256,
      "grad_norm": 7.78125,
      "learning_rate": 4.896718684649213e-05,
      "loss": 1.4902,
      "num_input_tokens_seen": 14219976,
      "step": 24525
    },
    {
      "epoch": 3.653559725945785,
      "grad_norm": 5.3125,
      "learning_rate": 4.8966262308842885e-05,
      "loss": 1.3382,
      "num_input_tokens_seen": 14222856,
      "step": 24530
    },
    {
      "epoch": 3.654304438486744,
      "grad_norm": 9.0,
      "learning_rate": 4.8965337366308e-05,
      "loss": 0.9712,
      "num_input_tokens_seen": 14225704,
      "step": 24535
    },
    {
      "epoch": 3.6550491510277032,
      "grad_norm": 17.75,
      "learning_rate": 4.896441201890309e-05,
      "loss": 1.2886,
      "num_input_tokens_seen": 14228520,
      "step": 24540
    },
    {
      "epoch": 3.6557938635686624,
      "grad_norm": 10.9375,
      "learning_rate": 4.896348626664381e-05,
      "loss": 0.9112,
      "num_input_tokens_seen": 14231624,
      "step": 24545
    },
    {
      "epoch": 3.6565385761096216,
      "grad_norm": 11.875,
      "learning_rate": 4.896256010954578e-05,
      "loss": 1.1216,
      "num_input_tokens_seen": 14234568,
      "step": 24550
    },
    {
      "epoch": 3.657283288650581,
      "grad_norm": 8.8125,
      "learning_rate": 4.896163354762464e-05,
      "loss": 1.0875,
      "num_input_tokens_seen": 14237704,
      "step": 24555
    },
    {
      "epoch": 3.65802800119154,
      "grad_norm": 10.8125,
      "learning_rate": 4.8960706580896066e-05,
      "loss": 1.2524,
      "num_input_tokens_seen": 14240616,
      "step": 24560
    },
    {
      "epoch": 3.6587727137324992,
      "grad_norm": 16.875,
      "learning_rate": 4.8959779209375703e-05,
      "loss": 1.2111,
      "num_input_tokens_seen": 14243656,
      "step": 24565
    },
    {
      "epoch": 3.6595174262734584,
      "grad_norm": 17.875,
      "learning_rate": 4.895885143307922e-05,
      "loss": 1.068,
      "num_input_tokens_seen": 14246440,
      "step": 24570
    },
    {
      "epoch": 3.6602621388144176,
      "grad_norm": 8.3125,
      "learning_rate": 4.8957923252022304e-05,
      "loss": 1.2523,
      "num_input_tokens_seen": 14249448,
      "step": 24575
    },
    {
      "epoch": 3.661006851355377,
      "grad_norm": 10.25,
      "learning_rate": 4.8956994666220615e-05,
      "loss": 1.0102,
      "num_input_tokens_seen": 14252392,
      "step": 24580
    },
    {
      "epoch": 3.661751563896336,
      "grad_norm": 9.6875,
      "learning_rate": 4.895606567568985e-05,
      "loss": 1.1051,
      "num_input_tokens_seen": 14255048,
      "step": 24585
    },
    {
      "epoch": 3.6624962764372953,
      "grad_norm": 6.5,
      "learning_rate": 4.8955136280445704e-05,
      "loss": 0.971,
      "num_input_tokens_seen": 14257832,
      "step": 24590
    },
    {
      "epoch": 3.6632409889782545,
      "grad_norm": 9.8125,
      "learning_rate": 4.895420648050388e-05,
      "loss": 1.1069,
      "num_input_tokens_seen": 14260712,
      "step": 24595
    },
    {
      "epoch": 3.6639857015192137,
      "grad_norm": 9.75,
      "learning_rate": 4.895327627588008e-05,
      "loss": 0.8763,
      "num_input_tokens_seen": 14263496,
      "step": 24600
    },
    {
      "epoch": 3.664730414060173,
      "grad_norm": 6.21875,
      "learning_rate": 4.8952345666590025e-05,
      "loss": 0.9773,
      "num_input_tokens_seen": 14266344,
      "step": 24605
    },
    {
      "epoch": 3.665475126601132,
      "grad_norm": 18.25,
      "learning_rate": 4.895141465264943e-05,
      "loss": 1.3743,
      "num_input_tokens_seen": 14269096,
      "step": 24610
    },
    {
      "epoch": 3.6662198391420913,
      "grad_norm": 8.625,
      "learning_rate": 4.895048323407403e-05,
      "loss": 0.8655,
      "num_input_tokens_seen": 14271912,
      "step": 24615
    },
    {
      "epoch": 3.6669645516830505,
      "grad_norm": 16.5,
      "learning_rate": 4.894955141087956e-05,
      "loss": 1.3547,
      "num_input_tokens_seen": 14274472,
      "step": 24620
    },
    {
      "epoch": 3.6677092642240097,
      "grad_norm": 14.375,
      "learning_rate": 4.894861918308176e-05,
      "loss": 1.498,
      "num_input_tokens_seen": 14277768,
      "step": 24625
    },
    {
      "epoch": 3.668453976764969,
      "grad_norm": 9.1875,
      "learning_rate": 4.894768655069638e-05,
      "loss": 1.2412,
      "num_input_tokens_seen": 14280872,
      "step": 24630
    },
    {
      "epoch": 3.669198689305928,
      "grad_norm": 12.125,
      "learning_rate": 4.8946753513739166e-05,
      "loss": 1.4591,
      "num_input_tokens_seen": 14283752,
      "step": 24635
    },
    {
      "epoch": 3.6699434018468873,
      "grad_norm": 18.625,
      "learning_rate": 4.89458200722259e-05,
      "loss": 0.9092,
      "num_input_tokens_seen": 14287016,
      "step": 24640
    },
    {
      "epoch": 3.670688114387846,
      "grad_norm": 9.875,
      "learning_rate": 4.894488622617234e-05,
      "loss": 1.1995,
      "num_input_tokens_seen": 14289864,
      "step": 24645
    },
    {
      "epoch": 3.6714328269288057,
      "grad_norm": 3.03125,
      "learning_rate": 4.894395197559426e-05,
      "loss": 0.8925,
      "num_input_tokens_seen": 14292776,
      "step": 24650
    },
    {
      "epoch": 3.6721775394697644,
      "grad_norm": 14.75,
      "learning_rate": 4.8943017320507444e-05,
      "loss": 1.414,
      "num_input_tokens_seen": 14295784,
      "step": 24655
    },
    {
      "epoch": 3.672922252010724,
      "grad_norm": 11.75,
      "learning_rate": 4.894208226092769e-05,
      "loss": 0.8573,
      "num_input_tokens_seen": 14298632,
      "step": 24660
    },
    {
      "epoch": 3.673666964551683,
      "grad_norm": 5.5,
      "learning_rate": 4.894114679687079e-05,
      "loss": 0.988,
      "num_input_tokens_seen": 14301576,
      "step": 24665
    },
    {
      "epoch": 3.6744116770926425,
      "grad_norm": 19.75,
      "learning_rate": 4.8940210928352545e-05,
      "loss": 1.3435,
      "num_input_tokens_seen": 14304296,
      "step": 24670
    },
    {
      "epoch": 3.6751563896336013,
      "grad_norm": 10.5,
      "learning_rate": 4.893927465538877e-05,
      "loss": 1.1502,
      "num_input_tokens_seen": 14306952,
      "step": 24675
    },
    {
      "epoch": 3.675901102174561,
      "grad_norm": 13.25,
      "learning_rate": 4.8938337977995286e-05,
      "loss": 0.8267,
      "num_input_tokens_seen": 14310024,
      "step": 24680
    },
    {
      "epoch": 3.6766458147155197,
      "grad_norm": 9.4375,
      "learning_rate": 4.89374008961879e-05,
      "loss": 0.9406,
      "num_input_tokens_seen": 14312840,
      "step": 24685
    },
    {
      "epoch": 3.677390527256479,
      "grad_norm": 11.125,
      "learning_rate": 4.8936463409982466e-05,
      "loss": 1.1302,
      "num_input_tokens_seen": 14315560,
      "step": 24690
    },
    {
      "epoch": 3.678135239797438,
      "grad_norm": 9.25,
      "learning_rate": 4.89355255193948e-05,
      "loss": 1.0405,
      "num_input_tokens_seen": 14318376,
      "step": 24695
    },
    {
      "epoch": 3.6788799523383973,
      "grad_norm": 10.1875,
      "learning_rate": 4.893458722444076e-05,
      "loss": 1.2554,
      "num_input_tokens_seen": 14321448,
      "step": 24700
    },
    {
      "epoch": 3.6796246648793565,
      "grad_norm": 8.9375,
      "learning_rate": 4.89336485251362e-05,
      "loss": 1.1592,
      "num_input_tokens_seen": 14324200,
      "step": 24705
    },
    {
      "epoch": 3.6803693774203157,
      "grad_norm": 16.375,
      "learning_rate": 4.893270942149697e-05,
      "loss": 0.9823,
      "num_input_tokens_seen": 14326984,
      "step": 24710
    },
    {
      "epoch": 3.681114089961275,
      "grad_norm": 13.1875,
      "learning_rate": 4.8931769913538945e-05,
      "loss": 1.1549,
      "num_input_tokens_seen": 14329768,
      "step": 24715
    },
    {
      "epoch": 3.681858802502234,
      "grad_norm": 10.75,
      "learning_rate": 4.893083000127798e-05,
      "loss": 0.9781,
      "num_input_tokens_seen": 14332424,
      "step": 24720
    },
    {
      "epoch": 3.6826035150431933,
      "grad_norm": 8.625,
      "learning_rate": 4.8929889684729966e-05,
      "loss": 1.1036,
      "num_input_tokens_seen": 14335400,
      "step": 24725
    },
    {
      "epoch": 3.6833482275841525,
      "grad_norm": 6.8125,
      "learning_rate": 4.892894896391079e-05,
      "loss": 1.0066,
      "num_input_tokens_seen": 14338216,
      "step": 24730
    },
    {
      "epoch": 3.6840929401251117,
      "grad_norm": 16.625,
      "learning_rate": 4.892800783883635e-05,
      "loss": 1.5126,
      "num_input_tokens_seen": 14341064,
      "step": 24735
    },
    {
      "epoch": 3.684837652666071,
      "grad_norm": 10.5625,
      "learning_rate": 4.892706630952253e-05,
      "loss": 1.0719,
      "num_input_tokens_seen": 14343944,
      "step": 24740
    },
    {
      "epoch": 3.68558236520703,
      "grad_norm": 8.8125,
      "learning_rate": 4.892612437598524e-05,
      "loss": 1.0812,
      "num_input_tokens_seen": 14346824,
      "step": 24745
    },
    {
      "epoch": 3.6863270777479893,
      "grad_norm": 6.71875,
      "learning_rate": 4.8925182038240395e-05,
      "loss": 0.8044,
      "num_input_tokens_seen": 14349736,
      "step": 24750
    },
    {
      "epoch": 3.6870717902889485,
      "grad_norm": 11.6875,
      "learning_rate": 4.892423929630392e-05,
      "loss": 1.0312,
      "num_input_tokens_seen": 14353096,
      "step": 24755
    },
    {
      "epoch": 3.6878165028299077,
      "grad_norm": 10.3125,
      "learning_rate": 4.892329615019173e-05,
      "loss": 0.9935,
      "num_input_tokens_seen": 14356104,
      "step": 24760
    },
    {
      "epoch": 3.688561215370867,
      "grad_norm": 7.09375,
      "learning_rate": 4.892235259991977e-05,
      "loss": 0.9109,
      "num_input_tokens_seen": 14359240,
      "step": 24765
    },
    {
      "epoch": 3.689305927911826,
      "grad_norm": 11.75,
      "learning_rate": 4.8921408645503986e-05,
      "loss": 0.9919,
      "num_input_tokens_seen": 14362216,
      "step": 24770
    },
    {
      "epoch": 3.6900506404527853,
      "grad_norm": 10.5625,
      "learning_rate": 4.892046428696031e-05,
      "loss": 0.9103,
      "num_input_tokens_seen": 14365096,
      "step": 24775
    },
    {
      "epoch": 3.6907953529937445,
      "grad_norm": 6.1875,
      "learning_rate": 4.8919519524304704e-05,
      "loss": 1.3413,
      "num_input_tokens_seen": 14368072,
      "step": 24780
    },
    {
      "epoch": 3.6915400655347037,
      "grad_norm": 18.5,
      "learning_rate": 4.891857435755312e-05,
      "loss": 1.3192,
      "num_input_tokens_seen": 14370728,
      "step": 24785
    },
    {
      "epoch": 3.692284778075663,
      "grad_norm": 14.0,
      "learning_rate": 4.891762878672153e-05,
      "loss": 0.8914,
      "num_input_tokens_seen": 14373960,
      "step": 24790
    },
    {
      "epoch": 3.693029490616622,
      "grad_norm": 8.3125,
      "learning_rate": 4.891668281182592e-05,
      "loss": 1.4007,
      "num_input_tokens_seen": 14376648,
      "step": 24795
    },
    {
      "epoch": 3.6937742031575813,
      "grad_norm": 5.75,
      "learning_rate": 4.8915736432882254e-05,
      "loss": 0.9579,
      "num_input_tokens_seen": 14380008,
      "step": 24800
    },
    {
      "epoch": 3.6945189156985405,
      "grad_norm": 12.9375,
      "learning_rate": 4.891478964990653e-05,
      "loss": 1.21,
      "num_input_tokens_seen": 14382760,
      "step": 24805
    },
    {
      "epoch": 3.6952636282394993,
      "grad_norm": 8.25,
      "learning_rate": 4.891384246291474e-05,
      "loss": 1.2318,
      "num_input_tokens_seen": 14386024,
      "step": 24810
    },
    {
      "epoch": 3.696008340780459,
      "grad_norm": 4.6875,
      "learning_rate": 4.891289487192289e-05,
      "loss": 0.62,
      "num_input_tokens_seen": 14389000,
      "step": 24815
    },
    {
      "epoch": 3.6967530533214177,
      "grad_norm": 8.6875,
      "learning_rate": 4.891194687694698e-05,
      "loss": 0.9701,
      "num_input_tokens_seen": 14391752,
      "step": 24820
    },
    {
      "epoch": 3.6974977658623773,
      "grad_norm": 16.5,
      "learning_rate": 4.8910998478003034e-05,
      "loss": 0.9861,
      "num_input_tokens_seen": 14394728,
      "step": 24825
    },
    {
      "epoch": 3.698242478403336,
      "grad_norm": 5.28125,
      "learning_rate": 4.891004967510707e-05,
      "loss": 1.3328,
      "num_input_tokens_seen": 14397480,
      "step": 24830
    },
    {
      "epoch": 3.6989871909442957,
      "grad_norm": 16.375,
      "learning_rate": 4.890910046827511e-05,
      "loss": 1.0386,
      "num_input_tokens_seen": 14400520,
      "step": 24835
    },
    {
      "epoch": 3.6997319034852545,
      "grad_norm": 21.875,
      "learning_rate": 4.890815085752322e-05,
      "loss": 1.065,
      "num_input_tokens_seen": 14403304,
      "step": 24840
    },
    {
      "epoch": 3.700476616026214,
      "grad_norm": 4.875,
      "learning_rate": 4.890720084286739e-05,
      "loss": 0.9215,
      "num_input_tokens_seen": 14406504,
      "step": 24845
    },
    {
      "epoch": 3.701221328567173,
      "grad_norm": 11.375,
      "learning_rate": 4.890625042432372e-05,
      "loss": 1.1238,
      "num_input_tokens_seen": 14409544,
      "step": 24850
    },
    {
      "epoch": 3.701966041108132,
      "grad_norm": 13.25,
      "learning_rate": 4.890529960190825e-05,
      "loss": 1.3962,
      "num_input_tokens_seen": 14412744,
      "step": 24855
    },
    {
      "epoch": 3.7027107536490913,
      "grad_norm": 6.28125,
      "learning_rate": 4.8904348375637025e-05,
      "loss": 1.2441,
      "num_input_tokens_seen": 14415464,
      "step": 24860
    },
    {
      "epoch": 3.7034554661900505,
      "grad_norm": 8.9375,
      "learning_rate": 4.8903396745526144e-05,
      "loss": 1.2516,
      "num_input_tokens_seen": 14418376,
      "step": 24865
    },
    {
      "epoch": 3.7042001787310097,
      "grad_norm": 7.125,
      "learning_rate": 4.8902444711591656e-05,
      "loss": 0.9169,
      "num_input_tokens_seen": 14421096,
      "step": 24870
    },
    {
      "epoch": 3.704944891271969,
      "grad_norm": 12.75,
      "learning_rate": 4.8901492273849666e-05,
      "loss": 1.184,
      "num_input_tokens_seen": 14423880,
      "step": 24875
    },
    {
      "epoch": 3.705689603812928,
      "grad_norm": 15.0,
      "learning_rate": 4.890053943231625e-05,
      "loss": 1.3966,
      "num_input_tokens_seen": 14426632,
      "step": 24880
    },
    {
      "epoch": 3.7064343163538873,
      "grad_norm": 6.1875,
      "learning_rate": 4.889958618700752e-05,
      "loss": 1.0967,
      "num_input_tokens_seen": 14429384,
      "step": 24885
    },
    {
      "epoch": 3.7071790288948465,
      "grad_norm": 12.375,
      "learning_rate": 4.8898632537939567e-05,
      "loss": 0.9249,
      "num_input_tokens_seen": 14432520,
      "step": 24890
    },
    {
      "epoch": 3.7079237414358057,
      "grad_norm": 12.9375,
      "learning_rate": 4.889767848512851e-05,
      "loss": 1.0211,
      "num_input_tokens_seen": 14435624,
      "step": 24895
    },
    {
      "epoch": 3.708668453976765,
      "grad_norm": 10.375,
      "learning_rate": 4.889672402859046e-05,
      "loss": 0.8615,
      "num_input_tokens_seen": 14438184,
      "step": 24900
    },
    {
      "epoch": 3.709413166517724,
      "grad_norm": 15.875,
      "learning_rate": 4.8895769168341546e-05,
      "loss": 1.0753,
      "num_input_tokens_seen": 14440808,
      "step": 24905
    },
    {
      "epoch": 3.7101578790586833,
      "grad_norm": 15.0,
      "learning_rate": 4.8894813904397895e-05,
      "loss": 1.1046,
      "num_input_tokens_seen": 14443720,
      "step": 24910
    },
    {
      "epoch": 3.7109025915996425,
      "grad_norm": 7.09375,
      "learning_rate": 4.889385823677565e-05,
      "loss": 0.9613,
      "num_input_tokens_seen": 14446504,
      "step": 24915
    },
    {
      "epoch": 3.7116473041406017,
      "grad_norm": 18.375,
      "learning_rate": 4.889290216549096e-05,
      "loss": 1.022,
      "num_input_tokens_seen": 14449576,
      "step": 24920
    },
    {
      "epoch": 3.712392016681561,
      "grad_norm": 5.4375,
      "learning_rate": 4.889194569055996e-05,
      "loss": 1.0508,
      "num_input_tokens_seen": 14452488,
      "step": 24925
    },
    {
      "epoch": 3.71313672922252,
      "grad_norm": 11.0,
      "learning_rate": 4.8890988811998835e-05,
      "loss": 1.0418,
      "num_input_tokens_seen": 14455272,
      "step": 24930
    },
    {
      "epoch": 3.7138814417634793,
      "grad_norm": 10.4375,
      "learning_rate": 4.889003152982373e-05,
      "loss": 1.329,
      "num_input_tokens_seen": 14458344,
      "step": 24935
    },
    {
      "epoch": 3.7146261543044385,
      "grad_norm": 8.0625,
      "learning_rate": 4.888907384405082e-05,
      "loss": 0.9001,
      "num_input_tokens_seen": 14461384,
      "step": 24940
    },
    {
      "epoch": 3.7153708668453977,
      "grad_norm": 15.9375,
      "learning_rate": 4.888811575469629e-05,
      "loss": 1.4418,
      "num_input_tokens_seen": 14464232,
      "step": 24945
    },
    {
      "epoch": 3.716115579386357,
      "grad_norm": 5.71875,
      "learning_rate": 4.8887157261776316e-05,
      "loss": 0.8903,
      "num_input_tokens_seen": 14467048,
      "step": 24950
    },
    {
      "epoch": 3.716860291927316,
      "grad_norm": 6.65625,
      "learning_rate": 4.888619836530711e-05,
      "loss": 1.3005,
      "num_input_tokens_seen": 14469832,
      "step": 24955
    },
    {
      "epoch": 3.7176050044682754,
      "grad_norm": 11.5625,
      "learning_rate": 4.8885239065304855e-05,
      "loss": 1.3513,
      "num_input_tokens_seen": 14472520,
      "step": 24960
    },
    {
      "epoch": 3.7183497170092346,
      "grad_norm": 6.375,
      "learning_rate": 4.8884279361785754e-05,
      "loss": 1.0189,
      "num_input_tokens_seen": 14475496,
      "step": 24965
    },
    {
      "epoch": 3.7190944295501938,
      "grad_norm": 9.4375,
      "learning_rate": 4.888331925476604e-05,
      "loss": 0.9025,
      "num_input_tokens_seen": 14478216,
      "step": 24970
    },
    {
      "epoch": 3.719839142091153,
      "grad_norm": 17.125,
      "learning_rate": 4.8882358744261914e-05,
      "loss": 1.0344,
      "num_input_tokens_seen": 14481192,
      "step": 24975
    },
    {
      "epoch": 3.720583854632112,
      "grad_norm": 13.25,
      "learning_rate": 4.888139783028961e-05,
      "loss": 1.2512,
      "num_input_tokens_seen": 14484104,
      "step": 24980
    },
    {
      "epoch": 3.721328567173071,
      "grad_norm": 10.5,
      "learning_rate": 4.888043651286537e-05,
      "loss": 0.9563,
      "num_input_tokens_seen": 14487304,
      "step": 24985
    },
    {
      "epoch": 3.7220732797140306,
      "grad_norm": 20.375,
      "learning_rate": 4.887947479200542e-05,
      "loss": 1.0433,
      "num_input_tokens_seen": 14490056,
      "step": 24990
    },
    {
      "epoch": 3.7228179922549893,
      "grad_norm": 12.25,
      "learning_rate": 4.887851266772601e-05,
      "loss": 1.0839,
      "num_input_tokens_seen": 14493384,
      "step": 24995
    },
    {
      "epoch": 3.723562704795949,
      "grad_norm": 9.3125,
      "learning_rate": 4.8877550140043404e-05,
      "loss": 0.875,
      "num_input_tokens_seen": 14496552,
      "step": 25000
    },
    {
      "epoch": 3.7243074173369077,
      "grad_norm": 9.0625,
      "learning_rate": 4.887658720897385e-05,
      "loss": 0.9435,
      "num_input_tokens_seen": 14499336,
      "step": 25005
    },
    {
      "epoch": 3.7250521298778674,
      "grad_norm": 19.0,
      "learning_rate": 4.8875623874533627e-05,
      "loss": 1.0955,
      "num_input_tokens_seen": 14502216,
      "step": 25010
    },
    {
      "epoch": 3.725796842418826,
      "grad_norm": 10.25,
      "learning_rate": 4.8874660136739e-05,
      "loss": 0.7827,
      "num_input_tokens_seen": 14505352,
      "step": 25015
    },
    {
      "epoch": 3.726541554959786,
      "grad_norm": 13.625,
      "learning_rate": 4.887369599560626e-05,
      "loss": 1.0318,
      "num_input_tokens_seen": 14508456,
      "step": 25020
    },
    {
      "epoch": 3.7272862675007445,
      "grad_norm": 16.375,
      "learning_rate": 4.8872731451151684e-05,
      "loss": 1.1162,
      "num_input_tokens_seen": 14511336,
      "step": 25025
    },
    {
      "epoch": 3.7280309800417037,
      "grad_norm": 9.625,
      "learning_rate": 4.887176650339158e-05,
      "loss": 0.9595,
      "num_input_tokens_seen": 14514248,
      "step": 25030
    },
    {
      "epoch": 3.728775692582663,
      "grad_norm": 11.5625,
      "learning_rate": 4.887080115234224e-05,
      "loss": 1.319,
      "num_input_tokens_seen": 14516904,
      "step": 25035
    },
    {
      "epoch": 3.729520405123622,
      "grad_norm": 6.59375,
      "learning_rate": 4.886983539801998e-05,
      "loss": 0.9647,
      "num_input_tokens_seen": 14520200,
      "step": 25040
    },
    {
      "epoch": 3.7302651176645814,
      "grad_norm": 18.375,
      "learning_rate": 4.886886924044111e-05,
      "loss": 0.7355,
      "num_input_tokens_seen": 14523144,
      "step": 25045
    },
    {
      "epoch": 3.7310098302055406,
      "grad_norm": 13.1875,
      "learning_rate": 4.8867902679621946e-05,
      "loss": 0.8396,
      "num_input_tokens_seen": 14526312,
      "step": 25050
    },
    {
      "epoch": 3.7317545427464998,
      "grad_norm": 6.96875,
      "learning_rate": 4.8866935715578835e-05,
      "loss": 1.3609,
      "num_input_tokens_seen": 14529096,
      "step": 25055
    },
    {
      "epoch": 3.732499255287459,
      "grad_norm": 7.75,
      "learning_rate": 4.88659683483281e-05,
      "loss": 1.0175,
      "num_input_tokens_seen": 14531752,
      "step": 25060
    },
    {
      "epoch": 3.733243967828418,
      "grad_norm": 9.0625,
      "learning_rate": 4.886500057788608e-05,
      "loss": 1.1265,
      "num_input_tokens_seen": 14534600,
      "step": 25065
    },
    {
      "epoch": 3.7339886803693774,
      "grad_norm": 10.4375,
      "learning_rate": 4.8864032404269126e-05,
      "loss": 1.0654,
      "num_input_tokens_seen": 14537448,
      "step": 25070
    },
    {
      "epoch": 3.7347333929103366,
      "grad_norm": 6.96875,
      "learning_rate": 4.886306382749361e-05,
      "loss": 1.105,
      "num_input_tokens_seen": 14540680,
      "step": 25075
    },
    {
      "epoch": 3.7354781054512958,
      "grad_norm": 13.1875,
      "learning_rate": 4.886209484757588e-05,
      "loss": 1.6643,
      "num_input_tokens_seen": 14543848,
      "step": 25080
    },
    {
      "epoch": 3.736222817992255,
      "grad_norm": 15.375,
      "learning_rate": 4.886112546453231e-05,
      "loss": 1.2768,
      "num_input_tokens_seen": 14546504,
      "step": 25085
    },
    {
      "epoch": 3.736967530533214,
      "grad_norm": 8.9375,
      "learning_rate": 4.886015567837927e-05,
      "loss": 1.1184,
      "num_input_tokens_seen": 14549480,
      "step": 25090
    },
    {
      "epoch": 3.7377122430741734,
      "grad_norm": 13.25,
      "learning_rate": 4.885918548913316e-05,
      "loss": 0.9781,
      "num_input_tokens_seen": 14552264,
      "step": 25095
    },
    {
      "epoch": 3.7384569556151326,
      "grad_norm": 12.75,
      "learning_rate": 4.885821489681036e-05,
      "loss": 1.1706,
      "num_input_tokens_seen": 14554952,
      "step": 25100
    },
    {
      "epoch": 3.739201668156092,
      "grad_norm": 10.25,
      "learning_rate": 4.885724390142726e-05,
      "loss": 0.8312,
      "num_input_tokens_seen": 14557992,
      "step": 25105
    },
    {
      "epoch": 3.739946380697051,
      "grad_norm": 17.5,
      "learning_rate": 4.885627250300028e-05,
      "loss": 0.9774,
      "num_input_tokens_seen": 14560648,
      "step": 25110
    },
    {
      "epoch": 3.74069109323801,
      "grad_norm": 8.75,
      "learning_rate": 4.885530070154582e-05,
      "loss": 1.0731,
      "num_input_tokens_seen": 14563336,
      "step": 25115
    },
    {
      "epoch": 3.7414358057789694,
      "grad_norm": 8.0,
      "learning_rate": 4.88543284970803e-05,
      "loss": 1.0529,
      "num_input_tokens_seen": 14565896,
      "step": 25120
    },
    {
      "epoch": 3.7421805183199286,
      "grad_norm": 10.3125,
      "learning_rate": 4.8853355889620143e-05,
      "loss": 0.7363,
      "num_input_tokens_seen": 14568808,
      "step": 25125
    },
    {
      "epoch": 3.742925230860888,
      "grad_norm": 14.0625,
      "learning_rate": 4.885238287918178e-05,
      "loss": 0.9659,
      "num_input_tokens_seen": 14571656,
      "step": 25130
    },
    {
      "epoch": 3.743669943401847,
      "grad_norm": 15.875,
      "learning_rate": 4.885140946578166e-05,
      "loss": 1.0622,
      "num_input_tokens_seen": 14574568,
      "step": 25135
    },
    {
      "epoch": 3.744414655942806,
      "grad_norm": 7.84375,
      "learning_rate": 4.885043564943621e-05,
      "loss": 0.9768,
      "num_input_tokens_seen": 14577576,
      "step": 25140
    },
    {
      "epoch": 3.7451593684837654,
      "grad_norm": 14.1875,
      "learning_rate": 4.884946143016189e-05,
      "loss": 1.243,
      "num_input_tokens_seen": 14580360,
      "step": 25145
    },
    {
      "epoch": 3.7459040810247246,
      "grad_norm": 11.625,
      "learning_rate": 4.884848680797516e-05,
      "loss": 1.0723,
      "num_input_tokens_seen": 14583368,
      "step": 25150
    },
    {
      "epoch": 3.746648793565684,
      "grad_norm": 9.625,
      "learning_rate": 4.884751178289249e-05,
      "loss": 1.2649,
      "num_input_tokens_seen": 14586088,
      "step": 25155
    },
    {
      "epoch": 3.7473935061066426,
      "grad_norm": 12.0,
      "learning_rate": 4.884653635493034e-05,
      "loss": 1.045,
      "num_input_tokens_seen": 14588904,
      "step": 25160
    },
    {
      "epoch": 3.748138218647602,
      "grad_norm": 10.0,
      "learning_rate": 4.8845560524105196e-05,
      "loss": 1.0042,
      "num_input_tokens_seen": 14591688,
      "step": 25165
    },
    {
      "epoch": 3.748882931188561,
      "grad_norm": 12.875,
      "learning_rate": 4.8844584290433536e-05,
      "loss": 1.0703,
      "num_input_tokens_seen": 14594376,
      "step": 25170
    },
    {
      "epoch": 3.7496276437295206,
      "grad_norm": 13.5,
      "learning_rate": 4.8843607653931865e-05,
      "loss": 1.01,
      "num_input_tokens_seen": 14597256,
      "step": 25175
    },
    {
      "epoch": 3.7503723562704794,
      "grad_norm": 6.9375,
      "learning_rate": 4.884263061461668e-05,
      "loss": 1.1094,
      "num_input_tokens_seen": 14599944,
      "step": 25180
    },
    {
      "epoch": 3.751117068811439,
      "grad_norm": 10.875,
      "learning_rate": 4.884165317250448e-05,
      "loss": 1.0364,
      "num_input_tokens_seen": 14602888,
      "step": 25185
    },
    {
      "epoch": 3.751861781352398,
      "grad_norm": 13.5,
      "learning_rate": 4.8840675327611785e-05,
      "loss": 0.885,
      "num_input_tokens_seen": 14605704,
      "step": 25190
    },
    {
      "epoch": 3.7526064938933574,
      "grad_norm": 13.75,
      "learning_rate": 4.8839697079955104e-05,
      "loss": 1.0338,
      "num_input_tokens_seen": 14608744,
      "step": 25195
    },
    {
      "epoch": 3.753351206434316,
      "grad_norm": 6.5625,
      "learning_rate": 4.883871842955097e-05,
      "loss": 1.2318,
      "num_input_tokens_seen": 14611432,
      "step": 25200
    },
    {
      "epoch": 3.7540959189752754,
      "grad_norm": 20.75,
      "learning_rate": 4.883773937641593e-05,
      "loss": 1.1507,
      "num_input_tokens_seen": 14614536,
      "step": 25205
    },
    {
      "epoch": 3.7548406315162346,
      "grad_norm": 6.875,
      "learning_rate": 4.8836759920566494e-05,
      "loss": 1.1223,
      "num_input_tokens_seen": 14617224,
      "step": 25210
    },
    {
      "epoch": 3.755585344057194,
      "grad_norm": 5.84375,
      "learning_rate": 4.8835780062019234e-05,
      "loss": 0.9414,
      "num_input_tokens_seen": 14619912,
      "step": 25215
    },
    {
      "epoch": 3.756330056598153,
      "grad_norm": 11.375,
      "learning_rate": 4.8834799800790694e-05,
      "loss": 1.297,
      "num_input_tokens_seen": 14622664,
      "step": 25220
    },
    {
      "epoch": 3.757074769139112,
      "grad_norm": 11.0,
      "learning_rate": 4.8833819136897436e-05,
      "loss": 1.4138,
      "num_input_tokens_seen": 14625480,
      "step": 25225
    },
    {
      "epoch": 3.7578194816800714,
      "grad_norm": 4.09375,
      "learning_rate": 4.883283807035602e-05,
      "loss": 0.9067,
      "num_input_tokens_seen": 14628424,
      "step": 25230
    },
    {
      "epoch": 3.7585641942210306,
      "grad_norm": 7.78125,
      "learning_rate": 4.883185660118304e-05,
      "loss": 1.2388,
      "num_input_tokens_seen": 14631176,
      "step": 25235
    },
    {
      "epoch": 3.75930890676199,
      "grad_norm": 10.8125,
      "learning_rate": 4.883087472939506e-05,
      "loss": 1.2186,
      "num_input_tokens_seen": 14633928,
      "step": 25240
    },
    {
      "epoch": 3.760053619302949,
      "grad_norm": 6.3125,
      "learning_rate": 4.882989245500867e-05,
      "loss": 0.8907,
      "num_input_tokens_seen": 14637096,
      "step": 25245
    },
    {
      "epoch": 3.760798331843908,
      "grad_norm": 12.3125,
      "learning_rate": 4.882890977804047e-05,
      "loss": 1.1452,
      "num_input_tokens_seen": 14639976,
      "step": 25250
    },
    {
      "epoch": 3.7615430443848674,
      "grad_norm": 7.71875,
      "learning_rate": 4.882792669850705e-05,
      "loss": 0.986,
      "num_input_tokens_seen": 14642792,
      "step": 25255
    },
    {
      "epoch": 3.7622877569258266,
      "grad_norm": 16.25,
      "learning_rate": 4.882694321642504e-05,
      "loss": 1.22,
      "num_input_tokens_seen": 14645704,
      "step": 25260
    },
    {
      "epoch": 3.763032469466786,
      "grad_norm": 7.875,
      "learning_rate": 4.8825959331811026e-05,
      "loss": 1.1851,
      "num_input_tokens_seen": 14648680,
      "step": 25265
    },
    {
      "epoch": 3.763777182007745,
      "grad_norm": 17.125,
      "learning_rate": 4.882497504468165e-05,
      "loss": 1.3354,
      "num_input_tokens_seen": 14651720,
      "step": 25270
    },
    {
      "epoch": 3.7645218945487042,
      "grad_norm": 9.9375,
      "learning_rate": 4.8823990355053536e-05,
      "loss": 1.1082,
      "num_input_tokens_seen": 14654344,
      "step": 25275
    },
    {
      "epoch": 3.7652666070896634,
      "grad_norm": 8.9375,
      "learning_rate": 4.8823005262943323e-05,
      "loss": 0.9235,
      "num_input_tokens_seen": 14657224,
      "step": 25280
    },
    {
      "epoch": 3.7660113196306226,
      "grad_norm": 7.0,
      "learning_rate": 4.882201976836764e-05,
      "loss": 0.9251,
      "num_input_tokens_seen": 14660200,
      "step": 25285
    },
    {
      "epoch": 3.766756032171582,
      "grad_norm": 15.125,
      "learning_rate": 4.8821033871343155e-05,
      "loss": 1.1706,
      "num_input_tokens_seen": 14662728,
      "step": 25290
    },
    {
      "epoch": 3.767500744712541,
      "grad_norm": 7.625,
      "learning_rate": 4.8820047571886504e-05,
      "loss": 0.7855,
      "num_input_tokens_seen": 14665800,
      "step": 25295
    },
    {
      "epoch": 3.7682454572535002,
      "grad_norm": 6.8125,
      "learning_rate": 4.8819060870014366e-05,
      "loss": 0.9879,
      "num_input_tokens_seen": 14668680,
      "step": 25300
    },
    {
      "epoch": 3.7689901697944594,
      "grad_norm": 10.6875,
      "learning_rate": 4.88180737657434e-05,
      "loss": 1.2448,
      "num_input_tokens_seen": 14671656,
      "step": 25305
    },
    {
      "epoch": 3.7697348823354186,
      "grad_norm": 9.0,
      "learning_rate": 4.881708625909028e-05,
      "loss": 1.0365,
      "num_input_tokens_seen": 14674952,
      "step": 25310
    },
    {
      "epoch": 3.770479594876378,
      "grad_norm": 8.4375,
      "learning_rate": 4.881609835007171e-05,
      "loss": 0.9086,
      "num_input_tokens_seen": 14677800,
      "step": 25315
    },
    {
      "epoch": 3.771224307417337,
      "grad_norm": 8.8125,
      "learning_rate": 4.881511003870435e-05,
      "loss": 0.9831,
      "num_input_tokens_seen": 14680552,
      "step": 25320
    },
    {
      "epoch": 3.7719690199582963,
      "grad_norm": 12.125,
      "learning_rate": 4.881412132500491e-05,
      "loss": 1.286,
      "num_input_tokens_seen": 14683368,
      "step": 25325
    },
    {
      "epoch": 3.7727137324992555,
      "grad_norm": 5.21875,
      "learning_rate": 4.8813132208990095e-05,
      "loss": 1.2349,
      "num_input_tokens_seen": 14686312,
      "step": 25330
    },
    {
      "epoch": 3.773458445040214,
      "grad_norm": 11.5,
      "learning_rate": 4.881214269067662e-05,
      "loss": 0.9996,
      "num_input_tokens_seen": 14689544,
      "step": 25335
    },
    {
      "epoch": 3.774203157581174,
      "grad_norm": 9.6875,
      "learning_rate": 4.881115277008119e-05,
      "loss": 1.2158,
      "num_input_tokens_seen": 14692488,
      "step": 25340
    },
    {
      "epoch": 3.7749478701221326,
      "grad_norm": 6.3125,
      "learning_rate": 4.881016244722054e-05,
      "loss": 1.2898,
      "num_input_tokens_seen": 14695368,
      "step": 25345
    },
    {
      "epoch": 3.7756925826630923,
      "grad_norm": 13.0,
      "learning_rate": 4.880917172211139e-05,
      "loss": 1.2239,
      "num_input_tokens_seen": 14698056,
      "step": 25350
    },
    {
      "epoch": 3.776437295204051,
      "grad_norm": 13.0,
      "learning_rate": 4.8808180594770486e-05,
      "loss": 1.0055,
      "num_input_tokens_seen": 14700872,
      "step": 25355
    },
    {
      "epoch": 3.7771820077450107,
      "grad_norm": 8.625,
      "learning_rate": 4.880718906521456e-05,
      "loss": 1.0468,
      "num_input_tokens_seen": 14703464,
      "step": 25360
    },
    {
      "epoch": 3.7779267202859694,
      "grad_norm": 15.5,
      "learning_rate": 4.880619713346039e-05,
      "loss": 1.478,
      "num_input_tokens_seen": 14706248,
      "step": 25365
    },
    {
      "epoch": 3.778671432826929,
      "grad_norm": 13.6875,
      "learning_rate": 4.8805204799524695e-05,
      "loss": 1.0686,
      "num_input_tokens_seen": 14709288,
      "step": 25370
    },
    {
      "epoch": 3.779416145367888,
      "grad_norm": 8.0625,
      "learning_rate": 4.880421206342427e-05,
      "loss": 1.0749,
      "num_input_tokens_seen": 14712104,
      "step": 25375
    },
    {
      "epoch": 3.780160857908847,
      "grad_norm": 10.6875,
      "learning_rate": 4.880321892517587e-05,
      "loss": 1.3709,
      "num_input_tokens_seen": 14715144,
      "step": 25380
    },
    {
      "epoch": 3.7809055704498062,
      "grad_norm": 12.6875,
      "learning_rate": 4.880222538479629e-05,
      "loss": 0.9465,
      "num_input_tokens_seen": 14717928,
      "step": 25385
    },
    {
      "epoch": 3.7816502829907654,
      "grad_norm": 5.875,
      "learning_rate": 4.880123144230229e-05,
      "loss": 0.8652,
      "num_input_tokens_seen": 14720968,
      "step": 25390
    },
    {
      "epoch": 3.7823949955317246,
      "grad_norm": 6.0,
      "learning_rate": 4.880023709771068e-05,
      "loss": 1.2419,
      "num_input_tokens_seen": 14724072,
      "step": 25395
    },
    {
      "epoch": 3.783139708072684,
      "grad_norm": 12.9375,
      "learning_rate": 4.8799242351038257e-05,
      "loss": 1.1828,
      "num_input_tokens_seen": 14726984,
      "step": 25400
    },
    {
      "epoch": 3.783884420613643,
      "grad_norm": 11.1875,
      "learning_rate": 4.8798247202301824e-05,
      "loss": 1.069,
      "num_input_tokens_seen": 14729832,
      "step": 25405
    },
    {
      "epoch": 3.7846291331546023,
      "grad_norm": 13.1875,
      "learning_rate": 4.879725165151818e-05,
      "loss": 1.0888,
      "num_input_tokens_seen": 14732552,
      "step": 25410
    },
    {
      "epoch": 3.7853738456955615,
      "grad_norm": 9.5,
      "learning_rate": 4.8796255698704165e-05,
      "loss": 1.0701,
      "num_input_tokens_seen": 14735496,
      "step": 25415
    },
    {
      "epoch": 3.7861185582365207,
      "grad_norm": 5.84375,
      "learning_rate": 4.87952593438766e-05,
      "loss": 1.0142,
      "num_input_tokens_seen": 14738344,
      "step": 25420
    },
    {
      "epoch": 3.78686327077748,
      "grad_norm": 11.6875,
      "learning_rate": 4.879426258705231e-05,
      "loss": 1.155,
      "num_input_tokens_seen": 14741448,
      "step": 25425
    },
    {
      "epoch": 3.787607983318439,
      "grad_norm": 12.0,
      "learning_rate": 4.879326542824813e-05,
      "loss": 0.9354,
      "num_input_tokens_seen": 14744264,
      "step": 25430
    },
    {
      "epoch": 3.7883526958593983,
      "grad_norm": 7.53125,
      "learning_rate": 4.8792267867480926e-05,
      "loss": 0.9935,
      "num_input_tokens_seen": 14747304,
      "step": 25435
    },
    {
      "epoch": 3.7890974084003575,
      "grad_norm": 19.375,
      "learning_rate": 4.879126990476752e-05,
      "loss": 1.187,
      "num_input_tokens_seen": 14750344,
      "step": 25440
    },
    {
      "epoch": 3.7898421209413167,
      "grad_norm": 9.8125,
      "learning_rate": 4.8790271540124796e-05,
      "loss": 0.7945,
      "num_input_tokens_seen": 14753576,
      "step": 25445
    },
    {
      "epoch": 3.790586833482276,
      "grad_norm": 9.5625,
      "learning_rate": 4.8789272773569625e-05,
      "loss": 0.7194,
      "num_input_tokens_seen": 14756200,
      "step": 25450
    },
    {
      "epoch": 3.791331546023235,
      "grad_norm": 7.8125,
      "learning_rate": 4.8788273605118855e-05,
      "loss": 1.135,
      "num_input_tokens_seen": 14759176,
      "step": 25455
    },
    {
      "epoch": 3.7920762585641943,
      "grad_norm": 9.6875,
      "learning_rate": 4.8787274034789386e-05,
      "loss": 1.0137,
      "num_input_tokens_seen": 14762280,
      "step": 25460
    },
    {
      "epoch": 3.7928209711051535,
      "grad_norm": 16.125,
      "learning_rate": 4.87862740625981e-05,
      "loss": 1.5782,
      "num_input_tokens_seen": 14765032,
      "step": 25465
    },
    {
      "epoch": 3.7935656836461127,
      "grad_norm": 12.1875,
      "learning_rate": 4.878527368856189e-05,
      "loss": 1.3188,
      "num_input_tokens_seen": 14767752,
      "step": 25470
    },
    {
      "epoch": 3.794310396187072,
      "grad_norm": 12.1875,
      "learning_rate": 4.878427291269765e-05,
      "loss": 1.1182,
      "num_input_tokens_seen": 14770536,
      "step": 25475
    },
    {
      "epoch": 3.795055108728031,
      "grad_norm": 11.375,
      "learning_rate": 4.878327173502229e-05,
      "loss": 1.0127,
      "num_input_tokens_seen": 14773384,
      "step": 25480
    },
    {
      "epoch": 3.7957998212689903,
      "grad_norm": 19.625,
      "learning_rate": 4.8782270155552735e-05,
      "loss": 1.3628,
      "num_input_tokens_seen": 14776328,
      "step": 25485
    },
    {
      "epoch": 3.7965445338099495,
      "grad_norm": 9.625,
      "learning_rate": 4.878126817430588e-05,
      "loss": 0.8422,
      "num_input_tokens_seen": 14779176,
      "step": 25490
    },
    {
      "epoch": 3.7972892463509087,
      "grad_norm": 11.9375,
      "learning_rate": 4.878026579129868e-05,
      "loss": 0.9716,
      "num_input_tokens_seen": 14782312,
      "step": 25495
    },
    {
      "epoch": 3.798033958891868,
      "grad_norm": 7.125,
      "learning_rate": 4.877926300654807e-05,
      "loss": 0.8591,
      "num_input_tokens_seen": 14785576,
      "step": 25500
    },
    {
      "epoch": 3.798778671432827,
      "grad_norm": 16.125,
      "learning_rate": 4.877825982007097e-05,
      "loss": 1.0296,
      "num_input_tokens_seen": 14788456,
      "step": 25505
    },
    {
      "epoch": 3.799523383973786,
      "grad_norm": 9.125,
      "learning_rate": 4.877725623188434e-05,
      "loss": 1.133,
      "num_input_tokens_seen": 14791272,
      "step": 25510
    },
    {
      "epoch": 3.8002680965147455,
      "grad_norm": 9.5,
      "learning_rate": 4.8776252242005124e-05,
      "loss": 1.3374,
      "num_input_tokens_seen": 14794056,
      "step": 25515
    },
    {
      "epoch": 3.8010128090557043,
      "grad_norm": 15.5,
      "learning_rate": 4.87752478504503e-05,
      "loss": 0.8235,
      "num_input_tokens_seen": 14797224,
      "step": 25520
    },
    {
      "epoch": 3.801757521596664,
      "grad_norm": 6.59375,
      "learning_rate": 4.8774243057236824e-05,
      "loss": 1.2547,
      "num_input_tokens_seen": 14800136,
      "step": 25525
    },
    {
      "epoch": 3.8025022341376227,
      "grad_norm": 5.0625,
      "learning_rate": 4.877323786238167e-05,
      "loss": 0.9868,
      "num_input_tokens_seen": 14803016,
      "step": 25530
    },
    {
      "epoch": 3.8032469466785823,
      "grad_norm": 10.8125,
      "learning_rate": 4.877223226590184e-05,
      "loss": 1.3068,
      "num_input_tokens_seen": 14805800,
      "step": 25535
    },
    {
      "epoch": 3.803991659219541,
      "grad_norm": 3.6875,
      "learning_rate": 4.877122626781429e-05,
      "loss": 1.1059,
      "num_input_tokens_seen": 14808936,
      "step": 25540
    },
    {
      "epoch": 3.8047363717605007,
      "grad_norm": 11.375,
      "learning_rate": 4.8770219868136036e-05,
      "loss": 0.8299,
      "num_input_tokens_seen": 14811496,
      "step": 25545
    },
    {
      "epoch": 3.8054810843014595,
      "grad_norm": 8.5625,
      "learning_rate": 4.876921306688408e-05,
      "loss": 0.9861,
      "num_input_tokens_seen": 14814568,
      "step": 25550
    },
    {
      "epoch": 3.8062257968424187,
      "grad_norm": 17.0,
      "learning_rate": 4.8768205864075425e-05,
      "loss": 1.1997,
      "num_input_tokens_seen": 14817544,
      "step": 25555
    },
    {
      "epoch": 3.806970509383378,
      "grad_norm": 14.0,
      "learning_rate": 4.876719825972709e-05,
      "loss": 1.0803,
      "num_input_tokens_seen": 14820776,
      "step": 25560
    },
    {
      "epoch": 3.807715221924337,
      "grad_norm": 9.1875,
      "learning_rate": 4.8766190253856106e-05,
      "loss": 0.8802,
      "num_input_tokens_seen": 14823976,
      "step": 25565
    },
    {
      "epoch": 3.8084599344652963,
      "grad_norm": 9.4375,
      "learning_rate": 4.876518184647948e-05,
      "loss": 0.9575,
      "num_input_tokens_seen": 14826568,
      "step": 25570
    },
    {
      "epoch": 3.8092046470062555,
      "grad_norm": 7.875,
      "learning_rate": 4.8764173037614256e-05,
      "loss": 0.9443,
      "num_input_tokens_seen": 14829224,
      "step": 25575
    },
    {
      "epoch": 3.8099493595472147,
      "grad_norm": 6.21875,
      "learning_rate": 4.876316382727749e-05,
      "loss": 0.86,
      "num_input_tokens_seen": 14832232,
      "step": 25580
    },
    {
      "epoch": 3.810694072088174,
      "grad_norm": 16.75,
      "learning_rate": 4.8762154215486225e-05,
      "loss": 1.3019,
      "num_input_tokens_seen": 14835048,
      "step": 25585
    },
    {
      "epoch": 3.811438784629133,
      "grad_norm": 11.0,
      "learning_rate": 4.876114420225751e-05,
      "loss": 0.8502,
      "num_input_tokens_seen": 14838280,
      "step": 25590
    },
    {
      "epoch": 3.8121834971700923,
      "grad_norm": 4.78125,
      "learning_rate": 4.876013378760842e-05,
      "loss": 0.8486,
      "num_input_tokens_seen": 14841096,
      "step": 25595
    },
    {
      "epoch": 3.8129282097110515,
      "grad_norm": 13.625,
      "learning_rate": 4.875912297155601e-05,
      "loss": 1.1989,
      "num_input_tokens_seen": 14844136,
      "step": 25600
    },
    {
      "epoch": 3.8136729222520107,
      "grad_norm": 10.5,
      "learning_rate": 4.875811175411737e-05,
      "loss": 0.8888,
      "num_input_tokens_seen": 14847112,
      "step": 25605
    },
    {
      "epoch": 3.81441763479297,
      "grad_norm": 14.1875,
      "learning_rate": 4.875710013530958e-05,
      "loss": 1.3522,
      "num_input_tokens_seen": 14850088,
      "step": 25610
    },
    {
      "epoch": 3.815162347333929,
      "grad_norm": 6.15625,
      "learning_rate": 4.8756088115149724e-05,
      "loss": 0.9578,
      "num_input_tokens_seen": 14852744,
      "step": 25615
    },
    {
      "epoch": 3.8159070598748883,
      "grad_norm": 7.0,
      "learning_rate": 4.8755075693654906e-05,
      "loss": 0.9908,
      "num_input_tokens_seen": 14855688,
      "step": 25620
    },
    {
      "epoch": 3.8166517724158475,
      "grad_norm": 6.15625,
      "learning_rate": 4.8754062870842234e-05,
      "loss": 1.1011,
      "num_input_tokens_seen": 14858728,
      "step": 25625
    },
    {
      "epoch": 3.8173964849568067,
      "grad_norm": 11.625,
      "learning_rate": 4.87530496467288e-05,
      "loss": 1.3207,
      "num_input_tokens_seen": 14861544,
      "step": 25630
    },
    {
      "epoch": 3.818141197497766,
      "grad_norm": 5.75,
      "learning_rate": 4.875203602133174e-05,
      "loss": 1.0995,
      "num_input_tokens_seen": 14864392,
      "step": 25635
    },
    {
      "epoch": 3.818885910038725,
      "grad_norm": 5.71875,
      "learning_rate": 4.875102199466817e-05,
      "loss": 1.0703,
      "num_input_tokens_seen": 14867208,
      "step": 25640
    },
    {
      "epoch": 3.8196306225796843,
      "grad_norm": 13.4375,
      "learning_rate": 4.875000756675523e-05,
      "loss": 1.1425,
      "num_input_tokens_seen": 14869800,
      "step": 25645
    },
    {
      "epoch": 3.8203753351206435,
      "grad_norm": 9.625,
      "learning_rate": 4.874899273761004e-05,
      "loss": 1.342,
      "num_input_tokens_seen": 14872552,
      "step": 25650
    },
    {
      "epoch": 3.8211200476616027,
      "grad_norm": 14.8125,
      "learning_rate": 4.8747977507249765e-05,
      "loss": 1.3378,
      "num_input_tokens_seen": 14875784,
      "step": 25655
    },
    {
      "epoch": 3.821864760202562,
      "grad_norm": 9.4375,
      "learning_rate": 4.874696187569154e-05,
      "loss": 0.8856,
      "num_input_tokens_seen": 14878824,
      "step": 25660
    },
    {
      "epoch": 3.822609472743521,
      "grad_norm": 9.9375,
      "learning_rate": 4.874594584295253e-05,
      "loss": 0.8687,
      "num_input_tokens_seen": 14881992,
      "step": 25665
    },
    {
      "epoch": 3.8233541852844803,
      "grad_norm": 6.8125,
      "learning_rate": 4.87449294090499e-05,
      "loss": 0.8224,
      "num_input_tokens_seen": 14884872,
      "step": 25670
    },
    {
      "epoch": 3.824098897825439,
      "grad_norm": 6.84375,
      "learning_rate": 4.874391257400083e-05,
      "loss": 0.9544,
      "num_input_tokens_seen": 14887688,
      "step": 25675
    },
    {
      "epoch": 3.8248436103663987,
      "grad_norm": 18.625,
      "learning_rate": 4.874289533782247e-05,
      "loss": 1.5036,
      "num_input_tokens_seen": 14890600,
      "step": 25680
    },
    {
      "epoch": 3.8255883229073575,
      "grad_norm": 12.5625,
      "learning_rate": 4.874187770053204e-05,
      "loss": 1.1263,
      "num_input_tokens_seen": 14893416,
      "step": 25685
    },
    {
      "epoch": 3.826333035448317,
      "grad_norm": 12.0,
      "learning_rate": 4.874085966214671e-05,
      "loss": 0.8728,
      "num_input_tokens_seen": 14896360,
      "step": 25690
    },
    {
      "epoch": 3.827077747989276,
      "grad_norm": 15.1875,
      "learning_rate": 4.873984122268369e-05,
      "loss": 1.2785,
      "num_input_tokens_seen": 14899368,
      "step": 25695
    },
    {
      "epoch": 3.8278224605302356,
      "grad_norm": 15.625,
      "learning_rate": 4.873882238216017e-05,
      "loss": 1.1005,
      "num_input_tokens_seen": 14902184,
      "step": 25700
    },
    {
      "epoch": 3.8285671730711943,
      "grad_norm": 15.6875,
      "learning_rate": 4.873780314059338e-05,
      "loss": 1.1045,
      "num_input_tokens_seen": 14905224,
      "step": 25705
    },
    {
      "epoch": 3.829311885612154,
      "grad_norm": 12.6875,
      "learning_rate": 4.873678349800054e-05,
      "loss": 1.6017,
      "num_input_tokens_seen": 14908200,
      "step": 25710
    },
    {
      "epoch": 3.8300565981531127,
      "grad_norm": 5.9375,
      "learning_rate": 4.873576345439886e-05,
      "loss": 0.9797,
      "num_input_tokens_seen": 14910888,
      "step": 25715
    },
    {
      "epoch": 3.830801310694072,
      "grad_norm": 9.8125,
      "learning_rate": 4.873474300980558e-05,
      "loss": 1.0813,
      "num_input_tokens_seen": 14913704,
      "step": 25720
    },
    {
      "epoch": 3.831546023235031,
      "grad_norm": 10.125,
      "learning_rate": 4.873372216423794e-05,
      "loss": 1.1062,
      "num_input_tokens_seen": 14916584,
      "step": 25725
    },
    {
      "epoch": 3.8322907357759903,
      "grad_norm": 10.625,
      "learning_rate": 4.8732700917713186e-05,
      "loss": 1.0217,
      "num_input_tokens_seen": 14919240,
      "step": 25730
    },
    {
      "epoch": 3.8330354483169495,
      "grad_norm": 8.3125,
      "learning_rate": 4.8731679270248575e-05,
      "loss": 0.8694,
      "num_input_tokens_seen": 14922120,
      "step": 25735
    },
    {
      "epoch": 3.8337801608579087,
      "grad_norm": 20.375,
      "learning_rate": 4.8730657221861354e-05,
      "loss": 1.3663,
      "num_input_tokens_seen": 14925192,
      "step": 25740
    },
    {
      "epoch": 3.834524873398868,
      "grad_norm": 8.625,
      "learning_rate": 4.8729634772568805e-05,
      "loss": 1.0121,
      "num_input_tokens_seen": 14928424,
      "step": 25745
    },
    {
      "epoch": 3.835269585939827,
      "grad_norm": 7.5625,
      "learning_rate": 4.872861192238819e-05,
      "loss": 1.0584,
      "num_input_tokens_seen": 14931432,
      "step": 25750
    },
    {
      "epoch": 3.8360142984807863,
      "grad_norm": 9.6875,
      "learning_rate": 4.87275886713368e-05,
      "loss": 1.2113,
      "num_input_tokens_seen": 14934376,
      "step": 25755
    },
    {
      "epoch": 3.8367590110217455,
      "grad_norm": 5.03125,
      "learning_rate": 4.872656501943191e-05,
      "loss": 0.9792,
      "num_input_tokens_seen": 14937352,
      "step": 25760
    },
    {
      "epoch": 3.8375037235627047,
      "grad_norm": 7.96875,
      "learning_rate": 4.872554096669082e-05,
      "loss": 1.0017,
      "num_input_tokens_seen": 14940200,
      "step": 25765
    },
    {
      "epoch": 3.838248436103664,
      "grad_norm": 9.1875,
      "learning_rate": 4.8724516513130826e-05,
      "loss": 1.086,
      "num_input_tokens_seen": 14943432,
      "step": 25770
    },
    {
      "epoch": 3.838993148644623,
      "grad_norm": 6.53125,
      "learning_rate": 4.872349165876924e-05,
      "loss": 1.1437,
      "num_input_tokens_seen": 14946120,
      "step": 25775
    },
    {
      "epoch": 3.8397378611855824,
      "grad_norm": 9.125,
      "learning_rate": 4.872246640362337e-05,
      "loss": 0.8256,
      "num_input_tokens_seen": 14949224,
      "step": 25780
    },
    {
      "epoch": 3.8404825737265416,
      "grad_norm": 8.875,
      "learning_rate": 4.872144074771054e-05,
      "loss": 1.0119,
      "num_input_tokens_seen": 14951912,
      "step": 25785
    },
    {
      "epoch": 3.8412272862675008,
      "grad_norm": 7.53125,
      "learning_rate": 4.872041469104809e-05,
      "loss": 0.9835,
      "num_input_tokens_seen": 14954792,
      "step": 25790
    },
    {
      "epoch": 3.84197199880846,
      "grad_norm": 14.625,
      "learning_rate": 4.871938823365333e-05,
      "loss": 2.0829,
      "num_input_tokens_seen": 14957928,
      "step": 25795
    },
    {
      "epoch": 3.842716711349419,
      "grad_norm": 12.375,
      "learning_rate": 4.871836137554362e-05,
      "loss": 0.9125,
      "num_input_tokens_seen": 14960904,
      "step": 25800
    },
    {
      "epoch": 3.8434614238903784,
      "grad_norm": 11.875,
      "learning_rate": 4.8717334116736293e-05,
      "loss": 0.8184,
      "num_input_tokens_seen": 14963944,
      "step": 25805
    },
    {
      "epoch": 3.8442061364313376,
      "grad_norm": 6.5625,
      "learning_rate": 4.8716306457248717e-05,
      "loss": 1.306,
      "num_input_tokens_seen": 14966728,
      "step": 25810
    },
    {
      "epoch": 3.8449508489722968,
      "grad_norm": 16.25,
      "learning_rate": 4.871527839709825e-05,
      "loss": 1.1604,
      "num_input_tokens_seen": 14969320,
      "step": 25815
    },
    {
      "epoch": 3.845695561513256,
      "grad_norm": 15.5625,
      "learning_rate": 4.871424993630226e-05,
      "loss": 1.1019,
      "num_input_tokens_seen": 14973576,
      "step": 25820
    },
    {
      "epoch": 3.846440274054215,
      "grad_norm": 19.0,
      "learning_rate": 4.871322107487811e-05,
      "loss": 0.9496,
      "num_input_tokens_seen": 14976488,
      "step": 25825
    },
    {
      "epoch": 3.8471849865951744,
      "grad_norm": 7.53125,
      "learning_rate": 4.8712191812843194e-05,
      "loss": 1.255,
      "num_input_tokens_seen": 14979528,
      "step": 25830
    },
    {
      "epoch": 3.8479296991361336,
      "grad_norm": 19.625,
      "learning_rate": 4.87111621502149e-05,
      "loss": 1.18,
      "num_input_tokens_seen": 14982536,
      "step": 25835
    },
    {
      "epoch": 3.848674411677093,
      "grad_norm": 16.125,
      "learning_rate": 4.871013208701062e-05,
      "loss": 0.9963,
      "num_input_tokens_seen": 14985640,
      "step": 25840
    },
    {
      "epoch": 3.849419124218052,
      "grad_norm": 5.28125,
      "learning_rate": 4.870910162324776e-05,
      "loss": 1.0278,
      "num_input_tokens_seen": 14988584,
      "step": 25845
    },
    {
      "epoch": 3.8501638367590107,
      "grad_norm": 12.75,
      "learning_rate": 4.8708070758943716e-05,
      "loss": 0.9914,
      "num_input_tokens_seen": 14991432,
      "step": 25850
    },
    {
      "epoch": 3.8509085492999704,
      "grad_norm": 24.25,
      "learning_rate": 4.870703949411591e-05,
      "loss": 1.5069,
      "num_input_tokens_seen": 14994536,
      "step": 25855
    },
    {
      "epoch": 3.851653261840929,
      "grad_norm": 6.03125,
      "learning_rate": 4.8706007828781776e-05,
      "loss": 0.9911,
      "num_input_tokens_seen": 14997512,
      "step": 25860
    },
    {
      "epoch": 3.852397974381889,
      "grad_norm": 17.75,
      "learning_rate": 4.8704975762958734e-05,
      "loss": 1.2301,
      "num_input_tokens_seen": 15000264,
      "step": 25865
    },
    {
      "epoch": 3.8531426869228476,
      "grad_norm": 6.21875,
      "learning_rate": 4.8703943296664214e-05,
      "loss": 1.1753,
      "num_input_tokens_seen": 15003272,
      "step": 25870
    },
    {
      "epoch": 3.853887399463807,
      "grad_norm": 7.28125,
      "learning_rate": 4.8702910429915663e-05,
      "loss": 1.1315,
      "num_input_tokens_seen": 15006024,
      "step": 25875
    },
    {
      "epoch": 3.854632112004766,
      "grad_norm": 9.8125,
      "learning_rate": 4.870187716273054e-05,
      "loss": 0.8271,
      "num_input_tokens_seen": 15008808,
      "step": 25880
    },
    {
      "epoch": 3.8553768245457256,
      "grad_norm": 18.875,
      "learning_rate": 4.870084349512628e-05,
      "loss": 1.1737,
      "num_input_tokens_seen": 15011752,
      "step": 25885
    },
    {
      "epoch": 3.8561215370866844,
      "grad_norm": 10.8125,
      "learning_rate": 4.8699809427120364e-05,
      "loss": 1.0592,
      "num_input_tokens_seen": 15014536,
      "step": 25890
    },
    {
      "epoch": 3.8568662496276436,
      "grad_norm": 18.0,
      "learning_rate": 4.869877495873025e-05,
      "loss": 1.4208,
      "num_input_tokens_seen": 15017736,
      "step": 25895
    },
    {
      "epoch": 3.8576109621686028,
      "grad_norm": 10.0,
      "learning_rate": 4.869774008997343e-05,
      "loss": 1.0327,
      "num_input_tokens_seen": 15020744,
      "step": 25900
    },
    {
      "epoch": 3.858355674709562,
      "grad_norm": 7.59375,
      "learning_rate": 4.869670482086737e-05,
      "loss": 1.1259,
      "num_input_tokens_seen": 15023688,
      "step": 25905
    },
    {
      "epoch": 3.859100387250521,
      "grad_norm": 14.75,
      "learning_rate": 4.869566915142956e-05,
      "loss": 1.4156,
      "num_input_tokens_seen": 15026664,
      "step": 25910
    },
    {
      "epoch": 3.8598450997914804,
      "grad_norm": 10.125,
      "learning_rate": 4.8694633081677507e-05,
      "loss": 1.257,
      "num_input_tokens_seen": 15029512,
      "step": 25915
    },
    {
      "epoch": 3.8605898123324396,
      "grad_norm": 8.3125,
      "learning_rate": 4.869359661162871e-05,
      "loss": 1.2572,
      "num_input_tokens_seen": 15032456,
      "step": 25920
    },
    {
      "epoch": 3.861334524873399,
      "grad_norm": 11.125,
      "learning_rate": 4.869255974130068e-05,
      "loss": 1.1424,
      "num_input_tokens_seen": 15035720,
      "step": 25925
    },
    {
      "epoch": 3.862079237414358,
      "grad_norm": 7.5625,
      "learning_rate": 4.869152247071094e-05,
      "loss": 0.8125,
      "num_input_tokens_seen": 15038536,
      "step": 25930
    },
    {
      "epoch": 3.862823949955317,
      "grad_norm": 20.125,
      "learning_rate": 4.8690484799877004e-05,
      "loss": 1.2907,
      "num_input_tokens_seen": 15041352,
      "step": 25935
    },
    {
      "epoch": 3.8635686624962764,
      "grad_norm": 9.25,
      "learning_rate": 4.86894467288164e-05,
      "loss": 1.0961,
      "num_input_tokens_seen": 15044616,
      "step": 25940
    },
    {
      "epoch": 3.8643133750372356,
      "grad_norm": 7.8125,
      "learning_rate": 4.868840825754667e-05,
      "loss": 1.0813,
      "num_input_tokens_seen": 15047496,
      "step": 25945
    },
    {
      "epoch": 3.865058087578195,
      "grad_norm": 4.125,
      "learning_rate": 4.868736938608536e-05,
      "loss": 0.9739,
      "num_input_tokens_seen": 15050376,
      "step": 25950
    },
    {
      "epoch": 3.865802800119154,
      "grad_norm": 11.375,
      "learning_rate": 4.8686330114450025e-05,
      "loss": 0.8984,
      "num_input_tokens_seen": 15053480,
      "step": 25955
    },
    {
      "epoch": 3.866547512660113,
      "grad_norm": 7.3125,
      "learning_rate": 4.868529044265821e-05,
      "loss": 1.1266,
      "num_input_tokens_seen": 15056328,
      "step": 25960
    },
    {
      "epoch": 3.8672922252010724,
      "grad_norm": 6.59375,
      "learning_rate": 4.868425037072749e-05,
      "loss": 0.9792,
      "num_input_tokens_seen": 15059080,
      "step": 25965
    },
    {
      "epoch": 3.8680369377420316,
      "grad_norm": 15.0,
      "learning_rate": 4.868320989867543e-05,
      "loss": 1.0103,
      "num_input_tokens_seen": 15061992,
      "step": 25970
    },
    {
      "epoch": 3.868781650282991,
      "grad_norm": 13.8125,
      "learning_rate": 4.868216902651961e-05,
      "loss": 1.5582,
      "num_input_tokens_seen": 15064968,
      "step": 25975
    },
    {
      "epoch": 3.86952636282395,
      "grad_norm": 8.125,
      "learning_rate": 4.8681127754277606e-05,
      "loss": 1.0694,
      "num_input_tokens_seen": 15068040,
      "step": 25980
    },
    {
      "epoch": 3.870271075364909,
      "grad_norm": 9.1875,
      "learning_rate": 4.868008608196702e-05,
      "loss": 0.9878,
      "num_input_tokens_seen": 15070792,
      "step": 25985
    },
    {
      "epoch": 3.8710157879058684,
      "grad_norm": 18.375,
      "learning_rate": 4.8679044009605455e-05,
      "loss": 1.3731,
      "num_input_tokens_seen": 15073672,
      "step": 25990
    },
    {
      "epoch": 3.8717605004468276,
      "grad_norm": 10.4375,
      "learning_rate": 4.867800153721051e-05,
      "loss": 1.0161,
      "num_input_tokens_seen": 15076872,
      "step": 25995
    },
    {
      "epoch": 3.872505212987787,
      "grad_norm": 13.6875,
      "learning_rate": 4.867695866479978e-05,
      "loss": 1.0936,
      "num_input_tokens_seen": 15079880,
      "step": 26000
    },
    {
      "epoch": 3.873249925528746,
      "grad_norm": 10.125,
      "learning_rate": 4.86759153923909e-05,
      "loss": 1.2715,
      "num_input_tokens_seen": 15082696,
      "step": 26005
    },
    {
      "epoch": 3.8739946380697052,
      "grad_norm": 8.125,
      "learning_rate": 4.86748717200015e-05,
      "loss": 0.7942,
      "num_input_tokens_seen": 15085608,
      "step": 26010
    },
    {
      "epoch": 3.8747393506106644,
      "grad_norm": 11.375,
      "learning_rate": 4.8673827647649206e-05,
      "loss": 1.1921,
      "num_input_tokens_seen": 15088328,
      "step": 26015
    },
    {
      "epoch": 3.8754840631516236,
      "grad_norm": 8.75,
      "learning_rate": 4.867278317535164e-05,
      "loss": 1.2804,
      "num_input_tokens_seen": 15091240,
      "step": 26020
    },
    {
      "epoch": 3.8762287756925824,
      "grad_norm": 6.25,
      "learning_rate": 4.867173830312648e-05,
      "loss": 0.9101,
      "num_input_tokens_seen": 15094088,
      "step": 26025
    },
    {
      "epoch": 3.876973488233542,
      "grad_norm": 12.6875,
      "learning_rate": 4.867069303099135e-05,
      "loss": 1.0838,
      "num_input_tokens_seen": 15097096,
      "step": 26030
    },
    {
      "epoch": 3.877718200774501,
      "grad_norm": 5.0625,
      "learning_rate": 4.8669647358963924e-05,
      "loss": 1.0391,
      "num_input_tokens_seen": 15099784,
      "step": 26035
    },
    {
      "epoch": 3.8784629133154604,
      "grad_norm": 8.5,
      "learning_rate": 4.866860128706186e-05,
      "loss": 1.3637,
      "num_input_tokens_seen": 15102888,
      "step": 26040
    },
    {
      "epoch": 3.879207625856419,
      "grad_norm": 7.09375,
      "learning_rate": 4.866755481530284e-05,
      "loss": 1.0124,
      "num_input_tokens_seen": 15105640,
      "step": 26045
    },
    {
      "epoch": 3.879952338397379,
      "grad_norm": 10.3125,
      "learning_rate": 4.866650794370452e-05,
      "loss": 1.1952,
      "num_input_tokens_seen": 15108584,
      "step": 26050
    },
    {
      "epoch": 3.8806970509383376,
      "grad_norm": 7.0,
      "learning_rate": 4.866546067228461e-05,
      "loss": 0.8688,
      "num_input_tokens_seen": 15111496,
      "step": 26055
    },
    {
      "epoch": 3.8814417634792973,
      "grad_norm": 8.6875,
      "learning_rate": 4.866441300106081e-05,
      "loss": 0.8549,
      "num_input_tokens_seen": 15114728,
      "step": 26060
    },
    {
      "epoch": 3.882186476020256,
      "grad_norm": 20.5,
      "learning_rate": 4.866336493005078e-05,
      "loss": 1.2595,
      "num_input_tokens_seen": 15117608,
      "step": 26065
    },
    {
      "epoch": 3.882931188561215,
      "grad_norm": 8.4375,
      "learning_rate": 4.866231645927226e-05,
      "loss": 1.2058,
      "num_input_tokens_seen": 15120456,
      "step": 26070
    },
    {
      "epoch": 3.8836759011021744,
      "grad_norm": 15.0625,
      "learning_rate": 4.866126758874295e-05,
      "loss": 1.214,
      "num_input_tokens_seen": 15123272,
      "step": 26075
    },
    {
      "epoch": 3.8844206136431336,
      "grad_norm": 10.625,
      "learning_rate": 4.8660218318480574e-05,
      "loss": 0.9658,
      "num_input_tokens_seen": 15126216,
      "step": 26080
    },
    {
      "epoch": 3.885165326184093,
      "grad_norm": 4.375,
      "learning_rate": 4.865916864850286e-05,
      "loss": 0.7769,
      "num_input_tokens_seen": 15128936,
      "step": 26085
    },
    {
      "epoch": 3.885910038725052,
      "grad_norm": 14.5,
      "learning_rate": 4.865811857882754e-05,
      "loss": 0.9955,
      "num_input_tokens_seen": 15131432,
      "step": 26090
    },
    {
      "epoch": 3.8866547512660112,
      "grad_norm": 8.6875,
      "learning_rate": 4.8657068109472345e-05,
      "loss": 0.953,
      "num_input_tokens_seen": 15134536,
      "step": 26095
    },
    {
      "epoch": 3.8873994638069704,
      "grad_norm": 17.0,
      "learning_rate": 4.8656017240455025e-05,
      "loss": 0.9549,
      "num_input_tokens_seen": 15137320,
      "step": 26100
    },
    {
      "epoch": 3.8881441763479296,
      "grad_norm": 9.9375,
      "learning_rate": 4.865496597179334e-05,
      "loss": 1.3217,
      "num_input_tokens_seen": 15140200,
      "step": 26105
    },
    {
      "epoch": 3.888888888888889,
      "grad_norm": 8.125,
      "learning_rate": 4.8653914303505054e-05,
      "loss": 1.0014,
      "num_input_tokens_seen": 15143272,
      "step": 26110
    },
    {
      "epoch": 3.889633601429848,
      "grad_norm": 7.75,
      "learning_rate": 4.865286223560792e-05,
      "loss": 1.1671,
      "num_input_tokens_seen": 15146088,
      "step": 26115
    },
    {
      "epoch": 3.8903783139708072,
      "grad_norm": 17.0,
      "learning_rate": 4.865180976811972e-05,
      "loss": 1.0278,
      "num_input_tokens_seen": 15148936,
      "step": 26120
    },
    {
      "epoch": 3.8911230265117664,
      "grad_norm": 17.375,
      "learning_rate": 4.8650756901058225e-05,
      "loss": 1.3319,
      "num_input_tokens_seen": 15152040,
      "step": 26125
    },
    {
      "epoch": 3.8918677390527256,
      "grad_norm": 20.75,
      "learning_rate": 4.864970363444124e-05,
      "loss": 1.1719,
      "num_input_tokens_seen": 15155080,
      "step": 26130
    },
    {
      "epoch": 3.892612451593685,
      "grad_norm": 5.875,
      "learning_rate": 4.864864996828654e-05,
      "loss": 1.0623,
      "num_input_tokens_seen": 15157768,
      "step": 26135
    },
    {
      "epoch": 3.893357164134644,
      "grad_norm": 9.6875,
      "learning_rate": 4.864759590261194e-05,
      "loss": 1.1062,
      "num_input_tokens_seen": 15160488,
      "step": 26140
    },
    {
      "epoch": 3.8941018766756033,
      "grad_norm": 16.5,
      "learning_rate": 4.8646541437435246e-05,
      "loss": 1.5562,
      "num_input_tokens_seen": 15163336,
      "step": 26145
    },
    {
      "epoch": 3.8948465892165625,
      "grad_norm": 9.375,
      "learning_rate": 4.8645486572774266e-05,
      "loss": 1.0345,
      "num_input_tokens_seen": 15166376,
      "step": 26150
    },
    {
      "epoch": 3.8955913017575217,
      "grad_norm": 13.75,
      "learning_rate": 4.8644431308646815e-05,
      "loss": 1.0392,
      "num_input_tokens_seen": 15169448,
      "step": 26155
    },
    {
      "epoch": 3.896336014298481,
      "grad_norm": 9.0,
      "learning_rate": 4.8643375645070735e-05,
      "loss": 1.0501,
      "num_input_tokens_seen": 15172392,
      "step": 26160
    },
    {
      "epoch": 3.89708072683944,
      "grad_norm": 13.3125,
      "learning_rate": 4.864231958206384e-05,
      "loss": 1.1872,
      "num_input_tokens_seen": 15176008,
      "step": 26165
    },
    {
      "epoch": 3.8978254393803993,
      "grad_norm": 10.625,
      "learning_rate": 4.8641263119644004e-05,
      "loss": 1.5491,
      "num_input_tokens_seen": 15178856,
      "step": 26170
    },
    {
      "epoch": 3.8985701519213585,
      "grad_norm": 8.1875,
      "learning_rate": 4.864020625782905e-05,
      "loss": 1.0762,
      "num_input_tokens_seen": 15181512,
      "step": 26175
    },
    {
      "epoch": 3.8993148644623177,
      "grad_norm": 8.625,
      "learning_rate": 4.863914899663683e-05,
      "loss": 0.9872,
      "num_input_tokens_seen": 15184392,
      "step": 26180
    },
    {
      "epoch": 3.900059577003277,
      "grad_norm": 9.0,
      "learning_rate": 4.8638091336085224e-05,
      "loss": 0.9577,
      "num_input_tokens_seen": 15187272,
      "step": 26185
    },
    {
      "epoch": 3.900804289544236,
      "grad_norm": 4.78125,
      "learning_rate": 4.863703327619208e-05,
      "loss": 1.068,
      "num_input_tokens_seen": 15189960,
      "step": 26190
    },
    {
      "epoch": 3.9015490020851953,
      "grad_norm": 11.0,
      "learning_rate": 4.863597481697528e-05,
      "loss": 1.0699,
      "num_input_tokens_seen": 15192648,
      "step": 26195
    },
    {
      "epoch": 3.902293714626154,
      "grad_norm": 14.5,
      "learning_rate": 4.8634915958452724e-05,
      "loss": 0.8546,
      "num_input_tokens_seen": 15195368,
      "step": 26200
    },
    {
      "epoch": 3.9030384271671137,
      "grad_norm": 5.53125,
      "learning_rate": 4.863385670064227e-05,
      "loss": 0.9282,
      "num_input_tokens_seen": 15198472,
      "step": 26205
    },
    {
      "epoch": 3.9037831397080724,
      "grad_norm": 7.6875,
      "learning_rate": 4.863279704356183e-05,
      "loss": 1.1005,
      "num_input_tokens_seen": 15201256,
      "step": 26210
    },
    {
      "epoch": 3.904527852249032,
      "grad_norm": 11.0,
      "learning_rate": 4.863173698722931e-05,
      "loss": 1.0975,
      "num_input_tokens_seen": 15204328,
      "step": 26215
    },
    {
      "epoch": 3.905272564789991,
      "grad_norm": 16.375,
      "learning_rate": 4.863067653166261e-05,
      "loss": 1.1089,
      "num_input_tokens_seen": 15207272,
      "step": 26220
    },
    {
      "epoch": 3.9060172773309505,
      "grad_norm": 7.4375,
      "learning_rate": 4.8629615676879634e-05,
      "loss": 0.9243,
      "num_input_tokens_seen": 15210376,
      "step": 26225
    },
    {
      "epoch": 3.9067619898719093,
      "grad_norm": 16.5,
      "learning_rate": 4.8628554422898334e-05,
      "loss": 1.2741,
      "num_input_tokens_seen": 15213256,
      "step": 26230
    },
    {
      "epoch": 3.907506702412869,
      "grad_norm": 9.4375,
      "learning_rate": 4.8627492769736616e-05,
      "loss": 1.4445,
      "num_input_tokens_seen": 15216072,
      "step": 26235
    },
    {
      "epoch": 3.9082514149538277,
      "grad_norm": 10.375,
      "learning_rate": 4.862643071741242e-05,
      "loss": 1.0956,
      "num_input_tokens_seen": 15219176,
      "step": 26240
    },
    {
      "epoch": 3.908996127494787,
      "grad_norm": 11.3125,
      "learning_rate": 4.8625368265943696e-05,
      "loss": 1.0172,
      "num_input_tokens_seen": 15222408,
      "step": 26245
    },
    {
      "epoch": 3.909740840035746,
      "grad_norm": 10.5625,
      "learning_rate": 4.8624305415348374e-05,
      "loss": 1.2203,
      "num_input_tokens_seen": 15225384,
      "step": 26250
    },
    {
      "epoch": 3.9104855525767053,
      "grad_norm": 17.375,
      "learning_rate": 4.8623242165644436e-05,
      "loss": 1.0379,
      "num_input_tokens_seen": 15228264,
      "step": 26255
    },
    {
      "epoch": 3.9112302651176645,
      "grad_norm": 9.0,
      "learning_rate": 4.8622178516849824e-05,
      "loss": 0.7972,
      "num_input_tokens_seen": 15231176,
      "step": 26260
    },
    {
      "epoch": 3.9119749776586237,
      "grad_norm": 19.0,
      "learning_rate": 4.862111446898252e-05,
      "loss": 1.0566,
      "num_input_tokens_seen": 15233960,
      "step": 26265
    },
    {
      "epoch": 3.912719690199583,
      "grad_norm": 16.875,
      "learning_rate": 4.862005002206049e-05,
      "loss": 1.0104,
      "num_input_tokens_seen": 15236584,
      "step": 26270
    },
    {
      "epoch": 3.913464402740542,
      "grad_norm": 20.25,
      "learning_rate": 4.8618985176101716e-05,
      "loss": 1.1478,
      "num_input_tokens_seen": 15239528,
      "step": 26275
    },
    {
      "epoch": 3.9142091152815013,
      "grad_norm": 8.375,
      "learning_rate": 4.86179199311242e-05,
      "loss": 1.1948,
      "num_input_tokens_seen": 15242728,
      "step": 26280
    },
    {
      "epoch": 3.9149538278224605,
      "grad_norm": 7.5625,
      "learning_rate": 4.861685428714593e-05,
      "loss": 1.2637,
      "num_input_tokens_seen": 15245352,
      "step": 26285
    },
    {
      "epoch": 3.9156985403634197,
      "grad_norm": 13.5,
      "learning_rate": 4.861578824418491e-05,
      "loss": 1.3541,
      "num_input_tokens_seen": 15248360,
      "step": 26290
    },
    {
      "epoch": 3.916443252904379,
      "grad_norm": 7.9375,
      "learning_rate": 4.861472180225915e-05,
      "loss": 1.2113,
      "num_input_tokens_seen": 15251112,
      "step": 26295
    },
    {
      "epoch": 3.917187965445338,
      "grad_norm": 6.4375,
      "learning_rate": 4.861365496138667e-05,
      "loss": 0.8087,
      "num_input_tokens_seen": 15253736,
      "step": 26300
    },
    {
      "epoch": 3.9179326779862973,
      "grad_norm": 8.75,
      "learning_rate": 4.861258772158548e-05,
      "loss": 0.7386,
      "num_input_tokens_seen": 15256648,
      "step": 26305
    },
    {
      "epoch": 3.9186773905272565,
      "grad_norm": 7.46875,
      "learning_rate": 4.861152008287362e-05,
      "loss": 1.1762,
      "num_input_tokens_seen": 15259624,
      "step": 26310
    },
    {
      "epoch": 3.9194221030682157,
      "grad_norm": 10.4375,
      "learning_rate": 4.861045204526913e-05,
      "loss": 0.9005,
      "num_input_tokens_seen": 15262696,
      "step": 26315
    },
    {
      "epoch": 3.920166815609175,
      "grad_norm": 24.125,
      "learning_rate": 4.8609383608790046e-05,
      "loss": 1.345,
      "num_input_tokens_seen": 15265352,
      "step": 26320
    },
    {
      "epoch": 3.920911528150134,
      "grad_norm": 13.1875,
      "learning_rate": 4.860831477345443e-05,
      "loss": 1.3619,
      "num_input_tokens_seen": 15268200,
      "step": 26325
    },
    {
      "epoch": 3.9216562406910933,
      "grad_norm": 12.125,
      "learning_rate": 4.860724553928032e-05,
      "loss": 1.0754,
      "num_input_tokens_seen": 15271240,
      "step": 26330
    },
    {
      "epoch": 3.9224009532320525,
      "grad_norm": 10.1875,
      "learning_rate": 4.86061759062858e-05,
      "loss": 0.9704,
      "num_input_tokens_seen": 15274472,
      "step": 26335
    },
    {
      "epoch": 3.9231456657730117,
      "grad_norm": 10.4375,
      "learning_rate": 4.8605105874488924e-05,
      "loss": 1.7127,
      "num_input_tokens_seen": 15277544,
      "step": 26340
    },
    {
      "epoch": 3.923890378313971,
      "grad_norm": 8.5,
      "learning_rate": 4.8604035443907775e-05,
      "loss": 1.1335,
      "num_input_tokens_seen": 15280456,
      "step": 26345
    },
    {
      "epoch": 3.92463509085493,
      "grad_norm": 17.75,
      "learning_rate": 4.860296461456044e-05,
      "loss": 1.5198,
      "num_input_tokens_seen": 15283560,
      "step": 26350
    },
    {
      "epoch": 3.9253798033958893,
      "grad_norm": 9.125,
      "learning_rate": 4.8601893386465e-05,
      "loss": 1.1945,
      "num_input_tokens_seen": 15286568,
      "step": 26355
    },
    {
      "epoch": 3.9261245159368485,
      "grad_norm": 16.375,
      "learning_rate": 4.860082175963957e-05,
      "loss": 1.2041,
      "num_input_tokens_seen": 15289160,
      "step": 26360
    },
    {
      "epoch": 3.9268692284778077,
      "grad_norm": 9.125,
      "learning_rate": 4.859974973410224e-05,
      "loss": 1.0412,
      "num_input_tokens_seen": 15292296,
      "step": 26365
    },
    {
      "epoch": 3.927613941018767,
      "grad_norm": 10.8125,
      "learning_rate": 4.8598677309871123e-05,
      "loss": 0.9022,
      "num_input_tokens_seen": 15295304,
      "step": 26370
    },
    {
      "epoch": 3.9283586535597257,
      "grad_norm": 13.1875,
      "learning_rate": 4.859760448696433e-05,
      "loss": 1.2407,
      "num_input_tokens_seen": 15297992,
      "step": 26375
    },
    {
      "epoch": 3.9291033661006853,
      "grad_norm": 10.8125,
      "learning_rate": 4.85965312654e-05,
      "loss": 0.8202,
      "num_input_tokens_seen": 15301128,
      "step": 26380
    },
    {
      "epoch": 3.929848078641644,
      "grad_norm": 5.40625,
      "learning_rate": 4.859545764519625e-05,
      "loss": 1.1909,
      "num_input_tokens_seen": 15304072,
      "step": 26385
    },
    {
      "epoch": 3.9305927911826037,
      "grad_norm": 7.1875,
      "learning_rate": 4.859438362637123e-05,
      "loss": 1.1117,
      "num_input_tokens_seen": 15306920,
      "step": 26390
    },
    {
      "epoch": 3.9313375037235625,
      "grad_norm": 11.75,
      "learning_rate": 4.8593309208943085e-05,
      "loss": 0.9162,
      "num_input_tokens_seen": 15309960,
      "step": 26395
    },
    {
      "epoch": 3.932082216264522,
      "grad_norm": 7.9375,
      "learning_rate": 4.859223439292995e-05,
      "loss": 1.0642,
      "num_input_tokens_seen": 15313032,
      "step": 26400
    },
    {
      "epoch": 3.932826928805481,
      "grad_norm": 17.25,
      "learning_rate": 4.859115917835e-05,
      "loss": 0.9027,
      "num_input_tokens_seen": 15315816,
      "step": 26405
    },
    {
      "epoch": 3.9335716413464406,
      "grad_norm": 11.25,
      "learning_rate": 4.859008356522139e-05,
      "loss": 1.2159,
      "num_input_tokens_seen": 15319080,
      "step": 26410
    },
    {
      "epoch": 3.9343163538873993,
      "grad_norm": 5.21875,
      "learning_rate": 4.8589007553562293e-05,
      "loss": 0.9707,
      "num_input_tokens_seen": 15322088,
      "step": 26415
    },
    {
      "epoch": 3.9350610664283585,
      "grad_norm": 10.8125,
      "learning_rate": 4.858793114339089e-05,
      "loss": 1.3925,
      "num_input_tokens_seen": 15325256,
      "step": 26420
    },
    {
      "epoch": 3.9358057789693177,
      "grad_norm": 9.1875,
      "learning_rate": 4.8586854334725365e-05,
      "loss": 0.932,
      "num_input_tokens_seen": 15328232,
      "step": 26425
    },
    {
      "epoch": 3.936550491510277,
      "grad_norm": 8.8125,
      "learning_rate": 4.8585777127583906e-05,
      "loss": 1.0168,
      "num_input_tokens_seen": 15330984,
      "step": 26430
    },
    {
      "epoch": 3.937295204051236,
      "grad_norm": 7.28125,
      "learning_rate": 4.858469952198471e-05,
      "loss": 1.0442,
      "num_input_tokens_seen": 15333768,
      "step": 26435
    },
    {
      "epoch": 3.9380399165921953,
      "grad_norm": 6.75,
      "learning_rate": 4.8583621517945995e-05,
      "loss": 0.9537,
      "num_input_tokens_seen": 15336712,
      "step": 26440
    },
    {
      "epoch": 3.9387846291331545,
      "grad_norm": 8.6875,
      "learning_rate": 4.858254311548596e-05,
      "loss": 1.181,
      "num_input_tokens_seen": 15339432,
      "step": 26445
    },
    {
      "epoch": 3.9395293416741137,
      "grad_norm": 8.6875,
      "learning_rate": 4.858146431462283e-05,
      "loss": 0.9066,
      "num_input_tokens_seen": 15342024,
      "step": 26450
    },
    {
      "epoch": 3.940274054215073,
      "grad_norm": 14.5625,
      "learning_rate": 4.858038511537482e-05,
      "loss": 1.0473,
      "num_input_tokens_seen": 15345096,
      "step": 26455
    },
    {
      "epoch": 3.941018766756032,
      "grad_norm": 14.25,
      "learning_rate": 4.857930551776017e-05,
      "loss": 1.1243,
      "num_input_tokens_seen": 15347848,
      "step": 26460
    },
    {
      "epoch": 3.9417634792969913,
      "grad_norm": 13.3125,
      "learning_rate": 4.857822552179713e-05,
      "loss": 1.3041,
      "num_input_tokens_seen": 15350792,
      "step": 26465
    },
    {
      "epoch": 3.9425081918379505,
      "grad_norm": 8.5625,
      "learning_rate": 4.857714512750392e-05,
      "loss": 0.8801,
      "num_input_tokens_seen": 15354056,
      "step": 26470
    },
    {
      "epoch": 3.9432529043789097,
      "grad_norm": 15.0,
      "learning_rate": 4.857606433489881e-05,
      "loss": 1.0247,
      "num_input_tokens_seen": 15357384,
      "step": 26475
    },
    {
      "epoch": 3.943997616919869,
      "grad_norm": 13.25,
      "learning_rate": 4.8574983144000055e-05,
      "loss": 1.2144,
      "num_input_tokens_seen": 15360392,
      "step": 26480
    },
    {
      "epoch": 3.944742329460828,
      "grad_norm": 10.9375,
      "learning_rate": 4.8573901554825915e-05,
      "loss": 1.0659,
      "num_input_tokens_seen": 15363496,
      "step": 26485
    },
    {
      "epoch": 3.9454870420017873,
      "grad_norm": 7.84375,
      "learning_rate": 4.857281956739468e-05,
      "loss": 0.9434,
      "num_input_tokens_seen": 15366248,
      "step": 26490
    },
    {
      "epoch": 3.9462317545427466,
      "grad_norm": 5.53125,
      "learning_rate": 4.8571737181724606e-05,
      "loss": 0.8226,
      "num_input_tokens_seen": 15369160,
      "step": 26495
    },
    {
      "epoch": 3.9469764670837058,
      "grad_norm": 19.125,
      "learning_rate": 4.8570654397834e-05,
      "loss": 1.285,
      "num_input_tokens_seen": 15372040,
      "step": 26500
    },
    {
      "epoch": 3.947721179624665,
      "grad_norm": 13.625,
      "learning_rate": 4.856957121574114e-05,
      "loss": 1.1173,
      "num_input_tokens_seen": 15375048,
      "step": 26505
    },
    {
      "epoch": 3.948465892165624,
      "grad_norm": 10.5625,
      "learning_rate": 4.856848763546433e-05,
      "loss": 1.0899,
      "num_input_tokens_seen": 15378056,
      "step": 26510
    },
    {
      "epoch": 3.9492106047065834,
      "grad_norm": 7.65625,
      "learning_rate": 4.856740365702187e-05,
      "loss": 1.1158,
      "num_input_tokens_seen": 15381000,
      "step": 26515
    },
    {
      "epoch": 3.9499553172475426,
      "grad_norm": 8.5625,
      "learning_rate": 4.8566319280432085e-05,
      "loss": 0.9171,
      "num_input_tokens_seen": 15384072,
      "step": 26520
    },
    {
      "epoch": 3.9507000297885018,
      "grad_norm": 10.6875,
      "learning_rate": 4.8565234505713276e-05,
      "loss": 0.9049,
      "num_input_tokens_seen": 15387144,
      "step": 26525
    },
    {
      "epoch": 3.951444742329461,
      "grad_norm": 11.25,
      "learning_rate": 4.856414933288379e-05,
      "loss": 1.1271,
      "num_input_tokens_seen": 15390184,
      "step": 26530
    },
    {
      "epoch": 3.95218945487042,
      "grad_norm": 16.125,
      "learning_rate": 4.856306376196195e-05,
      "loss": 0.9105,
      "num_input_tokens_seen": 15393416,
      "step": 26535
    },
    {
      "epoch": 3.9529341674113794,
      "grad_norm": 11.625,
      "learning_rate": 4.856197779296609e-05,
      "loss": 0.9606,
      "num_input_tokens_seen": 15396552,
      "step": 26540
    },
    {
      "epoch": 3.9536788799523386,
      "grad_norm": 7.3125,
      "learning_rate": 4.856089142591457e-05,
      "loss": 0.638,
      "num_input_tokens_seen": 15399176,
      "step": 26545
    },
    {
      "epoch": 3.9544235924932973,
      "grad_norm": 11.0,
      "learning_rate": 4.855980466082574e-05,
      "loss": 0.7648,
      "num_input_tokens_seen": 15402184,
      "step": 26550
    },
    {
      "epoch": 3.955168305034257,
      "grad_norm": 10.5625,
      "learning_rate": 4.855871749771794e-05,
      "loss": 1.3123,
      "num_input_tokens_seen": 15404840,
      "step": 26555
    },
    {
      "epoch": 3.9559130175752157,
      "grad_norm": 9.3125,
      "learning_rate": 4.855762993660956e-05,
      "loss": 1.4936,
      "num_input_tokens_seen": 15407560,
      "step": 26560
    },
    {
      "epoch": 3.9566577301161754,
      "grad_norm": 7.09375,
      "learning_rate": 4.855654197751896e-05,
      "loss": 1.0454,
      "num_input_tokens_seen": 15410408,
      "step": 26565
    },
    {
      "epoch": 3.957402442657134,
      "grad_norm": 8.75,
      "learning_rate": 4.855545362046454e-05,
      "loss": 0.8369,
      "num_input_tokens_seen": 15413192,
      "step": 26570
    },
    {
      "epoch": 3.958147155198094,
      "grad_norm": 10.625,
      "learning_rate": 4.855436486546466e-05,
      "loss": 0.9907,
      "num_input_tokens_seen": 15415944,
      "step": 26575
    },
    {
      "epoch": 3.9588918677390526,
      "grad_norm": 7.375,
      "learning_rate": 4.855327571253773e-05,
      "loss": 1.1559,
      "num_input_tokens_seen": 15418600,
      "step": 26580
    },
    {
      "epoch": 3.9596365802800118,
      "grad_norm": 14.125,
      "learning_rate": 4.855218616170214e-05,
      "loss": 1.0789,
      "num_input_tokens_seen": 15421416,
      "step": 26585
    },
    {
      "epoch": 3.960381292820971,
      "grad_norm": 7.71875,
      "learning_rate": 4.855109621297631e-05,
      "loss": 0.8193,
      "num_input_tokens_seen": 15424392,
      "step": 26590
    },
    {
      "epoch": 3.96112600536193,
      "grad_norm": 9.3125,
      "learning_rate": 4.855000586637864e-05,
      "loss": 0.7872,
      "num_input_tokens_seen": 15427336,
      "step": 26595
    },
    {
      "epoch": 3.9618707179028894,
      "grad_norm": 17.25,
      "learning_rate": 4.854891512192755e-05,
      "loss": 1.29,
      "num_input_tokens_seen": 15430536,
      "step": 26600
    },
    {
      "epoch": 3.9626154304438486,
      "grad_norm": 9.6875,
      "learning_rate": 4.8547823979641484e-05,
      "loss": 1.1247,
      "num_input_tokens_seen": 15433480,
      "step": 26605
    },
    {
      "epoch": 3.9633601429848078,
      "grad_norm": 11.4375,
      "learning_rate": 4.854673243953886e-05,
      "loss": 1.1624,
      "num_input_tokens_seen": 15436648,
      "step": 26610
    },
    {
      "epoch": 3.964104855525767,
      "grad_norm": 7.4375,
      "learning_rate": 4.854564050163812e-05,
      "loss": 1.006,
      "num_input_tokens_seen": 15439592,
      "step": 26615
    },
    {
      "epoch": 3.964849568066726,
      "grad_norm": 6.96875,
      "learning_rate": 4.854454816595773e-05,
      "loss": 0.8972,
      "num_input_tokens_seen": 15442408,
      "step": 26620
    },
    {
      "epoch": 3.9655942806076854,
      "grad_norm": 10.875,
      "learning_rate": 4.854345543251611e-05,
      "loss": 1.2071,
      "num_input_tokens_seen": 15445160,
      "step": 26625
    },
    {
      "epoch": 3.9663389931486446,
      "grad_norm": 7.53125,
      "learning_rate": 4.854236230133175e-05,
      "loss": 0.8914,
      "num_input_tokens_seen": 15447944,
      "step": 26630
    },
    {
      "epoch": 3.967083705689604,
      "grad_norm": 10.125,
      "learning_rate": 4.85412687724231e-05,
      "loss": 1.1952,
      "num_input_tokens_seen": 15450824,
      "step": 26635
    },
    {
      "epoch": 3.967828418230563,
      "grad_norm": 17.625,
      "learning_rate": 4.854017484580864e-05,
      "loss": 1.025,
      "num_input_tokens_seen": 15453640,
      "step": 26640
    },
    {
      "epoch": 3.968573130771522,
      "grad_norm": 12.9375,
      "learning_rate": 4.853908052150685e-05,
      "loss": 1.2581,
      "num_input_tokens_seen": 15456552,
      "step": 26645
    },
    {
      "epoch": 3.9693178433124814,
      "grad_norm": 13.0625,
      "learning_rate": 4.8537985799536226e-05,
      "loss": 1.227,
      "num_input_tokens_seen": 15459496,
      "step": 26650
    },
    {
      "epoch": 3.9700625558534406,
      "grad_norm": 5.5625,
      "learning_rate": 4.853689067991525e-05,
      "loss": 1.1934,
      "num_input_tokens_seen": 15462568,
      "step": 26655
    },
    {
      "epoch": 3.9708072683944,
      "grad_norm": 8.5625,
      "learning_rate": 4.853579516266243e-05,
      "loss": 1.3521,
      "num_input_tokens_seen": 15465672,
      "step": 26660
    },
    {
      "epoch": 3.971551980935359,
      "grad_norm": 5.34375,
      "learning_rate": 4.853469924779627e-05,
      "loss": 1.0513,
      "num_input_tokens_seen": 15468488,
      "step": 26665
    },
    {
      "epoch": 3.972296693476318,
      "grad_norm": 9.0625,
      "learning_rate": 4.853360293533529e-05,
      "loss": 0.893,
      "num_input_tokens_seen": 15471272,
      "step": 26670
    },
    {
      "epoch": 3.9730414060172774,
      "grad_norm": 12.8125,
      "learning_rate": 4.8532506225298004e-05,
      "loss": 0.9388,
      "num_input_tokens_seen": 15474024,
      "step": 26675
    },
    {
      "epoch": 3.9737861185582366,
      "grad_norm": 9.9375,
      "learning_rate": 4.853140911770294e-05,
      "loss": 1.1484,
      "num_input_tokens_seen": 15476872,
      "step": 26680
    },
    {
      "epoch": 3.974530831099196,
      "grad_norm": 8.375,
      "learning_rate": 4.853031161256863e-05,
      "loss": 0.917,
      "num_input_tokens_seen": 15479496,
      "step": 26685
    },
    {
      "epoch": 3.975275543640155,
      "grad_norm": 16.125,
      "learning_rate": 4.8529213709913626e-05,
      "loss": 1.1035,
      "num_input_tokens_seen": 15482664,
      "step": 26690
    },
    {
      "epoch": 3.976020256181114,
      "grad_norm": 7.34375,
      "learning_rate": 4.852811540975647e-05,
      "loss": 0.852,
      "num_input_tokens_seen": 15485512,
      "step": 26695
    },
    {
      "epoch": 3.9767649687220734,
      "grad_norm": 13.375,
      "learning_rate": 4.8527016712115725e-05,
      "loss": 1.1985,
      "num_input_tokens_seen": 15488552,
      "step": 26700
    },
    {
      "epoch": 3.9775096812630326,
      "grad_norm": 19.375,
      "learning_rate": 4.8525917617009945e-05,
      "loss": 1.1453,
      "num_input_tokens_seen": 15491240,
      "step": 26705
    },
    {
      "epoch": 3.978254393803992,
      "grad_norm": 6.09375,
      "learning_rate": 4.8524818124457684e-05,
      "loss": 1.1491,
      "num_input_tokens_seen": 15494184,
      "step": 26710
    },
    {
      "epoch": 3.9789991063449506,
      "grad_norm": 14.0625,
      "learning_rate": 4.852371823447753e-05,
      "loss": 1.2087,
      "num_input_tokens_seen": 15497064,
      "step": 26715
    },
    {
      "epoch": 3.9797438188859102,
      "grad_norm": 9.5625,
      "learning_rate": 4.852261794708808e-05,
      "loss": 1.0526,
      "num_input_tokens_seen": 15499880,
      "step": 26720
    },
    {
      "epoch": 3.980488531426869,
      "grad_norm": 7.5,
      "learning_rate": 4.8521517262307895e-05,
      "loss": 1.2082,
      "num_input_tokens_seen": 15502824,
      "step": 26725
    },
    {
      "epoch": 3.9812332439678286,
      "grad_norm": 5.9375,
      "learning_rate": 4.8520416180155594e-05,
      "loss": 1.042,
      "num_input_tokens_seen": 15505864,
      "step": 26730
    },
    {
      "epoch": 3.9819779565087874,
      "grad_norm": 17.375,
      "learning_rate": 4.8519314700649757e-05,
      "loss": 1.1696,
      "num_input_tokens_seen": 15508840,
      "step": 26735
    },
    {
      "epoch": 3.982722669049747,
      "grad_norm": 9.75,
      "learning_rate": 4.8518212823809e-05,
      "loss": 1.1505,
      "num_input_tokens_seen": 15511592,
      "step": 26740
    },
    {
      "epoch": 3.983467381590706,
      "grad_norm": 18.125,
      "learning_rate": 4.851711054965194e-05,
      "loss": 1.3133,
      "num_input_tokens_seen": 15514536,
      "step": 26745
    },
    {
      "epoch": 3.9842120941316654,
      "grad_norm": 9.3125,
      "learning_rate": 4.851600787819721e-05,
      "loss": 1.0577,
      "num_input_tokens_seen": 15517416,
      "step": 26750
    },
    {
      "epoch": 3.984956806672624,
      "grad_norm": 17.375,
      "learning_rate": 4.851490480946342e-05,
      "loss": 1.2733,
      "num_input_tokens_seen": 15520776,
      "step": 26755
    },
    {
      "epoch": 3.9857015192135834,
      "grad_norm": 12.0,
      "learning_rate": 4.851380134346921e-05,
      "loss": 1.2006,
      "num_input_tokens_seen": 15523464,
      "step": 26760
    },
    {
      "epoch": 3.9864462317545426,
      "grad_norm": 11.5625,
      "learning_rate": 4.851269748023323e-05,
      "loss": 1.2982,
      "num_input_tokens_seen": 15526312,
      "step": 26765
    },
    {
      "epoch": 3.987190944295502,
      "grad_norm": 13.875,
      "learning_rate": 4.851159321977412e-05,
      "loss": 1.2613,
      "num_input_tokens_seen": 15529352,
      "step": 26770
    },
    {
      "epoch": 3.987935656836461,
      "grad_norm": 7.0,
      "learning_rate": 4.851048856211054e-05,
      "loss": 1.0817,
      "num_input_tokens_seen": 15532104,
      "step": 26775
    },
    {
      "epoch": 3.98868036937742,
      "grad_norm": 9.5625,
      "learning_rate": 4.850938350726115e-05,
      "loss": 1.3152,
      "num_input_tokens_seen": 15534952,
      "step": 26780
    },
    {
      "epoch": 3.9894250819183794,
      "grad_norm": 8.0,
      "learning_rate": 4.8508278055244625e-05,
      "loss": 1.359,
      "num_input_tokens_seen": 15537672,
      "step": 26785
    },
    {
      "epoch": 3.9901697944593386,
      "grad_norm": 11.4375,
      "learning_rate": 4.8507172206079625e-05,
      "loss": 1.3451,
      "num_input_tokens_seen": 15541032,
      "step": 26790
    },
    {
      "epoch": 3.990914507000298,
      "grad_norm": 15.625,
      "learning_rate": 4.8506065959784854e-05,
      "loss": 1.4178,
      "num_input_tokens_seen": 15544008,
      "step": 26795
    },
    {
      "epoch": 3.991659219541257,
      "grad_norm": 6.03125,
      "learning_rate": 4.8504959316378974e-05,
      "loss": 1.1107,
      "num_input_tokens_seen": 15547048,
      "step": 26800
    },
    {
      "epoch": 3.9924039320822162,
      "grad_norm": 8.875,
      "learning_rate": 4.850385227588071e-05,
      "loss": 1.0191,
      "num_input_tokens_seen": 15549928,
      "step": 26805
    },
    {
      "epoch": 3.9931486446231754,
      "grad_norm": 9.0,
      "learning_rate": 4.8502744838308744e-05,
      "loss": 0.8394,
      "num_input_tokens_seen": 15552808,
      "step": 26810
    },
    {
      "epoch": 3.9938933571641346,
      "grad_norm": 10.1875,
      "learning_rate": 4.8501637003681786e-05,
      "loss": 1.0594,
      "num_input_tokens_seen": 15555560,
      "step": 26815
    },
    {
      "epoch": 3.994638069705094,
      "grad_norm": 8.0625,
      "learning_rate": 4.850052877201857e-05,
      "loss": 0.9402,
      "num_input_tokens_seen": 15558312,
      "step": 26820
    },
    {
      "epoch": 3.995382782246053,
      "grad_norm": 7.34375,
      "learning_rate": 4.8499420143337795e-05,
      "loss": 1.4599,
      "num_input_tokens_seen": 15561384,
      "step": 26825
    },
    {
      "epoch": 3.9961274947870122,
      "grad_norm": 8.875,
      "learning_rate": 4.84983111176582e-05,
      "loss": 1.3054,
      "num_input_tokens_seen": 15564296,
      "step": 26830
    },
    {
      "epoch": 3.9968722073279714,
      "grad_norm": 11.9375,
      "learning_rate": 4.8497201694998526e-05,
      "loss": 1.1134,
      "num_input_tokens_seen": 15567240,
      "step": 26835
    },
    {
      "epoch": 3.9976169198689306,
      "grad_norm": 12.6875,
      "learning_rate": 4.849609187537751e-05,
      "loss": 1.2281,
      "num_input_tokens_seen": 15570184,
      "step": 26840
    },
    {
      "epoch": 3.99836163240989,
      "grad_norm": 15.75,
      "learning_rate": 4.8494981658813895e-05,
      "loss": 1.1227,
      "num_input_tokens_seen": 15573320,
      "step": 26845
    },
    {
      "epoch": 3.999106344950849,
      "grad_norm": 7.1875,
      "learning_rate": 4.8493871045326455e-05,
      "loss": 1.0494,
      "num_input_tokens_seen": 15576104,
      "step": 26850
    },
    {
      "epoch": 3.9998510574918082,
      "grad_norm": 7.0625,
      "learning_rate": 4.849276003493394e-05,
      "loss": 1.3693,
      "num_input_tokens_seen": 15578728,
      "step": 26855
    },
    {
      "epoch": 4.0,
      "eval_loss": 1.0765974521636963,
      "eval_runtime": 43.727,
      "eval_samples_per_second": 68.242,
      "eval_steps_per_second": 17.06,
      "num_input_tokens_seen": 15578848,
      "step": 26856
    },
    {
      "epoch": 4.000595770032767,
      "grad_norm": 15.375,
      "learning_rate": 4.849164862765512e-05,
      "loss": 1.1187,
      "num_input_tokens_seen": 15581088,
      "step": 26860
    },
    {
      "epoch": 4.001340482573727,
      "grad_norm": 20.125,
      "learning_rate": 4.8490536823508767e-05,
      "loss": 1.2851,
      "num_input_tokens_seen": 15584096,
      "step": 26865
    },
    {
      "epoch": 4.002085195114685,
      "grad_norm": 11.4375,
      "learning_rate": 4.848942462251367e-05,
      "loss": 1.3869,
      "num_input_tokens_seen": 15586848,
      "step": 26870
    },
    {
      "epoch": 4.002829907655645,
      "grad_norm": 10.8125,
      "learning_rate": 4.848831202468862e-05,
      "loss": 1.085,
      "num_input_tokens_seen": 15589696,
      "step": 26875
    },
    {
      "epoch": 4.003574620196604,
      "grad_norm": 11.0625,
      "learning_rate": 4.848719903005241e-05,
      "loss": 0.8038,
      "num_input_tokens_seen": 15592512,
      "step": 26880
    },
    {
      "epoch": 4.0043193327375635,
      "grad_norm": 11.8125,
      "learning_rate": 4.848608563862385e-05,
      "loss": 1.2953,
      "num_input_tokens_seen": 15595296,
      "step": 26885
    },
    {
      "epoch": 4.005064045278522,
      "grad_norm": 9.125,
      "learning_rate": 4.848497185042173e-05,
      "loss": 1.2697,
      "num_input_tokens_seen": 15598176,
      "step": 26890
    },
    {
      "epoch": 4.005808757819482,
      "grad_norm": 15.375,
      "learning_rate": 4.84838576654649e-05,
      "loss": 1.1411,
      "num_input_tokens_seen": 15601120,
      "step": 26895
    },
    {
      "epoch": 4.006553470360441,
      "grad_norm": 10.5,
      "learning_rate": 4.848274308377214e-05,
      "loss": 0.9373,
      "num_input_tokens_seen": 15604064,
      "step": 26900
    },
    {
      "epoch": 4.0072981829014,
      "grad_norm": 10.0,
      "learning_rate": 4.8481628105362317e-05,
      "loss": 1.1296,
      "num_input_tokens_seen": 15607264,
      "step": 26905
    },
    {
      "epoch": 4.008042895442359,
      "grad_norm": 8.8125,
      "learning_rate": 4.848051273025425e-05,
      "loss": 1.2987,
      "num_input_tokens_seen": 15609920,
      "step": 26910
    },
    {
      "epoch": 4.008787607983319,
      "grad_norm": 8.25,
      "learning_rate": 4.8479396958466783e-05,
      "loss": 0.8199,
      "num_input_tokens_seen": 15612800,
      "step": 26915
    },
    {
      "epoch": 4.009532320524277,
      "grad_norm": 11.0625,
      "learning_rate": 4.8478280790018765e-05,
      "loss": 0.9712,
      "num_input_tokens_seen": 15615616,
      "step": 26920
    },
    {
      "epoch": 4.010277033065237,
      "grad_norm": 6.9375,
      "learning_rate": 4.847716422492906e-05,
      "loss": 0.9027,
      "num_input_tokens_seen": 15618528,
      "step": 26925
    },
    {
      "epoch": 4.011021745606196,
      "grad_norm": 9.25,
      "learning_rate": 4.847604726321652e-05,
      "loss": 1.1423,
      "num_input_tokens_seen": 15621408,
      "step": 26930
    },
    {
      "epoch": 4.0117664581471555,
      "grad_norm": 13.375,
      "learning_rate": 4.847492990490003e-05,
      "loss": 1.3264,
      "num_input_tokens_seen": 15624064,
      "step": 26935
    },
    {
      "epoch": 4.012511170688114,
      "grad_norm": 10.625,
      "learning_rate": 4.847381214999845e-05,
      "loss": 1.2311,
      "num_input_tokens_seen": 15627008,
      "step": 26940
    },
    {
      "epoch": 4.013255883229074,
      "grad_norm": 5.84375,
      "learning_rate": 4.847269399853068e-05,
      "loss": 1.0362,
      "num_input_tokens_seen": 15629888,
      "step": 26945
    },
    {
      "epoch": 4.014000595770033,
      "grad_norm": 10.0,
      "learning_rate": 4.84715754505156e-05,
      "loss": 1.1334,
      "num_input_tokens_seen": 15632736,
      "step": 26950
    },
    {
      "epoch": 4.014745308310992,
      "grad_norm": 10.9375,
      "learning_rate": 4.8470456505972105e-05,
      "loss": 0.9677,
      "num_input_tokens_seen": 15635680,
      "step": 26955
    },
    {
      "epoch": 4.015490020851951,
      "grad_norm": 9.75,
      "learning_rate": 4.8469337164919105e-05,
      "loss": 1.0791,
      "num_input_tokens_seen": 15638656,
      "step": 26960
    },
    {
      "epoch": 4.016234733392911,
      "grad_norm": 23.125,
      "learning_rate": 4.84682174273755e-05,
      "loss": 0.9305,
      "num_input_tokens_seen": 15641568,
      "step": 26965
    },
    {
      "epoch": 4.0169794459338695,
      "grad_norm": 9.5,
      "learning_rate": 4.846709729336022e-05,
      "loss": 1.2308,
      "num_input_tokens_seen": 15644768,
      "step": 26970
    },
    {
      "epoch": 4.017724158474829,
      "grad_norm": 12.0,
      "learning_rate": 4.846597676289218e-05,
      "loss": 0.9723,
      "num_input_tokens_seen": 15647904,
      "step": 26975
    },
    {
      "epoch": 4.018468871015788,
      "grad_norm": 10.5625,
      "learning_rate": 4.846485583599031e-05,
      "loss": 1.0945,
      "num_input_tokens_seen": 15650688,
      "step": 26980
    },
    {
      "epoch": 4.0192135835567475,
      "grad_norm": 10.4375,
      "learning_rate": 4.846373451267355e-05,
      "loss": 1.2323,
      "num_input_tokens_seen": 15653280,
      "step": 26985
    },
    {
      "epoch": 4.019958296097706,
      "grad_norm": 9.5,
      "learning_rate": 4.846261279296085e-05,
      "loss": 1.3804,
      "num_input_tokens_seen": 15656640,
      "step": 26990
    },
    {
      "epoch": 4.020703008638666,
      "grad_norm": 7.3125,
      "learning_rate": 4.8461490676871146e-05,
      "loss": 1.2316,
      "num_input_tokens_seen": 15659296,
      "step": 26995
    },
    {
      "epoch": 4.021447721179625,
      "grad_norm": 16.375,
      "learning_rate": 4.84603681644234e-05,
      "loss": 1.077,
      "num_input_tokens_seen": 15662080,
      "step": 27000
    },
    {
      "epoch": 4.022192433720583,
      "grad_norm": 8.5,
      "learning_rate": 4.8459245255636585e-05,
      "loss": 1.2646,
      "num_input_tokens_seen": 15664704,
      "step": 27005
    },
    {
      "epoch": 4.022937146261543,
      "grad_norm": 12.5,
      "learning_rate": 4.8458121950529654e-05,
      "loss": 1.0999,
      "num_input_tokens_seen": 15667392,
      "step": 27010
    },
    {
      "epoch": 4.023681858802502,
      "grad_norm": 12.4375,
      "learning_rate": 4.845699824912161e-05,
      "loss": 1.026,
      "num_input_tokens_seen": 15670272,
      "step": 27015
    },
    {
      "epoch": 4.0244265713434615,
      "grad_norm": 9.625,
      "learning_rate": 4.845587415143141e-05,
      "loss": 1.0079,
      "num_input_tokens_seen": 15673248,
      "step": 27020
    },
    {
      "epoch": 4.02517128388442,
      "grad_norm": 11.0625,
      "learning_rate": 4.845474965747806e-05,
      "loss": 0.9553,
      "num_input_tokens_seen": 15676032,
      "step": 27025
    },
    {
      "epoch": 4.02591599642538,
      "grad_norm": 16.875,
      "learning_rate": 4.8453624767280545e-05,
      "loss": 1.2613,
      "num_input_tokens_seen": 15679168,
      "step": 27030
    },
    {
      "epoch": 4.026660708966339,
      "grad_norm": 17.125,
      "learning_rate": 4.845249948085789e-05,
      "loss": 1.0684,
      "num_input_tokens_seen": 15681952,
      "step": 27035
    },
    {
      "epoch": 4.027405421507298,
      "grad_norm": 9.1875,
      "learning_rate": 4.8451373798229085e-05,
      "loss": 1.3557,
      "num_input_tokens_seen": 15684928,
      "step": 27040
    },
    {
      "epoch": 4.028150134048257,
      "grad_norm": 14.3125,
      "learning_rate": 4.845024771941316e-05,
      "loss": 0.7264,
      "num_input_tokens_seen": 15687808,
      "step": 27045
    },
    {
      "epoch": 4.028894846589217,
      "grad_norm": 5.8125,
      "learning_rate": 4.844912124442912e-05,
      "loss": 0.9923,
      "num_input_tokens_seen": 15690816,
      "step": 27050
    },
    {
      "epoch": 4.0296395591301755,
      "grad_norm": 6.71875,
      "learning_rate": 4.844799437329602e-05,
      "loss": 0.9985,
      "num_input_tokens_seen": 15693664,
      "step": 27055
    },
    {
      "epoch": 4.030384271671135,
      "grad_norm": 6.4375,
      "learning_rate": 4.844686710603289e-05,
      "loss": 0.9758,
      "num_input_tokens_seen": 15696512,
      "step": 27060
    },
    {
      "epoch": 4.031128984212094,
      "grad_norm": 6.375,
      "learning_rate": 4.844573944265876e-05,
      "loss": 0.8724,
      "num_input_tokens_seen": 15699392,
      "step": 27065
    },
    {
      "epoch": 4.0318736967530535,
      "grad_norm": 17.375,
      "learning_rate": 4.8444611383192695e-05,
      "loss": 1.6475,
      "num_input_tokens_seen": 15702528,
      "step": 27070
    },
    {
      "epoch": 4.032618409294012,
      "grad_norm": 9.625,
      "learning_rate": 4.844348292765375e-05,
      "loss": 0.7094,
      "num_input_tokens_seen": 15705536,
      "step": 27075
    },
    {
      "epoch": 4.033363121834972,
      "grad_norm": 9.3125,
      "learning_rate": 4.844235407606099e-05,
      "loss": 0.9882,
      "num_input_tokens_seen": 15708192,
      "step": 27080
    },
    {
      "epoch": 4.034107834375931,
      "grad_norm": 8.25,
      "learning_rate": 4.844122482843347e-05,
      "loss": 0.8821,
      "num_input_tokens_seen": 15711136,
      "step": 27085
    },
    {
      "epoch": 4.03485254691689,
      "grad_norm": 14.6875,
      "learning_rate": 4.8440095184790304e-05,
      "loss": 1.3123,
      "num_input_tokens_seen": 15713920,
      "step": 27090
    },
    {
      "epoch": 4.035597259457849,
      "grad_norm": 13.625,
      "learning_rate": 4.843896514515054e-05,
      "loss": 1.1786,
      "num_input_tokens_seen": 15716576,
      "step": 27095
    },
    {
      "epoch": 4.036341971998809,
      "grad_norm": 9.875,
      "learning_rate": 4.843783470953328e-05,
      "loss": 1.2181,
      "num_input_tokens_seen": 15719392,
      "step": 27100
    },
    {
      "epoch": 4.0370866845397675,
      "grad_norm": 8.25,
      "learning_rate": 4.843670387795763e-05,
      "loss": 1.0214,
      "num_input_tokens_seen": 15722368,
      "step": 27105
    },
    {
      "epoch": 4.037831397080727,
      "grad_norm": 10.5,
      "learning_rate": 4.843557265044268e-05,
      "loss": 1.2113,
      "num_input_tokens_seen": 15725568,
      "step": 27110
    },
    {
      "epoch": 4.038576109621686,
      "grad_norm": 10.8125,
      "learning_rate": 4.843444102700756e-05,
      "loss": 0.8754,
      "num_input_tokens_seen": 15728384,
      "step": 27115
    },
    {
      "epoch": 4.0393208221626455,
      "grad_norm": 11.5,
      "learning_rate": 4.843330900767137e-05,
      "loss": 1.1856,
      "num_input_tokens_seen": 15731104,
      "step": 27120
    },
    {
      "epoch": 4.040065534703604,
      "grad_norm": 13.3125,
      "learning_rate": 4.843217659245324e-05,
      "loss": 1.3779,
      "num_input_tokens_seen": 15733856,
      "step": 27125
    },
    {
      "epoch": 4.040810247244564,
      "grad_norm": 9.4375,
      "learning_rate": 4.843104378137231e-05,
      "loss": 1.1162,
      "num_input_tokens_seen": 15736704,
      "step": 27130
    },
    {
      "epoch": 4.041554959785523,
      "grad_norm": 3.828125,
      "learning_rate": 4.84299105744477e-05,
      "loss": 0.8108,
      "num_input_tokens_seen": 15739584,
      "step": 27135
    },
    {
      "epoch": 4.042299672326482,
      "grad_norm": 6.5,
      "learning_rate": 4.8428776971698566e-05,
      "loss": 0.6235,
      "num_input_tokens_seen": 15742464,
      "step": 27140
    },
    {
      "epoch": 4.043044384867441,
      "grad_norm": 12.6875,
      "learning_rate": 4.842764297314406e-05,
      "loss": 1.3741,
      "num_input_tokens_seen": 15745568,
      "step": 27145
    },
    {
      "epoch": 4.043789097408401,
      "grad_norm": 12.0,
      "learning_rate": 4.842650857880333e-05,
      "loss": 0.9245,
      "num_input_tokens_seen": 15748512,
      "step": 27150
    },
    {
      "epoch": 4.0445338099493595,
      "grad_norm": 9.75,
      "learning_rate": 4.842537378869556e-05,
      "loss": 0.9674,
      "num_input_tokens_seen": 15751328,
      "step": 27155
    },
    {
      "epoch": 4.045278522490319,
      "grad_norm": 14.5,
      "learning_rate": 4.84242386028399e-05,
      "loss": 1.2504,
      "num_input_tokens_seen": 15754176,
      "step": 27160
    },
    {
      "epoch": 4.046023235031278,
      "grad_norm": 8.3125,
      "learning_rate": 4.8423103021255535e-05,
      "loss": 0.9602,
      "num_input_tokens_seen": 15756992,
      "step": 27165
    },
    {
      "epoch": 4.046767947572237,
      "grad_norm": 4.4375,
      "learning_rate": 4.842196704396165e-05,
      "loss": 1.0468,
      "num_input_tokens_seen": 15759680,
      "step": 27170
    },
    {
      "epoch": 4.047512660113196,
      "grad_norm": 9.1875,
      "learning_rate": 4.842083067097744e-05,
      "loss": 1.3606,
      "num_input_tokens_seen": 15762848,
      "step": 27175
    },
    {
      "epoch": 4.048257372654155,
      "grad_norm": 8.25,
      "learning_rate": 4.84196939023221e-05,
      "loss": 1.1255,
      "num_input_tokens_seen": 15765888,
      "step": 27180
    },
    {
      "epoch": 4.049002085195115,
      "grad_norm": 11.25,
      "learning_rate": 4.841855673801483e-05,
      "loss": 1.0571,
      "num_input_tokens_seen": 15768896,
      "step": 27185
    },
    {
      "epoch": 4.0497467977360735,
      "grad_norm": 9.1875,
      "learning_rate": 4.8417419178074854e-05,
      "loss": 1.2881,
      "num_input_tokens_seen": 15772032,
      "step": 27190
    },
    {
      "epoch": 4.050491510277033,
      "grad_norm": 15.4375,
      "learning_rate": 4.841628122252138e-05,
      "loss": 1.2476,
      "num_input_tokens_seen": 15774848,
      "step": 27195
    },
    {
      "epoch": 4.051236222817992,
      "grad_norm": 15.25,
      "learning_rate": 4.841514287137362e-05,
      "loss": 1.2164,
      "num_input_tokens_seen": 15777792,
      "step": 27200
    },
    {
      "epoch": 4.0519809353589515,
      "grad_norm": 11.75,
      "learning_rate": 4.841400412465083e-05,
      "loss": 0.8497,
      "num_input_tokens_seen": 15780736,
      "step": 27205
    },
    {
      "epoch": 4.05272564789991,
      "grad_norm": 20.25,
      "learning_rate": 4.8412864982372244e-05,
      "loss": 1.0467,
      "num_input_tokens_seen": 15783488,
      "step": 27210
    },
    {
      "epoch": 4.05347036044087,
      "grad_norm": 5.59375,
      "learning_rate": 4.841172544455709e-05,
      "loss": 1.1176,
      "num_input_tokens_seen": 15786432,
      "step": 27215
    },
    {
      "epoch": 4.054215072981829,
      "grad_norm": 8.1875,
      "learning_rate": 4.841058551122463e-05,
      "loss": 0.9715,
      "num_input_tokens_seen": 15789184,
      "step": 27220
    },
    {
      "epoch": 4.054959785522788,
      "grad_norm": 9.6875,
      "learning_rate": 4.840944518239412e-05,
      "loss": 1.1099,
      "num_input_tokens_seen": 15792128,
      "step": 27225
    },
    {
      "epoch": 4.055704498063747,
      "grad_norm": 11.375,
      "learning_rate": 4.840830445808483e-05,
      "loss": 1.457,
      "num_input_tokens_seen": 15794944,
      "step": 27230
    },
    {
      "epoch": 4.056449210604707,
      "grad_norm": 9.125,
      "learning_rate": 4.840716333831602e-05,
      "loss": 1.3808,
      "num_input_tokens_seen": 15797408,
      "step": 27235
    },
    {
      "epoch": 4.0571939231456655,
      "grad_norm": 4.75,
      "learning_rate": 4.8406021823106985e-05,
      "loss": 1.2531,
      "num_input_tokens_seen": 15800544,
      "step": 27240
    },
    {
      "epoch": 4.057938635686625,
      "grad_norm": 10.875,
      "learning_rate": 4.8404879912477e-05,
      "loss": 1.1605,
      "num_input_tokens_seen": 15803520,
      "step": 27245
    },
    {
      "epoch": 4.058683348227584,
      "grad_norm": 4.4375,
      "learning_rate": 4.8403737606445355e-05,
      "loss": 1.2013,
      "num_input_tokens_seen": 15806592,
      "step": 27250
    },
    {
      "epoch": 4.059428060768544,
      "grad_norm": 11.0625,
      "learning_rate": 4.8402594905031346e-05,
      "loss": 0.9463,
      "num_input_tokens_seen": 15809408,
      "step": 27255
    },
    {
      "epoch": 4.060172773309502,
      "grad_norm": 4.875,
      "learning_rate": 4.840145180825428e-05,
      "loss": 1.1771,
      "num_input_tokens_seen": 15812320,
      "step": 27260
    },
    {
      "epoch": 4.060917485850462,
      "grad_norm": 5.59375,
      "learning_rate": 4.840030831613347e-05,
      "loss": 1.1273,
      "num_input_tokens_seen": 15815136,
      "step": 27265
    },
    {
      "epoch": 4.061662198391421,
      "grad_norm": 12.0,
      "learning_rate": 4.8399164428688244e-05,
      "loss": 0.8841,
      "num_input_tokens_seen": 15818080,
      "step": 27270
    },
    {
      "epoch": 4.06240691093238,
      "grad_norm": 13.4375,
      "learning_rate": 4.83980201459379e-05,
      "loss": 0.9489,
      "num_input_tokens_seen": 15820864,
      "step": 27275
    },
    {
      "epoch": 4.063151623473339,
      "grad_norm": 5.875,
      "learning_rate": 4.83968754679018e-05,
      "loss": 0.9105,
      "num_input_tokens_seen": 15823616,
      "step": 27280
    },
    {
      "epoch": 4.063896336014299,
      "grad_norm": 13.0,
      "learning_rate": 4.839573039459927e-05,
      "loss": 0.8107,
      "num_input_tokens_seen": 15826496,
      "step": 27285
    },
    {
      "epoch": 4.0646410485552575,
      "grad_norm": 11.8125,
      "learning_rate": 4.8394584926049644e-05,
      "loss": 1.0011,
      "num_input_tokens_seen": 15829344,
      "step": 27290
    },
    {
      "epoch": 4.065385761096217,
      "grad_norm": 5.21875,
      "learning_rate": 4.839343906227229e-05,
      "loss": 1.1031,
      "num_input_tokens_seen": 15832448,
      "step": 27295
    },
    {
      "epoch": 4.066130473637176,
      "grad_norm": 12.125,
      "learning_rate": 4.8392292803286554e-05,
      "loss": 0.9925,
      "num_input_tokens_seen": 15835488,
      "step": 27300
    },
    {
      "epoch": 4.066875186178136,
      "grad_norm": 10.375,
      "learning_rate": 4.839114614911181e-05,
      "loss": 1.415,
      "num_input_tokens_seen": 15838432,
      "step": 27305
    },
    {
      "epoch": 4.067619898719094,
      "grad_norm": 14.6875,
      "learning_rate": 4.838999909976742e-05,
      "loss": 0.8797,
      "num_input_tokens_seen": 15841504,
      "step": 27310
    },
    {
      "epoch": 4.068364611260054,
      "grad_norm": 16.5,
      "learning_rate": 4.838885165527277e-05,
      "loss": 1.0229,
      "num_input_tokens_seen": 15844256,
      "step": 27315
    },
    {
      "epoch": 4.069109323801013,
      "grad_norm": 9.8125,
      "learning_rate": 4.8387703815647245e-05,
      "loss": 1.1978,
      "num_input_tokens_seen": 15847488,
      "step": 27320
    },
    {
      "epoch": 4.069854036341972,
      "grad_norm": 10.5,
      "learning_rate": 4.838655558091024e-05,
      "loss": 1.2231,
      "num_input_tokens_seen": 15850240,
      "step": 27325
    },
    {
      "epoch": 4.070598748882931,
      "grad_norm": 13.875,
      "learning_rate": 4.8385406951081135e-05,
      "loss": 1.307,
      "num_input_tokens_seen": 15853504,
      "step": 27330
    },
    {
      "epoch": 4.071343461423891,
      "grad_norm": 11.9375,
      "learning_rate": 4.838425792617935e-05,
      "loss": 1.441,
      "num_input_tokens_seen": 15856512,
      "step": 27335
    },
    {
      "epoch": 4.07208817396485,
      "grad_norm": 10.625,
      "learning_rate": 4.8383108506224304e-05,
      "loss": 0.9456,
      "num_input_tokens_seen": 15859232,
      "step": 27340
    },
    {
      "epoch": 4.072832886505808,
      "grad_norm": 6.59375,
      "learning_rate": 4.8381958691235396e-05,
      "loss": 1.1745,
      "num_input_tokens_seen": 15862336,
      "step": 27345
    },
    {
      "epoch": 4.073577599046768,
      "grad_norm": 4.625,
      "learning_rate": 4.838080848123206e-05,
      "loss": 0.8955,
      "num_input_tokens_seen": 15865280,
      "step": 27350
    },
    {
      "epoch": 4.074322311587727,
      "grad_norm": 6.78125,
      "learning_rate": 4.837965787623373e-05,
      "loss": 0.7599,
      "num_input_tokens_seen": 15868192,
      "step": 27355
    },
    {
      "epoch": 4.075067024128686,
      "grad_norm": 4.84375,
      "learning_rate": 4.837850687625985e-05,
      "loss": 0.9797,
      "num_input_tokens_seen": 15871168,
      "step": 27360
    },
    {
      "epoch": 4.075811736669645,
      "grad_norm": 10.3125,
      "learning_rate": 4.8377355481329846e-05,
      "loss": 0.9879,
      "num_input_tokens_seen": 15874080,
      "step": 27365
    },
    {
      "epoch": 4.076556449210605,
      "grad_norm": 7.25,
      "learning_rate": 4.8376203691463184e-05,
      "loss": 0.8305,
      "num_input_tokens_seen": 15876896,
      "step": 27370
    },
    {
      "epoch": 4.0773011617515635,
      "grad_norm": 13.5,
      "learning_rate": 4.837505150667932e-05,
      "loss": 1.2221,
      "num_input_tokens_seen": 15879776,
      "step": 27375
    },
    {
      "epoch": 4.078045874292523,
      "grad_norm": 11.5625,
      "learning_rate": 4.837389892699772e-05,
      "loss": 1.0831,
      "num_input_tokens_seen": 15882336,
      "step": 27380
    },
    {
      "epoch": 4.078790586833482,
      "grad_norm": 18.875,
      "learning_rate": 4.837274595243785e-05,
      "loss": 1.1661,
      "num_input_tokens_seen": 15885120,
      "step": 27385
    },
    {
      "epoch": 4.079535299374442,
      "grad_norm": 5.96875,
      "learning_rate": 4.8371592583019196e-05,
      "loss": 1.1393,
      "num_input_tokens_seen": 15887840,
      "step": 27390
    },
    {
      "epoch": 4.0802800119154,
      "grad_norm": 13.125,
      "learning_rate": 4.8370438818761235e-05,
      "loss": 0.9734,
      "num_input_tokens_seen": 15890912,
      "step": 27395
    },
    {
      "epoch": 4.08102472445636,
      "grad_norm": 14.625,
      "learning_rate": 4.836928465968347e-05,
      "loss": 1.4506,
      "num_input_tokens_seen": 15893760,
      "step": 27400
    },
    {
      "epoch": 4.081769436997319,
      "grad_norm": 11.9375,
      "learning_rate": 4.836813010580538e-05,
      "loss": 1.0416,
      "num_input_tokens_seen": 15896352,
      "step": 27405
    },
    {
      "epoch": 4.082514149538278,
      "grad_norm": 5.09375,
      "learning_rate": 4.836697515714649e-05,
      "loss": 1.0631,
      "num_input_tokens_seen": 15899360,
      "step": 27410
    },
    {
      "epoch": 4.083258862079237,
      "grad_norm": 9.9375,
      "learning_rate": 4.8365819813726306e-05,
      "loss": 1.0653,
      "num_input_tokens_seen": 15902048,
      "step": 27415
    },
    {
      "epoch": 4.084003574620197,
      "grad_norm": 13.3125,
      "learning_rate": 4.8364664075564334e-05,
      "loss": 0.9635,
      "num_input_tokens_seen": 15904928,
      "step": 27420
    },
    {
      "epoch": 4.084748287161156,
      "grad_norm": 10.5625,
      "learning_rate": 4.836350794268012e-05,
      "loss": 0.9662,
      "num_input_tokens_seen": 15907872,
      "step": 27425
    },
    {
      "epoch": 4.085492999702115,
      "grad_norm": 15.0,
      "learning_rate": 4.836235141509318e-05,
      "loss": 1.054,
      "num_input_tokens_seen": 15910816,
      "step": 27430
    },
    {
      "epoch": 4.086237712243074,
      "grad_norm": 14.25,
      "learning_rate": 4.836119449282306e-05,
      "loss": 0.9869,
      "num_input_tokens_seen": 15913664,
      "step": 27435
    },
    {
      "epoch": 4.086982424784034,
      "grad_norm": 10.3125,
      "learning_rate": 4.8360037175889304e-05,
      "loss": 0.8446,
      "num_input_tokens_seen": 15916512,
      "step": 27440
    },
    {
      "epoch": 4.087727137324992,
      "grad_norm": 5.0,
      "learning_rate": 4.8358879464311455e-05,
      "loss": 1.2194,
      "num_input_tokens_seen": 15919552,
      "step": 27445
    },
    {
      "epoch": 4.088471849865952,
      "grad_norm": 10.25,
      "learning_rate": 4.835772135810909e-05,
      "loss": 0.6807,
      "num_input_tokens_seen": 15922592,
      "step": 27450
    },
    {
      "epoch": 4.089216562406911,
      "grad_norm": 9.625,
      "learning_rate": 4.8356562857301744e-05,
      "loss": 0.9177,
      "num_input_tokens_seen": 15925536,
      "step": 27455
    },
    {
      "epoch": 4.08996127494787,
      "grad_norm": 16.25,
      "learning_rate": 4.835540396190902e-05,
      "loss": 1.2057,
      "num_input_tokens_seen": 15928416,
      "step": 27460
    },
    {
      "epoch": 4.090705987488829,
      "grad_norm": 7.625,
      "learning_rate": 4.835424467195049e-05,
      "loss": 0.8477,
      "num_input_tokens_seen": 15931360,
      "step": 27465
    },
    {
      "epoch": 4.091450700029789,
      "grad_norm": 8.6875,
      "learning_rate": 4.835308498744572e-05,
      "loss": 1.1304,
      "num_input_tokens_seen": 15933984,
      "step": 27470
    },
    {
      "epoch": 4.092195412570748,
      "grad_norm": 12.6875,
      "learning_rate": 4.8351924908414314e-05,
      "loss": 1.1871,
      "num_input_tokens_seen": 15936896,
      "step": 27475
    },
    {
      "epoch": 4.092940125111707,
      "grad_norm": 13.125,
      "learning_rate": 4.835076443487587e-05,
      "loss": 0.7781,
      "num_input_tokens_seen": 15939936,
      "step": 27480
    },
    {
      "epoch": 4.093684837652666,
      "grad_norm": 8.3125,
      "learning_rate": 4.8349603566850003e-05,
      "loss": 0.8987,
      "num_input_tokens_seen": 15942784,
      "step": 27485
    },
    {
      "epoch": 4.094429550193626,
      "grad_norm": 9.6875,
      "learning_rate": 4.834844230435631e-05,
      "loss": 0.982,
      "num_input_tokens_seen": 15945600,
      "step": 27490
    },
    {
      "epoch": 4.095174262734584,
      "grad_norm": 17.25,
      "learning_rate": 4.8347280647414416e-05,
      "loss": 0.8592,
      "num_input_tokens_seen": 15948448,
      "step": 27495
    },
    {
      "epoch": 4.095918975275544,
      "grad_norm": 16.25,
      "learning_rate": 4.834611859604394e-05,
      "loss": 1.146,
      "num_input_tokens_seen": 15951232,
      "step": 27500
    },
    {
      "epoch": 4.096663687816503,
      "grad_norm": 14.8125,
      "learning_rate": 4.8344956150264524e-05,
      "loss": 0.8362,
      "num_input_tokens_seen": 15954240,
      "step": 27505
    },
    {
      "epoch": 4.0974084003574625,
      "grad_norm": 7.40625,
      "learning_rate": 4.83437933100958e-05,
      "loss": 0.7619,
      "num_input_tokens_seen": 15956864,
      "step": 27510
    },
    {
      "epoch": 4.098153112898421,
      "grad_norm": 11.6875,
      "learning_rate": 4.834263007555741e-05,
      "loss": 1.041,
      "num_input_tokens_seen": 15959904,
      "step": 27515
    },
    {
      "epoch": 4.09889782543938,
      "grad_norm": 9.8125,
      "learning_rate": 4.834146644666901e-05,
      "loss": 1.2542,
      "num_input_tokens_seen": 15962720,
      "step": 27520
    },
    {
      "epoch": 4.09964253798034,
      "grad_norm": 8.0625,
      "learning_rate": 4.834030242345026e-05,
      "loss": 1.0329,
      "num_input_tokens_seen": 15965408,
      "step": 27525
    },
    {
      "epoch": 4.100387250521298,
      "grad_norm": 10.625,
      "learning_rate": 4.8339138005920825e-05,
      "loss": 1.1367,
      "num_input_tokens_seen": 15968512,
      "step": 27530
    },
    {
      "epoch": 4.101131963062258,
      "grad_norm": 13.25,
      "learning_rate": 4.833797319410037e-05,
      "loss": 0.9049,
      "num_input_tokens_seen": 15971264,
      "step": 27535
    },
    {
      "epoch": 4.101876675603217,
      "grad_norm": 7.40625,
      "learning_rate": 4.833680798800858e-05,
      "loss": 1.2017,
      "num_input_tokens_seen": 15974144,
      "step": 27540
    },
    {
      "epoch": 4.102621388144176,
      "grad_norm": 11.375,
      "learning_rate": 4.833564238766513e-05,
      "loss": 1.1096,
      "num_input_tokens_seen": 15977216,
      "step": 27545
    },
    {
      "epoch": 4.103366100685135,
      "grad_norm": 15.3125,
      "learning_rate": 4.8334476393089726e-05,
      "loss": 1.1846,
      "num_input_tokens_seen": 15980096,
      "step": 27550
    },
    {
      "epoch": 4.104110813226095,
      "grad_norm": 8.625,
      "learning_rate": 4.8333310004302054e-05,
      "loss": 0.8135,
      "num_input_tokens_seen": 15982720,
      "step": 27555
    },
    {
      "epoch": 4.104855525767054,
      "grad_norm": 11.6875,
      "learning_rate": 4.833214322132183e-05,
      "loss": 1.0287,
      "num_input_tokens_seen": 15985344,
      "step": 27560
    },
    {
      "epoch": 4.105600238308013,
      "grad_norm": 16.875,
      "learning_rate": 4.8330976044168766e-05,
      "loss": 1.1968,
      "num_input_tokens_seen": 15988288,
      "step": 27565
    },
    {
      "epoch": 4.106344950848972,
      "grad_norm": 14.1875,
      "learning_rate": 4.832980847286256e-05,
      "loss": 1.1604,
      "num_input_tokens_seen": 15991616,
      "step": 27570
    },
    {
      "epoch": 4.107089663389932,
      "grad_norm": 4.6875,
      "learning_rate": 4.832864050742296e-05,
      "loss": 0.9778,
      "num_input_tokens_seen": 15994368,
      "step": 27575
    },
    {
      "epoch": 4.10783437593089,
      "grad_norm": 10.875,
      "learning_rate": 4.8327472147869684e-05,
      "loss": 1.3056,
      "num_input_tokens_seen": 15997184,
      "step": 27580
    },
    {
      "epoch": 4.10857908847185,
      "grad_norm": 11.4375,
      "learning_rate": 4.8326303394222476e-05,
      "loss": 1.1104,
      "num_input_tokens_seen": 16000288,
      "step": 27585
    },
    {
      "epoch": 4.109323801012809,
      "grad_norm": 9.75,
      "learning_rate": 4.832513424650108e-05,
      "loss": 1.6779,
      "num_input_tokens_seen": 16003040,
      "step": 27590
    },
    {
      "epoch": 4.1100685135537685,
      "grad_norm": 13.6875,
      "learning_rate": 4.8323964704725254e-05,
      "loss": 0.9076,
      "num_input_tokens_seen": 16005728,
      "step": 27595
    },
    {
      "epoch": 4.110813226094727,
      "grad_norm": 8.0625,
      "learning_rate": 4.8322794768914745e-05,
      "loss": 1.2933,
      "num_input_tokens_seen": 16008768,
      "step": 27600
    },
    {
      "epoch": 4.111557938635687,
      "grad_norm": 21.5,
      "learning_rate": 4.832162443908932e-05,
      "loss": 0.8451,
      "num_input_tokens_seen": 16011680,
      "step": 27605
    },
    {
      "epoch": 4.112302651176646,
      "grad_norm": 11.5,
      "learning_rate": 4.832045371526876e-05,
      "loss": 1.2339,
      "num_input_tokens_seen": 16014368,
      "step": 27610
    },
    {
      "epoch": 4.113047363717605,
      "grad_norm": 15.0625,
      "learning_rate": 4.8319282597472823e-05,
      "loss": 1.2165,
      "num_input_tokens_seen": 16017056,
      "step": 27615
    },
    {
      "epoch": 4.113792076258564,
      "grad_norm": 15.6875,
      "learning_rate": 4.8318111085721324e-05,
      "loss": 1.0431,
      "num_input_tokens_seen": 16020256,
      "step": 27620
    },
    {
      "epoch": 4.114536788799524,
      "grad_norm": 10.6875,
      "learning_rate": 4.8316939180034025e-05,
      "loss": 1.2657,
      "num_input_tokens_seen": 16023232,
      "step": 27625
    },
    {
      "epoch": 4.115281501340482,
      "grad_norm": 11.875,
      "learning_rate": 4.831576688043075e-05,
      "loss": 1.301,
      "num_input_tokens_seen": 16025728,
      "step": 27630
    },
    {
      "epoch": 4.116026213881442,
      "grad_norm": 21.125,
      "learning_rate": 4.831459418693128e-05,
      "loss": 1.1801,
      "num_input_tokens_seen": 16028384,
      "step": 27635
    },
    {
      "epoch": 4.116770926422401,
      "grad_norm": 8.5,
      "learning_rate": 4.8313421099555436e-05,
      "loss": 1.1114,
      "num_input_tokens_seen": 16031296,
      "step": 27640
    },
    {
      "epoch": 4.1175156389633605,
      "grad_norm": 17.125,
      "learning_rate": 4.831224761832304e-05,
      "loss": 0.9196,
      "num_input_tokens_seen": 16034048,
      "step": 27645
    },
    {
      "epoch": 4.118260351504319,
      "grad_norm": 10.125,
      "learning_rate": 4.831107374325391e-05,
      "loss": 1.095,
      "num_input_tokens_seen": 16037280,
      "step": 27650
    },
    {
      "epoch": 4.119005064045279,
      "grad_norm": 11.75,
      "learning_rate": 4.8309899474367894e-05,
      "loss": 0.9471,
      "num_input_tokens_seen": 16040000,
      "step": 27655
    },
    {
      "epoch": 4.119749776586238,
      "grad_norm": 11.5625,
      "learning_rate": 4.8308724811684805e-05,
      "loss": 1.3344,
      "num_input_tokens_seen": 16042720,
      "step": 27660
    },
    {
      "epoch": 4.120494489127197,
      "grad_norm": 5.4375,
      "learning_rate": 4.830754975522451e-05,
      "loss": 1.0013,
      "num_input_tokens_seen": 16045888,
      "step": 27665
    },
    {
      "epoch": 4.121239201668156,
      "grad_norm": 9.25,
      "learning_rate": 4.830637430500684e-05,
      "loss": 1.0636,
      "num_input_tokens_seen": 16048800,
      "step": 27670
    },
    {
      "epoch": 4.121983914209116,
      "grad_norm": 15.9375,
      "learning_rate": 4.830519846105167e-05,
      "loss": 1.4587,
      "num_input_tokens_seen": 16051872,
      "step": 27675
    },
    {
      "epoch": 4.1227286267500745,
      "grad_norm": 13.5,
      "learning_rate": 4.830402222337886e-05,
      "loss": 1.2491,
      "num_input_tokens_seen": 16054656,
      "step": 27680
    },
    {
      "epoch": 4.123473339291033,
      "grad_norm": 15.5625,
      "learning_rate": 4.830284559200828e-05,
      "loss": 1.2435,
      "num_input_tokens_seen": 16057472,
      "step": 27685
    },
    {
      "epoch": 4.124218051831993,
      "grad_norm": 10.75,
      "learning_rate": 4.83016685669598e-05,
      "loss": 0.8554,
      "num_input_tokens_seen": 16060512,
      "step": 27690
    },
    {
      "epoch": 4.124962764372952,
      "grad_norm": 11.4375,
      "learning_rate": 4.8300491148253315e-05,
      "loss": 1.1467,
      "num_input_tokens_seen": 16063392,
      "step": 27695
    },
    {
      "epoch": 4.125707476913911,
      "grad_norm": 4.5,
      "learning_rate": 4.829931333590872e-05,
      "loss": 0.8642,
      "num_input_tokens_seen": 16066464,
      "step": 27700
    },
    {
      "epoch": 4.12645218945487,
      "grad_norm": 11.9375,
      "learning_rate": 4.82981351299459e-05,
      "loss": 1.0062,
      "num_input_tokens_seen": 16069152,
      "step": 27705
    },
    {
      "epoch": 4.12719690199583,
      "grad_norm": 14.5,
      "learning_rate": 4.829695653038477e-05,
      "loss": 1.1141,
      "num_input_tokens_seen": 16071808,
      "step": 27710
    },
    {
      "epoch": 4.127941614536788,
      "grad_norm": 10.8125,
      "learning_rate": 4.829577753724523e-05,
      "loss": 1.064,
      "num_input_tokens_seen": 16074560,
      "step": 27715
    },
    {
      "epoch": 4.128686327077748,
      "grad_norm": 13.125,
      "learning_rate": 4.829459815054722e-05,
      "loss": 0.9287,
      "num_input_tokens_seen": 16077280,
      "step": 27720
    },
    {
      "epoch": 4.129431039618707,
      "grad_norm": 14.5625,
      "learning_rate": 4.829341837031064e-05,
      "loss": 0.8656,
      "num_input_tokens_seen": 16080320,
      "step": 27725
    },
    {
      "epoch": 4.1301757521596665,
      "grad_norm": 11.375,
      "learning_rate": 4.829223819655543e-05,
      "loss": 1.1867,
      "num_input_tokens_seen": 16083008,
      "step": 27730
    },
    {
      "epoch": 4.130920464700625,
      "grad_norm": 17.625,
      "learning_rate": 4.829105762930153e-05,
      "loss": 1.2709,
      "num_input_tokens_seen": 16085888,
      "step": 27735
    },
    {
      "epoch": 4.131665177241585,
      "grad_norm": 13.75,
      "learning_rate": 4.8289876668568886e-05,
      "loss": 1.2323,
      "num_input_tokens_seen": 16088800,
      "step": 27740
    },
    {
      "epoch": 4.132409889782544,
      "grad_norm": 15.125,
      "learning_rate": 4.828869531437744e-05,
      "loss": 1.2246,
      "num_input_tokens_seen": 16091840,
      "step": 27745
    },
    {
      "epoch": 4.133154602323503,
      "grad_norm": 4.84375,
      "learning_rate": 4.828751356674717e-05,
      "loss": 0.8554,
      "num_input_tokens_seen": 16094816,
      "step": 27750
    },
    {
      "epoch": 4.133899314864462,
      "grad_norm": 14.375,
      "learning_rate": 4.8286331425698014e-05,
      "loss": 1.326,
      "num_input_tokens_seen": 16097472,
      "step": 27755
    },
    {
      "epoch": 4.134644027405422,
      "grad_norm": 14.8125,
      "learning_rate": 4.828514889124995e-05,
      "loss": 1.4886,
      "num_input_tokens_seen": 16100576,
      "step": 27760
    },
    {
      "epoch": 4.1353887399463805,
      "grad_norm": 10.6875,
      "learning_rate": 4.828396596342298e-05,
      "loss": 1.1219,
      "num_input_tokens_seen": 16103680,
      "step": 27765
    },
    {
      "epoch": 4.13613345248734,
      "grad_norm": 11.0,
      "learning_rate": 4.828278264223706e-05,
      "loss": 1.174,
      "num_input_tokens_seen": 16106304,
      "step": 27770
    },
    {
      "epoch": 4.136878165028299,
      "grad_norm": 5.59375,
      "learning_rate": 4.828159892771219e-05,
      "loss": 1.1512,
      "num_input_tokens_seen": 16109376,
      "step": 27775
    },
    {
      "epoch": 4.1376228775692585,
      "grad_norm": 8.4375,
      "learning_rate": 4.828041481986837e-05,
      "loss": 0.9758,
      "num_input_tokens_seen": 16112160,
      "step": 27780
    },
    {
      "epoch": 4.138367590110217,
      "grad_norm": 11.625,
      "learning_rate": 4.82792303187256e-05,
      "loss": 0.9999,
      "num_input_tokens_seen": 16114880,
      "step": 27785
    },
    {
      "epoch": 4.139112302651177,
      "grad_norm": 14.0,
      "learning_rate": 4.82780454243039e-05,
      "loss": 1.0226,
      "num_input_tokens_seen": 16117568,
      "step": 27790
    },
    {
      "epoch": 4.139857015192136,
      "grad_norm": 7.8125,
      "learning_rate": 4.827686013662327e-05,
      "loss": 0.8065,
      "num_input_tokens_seen": 16120608,
      "step": 27795
    },
    {
      "epoch": 4.140601727733095,
      "grad_norm": 12.375,
      "learning_rate": 4.827567445570376e-05,
      "loss": 1.3566,
      "num_input_tokens_seen": 16123328,
      "step": 27800
    },
    {
      "epoch": 4.141346440274054,
      "grad_norm": 12.625,
      "learning_rate": 4.827448838156537e-05,
      "loss": 1.0768,
      "num_input_tokens_seen": 16126464,
      "step": 27805
    },
    {
      "epoch": 4.142091152815014,
      "grad_norm": 12.5625,
      "learning_rate": 4.827330191422817e-05,
      "loss": 1.2627,
      "num_input_tokens_seen": 16129312,
      "step": 27810
    },
    {
      "epoch": 4.1428358653559725,
      "grad_norm": 19.75,
      "learning_rate": 4.8272115053712185e-05,
      "loss": 1.0582,
      "num_input_tokens_seen": 16132224,
      "step": 27815
    },
    {
      "epoch": 4.143580577896932,
      "grad_norm": 20.375,
      "learning_rate": 4.8270927800037465e-05,
      "loss": 1.0835,
      "num_input_tokens_seen": 16135168,
      "step": 27820
    },
    {
      "epoch": 4.144325290437891,
      "grad_norm": 12.1875,
      "learning_rate": 4.826974015322407e-05,
      "loss": 1.1415,
      "num_input_tokens_seen": 16138688,
      "step": 27825
    },
    {
      "epoch": 4.1450700029788505,
      "grad_norm": 13.75,
      "learning_rate": 4.826855211329206e-05,
      "loss": 1.225,
      "num_input_tokens_seen": 16141920,
      "step": 27830
    },
    {
      "epoch": 4.145814715519809,
      "grad_norm": 5.3125,
      "learning_rate": 4.826736368026152e-05,
      "loss": 0.8067,
      "num_input_tokens_seen": 16144832,
      "step": 27835
    },
    {
      "epoch": 4.146559428060769,
      "grad_norm": 11.625,
      "learning_rate": 4.826617485415252e-05,
      "loss": 0.9042,
      "num_input_tokens_seen": 16147808,
      "step": 27840
    },
    {
      "epoch": 4.147304140601728,
      "grad_norm": 9.25,
      "learning_rate": 4.826498563498514e-05,
      "loss": 0.918,
      "num_input_tokens_seen": 16150688,
      "step": 27845
    },
    {
      "epoch": 4.148048853142687,
      "grad_norm": 8.5,
      "learning_rate": 4.826379602277947e-05,
      "loss": 0.9078,
      "num_input_tokens_seen": 16153664,
      "step": 27850
    },
    {
      "epoch": 4.148793565683646,
      "grad_norm": 8.5625,
      "learning_rate": 4.8262606017555616e-05,
      "loss": 0.7943,
      "num_input_tokens_seen": 16156672,
      "step": 27855
    },
    {
      "epoch": 4.149538278224606,
      "grad_norm": 16.5,
      "learning_rate": 4.826141561933367e-05,
      "loss": 1.2997,
      "num_input_tokens_seen": 16159584,
      "step": 27860
    },
    {
      "epoch": 4.1502829907655645,
      "grad_norm": 9.125,
      "learning_rate": 4.826022482813376e-05,
      "loss": 1.0567,
      "num_input_tokens_seen": 16162528,
      "step": 27865
    },
    {
      "epoch": 4.151027703306523,
      "grad_norm": 12.875,
      "learning_rate": 4.825903364397598e-05,
      "loss": 1.018,
      "num_input_tokens_seen": 16165312,
      "step": 27870
    },
    {
      "epoch": 4.151772415847483,
      "grad_norm": 16.5,
      "learning_rate": 4.8257842066880474e-05,
      "loss": 0.9398,
      "num_input_tokens_seen": 16168288,
      "step": 27875
    },
    {
      "epoch": 4.152517128388442,
      "grad_norm": 14.75,
      "learning_rate": 4.8256650096867364e-05,
      "loss": 1.1064,
      "num_input_tokens_seen": 16171200,
      "step": 27880
    },
    {
      "epoch": 4.153261840929401,
      "grad_norm": 10.875,
      "learning_rate": 4.8255457733956774e-05,
      "loss": 1.0106,
      "num_input_tokens_seen": 16174432,
      "step": 27885
    },
    {
      "epoch": 4.15400655347036,
      "grad_norm": 16.625,
      "learning_rate": 4.825426497816888e-05,
      "loss": 1.0195,
      "num_input_tokens_seen": 16177088,
      "step": 27890
    },
    {
      "epoch": 4.15475126601132,
      "grad_norm": 11.5,
      "learning_rate": 4.82530718295238e-05,
      "loss": 1.4313,
      "num_input_tokens_seen": 16179808,
      "step": 27895
    },
    {
      "epoch": 4.1554959785522785,
      "grad_norm": 11.5625,
      "learning_rate": 4.825187828804171e-05,
      "loss": 0.7901,
      "num_input_tokens_seen": 16182688,
      "step": 27900
    },
    {
      "epoch": 4.156240691093238,
      "grad_norm": 8.5,
      "learning_rate": 4.825068435374277e-05,
      "loss": 1.7228,
      "num_input_tokens_seen": 16185696,
      "step": 27905
    },
    {
      "epoch": 4.156985403634197,
      "grad_norm": 8.0625,
      "learning_rate": 4.824949002664715e-05,
      "loss": 0.9899,
      "num_input_tokens_seen": 16188768,
      "step": 27910
    },
    {
      "epoch": 4.1577301161751565,
      "grad_norm": 6.9375,
      "learning_rate": 4.824829530677503e-05,
      "loss": 1.1253,
      "num_input_tokens_seen": 16191712,
      "step": 27915
    },
    {
      "epoch": 4.158474828716115,
      "grad_norm": 12.3125,
      "learning_rate": 4.824710019414658e-05,
      "loss": 1.2659,
      "num_input_tokens_seen": 16194912,
      "step": 27920
    },
    {
      "epoch": 4.159219541257075,
      "grad_norm": 7.1875,
      "learning_rate": 4.8245904688781994e-05,
      "loss": 1.1125,
      "num_input_tokens_seen": 16197664,
      "step": 27925
    },
    {
      "epoch": 4.159964253798034,
      "grad_norm": 7.59375,
      "learning_rate": 4.8244708790701486e-05,
      "loss": 1.1179,
      "num_input_tokens_seen": 16200800,
      "step": 27930
    },
    {
      "epoch": 4.160708966338993,
      "grad_norm": 9.0,
      "learning_rate": 4.824351249992525e-05,
      "loss": 1.125,
      "num_input_tokens_seen": 16203936,
      "step": 27935
    },
    {
      "epoch": 4.161453678879952,
      "grad_norm": 11.125,
      "learning_rate": 4.824231581647348e-05,
      "loss": 1.2683,
      "num_input_tokens_seen": 16206912,
      "step": 27940
    },
    {
      "epoch": 4.162198391420912,
      "grad_norm": 18.625,
      "learning_rate": 4.824111874036642e-05,
      "loss": 1.0816,
      "num_input_tokens_seen": 16209760,
      "step": 27945
    },
    {
      "epoch": 4.1629431039618705,
      "grad_norm": 10.0,
      "learning_rate": 4.823992127162428e-05,
      "loss": 1.3208,
      "num_input_tokens_seen": 16212544,
      "step": 27950
    },
    {
      "epoch": 4.16368781650283,
      "grad_norm": 8.3125,
      "learning_rate": 4.8238723410267285e-05,
      "loss": 0.9205,
      "num_input_tokens_seen": 16215584,
      "step": 27955
    },
    {
      "epoch": 4.164432529043789,
      "grad_norm": 7.25,
      "learning_rate": 4.823752515631568e-05,
      "loss": 1.0284,
      "num_input_tokens_seen": 16218144,
      "step": 27960
    },
    {
      "epoch": 4.165177241584749,
      "grad_norm": 10.75,
      "learning_rate": 4.8236326509789695e-05,
      "loss": 1.0508,
      "num_input_tokens_seen": 16221312,
      "step": 27965
    },
    {
      "epoch": 4.165921954125707,
      "grad_norm": 8.125,
      "learning_rate": 4.8235127470709594e-05,
      "loss": 0.9805,
      "num_input_tokens_seen": 16224128,
      "step": 27970
    },
    {
      "epoch": 4.166666666666667,
      "grad_norm": 11.6875,
      "learning_rate": 4.8233928039095635e-05,
      "loss": 1.0859,
      "num_input_tokens_seen": 16226976,
      "step": 27975
    },
    {
      "epoch": 4.167411379207626,
      "grad_norm": 16.5,
      "learning_rate": 4.823272821496808e-05,
      "loss": 1.2013,
      "num_input_tokens_seen": 16229760,
      "step": 27980
    },
    {
      "epoch": 4.168156091748585,
      "grad_norm": 13.8125,
      "learning_rate": 4.823152799834718e-05,
      "loss": 1.1109,
      "num_input_tokens_seen": 16232928,
      "step": 27985
    },
    {
      "epoch": 4.168900804289544,
      "grad_norm": 18.625,
      "learning_rate": 4.823032738925324e-05,
      "loss": 1.5699,
      "num_input_tokens_seen": 16236096,
      "step": 27990
    },
    {
      "epoch": 4.169645516830504,
      "grad_norm": 7.8125,
      "learning_rate": 4.8229126387706516e-05,
      "loss": 1.131,
      "num_input_tokens_seen": 16238976,
      "step": 27995
    },
    {
      "epoch": 4.1703902293714625,
      "grad_norm": 10.1875,
      "learning_rate": 4.822792499372732e-05,
      "loss": 1.0695,
      "num_input_tokens_seen": 16242080,
      "step": 28000
    },
    {
      "epoch": 4.171134941912422,
      "grad_norm": 8.9375,
      "learning_rate": 4.822672320733594e-05,
      "loss": 0.9366,
      "num_input_tokens_seen": 16244928,
      "step": 28005
    },
    {
      "epoch": 4.171879654453381,
      "grad_norm": 7.0625,
      "learning_rate": 4.822552102855267e-05,
      "loss": 1.4077,
      "num_input_tokens_seen": 16247616,
      "step": 28010
    },
    {
      "epoch": 4.172624366994341,
      "grad_norm": 8.0625,
      "learning_rate": 4.822431845739783e-05,
      "loss": 1.1503,
      "num_input_tokens_seen": 16250624,
      "step": 28015
    },
    {
      "epoch": 4.173369079535299,
      "grad_norm": 8.5625,
      "learning_rate": 4.822311549389174e-05,
      "loss": 0.8932,
      "num_input_tokens_seen": 16253152,
      "step": 28020
    },
    {
      "epoch": 4.174113792076259,
      "grad_norm": 10.6875,
      "learning_rate": 4.8221912138054715e-05,
      "loss": 1.0962,
      "num_input_tokens_seen": 16256160,
      "step": 28025
    },
    {
      "epoch": 4.174858504617218,
      "grad_norm": 12.9375,
      "learning_rate": 4.822070838990708e-05,
      "loss": 1.0756,
      "num_input_tokens_seen": 16259040,
      "step": 28030
    },
    {
      "epoch": 4.1756032171581765,
      "grad_norm": 15.25,
      "learning_rate": 4.8219504249469186e-05,
      "loss": 1.1361,
      "num_input_tokens_seen": 16261984,
      "step": 28035
    },
    {
      "epoch": 4.176347929699136,
      "grad_norm": 11.1875,
      "learning_rate": 4.821829971676136e-05,
      "loss": 1.118,
      "num_input_tokens_seen": 16264928,
      "step": 28040
    },
    {
      "epoch": 4.177092642240095,
      "grad_norm": 9.5,
      "learning_rate": 4.8217094791803966e-05,
      "loss": 1.0383,
      "num_input_tokens_seen": 16267520,
      "step": 28045
    },
    {
      "epoch": 4.177837354781055,
      "grad_norm": 14.3125,
      "learning_rate": 4.821588947461734e-05,
      "loss": 1.1271,
      "num_input_tokens_seen": 16270240,
      "step": 28050
    },
    {
      "epoch": 4.178582067322013,
      "grad_norm": 14.625,
      "learning_rate": 4.821468376522186e-05,
      "loss": 1.0993,
      "num_input_tokens_seen": 16273024,
      "step": 28055
    },
    {
      "epoch": 4.179326779862973,
      "grad_norm": 14.25,
      "learning_rate": 4.82134776636379e-05,
      "loss": 1.0468,
      "num_input_tokens_seen": 16276160,
      "step": 28060
    },
    {
      "epoch": 4.180071492403932,
      "grad_norm": 6.15625,
      "learning_rate": 4.821227116988583e-05,
      "loss": 1.1924,
      "num_input_tokens_seen": 16279008,
      "step": 28065
    },
    {
      "epoch": 4.180816204944891,
      "grad_norm": 5.5625,
      "learning_rate": 4.8211064283986015e-05,
      "loss": 0.7706,
      "num_input_tokens_seen": 16281760,
      "step": 28070
    },
    {
      "epoch": 4.18156091748585,
      "grad_norm": 13.0,
      "learning_rate": 4.8209857005958866e-05,
      "loss": 0.9204,
      "num_input_tokens_seen": 16284736,
      "step": 28075
    },
    {
      "epoch": 4.18230563002681,
      "grad_norm": 6.4375,
      "learning_rate": 4.820864933582478e-05,
      "loss": 1.052,
      "num_input_tokens_seen": 16287712,
      "step": 28080
    },
    {
      "epoch": 4.1830503425677685,
      "grad_norm": 12.0625,
      "learning_rate": 4.8207441273604145e-05,
      "loss": 1.0767,
      "num_input_tokens_seen": 16290432,
      "step": 28085
    },
    {
      "epoch": 4.183795055108728,
      "grad_norm": 9.0,
      "learning_rate": 4.820623281931738e-05,
      "loss": 0.8584,
      "num_input_tokens_seen": 16293472,
      "step": 28090
    },
    {
      "epoch": 4.184539767649687,
      "grad_norm": 12.1875,
      "learning_rate": 4.8205023972984896e-05,
      "loss": 1.0251,
      "num_input_tokens_seen": 16296480,
      "step": 28095
    },
    {
      "epoch": 4.185284480190647,
      "grad_norm": 10.875,
      "learning_rate": 4.820381473462712e-05,
      "loss": 0.9897,
      "num_input_tokens_seen": 16299232,
      "step": 28100
    },
    {
      "epoch": 4.186029192731605,
      "grad_norm": 17.0,
      "learning_rate": 4.820260510426447e-05,
      "loss": 1.0128,
      "num_input_tokens_seen": 16302432,
      "step": 28105
    },
    {
      "epoch": 4.186773905272565,
      "grad_norm": 6.28125,
      "learning_rate": 4.820139508191739e-05,
      "loss": 0.9703,
      "num_input_tokens_seen": 16305408,
      "step": 28110
    },
    {
      "epoch": 4.187518617813524,
      "grad_norm": 13.5,
      "learning_rate": 4.820018466760633e-05,
      "loss": 1.0774,
      "num_input_tokens_seen": 16308384,
      "step": 28115
    },
    {
      "epoch": 4.188263330354483,
      "grad_norm": 8.875,
      "learning_rate": 4.819897386135172e-05,
      "loss": 1.3076,
      "num_input_tokens_seen": 16311168,
      "step": 28120
    },
    {
      "epoch": 4.189008042895442,
      "grad_norm": 11.125,
      "learning_rate": 4.819776266317403e-05,
      "loss": 1.1845,
      "num_input_tokens_seen": 16314144,
      "step": 28125
    },
    {
      "epoch": 4.189752755436402,
      "grad_norm": 10.375,
      "learning_rate": 4.819655107309371e-05,
      "loss": 1.0643,
      "num_input_tokens_seen": 16316864,
      "step": 28130
    },
    {
      "epoch": 4.190497467977361,
      "grad_norm": 12.8125,
      "learning_rate": 4.819533909113124e-05,
      "loss": 1.0205,
      "num_input_tokens_seen": 16319424,
      "step": 28135
    },
    {
      "epoch": 4.19124218051832,
      "grad_norm": 13.25,
      "learning_rate": 4.819412671730709e-05,
      "loss": 0.9053,
      "num_input_tokens_seen": 16322080,
      "step": 28140
    },
    {
      "epoch": 4.191986893059279,
      "grad_norm": 10.4375,
      "learning_rate": 4.8192913951641746e-05,
      "loss": 1.1698,
      "num_input_tokens_seen": 16324960,
      "step": 28145
    },
    {
      "epoch": 4.192731605600239,
      "grad_norm": 8.0,
      "learning_rate": 4.819170079415569e-05,
      "loss": 1.164,
      "num_input_tokens_seen": 16327712,
      "step": 28150
    },
    {
      "epoch": 4.193476318141197,
      "grad_norm": 14.5625,
      "learning_rate": 4.819048724486942e-05,
      "loss": 1.0949,
      "num_input_tokens_seen": 16330432,
      "step": 28155
    },
    {
      "epoch": 4.194221030682157,
      "grad_norm": 12.875,
      "learning_rate": 4.818927330380344e-05,
      "loss": 0.8292,
      "num_input_tokens_seen": 16333376,
      "step": 28160
    },
    {
      "epoch": 4.194965743223116,
      "grad_norm": 9.1875,
      "learning_rate": 4.8188058970978254e-05,
      "loss": 1.3704,
      "num_input_tokens_seen": 16336256,
      "step": 28165
    },
    {
      "epoch": 4.195710455764075,
      "grad_norm": 11.625,
      "learning_rate": 4.818684424641438e-05,
      "loss": 1.0989,
      "num_input_tokens_seen": 16338976,
      "step": 28170
    },
    {
      "epoch": 4.196455168305034,
      "grad_norm": 11.6875,
      "learning_rate": 4.8185629130132336e-05,
      "loss": 1.2045,
      "num_input_tokens_seen": 16342048,
      "step": 28175
    },
    {
      "epoch": 4.197199880845994,
      "grad_norm": 16.25,
      "learning_rate": 4.818441362215266e-05,
      "loss": 1.1535,
      "num_input_tokens_seen": 16344800,
      "step": 28180
    },
    {
      "epoch": 4.197944593386953,
      "grad_norm": 17.75,
      "learning_rate": 4.8183197722495877e-05,
      "loss": 1.2051,
      "num_input_tokens_seen": 16347616,
      "step": 28185
    },
    {
      "epoch": 4.198689305927912,
      "grad_norm": 10.875,
      "learning_rate": 4.8181981431182523e-05,
      "loss": 1.1337,
      "num_input_tokens_seen": 16350624,
      "step": 28190
    },
    {
      "epoch": 4.199434018468871,
      "grad_norm": 12.75,
      "learning_rate": 4.818076474823316e-05,
      "loss": 1.1353,
      "num_input_tokens_seen": 16353600,
      "step": 28195
    },
    {
      "epoch": 4.200178731009831,
      "grad_norm": 12.75,
      "learning_rate": 4.817954767366833e-05,
      "loss": 1.0522,
      "num_input_tokens_seen": 16356672,
      "step": 28200
    },
    {
      "epoch": 4.200923443550789,
      "grad_norm": 14.375,
      "learning_rate": 4.817833020750861e-05,
      "loss": 1.7381,
      "num_input_tokens_seen": 16359424,
      "step": 28205
    },
    {
      "epoch": 4.201668156091749,
      "grad_norm": 7.09375,
      "learning_rate": 4.8177112349774554e-05,
      "loss": 1.0296,
      "num_input_tokens_seen": 16362368,
      "step": 28210
    },
    {
      "epoch": 4.202412868632708,
      "grad_norm": 4.65625,
      "learning_rate": 4.817589410048674e-05,
      "loss": 0.7358,
      "num_input_tokens_seen": 16365120,
      "step": 28215
    },
    {
      "epoch": 4.203157581173667,
      "grad_norm": 11.375,
      "learning_rate": 4.817467545966575e-05,
      "loss": 0.8879,
      "num_input_tokens_seen": 16368160,
      "step": 28220
    },
    {
      "epoch": 4.203902293714626,
      "grad_norm": 10.5,
      "learning_rate": 4.8173456427332176e-05,
      "loss": 1.0732,
      "num_input_tokens_seen": 16371328,
      "step": 28225
    },
    {
      "epoch": 4.204647006255585,
      "grad_norm": 9.8125,
      "learning_rate": 4.817223700350661e-05,
      "loss": 1.1084,
      "num_input_tokens_seen": 16374400,
      "step": 28230
    },
    {
      "epoch": 4.205391718796545,
      "grad_norm": 12.625,
      "learning_rate": 4.817101718820965e-05,
      "loss": 1.1048,
      "num_input_tokens_seen": 16377056,
      "step": 28235
    },
    {
      "epoch": 4.206136431337503,
      "grad_norm": 22.5,
      "learning_rate": 4.8169796981461904e-05,
      "loss": 0.9398,
      "num_input_tokens_seen": 16379808,
      "step": 28240
    },
    {
      "epoch": 4.206881143878463,
      "grad_norm": 8.1875,
      "learning_rate": 4.816857638328398e-05,
      "loss": 1.0849,
      "num_input_tokens_seen": 16382784,
      "step": 28245
    },
    {
      "epoch": 4.207625856419422,
      "grad_norm": 7.1875,
      "learning_rate": 4.816735539369651e-05,
      "loss": 1.0459,
      "num_input_tokens_seen": 16385888,
      "step": 28250
    },
    {
      "epoch": 4.208370568960381,
      "grad_norm": 5.34375,
      "learning_rate": 4.816613401272011e-05,
      "loss": 0.83,
      "num_input_tokens_seen": 16388608,
      "step": 28255
    },
    {
      "epoch": 4.20911528150134,
      "grad_norm": 9.9375,
      "learning_rate": 4.816491224037543e-05,
      "loss": 1.1829,
      "num_input_tokens_seen": 16391712,
      "step": 28260
    },
    {
      "epoch": 4.2098599940423,
      "grad_norm": 9.5625,
      "learning_rate": 4.81636900766831e-05,
      "loss": 1.1853,
      "num_input_tokens_seen": 16394560,
      "step": 28265
    },
    {
      "epoch": 4.210604706583259,
      "grad_norm": 12.9375,
      "learning_rate": 4.816246752166377e-05,
      "loss": 1.1901,
      "num_input_tokens_seen": 16397216,
      "step": 28270
    },
    {
      "epoch": 4.211349419124218,
      "grad_norm": 11.1875,
      "learning_rate": 4.8161244575338086e-05,
      "loss": 1.0383,
      "num_input_tokens_seen": 16400288,
      "step": 28275
    },
    {
      "epoch": 4.212094131665177,
      "grad_norm": 5.9375,
      "learning_rate": 4.816002123772672e-05,
      "loss": 1.3749,
      "num_input_tokens_seen": 16403360,
      "step": 28280
    },
    {
      "epoch": 4.212838844206137,
      "grad_norm": 9.8125,
      "learning_rate": 4.815879750885033e-05,
      "loss": 0.9272,
      "num_input_tokens_seen": 16406080,
      "step": 28285
    },
    {
      "epoch": 4.213583556747095,
      "grad_norm": 15.375,
      "learning_rate": 4.81575733887296e-05,
      "loss": 1.0141,
      "num_input_tokens_seen": 16409024,
      "step": 28290
    },
    {
      "epoch": 4.214328269288055,
      "grad_norm": 13.3125,
      "learning_rate": 4.81563488773852e-05,
      "loss": 1.1574,
      "num_input_tokens_seen": 16411968,
      "step": 28295
    },
    {
      "epoch": 4.215072981829014,
      "grad_norm": 10.25,
      "learning_rate": 4.8155123974837824e-05,
      "loss": 0.8821,
      "num_input_tokens_seen": 16414784,
      "step": 28300
    },
    {
      "epoch": 4.2158176943699734,
      "grad_norm": 9.75,
      "learning_rate": 4.815389868110816e-05,
      "loss": 0.8621,
      "num_input_tokens_seen": 16417536,
      "step": 28305
    },
    {
      "epoch": 4.216562406910932,
      "grad_norm": 18.375,
      "learning_rate": 4.815267299621691e-05,
      "loss": 1.3758,
      "num_input_tokens_seen": 16420224,
      "step": 28310
    },
    {
      "epoch": 4.217307119451892,
      "grad_norm": 6.46875,
      "learning_rate": 4.815144692018477e-05,
      "loss": 0.7537,
      "num_input_tokens_seen": 16422944,
      "step": 28315
    },
    {
      "epoch": 4.218051831992851,
      "grad_norm": 13.75,
      "learning_rate": 4.815022045303248e-05,
      "loss": 0.8833,
      "num_input_tokens_seen": 16425472,
      "step": 28320
    },
    {
      "epoch": 4.21879654453381,
      "grad_norm": 7.1875,
      "learning_rate": 4.814899359478074e-05,
      "loss": 1.0329,
      "num_input_tokens_seen": 16428192,
      "step": 28325
    },
    {
      "epoch": 4.219541257074769,
      "grad_norm": 6.03125,
      "learning_rate": 4.814776634545028e-05,
      "loss": 1.1764,
      "num_input_tokens_seen": 16431200,
      "step": 28330
    },
    {
      "epoch": 4.220285969615729,
      "grad_norm": 11.6875,
      "learning_rate": 4.814653870506183e-05,
      "loss": 0.9085,
      "num_input_tokens_seen": 16434176,
      "step": 28335
    },
    {
      "epoch": 4.221030682156687,
      "grad_norm": 6.0,
      "learning_rate": 4.8145310673636143e-05,
      "loss": 1.4178,
      "num_input_tokens_seen": 16437120,
      "step": 28340
    },
    {
      "epoch": 4.221775394697647,
      "grad_norm": 8.1875,
      "learning_rate": 4.814408225119395e-05,
      "loss": 0.9714,
      "num_input_tokens_seen": 16440000,
      "step": 28345
    },
    {
      "epoch": 4.222520107238606,
      "grad_norm": 10.0625,
      "learning_rate": 4.8142853437756006e-05,
      "loss": 0.8342,
      "num_input_tokens_seen": 16442880,
      "step": 28350
    },
    {
      "epoch": 4.2232648197795655,
      "grad_norm": 14.5,
      "learning_rate": 4.814162423334309e-05,
      "loss": 0.8613,
      "num_input_tokens_seen": 16445888,
      "step": 28355
    },
    {
      "epoch": 4.224009532320524,
      "grad_norm": 14.0,
      "learning_rate": 4.814039463797594e-05,
      "loss": 1.2478,
      "num_input_tokens_seen": 16448448,
      "step": 28360
    },
    {
      "epoch": 4.224754244861484,
      "grad_norm": 4.875,
      "learning_rate": 4.813916465167534e-05,
      "loss": 0.9811,
      "num_input_tokens_seen": 16451232,
      "step": 28365
    },
    {
      "epoch": 4.225498957402443,
      "grad_norm": 12.625,
      "learning_rate": 4.813793427446207e-05,
      "loss": 1.2126,
      "num_input_tokens_seen": 16454144,
      "step": 28370
    },
    {
      "epoch": 4.226243669943402,
      "grad_norm": 9.75,
      "learning_rate": 4.813670350635693e-05,
      "loss": 1.1421,
      "num_input_tokens_seen": 16457152,
      "step": 28375
    },
    {
      "epoch": 4.226988382484361,
      "grad_norm": 13.125,
      "learning_rate": 4.8135472347380684e-05,
      "loss": 1.1957,
      "num_input_tokens_seen": 16460160,
      "step": 28380
    },
    {
      "epoch": 4.22773309502532,
      "grad_norm": 6.53125,
      "learning_rate": 4.8134240797554155e-05,
      "loss": 1.06,
      "num_input_tokens_seen": 16463392,
      "step": 28385
    },
    {
      "epoch": 4.2284778075662794,
      "grad_norm": 13.125,
      "learning_rate": 4.813300885689814e-05,
      "loss": 0.8458,
      "num_input_tokens_seen": 16466208,
      "step": 28390
    },
    {
      "epoch": 4.229222520107238,
      "grad_norm": 8.625,
      "learning_rate": 4.813177652543345e-05,
      "loss": 1.2333,
      "num_input_tokens_seen": 16468864,
      "step": 28395
    },
    {
      "epoch": 4.229967232648198,
      "grad_norm": 11.75,
      "learning_rate": 4.813054380318091e-05,
      "loss": 1.4404,
      "num_input_tokens_seen": 16472000,
      "step": 28400
    },
    {
      "epoch": 4.230711945189157,
      "grad_norm": 5.90625,
      "learning_rate": 4.8129310690161335e-05,
      "loss": 0.9906,
      "num_input_tokens_seen": 16474816,
      "step": 28405
    },
    {
      "epoch": 4.231456657730116,
      "grad_norm": 17.25,
      "learning_rate": 4.812807718639556e-05,
      "loss": 0.9522,
      "num_input_tokens_seen": 16477856,
      "step": 28410
    },
    {
      "epoch": 4.232201370271075,
      "grad_norm": 9.5,
      "learning_rate": 4.812684329190443e-05,
      "loss": 0.9992,
      "num_input_tokens_seen": 16480800,
      "step": 28415
    },
    {
      "epoch": 4.232946082812035,
      "grad_norm": 5.84375,
      "learning_rate": 4.8125609006708796e-05,
      "loss": 1.2935,
      "num_input_tokens_seen": 16484096,
      "step": 28420
    },
    {
      "epoch": 4.233690795352993,
      "grad_norm": 3.125,
      "learning_rate": 4.812437433082949e-05,
      "loss": 0.9967,
      "num_input_tokens_seen": 16486752,
      "step": 28425
    },
    {
      "epoch": 4.234435507893953,
      "grad_norm": 6.875,
      "learning_rate": 4.812313926428739e-05,
      "loss": 1.05,
      "num_input_tokens_seen": 16489696,
      "step": 28430
    },
    {
      "epoch": 4.235180220434912,
      "grad_norm": 10.0,
      "learning_rate": 4.812190380710335e-05,
      "loss": 1.6905,
      "num_input_tokens_seen": 16492544,
      "step": 28435
    },
    {
      "epoch": 4.2359249329758715,
      "grad_norm": 5.28125,
      "learning_rate": 4.812066795929825e-05,
      "loss": 1.2629,
      "num_input_tokens_seen": 16495520,
      "step": 28440
    },
    {
      "epoch": 4.23666964551683,
      "grad_norm": 8.1875,
      "learning_rate": 4.811943172089296e-05,
      "loss": 0.7945,
      "num_input_tokens_seen": 16498464,
      "step": 28445
    },
    {
      "epoch": 4.23741435805779,
      "grad_norm": 6.59375,
      "learning_rate": 4.811819509190837e-05,
      "loss": 1.0202,
      "num_input_tokens_seen": 16501216,
      "step": 28450
    },
    {
      "epoch": 4.238159070598749,
      "grad_norm": 8.0,
      "learning_rate": 4.811695807236537e-05,
      "loss": 0.8843,
      "num_input_tokens_seen": 16504384,
      "step": 28455
    },
    {
      "epoch": 4.238903783139708,
      "grad_norm": 8.0,
      "learning_rate": 4.8115720662284855e-05,
      "loss": 1.1011,
      "num_input_tokens_seen": 16507328,
      "step": 28460
    },
    {
      "epoch": 4.239648495680667,
      "grad_norm": 10.5,
      "learning_rate": 4.8114482861687734e-05,
      "loss": 1.3863,
      "num_input_tokens_seen": 16510560,
      "step": 28465
    },
    {
      "epoch": 4.240393208221627,
      "grad_norm": 7.03125,
      "learning_rate": 4.8113244670594926e-05,
      "loss": 1.2039,
      "num_input_tokens_seen": 16513312,
      "step": 28470
    },
    {
      "epoch": 4.2411379207625854,
      "grad_norm": 7.75,
      "learning_rate": 4.811200608902733e-05,
      "loss": 0.9096,
      "num_input_tokens_seen": 16516096,
      "step": 28475
    },
    {
      "epoch": 4.241882633303545,
      "grad_norm": 12.625,
      "learning_rate": 4.811076711700588e-05,
      "loss": 1.112,
      "num_input_tokens_seen": 16519168,
      "step": 28480
    },
    {
      "epoch": 4.242627345844504,
      "grad_norm": 11.375,
      "learning_rate": 4.810952775455152e-05,
      "loss": 0.7749,
      "num_input_tokens_seen": 16522240,
      "step": 28485
    },
    {
      "epoch": 4.2433720583854635,
      "grad_norm": 6.28125,
      "learning_rate": 4.810828800168517e-05,
      "loss": 0.7706,
      "num_input_tokens_seen": 16525344,
      "step": 28490
    },
    {
      "epoch": 4.244116770926422,
      "grad_norm": 18.875,
      "learning_rate": 4.810704785842778e-05,
      "loss": 1.0633,
      "num_input_tokens_seen": 16528288,
      "step": 28495
    },
    {
      "epoch": 4.244861483467382,
      "grad_norm": 5.0,
      "learning_rate": 4.81058073248003e-05,
      "loss": 1.3511,
      "num_input_tokens_seen": 16531296,
      "step": 28500
    },
    {
      "epoch": 4.245606196008341,
      "grad_norm": 21.75,
      "learning_rate": 4.810456640082369e-05,
      "loss": 1.2214,
      "num_input_tokens_seen": 16534336,
      "step": 28505
    },
    {
      "epoch": 4.2463509085493,
      "grad_norm": 17.375,
      "learning_rate": 4.810332508651891e-05,
      "loss": 1.0346,
      "num_input_tokens_seen": 16537216,
      "step": 28510
    },
    {
      "epoch": 4.247095621090259,
      "grad_norm": 6.4375,
      "learning_rate": 4.810208338190694e-05,
      "loss": 1.159,
      "num_input_tokens_seen": 16540192,
      "step": 28515
    },
    {
      "epoch": 4.247840333631219,
      "grad_norm": 14.5,
      "learning_rate": 4.810084128700875e-05,
      "loss": 1.1606,
      "num_input_tokens_seen": 16543040,
      "step": 28520
    },
    {
      "epoch": 4.2485850461721775,
      "grad_norm": 7.21875,
      "learning_rate": 4.809959880184532e-05,
      "loss": 1.0572,
      "num_input_tokens_seen": 16546176,
      "step": 28525
    },
    {
      "epoch": 4.249329758713137,
      "grad_norm": 11.9375,
      "learning_rate": 4.8098355926437655e-05,
      "loss": 1.1314,
      "num_input_tokens_seen": 16549184,
      "step": 28530
    },
    {
      "epoch": 4.250074471254096,
      "grad_norm": 13.9375,
      "learning_rate": 4.809711266080673e-05,
      "loss": 1.1763,
      "num_input_tokens_seen": 16551904,
      "step": 28535
    },
    {
      "epoch": 4.2508191837950555,
      "grad_norm": 7.25,
      "learning_rate": 4.809586900497357e-05,
      "loss": 0.8439,
      "num_input_tokens_seen": 16554688,
      "step": 28540
    },
    {
      "epoch": 4.251563896336014,
      "grad_norm": 9.3125,
      "learning_rate": 4.809462495895918e-05,
      "loss": 0.924,
      "num_input_tokens_seen": 16557696,
      "step": 28545
    },
    {
      "epoch": 4.252308608876973,
      "grad_norm": 14.75,
      "learning_rate": 4.809338052278456e-05,
      "loss": 1.0141,
      "num_input_tokens_seen": 16560352,
      "step": 28550
    },
    {
      "epoch": 4.253053321417933,
      "grad_norm": 5.25,
      "learning_rate": 4.809213569647076e-05,
      "loss": 1.0792,
      "num_input_tokens_seen": 16562880,
      "step": 28555
    },
    {
      "epoch": 4.253798033958892,
      "grad_norm": 11.9375,
      "learning_rate": 4.8090890480038796e-05,
      "loss": 1.1245,
      "num_input_tokens_seen": 16565792,
      "step": 28560
    },
    {
      "epoch": 4.254542746499851,
      "grad_norm": 18.25,
      "learning_rate": 4.80896448735097e-05,
      "loss": 1.3299,
      "num_input_tokens_seen": 16568704,
      "step": 28565
    },
    {
      "epoch": 4.25528745904081,
      "grad_norm": 12.25,
      "learning_rate": 4.8088398876904526e-05,
      "loss": 1.2028,
      "num_input_tokens_seen": 16571648,
      "step": 28570
    },
    {
      "epoch": 4.2560321715817695,
      "grad_norm": 13.4375,
      "learning_rate": 4.808715249024431e-05,
      "loss": 0.9708,
      "num_input_tokens_seen": 16575008,
      "step": 28575
    },
    {
      "epoch": 4.256776884122728,
      "grad_norm": 6.75,
      "learning_rate": 4.808590571355013e-05,
      "loss": 0.9286,
      "num_input_tokens_seen": 16577728,
      "step": 28580
    },
    {
      "epoch": 4.257521596663688,
      "grad_norm": 6.21875,
      "learning_rate": 4.808465854684303e-05,
      "loss": 0.9629,
      "num_input_tokens_seen": 16580608,
      "step": 28585
    },
    {
      "epoch": 4.258266309204647,
      "grad_norm": 5.125,
      "learning_rate": 4.8083410990144085e-05,
      "loss": 0.8669,
      "num_input_tokens_seen": 16583680,
      "step": 28590
    },
    {
      "epoch": 4.259011021745606,
      "grad_norm": 7.90625,
      "learning_rate": 4.808216304347438e-05,
      "loss": 0.7639,
      "num_input_tokens_seen": 16586912,
      "step": 28595
    },
    {
      "epoch": 4.259755734286565,
      "grad_norm": 7.3125,
      "learning_rate": 4.8080914706854985e-05,
      "loss": 1.0265,
      "num_input_tokens_seen": 16589792,
      "step": 28600
    },
    {
      "epoch": 4.260500446827525,
      "grad_norm": 4.40625,
      "learning_rate": 4.8079665980306986e-05,
      "loss": 1.2615,
      "num_input_tokens_seen": 16592928,
      "step": 28605
    },
    {
      "epoch": 4.2612451593684835,
      "grad_norm": 7.25,
      "learning_rate": 4.80784168638515e-05,
      "loss": 0.8934,
      "num_input_tokens_seen": 16595712,
      "step": 28610
    },
    {
      "epoch": 4.261989871909443,
      "grad_norm": 10.5,
      "learning_rate": 4.807716735750961e-05,
      "loss": 0.9064,
      "num_input_tokens_seen": 16598912,
      "step": 28615
    },
    {
      "epoch": 4.262734584450402,
      "grad_norm": 6.3125,
      "learning_rate": 4.8075917461302435e-05,
      "loss": 0.7992,
      "num_input_tokens_seen": 16602080,
      "step": 28620
    },
    {
      "epoch": 4.2634792969913615,
      "grad_norm": 9.5625,
      "learning_rate": 4.807466717525109e-05,
      "loss": 1.0672,
      "num_input_tokens_seen": 16604960,
      "step": 28625
    },
    {
      "epoch": 4.26422400953232,
      "grad_norm": 14.0,
      "learning_rate": 4.807341649937669e-05,
      "loss": 0.8787,
      "num_input_tokens_seen": 16607776,
      "step": 28630
    },
    {
      "epoch": 4.26496872207328,
      "grad_norm": 7.84375,
      "learning_rate": 4.8072165433700366e-05,
      "loss": 1.008,
      "num_input_tokens_seen": 16610624,
      "step": 28635
    },
    {
      "epoch": 4.265713434614239,
      "grad_norm": 13.625,
      "learning_rate": 4.807091397824327e-05,
      "loss": 1.2635,
      "num_input_tokens_seen": 16613504,
      "step": 28640
    },
    {
      "epoch": 4.266458147155198,
      "grad_norm": 9.875,
      "learning_rate": 4.806966213302652e-05,
      "loss": 1.2954,
      "num_input_tokens_seen": 16616544,
      "step": 28645
    },
    {
      "epoch": 4.267202859696157,
      "grad_norm": 5.6875,
      "learning_rate": 4.806840989807128e-05,
      "loss": 1.0747,
      "num_input_tokens_seen": 16619552,
      "step": 28650
    },
    {
      "epoch": 4.267947572237117,
      "grad_norm": 9.5,
      "learning_rate": 4.806715727339869e-05,
      "loss": 1.1557,
      "num_input_tokens_seen": 16622368,
      "step": 28655
    },
    {
      "epoch": 4.2686922847780755,
      "grad_norm": 9.375,
      "learning_rate": 4.8065904259029934e-05,
      "loss": 0.9973,
      "num_input_tokens_seen": 16625280,
      "step": 28660
    },
    {
      "epoch": 4.269436997319035,
      "grad_norm": 19.375,
      "learning_rate": 4.806465085498616e-05,
      "loss": 1.201,
      "num_input_tokens_seen": 16628256,
      "step": 28665
    },
    {
      "epoch": 4.270181709859994,
      "grad_norm": 12.6875,
      "learning_rate": 4.806339706128856e-05,
      "loss": 1.1969,
      "num_input_tokens_seen": 16631616,
      "step": 28670
    },
    {
      "epoch": 4.2709264224009535,
      "grad_norm": 14.875,
      "learning_rate": 4.8062142877958307e-05,
      "loss": 1.0119,
      "num_input_tokens_seen": 16634496,
      "step": 28675
    },
    {
      "epoch": 4.271671134941912,
      "grad_norm": 4.40625,
      "learning_rate": 4.8060888305016584e-05,
      "loss": 1.0368,
      "num_input_tokens_seen": 16637440,
      "step": 28680
    },
    {
      "epoch": 4.272415847482872,
      "grad_norm": 12.625,
      "learning_rate": 4.8059633342484586e-05,
      "loss": 1.4659,
      "num_input_tokens_seen": 16640192,
      "step": 28685
    },
    {
      "epoch": 4.273160560023831,
      "grad_norm": 11.3125,
      "learning_rate": 4.805837799038353e-05,
      "loss": 1.2086,
      "num_input_tokens_seen": 16643200,
      "step": 28690
    },
    {
      "epoch": 4.27390527256479,
      "grad_norm": 8.0,
      "learning_rate": 4.805712224873461e-05,
      "loss": 0.7856,
      "num_input_tokens_seen": 16646368,
      "step": 28695
    },
    {
      "epoch": 4.274649985105749,
      "grad_norm": 15.6875,
      "learning_rate": 4.805586611755905e-05,
      "loss": 0.8207,
      "num_input_tokens_seen": 16649216,
      "step": 28700
    },
    {
      "epoch": 4.275394697646709,
      "grad_norm": 5.71875,
      "learning_rate": 4.805460959687805e-05,
      "loss": 0.9566,
      "num_input_tokens_seen": 16652032,
      "step": 28705
    },
    {
      "epoch": 4.2761394101876675,
      "grad_norm": 8.5625,
      "learning_rate": 4.805335268671286e-05,
      "loss": 1.2128,
      "num_input_tokens_seen": 16654880,
      "step": 28710
    },
    {
      "epoch": 4.276884122728626,
      "grad_norm": 10.75,
      "learning_rate": 4.805209538708471e-05,
      "loss": 1.5503,
      "num_input_tokens_seen": 16657760,
      "step": 28715
    },
    {
      "epoch": 4.277628835269586,
      "grad_norm": 7.9375,
      "learning_rate": 4.805083769801484e-05,
      "loss": 1.0013,
      "num_input_tokens_seen": 16660640,
      "step": 28720
    },
    {
      "epoch": 4.278373547810546,
      "grad_norm": 14.625,
      "learning_rate": 4.804957961952449e-05,
      "loss": 1.1939,
      "num_input_tokens_seen": 16663552,
      "step": 28725
    },
    {
      "epoch": 4.279118260351504,
      "grad_norm": 9.1875,
      "learning_rate": 4.804832115163491e-05,
      "loss": 0.9225,
      "num_input_tokens_seen": 16666752,
      "step": 28730
    },
    {
      "epoch": 4.279862972892463,
      "grad_norm": 4.8125,
      "learning_rate": 4.804706229436739e-05,
      "loss": 0.7032,
      "num_input_tokens_seen": 16669696,
      "step": 28735
    },
    {
      "epoch": 4.280607685433423,
      "grad_norm": 10.8125,
      "learning_rate": 4.804580304774316e-05,
      "loss": 0.7401,
      "num_input_tokens_seen": 16672640,
      "step": 28740
    },
    {
      "epoch": 4.2813523979743815,
      "grad_norm": 15.375,
      "learning_rate": 4.804454341178352e-05,
      "loss": 1.0322,
      "num_input_tokens_seen": 16675520,
      "step": 28745
    },
    {
      "epoch": 4.282097110515341,
      "grad_norm": 12.9375,
      "learning_rate": 4.804328338650973e-05,
      "loss": 1.3602,
      "num_input_tokens_seen": 16678304,
      "step": 28750
    },
    {
      "epoch": 4.2828418230563,
      "grad_norm": 5.6875,
      "learning_rate": 4.804202297194309e-05,
      "loss": 0.9109,
      "num_input_tokens_seen": 16682304,
      "step": 28755
    },
    {
      "epoch": 4.2835865355972595,
      "grad_norm": 10.625,
      "learning_rate": 4.8040762168104895e-05,
      "loss": 0.9593,
      "num_input_tokens_seen": 16685152,
      "step": 28760
    },
    {
      "epoch": 4.284331248138218,
      "grad_norm": 7.0,
      "learning_rate": 4.803950097501644e-05,
      "loss": 1.1458,
      "num_input_tokens_seen": 16687904,
      "step": 28765
    },
    {
      "epoch": 4.285075960679178,
      "grad_norm": 4.78125,
      "learning_rate": 4.8038239392699033e-05,
      "loss": 1.2016,
      "num_input_tokens_seen": 16690752,
      "step": 28770
    },
    {
      "epoch": 4.285820673220137,
      "grad_norm": 8.4375,
      "learning_rate": 4.803697742117399e-05,
      "loss": 1.148,
      "num_input_tokens_seen": 16693760,
      "step": 28775
    },
    {
      "epoch": 4.286565385761096,
      "grad_norm": 11.5625,
      "learning_rate": 4.8035715060462614e-05,
      "loss": 1.5717,
      "num_input_tokens_seen": 16696480,
      "step": 28780
    },
    {
      "epoch": 4.287310098302055,
      "grad_norm": 13.375,
      "learning_rate": 4.803445231058625e-05,
      "loss": 1.0103,
      "num_input_tokens_seen": 16699328,
      "step": 28785
    },
    {
      "epoch": 4.288054810843015,
      "grad_norm": 14.5,
      "learning_rate": 4.803318917156624e-05,
      "loss": 0.8674,
      "num_input_tokens_seen": 16702336,
      "step": 28790
    },
    {
      "epoch": 4.2887995233839735,
      "grad_norm": 11.125,
      "learning_rate": 4.803192564342389e-05,
      "loss": 1.0336,
      "num_input_tokens_seen": 16705408,
      "step": 28795
    },
    {
      "epoch": 4.289544235924933,
      "grad_norm": 14.8125,
      "learning_rate": 4.803066172618058e-05,
      "loss": 1.1249,
      "num_input_tokens_seen": 16708448,
      "step": 28800
    },
    {
      "epoch": 4.290288948465892,
      "grad_norm": 7.5625,
      "learning_rate": 4.802939741985763e-05,
      "loss": 1.1219,
      "num_input_tokens_seen": 16711456,
      "step": 28805
    },
    {
      "epoch": 4.291033661006852,
      "grad_norm": 11.625,
      "learning_rate": 4.802813272447643e-05,
      "loss": 1.0009,
      "num_input_tokens_seen": 16714336,
      "step": 28810
    },
    {
      "epoch": 4.29177837354781,
      "grad_norm": 7.15625,
      "learning_rate": 4.8026867640058335e-05,
      "loss": 1.0885,
      "num_input_tokens_seen": 16717440,
      "step": 28815
    },
    {
      "epoch": 4.29252308608877,
      "grad_norm": 9.8125,
      "learning_rate": 4.8025602166624705e-05,
      "loss": 1.1929,
      "num_input_tokens_seen": 16720320,
      "step": 28820
    },
    {
      "epoch": 4.293267798629729,
      "grad_norm": 10.1875,
      "learning_rate": 4.8024336304196927e-05,
      "loss": 1.1022,
      "num_input_tokens_seen": 16722752,
      "step": 28825
    },
    {
      "epoch": 4.294012511170688,
      "grad_norm": 10.0625,
      "learning_rate": 4.802307005279639e-05,
      "loss": 1.3487,
      "num_input_tokens_seen": 16725728,
      "step": 28830
    },
    {
      "epoch": 4.294757223711647,
      "grad_norm": 11.4375,
      "learning_rate": 4.8021803412444496e-05,
      "loss": 1.4163,
      "num_input_tokens_seen": 16728992,
      "step": 28835
    },
    {
      "epoch": 4.295501936252607,
      "grad_norm": 22.0,
      "learning_rate": 4.8020536383162615e-05,
      "loss": 1.3095,
      "num_input_tokens_seen": 16732192,
      "step": 28840
    },
    {
      "epoch": 4.2962466487935655,
      "grad_norm": 9.6875,
      "learning_rate": 4.8019268964972184e-05,
      "loss": 1.1938,
      "num_input_tokens_seen": 16735104,
      "step": 28845
    },
    {
      "epoch": 4.296991361334525,
      "grad_norm": 8.6875,
      "learning_rate": 4.801800115789459e-05,
      "loss": 0.8361,
      "num_input_tokens_seen": 16737856,
      "step": 28850
    },
    {
      "epoch": 4.297736073875484,
      "grad_norm": 6.875,
      "learning_rate": 4.801673296195126e-05,
      "loss": 0.8286,
      "num_input_tokens_seen": 16740736,
      "step": 28855
    },
    {
      "epoch": 4.298480786416444,
      "grad_norm": 6.6875,
      "learning_rate": 4.801546437716362e-05,
      "loss": 0.9579,
      "num_input_tokens_seen": 16743488,
      "step": 28860
    },
    {
      "epoch": 4.299225498957402,
      "grad_norm": 7.75,
      "learning_rate": 4.801419540355311e-05,
      "loss": 0.8911,
      "num_input_tokens_seen": 16746496,
      "step": 28865
    },
    {
      "epoch": 4.299970211498362,
      "grad_norm": 9.9375,
      "learning_rate": 4.801292604114115e-05,
      "loss": 1.398,
      "num_input_tokens_seen": 16749280,
      "step": 28870
    },
    {
      "epoch": 4.300714924039321,
      "grad_norm": 12.5625,
      "learning_rate": 4.80116562899492e-05,
      "loss": 1.2449,
      "num_input_tokens_seen": 16752320,
      "step": 28875
    },
    {
      "epoch": 4.30145963658028,
      "grad_norm": 12.875,
      "learning_rate": 4.80103861499987e-05,
      "loss": 1.4827,
      "num_input_tokens_seen": 16755360,
      "step": 28880
    },
    {
      "epoch": 4.302204349121239,
      "grad_norm": 10.375,
      "learning_rate": 4.800911562131112e-05,
      "loss": 1.175,
      "num_input_tokens_seen": 16758144,
      "step": 28885
    },
    {
      "epoch": 4.302949061662199,
      "grad_norm": 9.0,
      "learning_rate": 4.800784470390791e-05,
      "loss": 0.9393,
      "num_input_tokens_seen": 16760800,
      "step": 28890
    },
    {
      "epoch": 4.303693774203158,
      "grad_norm": 9.875,
      "learning_rate": 4.800657339781055e-05,
      "loss": 1.3211,
      "num_input_tokens_seen": 16763680,
      "step": 28895
    },
    {
      "epoch": 4.304438486744116,
      "grad_norm": 21.0,
      "learning_rate": 4.800530170304051e-05,
      "loss": 1.2583,
      "num_input_tokens_seen": 16766368,
      "step": 28900
    },
    {
      "epoch": 4.305183199285076,
      "grad_norm": 13.75,
      "learning_rate": 4.800402961961928e-05,
      "loss": 0.9955,
      "num_input_tokens_seen": 16769152,
      "step": 28905
    },
    {
      "epoch": 4.305927911826035,
      "grad_norm": 15.3125,
      "learning_rate": 4.800275714756836e-05,
      "loss": 1.1018,
      "num_input_tokens_seen": 16771808,
      "step": 28910
    },
    {
      "epoch": 4.306672624366994,
      "grad_norm": 14.375,
      "learning_rate": 4.800148428690923e-05,
      "loss": 0.914,
      "num_input_tokens_seen": 16774784,
      "step": 28915
    },
    {
      "epoch": 4.307417336907953,
      "grad_norm": 11.0,
      "learning_rate": 4.80002110376634e-05,
      "loss": 0.774,
      "num_input_tokens_seen": 16777664,
      "step": 28920
    },
    {
      "epoch": 4.308162049448913,
      "grad_norm": 9.4375,
      "learning_rate": 4.7998937399852386e-05,
      "loss": 0.9591,
      "num_input_tokens_seen": 16780416,
      "step": 28925
    },
    {
      "epoch": 4.3089067619898715,
      "grad_norm": 6.03125,
      "learning_rate": 4.799766337349769e-05,
      "loss": 0.8569,
      "num_input_tokens_seen": 16783360,
      "step": 28930
    },
    {
      "epoch": 4.309651474530831,
      "grad_norm": 9.25,
      "learning_rate": 4.799638895862085e-05,
      "loss": 1.1254,
      "num_input_tokens_seen": 16786400,
      "step": 28935
    },
    {
      "epoch": 4.31039618707179,
      "grad_norm": 12.5625,
      "learning_rate": 4.79951141552434e-05,
      "loss": 1.1193,
      "num_input_tokens_seen": 16789248,
      "step": 28940
    },
    {
      "epoch": 4.31114089961275,
      "grad_norm": 18.625,
      "learning_rate": 4.799383896338686e-05,
      "loss": 1.3191,
      "num_input_tokens_seen": 16792192,
      "step": 28945
    },
    {
      "epoch": 4.311885612153708,
      "grad_norm": 14.0625,
      "learning_rate": 4.7992563383072775e-05,
      "loss": 0.738,
      "num_input_tokens_seen": 16794912,
      "step": 28950
    },
    {
      "epoch": 4.312630324694668,
      "grad_norm": 18.25,
      "learning_rate": 4.799128741432271e-05,
      "loss": 1.4396,
      "num_input_tokens_seen": 16797824,
      "step": 28955
    },
    {
      "epoch": 4.313375037235627,
      "grad_norm": 13.375,
      "learning_rate": 4.7990011057158207e-05,
      "loss": 1.213,
      "num_input_tokens_seen": 16800800,
      "step": 28960
    },
    {
      "epoch": 4.314119749776586,
      "grad_norm": 11.375,
      "learning_rate": 4.798873431160084e-05,
      "loss": 1.0291,
      "num_input_tokens_seen": 16803520,
      "step": 28965
    },
    {
      "epoch": 4.314864462317545,
      "grad_norm": 11.25,
      "learning_rate": 4.798745717767216e-05,
      "loss": 1.1508,
      "num_input_tokens_seen": 16806656,
      "step": 28970
    },
    {
      "epoch": 4.315609174858505,
      "grad_norm": 12.625,
      "learning_rate": 4.7986179655393756e-05,
      "loss": 1.0053,
      "num_input_tokens_seen": 16809216,
      "step": 28975
    },
    {
      "epoch": 4.316353887399464,
      "grad_norm": 17.0,
      "learning_rate": 4.798490174478721e-05,
      "loss": 1.0082,
      "num_input_tokens_seen": 16812064,
      "step": 28980
    },
    {
      "epoch": 4.317098599940423,
      "grad_norm": 10.75,
      "learning_rate": 4.7983623445874114e-05,
      "loss": 1.2711,
      "num_input_tokens_seen": 16814944,
      "step": 28985
    },
    {
      "epoch": 4.317843312481382,
      "grad_norm": 7.4375,
      "learning_rate": 4.798234475867606e-05,
      "loss": 0.9001,
      "num_input_tokens_seen": 16817952,
      "step": 28990
    },
    {
      "epoch": 4.318588025022342,
      "grad_norm": 12.625,
      "learning_rate": 4.7981065683214645e-05,
      "loss": 1.1517,
      "num_input_tokens_seen": 16820928,
      "step": 28995
    },
    {
      "epoch": 4.3193327375633,
      "grad_norm": 9.875,
      "learning_rate": 4.797978621951148e-05,
      "loss": 1.1257,
      "num_input_tokens_seen": 16823840,
      "step": 29000
    },
    {
      "epoch": 4.32007745010426,
      "grad_norm": 11.6875,
      "learning_rate": 4.797850636758819e-05,
      "loss": 1.2078,
      "num_input_tokens_seen": 16826560,
      "step": 29005
    },
    {
      "epoch": 4.320822162645219,
      "grad_norm": 4.40625,
      "learning_rate": 4.7977226127466386e-05,
      "loss": 1.1659,
      "num_input_tokens_seen": 16829184,
      "step": 29010
    },
    {
      "epoch": 4.321566875186178,
      "grad_norm": 15.625,
      "learning_rate": 4.7975945499167696e-05,
      "loss": 1.1984,
      "num_input_tokens_seen": 16832032,
      "step": 29015
    },
    {
      "epoch": 4.322311587727137,
      "grad_norm": 7.53125,
      "learning_rate": 4.797466448271376e-05,
      "loss": 1.1708,
      "num_input_tokens_seen": 16834976,
      "step": 29020
    },
    {
      "epoch": 4.323056300268097,
      "grad_norm": 17.0,
      "learning_rate": 4.7973383078126223e-05,
      "loss": 1.1857,
      "num_input_tokens_seen": 16837792,
      "step": 29025
    },
    {
      "epoch": 4.323801012809056,
      "grad_norm": 4.03125,
      "learning_rate": 4.797210128542673e-05,
      "loss": 0.9324,
      "num_input_tokens_seen": 16840864,
      "step": 29030
    },
    {
      "epoch": 4.324545725350015,
      "grad_norm": 13.9375,
      "learning_rate": 4.7970819104636924e-05,
      "loss": 1.356,
      "num_input_tokens_seen": 16843840,
      "step": 29035
    },
    {
      "epoch": 4.325290437890974,
      "grad_norm": 11.8125,
      "learning_rate": 4.796953653577848e-05,
      "loss": 1.1269,
      "num_input_tokens_seen": 16846560,
      "step": 29040
    },
    {
      "epoch": 4.326035150431934,
      "grad_norm": 10.9375,
      "learning_rate": 4.7968253578873054e-05,
      "loss": 0.743,
      "num_input_tokens_seen": 16849056,
      "step": 29045
    },
    {
      "epoch": 4.326779862972892,
      "grad_norm": 4.65625,
      "learning_rate": 4.796697023394234e-05,
      "loss": 0.8218,
      "num_input_tokens_seen": 16851904,
      "step": 29050
    },
    {
      "epoch": 4.327524575513852,
      "grad_norm": 19.25,
      "learning_rate": 4.7965686501008e-05,
      "loss": 1.3035,
      "num_input_tokens_seen": 16854592,
      "step": 29055
    },
    {
      "epoch": 4.328269288054811,
      "grad_norm": 12.25,
      "learning_rate": 4.7964402380091734e-05,
      "loss": 1.2882,
      "num_input_tokens_seen": 16857312,
      "step": 29060
    },
    {
      "epoch": 4.32901400059577,
      "grad_norm": 9.3125,
      "learning_rate": 4.7963117871215224e-05,
      "loss": 1.2604,
      "num_input_tokens_seen": 16860416,
      "step": 29065
    },
    {
      "epoch": 4.329758713136729,
      "grad_norm": 14.625,
      "learning_rate": 4.796183297440018e-05,
      "loss": 1.4723,
      "num_input_tokens_seen": 16863360,
      "step": 29070
    },
    {
      "epoch": 4.330503425677689,
      "grad_norm": 8.625,
      "learning_rate": 4.79605476896683e-05,
      "loss": 1.001,
      "num_input_tokens_seen": 16866208,
      "step": 29075
    },
    {
      "epoch": 4.331248138218648,
      "grad_norm": 10.1875,
      "learning_rate": 4.795926201704131e-05,
      "loss": 1.0038,
      "num_input_tokens_seen": 16868992,
      "step": 29080
    },
    {
      "epoch": 4.331992850759606,
      "grad_norm": 16.5,
      "learning_rate": 4.795797595654091e-05,
      "loss": 1.2826,
      "num_input_tokens_seen": 16872032,
      "step": 29085
    },
    {
      "epoch": 4.332737563300566,
      "grad_norm": 9.625,
      "learning_rate": 4.795668950818885e-05,
      "loss": 1.1984,
      "num_input_tokens_seen": 16875424,
      "step": 29090
    },
    {
      "epoch": 4.333482275841525,
      "grad_norm": 14.0,
      "learning_rate": 4.7955402672006854e-05,
      "loss": 1.1783,
      "num_input_tokens_seen": 16878080,
      "step": 29095
    },
    {
      "epoch": 4.334226988382484,
      "grad_norm": 11.0,
      "learning_rate": 4.7954115448016654e-05,
      "loss": 1.034,
      "num_input_tokens_seen": 16880800,
      "step": 29100
    },
    {
      "epoch": 4.334971700923443,
      "grad_norm": 18.375,
      "learning_rate": 4.795282783624001e-05,
      "loss": 1.0397,
      "num_input_tokens_seen": 16883552,
      "step": 29105
    },
    {
      "epoch": 4.335716413464403,
      "grad_norm": 11.5625,
      "learning_rate": 4.795153983669867e-05,
      "loss": 1.2512,
      "num_input_tokens_seen": 16886464,
      "step": 29110
    },
    {
      "epoch": 4.336461126005362,
      "grad_norm": 8.75,
      "learning_rate": 4.795025144941438e-05,
      "loss": 1.0551,
      "num_input_tokens_seen": 16889664,
      "step": 29115
    },
    {
      "epoch": 4.337205838546321,
      "grad_norm": 8.4375,
      "learning_rate": 4.794896267440893e-05,
      "loss": 1.4643,
      "num_input_tokens_seen": 16892320,
      "step": 29120
    },
    {
      "epoch": 4.33795055108728,
      "grad_norm": 9.625,
      "learning_rate": 4.794767351170406e-05,
      "loss": 1.5615,
      "num_input_tokens_seen": 16895104,
      "step": 29125
    },
    {
      "epoch": 4.33869526362824,
      "grad_norm": 13.3125,
      "learning_rate": 4.794638396132159e-05,
      "loss": 1.2991,
      "num_input_tokens_seen": 16897824,
      "step": 29130
    },
    {
      "epoch": 4.339439976169198,
      "grad_norm": 9.4375,
      "learning_rate": 4.7945094023283275e-05,
      "loss": 1.2017,
      "num_input_tokens_seen": 16901024,
      "step": 29135
    },
    {
      "epoch": 4.340184688710158,
      "grad_norm": 5.125,
      "learning_rate": 4.794380369761092e-05,
      "loss": 0.9478,
      "num_input_tokens_seen": 16903840,
      "step": 29140
    },
    {
      "epoch": 4.340929401251117,
      "grad_norm": 7.5625,
      "learning_rate": 4.794251298432632e-05,
      "loss": 1.0771,
      "num_input_tokens_seen": 16906752,
      "step": 29145
    },
    {
      "epoch": 4.3416741137920765,
      "grad_norm": 12.125,
      "learning_rate": 4.794122188345128e-05,
      "loss": 1.0284,
      "num_input_tokens_seen": 16910048,
      "step": 29150
    },
    {
      "epoch": 4.342418826333035,
      "grad_norm": 5.625,
      "learning_rate": 4.7939930395007615e-05,
      "loss": 1.1954,
      "num_input_tokens_seen": 16912800,
      "step": 29155
    },
    {
      "epoch": 4.343163538873995,
      "grad_norm": 5.75,
      "learning_rate": 4.7938638519017134e-05,
      "loss": 1.2081,
      "num_input_tokens_seen": 16915680,
      "step": 29160
    },
    {
      "epoch": 4.343908251414954,
      "grad_norm": 5.96875,
      "learning_rate": 4.793734625550167e-05,
      "loss": 1.1232,
      "num_input_tokens_seen": 16918848,
      "step": 29165
    },
    {
      "epoch": 4.344652963955913,
      "grad_norm": 10.8125,
      "learning_rate": 4.7936053604483065e-05,
      "loss": 1.1148,
      "num_input_tokens_seen": 16921600,
      "step": 29170
    },
    {
      "epoch": 4.345397676496872,
      "grad_norm": 12.75,
      "learning_rate": 4.793476056598314e-05,
      "loss": 1.057,
      "num_input_tokens_seen": 16924256,
      "step": 29175
    },
    {
      "epoch": 4.346142389037832,
      "grad_norm": 10.9375,
      "learning_rate": 4.7933467140023736e-05,
      "loss": 0.9359,
      "num_input_tokens_seen": 16927136,
      "step": 29180
    },
    {
      "epoch": 4.34688710157879,
      "grad_norm": 9.4375,
      "learning_rate": 4.793217332662672e-05,
      "loss": 1.141,
      "num_input_tokens_seen": 16929888,
      "step": 29185
    },
    {
      "epoch": 4.34763181411975,
      "grad_norm": 9.0625,
      "learning_rate": 4.7930879125813945e-05,
      "loss": 0.8735,
      "num_input_tokens_seen": 16932928,
      "step": 29190
    },
    {
      "epoch": 4.348376526660709,
      "grad_norm": 10.6875,
      "learning_rate": 4.792958453760728e-05,
      "loss": 1.0523,
      "num_input_tokens_seen": 16935968,
      "step": 29195
    },
    {
      "epoch": 4.3491212392016685,
      "grad_norm": 5.15625,
      "learning_rate": 4.792828956202857e-05,
      "loss": 1.1166,
      "num_input_tokens_seen": 16938720,
      "step": 29200
    },
    {
      "epoch": 4.349865951742627,
      "grad_norm": 13.1875,
      "learning_rate": 4.792699419909972e-05,
      "loss": 1.0554,
      "num_input_tokens_seen": 16941536,
      "step": 29205
    },
    {
      "epoch": 4.350610664283587,
      "grad_norm": 11.75,
      "learning_rate": 4.792569844884261e-05,
      "loss": 1.107,
      "num_input_tokens_seen": 16944448,
      "step": 29210
    },
    {
      "epoch": 4.351355376824546,
      "grad_norm": 11.875,
      "learning_rate": 4.792440231127912e-05,
      "loss": 0.914,
      "num_input_tokens_seen": 16947424,
      "step": 29215
    },
    {
      "epoch": 4.352100089365505,
      "grad_norm": 13.75,
      "learning_rate": 4.792310578643116e-05,
      "loss": 1.3679,
      "num_input_tokens_seen": 16950336,
      "step": 29220
    },
    {
      "epoch": 4.352844801906464,
      "grad_norm": 8.0625,
      "learning_rate": 4.7921808874320616e-05,
      "loss": 1.3065,
      "num_input_tokens_seen": 16953056,
      "step": 29225
    },
    {
      "epoch": 4.353589514447424,
      "grad_norm": 24.0,
      "learning_rate": 4.792051157496941e-05,
      "loss": 1.3366,
      "num_input_tokens_seen": 16955872,
      "step": 29230
    },
    {
      "epoch": 4.3543342269883825,
      "grad_norm": 13.375,
      "learning_rate": 4.791921388839946e-05,
      "loss": 1.1232,
      "num_input_tokens_seen": 16958848,
      "step": 29235
    },
    {
      "epoch": 4.355078939529342,
      "grad_norm": 4.34375,
      "learning_rate": 4.791791581463268e-05,
      "loss": 1.1665,
      "num_input_tokens_seen": 16961600,
      "step": 29240
    },
    {
      "epoch": 4.355823652070301,
      "grad_norm": 5.8125,
      "learning_rate": 4.791661735369101e-05,
      "loss": 0.8195,
      "num_input_tokens_seen": 16964480,
      "step": 29245
    },
    {
      "epoch": 4.35656836461126,
      "grad_norm": 8.4375,
      "learning_rate": 4.791531850559637e-05,
      "loss": 1.2286,
      "num_input_tokens_seen": 16967296,
      "step": 29250
    },
    {
      "epoch": 4.357313077152219,
      "grad_norm": 5.4375,
      "learning_rate": 4.791401927037073e-05,
      "loss": 0.5662,
      "num_input_tokens_seen": 16970240,
      "step": 29255
    },
    {
      "epoch": 4.358057789693178,
      "grad_norm": 4.71875,
      "learning_rate": 4.791271964803602e-05,
      "loss": 1.2258,
      "num_input_tokens_seen": 16973024,
      "step": 29260
    },
    {
      "epoch": 4.358802502234138,
      "grad_norm": 6.15625,
      "learning_rate": 4.791141963861419e-05,
      "loss": 1.1587,
      "num_input_tokens_seen": 16975584,
      "step": 29265
    },
    {
      "epoch": 4.359547214775096,
      "grad_norm": 11.1875,
      "learning_rate": 4.791011924212721e-05,
      "loss": 1.0971,
      "num_input_tokens_seen": 16978624,
      "step": 29270
    },
    {
      "epoch": 4.360291927316056,
      "grad_norm": 7.34375,
      "learning_rate": 4.790881845859707e-05,
      "loss": 1.2845,
      "num_input_tokens_seen": 16981376,
      "step": 29275
    },
    {
      "epoch": 4.361036639857015,
      "grad_norm": 8.0,
      "learning_rate": 4.790751728804571e-05,
      "loss": 0.9685,
      "num_input_tokens_seen": 16984160,
      "step": 29280
    },
    {
      "epoch": 4.3617813523979745,
      "grad_norm": 6.1875,
      "learning_rate": 4.790621573049513e-05,
      "loss": 0.8736,
      "num_input_tokens_seen": 16987072,
      "step": 29285
    },
    {
      "epoch": 4.362526064938933,
      "grad_norm": 12.4375,
      "learning_rate": 4.790491378596731e-05,
      "loss": 1.2692,
      "num_input_tokens_seen": 16990080,
      "step": 29290
    },
    {
      "epoch": 4.363270777479893,
      "grad_norm": 10.0,
      "learning_rate": 4.7903611454484266e-05,
      "loss": 0.7172,
      "num_input_tokens_seen": 16992992,
      "step": 29295
    },
    {
      "epoch": 4.364015490020852,
      "grad_norm": 5.40625,
      "learning_rate": 4.790230873606797e-05,
      "loss": 0.9616,
      "num_input_tokens_seen": 16996000,
      "step": 29300
    },
    {
      "epoch": 4.364760202561811,
      "grad_norm": 16.25,
      "learning_rate": 4.790100563074045e-05,
      "loss": 0.9708,
      "num_input_tokens_seen": 16998816,
      "step": 29305
    },
    {
      "epoch": 4.36550491510277,
      "grad_norm": 16.125,
      "learning_rate": 4.789970213852372e-05,
      "loss": 1.1822,
      "num_input_tokens_seen": 17001664,
      "step": 29310
    },
    {
      "epoch": 4.36624962764373,
      "grad_norm": 9.5,
      "learning_rate": 4.789839825943979e-05,
      "loss": 1.3092,
      "num_input_tokens_seen": 17004704,
      "step": 29315
    },
    {
      "epoch": 4.3669943401846885,
      "grad_norm": 8.625,
      "learning_rate": 4.78970939935107e-05,
      "loss": 1.0972,
      "num_input_tokens_seen": 17007680,
      "step": 29320
    },
    {
      "epoch": 4.367739052725648,
      "grad_norm": 15.8125,
      "learning_rate": 4.789578934075847e-05,
      "loss": 0.9446,
      "num_input_tokens_seen": 17010624,
      "step": 29325
    },
    {
      "epoch": 4.368483765266607,
      "grad_norm": 8.25,
      "learning_rate": 4.7894484301205156e-05,
      "loss": 1.2276,
      "num_input_tokens_seen": 17013696,
      "step": 29330
    },
    {
      "epoch": 4.3692284778075665,
      "grad_norm": 10.0625,
      "learning_rate": 4.78931788748728e-05,
      "loss": 1.1401,
      "num_input_tokens_seen": 17016512,
      "step": 29335
    },
    {
      "epoch": 4.369973190348525,
      "grad_norm": 9.5,
      "learning_rate": 4.789187306178345e-05,
      "loss": 1.0528,
      "num_input_tokens_seen": 17019392,
      "step": 29340
    },
    {
      "epoch": 4.370717902889485,
      "grad_norm": 6.53125,
      "learning_rate": 4.789056686195917e-05,
      "loss": 0.7922,
      "num_input_tokens_seen": 17022560,
      "step": 29345
    },
    {
      "epoch": 4.371462615430444,
      "grad_norm": 10.625,
      "learning_rate": 4.788926027542203e-05,
      "loss": 1.2277,
      "num_input_tokens_seen": 17025408,
      "step": 29350
    },
    {
      "epoch": 4.372207327971403,
      "grad_norm": 8.5625,
      "learning_rate": 4.7887953302194106e-05,
      "loss": 0.7299,
      "num_input_tokens_seen": 17028288,
      "step": 29355
    },
    {
      "epoch": 4.372952040512362,
      "grad_norm": 11.25,
      "learning_rate": 4.788664594229747e-05,
      "loss": 1.1032,
      "num_input_tokens_seen": 17031072,
      "step": 29360
    },
    {
      "epoch": 4.373696753053322,
      "grad_norm": 6.0625,
      "learning_rate": 4.788533819575421e-05,
      "loss": 1.0104,
      "num_input_tokens_seen": 17034176,
      "step": 29365
    },
    {
      "epoch": 4.3744414655942805,
      "grad_norm": 11.0625,
      "learning_rate": 4.7884030062586424e-05,
      "loss": 1.0689,
      "num_input_tokens_seen": 17037120,
      "step": 29370
    },
    {
      "epoch": 4.37518617813524,
      "grad_norm": 10.8125,
      "learning_rate": 4.78827215428162e-05,
      "loss": 0.8057,
      "num_input_tokens_seen": 17040160,
      "step": 29375
    },
    {
      "epoch": 4.375930890676199,
      "grad_norm": 9.75,
      "learning_rate": 4.7881412636465664e-05,
      "loss": 1.2034,
      "num_input_tokens_seen": 17043072,
      "step": 29380
    },
    {
      "epoch": 4.3766756032171585,
      "grad_norm": 7.5,
      "learning_rate": 4.7880103343556906e-05,
      "loss": 1.0139,
      "num_input_tokens_seen": 17045888,
      "step": 29385
    },
    {
      "epoch": 4.377420315758117,
      "grad_norm": 3.8125,
      "learning_rate": 4.787879366411206e-05,
      "loss": 0.7429,
      "num_input_tokens_seen": 17048672,
      "step": 29390
    },
    {
      "epoch": 4.378165028299077,
      "grad_norm": 11.3125,
      "learning_rate": 4.787748359815326e-05,
      "loss": 1.0616,
      "num_input_tokens_seen": 17051744,
      "step": 29395
    },
    {
      "epoch": 4.378909740840036,
      "grad_norm": 5.09375,
      "learning_rate": 4.787617314570261e-05,
      "loss": 1.1064,
      "num_input_tokens_seen": 17054592,
      "step": 29400
    },
    {
      "epoch": 4.379654453380995,
      "grad_norm": 8.5625,
      "learning_rate": 4.7874862306782276e-05,
      "loss": 1.3136,
      "num_input_tokens_seen": 17057312,
      "step": 29405
    },
    {
      "epoch": 4.380399165921954,
      "grad_norm": 10.25,
      "learning_rate": 4.787355108141439e-05,
      "loss": 1.4108,
      "num_input_tokens_seen": 17060224,
      "step": 29410
    },
    {
      "epoch": 4.381143878462913,
      "grad_norm": 12.1875,
      "learning_rate": 4.78722394696211e-05,
      "loss": 1.1457,
      "num_input_tokens_seen": 17063072,
      "step": 29415
    },
    {
      "epoch": 4.3818885910038725,
      "grad_norm": 17.5,
      "learning_rate": 4.787092747142458e-05,
      "loss": 1.2074,
      "num_input_tokens_seen": 17066240,
      "step": 29420
    },
    {
      "epoch": 4.382633303544832,
      "grad_norm": 13.25,
      "learning_rate": 4.7869615086846973e-05,
      "loss": 1.2343,
      "num_input_tokens_seen": 17068992,
      "step": 29425
    },
    {
      "epoch": 4.383378016085791,
      "grad_norm": 9.0,
      "learning_rate": 4.786830231591047e-05,
      "loss": 0.8436,
      "num_input_tokens_seen": 17071936,
      "step": 29430
    },
    {
      "epoch": 4.38412272862675,
      "grad_norm": 10.125,
      "learning_rate": 4.786698915863724e-05,
      "loss": 1.1076,
      "num_input_tokens_seen": 17074592,
      "step": 29435
    },
    {
      "epoch": 4.384867441167709,
      "grad_norm": 9.4375,
      "learning_rate": 4.7865675615049464e-05,
      "loss": 1.2257,
      "num_input_tokens_seen": 17077632,
      "step": 29440
    },
    {
      "epoch": 4.385612153708668,
      "grad_norm": 11.125,
      "learning_rate": 4.786436168516935e-05,
      "loss": 1.2466,
      "num_input_tokens_seen": 17080448,
      "step": 29445
    },
    {
      "epoch": 4.386356866249628,
      "grad_norm": 14.3125,
      "learning_rate": 4.786304736901908e-05,
      "loss": 0.9178,
      "num_input_tokens_seen": 17083296,
      "step": 29450
    },
    {
      "epoch": 4.3871015787905865,
      "grad_norm": 8.8125,
      "learning_rate": 4.7861732666620856e-05,
      "loss": 1.2337,
      "num_input_tokens_seen": 17086528,
      "step": 29455
    },
    {
      "epoch": 4.387846291331546,
      "grad_norm": 6.4375,
      "learning_rate": 4.78604175779969e-05,
      "loss": 0.8355,
      "num_input_tokens_seen": 17089696,
      "step": 29460
    },
    {
      "epoch": 4.388591003872505,
      "grad_norm": 14.0625,
      "learning_rate": 4.7859102103169415e-05,
      "loss": 1.0988,
      "num_input_tokens_seen": 17092736,
      "step": 29465
    },
    {
      "epoch": 4.3893357164134645,
      "grad_norm": 9.375,
      "learning_rate": 4.785778624216064e-05,
      "loss": 1.1364,
      "num_input_tokens_seen": 17095552,
      "step": 29470
    },
    {
      "epoch": 4.390080428954423,
      "grad_norm": 12.5625,
      "learning_rate": 4.7856469994992805e-05,
      "loss": 1.044,
      "num_input_tokens_seen": 17098368,
      "step": 29475
    },
    {
      "epoch": 4.390825141495383,
      "grad_norm": 9.3125,
      "learning_rate": 4.7855153361688124e-05,
      "loss": 0.8841,
      "num_input_tokens_seen": 17101216,
      "step": 29480
    },
    {
      "epoch": 4.391569854036342,
      "grad_norm": 7.46875,
      "learning_rate": 4.785383634226887e-05,
      "loss": 1.0582,
      "num_input_tokens_seen": 17104224,
      "step": 29485
    },
    {
      "epoch": 4.392314566577301,
      "grad_norm": 19.375,
      "learning_rate": 4.785251893675727e-05,
      "loss": 1.2259,
      "num_input_tokens_seen": 17107136,
      "step": 29490
    },
    {
      "epoch": 4.39305927911826,
      "grad_norm": 7.90625,
      "learning_rate": 4.785120114517559e-05,
      "loss": 1.1661,
      "num_input_tokens_seen": 17110048,
      "step": 29495
    },
    {
      "epoch": 4.39380399165922,
      "grad_norm": 8.1875,
      "learning_rate": 4.7849882967546086e-05,
      "loss": 0.914,
      "num_input_tokens_seen": 17113216,
      "step": 29500
    },
    {
      "epoch": 4.3945487042001785,
      "grad_norm": 13.9375,
      "learning_rate": 4.784856440389105e-05,
      "loss": 0.7835,
      "num_input_tokens_seen": 17115968,
      "step": 29505
    },
    {
      "epoch": 4.395293416741138,
      "grad_norm": 11.75,
      "learning_rate": 4.784724545423272e-05,
      "loss": 1.0617,
      "num_input_tokens_seen": 17118656,
      "step": 29510
    },
    {
      "epoch": 4.396038129282097,
      "grad_norm": 3.59375,
      "learning_rate": 4.7845926118593415e-05,
      "loss": 1.126,
      "num_input_tokens_seen": 17122144,
      "step": 29515
    },
    {
      "epoch": 4.396782841823057,
      "grad_norm": 6.4375,
      "learning_rate": 4.784460639699541e-05,
      "loss": 0.9417,
      "num_input_tokens_seen": 17125120,
      "step": 29520
    },
    {
      "epoch": 4.397527554364015,
      "grad_norm": 13.125,
      "learning_rate": 4.784328628946098e-05,
      "loss": 1.1845,
      "num_input_tokens_seen": 17127776,
      "step": 29525
    },
    {
      "epoch": 4.398272266904975,
      "grad_norm": 12.4375,
      "learning_rate": 4.784196579601246e-05,
      "loss": 0.9974,
      "num_input_tokens_seen": 17130496,
      "step": 29530
    },
    {
      "epoch": 4.399016979445934,
      "grad_norm": 8.75,
      "learning_rate": 4.784064491667214e-05,
      "loss": 1.2448,
      "num_input_tokens_seen": 17133344,
      "step": 29535
    },
    {
      "epoch": 4.399761691986893,
      "grad_norm": 15.5,
      "learning_rate": 4.7839323651462334e-05,
      "loss": 1.388,
      "num_input_tokens_seen": 17136224,
      "step": 29540
    },
    {
      "epoch": 4.400506404527852,
      "grad_norm": 10.4375,
      "learning_rate": 4.783800200040537e-05,
      "loss": 1.1386,
      "num_input_tokens_seen": 17139232,
      "step": 29545
    },
    {
      "epoch": 4.401251117068812,
      "grad_norm": 6.71875,
      "learning_rate": 4.783667996352357e-05,
      "loss": 0.9596,
      "num_input_tokens_seen": 17142208,
      "step": 29550
    },
    {
      "epoch": 4.4019958296097705,
      "grad_norm": 15.125,
      "learning_rate": 4.783535754083927e-05,
      "loss": 1.419,
      "num_input_tokens_seen": 17145024,
      "step": 29555
    },
    {
      "epoch": 4.40274054215073,
      "grad_norm": 5.40625,
      "learning_rate": 4.783403473237483e-05,
      "loss": 0.8437,
      "num_input_tokens_seen": 17147936,
      "step": 29560
    },
    {
      "epoch": 4.403485254691689,
      "grad_norm": 6.4375,
      "learning_rate": 4.783271153815257e-05,
      "loss": 1.1389,
      "num_input_tokens_seen": 17150720,
      "step": 29565
    },
    {
      "epoch": 4.404229967232649,
      "grad_norm": 12.0625,
      "learning_rate": 4.783138795819485e-05,
      "loss": 0.8972,
      "num_input_tokens_seen": 17153408,
      "step": 29570
    },
    {
      "epoch": 4.404974679773607,
      "grad_norm": 5.53125,
      "learning_rate": 4.783006399252404e-05,
      "loss": 0.9837,
      "num_input_tokens_seen": 17156576,
      "step": 29575
    },
    {
      "epoch": 4.405719392314566,
      "grad_norm": 17.0,
      "learning_rate": 4.782873964116251e-05,
      "loss": 1.38,
      "num_input_tokens_seen": 17159648,
      "step": 29580
    },
    {
      "epoch": 4.406464104855526,
      "grad_norm": 6.0625,
      "learning_rate": 4.782741490413262e-05,
      "loss": 0.9603,
      "num_input_tokens_seen": 17162432,
      "step": 29585
    },
    {
      "epoch": 4.407208817396485,
      "grad_norm": 20.125,
      "learning_rate": 4.782608978145675e-05,
      "loss": 1.1805,
      "num_input_tokens_seen": 17165152,
      "step": 29590
    },
    {
      "epoch": 4.407953529937444,
      "grad_norm": 12.1875,
      "learning_rate": 4.7824764273157295e-05,
      "loss": 1.0581,
      "num_input_tokens_seen": 17168000,
      "step": 29595
    },
    {
      "epoch": 4.408698242478403,
      "grad_norm": 7.40625,
      "learning_rate": 4.782343837925665e-05,
      "loss": 1.2584,
      "num_input_tokens_seen": 17170976,
      "step": 29600
    },
    {
      "epoch": 4.409442955019363,
      "grad_norm": 16.375,
      "learning_rate": 4.7822112099777205e-05,
      "loss": 1.2846,
      "num_input_tokens_seen": 17173600,
      "step": 29605
    },
    {
      "epoch": 4.410187667560321,
      "grad_norm": 10.3125,
      "learning_rate": 4.7820785434741375e-05,
      "loss": 1.0881,
      "num_input_tokens_seen": 17176512,
      "step": 29610
    },
    {
      "epoch": 4.410932380101281,
      "grad_norm": 13.25,
      "learning_rate": 4.7819458384171566e-05,
      "loss": 1.3098,
      "num_input_tokens_seen": 17179392,
      "step": 29615
    },
    {
      "epoch": 4.41167709264224,
      "grad_norm": 7.75,
      "learning_rate": 4.78181309480902e-05,
      "loss": 0.8878,
      "num_input_tokens_seen": 17182528,
      "step": 29620
    },
    {
      "epoch": 4.412421805183199,
      "grad_norm": 11.6875,
      "learning_rate": 4.781680312651971e-05,
      "loss": 0.9664,
      "num_input_tokens_seen": 17185696,
      "step": 29625
    },
    {
      "epoch": 4.413166517724158,
      "grad_norm": 11.625,
      "learning_rate": 4.781547491948252e-05,
      "loss": 1.4354,
      "num_input_tokens_seen": 17188832,
      "step": 29630
    },
    {
      "epoch": 4.413911230265118,
      "grad_norm": 12.5625,
      "learning_rate": 4.7814146327001067e-05,
      "loss": 1.074,
      "num_input_tokens_seen": 17191808,
      "step": 29635
    },
    {
      "epoch": 4.4146559428060765,
      "grad_norm": 14.375,
      "learning_rate": 4.7812817349097796e-05,
      "loss": 1.2618,
      "num_input_tokens_seen": 17195040,
      "step": 29640
    },
    {
      "epoch": 4.415400655347036,
      "grad_norm": 6.84375,
      "learning_rate": 4.7811487985795164e-05,
      "loss": 1.2856,
      "num_input_tokens_seen": 17197568,
      "step": 29645
    },
    {
      "epoch": 4.416145367887995,
      "grad_norm": 6.9375,
      "learning_rate": 4.781015823711563e-05,
      "loss": 1.038,
      "num_input_tokens_seen": 17200544,
      "step": 29650
    },
    {
      "epoch": 4.416890080428955,
      "grad_norm": 9.5625,
      "learning_rate": 4.780882810308165e-05,
      "loss": 0.921,
      "num_input_tokens_seen": 17203200,
      "step": 29655
    },
    {
      "epoch": 4.417634792969913,
      "grad_norm": 8.125,
      "learning_rate": 4.7807497583715704e-05,
      "loss": 0.909,
      "num_input_tokens_seen": 17205856,
      "step": 29660
    },
    {
      "epoch": 4.418379505510873,
      "grad_norm": 17.5,
      "learning_rate": 4.780616667904026e-05,
      "loss": 1.0069,
      "num_input_tokens_seen": 17208672,
      "step": 29665
    },
    {
      "epoch": 4.419124218051832,
      "grad_norm": 6.0625,
      "learning_rate": 4.7804835389077824e-05,
      "loss": 0.6387,
      "num_input_tokens_seen": 17211296,
      "step": 29670
    },
    {
      "epoch": 4.419868930592791,
      "grad_norm": 8.375,
      "learning_rate": 4.780350371385086e-05,
      "loss": 1.1391,
      "num_input_tokens_seen": 17214336,
      "step": 29675
    },
    {
      "epoch": 4.42061364313375,
      "grad_norm": 9.9375,
      "learning_rate": 4.7802171653381885e-05,
      "loss": 1.3308,
      "num_input_tokens_seen": 17217504,
      "step": 29680
    },
    {
      "epoch": 4.42135835567471,
      "grad_norm": 8.375,
      "learning_rate": 4.780083920769339e-05,
      "loss": 0.8506,
      "num_input_tokens_seen": 17220160,
      "step": 29685
    },
    {
      "epoch": 4.422103068215669,
      "grad_norm": 10.5625,
      "learning_rate": 4.779950637680789e-05,
      "loss": 0.8213,
      "num_input_tokens_seen": 17223168,
      "step": 29690
    },
    {
      "epoch": 4.422847780756628,
      "grad_norm": 14.0625,
      "learning_rate": 4.7798173160747906e-05,
      "loss": 1.0646,
      "num_input_tokens_seen": 17225888,
      "step": 29695
    },
    {
      "epoch": 4.423592493297587,
      "grad_norm": 11.1875,
      "learning_rate": 4.7796839559535955e-05,
      "loss": 0.9006,
      "num_input_tokens_seen": 17228576,
      "step": 29700
    },
    {
      "epoch": 4.424337205838547,
      "grad_norm": 10.4375,
      "learning_rate": 4.779550557319457e-05,
      "loss": 1.0999,
      "num_input_tokens_seen": 17231488,
      "step": 29705
    },
    {
      "epoch": 4.425081918379505,
      "grad_norm": 9.375,
      "learning_rate": 4.7794171201746285e-05,
      "loss": 1.3541,
      "num_input_tokens_seen": 17234208,
      "step": 29710
    },
    {
      "epoch": 4.425826630920465,
      "grad_norm": 8.0,
      "learning_rate": 4.779283644521365e-05,
      "loss": 1.1404,
      "num_input_tokens_seen": 17236896,
      "step": 29715
    },
    {
      "epoch": 4.426571343461424,
      "grad_norm": 10.0,
      "learning_rate": 4.7791501303619205e-05,
      "loss": 0.9578,
      "num_input_tokens_seen": 17239712,
      "step": 29720
    },
    {
      "epoch": 4.427316056002383,
      "grad_norm": 7.96875,
      "learning_rate": 4.7790165776985504e-05,
      "loss": 0.958,
      "num_input_tokens_seen": 17242688,
      "step": 29725
    },
    {
      "epoch": 4.428060768543342,
      "grad_norm": 6.625,
      "learning_rate": 4.7788829865335125e-05,
      "loss": 1.369,
      "num_input_tokens_seen": 17245440,
      "step": 29730
    },
    {
      "epoch": 4.428805481084302,
      "grad_norm": 17.0,
      "learning_rate": 4.778749356869062e-05,
      "loss": 1.2334,
      "num_input_tokens_seen": 17248256,
      "step": 29735
    },
    {
      "epoch": 4.429550193625261,
      "grad_norm": 9.3125,
      "learning_rate": 4.778615688707457e-05,
      "loss": 1.0075,
      "num_input_tokens_seen": 17251168,
      "step": 29740
    },
    {
      "epoch": 4.43029490616622,
      "grad_norm": 8.875,
      "learning_rate": 4.778481982050956e-05,
      "loss": 0.9957,
      "num_input_tokens_seen": 17253920,
      "step": 29745
    },
    {
      "epoch": 4.431039618707179,
      "grad_norm": 7.75,
      "learning_rate": 4.778348236901818e-05,
      "loss": 1.4124,
      "num_input_tokens_seen": 17256896,
      "step": 29750
    },
    {
      "epoch": 4.431784331248139,
      "grad_norm": 7.875,
      "learning_rate": 4.7782144532623016e-05,
      "loss": 1.2549,
      "num_input_tokens_seen": 17259808,
      "step": 29755
    },
    {
      "epoch": 4.432529043789097,
      "grad_norm": 13.0625,
      "learning_rate": 4.7780806311346684e-05,
      "loss": 0.9087,
      "num_input_tokens_seen": 17262592,
      "step": 29760
    },
    {
      "epoch": 4.433273756330056,
      "grad_norm": 9.4375,
      "learning_rate": 4.777946770521178e-05,
      "loss": 1.4379,
      "num_input_tokens_seen": 17265632,
      "step": 29765
    },
    {
      "epoch": 4.434018468871016,
      "grad_norm": 6.5,
      "learning_rate": 4.7778128714240915e-05,
      "loss": 0.8552,
      "num_input_tokens_seen": 17268672,
      "step": 29770
    },
    {
      "epoch": 4.434763181411975,
      "grad_norm": 13.75,
      "learning_rate": 4.7776789338456717e-05,
      "loss": 1.0154,
      "num_input_tokens_seen": 17271744,
      "step": 29775
    },
    {
      "epoch": 4.435507893952934,
      "grad_norm": 16.625,
      "learning_rate": 4.777544957788182e-05,
      "loss": 0.8308,
      "num_input_tokens_seen": 17274752,
      "step": 29780
    },
    {
      "epoch": 4.436252606493893,
      "grad_norm": 9.0,
      "learning_rate": 4.7774109432538843e-05,
      "loss": 1.2148,
      "num_input_tokens_seen": 17277600,
      "step": 29785
    },
    {
      "epoch": 4.436997319034853,
      "grad_norm": 9.3125,
      "learning_rate": 4.777276890245044e-05,
      "loss": 1.1218,
      "num_input_tokens_seen": 17280448,
      "step": 29790
    },
    {
      "epoch": 4.437742031575811,
      "grad_norm": 14.5625,
      "learning_rate": 4.7771427987639246e-05,
      "loss": 0.9738,
      "num_input_tokens_seen": 17283264,
      "step": 29795
    },
    {
      "epoch": 4.438486744116771,
      "grad_norm": 10.5,
      "learning_rate": 4.777008668812793e-05,
      "loss": 0.7442,
      "num_input_tokens_seen": 17286560,
      "step": 29800
    },
    {
      "epoch": 4.43923145665773,
      "grad_norm": 10.0625,
      "learning_rate": 4.776874500393912e-05,
      "loss": 1.1887,
      "num_input_tokens_seen": 17289376,
      "step": 29805
    },
    {
      "epoch": 4.439976169198689,
      "grad_norm": 12.125,
      "learning_rate": 4.7767402935095525e-05,
      "loss": 1.0568,
      "num_input_tokens_seen": 17292224,
      "step": 29810
    },
    {
      "epoch": 4.440720881739648,
      "grad_norm": 6.0,
      "learning_rate": 4.776606048161979e-05,
      "loss": 0.9529,
      "num_input_tokens_seen": 17295520,
      "step": 29815
    },
    {
      "epoch": 4.441465594280608,
      "grad_norm": 10.1875,
      "learning_rate": 4.77647176435346e-05,
      "loss": 0.8329,
      "num_input_tokens_seen": 17298464,
      "step": 29820
    },
    {
      "epoch": 4.442210306821567,
      "grad_norm": 6.75,
      "learning_rate": 4.7763374420862645e-05,
      "loss": 0.8495,
      "num_input_tokens_seen": 17301120,
      "step": 29825
    },
    {
      "epoch": 4.442955019362526,
      "grad_norm": 9.5,
      "learning_rate": 4.7762030813626615e-05,
      "loss": 1.0606,
      "num_input_tokens_seen": 17303840,
      "step": 29830
    },
    {
      "epoch": 4.443699731903485,
      "grad_norm": 9.1875,
      "learning_rate": 4.776068682184921e-05,
      "loss": 1.0602,
      "num_input_tokens_seen": 17306624,
      "step": 29835
    },
    {
      "epoch": 4.444444444444445,
      "grad_norm": 5.78125,
      "learning_rate": 4.7759342445553124e-05,
      "loss": 1.146,
      "num_input_tokens_seen": 17309472,
      "step": 29840
    },
    {
      "epoch": 4.445189156985403,
      "grad_norm": 6.34375,
      "learning_rate": 4.775799768476109e-05,
      "loss": 1.3604,
      "num_input_tokens_seen": 17312384,
      "step": 29845
    },
    {
      "epoch": 4.445933869526363,
      "grad_norm": 8.125,
      "learning_rate": 4.775665253949581e-05,
      "loss": 0.9755,
      "num_input_tokens_seen": 17315392,
      "step": 29850
    },
    {
      "epoch": 4.446678582067322,
      "grad_norm": 5.6875,
      "learning_rate": 4.775530700978002e-05,
      "loss": 0.8841,
      "num_input_tokens_seen": 17318272,
      "step": 29855
    },
    {
      "epoch": 4.4474232946082815,
      "grad_norm": 12.3125,
      "learning_rate": 4.775396109563644e-05,
      "loss": 1.1233,
      "num_input_tokens_seen": 17321152,
      "step": 29860
    },
    {
      "epoch": 4.44816800714924,
      "grad_norm": 8.5625,
      "learning_rate": 4.775261479708781e-05,
      "loss": 1.224,
      "num_input_tokens_seen": 17324128,
      "step": 29865
    },
    {
      "epoch": 4.4489127196902,
      "grad_norm": 5.15625,
      "learning_rate": 4.775126811415689e-05,
      "loss": 1.0259,
      "num_input_tokens_seen": 17327008,
      "step": 29870
    },
    {
      "epoch": 4.449657432231159,
      "grad_norm": 8.625,
      "learning_rate": 4.7749921046866407e-05,
      "loss": 1.1479,
      "num_input_tokens_seen": 17329728,
      "step": 29875
    },
    {
      "epoch": 4.450402144772118,
      "grad_norm": 7.34375,
      "learning_rate": 4.7748573595239134e-05,
      "loss": 0.8355,
      "num_input_tokens_seen": 17332480,
      "step": 29880
    },
    {
      "epoch": 4.451146857313077,
      "grad_norm": 10.875,
      "learning_rate": 4.7747225759297835e-05,
      "loss": 1.1156,
      "num_input_tokens_seen": 17335232,
      "step": 29885
    },
    {
      "epoch": 4.451891569854037,
      "grad_norm": 7.5625,
      "learning_rate": 4.774587753906526e-05,
      "loss": 0.7333,
      "num_input_tokens_seen": 17338016,
      "step": 29890
    },
    {
      "epoch": 4.452636282394995,
      "grad_norm": 17.125,
      "learning_rate": 4.774452893456423e-05,
      "loss": 1.1573,
      "num_input_tokens_seen": 17340736,
      "step": 29895
    },
    {
      "epoch": 4.453380994935955,
      "grad_norm": 12.875,
      "learning_rate": 4.774317994581748e-05,
      "loss": 1.4366,
      "num_input_tokens_seen": 17343584,
      "step": 29900
    },
    {
      "epoch": 4.454125707476914,
      "grad_norm": 12.6875,
      "learning_rate": 4.7741830572847826e-05,
      "loss": 1.2079,
      "num_input_tokens_seen": 17346432,
      "step": 29905
    },
    {
      "epoch": 4.4548704200178735,
      "grad_norm": 9.875,
      "learning_rate": 4.774048081567805e-05,
      "loss": 0.8838,
      "num_input_tokens_seen": 17349280,
      "step": 29910
    },
    {
      "epoch": 4.455615132558832,
      "grad_norm": 9.0,
      "learning_rate": 4.7739130674330966e-05,
      "loss": 1.2023,
      "num_input_tokens_seen": 17351936,
      "step": 29915
    },
    {
      "epoch": 4.456359845099792,
      "grad_norm": 10.9375,
      "learning_rate": 4.773778014882939e-05,
      "loss": 1.0687,
      "num_input_tokens_seen": 17354912,
      "step": 29920
    },
    {
      "epoch": 4.457104557640751,
      "grad_norm": 6.375,
      "learning_rate": 4.773642923919612e-05,
      "loss": 0.8655,
      "num_input_tokens_seen": 17357920,
      "step": 29925
    },
    {
      "epoch": 4.457849270181709,
      "grad_norm": 6.125,
      "learning_rate": 4.773507794545399e-05,
      "loss": 0.9589,
      "num_input_tokens_seen": 17360768,
      "step": 29930
    },
    {
      "epoch": 4.458593982722669,
      "grad_norm": 8.625,
      "learning_rate": 4.7733726267625824e-05,
      "loss": 1.1808,
      "num_input_tokens_seen": 17363456,
      "step": 29935
    },
    {
      "epoch": 4.459338695263629,
      "grad_norm": 8.0625,
      "learning_rate": 4.7732374205734456e-05,
      "loss": 1.1531,
      "num_input_tokens_seen": 17366464,
      "step": 29940
    },
    {
      "epoch": 4.4600834078045875,
      "grad_norm": 4.5,
      "learning_rate": 4.773102175980273e-05,
      "loss": 1.0136,
      "num_input_tokens_seen": 17369344,
      "step": 29945
    },
    {
      "epoch": 4.460828120345546,
      "grad_norm": 9.875,
      "learning_rate": 4.772966892985349e-05,
      "loss": 0.9368,
      "num_input_tokens_seen": 17372128,
      "step": 29950
    },
    {
      "epoch": 4.461572832886506,
      "grad_norm": 8.5,
      "learning_rate": 4.77283157159096e-05,
      "loss": 0.7226,
      "num_input_tokens_seen": 17375008,
      "step": 29955
    },
    {
      "epoch": 4.462317545427465,
      "grad_norm": 5.0,
      "learning_rate": 4.772696211799392e-05,
      "loss": 1.0743,
      "num_input_tokens_seen": 17377952,
      "step": 29960
    },
    {
      "epoch": 4.463062257968424,
      "grad_norm": 9.9375,
      "learning_rate": 4.7725608136129305e-05,
      "loss": 1.1359,
      "num_input_tokens_seen": 17380640,
      "step": 29965
    },
    {
      "epoch": 4.463806970509383,
      "grad_norm": 13.375,
      "learning_rate": 4.7724253770338645e-05,
      "loss": 1.1655,
      "num_input_tokens_seen": 17383840,
      "step": 29970
    },
    {
      "epoch": 4.464551683050343,
      "grad_norm": 12.0,
      "learning_rate": 4.772289902064481e-05,
      "loss": 1.1065,
      "num_input_tokens_seen": 17386592,
      "step": 29975
    },
    {
      "epoch": 4.465296395591301,
      "grad_norm": 23.125,
      "learning_rate": 4.772154388707069e-05,
      "loss": 0.954,
      "num_input_tokens_seen": 17389536,
      "step": 29980
    },
    {
      "epoch": 4.466041108132261,
      "grad_norm": 9.8125,
      "learning_rate": 4.7720188369639186e-05,
      "loss": 0.8369,
      "num_input_tokens_seen": 17392224,
      "step": 29985
    },
    {
      "epoch": 4.46678582067322,
      "grad_norm": 4.03125,
      "learning_rate": 4.771883246837318e-05,
      "loss": 0.9403,
      "num_input_tokens_seen": 17395168,
      "step": 29990
    },
    {
      "epoch": 4.4675305332141795,
      "grad_norm": 10.4375,
      "learning_rate": 4.77174761832956e-05,
      "loss": 0.9005,
      "num_input_tokens_seen": 17397984,
      "step": 29995
    },
    {
      "epoch": 4.468275245755138,
      "grad_norm": 8.9375,
      "learning_rate": 4.771611951442935e-05,
      "loss": 1.1427,
      "num_input_tokens_seen": 17400928,
      "step": 30000
    },
    {
      "epoch": 4.469019958296098,
      "grad_norm": 7.84375,
      "learning_rate": 4.771476246179734e-05,
      "loss": 1.3171,
      "num_input_tokens_seen": 17403424,
      "step": 30005
    },
    {
      "epoch": 4.469764670837057,
      "grad_norm": 7.84375,
      "learning_rate": 4.7713405025422505e-05,
      "loss": 0.9303,
      "num_input_tokens_seen": 17406432,
      "step": 30010
    },
    {
      "epoch": 4.470509383378016,
      "grad_norm": 11.0,
      "learning_rate": 4.771204720532778e-05,
      "loss": 1.0785,
      "num_input_tokens_seen": 17409152,
      "step": 30015
    },
    {
      "epoch": 4.471254095918975,
      "grad_norm": 12.0,
      "learning_rate": 4.7710689001536105e-05,
      "loss": 1.1915,
      "num_input_tokens_seen": 17411872,
      "step": 30020
    },
    {
      "epoch": 4.471998808459935,
      "grad_norm": 15.3125,
      "learning_rate": 4.7709330414070406e-05,
      "loss": 0.9956,
      "num_input_tokens_seen": 17414752,
      "step": 30025
    },
    {
      "epoch": 4.4727435210008935,
      "grad_norm": 5.625,
      "learning_rate": 4.770797144295366e-05,
      "loss": 1.0998,
      "num_input_tokens_seen": 17417472,
      "step": 30030
    },
    {
      "epoch": 4.473488233541853,
      "grad_norm": 6.53125,
      "learning_rate": 4.7706612088208826e-05,
      "loss": 0.6921,
      "num_input_tokens_seen": 17420320,
      "step": 30035
    },
    {
      "epoch": 4.474232946082812,
      "grad_norm": 8.625,
      "learning_rate": 4.770525234985884e-05,
      "loss": 1.4184,
      "num_input_tokens_seen": 17423072,
      "step": 30040
    },
    {
      "epoch": 4.4749776586237715,
      "grad_norm": 14.125,
      "learning_rate": 4.770389222792671e-05,
      "loss": 1.2059,
      "num_input_tokens_seen": 17425856,
      "step": 30045
    },
    {
      "epoch": 4.47572237116473,
      "grad_norm": 10.5625,
      "learning_rate": 4.770253172243538e-05,
      "loss": 0.9367,
      "num_input_tokens_seen": 17428704,
      "step": 30050
    },
    {
      "epoch": 4.47646708370569,
      "grad_norm": 6.34375,
      "learning_rate": 4.770117083340786e-05,
      "loss": 1.1524,
      "num_input_tokens_seen": 17431712,
      "step": 30055
    },
    {
      "epoch": 4.477211796246649,
      "grad_norm": 8.4375,
      "learning_rate": 4.769980956086714e-05,
      "loss": 0.9408,
      "num_input_tokens_seen": 17434720,
      "step": 30060
    },
    {
      "epoch": 4.477956508787608,
      "grad_norm": 13.375,
      "learning_rate": 4.769844790483619e-05,
      "loss": 1.1607,
      "num_input_tokens_seen": 17437280,
      "step": 30065
    },
    {
      "epoch": 4.478701221328567,
      "grad_norm": 8.3125,
      "learning_rate": 4.769708586533804e-05,
      "loss": 1.0969,
      "num_input_tokens_seen": 17440224,
      "step": 30070
    },
    {
      "epoch": 4.479445933869527,
      "grad_norm": 13.8125,
      "learning_rate": 4.7695723442395694e-05,
      "loss": 1.0523,
      "num_input_tokens_seen": 17443040,
      "step": 30075
    },
    {
      "epoch": 4.4801906464104855,
      "grad_norm": 12.1875,
      "learning_rate": 4.769436063603217e-05,
      "loss": 1.3266,
      "num_input_tokens_seen": 17446016,
      "step": 30080
    },
    {
      "epoch": 4.480935358951445,
      "grad_norm": 16.625,
      "learning_rate": 4.769299744627048e-05,
      "loss": 0.8509,
      "num_input_tokens_seen": 17448768,
      "step": 30085
    },
    {
      "epoch": 4.481680071492404,
      "grad_norm": 12.125,
      "learning_rate": 4.769163387313367e-05,
      "loss": 1.1414,
      "num_input_tokens_seen": 17451552,
      "step": 30090
    },
    {
      "epoch": 4.4824247840333635,
      "grad_norm": 16.875,
      "learning_rate": 4.7690269916644766e-05,
      "loss": 0.9536,
      "num_input_tokens_seen": 17454368,
      "step": 30095
    },
    {
      "epoch": 4.483169496574322,
      "grad_norm": 12.5,
      "learning_rate": 4.768890557682681e-05,
      "loss": 1.3679,
      "num_input_tokens_seen": 17457504,
      "step": 30100
    },
    {
      "epoch": 4.483914209115282,
      "grad_norm": 9.0625,
      "learning_rate": 4.768754085370286e-05,
      "loss": 1.0594,
      "num_input_tokens_seen": 17460384,
      "step": 30105
    },
    {
      "epoch": 4.484658921656241,
      "grad_norm": 14.1875,
      "learning_rate": 4.768617574729596e-05,
      "loss": 1.4148,
      "num_input_tokens_seen": 17462976,
      "step": 30110
    },
    {
      "epoch": 4.4854036341971995,
      "grad_norm": 12.125,
      "learning_rate": 4.768481025762918e-05,
      "loss": 1.1505,
      "num_input_tokens_seen": 17465856,
      "step": 30115
    },
    {
      "epoch": 4.486148346738159,
      "grad_norm": 23.125,
      "learning_rate": 4.768344438472559e-05,
      "loss": 0.985,
      "num_input_tokens_seen": 17468512,
      "step": 30120
    },
    {
      "epoch": 4.486893059279118,
      "grad_norm": 14.8125,
      "learning_rate": 4.768207812860826e-05,
      "loss": 1.1503,
      "num_input_tokens_seen": 17471424,
      "step": 30125
    },
    {
      "epoch": 4.4876377718200775,
      "grad_norm": 8.875,
      "learning_rate": 4.768071148930027e-05,
      "loss": 1.0152,
      "num_input_tokens_seen": 17474336,
      "step": 30130
    },
    {
      "epoch": 4.488382484361036,
      "grad_norm": 10.9375,
      "learning_rate": 4.7679344466824716e-05,
      "loss": 1.0483,
      "num_input_tokens_seen": 17477376,
      "step": 30135
    },
    {
      "epoch": 4.489127196901996,
      "grad_norm": 11.5,
      "learning_rate": 4.767797706120468e-05,
      "loss": 0.8339,
      "num_input_tokens_seen": 17480416,
      "step": 30140
    },
    {
      "epoch": 4.489871909442955,
      "grad_norm": 10.9375,
      "learning_rate": 4.767660927246328e-05,
      "loss": 1.2107,
      "num_input_tokens_seen": 17483424,
      "step": 30145
    },
    {
      "epoch": 4.490616621983914,
      "grad_norm": 9.1875,
      "learning_rate": 4.7675241100623604e-05,
      "loss": 1.1247,
      "num_input_tokens_seen": 17486496,
      "step": 30150
    },
    {
      "epoch": 4.491361334524873,
      "grad_norm": 8.5625,
      "learning_rate": 4.7673872545708784e-05,
      "loss": 0.9103,
      "num_input_tokens_seen": 17489376,
      "step": 30155
    },
    {
      "epoch": 4.492106047065833,
      "grad_norm": 10.0625,
      "learning_rate": 4.767250360774193e-05,
      "loss": 0.9456,
      "num_input_tokens_seen": 17492480,
      "step": 30160
    },
    {
      "epoch": 4.4928507596067915,
      "grad_norm": 8.375,
      "learning_rate": 4.767113428674616e-05,
      "loss": 0.9931,
      "num_input_tokens_seen": 17495328,
      "step": 30165
    },
    {
      "epoch": 4.493595472147751,
      "grad_norm": 20.125,
      "learning_rate": 4.766976458274464e-05,
      "loss": 1.1873,
      "num_input_tokens_seen": 17498208,
      "step": 30170
    },
    {
      "epoch": 4.49434018468871,
      "grad_norm": 11.1875,
      "learning_rate": 4.766839449576047e-05,
      "loss": 1.0245,
      "num_input_tokens_seen": 17501312,
      "step": 30175
    },
    {
      "epoch": 4.4950848972296695,
      "grad_norm": 7.75,
      "learning_rate": 4.766702402581682e-05,
      "loss": 0.9964,
      "num_input_tokens_seen": 17504384,
      "step": 30180
    },
    {
      "epoch": 4.495829609770628,
      "grad_norm": 16.5,
      "learning_rate": 4.766565317293683e-05,
      "loss": 1.3203,
      "num_input_tokens_seen": 17507136,
      "step": 30185
    },
    {
      "epoch": 4.496574322311588,
      "grad_norm": 6.4375,
      "learning_rate": 4.766428193714367e-05,
      "loss": 1.0876,
      "num_input_tokens_seen": 17510208,
      "step": 30190
    },
    {
      "epoch": 4.497319034852547,
      "grad_norm": 4.90625,
      "learning_rate": 4.766291031846051e-05,
      "loss": 1.4156,
      "num_input_tokens_seen": 17512960,
      "step": 30195
    },
    {
      "epoch": 4.498063747393506,
      "grad_norm": 7.625,
      "learning_rate": 4.76615383169105e-05,
      "loss": 0.9458,
      "num_input_tokens_seen": 17515840,
      "step": 30200
    },
    {
      "epoch": 4.498808459934465,
      "grad_norm": 16.125,
      "learning_rate": 4.766016593251684e-05,
      "loss": 1.2716,
      "num_input_tokens_seen": 17518592,
      "step": 30205
    },
    {
      "epoch": 4.499553172475425,
      "grad_norm": 8.6875,
      "learning_rate": 4.765879316530272e-05,
      "loss": 0.882,
      "num_input_tokens_seen": 17521824,
      "step": 30210
    },
    {
      "epoch": 4.5002978850163835,
      "grad_norm": 10.9375,
      "learning_rate": 4.76574200152913e-05,
      "loss": 1.0775,
      "num_input_tokens_seen": 17524832,
      "step": 30215
    },
    {
      "epoch": 4.501042597557343,
      "grad_norm": 7.78125,
      "learning_rate": 4.76560464825058e-05,
      "loss": 1.1663,
      "num_input_tokens_seen": 17527968,
      "step": 30220
    },
    {
      "epoch": 4.501787310098302,
      "grad_norm": 10.5,
      "learning_rate": 4.7654672566969424e-05,
      "loss": 0.8923,
      "num_input_tokens_seen": 17531296,
      "step": 30225
    },
    {
      "epoch": 4.5025320226392616,
      "grad_norm": 14.5625,
      "learning_rate": 4.765329826870538e-05,
      "loss": 1.1281,
      "num_input_tokens_seen": 17534336,
      "step": 30230
    },
    {
      "epoch": 4.50327673518022,
      "grad_norm": 12.5,
      "learning_rate": 4.765192358773689e-05,
      "loss": 1.2897,
      "num_input_tokens_seen": 17538464,
      "step": 30235
    },
    {
      "epoch": 4.50402144772118,
      "grad_norm": 13.6875,
      "learning_rate": 4.765054852408717e-05,
      "loss": 1.0318,
      "num_input_tokens_seen": 17541568,
      "step": 30240
    },
    {
      "epoch": 4.504766160262139,
      "grad_norm": 8.4375,
      "learning_rate": 4.7649173077779455e-05,
      "loss": 1.1556,
      "num_input_tokens_seen": 17544352,
      "step": 30245
    },
    {
      "epoch": 4.505510872803098,
      "grad_norm": 10.5,
      "learning_rate": 4.7647797248836975e-05,
      "loss": 1.048,
      "num_input_tokens_seen": 17547456,
      "step": 30250
    },
    {
      "epoch": 4.506255585344057,
      "grad_norm": 12.6875,
      "learning_rate": 4.7646421037282984e-05,
      "loss": 1.0484,
      "num_input_tokens_seen": 17550496,
      "step": 30255
    },
    {
      "epoch": 4.507000297885017,
      "grad_norm": 26.0,
      "learning_rate": 4.764504444314072e-05,
      "loss": 1.4433,
      "num_input_tokens_seen": 17553504,
      "step": 30260
    },
    {
      "epoch": 4.5077450104259755,
      "grad_norm": 16.375,
      "learning_rate": 4.7643667466433453e-05,
      "loss": 1.3894,
      "num_input_tokens_seen": 17556672,
      "step": 30265
    },
    {
      "epoch": 4.508489722966935,
      "grad_norm": 7.9375,
      "learning_rate": 4.7642290107184426e-05,
      "loss": 0.8397,
      "num_input_tokens_seen": 17559456,
      "step": 30270
    },
    {
      "epoch": 4.509234435507894,
      "grad_norm": 12.375,
      "learning_rate": 4.764091236541693e-05,
      "loss": 1.0705,
      "num_input_tokens_seen": 17562432,
      "step": 30275
    },
    {
      "epoch": 4.509979148048853,
      "grad_norm": 18.0,
      "learning_rate": 4.763953424115424e-05,
      "loss": 1.1566,
      "num_input_tokens_seen": 17565248,
      "step": 30280
    },
    {
      "epoch": 4.510723860589812,
      "grad_norm": 5.25,
      "learning_rate": 4.7638155734419616e-05,
      "loss": 0.9962,
      "num_input_tokens_seen": 17568032,
      "step": 30285
    },
    {
      "epoch": 4.511468573130772,
      "grad_norm": 22.5,
      "learning_rate": 4.763677684523636e-05,
      "loss": 1.1831,
      "num_input_tokens_seen": 17570912,
      "step": 30290
    },
    {
      "epoch": 4.512213285671731,
      "grad_norm": 9.0625,
      "learning_rate": 4.7635397573627774e-05,
      "loss": 1.0593,
      "num_input_tokens_seen": 17573728,
      "step": 30295
    },
    {
      "epoch": 4.5129579982126895,
      "grad_norm": 20.5,
      "learning_rate": 4.7634017919617143e-05,
      "loss": 1.1805,
      "num_input_tokens_seen": 17576800,
      "step": 30300
    },
    {
      "epoch": 4.513702710753649,
      "grad_norm": 9.3125,
      "learning_rate": 4.763263788322778e-05,
      "loss": 1.2141,
      "num_input_tokens_seen": 17580000,
      "step": 30305
    },
    {
      "epoch": 4.514447423294608,
      "grad_norm": 17.5,
      "learning_rate": 4.7631257464483014e-05,
      "loss": 0.912,
      "num_input_tokens_seen": 17583104,
      "step": 30310
    },
    {
      "epoch": 4.5151921358355676,
      "grad_norm": 13.0625,
      "learning_rate": 4.762987666340615e-05,
      "loss": 1.2272,
      "num_input_tokens_seen": 17586080,
      "step": 30315
    },
    {
      "epoch": 4.515936848376526,
      "grad_norm": 12.875,
      "learning_rate": 4.7628495480020516e-05,
      "loss": 1.0542,
      "num_input_tokens_seen": 17588992,
      "step": 30320
    },
    {
      "epoch": 4.516681560917486,
      "grad_norm": 13.1875,
      "learning_rate": 4.762711391434945e-05,
      "loss": 1.2487,
      "num_input_tokens_seen": 17591936,
      "step": 30325
    },
    {
      "epoch": 4.517426273458445,
      "grad_norm": 8.9375,
      "learning_rate": 4.76257319664163e-05,
      "loss": 1.1112,
      "num_input_tokens_seen": 17594496,
      "step": 30330
    },
    {
      "epoch": 4.518170985999404,
      "grad_norm": 19.25,
      "learning_rate": 4.76243496362444e-05,
      "loss": 1.2848,
      "num_input_tokens_seen": 17597216,
      "step": 30335
    },
    {
      "epoch": 4.518915698540363,
      "grad_norm": 11.25,
      "learning_rate": 4.76229669238571e-05,
      "loss": 1.067,
      "num_input_tokens_seen": 17599840,
      "step": 30340
    },
    {
      "epoch": 4.519660411081323,
      "grad_norm": 16.0,
      "learning_rate": 4.762158382927777e-05,
      "loss": 1.2509,
      "num_input_tokens_seen": 17602464,
      "step": 30345
    },
    {
      "epoch": 4.5204051236222815,
      "grad_norm": 22.25,
      "learning_rate": 4.762020035252978e-05,
      "loss": 1.4944,
      "num_input_tokens_seen": 17605408,
      "step": 30350
    },
    {
      "epoch": 4.521149836163241,
      "grad_norm": 5.40625,
      "learning_rate": 4.761881649363649e-05,
      "loss": 0.8268,
      "num_input_tokens_seen": 17608288,
      "step": 30355
    },
    {
      "epoch": 4.5218945487042,
      "grad_norm": 10.625,
      "learning_rate": 4.7617432252621285e-05,
      "loss": 1.2715,
      "num_input_tokens_seen": 17611232,
      "step": 30360
    },
    {
      "epoch": 4.52263926124516,
      "grad_norm": 9.5,
      "learning_rate": 4.7616047629507556e-05,
      "loss": 0.9765,
      "num_input_tokens_seen": 17614080,
      "step": 30365
    },
    {
      "epoch": 4.523383973786118,
      "grad_norm": 15.5,
      "learning_rate": 4.761466262431867e-05,
      "loss": 0.9312,
      "num_input_tokens_seen": 17616896,
      "step": 30370
    },
    {
      "epoch": 4.524128686327078,
      "grad_norm": 12.125,
      "learning_rate": 4.7613277237078055e-05,
      "loss": 1.1165,
      "num_input_tokens_seen": 17619840,
      "step": 30375
    },
    {
      "epoch": 4.524873398868037,
      "grad_norm": 15.8125,
      "learning_rate": 4.761189146780911e-05,
      "loss": 0.9122,
      "num_input_tokens_seen": 17622496,
      "step": 30380
    },
    {
      "epoch": 4.525618111408996,
      "grad_norm": 7.75,
      "learning_rate": 4.761050531653524e-05,
      "loss": 1.1212,
      "num_input_tokens_seen": 17625632,
      "step": 30385
    },
    {
      "epoch": 4.526362823949955,
      "grad_norm": 14.375,
      "learning_rate": 4.760911878327985e-05,
      "loss": 1.0375,
      "num_input_tokens_seen": 17628800,
      "step": 30390
    },
    {
      "epoch": 4.527107536490915,
      "grad_norm": 13.0625,
      "learning_rate": 4.760773186806639e-05,
      "loss": 1.1985,
      "num_input_tokens_seen": 17631872,
      "step": 30395
    },
    {
      "epoch": 4.5278522490318736,
      "grad_norm": 11.125,
      "learning_rate": 4.7606344570918264e-05,
      "loss": 1.1753,
      "num_input_tokens_seen": 17634656,
      "step": 30400
    },
    {
      "epoch": 4.528596961572833,
      "grad_norm": 7.96875,
      "learning_rate": 4.760495689185893e-05,
      "loss": 0.8012,
      "num_input_tokens_seen": 17637632,
      "step": 30405
    },
    {
      "epoch": 4.529341674113792,
      "grad_norm": 11.8125,
      "learning_rate": 4.760356883091183e-05,
      "loss": 0.8902,
      "num_input_tokens_seen": 17640384,
      "step": 30410
    },
    {
      "epoch": 4.530086386654752,
      "grad_norm": 9.25,
      "learning_rate": 4.7602180388100395e-05,
      "loss": 1.2312,
      "num_input_tokens_seen": 17643200,
      "step": 30415
    },
    {
      "epoch": 4.53083109919571,
      "grad_norm": 17.25,
      "learning_rate": 4.760079156344811e-05,
      "loss": 1.1476,
      "num_input_tokens_seen": 17646144,
      "step": 30420
    },
    {
      "epoch": 4.53157581173667,
      "grad_norm": 4.125,
      "learning_rate": 4.7599402356978406e-05,
      "loss": 0.8812,
      "num_input_tokens_seen": 17648992,
      "step": 30425
    },
    {
      "epoch": 4.532320524277629,
      "grad_norm": 7.125,
      "learning_rate": 4.759801276871478e-05,
      "loss": 0.9914,
      "num_input_tokens_seen": 17651968,
      "step": 30430
    },
    {
      "epoch": 4.533065236818588,
      "grad_norm": 5.84375,
      "learning_rate": 4.759662279868069e-05,
      "loss": 1.1017,
      "num_input_tokens_seen": 17655136,
      "step": 30435
    },
    {
      "epoch": 4.533809949359547,
      "grad_norm": 6.875,
      "learning_rate": 4.759523244689963e-05,
      "loss": 1.432,
      "num_input_tokens_seen": 17658112,
      "step": 30440
    },
    {
      "epoch": 4.534554661900506,
      "grad_norm": 8.3125,
      "learning_rate": 4.759384171339507e-05,
      "loss": 0.8961,
      "num_input_tokens_seen": 17661120,
      "step": 30445
    },
    {
      "epoch": 4.535299374441466,
      "grad_norm": 12.125,
      "learning_rate": 4.759245059819053e-05,
      "loss": 1.6184,
      "num_input_tokens_seen": 17663712,
      "step": 30450
    },
    {
      "epoch": 4.536044086982425,
      "grad_norm": 13.375,
      "learning_rate": 4.759105910130949e-05,
      "loss": 1.0544,
      "num_input_tokens_seen": 17666656,
      "step": 30455
    },
    {
      "epoch": 4.536788799523384,
      "grad_norm": 10.5625,
      "learning_rate": 4.758966722277547e-05,
      "loss": 0.9834,
      "num_input_tokens_seen": 17669568,
      "step": 30460
    },
    {
      "epoch": 4.537533512064343,
      "grad_norm": 21.125,
      "learning_rate": 4.758827496261199e-05,
      "loss": 1.4828,
      "num_input_tokens_seen": 17672672,
      "step": 30465
    },
    {
      "epoch": 4.538278224605302,
      "grad_norm": 14.6875,
      "learning_rate": 4.758688232084255e-05,
      "loss": 1.2686,
      "num_input_tokens_seen": 17676320,
      "step": 30470
    },
    {
      "epoch": 4.539022937146262,
      "grad_norm": 11.5625,
      "learning_rate": 4.7585489297490694e-05,
      "loss": 1.3765,
      "num_input_tokens_seen": 17679520,
      "step": 30475
    },
    {
      "epoch": 4.539767649687221,
      "grad_norm": 12.25,
      "learning_rate": 4.758409589257995e-05,
      "loss": 1.1626,
      "num_input_tokens_seen": 17682240,
      "step": 30480
    },
    {
      "epoch": 4.5405123622281796,
      "grad_norm": 10.3125,
      "learning_rate": 4.758270210613387e-05,
      "loss": 1.2873,
      "num_input_tokens_seen": 17685536,
      "step": 30485
    },
    {
      "epoch": 4.541257074769139,
      "grad_norm": 11.0625,
      "learning_rate": 4.758130793817598e-05,
      "loss": 1.0215,
      "num_input_tokens_seen": 17688416,
      "step": 30490
    },
    {
      "epoch": 4.542001787310098,
      "grad_norm": 7.46875,
      "learning_rate": 4.7579913388729844e-05,
      "loss": 0.9318,
      "num_input_tokens_seen": 17691328,
      "step": 30495
    },
    {
      "epoch": 4.542746499851058,
      "grad_norm": 5.90625,
      "learning_rate": 4.757851845781902e-05,
      "loss": 1.0688,
      "num_input_tokens_seen": 17693920,
      "step": 30500
    },
    {
      "epoch": 4.543491212392016,
      "grad_norm": 6.90625,
      "learning_rate": 4.757712314546707e-05,
      "loss": 0.839,
      "num_input_tokens_seen": 17696960,
      "step": 30505
    },
    {
      "epoch": 4.544235924932976,
      "grad_norm": 5.90625,
      "learning_rate": 4.7575727451697585e-05,
      "loss": 1.11,
      "num_input_tokens_seen": 17699840,
      "step": 30510
    },
    {
      "epoch": 4.544980637473935,
      "grad_norm": 19.25,
      "learning_rate": 4.757433137653411e-05,
      "loss": 1.5959,
      "num_input_tokens_seen": 17702720,
      "step": 30515
    },
    {
      "epoch": 4.545725350014894,
      "grad_norm": 8.375,
      "learning_rate": 4.757293492000027e-05,
      "loss": 1.1108,
      "num_input_tokens_seen": 17705888,
      "step": 30520
    },
    {
      "epoch": 4.546470062555853,
      "grad_norm": 6.53125,
      "learning_rate": 4.757153808211962e-05,
      "loss": 1.2939,
      "num_input_tokens_seen": 17708672,
      "step": 30525
    },
    {
      "epoch": 4.547214775096813,
      "grad_norm": 8.25,
      "learning_rate": 4.757014086291579e-05,
      "loss": 1.1074,
      "num_input_tokens_seen": 17711776,
      "step": 30530
    },
    {
      "epoch": 4.547959487637772,
      "grad_norm": 10.6875,
      "learning_rate": 4.7568743262412354e-05,
      "loss": 1.1164,
      "num_input_tokens_seen": 17714688,
      "step": 30535
    },
    {
      "epoch": 4.548704200178731,
      "grad_norm": 6.25,
      "learning_rate": 4.756734528063295e-05,
      "loss": 0.9907,
      "num_input_tokens_seen": 17717504,
      "step": 30540
    },
    {
      "epoch": 4.54944891271969,
      "grad_norm": 12.875,
      "learning_rate": 4.756594691760118e-05,
      "loss": 1.1625,
      "num_input_tokens_seen": 17720224,
      "step": 30545
    },
    {
      "epoch": 4.55019362526065,
      "grad_norm": 7.375,
      "learning_rate": 4.7564548173340664e-05,
      "loss": 0.9844,
      "num_input_tokens_seen": 17723168,
      "step": 30550
    },
    {
      "epoch": 4.550938337801608,
      "grad_norm": 20.0,
      "learning_rate": 4.7563149047875054e-05,
      "loss": 0.9906,
      "num_input_tokens_seen": 17725792,
      "step": 30555
    },
    {
      "epoch": 4.551683050342568,
      "grad_norm": 22.375,
      "learning_rate": 4.756174954122796e-05,
      "loss": 1.177,
      "num_input_tokens_seen": 17728832,
      "step": 30560
    },
    {
      "epoch": 4.552427762883527,
      "grad_norm": 5.125,
      "learning_rate": 4.7560349653423055e-05,
      "loss": 0.882,
      "num_input_tokens_seen": 17731584,
      "step": 30565
    },
    {
      "epoch": 4.553172475424486,
      "grad_norm": 12.0625,
      "learning_rate": 4.755894938448395e-05,
      "loss": 1.3166,
      "num_input_tokens_seen": 17734592,
      "step": 30570
    },
    {
      "epoch": 4.553917187965445,
      "grad_norm": 8.0,
      "learning_rate": 4.755754873443434e-05,
      "loss": 1.1067,
      "num_input_tokens_seen": 17737600,
      "step": 30575
    },
    {
      "epoch": 4.554661900506405,
      "grad_norm": 3.96875,
      "learning_rate": 4.7556147703297865e-05,
      "loss": 0.906,
      "num_input_tokens_seen": 17740384,
      "step": 30580
    },
    {
      "epoch": 4.555406613047364,
      "grad_norm": 7.0,
      "learning_rate": 4.75547462910982e-05,
      "loss": 1.0953,
      "num_input_tokens_seen": 17743296,
      "step": 30585
    },
    {
      "epoch": 4.556151325588323,
      "grad_norm": 6.03125,
      "learning_rate": 4.755334449785902e-05,
      "loss": 0.9983,
      "num_input_tokens_seen": 17746048,
      "step": 30590
    },
    {
      "epoch": 4.556896038129282,
      "grad_norm": 10.4375,
      "learning_rate": 4.755194232360401e-05,
      "loss": 1.2636,
      "num_input_tokens_seen": 17749184,
      "step": 30595
    },
    {
      "epoch": 4.557640750670242,
      "grad_norm": 4.96875,
      "learning_rate": 4.755053976835685e-05,
      "loss": 1.2473,
      "num_input_tokens_seen": 17752256,
      "step": 30600
    },
    {
      "epoch": 4.5583854632112,
      "grad_norm": 6.875,
      "learning_rate": 4.754913683214124e-05,
      "loss": 0.7985,
      "num_input_tokens_seen": 17755136,
      "step": 30605
    },
    {
      "epoch": 4.559130175752159,
      "grad_norm": 14.5625,
      "learning_rate": 4.754773351498088e-05,
      "loss": 1.0405,
      "num_input_tokens_seen": 17758048,
      "step": 30610
    },
    {
      "epoch": 4.559874888293119,
      "grad_norm": 12.5,
      "learning_rate": 4.754632981689949e-05,
      "loss": 0.8841,
      "num_input_tokens_seen": 17760832,
      "step": 30615
    },
    {
      "epoch": 4.5606196008340785,
      "grad_norm": 5.125,
      "learning_rate": 4.7544925737920766e-05,
      "loss": 0.9875,
      "num_input_tokens_seen": 17763648,
      "step": 30620
    },
    {
      "epoch": 4.561364313375037,
      "grad_norm": 12.875,
      "learning_rate": 4.754352127806843e-05,
      "loss": 1.0598,
      "num_input_tokens_seen": 17767008,
      "step": 30625
    },
    {
      "epoch": 4.562109025915996,
      "grad_norm": 11.75,
      "learning_rate": 4.754211643736622e-05,
      "loss": 0.9365,
      "num_input_tokens_seen": 17769728,
      "step": 30630
    },
    {
      "epoch": 4.562853738456956,
      "grad_norm": 13.5625,
      "learning_rate": 4.7540711215837866e-05,
      "loss": 1.1068,
      "num_input_tokens_seen": 17772736,
      "step": 30635
    },
    {
      "epoch": 4.563598450997915,
      "grad_norm": 10.5625,
      "learning_rate": 4.7539305613507096e-05,
      "loss": 1.4587,
      "num_input_tokens_seen": 17775680,
      "step": 30640
    },
    {
      "epoch": 4.564343163538874,
      "grad_norm": 10.6875,
      "learning_rate": 4.753789963039767e-05,
      "loss": 0.8796,
      "num_input_tokens_seen": 17778784,
      "step": 30645
    },
    {
      "epoch": 4.565087876079833,
      "grad_norm": 7.78125,
      "learning_rate": 4.753649326653334e-05,
      "loss": 0.9907,
      "num_input_tokens_seen": 17781632,
      "step": 30650
    },
    {
      "epoch": 4.565832588620792,
      "grad_norm": 9.8125,
      "learning_rate": 4.753508652193785e-05,
      "loss": 1.295,
      "num_input_tokens_seen": 17784576,
      "step": 30655
    },
    {
      "epoch": 4.566577301161751,
      "grad_norm": 9.75,
      "learning_rate": 4.7533679396634986e-05,
      "loss": 1.2091,
      "num_input_tokens_seen": 17787392,
      "step": 30660
    },
    {
      "epoch": 4.567322013702711,
      "grad_norm": 6.5625,
      "learning_rate": 4.7532271890648516e-05,
      "loss": 0.8913,
      "num_input_tokens_seen": 17790208,
      "step": 30665
    },
    {
      "epoch": 4.56806672624367,
      "grad_norm": 4.65625,
      "learning_rate": 4.753086400400221e-05,
      "loss": 1.0372,
      "num_input_tokens_seen": 17793088,
      "step": 30670
    },
    {
      "epoch": 4.568811438784629,
      "grad_norm": 14.8125,
      "learning_rate": 4.752945573671985e-05,
      "loss": 0.9093,
      "num_input_tokens_seen": 17795872,
      "step": 30675
    },
    {
      "epoch": 4.569556151325588,
      "grad_norm": 7.375,
      "learning_rate": 4.752804708882523e-05,
      "loss": 0.9171,
      "num_input_tokens_seen": 17798912,
      "step": 30680
    },
    {
      "epoch": 4.570300863866548,
      "grad_norm": 8.875,
      "learning_rate": 4.7526638060342164e-05,
      "loss": 1.1316,
      "num_input_tokens_seen": 17801696,
      "step": 30685
    },
    {
      "epoch": 4.571045576407506,
      "grad_norm": 7.21875,
      "learning_rate": 4.752522865129444e-05,
      "loss": 1.0749,
      "num_input_tokens_seen": 17804480,
      "step": 30690
    },
    {
      "epoch": 4.571790288948466,
      "grad_norm": 15.0625,
      "learning_rate": 4.7523818861705865e-05,
      "loss": 1.2293,
      "num_input_tokens_seen": 17807616,
      "step": 30695
    },
    {
      "epoch": 4.572535001489425,
      "grad_norm": 9.3125,
      "learning_rate": 4.752240869160026e-05,
      "loss": 0.8644,
      "num_input_tokens_seen": 17810368,
      "step": 30700
    },
    {
      "epoch": 4.5732797140303845,
      "grad_norm": 7.40625,
      "learning_rate": 4.752099814100146e-05,
      "loss": 0.8624,
      "num_input_tokens_seen": 17813568,
      "step": 30705
    },
    {
      "epoch": 4.574024426571343,
      "grad_norm": 10.0625,
      "learning_rate": 4.751958720993328e-05,
      "loss": 1.3132,
      "num_input_tokens_seen": 17816288,
      "step": 30710
    },
    {
      "epoch": 4.574769139112303,
      "grad_norm": 17.125,
      "learning_rate": 4.751817589841957e-05,
      "loss": 0.8564,
      "num_input_tokens_seen": 17819296,
      "step": 30715
    },
    {
      "epoch": 4.575513851653262,
      "grad_norm": 8.0625,
      "learning_rate": 4.7516764206484156e-05,
      "loss": 0.9285,
      "num_input_tokens_seen": 17822016,
      "step": 30720
    },
    {
      "epoch": 4.576258564194221,
      "grad_norm": 7.15625,
      "learning_rate": 4.75153521341509e-05,
      "loss": 0.8668,
      "num_input_tokens_seen": 17824864,
      "step": 30725
    },
    {
      "epoch": 4.57700327673518,
      "grad_norm": 3.984375,
      "learning_rate": 4.751393968144365e-05,
      "loss": 0.7342,
      "num_input_tokens_seen": 17827744,
      "step": 30730
    },
    {
      "epoch": 4.57774798927614,
      "grad_norm": 7.96875,
      "learning_rate": 4.7512526848386276e-05,
      "loss": 1.1811,
      "num_input_tokens_seen": 17830528,
      "step": 30735
    },
    {
      "epoch": 4.578492701817098,
      "grad_norm": 4.625,
      "learning_rate": 4.751111363500263e-05,
      "loss": 1.1933,
      "num_input_tokens_seen": 17833504,
      "step": 30740
    },
    {
      "epoch": 4.579237414358058,
      "grad_norm": 12.0625,
      "learning_rate": 4.750970004131662e-05,
      "loss": 1.0074,
      "num_input_tokens_seen": 17836640,
      "step": 30745
    },
    {
      "epoch": 4.579982126899017,
      "grad_norm": 19.375,
      "learning_rate": 4.7508286067352085e-05,
      "loss": 1.259,
      "num_input_tokens_seen": 17839520,
      "step": 30750
    },
    {
      "epoch": 4.5807268394399765,
      "grad_norm": 13.6875,
      "learning_rate": 4.750687171313294e-05,
      "loss": 0.8028,
      "num_input_tokens_seen": 17842336,
      "step": 30755
    },
    {
      "epoch": 4.581471551980935,
      "grad_norm": 7.59375,
      "learning_rate": 4.750545697868307e-05,
      "loss": 0.9054,
      "num_input_tokens_seen": 17845280,
      "step": 30760
    },
    {
      "epoch": 4.582216264521895,
      "grad_norm": 14.8125,
      "learning_rate": 4.750404186402639e-05,
      "loss": 1.2152,
      "num_input_tokens_seen": 17848032,
      "step": 30765
    },
    {
      "epoch": 4.582960977062854,
      "grad_norm": 13.0,
      "learning_rate": 4.7502626369186784e-05,
      "loss": 1.3069,
      "num_input_tokens_seen": 17850784,
      "step": 30770
    },
    {
      "epoch": 4.583705689603813,
      "grad_norm": 6.3125,
      "learning_rate": 4.750121049418817e-05,
      "loss": 1.0322,
      "num_input_tokens_seen": 17853760,
      "step": 30775
    },
    {
      "epoch": 4.584450402144772,
      "grad_norm": 5.15625,
      "learning_rate": 4.749979423905449e-05,
      "loss": 0.9579,
      "num_input_tokens_seen": 17856640,
      "step": 30780
    },
    {
      "epoch": 4.585195114685732,
      "grad_norm": 9.0625,
      "learning_rate": 4.749837760380965e-05,
      "loss": 1.041,
      "num_input_tokens_seen": 17859520,
      "step": 30785
    },
    {
      "epoch": 4.5859398272266905,
      "grad_norm": 18.75,
      "learning_rate": 4.749696058847758e-05,
      "loss": 1.1158,
      "num_input_tokens_seen": 17862400,
      "step": 30790
    },
    {
      "epoch": 4.586684539767649,
      "grad_norm": 6.96875,
      "learning_rate": 4.749554319308223e-05,
      "loss": 0.7539,
      "num_input_tokens_seen": 17865152,
      "step": 30795
    },
    {
      "epoch": 4.587429252308609,
      "grad_norm": 5.53125,
      "learning_rate": 4.7494125417647536e-05,
      "loss": 0.8736,
      "num_input_tokens_seen": 17868160,
      "step": 30800
    },
    {
      "epoch": 4.5881739648495685,
      "grad_norm": 8.3125,
      "learning_rate": 4.749270726219746e-05,
      "loss": 1.0883,
      "num_input_tokens_seen": 17871232,
      "step": 30805
    },
    {
      "epoch": 4.588918677390527,
      "grad_norm": 6.71875,
      "learning_rate": 4.7491288726755954e-05,
      "loss": 0.7451,
      "num_input_tokens_seen": 17874208,
      "step": 30810
    },
    {
      "epoch": 4.589663389931486,
      "grad_norm": 9.75,
      "learning_rate": 4.7489869811346984e-05,
      "loss": 0.8734,
      "num_input_tokens_seen": 17877216,
      "step": 30815
    },
    {
      "epoch": 4.590408102472446,
      "grad_norm": 16.625,
      "learning_rate": 4.748845051599452e-05,
      "loss": 1.1486,
      "num_input_tokens_seen": 17880320,
      "step": 30820
    },
    {
      "epoch": 4.591152815013404,
      "grad_norm": 12.1875,
      "learning_rate": 4.748703084072255e-05,
      "loss": 1.1038,
      "num_input_tokens_seen": 17883264,
      "step": 30825
    },
    {
      "epoch": 4.591897527554364,
      "grad_norm": 8.0625,
      "learning_rate": 4.748561078555504e-05,
      "loss": 0.9439,
      "num_input_tokens_seen": 17886496,
      "step": 30830
    },
    {
      "epoch": 4.592642240095323,
      "grad_norm": 9.5,
      "learning_rate": 4.748419035051599e-05,
      "loss": 0.9859,
      "num_input_tokens_seen": 17889344,
      "step": 30835
    },
    {
      "epoch": 4.5933869526362825,
      "grad_norm": 17.625,
      "learning_rate": 4.748276953562939e-05,
      "loss": 1.0197,
      "num_input_tokens_seen": 17892352,
      "step": 30840
    },
    {
      "epoch": 4.594131665177241,
      "grad_norm": 6.625,
      "learning_rate": 4.7481348340919255e-05,
      "loss": 1.2693,
      "num_input_tokens_seen": 17895424,
      "step": 30845
    },
    {
      "epoch": 4.594876377718201,
      "grad_norm": 10.4375,
      "learning_rate": 4.747992676640959e-05,
      "loss": 1.2473,
      "num_input_tokens_seen": 17897984,
      "step": 30850
    },
    {
      "epoch": 4.59562109025916,
      "grad_norm": 15.0625,
      "learning_rate": 4.7478504812124416e-05,
      "loss": 1.0191,
      "num_input_tokens_seen": 17900960,
      "step": 30855
    },
    {
      "epoch": 4.596365802800119,
      "grad_norm": 11.625,
      "learning_rate": 4.7477082478087734e-05,
      "loss": 1.1005,
      "num_input_tokens_seen": 17903776,
      "step": 30860
    },
    {
      "epoch": 4.597110515341078,
      "grad_norm": 13.625,
      "learning_rate": 4.74756597643236e-05,
      "loss": 0.8158,
      "num_input_tokens_seen": 17906368,
      "step": 30865
    },
    {
      "epoch": 4.597855227882038,
      "grad_norm": 12.75,
      "learning_rate": 4.747423667085603e-05,
      "loss": 1.0216,
      "num_input_tokens_seen": 17909504,
      "step": 30870
    },
    {
      "epoch": 4.5985999404229965,
      "grad_norm": 11.0,
      "learning_rate": 4.7472813197709084e-05,
      "loss": 0.8364,
      "num_input_tokens_seen": 17912736,
      "step": 30875
    },
    {
      "epoch": 4.599344652963956,
      "grad_norm": 12.625,
      "learning_rate": 4.747138934490679e-05,
      "loss": 1.2461,
      "num_input_tokens_seen": 17915552,
      "step": 30880
    },
    {
      "epoch": 4.600089365504915,
      "grad_norm": 15.75,
      "learning_rate": 4.746996511247321e-05,
      "loss": 1.0792,
      "num_input_tokens_seen": 17918336,
      "step": 30885
    },
    {
      "epoch": 4.6008340780458745,
      "grad_norm": 10.3125,
      "learning_rate": 4.746854050043241e-05,
      "loss": 0.972,
      "num_input_tokens_seen": 17921312,
      "step": 30890
    },
    {
      "epoch": 4.601578790586833,
      "grad_norm": 7.1875,
      "learning_rate": 4.7467115508808456e-05,
      "loss": 1.0593,
      "num_input_tokens_seen": 17924160,
      "step": 30895
    },
    {
      "epoch": 4.602323503127793,
      "grad_norm": 11.125,
      "learning_rate": 4.746569013762543e-05,
      "loss": 1.1256,
      "num_input_tokens_seen": 17927008,
      "step": 30900
    },
    {
      "epoch": 4.603068215668752,
      "grad_norm": 13.5,
      "learning_rate": 4.7464264386907385e-05,
      "loss": 1.3043,
      "num_input_tokens_seen": 17929824,
      "step": 30905
    },
    {
      "epoch": 4.603812928209711,
      "grad_norm": 11.4375,
      "learning_rate": 4.746283825667843e-05,
      "loss": 0.7922,
      "num_input_tokens_seen": 17932704,
      "step": 30910
    },
    {
      "epoch": 4.60455764075067,
      "grad_norm": 8.6875,
      "learning_rate": 4.746141174696266e-05,
      "loss": 0.8182,
      "num_input_tokens_seen": 17935552,
      "step": 30915
    },
    {
      "epoch": 4.60530235329163,
      "grad_norm": 17.125,
      "learning_rate": 4.745998485778416e-05,
      "loss": 0.9096,
      "num_input_tokens_seen": 17938720,
      "step": 30920
    },
    {
      "epoch": 4.6060470658325885,
      "grad_norm": 7.9375,
      "learning_rate": 4.7458557589167044e-05,
      "loss": 0.9952,
      "num_input_tokens_seen": 17941408,
      "step": 30925
    },
    {
      "epoch": 4.606791778373548,
      "grad_norm": 8.75,
      "learning_rate": 4.7457129941135424e-05,
      "loss": 1.191,
      "num_input_tokens_seen": 17944256,
      "step": 30930
    },
    {
      "epoch": 4.607536490914507,
      "grad_norm": 11.9375,
      "learning_rate": 4.7455701913713424e-05,
      "loss": 1.1874,
      "num_input_tokens_seen": 17946880,
      "step": 30935
    },
    {
      "epoch": 4.6082812034554665,
      "grad_norm": 16.5,
      "learning_rate": 4.745427350692515e-05,
      "loss": 0.8885,
      "num_input_tokens_seen": 17949920,
      "step": 30940
    },
    {
      "epoch": 4.609025915996425,
      "grad_norm": 7.0625,
      "learning_rate": 4.7452844720794756e-05,
      "loss": 0.8803,
      "num_input_tokens_seen": 17952832,
      "step": 30945
    },
    {
      "epoch": 4.609770628537385,
      "grad_norm": 4.78125,
      "learning_rate": 4.745141555534637e-05,
      "loss": 1.0137,
      "num_input_tokens_seen": 17956288,
      "step": 30950
    },
    {
      "epoch": 4.610515341078344,
      "grad_norm": 7.0,
      "learning_rate": 4.744998601060414e-05,
      "loss": 1.0344,
      "num_input_tokens_seen": 17959040,
      "step": 30955
    },
    {
      "epoch": 4.6112600536193025,
      "grad_norm": 17.625,
      "learning_rate": 4.74485560865922e-05,
      "loss": 1.0795,
      "num_input_tokens_seen": 17961984,
      "step": 30960
    },
    {
      "epoch": 4.612004766160262,
      "grad_norm": 9.75,
      "learning_rate": 4.744712578333473e-05,
      "loss": 1.1649,
      "num_input_tokens_seen": 17964704,
      "step": 30965
    },
    {
      "epoch": 4.612749478701222,
      "grad_norm": 8.5625,
      "learning_rate": 4.744569510085589e-05,
      "loss": 0.6906,
      "num_input_tokens_seen": 17967296,
      "step": 30970
    },
    {
      "epoch": 4.6134941912421805,
      "grad_norm": 9.625,
      "learning_rate": 4.7444264039179845e-05,
      "loss": 1.1329,
      "num_input_tokens_seen": 17970272,
      "step": 30975
    },
    {
      "epoch": 4.614238903783139,
      "grad_norm": 14.4375,
      "learning_rate": 4.744283259833076e-05,
      "loss": 0.9067,
      "num_input_tokens_seen": 17973184,
      "step": 30980
    },
    {
      "epoch": 4.614983616324099,
      "grad_norm": 17.25,
      "learning_rate": 4.744140077833283e-05,
      "loss": 1.099,
      "num_input_tokens_seen": 17975936,
      "step": 30985
    },
    {
      "epoch": 4.615728328865059,
      "grad_norm": 11.5625,
      "learning_rate": 4.743996857921024e-05,
      "loss": 0.9031,
      "num_input_tokens_seen": 17978720,
      "step": 30990
    },
    {
      "epoch": 4.616473041406017,
      "grad_norm": 10.0,
      "learning_rate": 4.7438536000987195e-05,
      "loss": 1.003,
      "num_input_tokens_seen": 17981536,
      "step": 30995
    },
    {
      "epoch": 4.617217753946976,
      "grad_norm": 20.125,
      "learning_rate": 4.743710304368788e-05,
      "loss": 0.9728,
      "num_input_tokens_seen": 17984320,
      "step": 31000
    },
    {
      "epoch": 4.617962466487936,
      "grad_norm": 17.375,
      "learning_rate": 4.743566970733652e-05,
      "loss": 0.9825,
      "num_input_tokens_seen": 17987200,
      "step": 31005
    },
    {
      "epoch": 4.6187071790288945,
      "grad_norm": 6.84375,
      "learning_rate": 4.7434235991957326e-05,
      "loss": 0.7149,
      "num_input_tokens_seen": 17990304,
      "step": 31010
    },
    {
      "epoch": 4.619451891569854,
      "grad_norm": 6.65625,
      "learning_rate": 4.743280189757451e-05,
      "loss": 1.1371,
      "num_input_tokens_seen": 17993088,
      "step": 31015
    },
    {
      "epoch": 4.620196604110813,
      "grad_norm": 11.4375,
      "learning_rate": 4.7431367424212305e-05,
      "loss": 0.8307,
      "num_input_tokens_seen": 17995648,
      "step": 31020
    },
    {
      "epoch": 4.6209413166517725,
      "grad_norm": 11.0625,
      "learning_rate": 4.7429932571894954e-05,
      "loss": 1.0523,
      "num_input_tokens_seen": 17998560,
      "step": 31025
    },
    {
      "epoch": 4.621686029192731,
      "grad_norm": 6.71875,
      "learning_rate": 4.742849734064668e-05,
      "loss": 0.556,
      "num_input_tokens_seen": 18001504,
      "step": 31030
    },
    {
      "epoch": 4.622430741733691,
      "grad_norm": 7.21875,
      "learning_rate": 4.742706173049174e-05,
      "loss": 1.2524,
      "num_input_tokens_seen": 18004608,
      "step": 31035
    },
    {
      "epoch": 4.62317545427465,
      "grad_norm": 11.6875,
      "learning_rate": 4.7425625741454394e-05,
      "loss": 0.9861,
      "num_input_tokens_seen": 18007584,
      "step": 31040
    },
    {
      "epoch": 4.623920166815609,
      "grad_norm": 11.875,
      "learning_rate": 4.7424189373558886e-05,
      "loss": 1.2417,
      "num_input_tokens_seen": 18010496,
      "step": 31045
    },
    {
      "epoch": 4.624664879356568,
      "grad_norm": 15.375,
      "learning_rate": 4.742275262682949e-05,
      "loss": 1.0448,
      "num_input_tokens_seen": 18013216,
      "step": 31050
    },
    {
      "epoch": 4.625409591897528,
      "grad_norm": 8.8125,
      "learning_rate": 4.7421315501290484e-05,
      "loss": 0.9953,
      "num_input_tokens_seen": 18016000,
      "step": 31055
    },
    {
      "epoch": 4.6261543044384865,
      "grad_norm": 15.6875,
      "learning_rate": 4.7419877996966134e-05,
      "loss": 1.2401,
      "num_input_tokens_seen": 18019040,
      "step": 31060
    },
    {
      "epoch": 4.626899016979446,
      "grad_norm": 12.0,
      "learning_rate": 4.7418440113880736e-05,
      "loss": 1.201,
      "num_input_tokens_seen": 18022240,
      "step": 31065
    },
    {
      "epoch": 4.627643729520405,
      "grad_norm": 8.5,
      "learning_rate": 4.7417001852058576e-05,
      "loss": 0.9278,
      "num_input_tokens_seen": 18025312,
      "step": 31070
    },
    {
      "epoch": 4.628388442061365,
      "grad_norm": 13.5,
      "learning_rate": 4.741556321152395e-05,
      "loss": 0.8505,
      "num_input_tokens_seen": 18028096,
      "step": 31075
    },
    {
      "epoch": 4.629133154602323,
      "grad_norm": 6.96875,
      "learning_rate": 4.741412419230117e-05,
      "loss": 1.1887,
      "num_input_tokens_seen": 18031040,
      "step": 31080
    },
    {
      "epoch": 4.629877867143283,
      "grad_norm": 7.40625,
      "learning_rate": 4.741268479441453e-05,
      "loss": 1.1178,
      "num_input_tokens_seen": 18034304,
      "step": 31085
    },
    {
      "epoch": 4.630622579684242,
      "grad_norm": 13.3125,
      "learning_rate": 4.7411245017888374e-05,
      "loss": 0.8993,
      "num_input_tokens_seen": 18036960,
      "step": 31090
    },
    {
      "epoch": 4.631367292225201,
      "grad_norm": 11.0625,
      "learning_rate": 4.7409804862747007e-05,
      "loss": 0.8549,
      "num_input_tokens_seen": 18039808,
      "step": 31095
    },
    {
      "epoch": 4.63211200476616,
      "grad_norm": 6.625,
      "learning_rate": 4.740836432901476e-05,
      "loss": 0.6982,
      "num_input_tokens_seen": 18042208,
      "step": 31100
    },
    {
      "epoch": 4.63285671730712,
      "grad_norm": 13.8125,
      "learning_rate": 4.7406923416715976e-05,
      "loss": 0.9212,
      "num_input_tokens_seen": 18045024,
      "step": 31105
    },
    {
      "epoch": 4.6336014298480785,
      "grad_norm": 15.0625,
      "learning_rate": 4.7405482125875e-05,
      "loss": 1.5501,
      "num_input_tokens_seen": 18047936,
      "step": 31110
    },
    {
      "epoch": 4.634346142389038,
      "grad_norm": 11.125,
      "learning_rate": 4.7404040456516164e-05,
      "loss": 0.9639,
      "num_input_tokens_seen": 18050688,
      "step": 31115
    },
    {
      "epoch": 4.635090854929997,
      "grad_norm": 9.875,
      "learning_rate": 4.7402598408663846e-05,
      "loss": 1.1676,
      "num_input_tokens_seen": 18054144,
      "step": 31120
    },
    {
      "epoch": 4.635835567470957,
      "grad_norm": 6.28125,
      "learning_rate": 4.740115598234239e-05,
      "loss": 1.2484,
      "num_input_tokens_seen": 18056928,
      "step": 31125
    },
    {
      "epoch": 4.636580280011915,
      "grad_norm": 12.0,
      "learning_rate": 4.739971317757617e-05,
      "loss": 0.9412,
      "num_input_tokens_seen": 18059520,
      "step": 31130
    },
    {
      "epoch": 4.637324992552875,
      "grad_norm": 6.4375,
      "learning_rate": 4.7398269994389567e-05,
      "loss": 1.0756,
      "num_input_tokens_seen": 18062400,
      "step": 31135
    },
    {
      "epoch": 4.638069705093834,
      "grad_norm": 17.125,
      "learning_rate": 4.739682643280695e-05,
      "loss": 1.0418,
      "num_input_tokens_seen": 18065120,
      "step": 31140
    },
    {
      "epoch": 4.6388144176347925,
      "grad_norm": 3.015625,
      "learning_rate": 4.739538249285272e-05,
      "loss": 0.7486,
      "num_input_tokens_seen": 18068000,
      "step": 31145
    },
    {
      "epoch": 4.639559130175752,
      "grad_norm": 4.96875,
      "learning_rate": 4.739393817455127e-05,
      "loss": 1.3153,
      "num_input_tokens_seen": 18070720,
      "step": 31150
    },
    {
      "epoch": 4.640303842716712,
      "grad_norm": 7.65625,
      "learning_rate": 4.739249347792698e-05,
      "loss": 1.3004,
      "num_input_tokens_seen": 18073600,
      "step": 31155
    },
    {
      "epoch": 4.641048555257671,
      "grad_norm": 9.3125,
      "learning_rate": 4.739104840300428e-05,
      "loss": 1.2046,
      "num_input_tokens_seen": 18076736,
      "step": 31160
    },
    {
      "epoch": 4.641793267798629,
      "grad_norm": 9.375,
      "learning_rate": 4.738960294980757e-05,
      "loss": 1.0437,
      "num_input_tokens_seen": 18079840,
      "step": 31165
    },
    {
      "epoch": 4.642537980339589,
      "grad_norm": 10.1875,
      "learning_rate": 4.738815711836128e-05,
      "loss": 1.0464,
      "num_input_tokens_seen": 18082592,
      "step": 31170
    },
    {
      "epoch": 4.643282692880548,
      "grad_norm": 7.625,
      "learning_rate": 4.738671090868982e-05,
      "loss": 0.8017,
      "num_input_tokens_seen": 18085664,
      "step": 31175
    },
    {
      "epoch": 4.644027405421507,
      "grad_norm": 14.5625,
      "learning_rate": 4.738526432081765e-05,
      "loss": 1.0638,
      "num_input_tokens_seen": 18088768,
      "step": 31180
    },
    {
      "epoch": 4.644772117962466,
      "grad_norm": 10.125,
      "learning_rate": 4.738381735476917e-05,
      "loss": 1.2295,
      "num_input_tokens_seen": 18091456,
      "step": 31185
    },
    {
      "epoch": 4.645516830503426,
      "grad_norm": 8.9375,
      "learning_rate": 4.738237001056886e-05,
      "loss": 1.1605,
      "num_input_tokens_seen": 18094176,
      "step": 31190
    },
    {
      "epoch": 4.6462615430443845,
      "grad_norm": 4.9375,
      "learning_rate": 4.738092228824115e-05,
      "loss": 0.8983,
      "num_input_tokens_seen": 18097120,
      "step": 31195
    },
    {
      "epoch": 4.647006255585344,
      "grad_norm": 26.125,
      "learning_rate": 4.7379474187810506e-05,
      "loss": 1.2719,
      "num_input_tokens_seen": 18099872,
      "step": 31200
    },
    {
      "epoch": 4.647750968126303,
      "grad_norm": 6.46875,
      "learning_rate": 4.7378025709301386e-05,
      "loss": 1.0334,
      "num_input_tokens_seen": 18102944,
      "step": 31205
    },
    {
      "epoch": 4.648495680667263,
      "grad_norm": 11.125,
      "learning_rate": 4.7376576852738274e-05,
      "loss": 0.9942,
      "num_input_tokens_seen": 18105792,
      "step": 31210
    },
    {
      "epoch": 4.649240393208221,
      "grad_norm": 9.5625,
      "learning_rate": 4.7375127618145645e-05,
      "loss": 0.9894,
      "num_input_tokens_seen": 18108640,
      "step": 31215
    },
    {
      "epoch": 4.649985105749181,
      "grad_norm": 9.5,
      "learning_rate": 4.737367800554796e-05,
      "loss": 0.9985,
      "num_input_tokens_seen": 18111712,
      "step": 31220
    },
    {
      "epoch": 4.65072981829014,
      "grad_norm": 18.625,
      "learning_rate": 4.737222801496973e-05,
      "loss": 1.5219,
      "num_input_tokens_seen": 18114656,
      "step": 31225
    },
    {
      "epoch": 4.651474530831099,
      "grad_norm": 13.9375,
      "learning_rate": 4.737077764643545e-05,
      "loss": 0.9853,
      "num_input_tokens_seen": 18117728,
      "step": 31230
    },
    {
      "epoch": 4.652219243372058,
      "grad_norm": 8.5,
      "learning_rate": 4.736932689996962e-05,
      "loss": 0.983,
      "num_input_tokens_seen": 18120416,
      "step": 31235
    },
    {
      "epoch": 4.652963955913018,
      "grad_norm": 20.875,
      "learning_rate": 4.7367875775596746e-05,
      "loss": 1.425,
      "num_input_tokens_seen": 18123200,
      "step": 31240
    },
    {
      "epoch": 4.653708668453977,
      "grad_norm": 13.625,
      "learning_rate": 4.7366424273341334e-05,
      "loss": 1.1923,
      "num_input_tokens_seen": 18126048,
      "step": 31245
    },
    {
      "epoch": 4.654453380994936,
      "grad_norm": 11.75,
      "learning_rate": 4.7364972393227925e-05,
      "loss": 0.8844,
      "num_input_tokens_seen": 18129024,
      "step": 31250
    },
    {
      "epoch": 4.655198093535895,
      "grad_norm": 5.46875,
      "learning_rate": 4.736352013528104e-05,
      "loss": 1.0046,
      "num_input_tokens_seen": 18132544,
      "step": 31255
    },
    {
      "epoch": 4.655942806076855,
      "grad_norm": 11.5625,
      "learning_rate": 4.736206749952521e-05,
      "loss": 0.956,
      "num_input_tokens_seen": 18135392,
      "step": 31260
    },
    {
      "epoch": 4.656687518617813,
      "grad_norm": 8.4375,
      "learning_rate": 4.736061448598498e-05,
      "loss": 1.1579,
      "num_input_tokens_seen": 18138272,
      "step": 31265
    },
    {
      "epoch": 4.657432231158773,
      "grad_norm": 9.5,
      "learning_rate": 4.7359161094684886e-05,
      "loss": 0.7962,
      "num_input_tokens_seen": 18141376,
      "step": 31270
    },
    {
      "epoch": 4.658176943699732,
      "grad_norm": 9.375,
      "learning_rate": 4.735770732564949e-05,
      "loss": 1.0231,
      "num_input_tokens_seen": 18144640,
      "step": 31275
    },
    {
      "epoch": 4.658921656240691,
      "grad_norm": 8.25,
      "learning_rate": 4.735625317890336e-05,
      "loss": 1.0423,
      "num_input_tokens_seen": 18147872,
      "step": 31280
    },
    {
      "epoch": 4.65966636878165,
      "grad_norm": 8.0625,
      "learning_rate": 4.735479865447105e-05,
      "loss": 1.2198,
      "num_input_tokens_seen": 18150656,
      "step": 31285
    },
    {
      "epoch": 4.66041108132261,
      "grad_norm": 13.875,
      "learning_rate": 4.735334375237714e-05,
      "loss": 1.0014,
      "num_input_tokens_seen": 18153568,
      "step": 31290
    },
    {
      "epoch": 4.661155793863569,
      "grad_norm": 16.125,
      "learning_rate": 4.7351888472646204e-05,
      "loss": 1.0561,
      "num_input_tokens_seen": 18156384,
      "step": 31295
    },
    {
      "epoch": 4.661900506404528,
      "grad_norm": 18.625,
      "learning_rate": 4.735043281530283e-05,
      "loss": 0.9978,
      "num_input_tokens_seen": 18159264,
      "step": 31300
    },
    {
      "epoch": 4.662645218945487,
      "grad_norm": 9.4375,
      "learning_rate": 4.7348976780371615e-05,
      "loss": 0.9663,
      "num_input_tokens_seen": 18162272,
      "step": 31305
    },
    {
      "epoch": 4.663389931486446,
      "grad_norm": 11.0,
      "learning_rate": 4.734752036787714e-05,
      "loss": 0.9659,
      "num_input_tokens_seen": 18165088,
      "step": 31310
    },
    {
      "epoch": 4.664134644027405,
      "grad_norm": 6.21875,
      "learning_rate": 4.734606357784403e-05,
      "loss": 0.9097,
      "num_input_tokens_seen": 18167808,
      "step": 31315
    },
    {
      "epoch": 4.664879356568365,
      "grad_norm": 11.5,
      "learning_rate": 4.734460641029689e-05,
      "loss": 0.9861,
      "num_input_tokens_seen": 18170912,
      "step": 31320
    },
    {
      "epoch": 4.665624069109324,
      "grad_norm": 10.3125,
      "learning_rate": 4.7343148865260326e-05,
      "loss": 1.3013,
      "num_input_tokens_seen": 18173728,
      "step": 31325
    },
    {
      "epoch": 4.666368781650283,
      "grad_norm": 11.4375,
      "learning_rate": 4.7341690942758974e-05,
      "loss": 1.1801,
      "num_input_tokens_seen": 18176576,
      "step": 31330
    },
    {
      "epoch": 4.667113494191242,
      "grad_norm": 8.0625,
      "learning_rate": 4.734023264281746e-05,
      "loss": 0.554,
      "num_input_tokens_seen": 18179520,
      "step": 31335
    },
    {
      "epoch": 4.667858206732202,
      "grad_norm": 11.5625,
      "learning_rate": 4.7338773965460426e-05,
      "loss": 1.1568,
      "num_input_tokens_seen": 18182496,
      "step": 31340
    },
    {
      "epoch": 4.668602919273161,
      "grad_norm": 9.3125,
      "learning_rate": 4.733731491071251e-05,
      "loss": 0.9848,
      "num_input_tokens_seen": 18185344,
      "step": 31345
    },
    {
      "epoch": 4.669347631814119,
      "grad_norm": 9.125,
      "learning_rate": 4.7335855478598354e-05,
      "loss": 0.8831,
      "num_input_tokens_seen": 18188288,
      "step": 31350
    },
    {
      "epoch": 4.670092344355079,
      "grad_norm": 5.875,
      "learning_rate": 4.7334395669142616e-05,
      "loss": 0.9295,
      "num_input_tokens_seen": 18191456,
      "step": 31355
    },
    {
      "epoch": 4.670837056896038,
      "grad_norm": 14.5625,
      "learning_rate": 4.733293548236998e-05,
      "loss": 1.0486,
      "num_input_tokens_seen": 18194400,
      "step": 31360
    },
    {
      "epoch": 4.671581769436997,
      "grad_norm": 7.0,
      "learning_rate": 4.7331474918305086e-05,
      "loss": 0.8233,
      "num_input_tokens_seen": 18197440,
      "step": 31365
    },
    {
      "epoch": 4.672326481977956,
      "grad_norm": 10.8125,
      "learning_rate": 4.733001397697262e-05,
      "loss": 1.2653,
      "num_input_tokens_seen": 18200128,
      "step": 31370
    },
    {
      "epoch": 4.673071194518916,
      "grad_norm": 13.9375,
      "learning_rate": 4.732855265839726e-05,
      "loss": 1.3165,
      "num_input_tokens_seen": 18203040,
      "step": 31375
    },
    {
      "epoch": 4.673815907059875,
      "grad_norm": 12.0625,
      "learning_rate": 4.7327090962603704e-05,
      "loss": 0.9592,
      "num_input_tokens_seen": 18205856,
      "step": 31380
    },
    {
      "epoch": 4.674560619600834,
      "grad_norm": 5.5625,
      "learning_rate": 4.7325628889616644e-05,
      "loss": 0.7618,
      "num_input_tokens_seen": 18208928,
      "step": 31385
    },
    {
      "epoch": 4.675305332141793,
      "grad_norm": 9.6875,
      "learning_rate": 4.732416643946076e-05,
      "loss": 0.7857,
      "num_input_tokens_seen": 18211968,
      "step": 31390
    },
    {
      "epoch": 4.676050044682753,
      "grad_norm": 13.0625,
      "learning_rate": 4.732270361216078e-05,
      "loss": 0.7098,
      "num_input_tokens_seen": 18214880,
      "step": 31395
    },
    {
      "epoch": 4.676794757223711,
      "grad_norm": 9.25,
      "learning_rate": 4.732124040774142e-05,
      "loss": 1.1124,
      "num_input_tokens_seen": 18217952,
      "step": 31400
    },
    {
      "epoch": 4.677539469764671,
      "grad_norm": 12.3125,
      "learning_rate": 4.731977682622737e-05,
      "loss": 1.4591,
      "num_input_tokens_seen": 18220832,
      "step": 31405
    },
    {
      "epoch": 4.67828418230563,
      "grad_norm": 8.625,
      "learning_rate": 4.731831286764339e-05,
      "loss": 0.8953,
      "num_input_tokens_seen": 18223808,
      "step": 31410
    },
    {
      "epoch": 4.6790288948465895,
      "grad_norm": 10.6875,
      "learning_rate": 4.731684853201419e-05,
      "loss": 1.1675,
      "num_input_tokens_seen": 18226720,
      "step": 31415
    },
    {
      "epoch": 4.679773607387548,
      "grad_norm": 16.75,
      "learning_rate": 4.7315383819364526e-05,
      "loss": 1.0189,
      "num_input_tokens_seen": 18229536,
      "step": 31420
    },
    {
      "epoch": 4.680518319928508,
      "grad_norm": 11.0,
      "learning_rate": 4.731391872971912e-05,
      "loss": 1.2128,
      "num_input_tokens_seen": 18232672,
      "step": 31425
    },
    {
      "epoch": 4.681263032469467,
      "grad_norm": 9.5,
      "learning_rate": 4.731245326310274e-05,
      "loss": 0.8229,
      "num_input_tokens_seen": 18235616,
      "step": 31430
    },
    {
      "epoch": 4.682007745010426,
      "grad_norm": 12.9375,
      "learning_rate": 4.731098741954014e-05,
      "loss": 1.0976,
      "num_input_tokens_seen": 18238656,
      "step": 31435
    },
    {
      "epoch": 4.682752457551385,
      "grad_norm": 9.75,
      "learning_rate": 4.730952119905609e-05,
      "loss": 1.157,
      "num_input_tokens_seen": 18241472,
      "step": 31440
    },
    {
      "epoch": 4.683497170092345,
      "grad_norm": 6.375,
      "learning_rate": 4.730805460167534e-05,
      "loss": 0.9015,
      "num_input_tokens_seen": 18244128,
      "step": 31445
    },
    {
      "epoch": 4.684241882633303,
      "grad_norm": 12.8125,
      "learning_rate": 4.730658762742269e-05,
      "loss": 0.9754,
      "num_input_tokens_seen": 18246752,
      "step": 31450
    },
    {
      "epoch": 4.684986595174263,
      "grad_norm": 14.5625,
      "learning_rate": 4.730512027632292e-05,
      "loss": 1.3926,
      "num_input_tokens_seen": 18249440,
      "step": 31455
    },
    {
      "epoch": 4.685731307715222,
      "grad_norm": 10.6875,
      "learning_rate": 4.7303652548400803e-05,
      "loss": 1.078,
      "num_input_tokens_seen": 18252256,
      "step": 31460
    },
    {
      "epoch": 4.6864760202561815,
      "grad_norm": 5.03125,
      "learning_rate": 4.730218444368114e-05,
      "loss": 0.9874,
      "num_input_tokens_seen": 18255104,
      "step": 31465
    },
    {
      "epoch": 4.68722073279714,
      "grad_norm": 10.25,
      "learning_rate": 4.7300715962188744e-05,
      "loss": 1.0386,
      "num_input_tokens_seen": 18257984,
      "step": 31470
    },
    {
      "epoch": 4.687965445338099,
      "grad_norm": 7.09375,
      "learning_rate": 4.7299247103948417e-05,
      "loss": 0.8031,
      "num_input_tokens_seen": 18260992,
      "step": 31475
    },
    {
      "epoch": 4.688710157879059,
      "grad_norm": 11.75,
      "learning_rate": 4.729777786898498e-05,
      "loss": 1.1018,
      "num_input_tokens_seen": 18263968,
      "step": 31480
    },
    {
      "epoch": 4.689454870420018,
      "grad_norm": 7.03125,
      "learning_rate": 4.729630825732324e-05,
      "loss": 1.0259,
      "num_input_tokens_seen": 18266976,
      "step": 31485
    },
    {
      "epoch": 4.690199582960977,
      "grad_norm": 5.59375,
      "learning_rate": 4.729483826898804e-05,
      "loss": 1.0631,
      "num_input_tokens_seen": 18270144,
      "step": 31490
    },
    {
      "epoch": 4.690944295501936,
      "grad_norm": 10.0625,
      "learning_rate": 4.729336790400421e-05,
      "loss": 1.0328,
      "num_input_tokens_seen": 18273344,
      "step": 31495
    },
    {
      "epoch": 4.6916890080428955,
      "grad_norm": 5.09375,
      "learning_rate": 4.729189716239657e-05,
      "loss": 0.9971,
      "num_input_tokens_seen": 18276288,
      "step": 31500
    },
    {
      "epoch": 4.692433720583855,
      "grad_norm": 8.9375,
      "learning_rate": 4.7290426044189995e-05,
      "loss": 1.1125,
      "num_input_tokens_seen": 18279136,
      "step": 31505
    },
    {
      "epoch": 4.693178433124814,
      "grad_norm": 12.125,
      "learning_rate": 4.728895454940933e-05,
      "loss": 1.2634,
      "num_input_tokens_seen": 18282016,
      "step": 31510
    },
    {
      "epoch": 4.693923145665773,
      "grad_norm": 7.6875,
      "learning_rate": 4.728748267807942e-05,
      "loss": 1.4992,
      "num_input_tokens_seen": 18284768,
      "step": 31515
    },
    {
      "epoch": 4.694667858206732,
      "grad_norm": 7.84375,
      "learning_rate": 4.728601043022515e-05,
      "loss": 0.7904,
      "num_input_tokens_seen": 18287680,
      "step": 31520
    },
    {
      "epoch": 4.695412570747691,
      "grad_norm": 8.8125,
      "learning_rate": 4.728453780587139e-05,
      "loss": 0.908,
      "num_input_tokens_seen": 18290624,
      "step": 31525
    },
    {
      "epoch": 4.696157283288651,
      "grad_norm": 21.375,
      "learning_rate": 4.7283064805043e-05,
      "loss": 1.8306,
      "num_input_tokens_seen": 18293408,
      "step": 31530
    },
    {
      "epoch": 4.696901995829609,
      "grad_norm": 10.0,
      "learning_rate": 4.7281591427764886e-05,
      "loss": 1.2398,
      "num_input_tokens_seen": 18296416,
      "step": 31535
    },
    {
      "epoch": 4.697646708370569,
      "grad_norm": 6.8125,
      "learning_rate": 4.728011767406193e-05,
      "loss": 1.0164,
      "num_input_tokens_seen": 18299200,
      "step": 31540
    },
    {
      "epoch": 4.698391420911528,
      "grad_norm": 10.375,
      "learning_rate": 4.7278643543959025e-05,
      "loss": 0.9907,
      "num_input_tokens_seen": 18302080,
      "step": 31545
    },
    {
      "epoch": 4.6991361334524875,
      "grad_norm": 8.25,
      "learning_rate": 4.727716903748108e-05,
      "loss": 1.1822,
      "num_input_tokens_seen": 18305088,
      "step": 31550
    },
    {
      "epoch": 4.699880845993446,
      "grad_norm": 5.71875,
      "learning_rate": 4.727569415465302e-05,
      "loss": 0.822,
      "num_input_tokens_seen": 18308160,
      "step": 31555
    },
    {
      "epoch": 4.700625558534406,
      "grad_norm": 15.3125,
      "learning_rate": 4.727421889549973e-05,
      "loss": 1.4098,
      "num_input_tokens_seen": 18310880,
      "step": 31560
    },
    {
      "epoch": 4.701370271075365,
      "grad_norm": 5.90625,
      "learning_rate": 4.727274326004616e-05,
      "loss": 1.0697,
      "num_input_tokens_seen": 18313536,
      "step": 31565
    },
    {
      "epoch": 4.702114983616324,
      "grad_norm": 12.0625,
      "learning_rate": 4.727126724831723e-05,
      "loss": 1.3147,
      "num_input_tokens_seen": 18316320,
      "step": 31570
    },
    {
      "epoch": 4.702859696157283,
      "grad_norm": 18.875,
      "learning_rate": 4.726979086033787e-05,
      "loss": 1.3853,
      "num_input_tokens_seen": 18319392,
      "step": 31575
    },
    {
      "epoch": 4.703604408698243,
      "grad_norm": 8.4375,
      "learning_rate": 4.726831409613303e-05,
      "loss": 1.3889,
      "num_input_tokens_seen": 18322432,
      "step": 31580
    },
    {
      "epoch": 4.7043491212392015,
      "grad_norm": 15.5,
      "learning_rate": 4.7266836955727655e-05,
      "loss": 1.6124,
      "num_input_tokens_seen": 18325344,
      "step": 31585
    },
    {
      "epoch": 4.705093833780161,
      "grad_norm": 18.375,
      "learning_rate": 4.72653594391467e-05,
      "loss": 1.3059,
      "num_input_tokens_seen": 18328192,
      "step": 31590
    },
    {
      "epoch": 4.70583854632112,
      "grad_norm": 6.1875,
      "learning_rate": 4.7263881546415135e-05,
      "loss": 1.2827,
      "num_input_tokens_seen": 18330976,
      "step": 31595
    },
    {
      "epoch": 4.7065832588620795,
      "grad_norm": 8.625,
      "learning_rate": 4.726240327755791e-05,
      "loss": 1.0319,
      "num_input_tokens_seen": 18333760,
      "step": 31600
    },
    {
      "epoch": 4.707327971403038,
      "grad_norm": 7.0625,
      "learning_rate": 4.726092463260001e-05,
      "loss": 0.9944,
      "num_input_tokens_seen": 18336672,
      "step": 31605
    },
    {
      "epoch": 4.708072683943998,
      "grad_norm": 9.875,
      "learning_rate": 4.7259445611566414e-05,
      "loss": 0.888,
      "num_input_tokens_seen": 18339648,
      "step": 31610
    },
    {
      "epoch": 4.708817396484957,
      "grad_norm": 8.5625,
      "learning_rate": 4.7257966214482106e-05,
      "loss": 0.7645,
      "num_input_tokens_seen": 18342496,
      "step": 31615
    },
    {
      "epoch": 4.709562109025916,
      "grad_norm": 13.9375,
      "learning_rate": 4.725648644137208e-05,
      "loss": 1.207,
      "num_input_tokens_seen": 18345536,
      "step": 31620
    },
    {
      "epoch": 4.710306821566875,
      "grad_norm": 9.5625,
      "learning_rate": 4.725500629226134e-05,
      "loss": 1.3978,
      "num_input_tokens_seen": 18348352,
      "step": 31625
    },
    {
      "epoch": 4.711051534107835,
      "grad_norm": 5.5625,
      "learning_rate": 4.725352576717489e-05,
      "loss": 0.9718,
      "num_input_tokens_seen": 18351328,
      "step": 31630
    },
    {
      "epoch": 4.7117962466487935,
      "grad_norm": 12.125,
      "learning_rate": 4.7252044866137736e-05,
      "loss": 0.9912,
      "num_input_tokens_seen": 18354112,
      "step": 31635
    },
    {
      "epoch": 4.712540959189753,
      "grad_norm": 5.84375,
      "learning_rate": 4.72505635891749e-05,
      "loss": 1.0866,
      "num_input_tokens_seen": 18356864,
      "step": 31640
    },
    {
      "epoch": 4.713285671730712,
      "grad_norm": 8.9375,
      "learning_rate": 4.7249081936311415e-05,
      "loss": 1.1464,
      "num_input_tokens_seen": 18359808,
      "step": 31645
    },
    {
      "epoch": 4.7140303842716715,
      "grad_norm": 8.3125,
      "learning_rate": 4.7247599907572285e-05,
      "loss": 1.2513,
      "num_input_tokens_seen": 18362720,
      "step": 31650
    },
    {
      "epoch": 4.71477509681263,
      "grad_norm": 13.9375,
      "learning_rate": 4.724611750298258e-05,
      "loss": 0.9488,
      "num_input_tokens_seen": 18365888,
      "step": 31655
    },
    {
      "epoch": 4.715519809353589,
      "grad_norm": 8.5625,
      "learning_rate": 4.724463472256733e-05,
      "loss": 1.2193,
      "num_input_tokens_seen": 18368512,
      "step": 31660
    },
    {
      "epoch": 4.716264521894549,
      "grad_norm": 9.6875,
      "learning_rate": 4.724315156635157e-05,
      "loss": 0.8868,
      "num_input_tokens_seen": 18371392,
      "step": 31665
    },
    {
      "epoch": 4.717009234435508,
      "grad_norm": 6.1875,
      "learning_rate": 4.7241668034360384e-05,
      "loss": 1.0286,
      "num_input_tokens_seen": 18374112,
      "step": 31670
    },
    {
      "epoch": 4.717753946976467,
      "grad_norm": 10.3125,
      "learning_rate": 4.724018412661883e-05,
      "loss": 1.1605,
      "num_input_tokens_seen": 18376864,
      "step": 31675
    },
    {
      "epoch": 4.718498659517426,
      "grad_norm": 7.5,
      "learning_rate": 4.7238699843151954e-05,
      "loss": 1.1779,
      "num_input_tokens_seen": 18379808,
      "step": 31680
    },
    {
      "epoch": 4.7192433720583855,
      "grad_norm": 6.65625,
      "learning_rate": 4.723721518398485e-05,
      "loss": 1.1354,
      "num_input_tokens_seen": 18382560,
      "step": 31685
    },
    {
      "epoch": 4.719988084599344,
      "grad_norm": 9.25,
      "learning_rate": 4.72357301491426e-05,
      "loss": 1.0715,
      "num_input_tokens_seen": 18385696,
      "step": 31690
    },
    {
      "epoch": 4.720732797140304,
      "grad_norm": 8.375,
      "learning_rate": 4.723424473865029e-05,
      "loss": 1.0375,
      "num_input_tokens_seen": 18388896,
      "step": 31695
    },
    {
      "epoch": 4.721477509681263,
      "grad_norm": 4.6875,
      "learning_rate": 4.7232758952533006e-05,
      "loss": 0.9883,
      "num_input_tokens_seen": 18391712,
      "step": 31700
    },
    {
      "epoch": 4.722222222222222,
      "grad_norm": 16.125,
      "learning_rate": 4.723127279081586e-05,
      "loss": 1.0559,
      "num_input_tokens_seen": 18394304,
      "step": 31705
    },
    {
      "epoch": 4.722966934763181,
      "grad_norm": 7.34375,
      "learning_rate": 4.7229786253523956e-05,
      "loss": 0.816,
      "num_input_tokens_seen": 18397280,
      "step": 31710
    },
    {
      "epoch": 4.723711647304141,
      "grad_norm": 12.5625,
      "learning_rate": 4.7228299340682405e-05,
      "loss": 0.8293,
      "num_input_tokens_seen": 18399968,
      "step": 31715
    },
    {
      "epoch": 4.7244563598450995,
      "grad_norm": 11.8125,
      "learning_rate": 4.7226812052316325e-05,
      "loss": 0.8118,
      "num_input_tokens_seen": 18402592,
      "step": 31720
    },
    {
      "epoch": 4.725201072386059,
      "grad_norm": 16.0,
      "learning_rate": 4.722532438845085e-05,
      "loss": 1.1639,
      "num_input_tokens_seen": 18405120,
      "step": 31725
    },
    {
      "epoch": 4.725945784927018,
      "grad_norm": 7.0625,
      "learning_rate": 4.7223836349111106e-05,
      "loss": 1.1536,
      "num_input_tokens_seen": 18408096,
      "step": 31730
    },
    {
      "epoch": 4.7266904974679775,
      "grad_norm": 22.875,
      "learning_rate": 4.722234793432224e-05,
      "loss": 1.3522,
      "num_input_tokens_seen": 18410880,
      "step": 31735
    },
    {
      "epoch": 4.727435210008936,
      "grad_norm": 13.75,
      "learning_rate": 4.722085914410938e-05,
      "loss": 1.3024,
      "num_input_tokens_seen": 18413600,
      "step": 31740
    },
    {
      "epoch": 4.728179922549896,
      "grad_norm": 10.125,
      "learning_rate": 4.7219369978497705e-05,
      "loss": 1.152,
      "num_input_tokens_seen": 18416512,
      "step": 31745
    },
    {
      "epoch": 4.728924635090855,
      "grad_norm": 8.1875,
      "learning_rate": 4.7217880437512344e-05,
      "loss": 0.9393,
      "num_input_tokens_seen": 18419424,
      "step": 31750
    },
    {
      "epoch": 4.729669347631814,
      "grad_norm": 7.71875,
      "learning_rate": 4.7216390521178475e-05,
      "loss": 1.2999,
      "num_input_tokens_seen": 18422208,
      "step": 31755
    },
    {
      "epoch": 4.730414060172773,
      "grad_norm": 7.90625,
      "learning_rate": 4.721490022952128e-05,
      "loss": 0.8376,
      "num_input_tokens_seen": 18424960,
      "step": 31760
    },
    {
      "epoch": 4.731158772713733,
      "grad_norm": 5.65625,
      "learning_rate": 4.721340956256591e-05,
      "loss": 0.7147,
      "num_input_tokens_seen": 18427744,
      "step": 31765
    },
    {
      "epoch": 4.7319034852546915,
      "grad_norm": 8.25,
      "learning_rate": 4.721191852033757e-05,
      "loss": 1.0705,
      "num_input_tokens_seen": 18430912,
      "step": 31770
    },
    {
      "epoch": 4.732648197795651,
      "grad_norm": 8.6875,
      "learning_rate": 4.7210427102861437e-05,
      "loss": 1.0092,
      "num_input_tokens_seen": 18434016,
      "step": 31775
    },
    {
      "epoch": 4.73339291033661,
      "grad_norm": 7.4375,
      "learning_rate": 4.720893531016271e-05,
      "loss": 1.2493,
      "num_input_tokens_seen": 18436896,
      "step": 31780
    },
    {
      "epoch": 4.73413762287757,
      "grad_norm": 13.6875,
      "learning_rate": 4.72074431422666e-05,
      "loss": 1.1603,
      "num_input_tokens_seen": 18439808,
      "step": 31785
    },
    {
      "epoch": 4.734882335418528,
      "grad_norm": 8.0625,
      "learning_rate": 4.72059505991983e-05,
      "loss": 0.7431,
      "num_input_tokens_seen": 18443104,
      "step": 31790
    },
    {
      "epoch": 4.735627047959488,
      "grad_norm": 17.875,
      "learning_rate": 4.720445768098304e-05,
      "loss": 1.5713,
      "num_input_tokens_seen": 18446208,
      "step": 31795
    },
    {
      "epoch": 4.736371760500447,
      "grad_norm": 14.625,
      "learning_rate": 4.720296438764604e-05,
      "loss": 0.9077,
      "num_input_tokens_seen": 18448896,
      "step": 31800
    },
    {
      "epoch": 4.737116473041406,
      "grad_norm": 5.5,
      "learning_rate": 4.7201470719212514e-05,
      "loss": 1.0756,
      "num_input_tokens_seen": 18451616,
      "step": 31805
    },
    {
      "epoch": 4.737861185582365,
      "grad_norm": 14.0,
      "learning_rate": 4.7199976675707716e-05,
      "loss": 0.9214,
      "num_input_tokens_seen": 18454464,
      "step": 31810
    },
    {
      "epoch": 4.738605898123325,
      "grad_norm": 5.875,
      "learning_rate": 4.719848225715686e-05,
      "loss": 0.9713,
      "num_input_tokens_seen": 18457376,
      "step": 31815
    },
    {
      "epoch": 4.7393506106642835,
      "grad_norm": 10.375,
      "learning_rate": 4.719698746358522e-05,
      "loss": 1.0353,
      "num_input_tokens_seen": 18460288,
      "step": 31820
    },
    {
      "epoch": 4.740095323205242,
      "grad_norm": 11.3125,
      "learning_rate": 4.719549229501803e-05,
      "loss": 1.378,
      "num_input_tokens_seen": 18463040,
      "step": 31825
    },
    {
      "epoch": 4.740840035746202,
      "grad_norm": 19.625,
      "learning_rate": 4.7193996751480555e-05,
      "loss": 1.0607,
      "num_input_tokens_seen": 18466112,
      "step": 31830
    },
    {
      "epoch": 4.741584748287162,
      "grad_norm": 8.625,
      "learning_rate": 4.7192500832998065e-05,
      "loss": 1.4154,
      "num_input_tokens_seen": 18468992,
      "step": 31835
    },
    {
      "epoch": 4.74232946082812,
      "grad_norm": 5.09375,
      "learning_rate": 4.719100453959583e-05,
      "loss": 0.8378,
      "num_input_tokens_seen": 18471712,
      "step": 31840
    },
    {
      "epoch": 4.743074173369079,
      "grad_norm": 14.5,
      "learning_rate": 4.718950787129912e-05,
      "loss": 0.7931,
      "num_input_tokens_seen": 18474368,
      "step": 31845
    },
    {
      "epoch": 4.743818885910039,
      "grad_norm": 13.0625,
      "learning_rate": 4.7188010828133233e-05,
      "loss": 1.0047,
      "num_input_tokens_seen": 18477280,
      "step": 31850
    },
    {
      "epoch": 4.744563598450998,
      "grad_norm": 15.0,
      "learning_rate": 4.7186513410123455e-05,
      "loss": 1.1596,
      "num_input_tokens_seen": 18480256,
      "step": 31855
    },
    {
      "epoch": 4.745308310991957,
      "grad_norm": 7.625,
      "learning_rate": 4.718501561729508e-05,
      "loss": 1.2468,
      "num_input_tokens_seen": 18483104,
      "step": 31860
    },
    {
      "epoch": 4.746053023532916,
      "grad_norm": 8.0625,
      "learning_rate": 4.718351744967342e-05,
      "loss": 1.1745,
      "num_input_tokens_seen": 18485952,
      "step": 31865
    },
    {
      "epoch": 4.746797736073876,
      "grad_norm": 8.6875,
      "learning_rate": 4.7182018907283776e-05,
      "loss": 0.9852,
      "num_input_tokens_seen": 18488928,
      "step": 31870
    },
    {
      "epoch": 4.747542448614834,
      "grad_norm": 4.34375,
      "learning_rate": 4.718051999015146e-05,
      "loss": 1.0293,
      "num_input_tokens_seen": 18491840,
      "step": 31875
    },
    {
      "epoch": 4.748287161155794,
      "grad_norm": 8.4375,
      "learning_rate": 4.7179020698301814e-05,
      "loss": 1.2171,
      "num_input_tokens_seen": 18494752,
      "step": 31880
    },
    {
      "epoch": 4.749031873696753,
      "grad_norm": 14.125,
      "learning_rate": 4.717752103176014e-05,
      "loss": 1.2512,
      "num_input_tokens_seen": 18497888,
      "step": 31885
    },
    {
      "epoch": 4.749776586237712,
      "grad_norm": 16.25,
      "learning_rate": 4.71760209905518e-05,
      "loss": 1.0549,
      "num_input_tokens_seen": 18500576,
      "step": 31890
    },
    {
      "epoch": 4.750521298778671,
      "grad_norm": 26.25,
      "learning_rate": 4.717452057470212e-05,
      "loss": 1.547,
      "num_input_tokens_seen": 18503136,
      "step": 31895
    },
    {
      "epoch": 4.751266011319631,
      "grad_norm": 11.8125,
      "learning_rate": 4.7173019784236455e-05,
      "loss": 1.115,
      "num_input_tokens_seen": 18505920,
      "step": 31900
    },
    {
      "epoch": 4.7520107238605895,
      "grad_norm": 5.875,
      "learning_rate": 4.717151861918015e-05,
      "loss": 1.0583,
      "num_input_tokens_seen": 18509024,
      "step": 31905
    },
    {
      "epoch": 4.752755436401549,
      "grad_norm": 8.5625,
      "learning_rate": 4.717001707955858e-05,
      "loss": 1.2853,
      "num_input_tokens_seen": 18511968,
      "step": 31910
    },
    {
      "epoch": 4.753500148942508,
      "grad_norm": 11.5,
      "learning_rate": 4.71685151653971e-05,
      "loss": 0.863,
      "num_input_tokens_seen": 18514848,
      "step": 31915
    },
    {
      "epoch": 4.754244861483468,
      "grad_norm": 7.9375,
      "learning_rate": 4.716701287672109e-05,
      "loss": 0.8406,
      "num_input_tokens_seen": 18517728,
      "step": 31920
    },
    {
      "epoch": 4.754989574024426,
      "grad_norm": 13.4375,
      "learning_rate": 4.716551021355593e-05,
      "loss": 1.2035,
      "num_input_tokens_seen": 18520480,
      "step": 31925
    },
    {
      "epoch": 4.755734286565386,
      "grad_norm": 14.0,
      "learning_rate": 4.716400717592699e-05,
      "loss": 1.2394,
      "num_input_tokens_seen": 18523040,
      "step": 31930
    },
    {
      "epoch": 4.756478999106345,
      "grad_norm": 9.5625,
      "learning_rate": 4.716250376385969e-05,
      "loss": 0.8873,
      "num_input_tokens_seen": 18525792,
      "step": 31935
    },
    {
      "epoch": 4.757223711647304,
      "grad_norm": 10.5,
      "learning_rate": 4.716099997737941e-05,
      "loss": 1.099,
      "num_input_tokens_seen": 18528608,
      "step": 31940
    },
    {
      "epoch": 4.757968424188263,
      "grad_norm": 14.0625,
      "learning_rate": 4.7159495816511546e-05,
      "loss": 0.6901,
      "num_input_tokens_seen": 18531296,
      "step": 31945
    },
    {
      "epoch": 4.758713136729223,
      "grad_norm": 9.4375,
      "learning_rate": 4.7157991281281536e-05,
      "loss": 1.1896,
      "num_input_tokens_seen": 18534304,
      "step": 31950
    },
    {
      "epoch": 4.759457849270182,
      "grad_norm": 13.375,
      "learning_rate": 4.715648637171478e-05,
      "loss": 1.2238,
      "num_input_tokens_seen": 18537280,
      "step": 31955
    },
    {
      "epoch": 4.760202561811141,
      "grad_norm": 10.625,
      "learning_rate": 4.71549810878367e-05,
      "loss": 0.9816,
      "num_input_tokens_seen": 18540544,
      "step": 31960
    },
    {
      "epoch": 4.7609472743521,
      "grad_norm": 4.59375,
      "learning_rate": 4.7153475429672736e-05,
      "loss": 1.2368,
      "num_input_tokens_seen": 18543552,
      "step": 31965
    },
    {
      "epoch": 4.76169198689306,
      "grad_norm": 12.75,
      "learning_rate": 4.715196939724832e-05,
      "loss": 0.8926,
      "num_input_tokens_seen": 18546528,
      "step": 31970
    },
    {
      "epoch": 4.762436699434018,
      "grad_norm": 12.25,
      "learning_rate": 4.71504629905889e-05,
      "loss": 1.3738,
      "num_input_tokens_seen": 18549728,
      "step": 31975
    },
    {
      "epoch": 4.763181411974978,
      "grad_norm": 6.1875,
      "learning_rate": 4.714895620971992e-05,
      "loss": 1.2653,
      "num_input_tokens_seen": 18552768,
      "step": 31980
    },
    {
      "epoch": 4.763926124515937,
      "grad_norm": 16.625,
      "learning_rate": 4.714744905466683e-05,
      "loss": 0.9992,
      "num_input_tokens_seen": 18555552,
      "step": 31985
    },
    {
      "epoch": 4.764670837056896,
      "grad_norm": 12.1875,
      "learning_rate": 4.71459415254551e-05,
      "loss": 1.2961,
      "num_input_tokens_seen": 18558240,
      "step": 31990
    },
    {
      "epoch": 4.765415549597855,
      "grad_norm": 9.0625,
      "learning_rate": 4.71444336221102e-05,
      "loss": 0.821,
      "num_input_tokens_seen": 18561184,
      "step": 31995
    },
    {
      "epoch": 4.766160262138815,
      "grad_norm": 6.46875,
      "learning_rate": 4.71429253446576e-05,
      "loss": 0.9859,
      "num_input_tokens_seen": 18564192,
      "step": 32000
    },
    {
      "epoch": 4.766904974679774,
      "grad_norm": 10.625,
      "learning_rate": 4.714141669312278e-05,
      "loss": 0.981,
      "num_input_tokens_seen": 18567040,
      "step": 32005
    },
    {
      "epoch": 4.767649687220732,
      "grad_norm": 11.6875,
      "learning_rate": 4.713990766753123e-05,
      "loss": 1.1159,
      "num_input_tokens_seen": 18570048,
      "step": 32010
    },
    {
      "epoch": 4.768394399761692,
      "grad_norm": 6.6875,
      "learning_rate": 4.7138398267908434e-05,
      "loss": 1.0956,
      "num_input_tokens_seen": 18573312,
      "step": 32015
    },
    {
      "epoch": 4.769139112302652,
      "grad_norm": 11.0,
      "learning_rate": 4.713688849427991e-05,
      "loss": 1.5901,
      "num_input_tokens_seen": 18576096,
      "step": 32020
    },
    {
      "epoch": 4.76988382484361,
      "grad_norm": 13.0625,
      "learning_rate": 4.713537834667115e-05,
      "loss": 1.2052,
      "num_input_tokens_seen": 18579264,
      "step": 32025
    },
    {
      "epoch": 4.770628537384569,
      "grad_norm": 15.25,
      "learning_rate": 4.713386782510766e-05,
      "loss": 0.9797,
      "num_input_tokens_seen": 18581888,
      "step": 32030
    },
    {
      "epoch": 4.771373249925529,
      "grad_norm": 7.40625,
      "learning_rate": 4.713235692961498e-05,
      "loss": 1.1956,
      "num_input_tokens_seen": 18584928,
      "step": 32035
    },
    {
      "epoch": 4.772117962466488,
      "grad_norm": 13.625,
      "learning_rate": 4.713084566021863e-05,
      "loss": 1.3146,
      "num_input_tokens_seen": 18587584,
      "step": 32040
    },
    {
      "epoch": 4.772862675007447,
      "grad_norm": 8.375,
      "learning_rate": 4.7129334016944124e-05,
      "loss": 0.9625,
      "num_input_tokens_seen": 18590272,
      "step": 32045
    },
    {
      "epoch": 4.773607387548406,
      "grad_norm": 10.0625,
      "learning_rate": 4.7127821999817014e-05,
      "loss": 1.049,
      "num_input_tokens_seen": 18593376,
      "step": 32050
    },
    {
      "epoch": 4.774352100089366,
      "grad_norm": 8.25,
      "learning_rate": 4.712630960886284e-05,
      "loss": 0.9931,
      "num_input_tokens_seen": 18596224,
      "step": 32055
    },
    {
      "epoch": 4.775096812630324,
      "grad_norm": 5.1875,
      "learning_rate": 4.7124796844107155e-05,
      "loss": 1.06,
      "num_input_tokens_seen": 18599104,
      "step": 32060
    },
    {
      "epoch": 4.775841525171284,
      "grad_norm": 7.65625,
      "learning_rate": 4.7123283705575514e-05,
      "loss": 1.2002,
      "num_input_tokens_seen": 18601920,
      "step": 32065
    },
    {
      "epoch": 4.776586237712243,
      "grad_norm": 12.125,
      "learning_rate": 4.712177019329348e-05,
      "loss": 0.98,
      "num_input_tokens_seen": 18605088,
      "step": 32070
    },
    {
      "epoch": 4.777330950253202,
      "grad_norm": 13.25,
      "learning_rate": 4.712025630728662e-05,
      "loss": 1.2925,
      "num_input_tokens_seen": 18607712,
      "step": 32075
    },
    {
      "epoch": 4.778075662794161,
      "grad_norm": 10.0,
      "learning_rate": 4.711874204758051e-05,
      "loss": 1.2274,
      "num_input_tokens_seen": 18610688,
      "step": 32080
    },
    {
      "epoch": 4.778820375335121,
      "grad_norm": 4.09375,
      "learning_rate": 4.7117227414200735e-05,
      "loss": 1.0,
      "num_input_tokens_seen": 18613824,
      "step": 32085
    },
    {
      "epoch": 4.77956508787608,
      "grad_norm": 12.3125,
      "learning_rate": 4.711571240717289e-05,
      "loss": 1.138,
      "num_input_tokens_seen": 18616704,
      "step": 32090
    },
    {
      "epoch": 4.780309800417039,
      "grad_norm": 14.875,
      "learning_rate": 4.7114197026522555e-05,
      "loss": 1.293,
      "num_input_tokens_seen": 18619648,
      "step": 32095
    },
    {
      "epoch": 4.781054512957998,
      "grad_norm": 9.3125,
      "learning_rate": 4.711268127227534e-05,
      "loss": 0.991,
      "num_input_tokens_seen": 18622656,
      "step": 32100
    },
    {
      "epoch": 4.781799225498958,
      "grad_norm": 10.625,
      "learning_rate": 4.711116514445685e-05,
      "loss": 1.3156,
      "num_input_tokens_seen": 18625536,
      "step": 32105
    },
    {
      "epoch": 4.782543938039916,
      "grad_norm": 12.5625,
      "learning_rate": 4.710964864309269e-05,
      "loss": 1.0763,
      "num_input_tokens_seen": 18628480,
      "step": 32110
    },
    {
      "epoch": 4.783288650580876,
      "grad_norm": 18.625,
      "learning_rate": 4.710813176820848e-05,
      "loss": 1.1301,
      "num_input_tokens_seen": 18631488,
      "step": 32115
    },
    {
      "epoch": 4.784033363121835,
      "grad_norm": 9.4375,
      "learning_rate": 4.710661451982987e-05,
      "loss": 1.3366,
      "num_input_tokens_seen": 18634720,
      "step": 32120
    },
    {
      "epoch": 4.7847780756627944,
      "grad_norm": 5.5625,
      "learning_rate": 4.7105096897982473e-05,
      "loss": 0.8861,
      "num_input_tokens_seen": 18637664,
      "step": 32125
    },
    {
      "epoch": 4.785522788203753,
      "grad_norm": 16.75,
      "learning_rate": 4.710357890269193e-05,
      "loss": 1.4955,
      "num_input_tokens_seen": 18640704,
      "step": 32130
    },
    {
      "epoch": 4.786267500744713,
      "grad_norm": 7.3125,
      "learning_rate": 4.710206053398388e-05,
      "loss": 1.2469,
      "num_input_tokens_seen": 18643264,
      "step": 32135
    },
    {
      "epoch": 4.787012213285672,
      "grad_norm": 17.375,
      "learning_rate": 4.710054179188399e-05,
      "loss": 1.3293,
      "num_input_tokens_seen": 18645952,
      "step": 32140
    },
    {
      "epoch": 4.787756925826631,
      "grad_norm": 9.5,
      "learning_rate": 4.7099022676417904e-05,
      "loss": 1.0371,
      "num_input_tokens_seen": 18649024,
      "step": 32145
    },
    {
      "epoch": 4.78850163836759,
      "grad_norm": 11.75,
      "learning_rate": 4.709750318761129e-05,
      "loss": 1.0912,
      "num_input_tokens_seen": 18652192,
      "step": 32150
    },
    {
      "epoch": 4.78924635090855,
      "grad_norm": 12.0,
      "learning_rate": 4.709598332548982e-05,
      "loss": 1.3101,
      "num_input_tokens_seen": 18655008,
      "step": 32155
    },
    {
      "epoch": 4.789991063449508,
      "grad_norm": 14.8125,
      "learning_rate": 4.709446309007917e-05,
      "loss": 1.2489,
      "num_input_tokens_seen": 18657728,
      "step": 32160
    },
    {
      "epoch": 4.790735775990468,
      "grad_norm": 10.4375,
      "learning_rate": 4.709294248140502e-05,
      "loss": 0.9208,
      "num_input_tokens_seen": 18660864,
      "step": 32165
    },
    {
      "epoch": 4.791480488531427,
      "grad_norm": 16.375,
      "learning_rate": 4.709142149949306e-05,
      "loss": 1.52,
      "num_input_tokens_seen": 18663680,
      "step": 32170
    },
    {
      "epoch": 4.792225201072386,
      "grad_norm": 11.375,
      "learning_rate": 4.708990014436899e-05,
      "loss": 1.2093,
      "num_input_tokens_seen": 18666400,
      "step": 32175
    },
    {
      "epoch": 4.792969913613345,
      "grad_norm": 7.90625,
      "learning_rate": 4.708837841605851e-05,
      "loss": 1.0666,
      "num_input_tokens_seen": 18668960,
      "step": 32180
    },
    {
      "epoch": 4.793714626154305,
      "grad_norm": 13.4375,
      "learning_rate": 4.7086856314587316e-05,
      "loss": 1.0417,
      "num_input_tokens_seen": 18672064,
      "step": 32185
    },
    {
      "epoch": 4.794459338695264,
      "grad_norm": 13.9375,
      "learning_rate": 4.708533383998114e-05,
      "loss": 1.1976,
      "num_input_tokens_seen": 18674816,
      "step": 32190
    },
    {
      "epoch": 4.795204051236222,
      "grad_norm": 7.0625,
      "learning_rate": 4.7083810992265696e-05,
      "loss": 1.0479,
      "num_input_tokens_seen": 18677632,
      "step": 32195
    },
    {
      "epoch": 4.795948763777182,
      "grad_norm": 6.46875,
      "learning_rate": 4.7082287771466705e-05,
      "loss": 1.1438,
      "num_input_tokens_seen": 18680512,
      "step": 32200
    },
    {
      "epoch": 4.796693476318142,
      "grad_norm": 7.9375,
      "learning_rate": 4.7080764177609914e-05,
      "loss": 1.2732,
      "num_input_tokens_seen": 18683328,
      "step": 32205
    },
    {
      "epoch": 4.7974381888591004,
      "grad_norm": 5.0625,
      "learning_rate": 4.7079240210721046e-05,
      "loss": 0.8066,
      "num_input_tokens_seen": 18686144,
      "step": 32210
    },
    {
      "epoch": 4.798182901400059,
      "grad_norm": 8.0625,
      "learning_rate": 4.707771587082586e-05,
      "loss": 0.8979,
      "num_input_tokens_seen": 18688832,
      "step": 32215
    },
    {
      "epoch": 4.798927613941019,
      "grad_norm": 11.8125,
      "learning_rate": 4.70761911579501e-05,
      "loss": 1.0697,
      "num_input_tokens_seen": 18691712,
      "step": 32220
    },
    {
      "epoch": 4.799672326481978,
      "grad_norm": 7.4375,
      "learning_rate": 4.707466607211953e-05,
      "loss": 0.9118,
      "num_input_tokens_seen": 18694912,
      "step": 32225
    },
    {
      "epoch": 4.800417039022937,
      "grad_norm": 15.75,
      "learning_rate": 4.707314061335991e-05,
      "loss": 1.1166,
      "num_input_tokens_seen": 18697952,
      "step": 32230
    },
    {
      "epoch": 4.801161751563896,
      "grad_norm": 9.75,
      "learning_rate": 4.707161478169702e-05,
      "loss": 1.4472,
      "num_input_tokens_seen": 18700768,
      "step": 32235
    },
    {
      "epoch": 4.801906464104856,
      "grad_norm": 13.5625,
      "learning_rate": 4.7070088577156625e-05,
      "loss": 1.2091,
      "num_input_tokens_seen": 18703968,
      "step": 32240
    },
    {
      "epoch": 4.802651176645814,
      "grad_norm": 15.125,
      "learning_rate": 4.706856199976451e-05,
      "loss": 1.06,
      "num_input_tokens_seen": 18706784,
      "step": 32245
    },
    {
      "epoch": 4.803395889186774,
      "grad_norm": 14.1875,
      "learning_rate": 4.706703504954647e-05,
      "loss": 0.9379,
      "num_input_tokens_seen": 18709376,
      "step": 32250
    },
    {
      "epoch": 4.804140601727733,
      "grad_norm": 8.3125,
      "learning_rate": 4.706550772652831e-05,
      "loss": 1.103,
      "num_input_tokens_seen": 18712096,
      "step": 32255
    },
    {
      "epoch": 4.8048853142686925,
      "grad_norm": 12.875,
      "learning_rate": 4.7063980030735824e-05,
      "loss": 1.1278,
      "num_input_tokens_seen": 18715008,
      "step": 32260
    },
    {
      "epoch": 4.805630026809651,
      "grad_norm": 12.0625,
      "learning_rate": 4.7062451962194806e-05,
      "loss": 0.8119,
      "num_input_tokens_seen": 18717856,
      "step": 32265
    },
    {
      "epoch": 4.806374739350611,
      "grad_norm": 4.28125,
      "learning_rate": 4.70609235209311e-05,
      "loss": 0.703,
      "num_input_tokens_seen": 18720512,
      "step": 32270
    },
    {
      "epoch": 4.80711945189157,
      "grad_norm": 10.9375,
      "learning_rate": 4.705939470697051e-05,
      "loss": 1.3116,
      "num_input_tokens_seen": 18723264,
      "step": 32275
    },
    {
      "epoch": 4.807864164432529,
      "grad_norm": 10.1875,
      "learning_rate": 4.7057865520338865e-05,
      "loss": 0.923,
      "num_input_tokens_seen": 18726496,
      "step": 32280
    },
    {
      "epoch": 4.808608876973488,
      "grad_norm": 13.4375,
      "learning_rate": 4.7056335961061994e-05,
      "loss": 1.1025,
      "num_input_tokens_seen": 18729280,
      "step": 32285
    },
    {
      "epoch": 4.809353589514448,
      "grad_norm": 9.625,
      "learning_rate": 4.705480602916575e-05,
      "loss": 1.0419,
      "num_input_tokens_seen": 18732160,
      "step": 32290
    },
    {
      "epoch": 4.8100983020554064,
      "grad_norm": 18.25,
      "learning_rate": 4.705327572467597e-05,
      "loss": 1.3719,
      "num_input_tokens_seen": 18734944,
      "step": 32295
    },
    {
      "epoch": 4.810843014596366,
      "grad_norm": 12.9375,
      "learning_rate": 4.705174504761851e-05,
      "loss": 1.4339,
      "num_input_tokens_seen": 18737440,
      "step": 32300
    },
    {
      "epoch": 4.811587727137325,
      "grad_norm": 10.875,
      "learning_rate": 4.705021399801924e-05,
      "loss": 1.0734,
      "num_input_tokens_seen": 18740256,
      "step": 32305
    },
    {
      "epoch": 4.8123324396782845,
      "grad_norm": 11.625,
      "learning_rate": 4.704868257590401e-05,
      "loss": 0.9672,
      "num_input_tokens_seen": 18743104,
      "step": 32310
    },
    {
      "epoch": 4.813077152219243,
      "grad_norm": 13.1875,
      "learning_rate": 4.7047150781298693e-05,
      "loss": 0.9246,
      "num_input_tokens_seen": 18745888,
      "step": 32315
    },
    {
      "epoch": 4.813821864760203,
      "grad_norm": 12.875,
      "learning_rate": 4.704561861422917e-05,
      "loss": 1.4729,
      "num_input_tokens_seen": 18748960,
      "step": 32320
    },
    {
      "epoch": 4.814566577301162,
      "grad_norm": 16.875,
      "learning_rate": 4.704408607472134e-05,
      "loss": 0.9844,
      "num_input_tokens_seen": 18752000,
      "step": 32325
    },
    {
      "epoch": 4.815311289842121,
      "grad_norm": 6.84375,
      "learning_rate": 4.704255316280106e-05,
      "loss": 1.1434,
      "num_input_tokens_seen": 18754944,
      "step": 32330
    },
    {
      "epoch": 4.81605600238308,
      "grad_norm": 17.75,
      "learning_rate": 4.704101987849426e-05,
      "loss": 1.2838,
      "num_input_tokens_seen": 18757696,
      "step": 32335
    },
    {
      "epoch": 4.816800714924039,
      "grad_norm": 9.0,
      "learning_rate": 4.7039486221826834e-05,
      "loss": 1.2778,
      "num_input_tokens_seen": 18760576,
      "step": 32340
    },
    {
      "epoch": 4.8175454274649985,
      "grad_norm": 10.0,
      "learning_rate": 4.703795219282469e-05,
      "loss": 1.0787,
      "num_input_tokens_seen": 18763616,
      "step": 32345
    },
    {
      "epoch": 4.818290140005958,
      "grad_norm": 7.5625,
      "learning_rate": 4.7036417791513735e-05,
      "loss": 0.8338,
      "num_input_tokens_seen": 18766688,
      "step": 32350
    },
    {
      "epoch": 4.819034852546917,
      "grad_norm": 10.8125,
      "learning_rate": 4.7034883017919896e-05,
      "loss": 1.3222,
      "num_input_tokens_seen": 18769504,
      "step": 32355
    },
    {
      "epoch": 4.819779565087876,
      "grad_norm": 9.3125,
      "learning_rate": 4.70333478720691e-05,
      "loss": 0.9731,
      "num_input_tokens_seen": 18772384,
      "step": 32360
    },
    {
      "epoch": 4.820524277628835,
      "grad_norm": 10.5625,
      "learning_rate": 4.703181235398729e-05,
      "loss": 0.9838,
      "num_input_tokens_seen": 18775392,
      "step": 32365
    },
    {
      "epoch": 4.821268990169795,
      "grad_norm": 4.5,
      "learning_rate": 4.7030276463700405e-05,
      "loss": 0.7825,
      "num_input_tokens_seen": 18778304,
      "step": 32370
    },
    {
      "epoch": 4.822013702710754,
      "grad_norm": 12.0625,
      "learning_rate": 4.702874020123439e-05,
      "loss": 1.3003,
      "num_input_tokens_seen": 18781376,
      "step": 32375
    },
    {
      "epoch": 4.8227584152517124,
      "grad_norm": 16.75,
      "learning_rate": 4.70272035666152e-05,
      "loss": 1.3822,
      "num_input_tokens_seen": 18784288,
      "step": 32380
    },
    {
      "epoch": 4.823503127792672,
      "grad_norm": 6.78125,
      "learning_rate": 4.702566655986879e-05,
      "loss": 1.2337,
      "num_input_tokens_seen": 18787488,
      "step": 32385
    },
    {
      "epoch": 4.824247840333631,
      "grad_norm": 14.3125,
      "learning_rate": 4.702412918102113e-05,
      "loss": 1.1201,
      "num_input_tokens_seen": 18790528,
      "step": 32390
    },
    {
      "epoch": 4.8249925528745905,
      "grad_norm": 9.75,
      "learning_rate": 4.702259143009819e-05,
      "loss": 0.9766,
      "num_input_tokens_seen": 18793472,
      "step": 32395
    },
    {
      "epoch": 4.825737265415549,
      "grad_norm": 9.8125,
      "learning_rate": 4.702105330712595e-05,
      "loss": 1.0747,
      "num_input_tokens_seen": 18796416,
      "step": 32400
    },
    {
      "epoch": 4.826481977956509,
      "grad_norm": 13.0,
      "learning_rate": 4.70195148121304e-05,
      "loss": 1.3814,
      "num_input_tokens_seen": 18799104,
      "step": 32405
    },
    {
      "epoch": 4.827226690497468,
      "grad_norm": 9.4375,
      "learning_rate": 4.7017975945137524e-05,
      "loss": 1.0229,
      "num_input_tokens_seen": 18801984,
      "step": 32410
    },
    {
      "epoch": 4.827971403038427,
      "grad_norm": 9.4375,
      "learning_rate": 4.701643670617333e-05,
      "loss": 1.0694,
      "num_input_tokens_seen": 18804736,
      "step": 32415
    },
    {
      "epoch": 4.828716115579386,
      "grad_norm": 7.4375,
      "learning_rate": 4.70148970952638e-05,
      "loss": 1.1008,
      "num_input_tokens_seen": 18807744,
      "step": 32420
    },
    {
      "epoch": 4.829460828120346,
      "grad_norm": 4.75,
      "learning_rate": 4.701335711243497e-05,
      "loss": 0.9366,
      "num_input_tokens_seen": 18810624,
      "step": 32425
    },
    {
      "epoch": 4.8302055406613045,
      "grad_norm": 10.375,
      "learning_rate": 4.7011816757712835e-05,
      "loss": 1.221,
      "num_input_tokens_seen": 18813664,
      "step": 32430
    },
    {
      "epoch": 4.830950253202264,
      "grad_norm": 6.5625,
      "learning_rate": 4.701027603112343e-05,
      "loss": 1.4038,
      "num_input_tokens_seen": 18816672,
      "step": 32435
    },
    {
      "epoch": 4.831694965743223,
      "grad_norm": 12.1875,
      "learning_rate": 4.7008734932692795e-05,
      "loss": 1.0985,
      "num_input_tokens_seen": 18819584,
      "step": 32440
    },
    {
      "epoch": 4.8324396782841825,
      "grad_norm": 8.1875,
      "learning_rate": 4.700719346244694e-05,
      "loss": 1.056,
      "num_input_tokens_seen": 18822272,
      "step": 32445
    },
    {
      "epoch": 4.833184390825141,
      "grad_norm": 12.75,
      "learning_rate": 4.7005651620411914e-05,
      "loss": 0.9381,
      "num_input_tokens_seen": 18824928,
      "step": 32450
    },
    {
      "epoch": 4.833929103366101,
      "grad_norm": 9.25,
      "learning_rate": 4.7004109406613786e-05,
      "loss": 1.04,
      "num_input_tokens_seen": 18827712,
      "step": 32455
    },
    {
      "epoch": 4.83467381590706,
      "grad_norm": 12.375,
      "learning_rate": 4.700256682107858e-05,
      "loss": 1.4945,
      "num_input_tokens_seen": 18830784,
      "step": 32460
    },
    {
      "epoch": 4.835418528448019,
      "grad_norm": 13.5,
      "learning_rate": 4.700102386383237e-05,
      "loss": 1.352,
      "num_input_tokens_seen": 18833792,
      "step": 32465
    },
    {
      "epoch": 4.836163240988978,
      "grad_norm": 13.0625,
      "learning_rate": 4.699948053490123e-05,
      "loss": 1.2531,
      "num_input_tokens_seen": 18836928,
      "step": 32470
    },
    {
      "epoch": 4.836907953529938,
      "grad_norm": 8.625,
      "learning_rate": 4.699793683431122e-05,
      "loss": 0.8234,
      "num_input_tokens_seen": 18839840,
      "step": 32475
    },
    {
      "epoch": 4.8376526660708965,
      "grad_norm": 9.75,
      "learning_rate": 4.699639276208843e-05,
      "loss": 1.2895,
      "num_input_tokens_seen": 18842944,
      "step": 32480
    },
    {
      "epoch": 4.838397378611856,
      "grad_norm": 8.0625,
      "learning_rate": 4.699484831825894e-05,
      "loss": 1.3335,
      "num_input_tokens_seen": 18845888,
      "step": 32485
    },
    {
      "epoch": 4.839142091152815,
      "grad_norm": 11.75,
      "learning_rate": 4.699330350284884e-05,
      "loss": 1.191,
      "num_input_tokens_seen": 18848672,
      "step": 32490
    },
    {
      "epoch": 4.8398868036937746,
      "grad_norm": 4.46875,
      "learning_rate": 4.6991758315884225e-05,
      "loss": 0.6787,
      "num_input_tokens_seen": 18851712,
      "step": 32495
    },
    {
      "epoch": 4.840631516234733,
      "grad_norm": 8.1875,
      "learning_rate": 4.699021275739121e-05,
      "loss": 1.1427,
      "num_input_tokens_seen": 18854720,
      "step": 32500
    },
    {
      "epoch": 4.841376228775693,
      "grad_norm": 10.5625,
      "learning_rate": 4.69886668273959e-05,
      "loss": 1.0827,
      "num_input_tokens_seen": 18857408,
      "step": 32505
    },
    {
      "epoch": 4.842120941316652,
      "grad_norm": 13.25,
      "learning_rate": 4.698712052592441e-05,
      "loss": 1.5324,
      "num_input_tokens_seen": 18860224,
      "step": 32510
    },
    {
      "epoch": 4.842865653857611,
      "grad_norm": 13.375,
      "learning_rate": 4.6985573853002875e-05,
      "loss": 1.12,
      "num_input_tokens_seen": 18863104,
      "step": 32515
    },
    {
      "epoch": 4.84361036639857,
      "grad_norm": 6.28125,
      "learning_rate": 4.698402680865741e-05,
      "loss": 1.0917,
      "num_input_tokens_seen": 18866176,
      "step": 32520
    },
    {
      "epoch": 4.844355078939529,
      "grad_norm": 8.0,
      "learning_rate": 4.6982479392914144e-05,
      "loss": 1.14,
      "num_input_tokens_seen": 18869216,
      "step": 32525
    },
    {
      "epoch": 4.8450997914804885,
      "grad_norm": 7.84375,
      "learning_rate": 4.698093160579924e-05,
      "loss": 0.983,
      "num_input_tokens_seen": 18872032,
      "step": 32530
    },
    {
      "epoch": 4.845844504021448,
      "grad_norm": 8.8125,
      "learning_rate": 4.697938344733884e-05,
      "loss": 0.9942,
      "num_input_tokens_seen": 18875008,
      "step": 32535
    },
    {
      "epoch": 4.846589216562407,
      "grad_norm": 13.0,
      "learning_rate": 4.6977834917559095e-05,
      "loss": 0.9527,
      "num_input_tokens_seen": 18877792,
      "step": 32540
    },
    {
      "epoch": 4.847333929103366,
      "grad_norm": 15.875,
      "learning_rate": 4.697628601648616e-05,
      "loss": 1.6201,
      "num_input_tokens_seen": 18880672,
      "step": 32545
    },
    {
      "epoch": 4.848078641644325,
      "grad_norm": 8.25,
      "learning_rate": 4.697473674414621e-05,
      "loss": 0.8327,
      "num_input_tokens_seen": 18883328,
      "step": 32550
    },
    {
      "epoch": 4.848823354185284,
      "grad_norm": 7.84375,
      "learning_rate": 4.697318710056542e-05,
      "loss": 0.9375,
      "num_input_tokens_seen": 18886144,
      "step": 32555
    },
    {
      "epoch": 4.849568066726244,
      "grad_norm": 16.0,
      "learning_rate": 4.697163708576997e-05,
      "loss": 0.9517,
      "num_input_tokens_seen": 18888928,
      "step": 32560
    },
    {
      "epoch": 4.8503127792672025,
      "grad_norm": 15.3125,
      "learning_rate": 4.697008669978603e-05,
      "loss": 1.4703,
      "num_input_tokens_seen": 18891680,
      "step": 32565
    },
    {
      "epoch": 4.851057491808162,
      "grad_norm": 13.4375,
      "learning_rate": 4.696853594263981e-05,
      "loss": 1.0707,
      "num_input_tokens_seen": 18894528,
      "step": 32570
    },
    {
      "epoch": 4.851802204349121,
      "grad_norm": 8.5625,
      "learning_rate": 4.6966984814357515e-05,
      "loss": 0.8638,
      "num_input_tokens_seen": 18897248,
      "step": 32575
    },
    {
      "epoch": 4.8525469168900806,
      "grad_norm": 7.90625,
      "learning_rate": 4.6965433314965325e-05,
      "loss": 0.9319,
      "num_input_tokens_seen": 18900352,
      "step": 32580
    },
    {
      "epoch": 4.853291629431039,
      "grad_norm": 7.53125,
      "learning_rate": 4.6963881444489464e-05,
      "loss": 0.9409,
      "num_input_tokens_seen": 18903456,
      "step": 32585
    },
    {
      "epoch": 4.854036341971999,
      "grad_norm": 10.6875,
      "learning_rate": 4.696232920295616e-05,
      "loss": 1.2001,
      "num_input_tokens_seen": 18906336,
      "step": 32590
    },
    {
      "epoch": 4.854781054512958,
      "grad_norm": 8.625,
      "learning_rate": 4.696077659039161e-05,
      "loss": 1.0266,
      "num_input_tokens_seen": 18909536,
      "step": 32595
    },
    {
      "epoch": 4.855525767053917,
      "grad_norm": 10.0625,
      "learning_rate": 4.6959223606822066e-05,
      "loss": 1.1597,
      "num_input_tokens_seen": 18912096,
      "step": 32600
    },
    {
      "epoch": 4.856270479594876,
      "grad_norm": 14.0,
      "learning_rate": 4.695767025227376e-05,
      "loss": 1.0674,
      "num_input_tokens_seen": 18915104,
      "step": 32605
    },
    {
      "epoch": 4.857015192135836,
      "grad_norm": 10.875,
      "learning_rate": 4.6956116526772934e-05,
      "loss": 1.125,
      "num_input_tokens_seen": 18917920,
      "step": 32610
    },
    {
      "epoch": 4.8577599046767945,
      "grad_norm": 6.625,
      "learning_rate": 4.6954562430345825e-05,
      "loss": 1.0455,
      "num_input_tokens_seen": 18921120,
      "step": 32615
    },
    {
      "epoch": 4.858504617217754,
      "grad_norm": 6.21875,
      "learning_rate": 4.695300796301871e-05,
      "loss": 1.5237,
      "num_input_tokens_seen": 18924256,
      "step": 32620
    },
    {
      "epoch": 4.859249329758713,
      "grad_norm": 15.375,
      "learning_rate": 4.695145312481783e-05,
      "loss": 0.9315,
      "num_input_tokens_seen": 18927168,
      "step": 32625
    },
    {
      "epoch": 4.859994042299673,
      "grad_norm": 16.875,
      "learning_rate": 4.694989791576946e-05,
      "loss": 1.6171,
      "num_input_tokens_seen": 18930080,
      "step": 32630
    },
    {
      "epoch": 4.860738754840631,
      "grad_norm": 13.25,
      "learning_rate": 4.6948342335899874e-05,
      "loss": 1.5368,
      "num_input_tokens_seen": 18933120,
      "step": 32635
    },
    {
      "epoch": 4.861483467381591,
      "grad_norm": 12.1875,
      "learning_rate": 4.694678638523535e-05,
      "loss": 1.0752,
      "num_input_tokens_seen": 18935904,
      "step": 32640
    },
    {
      "epoch": 4.86222817992255,
      "grad_norm": 12.9375,
      "learning_rate": 4.694523006380218e-05,
      "loss": 1.312,
      "num_input_tokens_seen": 18939104,
      "step": 32645
    },
    {
      "epoch": 4.862972892463509,
      "grad_norm": 8.625,
      "learning_rate": 4.694367337162665e-05,
      "loss": 0.5982,
      "num_input_tokens_seen": 18941696,
      "step": 32650
    },
    {
      "epoch": 4.863717605004468,
      "grad_norm": 7.25,
      "learning_rate": 4.694211630873506e-05,
      "loss": 1.2263,
      "num_input_tokens_seen": 18944800,
      "step": 32655
    },
    {
      "epoch": 4.864462317545428,
      "grad_norm": 10.25,
      "learning_rate": 4.694055887515372e-05,
      "loss": 1.1097,
      "num_input_tokens_seen": 18947936,
      "step": 32660
    },
    {
      "epoch": 4.8652070300863866,
      "grad_norm": 12.8125,
      "learning_rate": 4.6939001070908925e-05,
      "loss": 1.1096,
      "num_input_tokens_seen": 18950912,
      "step": 32665
    },
    {
      "epoch": 4.865951742627346,
      "grad_norm": 15.5,
      "learning_rate": 4.693744289602702e-05,
      "loss": 1.1578,
      "num_input_tokens_seen": 18953760,
      "step": 32670
    },
    {
      "epoch": 4.866696455168305,
      "grad_norm": 9.5,
      "learning_rate": 4.693588435053431e-05,
      "loss": 1.2836,
      "num_input_tokens_seen": 18956768,
      "step": 32675
    },
    {
      "epoch": 4.867441167709265,
      "grad_norm": 6.9375,
      "learning_rate": 4.693432543445712e-05,
      "loss": 1.0681,
      "num_input_tokens_seen": 18959584,
      "step": 32680
    },
    {
      "epoch": 4.868185880250223,
      "grad_norm": 11.375,
      "learning_rate": 4.6932766147821804e-05,
      "loss": 1.329,
      "num_input_tokens_seen": 18962208,
      "step": 32685
    },
    {
      "epoch": 4.868930592791182,
      "grad_norm": 7.15625,
      "learning_rate": 4.693120649065469e-05,
      "loss": 1.0398,
      "num_input_tokens_seen": 18965408,
      "step": 32690
    },
    {
      "epoch": 4.869675305332142,
      "grad_norm": 7.375,
      "learning_rate": 4.6929646462982135e-05,
      "loss": 0.9506,
      "num_input_tokens_seen": 18968160,
      "step": 32695
    },
    {
      "epoch": 4.870420017873101,
      "grad_norm": 12.8125,
      "learning_rate": 4.692808606483049e-05,
      "loss": 0.9761,
      "num_input_tokens_seen": 18971552,
      "step": 32700
    },
    {
      "epoch": 4.87116473041406,
      "grad_norm": 18.0,
      "learning_rate": 4.692652529622612e-05,
      "loss": 1.2074,
      "num_input_tokens_seen": 18974688,
      "step": 32705
    },
    {
      "epoch": 4.871909442955019,
      "grad_norm": 10.875,
      "learning_rate": 4.692496415719539e-05,
      "loss": 1.2121,
      "num_input_tokens_seen": 18977504,
      "step": 32710
    },
    {
      "epoch": 4.872654155495979,
      "grad_norm": 9.0,
      "learning_rate": 4.692340264776467e-05,
      "loss": 0.8141,
      "num_input_tokens_seen": 18980288,
      "step": 32715
    },
    {
      "epoch": 4.873398868036938,
      "grad_norm": 11.5625,
      "learning_rate": 4.6921840767960346e-05,
      "loss": 1.2991,
      "num_input_tokens_seen": 18982976,
      "step": 32720
    },
    {
      "epoch": 4.874143580577897,
      "grad_norm": 11.75,
      "learning_rate": 4.69202785178088e-05,
      "loss": 1.5026,
      "num_input_tokens_seen": 18985856,
      "step": 32725
    },
    {
      "epoch": 4.874888293118856,
      "grad_norm": 5.21875,
      "learning_rate": 4.6918715897336434e-05,
      "loss": 1.1623,
      "num_input_tokens_seen": 18988608,
      "step": 32730
    },
    {
      "epoch": 4.875633005659815,
      "grad_norm": 11.75,
      "learning_rate": 4.691715290656964e-05,
      "loss": 1.27,
      "num_input_tokens_seen": 18991200,
      "step": 32735
    },
    {
      "epoch": 4.876377718200774,
      "grad_norm": 10.625,
      "learning_rate": 4.6915589545534814e-05,
      "loss": 1.1763,
      "num_input_tokens_seen": 18994272,
      "step": 32740
    },
    {
      "epoch": 4.877122430741734,
      "grad_norm": 7.3125,
      "learning_rate": 4.691402581425839e-05,
      "loss": 0.9632,
      "num_input_tokens_seen": 18997152,
      "step": 32745
    },
    {
      "epoch": 4.8778671432826926,
      "grad_norm": 28.625,
      "learning_rate": 4.691246171276676e-05,
      "loss": 1.5666,
      "num_input_tokens_seen": 19000096,
      "step": 32750
    },
    {
      "epoch": 4.878611855823652,
      "grad_norm": 11.8125,
      "learning_rate": 4.691089724108636e-05,
      "loss": 1.1818,
      "num_input_tokens_seen": 19003424,
      "step": 32755
    },
    {
      "epoch": 4.879356568364611,
      "grad_norm": 15.1875,
      "learning_rate": 4.6909332399243636e-05,
      "loss": 1.2352,
      "num_input_tokens_seen": 19006304,
      "step": 32760
    },
    {
      "epoch": 4.880101280905571,
      "grad_norm": 10.4375,
      "learning_rate": 4.690776718726499e-05,
      "loss": 1.1631,
      "num_input_tokens_seen": 19009248,
      "step": 32765
    },
    {
      "epoch": 4.880845993446529,
      "grad_norm": 17.625,
      "learning_rate": 4.690620160517689e-05,
      "loss": 1.7169,
      "num_input_tokens_seen": 19012000,
      "step": 32770
    },
    {
      "epoch": 4.881590705987489,
      "grad_norm": 10.0625,
      "learning_rate": 4.690463565300579e-05,
      "loss": 1.1791,
      "num_input_tokens_seen": 19014816,
      "step": 32775
    },
    {
      "epoch": 4.882335418528448,
      "grad_norm": 8.625,
      "learning_rate": 4.690306933077811e-05,
      "loss": 0.8503,
      "num_input_tokens_seen": 19017664,
      "step": 32780
    },
    {
      "epoch": 4.883080131069407,
      "grad_norm": 8.0625,
      "learning_rate": 4.6901502638520355e-05,
      "loss": 1.0484,
      "num_input_tokens_seen": 19020256,
      "step": 32785
    },
    {
      "epoch": 4.883824843610366,
      "grad_norm": 11.6875,
      "learning_rate": 4.689993557625897e-05,
      "loss": 1.1066,
      "num_input_tokens_seen": 19023264,
      "step": 32790
    },
    {
      "epoch": 4.884569556151326,
      "grad_norm": 7.625,
      "learning_rate": 4.689836814402042e-05,
      "loss": 0.9301,
      "num_input_tokens_seen": 19025984,
      "step": 32795
    },
    {
      "epoch": 4.885314268692285,
      "grad_norm": 13.6875,
      "learning_rate": 4.689680034183121e-05,
      "loss": 1.4774,
      "num_input_tokens_seen": 19028672,
      "step": 32800
    },
    {
      "epoch": 4.886058981233244,
      "grad_norm": 7.8125,
      "learning_rate": 4.689523216971781e-05,
      "loss": 1.3179,
      "num_input_tokens_seen": 19031456,
      "step": 32805
    },
    {
      "epoch": 4.886803693774203,
      "grad_norm": 5.53125,
      "learning_rate": 4.689366362770671e-05,
      "loss": 0.7466,
      "num_input_tokens_seen": 19034240,
      "step": 32810
    },
    {
      "epoch": 4.887548406315163,
      "grad_norm": 8.875,
      "learning_rate": 4.689209471582442e-05,
      "loss": 1.2846,
      "num_input_tokens_seen": 19036928,
      "step": 32815
    },
    {
      "epoch": 4.888293118856121,
      "grad_norm": 5.9375,
      "learning_rate": 4.689052543409743e-05,
      "loss": 0.9011,
      "num_input_tokens_seen": 19039584,
      "step": 32820
    },
    {
      "epoch": 4.889037831397081,
      "grad_norm": 5.625,
      "learning_rate": 4.6888955782552274e-05,
      "loss": 0.9291,
      "num_input_tokens_seen": 19042560,
      "step": 32825
    },
    {
      "epoch": 4.88978254393804,
      "grad_norm": 16.625,
      "learning_rate": 4.688738576121545e-05,
      "loss": 1.1236,
      "num_input_tokens_seen": 19045408,
      "step": 32830
    },
    {
      "epoch": 4.890527256478999,
      "grad_norm": 6.75,
      "learning_rate": 4.68858153701135e-05,
      "loss": 0.9803,
      "num_input_tokens_seen": 19048224,
      "step": 32835
    },
    {
      "epoch": 4.891271969019958,
      "grad_norm": 8.4375,
      "learning_rate": 4.688424460927293e-05,
      "loss": 1.1855,
      "num_input_tokens_seen": 19051168,
      "step": 32840
    },
    {
      "epoch": 4.892016681560918,
      "grad_norm": 11.9375,
      "learning_rate": 4.688267347872029e-05,
      "loss": 1.1105,
      "num_input_tokens_seen": 19054080,
      "step": 32845
    },
    {
      "epoch": 4.892761394101877,
      "grad_norm": 16.375,
      "learning_rate": 4.6881101978482124e-05,
      "loss": 0.9176,
      "num_input_tokens_seen": 19056768,
      "step": 32850
    },
    {
      "epoch": 4.893506106642836,
      "grad_norm": 11.8125,
      "learning_rate": 4.687953010858498e-05,
      "loss": 1.1837,
      "num_input_tokens_seen": 19059584,
      "step": 32855
    },
    {
      "epoch": 4.894250819183795,
      "grad_norm": 10.4375,
      "learning_rate": 4.6877957869055414e-05,
      "loss": 1.0204,
      "num_input_tokens_seen": 19062304,
      "step": 32860
    },
    {
      "epoch": 4.894995531724755,
      "grad_norm": 9.125,
      "learning_rate": 4.6876385259919984e-05,
      "loss": 1.1716,
      "num_input_tokens_seen": 19065184,
      "step": 32865
    },
    {
      "epoch": 4.895740244265713,
      "grad_norm": 14.8125,
      "learning_rate": 4.687481228120526e-05,
      "loss": 1.2818,
      "num_input_tokens_seen": 19068256,
      "step": 32870
    },
    {
      "epoch": 4.896484956806672,
      "grad_norm": 5.53125,
      "learning_rate": 4.687323893293781e-05,
      "loss": 0.9211,
      "num_input_tokens_seen": 19071104,
      "step": 32875
    },
    {
      "epoch": 4.897229669347632,
      "grad_norm": 14.5,
      "learning_rate": 4.687166521514423e-05,
      "loss": 1.3833,
      "num_input_tokens_seen": 19074144,
      "step": 32880
    },
    {
      "epoch": 4.8979743818885915,
      "grad_norm": 7.21875,
      "learning_rate": 4.687009112785109e-05,
      "loss": 1.4602,
      "num_input_tokens_seen": 19077184,
      "step": 32885
    },
    {
      "epoch": 4.89871909442955,
      "grad_norm": 9.1875,
      "learning_rate": 4.686851667108499e-05,
      "loss": 1.1745,
      "num_input_tokens_seen": 19080128,
      "step": 32890
    },
    {
      "epoch": 4.899463806970509,
      "grad_norm": 5.875,
      "learning_rate": 4.686694184487253e-05,
      "loss": 0.9601,
      "num_input_tokens_seen": 19083104,
      "step": 32895
    },
    {
      "epoch": 4.900208519511469,
      "grad_norm": 15.5,
      "learning_rate": 4.68653666492403e-05,
      "loss": 1.2494,
      "num_input_tokens_seen": 19085856,
      "step": 32900
    },
    {
      "epoch": 4.900953232052427,
      "grad_norm": 10.375,
      "learning_rate": 4.686379108421493e-05,
      "loss": 1.1571,
      "num_input_tokens_seen": 19088704,
      "step": 32905
    },
    {
      "epoch": 4.901697944593387,
      "grad_norm": 13.3125,
      "learning_rate": 4.686221514982303e-05,
      "loss": 1.1794,
      "num_input_tokens_seen": 19091776,
      "step": 32910
    },
    {
      "epoch": 4.902442657134346,
      "grad_norm": 17.625,
      "learning_rate": 4.686063884609122e-05,
      "loss": 1.2417,
      "num_input_tokens_seen": 19094976,
      "step": 32915
    },
    {
      "epoch": 4.903187369675305,
      "grad_norm": 12.125,
      "learning_rate": 4.685906217304615e-05,
      "loss": 1.105,
      "num_input_tokens_seen": 19097856,
      "step": 32920
    },
    {
      "epoch": 4.903932082216264,
      "grad_norm": 13.3125,
      "learning_rate": 4.685748513071443e-05,
      "loss": 1.0343,
      "num_input_tokens_seen": 19100672,
      "step": 32925
    },
    {
      "epoch": 4.904676794757224,
      "grad_norm": 10.875,
      "learning_rate": 4.685590771912272e-05,
      "loss": 1.0942,
      "num_input_tokens_seen": 19103680,
      "step": 32930
    },
    {
      "epoch": 4.905421507298183,
      "grad_norm": 13.625,
      "learning_rate": 4.685432993829765e-05,
      "loss": 1.0889,
      "num_input_tokens_seen": 19106464,
      "step": 32935
    },
    {
      "epoch": 4.906166219839142,
      "grad_norm": 8.125,
      "learning_rate": 4.6852751788265895e-05,
      "loss": 1.2101,
      "num_input_tokens_seen": 19109632,
      "step": 32940
    },
    {
      "epoch": 4.906910932380101,
      "grad_norm": 9.375,
      "learning_rate": 4.6851173269054116e-05,
      "loss": 1.1189,
      "num_input_tokens_seen": 19112416,
      "step": 32945
    },
    {
      "epoch": 4.907655644921061,
      "grad_norm": 10.875,
      "learning_rate": 4.6849594380688966e-05,
      "loss": 1.2586,
      "num_input_tokens_seen": 19115552,
      "step": 32950
    },
    {
      "epoch": 4.908400357462019,
      "grad_norm": 7.8125,
      "learning_rate": 4.684801512319712e-05,
      "loss": 1.1307,
      "num_input_tokens_seen": 19118368,
      "step": 32955
    },
    {
      "epoch": 4.909145070002979,
      "grad_norm": 5.5,
      "learning_rate": 4.6846435496605275e-05,
      "loss": 1.0164,
      "num_input_tokens_seen": 19121184,
      "step": 32960
    },
    {
      "epoch": 4.909889782543938,
      "grad_norm": 4.1875,
      "learning_rate": 4.6844855500940096e-05,
      "loss": 0.8694,
      "num_input_tokens_seen": 19123968,
      "step": 32965
    },
    {
      "epoch": 4.9106344950848975,
      "grad_norm": 8.125,
      "learning_rate": 4.684327513622829e-05,
      "loss": 0.9616,
      "num_input_tokens_seen": 19126976,
      "step": 32970
    },
    {
      "epoch": 4.911379207625856,
      "grad_norm": 8.125,
      "learning_rate": 4.684169440249656e-05,
      "loss": 0.7865,
      "num_input_tokens_seen": 19130016,
      "step": 32975
    },
    {
      "epoch": 4.912123920166816,
      "grad_norm": 6.03125,
      "learning_rate": 4.684011329977159e-05,
      "loss": 0.9382,
      "num_input_tokens_seen": 19132928,
      "step": 32980
    },
    {
      "epoch": 4.912868632707775,
      "grad_norm": 18.5,
      "learning_rate": 4.6838531828080104e-05,
      "loss": 1.2112,
      "num_input_tokens_seen": 19136192,
      "step": 32985
    },
    {
      "epoch": 4.913613345248734,
      "grad_norm": 7.3125,
      "learning_rate": 4.6836949987448824e-05,
      "loss": 1.0254,
      "num_input_tokens_seen": 19139296,
      "step": 32990
    },
    {
      "epoch": 4.914358057789693,
      "grad_norm": 10.1875,
      "learning_rate": 4.6835367777904466e-05,
      "loss": 1.1043,
      "num_input_tokens_seen": 19142240,
      "step": 32995
    },
    {
      "epoch": 4.915102770330653,
      "grad_norm": 6.9375,
      "learning_rate": 4.6833785199473756e-05,
      "loss": 1.0063,
      "num_input_tokens_seen": 19145184,
      "step": 33000
    },
    {
      "epoch": 4.915847482871611,
      "grad_norm": 7.96875,
      "learning_rate": 4.683220225218344e-05,
      "loss": 1.0442,
      "num_input_tokens_seen": 19148032,
      "step": 33005
    },
    {
      "epoch": 4.916592195412571,
      "grad_norm": 14.8125,
      "learning_rate": 4.683061893606026e-05,
      "loss": 0.9081,
      "num_input_tokens_seen": 19150720,
      "step": 33010
    },
    {
      "epoch": 4.91733690795353,
      "grad_norm": 7.34375,
      "learning_rate": 4.682903525113096e-05,
      "loss": 1.4265,
      "num_input_tokens_seen": 19153408,
      "step": 33015
    },
    {
      "epoch": 4.9180816204944895,
      "grad_norm": 13.75,
      "learning_rate": 4.682745119742229e-05,
      "loss": 1.0292,
      "num_input_tokens_seen": 19156544,
      "step": 33020
    },
    {
      "epoch": 4.918826333035448,
      "grad_norm": 13.1875,
      "learning_rate": 4.682586677496102e-05,
      "loss": 1.3189,
      "num_input_tokens_seen": 19159360,
      "step": 33025
    },
    {
      "epoch": 4.919571045576408,
      "grad_norm": 11.6875,
      "learning_rate": 4.6824281983773914e-05,
      "loss": 1.0961,
      "num_input_tokens_seen": 19162272,
      "step": 33030
    },
    {
      "epoch": 4.920315758117367,
      "grad_norm": 15.1875,
      "learning_rate": 4.682269682388775e-05,
      "loss": 1.07,
      "num_input_tokens_seen": 19165024,
      "step": 33035
    },
    {
      "epoch": 4.921060470658325,
      "grad_norm": 10.75,
      "learning_rate": 4.6821111295329294e-05,
      "loss": 1.0383,
      "num_input_tokens_seen": 19167872,
      "step": 33040
    },
    {
      "epoch": 4.921805183199285,
      "grad_norm": 8.0625,
      "learning_rate": 4.681952539812534e-05,
      "loss": 1.0761,
      "num_input_tokens_seen": 19170848,
      "step": 33045
    },
    {
      "epoch": 4.922549895740245,
      "grad_norm": 7.1875,
      "learning_rate": 4.681793913230269e-05,
      "loss": 0.9838,
      "num_input_tokens_seen": 19173760,
      "step": 33050
    },
    {
      "epoch": 4.9232946082812035,
      "grad_norm": 21.5,
      "learning_rate": 4.6816352497888125e-05,
      "loss": 1.3965,
      "num_input_tokens_seen": 19176672,
      "step": 33055
    },
    {
      "epoch": 4.924039320822162,
      "grad_norm": 10.625,
      "learning_rate": 4.6814765494908465e-05,
      "loss": 1.2587,
      "num_input_tokens_seen": 19179680,
      "step": 33060
    },
    {
      "epoch": 4.924784033363122,
      "grad_norm": 13.8125,
      "learning_rate": 4.681317812339051e-05,
      "loss": 1.11,
      "num_input_tokens_seen": 19182624,
      "step": 33065
    },
    {
      "epoch": 4.9255287459040815,
      "grad_norm": 13.0625,
      "learning_rate": 4.681159038336108e-05,
      "loss": 1.0893,
      "num_input_tokens_seen": 19185344,
      "step": 33070
    },
    {
      "epoch": 4.92627345844504,
      "grad_norm": 4.09375,
      "learning_rate": 4.6810002274847e-05,
      "loss": 0.9147,
      "num_input_tokens_seen": 19188288,
      "step": 33075
    },
    {
      "epoch": 4.927018170985999,
      "grad_norm": 9.5,
      "learning_rate": 4.680841379787509e-05,
      "loss": 1.3063,
      "num_input_tokens_seen": 19191264,
      "step": 33080
    },
    {
      "epoch": 4.927762883526959,
      "grad_norm": 9.4375,
      "learning_rate": 4.6806824952472204e-05,
      "loss": 0.88,
      "num_input_tokens_seen": 19194368,
      "step": 33085
    },
    {
      "epoch": 4.928507596067917,
      "grad_norm": 10.125,
      "learning_rate": 4.6805235738665164e-05,
      "loss": 1.006,
      "num_input_tokens_seen": 19197216,
      "step": 33090
    },
    {
      "epoch": 4.929252308608877,
      "grad_norm": 6.5,
      "learning_rate": 4.680364615648084e-05,
      "loss": 0.8675,
      "num_input_tokens_seen": 19200384,
      "step": 33095
    },
    {
      "epoch": 4.929997021149836,
      "grad_norm": 11.6875,
      "learning_rate": 4.680205620594606e-05,
      "loss": 1.1349,
      "num_input_tokens_seen": 19203488,
      "step": 33100
    },
    {
      "epoch": 4.9307417336907955,
      "grad_norm": 8.3125,
      "learning_rate": 4.680046588708772e-05,
      "loss": 1.0653,
      "num_input_tokens_seen": 19206784,
      "step": 33105
    },
    {
      "epoch": 4.931486446231754,
      "grad_norm": 11.6875,
      "learning_rate": 4.679887519993265e-05,
      "loss": 1.2741,
      "num_input_tokens_seen": 19209696,
      "step": 33110
    },
    {
      "epoch": 4.932231158772714,
      "grad_norm": 9.25,
      "learning_rate": 4.679728414450774e-05,
      "loss": 0.9881,
      "num_input_tokens_seen": 19212544,
      "step": 33115
    },
    {
      "epoch": 4.932975871313673,
      "grad_norm": 10.375,
      "learning_rate": 4.679569272083987e-05,
      "loss": 0.9737,
      "num_input_tokens_seen": 19215648,
      "step": 33120
    },
    {
      "epoch": 4.933720583854632,
      "grad_norm": 5.90625,
      "learning_rate": 4.6794100928955934e-05,
      "loss": 0.8248,
      "num_input_tokens_seen": 19218208,
      "step": 33125
    },
    {
      "epoch": 4.934465296395591,
      "grad_norm": 13.375,
      "learning_rate": 4.67925087688828e-05,
      "loss": 1.1321,
      "num_input_tokens_seen": 19221376,
      "step": 33130
    },
    {
      "epoch": 4.935210008936551,
      "grad_norm": 13.625,
      "learning_rate": 4.679091624064738e-05,
      "loss": 1.0848,
      "num_input_tokens_seen": 19224160,
      "step": 33135
    },
    {
      "epoch": 4.9359547214775095,
      "grad_norm": 18.625,
      "learning_rate": 4.678932334427658e-05,
      "loss": 1.0047,
      "num_input_tokens_seen": 19227136,
      "step": 33140
    },
    {
      "epoch": 4.936699434018469,
      "grad_norm": 9.375,
      "learning_rate": 4.678773007979731e-05,
      "loss": 0.8618,
      "num_input_tokens_seen": 19229856,
      "step": 33145
    },
    {
      "epoch": 4.937444146559428,
      "grad_norm": 14.1875,
      "learning_rate": 4.678613644723649e-05,
      "loss": 1.1825,
      "num_input_tokens_seen": 19232576,
      "step": 33150
    },
    {
      "epoch": 4.9381888591003875,
      "grad_norm": 11.5625,
      "learning_rate": 4.6784542446621026e-05,
      "loss": 0.9248,
      "num_input_tokens_seen": 19235488,
      "step": 33155
    },
    {
      "epoch": 4.938933571641346,
      "grad_norm": 11.75,
      "learning_rate": 4.678294807797786e-05,
      "loss": 0.8683,
      "num_input_tokens_seen": 19238304,
      "step": 33160
    },
    {
      "epoch": 4.939678284182306,
      "grad_norm": 8.9375,
      "learning_rate": 4.6781353341333926e-05,
      "loss": 1.0517,
      "num_input_tokens_seen": 19241216,
      "step": 33165
    },
    {
      "epoch": 4.940422996723265,
      "grad_norm": 9.5625,
      "learning_rate": 4.6779758236716165e-05,
      "loss": 1.0143,
      "num_input_tokens_seen": 19244064,
      "step": 33170
    },
    {
      "epoch": 4.941167709264224,
      "grad_norm": 9.125,
      "learning_rate": 4.677816276415153e-05,
      "loss": 1.3829,
      "num_input_tokens_seen": 19246784,
      "step": 33175
    },
    {
      "epoch": 4.941912421805183,
      "grad_norm": 11.75,
      "learning_rate": 4.677656692366696e-05,
      "loss": 0.9172,
      "num_input_tokens_seen": 19249696,
      "step": 33180
    },
    {
      "epoch": 4.942657134346143,
      "grad_norm": 4.46875,
      "learning_rate": 4.677497071528944e-05,
      "loss": 0.97,
      "num_input_tokens_seen": 19252704,
      "step": 33185
    },
    {
      "epoch": 4.9434018468871015,
      "grad_norm": 8.8125,
      "learning_rate": 4.67733741390459e-05,
      "loss": 1.1841,
      "num_input_tokens_seen": 19255424,
      "step": 33190
    },
    {
      "epoch": 4.944146559428061,
      "grad_norm": 19.0,
      "learning_rate": 4.677177719496335e-05,
      "loss": 1.0014,
      "num_input_tokens_seen": 19258400,
      "step": 33195
    },
    {
      "epoch": 4.94489127196902,
      "grad_norm": 7.21875,
      "learning_rate": 4.677017988306874e-05,
      "loss": 0.9255,
      "num_input_tokens_seen": 19261216,
      "step": 33200
    },
    {
      "epoch": 4.945635984509979,
      "grad_norm": 9.0,
      "learning_rate": 4.676858220338908e-05,
      "loss": 1.1264,
      "num_input_tokens_seen": 19264256,
      "step": 33205
    },
    {
      "epoch": 4.946380697050938,
      "grad_norm": 9.25,
      "learning_rate": 4.676698415595134e-05,
      "loss": 1.4472,
      "num_input_tokens_seen": 19267360,
      "step": 33210
    },
    {
      "epoch": 4.947125409591898,
      "grad_norm": 7.875,
      "learning_rate": 4.676538574078253e-05,
      "loss": 0.9545,
      "num_input_tokens_seen": 19270208,
      "step": 33215
    },
    {
      "epoch": 4.947870122132857,
      "grad_norm": 11.4375,
      "learning_rate": 4.676378695790964e-05,
      "loss": 0.905,
      "num_input_tokens_seen": 19273056,
      "step": 33220
    },
    {
      "epoch": 4.9486148346738155,
      "grad_norm": 4.4375,
      "learning_rate": 4.67621878073597e-05,
      "loss": 1.3302,
      "num_input_tokens_seen": 19276000,
      "step": 33225
    },
    {
      "epoch": 4.949359547214775,
      "grad_norm": 5.15625,
      "learning_rate": 4.676058828915971e-05,
      "loss": 1.0345,
      "num_input_tokens_seen": 19278752,
      "step": 33230
    },
    {
      "epoch": 4.950104259755735,
      "grad_norm": 15.875,
      "learning_rate": 4.67589884033367e-05,
      "loss": 0.8804,
      "num_input_tokens_seen": 19281632,
      "step": 33235
    },
    {
      "epoch": 4.9508489722966935,
      "grad_norm": 11.5,
      "learning_rate": 4.675738814991769e-05,
      "loss": 1.2613,
      "num_input_tokens_seen": 19284736,
      "step": 33240
    },
    {
      "epoch": 4.951593684837652,
      "grad_norm": 9.6875,
      "learning_rate": 4.6755787528929726e-05,
      "loss": 1.5332,
      "num_input_tokens_seen": 19287616,
      "step": 33245
    },
    {
      "epoch": 4.952338397378612,
      "grad_norm": 6.375,
      "learning_rate": 4.675418654039984e-05,
      "loss": 1.0187,
      "num_input_tokens_seen": 19290304,
      "step": 33250
    },
    {
      "epoch": 4.953083109919571,
      "grad_norm": 11.125,
      "learning_rate": 4.6752585184355084e-05,
      "loss": 1.1155,
      "num_input_tokens_seen": 19293568,
      "step": 33255
    },
    {
      "epoch": 4.95382782246053,
      "grad_norm": 18.0,
      "learning_rate": 4.675098346082251e-05,
      "loss": 1.2437,
      "num_input_tokens_seen": 19296704,
      "step": 33260
    },
    {
      "epoch": 4.954572535001489,
      "grad_norm": 6.6875,
      "learning_rate": 4.674938136982918e-05,
      "loss": 1.0299,
      "num_input_tokens_seen": 19299680,
      "step": 33265
    },
    {
      "epoch": 4.955317247542449,
      "grad_norm": 5.84375,
      "learning_rate": 4.674777891140215e-05,
      "loss": 1.1374,
      "num_input_tokens_seen": 19302528,
      "step": 33270
    },
    {
      "epoch": 4.9560619600834075,
      "grad_norm": 7.1875,
      "learning_rate": 4.6746176085568506e-05,
      "loss": 1.1549,
      "num_input_tokens_seen": 19305696,
      "step": 33275
    },
    {
      "epoch": 4.956806672624367,
      "grad_norm": 8.125,
      "learning_rate": 4.674457289235531e-05,
      "loss": 1.0066,
      "num_input_tokens_seen": 19308544,
      "step": 33280
    },
    {
      "epoch": 4.957551385165326,
      "grad_norm": 10.8125,
      "learning_rate": 4.674296933178967e-05,
      "loss": 1.2484,
      "num_input_tokens_seen": 19311680,
      "step": 33285
    },
    {
      "epoch": 4.9582960977062855,
      "grad_norm": 8.1875,
      "learning_rate": 4.674136540389864e-05,
      "loss": 0.6662,
      "num_input_tokens_seen": 19314592,
      "step": 33290
    },
    {
      "epoch": 4.959040810247244,
      "grad_norm": 9.5625,
      "learning_rate": 4.6739761108709356e-05,
      "loss": 1.1517,
      "num_input_tokens_seen": 19317440,
      "step": 33295
    },
    {
      "epoch": 4.959785522788204,
      "grad_norm": 17.5,
      "learning_rate": 4.673815644624889e-05,
      "loss": 1.3554,
      "num_input_tokens_seen": 19320512,
      "step": 33300
    },
    {
      "epoch": 4.960530235329163,
      "grad_norm": 12.375,
      "learning_rate": 4.673655141654438e-05,
      "loss": 0.9811,
      "num_input_tokens_seen": 19323488,
      "step": 33305
    },
    {
      "epoch": 4.961274947870122,
      "grad_norm": 9.8125,
      "learning_rate": 4.673494601962292e-05,
      "loss": 1.0562,
      "num_input_tokens_seen": 19326272,
      "step": 33310
    },
    {
      "epoch": 4.962019660411081,
      "grad_norm": 7.65625,
      "learning_rate": 4.673334025551164e-05,
      "loss": 1.1289,
      "num_input_tokens_seen": 19329248,
      "step": 33315
    },
    {
      "epoch": 4.962764372952041,
      "grad_norm": 5.5625,
      "learning_rate": 4.6731734124237654e-05,
      "loss": 0.7481,
      "num_input_tokens_seen": 19332352,
      "step": 33320
    },
    {
      "epoch": 4.9635090854929995,
      "grad_norm": 18.125,
      "learning_rate": 4.6730127625828113e-05,
      "loss": 1.2127,
      "num_input_tokens_seen": 19335168,
      "step": 33325
    },
    {
      "epoch": 4.964253798033959,
      "grad_norm": 13.1875,
      "learning_rate": 4.672852076031015e-05,
      "loss": 1.348,
      "num_input_tokens_seen": 19337984,
      "step": 33330
    },
    {
      "epoch": 4.964998510574918,
      "grad_norm": 11.0,
      "learning_rate": 4.6726913527710915e-05,
      "loss": 1.3096,
      "num_input_tokens_seen": 19340896,
      "step": 33335
    },
    {
      "epoch": 4.965743223115878,
      "grad_norm": 10.25,
      "learning_rate": 4.672530592805756e-05,
      "loss": 0.9961,
      "num_input_tokens_seen": 19343520,
      "step": 33340
    },
    {
      "epoch": 4.966487935656836,
      "grad_norm": 18.625,
      "learning_rate": 4.672369796137724e-05,
      "loss": 1.0993,
      "num_input_tokens_seen": 19346432,
      "step": 33345
    },
    {
      "epoch": 4.967232648197796,
      "grad_norm": 7.5,
      "learning_rate": 4.672208962769713e-05,
      "loss": 0.8983,
      "num_input_tokens_seen": 19349312,
      "step": 33350
    },
    {
      "epoch": 4.967977360738755,
      "grad_norm": 7.15625,
      "learning_rate": 4.672048092704438e-05,
      "loss": 1.077,
      "num_input_tokens_seen": 19352128,
      "step": 33355
    },
    {
      "epoch": 4.968722073279714,
      "grad_norm": 15.25,
      "learning_rate": 4.671887185944618e-05,
      "loss": 1.0722,
      "num_input_tokens_seen": 19355264,
      "step": 33360
    },
    {
      "epoch": 4.969466785820673,
      "grad_norm": 9.375,
      "learning_rate": 4.671726242492972e-05,
      "loss": 1.1402,
      "num_input_tokens_seen": 19358368,
      "step": 33365
    },
    {
      "epoch": 4.970211498361633,
      "grad_norm": 14.3125,
      "learning_rate": 4.671565262352219e-05,
      "loss": 1.1924,
      "num_input_tokens_seen": 19361344,
      "step": 33370
    },
    {
      "epoch": 4.9709562109025915,
      "grad_norm": 19.375,
      "learning_rate": 4.671404245525077e-05,
      "loss": 0.9911,
      "num_input_tokens_seen": 19364224,
      "step": 33375
    },
    {
      "epoch": 4.971700923443551,
      "grad_norm": 9.0625,
      "learning_rate": 4.671243192014267e-05,
      "loss": 0.9343,
      "num_input_tokens_seen": 19367168,
      "step": 33380
    },
    {
      "epoch": 4.97244563598451,
      "grad_norm": 10.9375,
      "learning_rate": 4.6710821018225104e-05,
      "loss": 0.9213,
      "num_input_tokens_seen": 19370144,
      "step": 33385
    },
    {
      "epoch": 4.973190348525469,
      "grad_norm": 7.0625,
      "learning_rate": 4.670920974952529e-05,
      "loss": 1.1747,
      "num_input_tokens_seen": 19372992,
      "step": 33390
    },
    {
      "epoch": 4.973935061066428,
      "grad_norm": 5.84375,
      "learning_rate": 4.6707598114070436e-05,
      "loss": 1.0545,
      "num_input_tokens_seen": 19375648,
      "step": 33395
    },
    {
      "epoch": 4.974679773607388,
      "grad_norm": 6.28125,
      "learning_rate": 4.6705986111887765e-05,
      "loss": 0.7803,
      "num_input_tokens_seen": 19378432,
      "step": 33400
    },
    {
      "epoch": 4.975424486148347,
      "grad_norm": 13.4375,
      "learning_rate": 4.6704373743004534e-05,
      "loss": 1.1148,
      "num_input_tokens_seen": 19381472,
      "step": 33405
    },
    {
      "epoch": 4.9761691986893055,
      "grad_norm": 15.3125,
      "learning_rate": 4.670276100744796e-05,
      "loss": 1.2466,
      "num_input_tokens_seen": 19384384,
      "step": 33410
    },
    {
      "epoch": 4.976913911230265,
      "grad_norm": 13.875,
      "learning_rate": 4.67011479052453e-05,
      "loss": 1.0562,
      "num_input_tokens_seen": 19387200,
      "step": 33415
    },
    {
      "epoch": 4.977658623771224,
      "grad_norm": 7.25,
      "learning_rate": 4.66995344364238e-05,
      "loss": 0.9402,
      "num_input_tokens_seen": 19390080,
      "step": 33420
    },
    {
      "epoch": 4.978403336312184,
      "grad_norm": 19.125,
      "learning_rate": 4.6697920601010724e-05,
      "loss": 1.2736,
      "num_input_tokens_seen": 19392896,
      "step": 33425
    },
    {
      "epoch": 4.979148048853142,
      "grad_norm": 6.3125,
      "learning_rate": 4.669630639903333e-05,
      "loss": 0.9673,
      "num_input_tokens_seen": 19395776,
      "step": 33430
    },
    {
      "epoch": 4.979892761394102,
      "grad_norm": 5.96875,
      "learning_rate": 4.669469183051889e-05,
      "loss": 0.9463,
      "num_input_tokens_seen": 19398816,
      "step": 33435
    },
    {
      "epoch": 4.980637473935061,
      "grad_norm": 11.75,
      "learning_rate": 4.669307689549468e-05,
      "loss": 1.0541,
      "num_input_tokens_seen": 19401760,
      "step": 33440
    },
    {
      "epoch": 4.98138218647602,
      "grad_norm": 7.9375,
      "learning_rate": 4.6691461593987985e-05,
      "loss": 1.1195,
      "num_input_tokens_seen": 19404640,
      "step": 33445
    },
    {
      "epoch": 4.982126899016979,
      "grad_norm": 14.9375,
      "learning_rate": 4.668984592602609e-05,
      "loss": 0.897,
      "num_input_tokens_seen": 19407328,
      "step": 33450
    },
    {
      "epoch": 4.982871611557939,
      "grad_norm": 12.9375,
      "learning_rate": 4.66882298916363e-05,
      "loss": 1.0183,
      "num_input_tokens_seen": 19410400,
      "step": 33455
    },
    {
      "epoch": 4.9836163240988975,
      "grad_norm": 13.75,
      "learning_rate": 4.66866134908459e-05,
      "loss": 1.1674,
      "num_input_tokens_seen": 19413536,
      "step": 33460
    },
    {
      "epoch": 4.984361036639857,
      "grad_norm": 8.875,
      "learning_rate": 4.668499672368221e-05,
      "loss": 1.2185,
      "num_input_tokens_seen": 19416384,
      "step": 33465
    },
    {
      "epoch": 4.985105749180816,
      "grad_norm": 10.25,
      "learning_rate": 4.668337959017254e-05,
      "loss": 0.8725,
      "num_input_tokens_seen": 19419200,
      "step": 33470
    },
    {
      "epoch": 4.985850461721776,
      "grad_norm": 10.625,
      "learning_rate": 4.668176209034421e-05,
      "loss": 0.9172,
      "num_input_tokens_seen": 19422464,
      "step": 33475
    },
    {
      "epoch": 4.986595174262734,
      "grad_norm": 11.8125,
      "learning_rate": 4.668014422422455e-05,
      "loss": 1.0749,
      "num_input_tokens_seen": 19425312,
      "step": 33480
    },
    {
      "epoch": 4.987339886803694,
      "grad_norm": 12.0625,
      "learning_rate": 4.6678525991840886e-05,
      "loss": 0.833,
      "num_input_tokens_seen": 19428096,
      "step": 33485
    },
    {
      "epoch": 4.988084599344653,
      "grad_norm": 10.5,
      "learning_rate": 4.667690739322055e-05,
      "loss": 1.0869,
      "num_input_tokens_seen": 19431104,
      "step": 33490
    },
    {
      "epoch": 4.988829311885612,
      "grad_norm": 10.25,
      "learning_rate": 4.667528842839091e-05,
      "loss": 0.8506,
      "num_input_tokens_seen": 19433952,
      "step": 33495
    },
    {
      "epoch": 4.989574024426571,
      "grad_norm": 17.25,
      "learning_rate": 4.6673669097379294e-05,
      "loss": 1.3396,
      "num_input_tokens_seen": 19437088,
      "step": 33500
    },
    {
      "epoch": 4.990318736967531,
      "grad_norm": 10.1875,
      "learning_rate": 4.6672049400213056e-05,
      "loss": 1.0205,
      "num_input_tokens_seen": 19440000,
      "step": 33505
    },
    {
      "epoch": 4.99106344950849,
      "grad_norm": 15.1875,
      "learning_rate": 4.6670429336919585e-05,
      "loss": 0.9801,
      "num_input_tokens_seen": 19442784,
      "step": 33510
    },
    {
      "epoch": 4.991808162049449,
      "grad_norm": 16.25,
      "learning_rate": 4.666880890752623e-05,
      "loss": 1.3242,
      "num_input_tokens_seen": 19445664,
      "step": 33515
    },
    {
      "epoch": 4.992552874590408,
      "grad_norm": 9.5,
      "learning_rate": 4.6667188112060365e-05,
      "loss": 1.021,
      "num_input_tokens_seen": 19448128,
      "step": 33520
    },
    {
      "epoch": 4.993297587131368,
      "grad_norm": 9.3125,
      "learning_rate": 4.666556695054939e-05,
      "loss": 1.3122,
      "num_input_tokens_seen": 19451040,
      "step": 33525
    },
    {
      "epoch": 4.994042299672326,
      "grad_norm": 17.5,
      "learning_rate": 4.666394542302068e-05,
      "loss": 1.3188,
      "num_input_tokens_seen": 19453888,
      "step": 33530
    },
    {
      "epoch": 4.994787012213286,
      "grad_norm": 10.125,
      "learning_rate": 4.6662323529501625e-05,
      "loss": 0.9788,
      "num_input_tokens_seen": 19457056,
      "step": 33535
    },
    {
      "epoch": 4.995531724754245,
      "grad_norm": 12.1875,
      "learning_rate": 4.666070127001963e-05,
      "loss": 1.0832,
      "num_input_tokens_seen": 19459936,
      "step": 33540
    },
    {
      "epoch": 4.996276437295204,
      "grad_norm": 11.875,
      "learning_rate": 4.6659078644602103e-05,
      "loss": 1.2741,
      "num_input_tokens_seen": 19462848,
      "step": 33545
    },
    {
      "epoch": 4.997021149836163,
      "grad_norm": 14.75,
      "learning_rate": 4.665745565327646e-05,
      "loss": 1.2834,
      "num_input_tokens_seen": 19465696,
      "step": 33550
    },
    {
      "epoch": 4.997765862377122,
      "grad_norm": 11.6875,
      "learning_rate": 4.665583229607011e-05,
      "loss": 0.7386,
      "num_input_tokens_seen": 19468576,
      "step": 33555
    },
    {
      "epoch": 4.998510574918082,
      "grad_norm": 7.5,
      "learning_rate": 4.6654208573010484e-05,
      "loss": 1.1448,
      "num_input_tokens_seen": 19471424,
      "step": 33560
    },
    {
      "epoch": 4.999255287459041,
      "grad_norm": 9.375,
      "learning_rate": 4.665258448412502e-05,
      "loss": 1.0741,
      "num_input_tokens_seen": 19474208,
      "step": 33565
    },
    {
      "epoch": 5.0,
      "grad_norm": 10.1875,
      "learning_rate": 4.665096002944114e-05,
      "loss": 1.201,
      "num_input_tokens_seen": 19476576,
      "step": 33570
    },
    {
      "epoch": 5.0,
      "eval_loss": 1.087706446647644,
      "eval_runtime": 43.7535,
      "eval_samples_per_second": 68.2,
      "eval_steps_per_second": 17.05,
      "num_input_tokens_seen": 19476576,
      "step": 33570
    },
    {
      "epoch": 5.000744712540959,
      "grad_norm": 10.1875,
      "learning_rate": 4.6649335208986294e-05,
      "loss": 1.1532,
      "num_input_tokens_seen": 19479616,
      "step": 33575
    },
    {
      "epoch": 5.001489425081918,
      "grad_norm": 10.0625,
      "learning_rate": 4.6647710022787935e-05,
      "loss": 0.9242,
      "num_input_tokens_seen": 19482464,
      "step": 33580
    },
    {
      "epoch": 5.002234137622877,
      "grad_norm": 8.4375,
      "learning_rate": 4.664608447087352e-05,
      "loss": 1.1353,
      "num_input_tokens_seen": 19485504,
      "step": 33585
    },
    {
      "epoch": 5.002978850163837,
      "grad_norm": 7.53125,
      "learning_rate": 4.664445855327051e-05,
      "loss": 1.0966,
      "num_input_tokens_seen": 19488544,
      "step": 33590
    },
    {
      "epoch": 5.003723562704796,
      "grad_norm": 13.5625,
      "learning_rate": 4.664283227000636e-05,
      "loss": 1.5195,
      "num_input_tokens_seen": 19491520,
      "step": 33595
    },
    {
      "epoch": 5.004468275245755,
      "grad_norm": 6.84375,
      "learning_rate": 4.664120562110857e-05,
      "loss": 1.1756,
      "num_input_tokens_seen": 19494336,
      "step": 33600
    },
    {
      "epoch": 5.005212987786714,
      "grad_norm": 13.625,
      "learning_rate": 4.6639578606604596e-05,
      "loss": 0.8733,
      "num_input_tokens_seen": 19497248,
      "step": 33605
    },
    {
      "epoch": 5.005957700327674,
      "grad_norm": 14.6875,
      "learning_rate": 4.6637951226521935e-05,
      "loss": 1.2068,
      "num_input_tokens_seen": 19500288,
      "step": 33610
    },
    {
      "epoch": 5.006702412868632,
      "grad_norm": 9.875,
      "learning_rate": 4.663632348088809e-05,
      "loss": 1.1022,
      "num_input_tokens_seen": 19503488,
      "step": 33615
    },
    {
      "epoch": 5.007447125409592,
      "grad_norm": 6.4375,
      "learning_rate": 4.663469536973054e-05,
      "loss": 0.9857,
      "num_input_tokens_seen": 19506432,
      "step": 33620
    },
    {
      "epoch": 5.008191837950551,
      "grad_norm": 9.5,
      "learning_rate": 4.6633066893076804e-05,
      "loss": 0.8349,
      "num_input_tokens_seen": 19509248,
      "step": 33625
    },
    {
      "epoch": 5.00893655049151,
      "grad_norm": 8.0,
      "learning_rate": 4.663143805095439e-05,
      "loss": 1.087,
      "num_input_tokens_seen": 19511936,
      "step": 33630
    },
    {
      "epoch": 5.009681263032469,
      "grad_norm": 8.9375,
      "learning_rate": 4.662980884339081e-05,
      "loss": 1.0698,
      "num_input_tokens_seen": 19514784,
      "step": 33635
    },
    {
      "epoch": 5.010425975573429,
      "grad_norm": 11.9375,
      "learning_rate": 4.66281792704136e-05,
      "loss": 1.0714,
      "num_input_tokens_seen": 19517664,
      "step": 33640
    },
    {
      "epoch": 5.011170688114388,
      "grad_norm": 7.53125,
      "learning_rate": 4.6626549332050284e-05,
      "loss": 0.8366,
      "num_input_tokens_seen": 19520544,
      "step": 33645
    },
    {
      "epoch": 5.011915400655347,
      "grad_norm": 4.5625,
      "learning_rate": 4.6624919028328394e-05,
      "loss": 0.7957,
      "num_input_tokens_seen": 19523232,
      "step": 33650
    },
    {
      "epoch": 5.012660113196306,
      "grad_norm": 8.8125,
      "learning_rate": 4.6623288359275474e-05,
      "loss": 1.0876,
      "num_input_tokens_seen": 19526272,
      "step": 33655
    },
    {
      "epoch": 5.013404825737266,
      "grad_norm": 9.5625,
      "learning_rate": 4.662165732491907e-05,
      "loss": 1.093,
      "num_input_tokens_seen": 19529376,
      "step": 33660
    },
    {
      "epoch": 5.014149538278224,
      "grad_norm": 16.375,
      "learning_rate": 4.662002592528675e-05,
      "loss": 1.1999,
      "num_input_tokens_seen": 19532672,
      "step": 33665
    },
    {
      "epoch": 5.014894250819184,
      "grad_norm": 11.8125,
      "learning_rate": 4.661839416040606e-05,
      "loss": 1.6044,
      "num_input_tokens_seen": 19535456,
      "step": 33670
    },
    {
      "epoch": 5.015638963360143,
      "grad_norm": 9.9375,
      "learning_rate": 4.6616762030304576e-05,
      "loss": 1.0313,
      "num_input_tokens_seen": 19538592,
      "step": 33675
    },
    {
      "epoch": 5.0163836759011025,
      "grad_norm": 11.3125,
      "learning_rate": 4.661512953500987e-05,
      "loss": 1.1462,
      "num_input_tokens_seen": 19541632,
      "step": 33680
    },
    {
      "epoch": 5.017128388442061,
      "grad_norm": 14.5,
      "learning_rate": 4.661349667454951e-05,
      "loss": 1.2084,
      "num_input_tokens_seen": 19544928,
      "step": 33685
    },
    {
      "epoch": 5.017873100983021,
      "grad_norm": 19.375,
      "learning_rate": 4.6611863448951096e-05,
      "loss": 1.0158,
      "num_input_tokens_seen": 19547680,
      "step": 33690
    },
    {
      "epoch": 5.01861781352398,
      "grad_norm": 13.6875,
      "learning_rate": 4.661022985824222e-05,
      "loss": 0.741,
      "num_input_tokens_seen": 19550656,
      "step": 33695
    },
    {
      "epoch": 5.019362526064939,
      "grad_norm": 8.625,
      "learning_rate": 4.660859590245046e-05,
      "loss": 1.1545,
      "num_input_tokens_seen": 19553312,
      "step": 33700
    },
    {
      "epoch": 5.020107238605898,
      "grad_norm": 9.3125,
      "learning_rate": 4.6606961581603446e-05,
      "loss": 1.2336,
      "num_input_tokens_seen": 19556480,
      "step": 33705
    },
    {
      "epoch": 5.020851951146858,
      "grad_norm": 13.0,
      "learning_rate": 4.6605326895728773e-05,
      "loss": 0.9826,
      "num_input_tokens_seen": 19559264,
      "step": 33710
    },
    {
      "epoch": 5.021596663687816,
      "grad_norm": 11.5,
      "learning_rate": 4.6603691844854065e-05,
      "loss": 0.9789,
      "num_input_tokens_seen": 19561984,
      "step": 33715
    },
    {
      "epoch": 5.022341376228776,
      "grad_norm": 14.6875,
      "learning_rate": 4.660205642900693e-05,
      "loss": 1.2535,
      "num_input_tokens_seen": 19565056,
      "step": 33720
    },
    {
      "epoch": 5.023086088769735,
      "grad_norm": 14.125,
      "learning_rate": 4.660042064821501e-05,
      "loss": 1.3886,
      "num_input_tokens_seen": 19567872,
      "step": 33725
    },
    {
      "epoch": 5.0238308013106945,
      "grad_norm": 10.4375,
      "learning_rate": 4.659878450250595e-05,
      "loss": 0.9705,
      "num_input_tokens_seen": 19570784,
      "step": 33730
    },
    {
      "epoch": 5.024575513851653,
      "grad_norm": 5.5625,
      "learning_rate": 4.6597147991907365e-05,
      "loss": 0.9895,
      "num_input_tokens_seen": 19573472,
      "step": 33735
    },
    {
      "epoch": 5.025320226392613,
      "grad_norm": 12.3125,
      "learning_rate": 4.659551111644692e-05,
      "loss": 1.3287,
      "num_input_tokens_seen": 19576512,
      "step": 33740
    },
    {
      "epoch": 5.026064938933572,
      "grad_norm": 8.75,
      "learning_rate": 4.659387387615226e-05,
      "loss": 1.0722,
      "num_input_tokens_seen": 19579136,
      "step": 33745
    },
    {
      "epoch": 5.02680965147453,
      "grad_norm": 14.0625,
      "learning_rate": 4.659223627105105e-05,
      "loss": 1.5768,
      "num_input_tokens_seen": 19581824,
      "step": 33750
    },
    {
      "epoch": 5.02755436401549,
      "grad_norm": 7.6875,
      "learning_rate": 4.659059830117095e-05,
      "loss": 0.8089,
      "num_input_tokens_seen": 19584832,
      "step": 33755
    },
    {
      "epoch": 5.028299076556449,
      "grad_norm": 13.125,
      "learning_rate": 4.658895996653964e-05,
      "loss": 1.1175,
      "num_input_tokens_seen": 19587648,
      "step": 33760
    },
    {
      "epoch": 5.0290437890974085,
      "grad_norm": 8.0,
      "learning_rate": 4.658732126718479e-05,
      "loss": 1.0022,
      "num_input_tokens_seen": 19590400,
      "step": 33765
    },
    {
      "epoch": 5.029788501638367,
      "grad_norm": 12.125,
      "learning_rate": 4.6585682203134094e-05,
      "loss": 1.2492,
      "num_input_tokens_seen": 19593376,
      "step": 33770
    },
    {
      "epoch": 5.030533214179327,
      "grad_norm": 16.0,
      "learning_rate": 4.658404277441523e-05,
      "loss": 1.2124,
      "num_input_tokens_seen": 19596064,
      "step": 33775
    },
    {
      "epoch": 5.031277926720286,
      "grad_norm": 10.6875,
      "learning_rate": 4.65824029810559e-05,
      "loss": 0.9151,
      "num_input_tokens_seen": 19598976,
      "step": 33780
    },
    {
      "epoch": 5.032022639261245,
      "grad_norm": 13.875,
      "learning_rate": 4.658076282308381e-05,
      "loss": 1.0724,
      "num_input_tokens_seen": 19601856,
      "step": 33785
    },
    {
      "epoch": 5.032767351802204,
      "grad_norm": 11.0,
      "learning_rate": 4.657912230052667e-05,
      "loss": 0.9392,
      "num_input_tokens_seen": 19604608,
      "step": 33790
    },
    {
      "epoch": 5.033512064343164,
      "grad_norm": 7.09375,
      "learning_rate": 4.657748141341218e-05,
      "loss": 0.9895,
      "num_input_tokens_seen": 19608000,
      "step": 33795
    },
    {
      "epoch": 5.034256776884122,
      "grad_norm": 9.0625,
      "learning_rate": 4.657584016176808e-05,
      "loss": 1.2023,
      "num_input_tokens_seen": 19611008,
      "step": 33800
    },
    {
      "epoch": 5.035001489425082,
      "grad_norm": 16.875,
      "learning_rate": 4.657419854562208e-05,
      "loss": 1.1301,
      "num_input_tokens_seen": 19614368,
      "step": 33805
    },
    {
      "epoch": 5.035746201966041,
      "grad_norm": 11.25,
      "learning_rate": 4.657255656500193e-05,
      "loss": 0.8767,
      "num_input_tokens_seen": 19617120,
      "step": 33810
    },
    {
      "epoch": 5.0364909145070005,
      "grad_norm": 12.5625,
      "learning_rate": 4.657091421993536e-05,
      "loss": 0.956,
      "num_input_tokens_seen": 19620160,
      "step": 33815
    },
    {
      "epoch": 5.037235627047959,
      "grad_norm": 3.859375,
      "learning_rate": 4.656927151045012e-05,
      "loss": 1.0228,
      "num_input_tokens_seen": 19622976,
      "step": 33820
    },
    {
      "epoch": 5.037980339588919,
      "grad_norm": 11.625,
      "learning_rate": 4.656762843657396e-05,
      "loss": 1.0252,
      "num_input_tokens_seen": 19625952,
      "step": 33825
    },
    {
      "epoch": 5.038725052129878,
      "grad_norm": 12.6875,
      "learning_rate": 4.656598499833463e-05,
      "loss": 1.1021,
      "num_input_tokens_seen": 19628576,
      "step": 33830
    },
    {
      "epoch": 5.039469764670837,
      "grad_norm": 11.8125,
      "learning_rate": 4.6564341195759915e-05,
      "loss": 0.9005,
      "num_input_tokens_seen": 19631616,
      "step": 33835
    },
    {
      "epoch": 5.040214477211796,
      "grad_norm": 11.125,
      "learning_rate": 4.656269702887757e-05,
      "loss": 1.0748,
      "num_input_tokens_seen": 19634176,
      "step": 33840
    },
    {
      "epoch": 5.040959189752756,
      "grad_norm": 13.1875,
      "learning_rate": 4.656105249771536e-05,
      "loss": 0.8913,
      "num_input_tokens_seen": 19636928,
      "step": 33845
    },
    {
      "epoch": 5.0417039022937145,
      "grad_norm": 10.0625,
      "learning_rate": 4.65594076023011e-05,
      "loss": 1.116,
      "num_input_tokens_seen": 19639744,
      "step": 33850
    },
    {
      "epoch": 5.042448614834674,
      "grad_norm": 10.1875,
      "learning_rate": 4.655776234266255e-05,
      "loss": 0.8954,
      "num_input_tokens_seen": 19642944,
      "step": 33855
    },
    {
      "epoch": 5.043193327375633,
      "grad_norm": 7.4375,
      "learning_rate": 4.655611671882752e-05,
      "loss": 0.8658,
      "num_input_tokens_seen": 19645920,
      "step": 33860
    },
    {
      "epoch": 5.0439380399165925,
      "grad_norm": 9.9375,
      "learning_rate": 4.655447073082381e-05,
      "loss": 0.7074,
      "num_input_tokens_seen": 19648576,
      "step": 33865
    },
    {
      "epoch": 5.044682752457551,
      "grad_norm": 9.625,
      "learning_rate": 4.6552824378679216e-05,
      "loss": 1.5085,
      "num_input_tokens_seen": 19651328,
      "step": 33870
    },
    {
      "epoch": 5.045427464998511,
      "grad_norm": 16.75,
      "learning_rate": 4.655117766242156e-05,
      "loss": 1.3381,
      "num_input_tokens_seen": 19653952,
      "step": 33875
    },
    {
      "epoch": 5.04617217753947,
      "grad_norm": 6.3125,
      "learning_rate": 4.654953058207866e-05,
      "loss": 1.1209,
      "num_input_tokens_seen": 19657056,
      "step": 33880
    },
    {
      "epoch": 5.046916890080429,
      "grad_norm": 9.0,
      "learning_rate": 4.654788313767835e-05,
      "loss": 1.3581,
      "num_input_tokens_seen": 19660480,
      "step": 33885
    },
    {
      "epoch": 5.047661602621388,
      "grad_norm": 9.75,
      "learning_rate": 4.654623532924845e-05,
      "loss": 0.7727,
      "num_input_tokens_seen": 19663104,
      "step": 33890
    },
    {
      "epoch": 5.048406315162348,
      "grad_norm": 5.15625,
      "learning_rate": 4.6544587156816806e-05,
      "loss": 1.1519,
      "num_input_tokens_seen": 19665888,
      "step": 33895
    },
    {
      "epoch": 5.0491510277033065,
      "grad_norm": 10.5,
      "learning_rate": 4.6542938620411256e-05,
      "loss": 0.8123,
      "num_input_tokens_seen": 19668640,
      "step": 33900
    },
    {
      "epoch": 5.049895740244266,
      "grad_norm": 10.625,
      "learning_rate": 4.654128972005966e-05,
      "loss": 1.1585,
      "num_input_tokens_seen": 19671456,
      "step": 33905
    },
    {
      "epoch": 5.050640452785225,
      "grad_norm": 16.625,
      "learning_rate": 4.653964045578986e-05,
      "loss": 1.3857,
      "num_input_tokens_seen": 19674400,
      "step": 33910
    },
    {
      "epoch": 5.0513851653261845,
      "grad_norm": 12.4375,
      "learning_rate": 4.6537990827629726e-05,
      "loss": 1.0696,
      "num_input_tokens_seen": 19677312,
      "step": 33915
    },
    {
      "epoch": 5.052129877867143,
      "grad_norm": 15.8125,
      "learning_rate": 4.653634083560713e-05,
      "loss": 1.5085,
      "num_input_tokens_seen": 19680544,
      "step": 33920
    },
    {
      "epoch": 5.052874590408102,
      "grad_norm": 9.375,
      "learning_rate": 4.653469047974994e-05,
      "loss": 0.9581,
      "num_input_tokens_seen": 19683264,
      "step": 33925
    },
    {
      "epoch": 5.053619302949062,
      "grad_norm": 8.9375,
      "learning_rate": 4.653303976008604e-05,
      "loss": 0.9256,
      "num_input_tokens_seen": 19686080,
      "step": 33930
    },
    {
      "epoch": 5.0543640154900205,
      "grad_norm": 14.0,
      "learning_rate": 4.6531388676643325e-05,
      "loss": 0.9672,
      "num_input_tokens_seen": 19688896,
      "step": 33935
    },
    {
      "epoch": 5.05510872803098,
      "grad_norm": 9.5625,
      "learning_rate": 4.6529737229449676e-05,
      "loss": 0.7706,
      "num_input_tokens_seen": 19691616,
      "step": 33940
    },
    {
      "epoch": 5.055853440571939,
      "grad_norm": 13.375,
      "learning_rate": 4.6528085418533004e-05,
      "loss": 1.1776,
      "num_input_tokens_seen": 19694784,
      "step": 33945
    },
    {
      "epoch": 5.0565981531128985,
      "grad_norm": 5.875,
      "learning_rate": 4.652643324392121e-05,
      "loss": 0.9502,
      "num_input_tokens_seen": 19697888,
      "step": 33950
    },
    {
      "epoch": 5.057342865653857,
      "grad_norm": 5.90625,
      "learning_rate": 4.65247807056422e-05,
      "loss": 0.8479,
      "num_input_tokens_seen": 19700992,
      "step": 33955
    },
    {
      "epoch": 5.058087578194817,
      "grad_norm": 9.75,
      "learning_rate": 4.65231278037239e-05,
      "loss": 1.4116,
      "num_input_tokens_seen": 19703872,
      "step": 33960
    },
    {
      "epoch": 5.058832290735776,
      "grad_norm": 11.6875,
      "learning_rate": 4.652147453819423e-05,
      "loss": 1.1875,
      "num_input_tokens_seen": 19706944,
      "step": 33965
    },
    {
      "epoch": 5.059577003276735,
      "grad_norm": 11.625,
      "learning_rate": 4.651982090908112e-05,
      "loss": 0.7163,
      "num_input_tokens_seen": 19709664,
      "step": 33970
    },
    {
      "epoch": 5.060321715817694,
      "grad_norm": 10.5,
      "learning_rate": 4.6518166916412506e-05,
      "loss": 0.9347,
      "num_input_tokens_seen": 19712416,
      "step": 33975
    },
    {
      "epoch": 5.061066428358654,
      "grad_norm": 7.03125,
      "learning_rate": 4.651651256021634e-05,
      "loss": 1.0606,
      "num_input_tokens_seen": 19715360,
      "step": 33980
    },
    {
      "epoch": 5.0618111408996125,
      "grad_norm": 7.0625,
      "learning_rate": 4.651485784052055e-05,
      "loss": 1.0197,
      "num_input_tokens_seen": 19718240,
      "step": 33985
    },
    {
      "epoch": 5.062555853440572,
      "grad_norm": 5.9375,
      "learning_rate": 4.6513202757353116e-05,
      "loss": 0.7679,
      "num_input_tokens_seen": 19721216,
      "step": 33990
    },
    {
      "epoch": 5.063300565981531,
      "grad_norm": 15.4375,
      "learning_rate": 4.6511547310741984e-05,
      "loss": 0.9937,
      "num_input_tokens_seen": 19724064,
      "step": 33995
    },
    {
      "epoch": 5.0640452785224905,
      "grad_norm": 14.6875,
      "learning_rate": 4.650989150071512e-05,
      "loss": 1.1947,
      "num_input_tokens_seen": 19726880,
      "step": 34000
    },
    {
      "epoch": 5.064789991063449,
      "grad_norm": 12.8125,
      "learning_rate": 4.6508235327300496e-05,
      "loss": 1.3426,
      "num_input_tokens_seen": 19729664,
      "step": 34005
    },
    {
      "epoch": 5.065534703604409,
      "grad_norm": 5.1875,
      "learning_rate": 4.65065787905261e-05,
      "loss": 1.1801,
      "num_input_tokens_seen": 19732704,
      "step": 34010
    },
    {
      "epoch": 5.066279416145368,
      "grad_norm": 10.3125,
      "learning_rate": 4.650492189041992e-05,
      "loss": 1.0541,
      "num_input_tokens_seen": 19735488,
      "step": 34015
    },
    {
      "epoch": 5.067024128686327,
      "grad_norm": 12.5625,
      "learning_rate": 4.650326462700993e-05,
      "loss": 0.9046,
      "num_input_tokens_seen": 19738144,
      "step": 34020
    },
    {
      "epoch": 5.067768841227286,
      "grad_norm": 7.25,
      "learning_rate": 4.650160700032416e-05,
      "loss": 1.1987,
      "num_input_tokens_seen": 19741152,
      "step": 34025
    },
    {
      "epoch": 5.068513553768246,
      "grad_norm": 27.125,
      "learning_rate": 4.649994901039057e-05,
      "loss": 1.0686,
      "num_input_tokens_seen": 19744352,
      "step": 34030
    },
    {
      "epoch": 5.0692582663092045,
      "grad_norm": 10.8125,
      "learning_rate": 4.6498290657237205e-05,
      "loss": 1.0628,
      "num_input_tokens_seen": 19747136,
      "step": 34035
    },
    {
      "epoch": 5.070002978850164,
      "grad_norm": 15.1875,
      "learning_rate": 4.649663194089207e-05,
      "loss": 1.1932,
      "num_input_tokens_seen": 19749984,
      "step": 34040
    },
    {
      "epoch": 5.070747691391123,
      "grad_norm": 4.15625,
      "learning_rate": 4.649497286138318e-05,
      "loss": 1.0263,
      "num_input_tokens_seen": 19753184,
      "step": 34045
    },
    {
      "epoch": 5.071492403932083,
      "grad_norm": 12.9375,
      "learning_rate": 4.6493313418738564e-05,
      "loss": 1.1331,
      "num_input_tokens_seen": 19755776,
      "step": 34050
    },
    {
      "epoch": 5.072237116473041,
      "grad_norm": 6.3125,
      "learning_rate": 4.649165361298628e-05,
      "loss": 1.4401,
      "num_input_tokens_seen": 19758496,
      "step": 34055
    },
    {
      "epoch": 5.072981829014001,
      "grad_norm": 12.6875,
      "learning_rate": 4.6489993444154334e-05,
      "loss": 1.0331,
      "num_input_tokens_seen": 19761440,
      "step": 34060
    },
    {
      "epoch": 5.07372654155496,
      "grad_norm": 6.0,
      "learning_rate": 4.64883329122708e-05,
      "loss": 0.7886,
      "num_input_tokens_seen": 19764160,
      "step": 34065
    },
    {
      "epoch": 5.074471254095919,
      "grad_norm": 18.75,
      "learning_rate": 4.648667201736372e-05,
      "loss": 1.1113,
      "num_input_tokens_seen": 19767072,
      "step": 34070
    },
    {
      "epoch": 5.075215966636878,
      "grad_norm": 15.0,
      "learning_rate": 4.648501075946116e-05,
      "loss": 1.1082,
      "num_input_tokens_seen": 19770208,
      "step": 34075
    },
    {
      "epoch": 5.075960679177838,
      "grad_norm": 10.5625,
      "learning_rate": 4.648334913859117e-05,
      "loss": 1.511,
      "num_input_tokens_seen": 19773056,
      "step": 34080
    },
    {
      "epoch": 5.0767053917187965,
      "grad_norm": 13.5625,
      "learning_rate": 4.648168715478183e-05,
      "loss": 1.453,
      "num_input_tokens_seen": 19776448,
      "step": 34085
    },
    {
      "epoch": 5.077450104259755,
      "grad_norm": 10.75,
      "learning_rate": 4.648002480806123e-05,
      "loss": 1.2246,
      "num_input_tokens_seen": 19779296,
      "step": 34090
    },
    {
      "epoch": 5.078194816800715,
      "grad_norm": 6.9375,
      "learning_rate": 4.647836209845744e-05,
      "loss": 1.0347,
      "num_input_tokens_seen": 19782208,
      "step": 34095
    },
    {
      "epoch": 5.078939529341674,
      "grad_norm": 9.375,
      "learning_rate": 4.647669902599854e-05,
      "loss": 0.9844,
      "num_input_tokens_seen": 19785024,
      "step": 34100
    },
    {
      "epoch": 5.079684241882633,
      "grad_norm": 9.25,
      "learning_rate": 4.6475035590712646e-05,
      "loss": 0.901,
      "num_input_tokens_seen": 19788064,
      "step": 34105
    },
    {
      "epoch": 5.080428954423592,
      "grad_norm": 9.3125,
      "learning_rate": 4.6473371792627854e-05,
      "loss": 0.8963,
      "num_input_tokens_seen": 19790816,
      "step": 34110
    },
    {
      "epoch": 5.081173666964552,
      "grad_norm": 11.0625,
      "learning_rate": 4.6471707631772267e-05,
      "loss": 0.9523,
      "num_input_tokens_seen": 19793760,
      "step": 34115
    },
    {
      "epoch": 5.0819183795055105,
      "grad_norm": 7.90625,
      "learning_rate": 4.6470043108174e-05,
      "loss": 1.4892,
      "num_input_tokens_seen": 19796768,
      "step": 34120
    },
    {
      "epoch": 5.08266309204647,
      "grad_norm": 12.875,
      "learning_rate": 4.6468378221861175e-05,
      "loss": 1.0625,
      "num_input_tokens_seen": 19799456,
      "step": 34125
    },
    {
      "epoch": 5.083407804587429,
      "grad_norm": 10.625,
      "learning_rate": 4.646671297286193e-05,
      "loss": 0.9187,
      "num_input_tokens_seen": 19802464,
      "step": 34130
    },
    {
      "epoch": 5.084152517128389,
      "grad_norm": 22.75,
      "learning_rate": 4.646504736120438e-05,
      "loss": 1.2484,
      "num_input_tokens_seen": 19805088,
      "step": 34135
    },
    {
      "epoch": 5.084897229669347,
      "grad_norm": 8.4375,
      "learning_rate": 4.646338138691667e-05,
      "loss": 0.8842,
      "num_input_tokens_seen": 19808000,
      "step": 34140
    },
    {
      "epoch": 5.085641942210307,
      "grad_norm": 10.875,
      "learning_rate": 4.646171505002694e-05,
      "loss": 0.9619,
      "num_input_tokens_seen": 19811072,
      "step": 34145
    },
    {
      "epoch": 5.086386654751266,
      "grad_norm": 10.0625,
      "learning_rate": 4.646004835056336e-05,
      "loss": 0.9935,
      "num_input_tokens_seen": 19814112,
      "step": 34150
    },
    {
      "epoch": 5.087131367292225,
      "grad_norm": 11.9375,
      "learning_rate": 4.645838128855406e-05,
      "loss": 1.1744,
      "num_input_tokens_seen": 19816992,
      "step": 34155
    },
    {
      "epoch": 5.087876079833184,
      "grad_norm": 12.375,
      "learning_rate": 4.6456713864027234e-05,
      "loss": 1.0994,
      "num_input_tokens_seen": 19819872,
      "step": 34160
    },
    {
      "epoch": 5.088620792374144,
      "grad_norm": 15.1875,
      "learning_rate": 4.645504607701102e-05,
      "loss": 1.221,
      "num_input_tokens_seen": 19822848,
      "step": 34165
    },
    {
      "epoch": 5.0893655049151025,
      "grad_norm": 13.0,
      "learning_rate": 4.645337792753362e-05,
      "loss": 1.2986,
      "num_input_tokens_seen": 19825664,
      "step": 34170
    },
    {
      "epoch": 5.090110217456062,
      "grad_norm": 5.40625,
      "learning_rate": 4.64517094156232e-05,
      "loss": 1.2791,
      "num_input_tokens_seen": 19828544,
      "step": 34175
    },
    {
      "epoch": 5.090854929997021,
      "grad_norm": 17.125,
      "learning_rate": 4.645004054130795e-05,
      "loss": 1.2078,
      "num_input_tokens_seen": 19831520,
      "step": 34180
    },
    {
      "epoch": 5.091599642537981,
      "grad_norm": 11.0625,
      "learning_rate": 4.644837130461607e-05,
      "loss": 1.0185,
      "num_input_tokens_seen": 19834208,
      "step": 34185
    },
    {
      "epoch": 5.092344355078939,
      "grad_norm": 15.4375,
      "learning_rate": 4.644670170557575e-05,
      "loss": 1.2865,
      "num_input_tokens_seen": 19836992,
      "step": 34190
    },
    {
      "epoch": 5.093089067619899,
      "grad_norm": 6.78125,
      "learning_rate": 4.644503174421521e-05,
      "loss": 0.9297,
      "num_input_tokens_seen": 19839680,
      "step": 34195
    },
    {
      "epoch": 5.093833780160858,
      "grad_norm": 11.75,
      "learning_rate": 4.644336142056265e-05,
      "loss": 1.0815,
      "num_input_tokens_seen": 19842400,
      "step": 34200
    },
    {
      "epoch": 5.094578492701817,
      "grad_norm": 14.875,
      "learning_rate": 4.644169073464629e-05,
      "loss": 0.9603,
      "num_input_tokens_seen": 19845280,
      "step": 34205
    },
    {
      "epoch": 5.095323205242776,
      "grad_norm": 9.5,
      "learning_rate": 4.644001968649436e-05,
      "loss": 1.0706,
      "num_input_tokens_seen": 19848256,
      "step": 34210
    },
    {
      "epoch": 5.096067917783736,
      "grad_norm": 5.1875,
      "learning_rate": 4.643834827613508e-05,
      "loss": 1.0255,
      "num_input_tokens_seen": 19850944,
      "step": 34215
    },
    {
      "epoch": 5.096812630324695,
      "grad_norm": 6.46875,
      "learning_rate": 4.643667650359671e-05,
      "loss": 0.8638,
      "num_input_tokens_seen": 19853856,
      "step": 34220
    },
    {
      "epoch": 5.097557342865654,
      "grad_norm": 11.625,
      "learning_rate": 4.643500436890746e-05,
      "loss": 1.2535,
      "num_input_tokens_seen": 19856896,
      "step": 34225
    },
    {
      "epoch": 5.098302055406613,
      "grad_norm": 5.375,
      "learning_rate": 4.6433331872095615e-05,
      "loss": 1.0146,
      "num_input_tokens_seen": 19859808,
      "step": 34230
    },
    {
      "epoch": 5.099046767947573,
      "grad_norm": 12.625,
      "learning_rate": 4.643165901318941e-05,
      "loss": 1.2268,
      "num_input_tokens_seen": 19862752,
      "step": 34235
    },
    {
      "epoch": 5.099791480488531,
      "grad_norm": 7.21875,
      "learning_rate": 4.6429985792217095e-05,
      "loss": 0.77,
      "num_input_tokens_seen": 19865280,
      "step": 34240
    },
    {
      "epoch": 5.100536193029491,
      "grad_norm": 7.09375,
      "learning_rate": 4.642831220920696e-05,
      "loss": 0.9466,
      "num_input_tokens_seen": 19868128,
      "step": 34245
    },
    {
      "epoch": 5.10128090557045,
      "grad_norm": 15.125,
      "learning_rate": 4.642663826418726e-05,
      "loss": 1.2399,
      "num_input_tokens_seen": 19871328,
      "step": 34250
    },
    {
      "epoch": 5.102025618111409,
      "grad_norm": 12.3125,
      "learning_rate": 4.64249639571863e-05,
      "loss": 0.9856,
      "num_input_tokens_seen": 19874592,
      "step": 34255
    },
    {
      "epoch": 5.102770330652368,
      "grad_norm": 13.8125,
      "learning_rate": 4.642328928823234e-05,
      "loss": 1.3131,
      "num_input_tokens_seen": 19877728,
      "step": 34260
    },
    {
      "epoch": 5.103515043193327,
      "grad_norm": 13.625,
      "learning_rate": 4.6421614257353676e-05,
      "loss": 0.7964,
      "num_input_tokens_seen": 19880736,
      "step": 34265
    },
    {
      "epoch": 5.104259755734287,
      "grad_norm": 12.0625,
      "learning_rate": 4.6419938864578615e-05,
      "loss": 1.1017,
      "num_input_tokens_seen": 19883712,
      "step": 34270
    },
    {
      "epoch": 5.105004468275245,
      "grad_norm": 10.3125,
      "learning_rate": 4.641826310993546e-05,
      "loss": 1.0666,
      "num_input_tokens_seen": 19886720,
      "step": 34275
    },
    {
      "epoch": 5.105749180816205,
      "grad_norm": 6.71875,
      "learning_rate": 4.641658699345251e-05,
      "loss": 0.9377,
      "num_input_tokens_seen": 19889568,
      "step": 34280
    },
    {
      "epoch": 5.106493893357164,
      "grad_norm": 15.125,
      "learning_rate": 4.64149105151581e-05,
      "loss": 1.1421,
      "num_input_tokens_seen": 19892544,
      "step": 34285
    },
    {
      "epoch": 5.107238605898123,
      "grad_norm": 5.40625,
      "learning_rate": 4.641323367508054e-05,
      "loss": 1.1028,
      "num_input_tokens_seen": 19895168,
      "step": 34290
    },
    {
      "epoch": 5.107983318439082,
      "grad_norm": 7.34375,
      "learning_rate": 4.641155647324816e-05,
      "loss": 0.9395,
      "num_input_tokens_seen": 19897824,
      "step": 34295
    },
    {
      "epoch": 5.108728030980042,
      "grad_norm": 9.375,
      "learning_rate": 4.6409878909689286e-05,
      "loss": 0.9918,
      "num_input_tokens_seen": 19900480,
      "step": 34300
    },
    {
      "epoch": 5.109472743521001,
      "grad_norm": 6.53125,
      "learning_rate": 4.6408200984432276e-05,
      "loss": 0.9431,
      "num_input_tokens_seen": 19903328,
      "step": 34305
    },
    {
      "epoch": 5.11021745606196,
      "grad_norm": 6.5,
      "learning_rate": 4.640652269750547e-05,
      "loss": 1.2139,
      "num_input_tokens_seen": 19906048,
      "step": 34310
    },
    {
      "epoch": 5.110962168602919,
      "grad_norm": 4.65625,
      "learning_rate": 4.640484404893722e-05,
      "loss": 0.8617,
      "num_input_tokens_seen": 19909248,
      "step": 34315
    },
    {
      "epoch": 5.111706881143879,
      "grad_norm": 8.375,
      "learning_rate": 4.640316503875588e-05,
      "loss": 1.2299,
      "num_input_tokens_seen": 19912352,
      "step": 34320
    },
    {
      "epoch": 5.112451593684837,
      "grad_norm": 6.28125,
      "learning_rate": 4.640148566698982e-05,
      "loss": 0.9158,
      "num_input_tokens_seen": 19915168,
      "step": 34325
    },
    {
      "epoch": 5.113196306225797,
      "grad_norm": 14.375,
      "learning_rate": 4.639980593366742e-05,
      "loss": 1.3084,
      "num_input_tokens_seen": 19917888,
      "step": 34330
    },
    {
      "epoch": 5.113941018766756,
      "grad_norm": 15.0,
      "learning_rate": 4.639812583881704e-05,
      "loss": 1.4399,
      "num_input_tokens_seen": 19920640,
      "step": 34335
    },
    {
      "epoch": 5.114685731307715,
      "grad_norm": 18.375,
      "learning_rate": 4.6396445382467067e-05,
      "loss": 0.9424,
      "num_input_tokens_seen": 19923776,
      "step": 34340
    },
    {
      "epoch": 5.115430443848674,
      "grad_norm": 9.375,
      "learning_rate": 4.639476456464591e-05,
      "loss": 1.1249,
      "num_input_tokens_seen": 19926720,
      "step": 34345
    },
    {
      "epoch": 5.116175156389634,
      "grad_norm": 6.0,
      "learning_rate": 4.639308338538194e-05,
      "loss": 1.0395,
      "num_input_tokens_seen": 19929536,
      "step": 34350
    },
    {
      "epoch": 5.116919868930593,
      "grad_norm": 7.125,
      "learning_rate": 4.639140184470357e-05,
      "loss": 1.1277,
      "num_input_tokens_seen": 19932448,
      "step": 34355
    },
    {
      "epoch": 5.117664581471552,
      "grad_norm": 7.59375,
      "learning_rate": 4.638971994263921e-05,
      "loss": 0.9298,
      "num_input_tokens_seen": 19935232,
      "step": 34360
    },
    {
      "epoch": 5.118409294012511,
      "grad_norm": 18.0,
      "learning_rate": 4.6388037679217274e-05,
      "loss": 1.0809,
      "num_input_tokens_seen": 19938464,
      "step": 34365
    },
    {
      "epoch": 5.119154006553471,
      "grad_norm": 4.75,
      "learning_rate": 4.638635505446617e-05,
      "loss": 1.364,
      "num_input_tokens_seen": 19941504,
      "step": 34370
    },
    {
      "epoch": 5.119898719094429,
      "grad_norm": 11.4375,
      "learning_rate": 4.638467206841434e-05,
      "loss": 1.2078,
      "num_input_tokens_seen": 19944096,
      "step": 34375
    },
    {
      "epoch": 5.120643431635389,
      "grad_norm": 11.625,
      "learning_rate": 4.6382988721090214e-05,
      "loss": 0.7936,
      "num_input_tokens_seen": 19946880,
      "step": 34380
    },
    {
      "epoch": 5.121388144176348,
      "grad_norm": 9.375,
      "learning_rate": 4.638130501252221e-05,
      "loss": 1.1298,
      "num_input_tokens_seen": 19949824,
      "step": 34385
    },
    {
      "epoch": 5.1221328567173074,
      "grad_norm": 9.3125,
      "learning_rate": 4.6379620942738814e-05,
      "loss": 1.2923,
      "num_input_tokens_seen": 19952544,
      "step": 34390
    },
    {
      "epoch": 5.122877569258266,
      "grad_norm": 21.125,
      "learning_rate": 4.637793651176843e-05,
      "loss": 1.3913,
      "num_input_tokens_seen": 19955680,
      "step": 34395
    },
    {
      "epoch": 5.123622281799226,
      "grad_norm": 15.0625,
      "learning_rate": 4.637625171963954e-05,
      "loss": 1.2223,
      "num_input_tokens_seen": 19958880,
      "step": 34400
    },
    {
      "epoch": 5.124366994340185,
      "grad_norm": 13.8125,
      "learning_rate": 4.63745665663806e-05,
      "loss": 0.9608,
      "num_input_tokens_seen": 19961856,
      "step": 34405
    },
    {
      "epoch": 5.125111706881144,
      "grad_norm": 9.9375,
      "learning_rate": 4.63728810520201e-05,
      "loss": 0.9827,
      "num_input_tokens_seen": 19964704,
      "step": 34410
    },
    {
      "epoch": 5.125856419422103,
      "grad_norm": 9.1875,
      "learning_rate": 4.637119517658648e-05,
      "loss": 0.9312,
      "num_input_tokens_seen": 19967360,
      "step": 34415
    },
    {
      "epoch": 5.126601131963063,
      "grad_norm": 14.0625,
      "learning_rate": 4.636950894010825e-05,
      "loss": 1.3201,
      "num_input_tokens_seen": 19970496,
      "step": 34420
    },
    {
      "epoch": 5.127345844504021,
      "grad_norm": 6.96875,
      "learning_rate": 4.636782234261388e-05,
      "loss": 1.0288,
      "num_input_tokens_seen": 19973312,
      "step": 34425
    },
    {
      "epoch": 5.128090557044981,
      "grad_norm": 9.125,
      "learning_rate": 4.6366135384131866e-05,
      "loss": 1.2489,
      "num_input_tokens_seen": 19976448,
      "step": 34430
    },
    {
      "epoch": 5.12883526958594,
      "grad_norm": 17.75,
      "learning_rate": 4.6364448064690716e-05,
      "loss": 1.3464,
      "num_input_tokens_seen": 19979296,
      "step": 34435
    },
    {
      "epoch": 5.129579982126899,
      "grad_norm": 7.6875,
      "learning_rate": 4.636276038431892e-05,
      "loss": 0.9398,
      "num_input_tokens_seen": 19982240,
      "step": 34440
    },
    {
      "epoch": 5.130324694667858,
      "grad_norm": 7.90625,
      "learning_rate": 4.636107234304501e-05,
      "loss": 1.2506,
      "num_input_tokens_seen": 19985088,
      "step": 34445
    },
    {
      "epoch": 5.131069407208817,
      "grad_norm": 8.125,
      "learning_rate": 4.635938394089748e-05,
      "loss": 1.0307,
      "num_input_tokens_seen": 19987936,
      "step": 34450
    },
    {
      "epoch": 5.131814119749777,
      "grad_norm": 12.9375,
      "learning_rate": 4.635769517790488e-05,
      "loss": 0.9656,
      "num_input_tokens_seen": 19991008,
      "step": 34455
    },
    {
      "epoch": 5.132558832290735,
      "grad_norm": 8.5625,
      "learning_rate": 4.635600605409572e-05,
      "loss": 1.353,
      "num_input_tokens_seen": 19993984,
      "step": 34460
    },
    {
      "epoch": 5.133303544831695,
      "grad_norm": 7.84375,
      "learning_rate": 4.6354316569498545e-05,
      "loss": 1.1752,
      "num_input_tokens_seen": 19997120,
      "step": 34465
    },
    {
      "epoch": 5.134048257372654,
      "grad_norm": 18.625,
      "learning_rate": 4.63526267241419e-05,
      "loss": 0.8601,
      "num_input_tokens_seen": 20000192,
      "step": 34470
    },
    {
      "epoch": 5.1347929699136134,
      "grad_norm": 7.375,
      "learning_rate": 4.6350936518054325e-05,
      "loss": 1.031,
      "num_input_tokens_seen": 20003168,
      "step": 34475
    },
    {
      "epoch": 5.135537682454572,
      "grad_norm": 8.3125,
      "learning_rate": 4.634924595126437e-05,
      "loss": 0.8587,
      "num_input_tokens_seen": 20006048,
      "step": 34480
    },
    {
      "epoch": 5.136282394995532,
      "grad_norm": 12.5,
      "learning_rate": 4.6347555023800616e-05,
      "loss": 0.9189,
      "num_input_tokens_seen": 20009120,
      "step": 34485
    },
    {
      "epoch": 5.137027107536491,
      "grad_norm": 8.75,
      "learning_rate": 4.634586373569161e-05,
      "loss": 1.2505,
      "num_input_tokens_seen": 20012064,
      "step": 34490
    },
    {
      "epoch": 5.13777182007745,
      "grad_norm": 17.0,
      "learning_rate": 4.634417208696593e-05,
      "loss": 1.1055,
      "num_input_tokens_seen": 20015040,
      "step": 34495
    },
    {
      "epoch": 5.138516532618409,
      "grad_norm": 6.25,
      "learning_rate": 4.634248007765216e-05,
      "loss": 0.8843,
      "num_input_tokens_seen": 20017920,
      "step": 34500
    },
    {
      "epoch": 5.139261245159369,
      "grad_norm": 9.375,
      "learning_rate": 4.6340787707778874e-05,
      "loss": 0.8355,
      "num_input_tokens_seen": 20020672,
      "step": 34505
    },
    {
      "epoch": 5.140005957700327,
      "grad_norm": 10.3125,
      "learning_rate": 4.633909497737468e-05,
      "loss": 1.3391,
      "num_input_tokens_seen": 20023520,
      "step": 34510
    },
    {
      "epoch": 5.140750670241287,
      "grad_norm": 10.25,
      "learning_rate": 4.6337401886468156e-05,
      "loss": 1.1792,
      "num_input_tokens_seen": 20026432,
      "step": 34515
    },
    {
      "epoch": 5.141495382782246,
      "grad_norm": 5.875,
      "learning_rate": 4.633570843508792e-05,
      "loss": 1.0122,
      "num_input_tokens_seen": 20029056,
      "step": 34520
    },
    {
      "epoch": 5.1422400953232055,
      "grad_norm": 10.4375,
      "learning_rate": 4.633401462326257e-05,
      "loss": 1.2333,
      "num_input_tokens_seen": 20031872,
      "step": 34525
    },
    {
      "epoch": 5.142984807864164,
      "grad_norm": 11.75,
      "learning_rate": 4.633232045102072e-05,
      "loss": 1.3692,
      "num_input_tokens_seen": 20034560,
      "step": 34530
    },
    {
      "epoch": 5.143729520405124,
      "grad_norm": 20.375,
      "learning_rate": 4.6330625918391e-05,
      "loss": 1.4364,
      "num_input_tokens_seen": 20037472,
      "step": 34535
    },
    {
      "epoch": 5.144474232946083,
      "grad_norm": 6.625,
      "learning_rate": 4.6328931025402045e-05,
      "loss": 1.2131,
      "num_input_tokens_seen": 20040160,
      "step": 34540
    },
    {
      "epoch": 5.145218945487042,
      "grad_norm": 7.59375,
      "learning_rate": 4.6327235772082466e-05,
      "loss": 1.0037,
      "num_input_tokens_seen": 20043040,
      "step": 34545
    },
    {
      "epoch": 5.145963658028001,
      "grad_norm": 12.125,
      "learning_rate": 4.632554015846092e-05,
      "loss": 1.0095,
      "num_input_tokens_seen": 20045856,
      "step": 34550
    },
    {
      "epoch": 5.146708370568961,
      "grad_norm": 13.125,
      "learning_rate": 4.6323844184566045e-05,
      "loss": 0.9782,
      "num_input_tokens_seen": 20049056,
      "step": 34555
    },
    {
      "epoch": 5.1474530831099194,
      "grad_norm": 14.875,
      "learning_rate": 4.6322147850426504e-05,
      "loss": 1.0674,
      "num_input_tokens_seen": 20051936,
      "step": 34560
    },
    {
      "epoch": 5.148197795650879,
      "grad_norm": 8.0625,
      "learning_rate": 4.6320451156070934e-05,
      "loss": 1.0299,
      "num_input_tokens_seen": 20054688,
      "step": 34565
    },
    {
      "epoch": 5.148942508191838,
      "grad_norm": 16.75,
      "learning_rate": 4.6318754101528014e-05,
      "loss": 1.1476,
      "num_input_tokens_seen": 20057696,
      "step": 34570
    },
    {
      "epoch": 5.1496872207327975,
      "grad_norm": 9.875,
      "learning_rate": 4.631705668682641e-05,
      "loss": 1.1115,
      "num_input_tokens_seen": 20060576,
      "step": 34575
    },
    {
      "epoch": 5.150431933273756,
      "grad_norm": 20.125,
      "learning_rate": 4.631535891199481e-05,
      "loss": 1.3601,
      "num_input_tokens_seen": 20063552,
      "step": 34580
    },
    {
      "epoch": 5.151176645814716,
      "grad_norm": 16.875,
      "learning_rate": 4.6313660777061874e-05,
      "loss": 1.4167,
      "num_input_tokens_seen": 20066944,
      "step": 34585
    },
    {
      "epoch": 5.151921358355675,
      "grad_norm": 12.1875,
      "learning_rate": 4.63119622820563e-05,
      "loss": 1.1251,
      "num_input_tokens_seen": 20070112,
      "step": 34590
    },
    {
      "epoch": 5.152666070896634,
      "grad_norm": 13.4375,
      "learning_rate": 4.6310263427006786e-05,
      "loss": 1.1233,
      "num_input_tokens_seen": 20072928,
      "step": 34595
    },
    {
      "epoch": 5.153410783437593,
      "grad_norm": 11.25,
      "learning_rate": 4.6308564211942044e-05,
      "loss": 1.4157,
      "num_input_tokens_seen": 20075712,
      "step": 34600
    },
    {
      "epoch": 5.154155495978552,
      "grad_norm": 5.625,
      "learning_rate": 4.6306864636890745e-05,
      "loss": 0.8884,
      "num_input_tokens_seen": 20078432,
      "step": 34605
    },
    {
      "epoch": 5.1549002085195115,
      "grad_norm": 18.0,
      "learning_rate": 4.6305164701881634e-05,
      "loss": 1.265,
      "num_input_tokens_seen": 20081248,
      "step": 34610
    },
    {
      "epoch": 5.15564492106047,
      "grad_norm": 9.4375,
      "learning_rate": 4.6303464406943416e-05,
      "loss": 0.9472,
      "num_input_tokens_seen": 20084256,
      "step": 34615
    },
    {
      "epoch": 5.15638963360143,
      "grad_norm": 8.5,
      "learning_rate": 4.630176375210482e-05,
      "loss": 1.3838,
      "num_input_tokens_seen": 20086976,
      "step": 34620
    },
    {
      "epoch": 5.157134346142389,
      "grad_norm": 12.0,
      "learning_rate": 4.630006273739458e-05,
      "loss": 1.2042,
      "num_input_tokens_seen": 20089824,
      "step": 34625
    },
    {
      "epoch": 5.157879058683348,
      "grad_norm": 11.6875,
      "learning_rate": 4.629836136284143e-05,
      "loss": 1.1753,
      "num_input_tokens_seen": 20092640,
      "step": 34630
    },
    {
      "epoch": 5.158623771224307,
      "grad_norm": 13.1875,
      "learning_rate": 4.62966596284741e-05,
      "loss": 1.2502,
      "num_input_tokens_seen": 20095680,
      "step": 34635
    },
    {
      "epoch": 5.159368483765267,
      "grad_norm": 13.25,
      "learning_rate": 4.629495753432136e-05,
      "loss": 1.0027,
      "num_input_tokens_seen": 20098560,
      "step": 34640
    },
    {
      "epoch": 5.1601131963062254,
      "grad_norm": 7.09375,
      "learning_rate": 4.629325508041195e-05,
      "loss": 0.9555,
      "num_input_tokens_seen": 20101536,
      "step": 34645
    },
    {
      "epoch": 5.160857908847185,
      "grad_norm": 5.6875,
      "learning_rate": 4.629155226677464e-05,
      "loss": 1.2731,
      "num_input_tokens_seen": 20104480,
      "step": 34650
    },
    {
      "epoch": 5.161602621388144,
      "grad_norm": 3.875,
      "learning_rate": 4.628984909343819e-05,
      "loss": 1.3763,
      "num_input_tokens_seen": 20107136,
      "step": 34655
    },
    {
      "epoch": 5.1623473339291035,
      "grad_norm": 19.5,
      "learning_rate": 4.6288145560431385e-05,
      "loss": 1.4789,
      "num_input_tokens_seen": 20110336,
      "step": 34660
    },
    {
      "epoch": 5.163092046470062,
      "grad_norm": 14.25,
      "learning_rate": 4.628644166778299e-05,
      "loss": 0.9427,
      "num_input_tokens_seen": 20113504,
      "step": 34665
    },
    {
      "epoch": 5.163836759011022,
      "grad_norm": 9.0625,
      "learning_rate": 4.6284737415521806e-05,
      "loss": 1.056,
      "num_input_tokens_seen": 20116544,
      "step": 34670
    },
    {
      "epoch": 5.164581471551981,
      "grad_norm": 13.875,
      "learning_rate": 4.628303280367661e-05,
      "loss": 1.1322,
      "num_input_tokens_seen": 20119424,
      "step": 34675
    },
    {
      "epoch": 5.16532618409294,
      "grad_norm": 9.375,
      "learning_rate": 4.6281327832276204e-05,
      "loss": 0.9868,
      "num_input_tokens_seen": 20122400,
      "step": 34680
    },
    {
      "epoch": 5.166070896633899,
      "grad_norm": 7.78125,
      "learning_rate": 4.627962250134939e-05,
      "loss": 0.977,
      "num_input_tokens_seen": 20125312,
      "step": 34685
    },
    {
      "epoch": 5.166815609174859,
      "grad_norm": 10.125,
      "learning_rate": 4.627791681092499e-05,
      "loss": 0.9642,
      "num_input_tokens_seen": 20128064,
      "step": 34690
    },
    {
      "epoch": 5.1675603217158175,
      "grad_norm": 7.3125,
      "learning_rate": 4.627621076103181e-05,
      "loss": 0.948,
      "num_input_tokens_seen": 20130816,
      "step": 34695
    },
    {
      "epoch": 5.168305034256777,
      "grad_norm": 17.375,
      "learning_rate": 4.627450435169868e-05,
      "loss": 1.1475,
      "num_input_tokens_seen": 20133856,
      "step": 34700
    },
    {
      "epoch": 5.169049746797736,
      "grad_norm": 8.75,
      "learning_rate": 4.627279758295441e-05,
      "loss": 1.0033,
      "num_input_tokens_seen": 20136704,
      "step": 34705
    },
    {
      "epoch": 5.1697944593386955,
      "grad_norm": 9.25,
      "learning_rate": 4.627109045482785e-05,
      "loss": 1.1196,
      "num_input_tokens_seen": 20139456,
      "step": 34710
    },
    {
      "epoch": 5.170539171879654,
      "grad_norm": 9.625,
      "learning_rate": 4.626938296734784e-05,
      "loss": 1.0215,
      "num_input_tokens_seen": 20142144,
      "step": 34715
    },
    {
      "epoch": 5.171283884420614,
      "grad_norm": 7.3125,
      "learning_rate": 4.626767512054321e-05,
      "loss": 1.3788,
      "num_input_tokens_seen": 20144896,
      "step": 34720
    },
    {
      "epoch": 5.172028596961573,
      "grad_norm": 8.1875,
      "learning_rate": 4.626596691444284e-05,
      "loss": 0.9728,
      "num_input_tokens_seen": 20147488,
      "step": 34725
    },
    {
      "epoch": 5.172773309502532,
      "grad_norm": 6.09375,
      "learning_rate": 4.626425834907556e-05,
      "loss": 1.0633,
      "num_input_tokens_seen": 20150528,
      "step": 34730
    },
    {
      "epoch": 5.173518022043491,
      "grad_norm": 12.375,
      "learning_rate": 4.6262549424470253e-05,
      "loss": 0.8866,
      "num_input_tokens_seen": 20153664,
      "step": 34735
    },
    {
      "epoch": 5.174262734584451,
      "grad_norm": 6.71875,
      "learning_rate": 4.626084014065578e-05,
      "loss": 0.9068,
      "num_input_tokens_seen": 20156128,
      "step": 34740
    },
    {
      "epoch": 5.1750074471254095,
      "grad_norm": 14.0,
      "learning_rate": 4.625913049766103e-05,
      "loss": 1.0949,
      "num_input_tokens_seen": 20159008,
      "step": 34745
    },
    {
      "epoch": 5.175752159666369,
      "grad_norm": 11.4375,
      "learning_rate": 4.625742049551487e-05,
      "loss": 1.033,
      "num_input_tokens_seen": 20162080,
      "step": 34750
    },
    {
      "epoch": 5.176496872207328,
      "grad_norm": 13.125,
      "learning_rate": 4.6255710134246197e-05,
      "loss": 1.172,
      "num_input_tokens_seen": 20165088,
      "step": 34755
    },
    {
      "epoch": 5.1772415847482876,
      "grad_norm": 14.5625,
      "learning_rate": 4.6253999413883905e-05,
      "loss": 1.3212,
      "num_input_tokens_seen": 20167904,
      "step": 34760
    },
    {
      "epoch": 5.177986297289246,
      "grad_norm": 11.75,
      "learning_rate": 4.6252288334456887e-05,
      "loss": 1.1052,
      "num_input_tokens_seen": 20170880,
      "step": 34765
    },
    {
      "epoch": 5.178731009830206,
      "grad_norm": 13.4375,
      "learning_rate": 4.625057689599407e-05,
      "loss": 1.3483,
      "num_input_tokens_seen": 20173568,
      "step": 34770
    },
    {
      "epoch": 5.179475722371165,
      "grad_norm": 12.0,
      "learning_rate": 4.6248865098524346e-05,
      "loss": 0.8945,
      "num_input_tokens_seen": 20176320,
      "step": 34775
    },
    {
      "epoch": 5.180220434912124,
      "grad_norm": 9.0625,
      "learning_rate": 4.6247152942076646e-05,
      "loss": 1.062,
      "num_input_tokens_seen": 20179040,
      "step": 34780
    },
    {
      "epoch": 5.180965147453083,
      "grad_norm": 15.1875,
      "learning_rate": 4.624544042667989e-05,
      "loss": 1.4312,
      "num_input_tokens_seen": 20181920,
      "step": 34785
    },
    {
      "epoch": 5.181709859994042,
      "grad_norm": 11.0625,
      "learning_rate": 4.624372755236301e-05,
      "loss": 1.1526,
      "num_input_tokens_seen": 20184576,
      "step": 34790
    },
    {
      "epoch": 5.1824545725350015,
      "grad_norm": 15.5625,
      "learning_rate": 4.624201431915495e-05,
      "loss": 1.2756,
      "num_input_tokens_seen": 20187840,
      "step": 34795
    },
    {
      "epoch": 5.18319928507596,
      "grad_norm": 9.625,
      "learning_rate": 4.624030072708464e-05,
      "loss": 0.9695,
      "num_input_tokens_seen": 20190944,
      "step": 34800
    },
    {
      "epoch": 5.18394399761692,
      "grad_norm": 17.5,
      "learning_rate": 4.623858677618104e-05,
      "loss": 0.9668,
      "num_input_tokens_seen": 20194016,
      "step": 34805
    },
    {
      "epoch": 5.184688710157879,
      "grad_norm": 8.4375,
      "learning_rate": 4.62368724664731e-05,
      "loss": 0.9521,
      "num_input_tokens_seen": 20197024,
      "step": 34810
    },
    {
      "epoch": 5.185433422698838,
      "grad_norm": 10.5,
      "learning_rate": 4.623515779798979e-05,
      "loss": 1.3525,
      "num_input_tokens_seen": 20199968,
      "step": 34815
    },
    {
      "epoch": 5.186178135239797,
      "grad_norm": 8.5625,
      "learning_rate": 4.623344277076007e-05,
      "loss": 1.2444,
      "num_input_tokens_seen": 20202688,
      "step": 34820
    },
    {
      "epoch": 5.186922847780757,
      "grad_norm": 15.8125,
      "learning_rate": 4.6231727384812916e-05,
      "loss": 1.0619,
      "num_input_tokens_seen": 20205600,
      "step": 34825
    },
    {
      "epoch": 5.1876675603217155,
      "grad_norm": 10.3125,
      "learning_rate": 4.62300116401773e-05,
      "loss": 1.032,
      "num_input_tokens_seen": 20208288,
      "step": 34830
    },
    {
      "epoch": 5.188412272862675,
      "grad_norm": 6.25,
      "learning_rate": 4.622829553688222e-05,
      "loss": 0.6631,
      "num_input_tokens_seen": 20211136,
      "step": 34835
    },
    {
      "epoch": 5.189156985403634,
      "grad_norm": 7.625,
      "learning_rate": 4.622657907495667e-05,
      "loss": 1.0199,
      "num_input_tokens_seen": 20213920,
      "step": 34840
    },
    {
      "epoch": 5.1899016979445936,
      "grad_norm": 9.75,
      "learning_rate": 4.6224862254429623e-05,
      "loss": 0.8793,
      "num_input_tokens_seen": 20216896,
      "step": 34845
    },
    {
      "epoch": 5.190646410485552,
      "grad_norm": 18.375,
      "learning_rate": 4.622314507533011e-05,
      "loss": 1.2903,
      "num_input_tokens_seen": 20219808,
      "step": 34850
    },
    {
      "epoch": 5.191391123026512,
      "grad_norm": 5.96875,
      "learning_rate": 4.622142753768713e-05,
      "loss": 1.1327,
      "num_input_tokens_seen": 20222560,
      "step": 34855
    },
    {
      "epoch": 5.192135835567471,
      "grad_norm": 6.65625,
      "learning_rate": 4.6219709641529695e-05,
      "loss": 1.1044,
      "num_input_tokens_seen": 20225376,
      "step": 34860
    },
    {
      "epoch": 5.19288054810843,
      "grad_norm": 15.9375,
      "learning_rate": 4.621799138688684e-05,
      "loss": 1.1085,
      "num_input_tokens_seen": 20227936,
      "step": 34865
    },
    {
      "epoch": 5.193625260649389,
      "grad_norm": 9.75,
      "learning_rate": 4.6216272773787586e-05,
      "loss": 0.9485,
      "num_input_tokens_seen": 20230624,
      "step": 34870
    },
    {
      "epoch": 5.194369973190349,
      "grad_norm": 10.125,
      "learning_rate": 4.621455380226096e-05,
      "loss": 1.2601,
      "num_input_tokens_seen": 20233504,
      "step": 34875
    },
    {
      "epoch": 5.1951146857313075,
      "grad_norm": 17.75,
      "learning_rate": 4.6212834472336016e-05,
      "loss": 1.4835,
      "num_input_tokens_seen": 20236448,
      "step": 34880
    },
    {
      "epoch": 5.195859398272267,
      "grad_norm": 23.5,
      "learning_rate": 4.6211114784041784e-05,
      "loss": 1.048,
      "num_input_tokens_seen": 20239552,
      "step": 34885
    },
    {
      "epoch": 5.196604110813226,
      "grad_norm": 8.125,
      "learning_rate": 4.620939473740733e-05,
      "loss": 1.1743,
      "num_input_tokens_seen": 20242336,
      "step": 34890
    },
    {
      "epoch": 5.197348823354186,
      "grad_norm": 8.625,
      "learning_rate": 4.6207674332461716e-05,
      "loss": 1.0057,
      "num_input_tokens_seen": 20244992,
      "step": 34895
    },
    {
      "epoch": 5.198093535895144,
      "grad_norm": 10.0625,
      "learning_rate": 4.620595356923399e-05,
      "loss": 0.9988,
      "num_input_tokens_seen": 20248064,
      "step": 34900
    },
    {
      "epoch": 5.198838248436104,
      "grad_norm": 8.8125,
      "learning_rate": 4.620423244775323e-05,
      "loss": 0.7672,
      "num_input_tokens_seen": 20250976,
      "step": 34905
    },
    {
      "epoch": 5.199582960977063,
      "grad_norm": 9.625,
      "learning_rate": 4.6202510968048515e-05,
      "loss": 1.2901,
      "num_input_tokens_seen": 20253792,
      "step": 34910
    },
    {
      "epoch": 5.200327673518022,
      "grad_norm": 11.125,
      "learning_rate": 4.620078913014893e-05,
      "loss": 0.9875,
      "num_input_tokens_seen": 20257120,
      "step": 34915
    },
    {
      "epoch": 5.201072386058981,
      "grad_norm": 10.5625,
      "learning_rate": 4.619906693408357e-05,
      "loss": 0.913,
      "num_input_tokens_seen": 20260128,
      "step": 34920
    },
    {
      "epoch": 5.201817098599941,
      "grad_norm": 7.8125,
      "learning_rate": 4.61973443798815e-05,
      "loss": 0.9807,
      "num_input_tokens_seen": 20263008,
      "step": 34925
    },
    {
      "epoch": 5.2025618111408996,
      "grad_norm": 5.1875,
      "learning_rate": 4.6195621467571856e-05,
      "loss": 1.0314,
      "num_input_tokens_seen": 20265760,
      "step": 34930
    },
    {
      "epoch": 5.203306523681859,
      "grad_norm": 15.4375,
      "learning_rate": 4.619389819718371e-05,
      "loss": 1.0244,
      "num_input_tokens_seen": 20268736,
      "step": 34935
    },
    {
      "epoch": 5.204051236222818,
      "grad_norm": 6.21875,
      "learning_rate": 4.619217456874622e-05,
      "loss": 1.0523,
      "num_input_tokens_seen": 20271616,
      "step": 34940
    },
    {
      "epoch": 5.204795948763778,
      "grad_norm": 9.75,
      "learning_rate": 4.619045058228847e-05,
      "loss": 1.33,
      "num_input_tokens_seen": 20274304,
      "step": 34945
    },
    {
      "epoch": 5.205540661304736,
      "grad_norm": 6.15625,
      "learning_rate": 4.6188726237839586e-05,
      "loss": 0.9558,
      "num_input_tokens_seen": 20276864,
      "step": 34950
    },
    {
      "epoch": 5.206285373845695,
      "grad_norm": 8.0625,
      "learning_rate": 4.6187001535428716e-05,
      "loss": 1.431,
      "num_input_tokens_seen": 20280096,
      "step": 34955
    },
    {
      "epoch": 5.207030086386655,
      "grad_norm": 15.375,
      "learning_rate": 4.618527647508498e-05,
      "loss": 1.3251,
      "num_input_tokens_seen": 20283296,
      "step": 34960
    },
    {
      "epoch": 5.2077747989276135,
      "grad_norm": 16.5,
      "learning_rate": 4.618355105683754e-05,
      "loss": 1.7505,
      "num_input_tokens_seen": 20286368,
      "step": 34965
    },
    {
      "epoch": 5.208519511468573,
      "grad_norm": 3.640625,
      "learning_rate": 4.618182528071553e-05,
      "loss": 0.6226,
      "num_input_tokens_seen": 20289248,
      "step": 34970
    },
    {
      "epoch": 5.209264224009532,
      "grad_norm": 6.34375,
      "learning_rate": 4.618009914674811e-05,
      "loss": 0.8481,
      "num_input_tokens_seen": 20291872,
      "step": 34975
    },
    {
      "epoch": 5.210008936550492,
      "grad_norm": 17.75,
      "learning_rate": 4.617837265496444e-05,
      "loss": 1.4067,
      "num_input_tokens_seen": 20294944,
      "step": 34980
    },
    {
      "epoch": 5.21075364909145,
      "grad_norm": 8.8125,
      "learning_rate": 4.617664580539369e-05,
      "loss": 0.7818,
      "num_input_tokens_seen": 20297792,
      "step": 34985
    },
    {
      "epoch": 5.21149836163241,
      "grad_norm": 10.1875,
      "learning_rate": 4.6174918598065034e-05,
      "loss": 1.2314,
      "num_input_tokens_seen": 20300640,
      "step": 34990
    },
    {
      "epoch": 5.212243074173369,
      "grad_norm": 12.8125,
      "learning_rate": 4.617319103300764e-05,
      "loss": 1.6101,
      "num_input_tokens_seen": 20303648,
      "step": 34995
    },
    {
      "epoch": 5.212987786714328,
      "grad_norm": 10.5625,
      "learning_rate": 4.6171463110250725e-05,
      "loss": 0.987,
      "num_input_tokens_seen": 20306784,
      "step": 35000
    },
    {
      "epoch": 5.213732499255287,
      "grad_norm": 13.4375,
      "learning_rate": 4.616973482982344e-05,
      "loss": 0.8564,
      "num_input_tokens_seen": 20309632,
      "step": 35005
    },
    {
      "epoch": 5.214477211796247,
      "grad_norm": 9.8125,
      "learning_rate": 4.616800619175501e-05,
      "loss": 0.8847,
      "num_input_tokens_seen": 20312352,
      "step": 35010
    },
    {
      "epoch": 5.2152219243372056,
      "grad_norm": 11.75,
      "learning_rate": 4.616627719607462e-05,
      "loss": 1.1261,
      "num_input_tokens_seen": 20315552,
      "step": 35015
    },
    {
      "epoch": 5.215966636878165,
      "grad_norm": 9.8125,
      "learning_rate": 4.6164547842811494e-05,
      "loss": 1.1373,
      "num_input_tokens_seen": 20318656,
      "step": 35020
    },
    {
      "epoch": 5.216711349419124,
      "grad_norm": 14.5625,
      "learning_rate": 4.6162818131994845e-05,
      "loss": 1.3374,
      "num_input_tokens_seen": 20321440,
      "step": 35025
    },
    {
      "epoch": 5.217456061960084,
      "grad_norm": 5.65625,
      "learning_rate": 4.616108806365389e-05,
      "loss": 1.3503,
      "num_input_tokens_seen": 20324608,
      "step": 35030
    },
    {
      "epoch": 5.218200774501042,
      "grad_norm": 7.78125,
      "learning_rate": 4.6159357637817855e-05,
      "loss": 1.027,
      "num_input_tokens_seen": 20327424,
      "step": 35035
    },
    {
      "epoch": 5.218945487042002,
      "grad_norm": 5.9375,
      "learning_rate": 4.6157626854515986e-05,
      "loss": 1.0002,
      "num_input_tokens_seen": 20330336,
      "step": 35040
    },
    {
      "epoch": 5.219690199582961,
      "grad_norm": 18.125,
      "learning_rate": 4.615589571377752e-05,
      "loss": 1.3366,
      "num_input_tokens_seen": 20333376,
      "step": 35045
    },
    {
      "epoch": 5.22043491212392,
      "grad_norm": 8.8125,
      "learning_rate": 4.6154164215631685e-05,
      "loss": 1.5293,
      "num_input_tokens_seen": 20336224,
      "step": 35050
    },
    {
      "epoch": 5.221179624664879,
      "grad_norm": 17.375,
      "learning_rate": 4.615243236010775e-05,
      "loss": 1.3574,
      "num_input_tokens_seen": 20339072,
      "step": 35055
    },
    {
      "epoch": 5.221924337205839,
      "grad_norm": 6.90625,
      "learning_rate": 4.615070014723497e-05,
      "loss": 0.7741,
      "num_input_tokens_seen": 20341824,
      "step": 35060
    },
    {
      "epoch": 5.222669049746798,
      "grad_norm": 15.3125,
      "learning_rate": 4.614896757704261e-05,
      "loss": 1.1419,
      "num_input_tokens_seen": 20344864,
      "step": 35065
    },
    {
      "epoch": 5.223413762287757,
      "grad_norm": 11.25,
      "learning_rate": 4.614723464955993e-05,
      "loss": 1.1576,
      "num_input_tokens_seen": 20347808,
      "step": 35070
    },
    {
      "epoch": 5.224158474828716,
      "grad_norm": 12.6875,
      "learning_rate": 4.6145501364816226e-05,
      "loss": 0.9137,
      "num_input_tokens_seen": 20350560,
      "step": 35075
    },
    {
      "epoch": 5.224903187369676,
      "grad_norm": 9.5,
      "learning_rate": 4.614376772284075e-05,
      "loss": 1.1084,
      "num_input_tokens_seen": 20353824,
      "step": 35080
    },
    {
      "epoch": 5.225647899910634,
      "grad_norm": 8.5,
      "learning_rate": 4.6142033723662825e-05,
      "loss": 1.0846,
      "num_input_tokens_seen": 20356704,
      "step": 35085
    },
    {
      "epoch": 5.226392612451594,
      "grad_norm": 6.4375,
      "learning_rate": 4.614029936731172e-05,
      "loss": 1.004,
      "num_input_tokens_seen": 20359808,
      "step": 35090
    },
    {
      "epoch": 5.227137324992553,
      "grad_norm": 8.5,
      "learning_rate": 4.613856465381674e-05,
      "loss": 0.983,
      "num_input_tokens_seen": 20362880,
      "step": 35095
    },
    {
      "epoch": 5.227882037533512,
      "grad_norm": 10.375,
      "learning_rate": 4.6136829583207197e-05,
      "loss": 0.9156,
      "num_input_tokens_seen": 20365728,
      "step": 35100
    },
    {
      "epoch": 5.228626750074471,
      "grad_norm": 14.5625,
      "learning_rate": 4.6135094155512405e-05,
      "loss": 0.8024,
      "num_input_tokens_seen": 20368416,
      "step": 35105
    },
    {
      "epoch": 5.229371462615431,
      "grad_norm": 9.4375,
      "learning_rate": 4.613335837076168e-05,
      "loss": 1.1914,
      "num_input_tokens_seen": 20371008,
      "step": 35110
    },
    {
      "epoch": 5.23011617515639,
      "grad_norm": 18.125,
      "learning_rate": 4.6131622228984336e-05,
      "loss": 1.2433,
      "num_input_tokens_seen": 20374112,
      "step": 35115
    },
    {
      "epoch": 5.230860887697349,
      "grad_norm": 5.625,
      "learning_rate": 4.6129885730209715e-05,
      "loss": 0.8992,
      "num_input_tokens_seen": 20377056,
      "step": 35120
    },
    {
      "epoch": 5.231605600238308,
      "grad_norm": 11.625,
      "learning_rate": 4.612814887446715e-05,
      "loss": 0.9341,
      "num_input_tokens_seen": 20380064,
      "step": 35125
    },
    {
      "epoch": 5.232350312779268,
      "grad_norm": 10.75,
      "learning_rate": 4.6126411661785984e-05,
      "loss": 1.2616,
      "num_input_tokens_seen": 20383104,
      "step": 35130
    },
    {
      "epoch": 5.233095025320226,
      "grad_norm": 9.75,
      "learning_rate": 4.612467409219556e-05,
      "loss": 1.0436,
      "num_input_tokens_seen": 20386048,
      "step": 35135
    },
    {
      "epoch": 5.233839737861185,
      "grad_norm": 8.375,
      "learning_rate": 4.612293616572525e-05,
      "loss": 0.9457,
      "num_input_tokens_seen": 20388800,
      "step": 35140
    },
    {
      "epoch": 5.234584450402145,
      "grad_norm": 21.0,
      "learning_rate": 4.612119788240439e-05,
      "loss": 1.0895,
      "num_input_tokens_seen": 20391616,
      "step": 35145
    },
    {
      "epoch": 5.235329162943104,
      "grad_norm": 11.9375,
      "learning_rate": 4.6119459242262366e-05,
      "loss": 1.0673,
      "num_input_tokens_seen": 20394368,
      "step": 35150
    },
    {
      "epoch": 5.236073875484063,
      "grad_norm": 7.9375,
      "learning_rate": 4.611772024532854e-05,
      "loss": 1.3352,
      "num_input_tokens_seen": 20397344,
      "step": 35155
    },
    {
      "epoch": 5.236818588025022,
      "grad_norm": 11.0,
      "learning_rate": 4.611598089163229e-05,
      "loss": 1.3047,
      "num_input_tokens_seen": 20400576,
      "step": 35160
    },
    {
      "epoch": 5.237563300565982,
      "grad_norm": 12.125,
      "learning_rate": 4.611424118120301e-05,
      "loss": 1.0306,
      "num_input_tokens_seen": 20403680,
      "step": 35165
    },
    {
      "epoch": 5.23830801310694,
      "grad_norm": 13.375,
      "learning_rate": 4.611250111407008e-05,
      "loss": 1.1641,
      "num_input_tokens_seen": 20406464,
      "step": 35170
    },
    {
      "epoch": 5.2390527256479,
      "grad_norm": 3.90625,
      "learning_rate": 4.6110760690262907e-05,
      "loss": 1.2209,
      "num_input_tokens_seen": 20409312,
      "step": 35175
    },
    {
      "epoch": 5.239797438188859,
      "grad_norm": 5.5,
      "learning_rate": 4.610901990981088e-05,
      "loss": 1.1559,
      "num_input_tokens_seen": 20412384,
      "step": 35180
    },
    {
      "epoch": 5.240542150729818,
      "grad_norm": 4.625,
      "learning_rate": 4.6107278772743426e-05,
      "loss": 0.9295,
      "num_input_tokens_seen": 20415072,
      "step": 35185
    },
    {
      "epoch": 5.241286863270777,
      "grad_norm": 17.5,
      "learning_rate": 4.610553727908994e-05,
      "loss": 1.3339,
      "num_input_tokens_seen": 20417984,
      "step": 35190
    },
    {
      "epoch": 5.242031575811737,
      "grad_norm": 9.625,
      "learning_rate": 4.6103795428879856e-05,
      "loss": 0.7759,
      "num_input_tokens_seen": 20420864,
      "step": 35195
    },
    {
      "epoch": 5.242776288352696,
      "grad_norm": 9.6875,
      "learning_rate": 4.6102053222142595e-05,
      "loss": 1.0835,
      "num_input_tokens_seen": 20423712,
      "step": 35200
    },
    {
      "epoch": 5.243521000893655,
      "grad_norm": 17.25,
      "learning_rate": 4.61003106589076e-05,
      "loss": 0.9047,
      "num_input_tokens_seen": 20426816,
      "step": 35205
    },
    {
      "epoch": 5.244265713434614,
      "grad_norm": 9.5,
      "learning_rate": 4.6098567739204294e-05,
      "loss": 1.1957,
      "num_input_tokens_seen": 20429792,
      "step": 35210
    },
    {
      "epoch": 5.245010425975574,
      "grad_norm": 12.9375,
      "learning_rate": 4.609682446306213e-05,
      "loss": 0.9739,
      "num_input_tokens_seen": 20432576,
      "step": 35215
    },
    {
      "epoch": 5.245755138516532,
      "grad_norm": 10.125,
      "learning_rate": 4.6095080830510564e-05,
      "loss": 1.5416,
      "num_input_tokens_seen": 20435328,
      "step": 35220
    },
    {
      "epoch": 5.246499851057492,
      "grad_norm": 11.25,
      "learning_rate": 4.6093336841579044e-05,
      "loss": 0.9515,
      "num_input_tokens_seen": 20438176,
      "step": 35225
    },
    {
      "epoch": 5.247244563598451,
      "grad_norm": 10.125,
      "learning_rate": 4.609159249629704e-05,
      "loss": 1.1113,
      "num_input_tokens_seen": 20441088,
      "step": 35230
    },
    {
      "epoch": 5.2479892761394105,
      "grad_norm": 7.375,
      "learning_rate": 4.6089847794694005e-05,
      "loss": 0.9348,
      "num_input_tokens_seen": 20443808,
      "step": 35235
    },
    {
      "epoch": 5.248733988680369,
      "grad_norm": 8.3125,
      "learning_rate": 4.6088102736799445e-05,
      "loss": 1.14,
      "num_input_tokens_seen": 20446976,
      "step": 35240
    },
    {
      "epoch": 5.249478701221329,
      "grad_norm": 7.0625,
      "learning_rate": 4.60863573226428e-05,
      "loss": 1.0483,
      "num_input_tokens_seen": 20449856,
      "step": 35245
    },
    {
      "epoch": 5.250223413762288,
      "grad_norm": 19.625,
      "learning_rate": 4.6084611552253595e-05,
      "loss": 0.8637,
      "num_input_tokens_seen": 20452640,
      "step": 35250
    },
    {
      "epoch": 5.250968126303247,
      "grad_norm": 6.5,
      "learning_rate": 4.6082865425661307e-05,
      "loss": 1.5435,
      "num_input_tokens_seen": 20455584,
      "step": 35255
    },
    {
      "epoch": 5.251712838844206,
      "grad_norm": 11.3125,
      "learning_rate": 4.608111894289543e-05,
      "loss": 1.2098,
      "num_input_tokens_seen": 20458624,
      "step": 35260
    },
    {
      "epoch": 5.252457551385166,
      "grad_norm": 15.375,
      "learning_rate": 4.607937210398548e-05,
      "loss": 1.4288,
      "num_input_tokens_seen": 20461536,
      "step": 35265
    },
    {
      "epoch": 5.253202263926124,
      "grad_norm": 10.9375,
      "learning_rate": 4.607762490896096e-05,
      "loss": 1.4703,
      "num_input_tokens_seen": 20464320,
      "step": 35270
    },
    {
      "epoch": 5.253946976467084,
      "grad_norm": 13.0625,
      "learning_rate": 4.6075877357851384e-05,
      "loss": 1.0452,
      "num_input_tokens_seen": 20467200,
      "step": 35275
    },
    {
      "epoch": 5.254691689008043,
      "grad_norm": 10.8125,
      "learning_rate": 4.607412945068629e-05,
      "loss": 1.0163,
      "num_input_tokens_seen": 20470048,
      "step": 35280
    },
    {
      "epoch": 5.2554364015490025,
      "grad_norm": 5.34375,
      "learning_rate": 4.607238118749519e-05,
      "loss": 1.1438,
      "num_input_tokens_seen": 20472896,
      "step": 35285
    },
    {
      "epoch": 5.256181114089961,
      "grad_norm": 7.90625,
      "learning_rate": 4.607063256830763e-05,
      "loss": 1.0157,
      "num_input_tokens_seen": 20475776,
      "step": 35290
    },
    {
      "epoch": 5.256925826630921,
      "grad_norm": 10.625,
      "learning_rate": 4.6068883593153147e-05,
      "loss": 0.7509,
      "num_input_tokens_seen": 20479168,
      "step": 35295
    },
    {
      "epoch": 5.25767053917188,
      "grad_norm": 9.75,
      "learning_rate": 4.606713426206129e-05,
      "loss": 1.41,
      "num_input_tokens_seen": 20481984,
      "step": 35300
    },
    {
      "epoch": 5.258415251712838,
      "grad_norm": 12.125,
      "learning_rate": 4.606538457506161e-05,
      "loss": 0.7976,
      "num_input_tokens_seen": 20484864,
      "step": 35305
    },
    {
      "epoch": 5.259159964253798,
      "grad_norm": 11.125,
      "learning_rate": 4.606363453218367e-05,
      "loss": 1.2036,
      "num_input_tokens_seen": 20487712,
      "step": 35310
    },
    {
      "epoch": 5.259904676794757,
      "grad_norm": 7.84375,
      "learning_rate": 4.606188413345704e-05,
      "loss": 1.0337,
      "num_input_tokens_seen": 20490592,
      "step": 35315
    },
    {
      "epoch": 5.2606493893357165,
      "grad_norm": 7.4375,
      "learning_rate": 4.6060133378911265e-05,
      "loss": 1.3186,
      "num_input_tokens_seen": 20493824,
      "step": 35320
    },
    {
      "epoch": 5.261394101876675,
      "grad_norm": 9.5,
      "learning_rate": 4.605838226857595e-05,
      "loss": 0.8977,
      "num_input_tokens_seen": 20496704,
      "step": 35325
    },
    {
      "epoch": 5.262138814417635,
      "grad_norm": 6.375,
      "learning_rate": 4.605663080248067e-05,
      "loss": 0.9911,
      "num_input_tokens_seen": 20499584,
      "step": 35330
    },
    {
      "epoch": 5.262883526958594,
      "grad_norm": 10.75,
      "learning_rate": 4.6054878980655015e-05,
      "loss": 1.1498,
      "num_input_tokens_seen": 20502592,
      "step": 35335
    },
    {
      "epoch": 5.263628239499553,
      "grad_norm": 14.3125,
      "learning_rate": 4.605312680312858e-05,
      "loss": 1.3106,
      "num_input_tokens_seen": 20505696,
      "step": 35340
    },
    {
      "epoch": 5.264372952040512,
      "grad_norm": 9.0,
      "learning_rate": 4.605137426993096e-05,
      "loss": 0.9781,
      "num_input_tokens_seen": 20508608,
      "step": 35345
    },
    {
      "epoch": 5.265117664581472,
      "grad_norm": 13.3125,
      "learning_rate": 4.6049621381091776e-05,
      "loss": 0.8604,
      "num_input_tokens_seen": 20512000,
      "step": 35350
    },
    {
      "epoch": 5.26586237712243,
      "grad_norm": 7.8125,
      "learning_rate": 4.604786813664063e-05,
      "loss": 1.1331,
      "num_input_tokens_seen": 20514720,
      "step": 35355
    },
    {
      "epoch": 5.26660708966339,
      "grad_norm": 13.5625,
      "learning_rate": 4.6046114536607133e-05,
      "loss": 0.8047,
      "num_input_tokens_seen": 20517248,
      "step": 35360
    },
    {
      "epoch": 5.267351802204349,
      "grad_norm": 13.9375,
      "learning_rate": 4.604436058102093e-05,
      "loss": 1.0411,
      "num_input_tokens_seen": 20520096,
      "step": 35365
    },
    {
      "epoch": 5.2680965147453085,
      "grad_norm": 8.75,
      "learning_rate": 4.6042606269911645e-05,
      "loss": 1.0635,
      "num_input_tokens_seen": 20523136,
      "step": 35370
    },
    {
      "epoch": 5.268841227286267,
      "grad_norm": 6.1875,
      "learning_rate": 4.604085160330891e-05,
      "loss": 0.7853,
      "num_input_tokens_seen": 20526048,
      "step": 35375
    },
    {
      "epoch": 5.269585939827227,
      "grad_norm": 7.34375,
      "learning_rate": 4.603909658124238e-05,
      "loss": 1.1191,
      "num_input_tokens_seen": 20528928,
      "step": 35380
    },
    {
      "epoch": 5.270330652368186,
      "grad_norm": 10.3125,
      "learning_rate": 4.6037341203741686e-05,
      "loss": 0.8526,
      "num_input_tokens_seen": 20531808,
      "step": 35385
    },
    {
      "epoch": 5.271075364909145,
      "grad_norm": 6.28125,
      "learning_rate": 4.6035585470836494e-05,
      "loss": 0.9992,
      "num_input_tokens_seen": 20534464,
      "step": 35390
    },
    {
      "epoch": 5.271820077450104,
      "grad_norm": 7.875,
      "learning_rate": 4.603382938255647e-05,
      "loss": 1.0333,
      "num_input_tokens_seen": 20537376,
      "step": 35395
    },
    {
      "epoch": 5.272564789991064,
      "grad_norm": 11.9375,
      "learning_rate": 4.603207293893128e-05,
      "loss": 1.2131,
      "num_input_tokens_seen": 20540320,
      "step": 35400
    },
    {
      "epoch": 5.2733095025320225,
      "grad_norm": 12.4375,
      "learning_rate": 4.6030316139990595e-05,
      "loss": 1.1702,
      "num_input_tokens_seen": 20543264,
      "step": 35405
    },
    {
      "epoch": 5.274054215072982,
      "grad_norm": 12.6875,
      "learning_rate": 4.602855898576408e-05,
      "loss": 1.1851,
      "num_input_tokens_seen": 20545952,
      "step": 35410
    },
    {
      "epoch": 5.274798927613941,
      "grad_norm": 11.1875,
      "learning_rate": 4.6026801476281436e-05,
      "loss": 1.0636,
      "num_input_tokens_seen": 20548736,
      "step": 35415
    },
    {
      "epoch": 5.2755436401549005,
      "grad_norm": 5.1875,
      "learning_rate": 4.602504361157236e-05,
      "loss": 0.691,
      "num_input_tokens_seen": 20551392,
      "step": 35420
    },
    {
      "epoch": 5.276288352695859,
      "grad_norm": 14.5,
      "learning_rate": 4.602328539166654e-05,
      "loss": 1.0864,
      "num_input_tokens_seen": 20554048,
      "step": 35425
    },
    {
      "epoch": 5.277033065236819,
      "grad_norm": 22.625,
      "learning_rate": 4.602152681659368e-05,
      "loss": 1.3212,
      "num_input_tokens_seen": 20557088,
      "step": 35430
    },
    {
      "epoch": 5.277777777777778,
      "grad_norm": 9.8125,
      "learning_rate": 4.601976788638349e-05,
      "loss": 1.1184,
      "num_input_tokens_seen": 20560032,
      "step": 35435
    },
    {
      "epoch": 5.278522490318737,
      "grad_norm": 9.9375,
      "learning_rate": 4.601800860106568e-05,
      "loss": 0.9107,
      "num_input_tokens_seen": 20563008,
      "step": 35440
    },
    {
      "epoch": 5.279267202859696,
      "grad_norm": 10.5625,
      "learning_rate": 4.601624896066998e-05,
      "loss": 1.2885,
      "num_input_tokens_seen": 20565920,
      "step": 35445
    },
    {
      "epoch": 5.280011915400656,
      "grad_norm": 7.5625,
      "learning_rate": 4.601448896522611e-05,
      "loss": 1.0195,
      "num_input_tokens_seen": 20568768,
      "step": 35450
    },
    {
      "epoch": 5.2807566279416145,
      "grad_norm": 7.6875,
      "learning_rate": 4.601272861476381e-05,
      "loss": 1.1583,
      "num_input_tokens_seen": 20571648,
      "step": 35455
    },
    {
      "epoch": 5.281501340482574,
      "grad_norm": 12.75,
      "learning_rate": 4.601096790931282e-05,
      "loss": 1.1429,
      "num_input_tokens_seen": 20574464,
      "step": 35460
    },
    {
      "epoch": 5.282246053023533,
      "grad_norm": 10.8125,
      "learning_rate": 4.6009206848902874e-05,
      "loss": 0.9531,
      "num_input_tokens_seen": 20577216,
      "step": 35465
    },
    {
      "epoch": 5.282990765564492,
      "grad_norm": 17.5,
      "learning_rate": 4.6007445433563734e-05,
      "loss": 1.2514,
      "num_input_tokens_seen": 20579936,
      "step": 35470
    },
    {
      "epoch": 5.283735478105451,
      "grad_norm": 11.375,
      "learning_rate": 4.600568366332516e-05,
      "loss": 1.3047,
      "num_input_tokens_seen": 20582784,
      "step": 35475
    },
    {
      "epoch": 5.284480190646411,
      "grad_norm": 15.8125,
      "learning_rate": 4.60039215382169e-05,
      "loss": 1.0135,
      "num_input_tokens_seen": 20585632,
      "step": 35480
    },
    {
      "epoch": 5.28522490318737,
      "grad_norm": 15.0,
      "learning_rate": 4.6002159058268744e-05,
      "loss": 1.6158,
      "num_input_tokens_seen": 20588992,
      "step": 35485
    },
    {
      "epoch": 5.2859696157283285,
      "grad_norm": 12.4375,
      "learning_rate": 4.600039622351045e-05,
      "loss": 1.0386,
      "num_input_tokens_seen": 20591872,
      "step": 35490
    },
    {
      "epoch": 5.286714328269288,
      "grad_norm": 10.0625,
      "learning_rate": 4.59986330339718e-05,
      "loss": 1.1131,
      "num_input_tokens_seen": 20594624,
      "step": 35495
    },
    {
      "epoch": 5.287459040810247,
      "grad_norm": 15.4375,
      "learning_rate": 4.59968694896826e-05,
      "loss": 1.4072,
      "num_input_tokens_seen": 20597408,
      "step": 35500
    },
    {
      "epoch": 5.2882037533512065,
      "grad_norm": 41.0,
      "learning_rate": 4.599510559067263e-05,
      "loss": 1.097,
      "num_input_tokens_seen": 20600320,
      "step": 35505
    },
    {
      "epoch": 5.288948465892165,
      "grad_norm": 8.125,
      "learning_rate": 4.599334133697167e-05,
      "loss": 1.0731,
      "num_input_tokens_seen": 20603168,
      "step": 35510
    },
    {
      "epoch": 5.289693178433125,
      "grad_norm": 9.8125,
      "learning_rate": 4.5991576728609565e-05,
      "loss": 1.4178,
      "num_input_tokens_seen": 20606208,
      "step": 35515
    },
    {
      "epoch": 5.290437890974084,
      "grad_norm": 8.375,
      "learning_rate": 4.5989811765616094e-05,
      "loss": 0.8451,
      "num_input_tokens_seen": 20609088,
      "step": 35520
    },
    {
      "epoch": 5.291182603515043,
      "grad_norm": 5.9375,
      "learning_rate": 4.5988046448021096e-05,
      "loss": 0.8646,
      "num_input_tokens_seen": 20612064,
      "step": 35525
    },
    {
      "epoch": 5.291927316056002,
      "grad_norm": 5.84375,
      "learning_rate": 4.598628077585438e-05,
      "loss": 1.0653,
      "num_input_tokens_seen": 20614880,
      "step": 35530
    },
    {
      "epoch": 5.292672028596962,
      "grad_norm": 16.375,
      "learning_rate": 4.598451474914578e-05,
      "loss": 1.0396,
      "num_input_tokens_seen": 20617728,
      "step": 35535
    },
    {
      "epoch": 5.2934167411379205,
      "grad_norm": 18.0,
      "learning_rate": 4.598274836792513e-05,
      "loss": 1.2494,
      "num_input_tokens_seen": 20620576,
      "step": 35540
    },
    {
      "epoch": 5.29416145367888,
      "grad_norm": 10.125,
      "learning_rate": 4.5980981632222275e-05,
      "loss": 1.2788,
      "num_input_tokens_seen": 20623488,
      "step": 35545
    },
    {
      "epoch": 5.294906166219839,
      "grad_norm": 15.0625,
      "learning_rate": 4.5979214542067056e-05,
      "loss": 1.1318,
      "num_input_tokens_seen": 20626528,
      "step": 35550
    },
    {
      "epoch": 5.2956508787607985,
      "grad_norm": 14.25,
      "learning_rate": 4.597744709748933e-05,
      "loss": 1.1184,
      "num_input_tokens_seen": 20629344,
      "step": 35555
    },
    {
      "epoch": 5.296395591301757,
      "grad_norm": 8.9375,
      "learning_rate": 4.597567929851896e-05,
      "loss": 0.7189,
      "num_input_tokens_seen": 20632064,
      "step": 35560
    },
    {
      "epoch": 5.297140303842717,
      "grad_norm": 7.90625,
      "learning_rate": 4.59739111451858e-05,
      "loss": 0.9774,
      "num_input_tokens_seen": 20634816,
      "step": 35565
    },
    {
      "epoch": 5.297885016383676,
      "grad_norm": 14.5625,
      "learning_rate": 4.5972142637519735e-05,
      "loss": 1.3832,
      "num_input_tokens_seen": 20637600,
      "step": 35570
    },
    {
      "epoch": 5.298629728924635,
      "grad_norm": 11.0,
      "learning_rate": 4.597037377555063e-05,
      "loss": 1.3312,
      "num_input_tokens_seen": 20640224,
      "step": 35575
    },
    {
      "epoch": 5.299374441465594,
      "grad_norm": 22.25,
      "learning_rate": 4.5968604559308374e-05,
      "loss": 1.4438,
      "num_input_tokens_seen": 20642752,
      "step": 35580
    },
    {
      "epoch": 5.300119154006554,
      "grad_norm": 5.34375,
      "learning_rate": 4.596683498882286e-05,
      "loss": 0.9914,
      "num_input_tokens_seen": 20645600,
      "step": 35585
    },
    {
      "epoch": 5.3008638665475125,
      "grad_norm": 11.1875,
      "learning_rate": 4.596506506412398e-05,
      "loss": 0.9098,
      "num_input_tokens_seen": 20648256,
      "step": 35590
    },
    {
      "epoch": 5.301608579088472,
      "grad_norm": 5.5,
      "learning_rate": 4.596329478524163e-05,
      "loss": 0.8646,
      "num_input_tokens_seen": 20651136,
      "step": 35595
    },
    {
      "epoch": 5.302353291629431,
      "grad_norm": 10.5,
      "learning_rate": 4.596152415220572e-05,
      "loss": 0.9002,
      "num_input_tokens_seen": 20654400,
      "step": 35600
    },
    {
      "epoch": 5.303098004170391,
      "grad_norm": 7.25,
      "learning_rate": 4.595975316504616e-05,
      "loss": 1.1759,
      "num_input_tokens_seen": 20657376,
      "step": 35605
    },
    {
      "epoch": 5.303842716711349,
      "grad_norm": 13.3125,
      "learning_rate": 4.595798182379288e-05,
      "loss": 1.3435,
      "num_input_tokens_seen": 20660288,
      "step": 35610
    },
    {
      "epoch": 5.304587429252309,
      "grad_norm": 7.9375,
      "learning_rate": 4.595621012847579e-05,
      "loss": 0.941,
      "num_input_tokens_seen": 20663424,
      "step": 35615
    },
    {
      "epoch": 5.305332141793268,
      "grad_norm": 9.6875,
      "learning_rate": 4.5954438079124836e-05,
      "loss": 1.1227,
      "num_input_tokens_seen": 20666528,
      "step": 35620
    },
    {
      "epoch": 5.306076854334227,
      "grad_norm": 7.875,
      "learning_rate": 4.595266567576995e-05,
      "loss": 1.1682,
      "num_input_tokens_seen": 20669536,
      "step": 35625
    },
    {
      "epoch": 5.306821566875186,
      "grad_norm": 8.25,
      "learning_rate": 4.595089291844106e-05,
      "loss": 1.3247,
      "num_input_tokens_seen": 20672544,
      "step": 35630
    },
    {
      "epoch": 5.307566279416146,
      "grad_norm": 8.75,
      "learning_rate": 4.594911980716814e-05,
      "loss": 0.8261,
      "num_input_tokens_seen": 20675392,
      "step": 35635
    },
    {
      "epoch": 5.3083109919571045,
      "grad_norm": 8.0625,
      "learning_rate": 4.594734634198112e-05,
      "loss": 1.0423,
      "num_input_tokens_seen": 20678432,
      "step": 35640
    },
    {
      "epoch": 5.309055704498064,
      "grad_norm": 15.1875,
      "learning_rate": 4.594557252290998e-05,
      "loss": 1.1719,
      "num_input_tokens_seen": 20680992,
      "step": 35645
    },
    {
      "epoch": 5.309800417039023,
      "grad_norm": 8.4375,
      "learning_rate": 4.594379834998469e-05,
      "loss": 1.1657,
      "num_input_tokens_seen": 20683616,
      "step": 35650
    },
    {
      "epoch": 5.310545129579982,
      "grad_norm": 19.875,
      "learning_rate": 4.594202382323521e-05,
      "loss": 1.0832,
      "num_input_tokens_seen": 20686528,
      "step": 35655
    },
    {
      "epoch": 5.311289842120941,
      "grad_norm": 10.5,
      "learning_rate": 4.594024894269151e-05,
      "loss": 0.7605,
      "num_input_tokens_seen": 20689312,
      "step": 35660
    },
    {
      "epoch": 5.3120345546619,
      "grad_norm": 16.5,
      "learning_rate": 4.59384737083836e-05,
      "loss": 0.9711,
      "num_input_tokens_seen": 20692288,
      "step": 35665
    },
    {
      "epoch": 5.31277926720286,
      "grad_norm": 9.0625,
      "learning_rate": 4.5936698120341445e-05,
      "loss": 1.1666,
      "num_input_tokens_seen": 20695296,
      "step": 35670
    },
    {
      "epoch": 5.3135239797438185,
      "grad_norm": 10.5,
      "learning_rate": 4.593492217859506e-05,
      "loss": 1.1462,
      "num_input_tokens_seen": 20698464,
      "step": 35675
    },
    {
      "epoch": 5.314268692284778,
      "grad_norm": 7.8125,
      "learning_rate": 4.593314588317445e-05,
      "loss": 0.7973,
      "num_input_tokens_seen": 20701824,
      "step": 35680
    },
    {
      "epoch": 5.315013404825737,
      "grad_norm": 13.9375,
      "learning_rate": 4.5931369234109614e-05,
      "loss": 1.2637,
      "num_input_tokens_seen": 20704992,
      "step": 35685
    },
    {
      "epoch": 5.315758117366697,
      "grad_norm": 6.78125,
      "learning_rate": 4.592959223143056e-05,
      "loss": 1.037,
      "num_input_tokens_seen": 20707840,
      "step": 35690
    },
    {
      "epoch": 5.316502829907655,
      "grad_norm": 7.59375,
      "learning_rate": 4.592781487516732e-05,
      "loss": 1.1781,
      "num_input_tokens_seen": 20710816,
      "step": 35695
    },
    {
      "epoch": 5.317247542448615,
      "grad_norm": 9.125,
      "learning_rate": 4.592603716534992e-05,
      "loss": 0.8962,
      "num_input_tokens_seen": 20713472,
      "step": 35700
    },
    {
      "epoch": 5.317992254989574,
      "grad_norm": 5.03125,
      "learning_rate": 4.5924259102008386e-05,
      "loss": 1.3895,
      "num_input_tokens_seen": 20716352,
      "step": 35705
    },
    {
      "epoch": 5.318736967530533,
      "grad_norm": 7.53125,
      "learning_rate": 4.592248068517276e-05,
      "loss": 1.0144,
      "num_input_tokens_seen": 20719392,
      "step": 35710
    },
    {
      "epoch": 5.319481680071492,
      "grad_norm": 19.0,
      "learning_rate": 4.59207019148731e-05,
      "loss": 1.1448,
      "num_input_tokens_seen": 20722528,
      "step": 35715
    },
    {
      "epoch": 5.320226392612452,
      "grad_norm": 15.375,
      "learning_rate": 4.591892279113943e-05,
      "loss": 1.0059,
      "num_input_tokens_seen": 20725408,
      "step": 35720
    },
    {
      "epoch": 5.3209711051534105,
      "grad_norm": 15.75,
      "learning_rate": 4.591714331400183e-05,
      "loss": 1.1213,
      "num_input_tokens_seen": 20728064,
      "step": 35725
    },
    {
      "epoch": 5.32171581769437,
      "grad_norm": 16.125,
      "learning_rate": 4.5915363483490346e-05,
      "loss": 1.2811,
      "num_input_tokens_seen": 20731072,
      "step": 35730
    },
    {
      "epoch": 5.322460530235329,
      "grad_norm": 10.6875,
      "learning_rate": 4.591358329963505e-05,
      "loss": 1.0959,
      "num_input_tokens_seen": 20733792,
      "step": 35735
    },
    {
      "epoch": 5.323205242776289,
      "grad_norm": 12.375,
      "learning_rate": 4.5911802762466034e-05,
      "loss": 1.2191,
      "num_input_tokens_seen": 20736768,
      "step": 35740
    },
    {
      "epoch": 5.323949955317247,
      "grad_norm": 7.65625,
      "learning_rate": 4.5910021872013355e-05,
      "loss": 1.2161,
      "num_input_tokens_seen": 20739616,
      "step": 35745
    },
    {
      "epoch": 5.324694667858207,
      "grad_norm": 15.5,
      "learning_rate": 4.590824062830711e-05,
      "loss": 0.87,
      "num_input_tokens_seen": 20742496,
      "step": 35750
    },
    {
      "epoch": 5.325439380399166,
      "grad_norm": 8.4375,
      "learning_rate": 4.590645903137739e-05,
      "loss": 0.8352,
      "num_input_tokens_seen": 20745568,
      "step": 35755
    },
    {
      "epoch": 5.326184092940125,
      "grad_norm": 9.1875,
      "learning_rate": 4.590467708125429e-05,
      "loss": 1.1057,
      "num_input_tokens_seen": 20748448,
      "step": 35760
    },
    {
      "epoch": 5.326928805481084,
      "grad_norm": 11.4375,
      "learning_rate": 4.590289477796792e-05,
      "loss": 1.3251,
      "num_input_tokens_seen": 20751264,
      "step": 35765
    },
    {
      "epoch": 5.327673518022044,
      "grad_norm": 9.5625,
      "learning_rate": 4.590111212154839e-05,
      "loss": 1.0671,
      "num_input_tokens_seen": 20754400,
      "step": 35770
    },
    {
      "epoch": 5.328418230563003,
      "grad_norm": 21.125,
      "learning_rate": 4.589932911202581e-05,
      "loss": 1.237,
      "num_input_tokens_seen": 20756992,
      "step": 35775
    },
    {
      "epoch": 5.329162943103962,
      "grad_norm": 10.1875,
      "learning_rate": 4.5897545749430305e-05,
      "loss": 1.2398,
      "num_input_tokens_seen": 20759840,
      "step": 35780
    },
    {
      "epoch": 5.329907655644921,
      "grad_norm": 8.4375,
      "learning_rate": 4.5895762033792e-05,
      "loss": 1.1445,
      "num_input_tokens_seen": 20762848,
      "step": 35785
    },
    {
      "epoch": 5.330652368185881,
      "grad_norm": 10.5625,
      "learning_rate": 4.589397796514104e-05,
      "loss": 0.9468,
      "num_input_tokens_seen": 20765728,
      "step": 35790
    },
    {
      "epoch": 5.331397080726839,
      "grad_norm": 17.875,
      "learning_rate": 4.5892193543507556e-05,
      "loss": 1.1913,
      "num_input_tokens_seen": 20768480,
      "step": 35795
    },
    {
      "epoch": 5.332141793267799,
      "grad_norm": 7.28125,
      "learning_rate": 4.58904087689217e-05,
      "loss": 0.7558,
      "num_input_tokens_seen": 20771648,
      "step": 35800
    },
    {
      "epoch": 5.332886505808758,
      "grad_norm": 13.4375,
      "learning_rate": 4.5888623641413615e-05,
      "loss": 1.173,
      "num_input_tokens_seen": 20774528,
      "step": 35805
    },
    {
      "epoch": 5.333631218349717,
      "grad_norm": 12.875,
      "learning_rate": 4.588683816101347e-05,
      "loss": 1.1306,
      "num_input_tokens_seen": 20777472,
      "step": 35810
    },
    {
      "epoch": 5.334375930890676,
      "grad_norm": 10.5625,
      "learning_rate": 4.588505232775141e-05,
      "loss": 0.9471,
      "num_input_tokens_seen": 20780320,
      "step": 35815
    },
    {
      "epoch": 5.335120643431635,
      "grad_norm": 6.46875,
      "learning_rate": 4.588326614165763e-05,
      "loss": 0.9614,
      "num_input_tokens_seen": 20783200,
      "step": 35820
    },
    {
      "epoch": 5.335865355972595,
      "grad_norm": 10.5625,
      "learning_rate": 4.5881479602762286e-05,
      "loss": 1.0808,
      "num_input_tokens_seen": 20786176,
      "step": 35825
    },
    {
      "epoch": 5.336610068513554,
      "grad_norm": 8.125,
      "learning_rate": 4.587969271109557e-05,
      "loss": 1.2718,
      "num_input_tokens_seen": 20788928,
      "step": 35830
    },
    {
      "epoch": 5.337354781054513,
      "grad_norm": 4.28125,
      "learning_rate": 4.5877905466687666e-05,
      "loss": 0.959,
      "num_input_tokens_seen": 20792032,
      "step": 35835
    },
    {
      "epoch": 5.338099493595472,
      "grad_norm": 14.1875,
      "learning_rate": 4.5876117869568766e-05,
      "loss": 1.0104,
      "num_input_tokens_seen": 20795456,
      "step": 35840
    },
    {
      "epoch": 5.338844206136431,
      "grad_norm": 7.90625,
      "learning_rate": 4.587432991976908e-05,
      "loss": 1.0217,
      "num_input_tokens_seen": 20798624,
      "step": 35845
    },
    {
      "epoch": 5.33958891867739,
      "grad_norm": 8.4375,
      "learning_rate": 4.58725416173188e-05,
      "loss": 1.1264,
      "num_input_tokens_seen": 20801600,
      "step": 35850
    },
    {
      "epoch": 5.34033363121835,
      "grad_norm": 4.65625,
      "learning_rate": 4.587075296224814e-05,
      "loss": 0.9283,
      "num_input_tokens_seen": 20804416,
      "step": 35855
    },
    {
      "epoch": 5.341078343759309,
      "grad_norm": 6.875,
      "learning_rate": 4.586896395458733e-05,
      "loss": 0.9905,
      "num_input_tokens_seen": 20807296,
      "step": 35860
    },
    {
      "epoch": 5.341823056300268,
      "grad_norm": 7.8125,
      "learning_rate": 4.586717459436658e-05,
      "loss": 0.8973,
      "num_input_tokens_seen": 20810240,
      "step": 35865
    },
    {
      "epoch": 5.342567768841227,
      "grad_norm": 12.625,
      "learning_rate": 4.586538488161612e-05,
      "loss": 1.0018,
      "num_input_tokens_seen": 20813056,
      "step": 35870
    },
    {
      "epoch": 5.343312481382187,
      "grad_norm": 13.875,
      "learning_rate": 4.58635948163662e-05,
      "loss": 1.1592,
      "num_input_tokens_seen": 20815936,
      "step": 35875
    },
    {
      "epoch": 5.344057193923145,
      "grad_norm": 7.75,
      "learning_rate": 4.586180439864704e-05,
      "loss": 0.9814,
      "num_input_tokens_seen": 20818720,
      "step": 35880
    },
    {
      "epoch": 5.344801906464105,
      "grad_norm": 15.0625,
      "learning_rate": 4.586001362848889e-05,
      "loss": 1.2646,
      "num_input_tokens_seen": 20821728,
      "step": 35885
    },
    {
      "epoch": 5.345546619005064,
      "grad_norm": 13.75,
      "learning_rate": 4.5858222505922026e-05,
      "loss": 1.1308,
      "num_input_tokens_seen": 20824320,
      "step": 35890
    },
    {
      "epoch": 5.346291331546023,
      "grad_norm": 9.3125,
      "learning_rate": 4.585643103097669e-05,
      "loss": 1.2528,
      "num_input_tokens_seen": 20827360,
      "step": 35895
    },
    {
      "epoch": 5.347036044086982,
      "grad_norm": 6.84375,
      "learning_rate": 4.5854639203683146e-05,
      "loss": 1.1848,
      "num_input_tokens_seen": 20830208,
      "step": 35900
    },
    {
      "epoch": 5.347780756627942,
      "grad_norm": 14.25,
      "learning_rate": 4.5852847024071664e-05,
      "loss": 1.1064,
      "num_input_tokens_seen": 20832864,
      "step": 35905
    },
    {
      "epoch": 5.348525469168901,
      "grad_norm": 6.71875,
      "learning_rate": 4.585105449217253e-05,
      "loss": 0.8451,
      "num_input_tokens_seen": 20836000,
      "step": 35910
    },
    {
      "epoch": 5.34927018170986,
      "grad_norm": 6.5,
      "learning_rate": 4.5849261608016026e-05,
      "loss": 0.873,
      "num_input_tokens_seen": 20838944,
      "step": 35915
    },
    {
      "epoch": 5.350014894250819,
      "grad_norm": 11.5,
      "learning_rate": 4.584746837163243e-05,
      "loss": 1.3639,
      "num_input_tokens_seen": 20842304,
      "step": 35920
    },
    {
      "epoch": 5.350759606791779,
      "grad_norm": 15.1875,
      "learning_rate": 4.584567478305205e-05,
      "loss": 1.0321,
      "num_input_tokens_seen": 20844992,
      "step": 35925
    },
    {
      "epoch": 5.351504319332737,
      "grad_norm": 15.5625,
      "learning_rate": 4.584388084230518e-05,
      "loss": 0.8326,
      "num_input_tokens_seen": 20847808,
      "step": 35930
    },
    {
      "epoch": 5.352249031873697,
      "grad_norm": 11.75,
      "learning_rate": 4.584208654942212e-05,
      "loss": 1.0664,
      "num_input_tokens_seen": 20850752,
      "step": 35935
    },
    {
      "epoch": 5.352993744414656,
      "grad_norm": 6.9375,
      "learning_rate": 4.584029190443321e-05,
      "loss": 1.0168,
      "num_input_tokens_seen": 20853664,
      "step": 35940
    },
    {
      "epoch": 5.3537384569556155,
      "grad_norm": 7.75,
      "learning_rate": 4.583849690736873e-05,
      "loss": 0.9226,
      "num_input_tokens_seen": 20856704,
      "step": 35945
    },
    {
      "epoch": 5.354483169496574,
      "grad_norm": 13.75,
      "learning_rate": 4.583670155825903e-05,
      "loss": 1.1473,
      "num_input_tokens_seen": 20859584,
      "step": 35950
    },
    {
      "epoch": 5.355227882037534,
      "grad_norm": 15.0,
      "learning_rate": 4.5834905857134436e-05,
      "loss": 1.1819,
      "num_input_tokens_seen": 20862464,
      "step": 35955
    },
    {
      "epoch": 5.355972594578493,
      "grad_norm": 8.75,
      "learning_rate": 4.583310980402529e-05,
      "loss": 1.1193,
      "num_input_tokens_seen": 20865248,
      "step": 35960
    },
    {
      "epoch": 5.356717307119452,
      "grad_norm": 10.6875,
      "learning_rate": 4.5831313398961915e-05,
      "loss": 1.1803,
      "num_input_tokens_seen": 20868064,
      "step": 35965
    },
    {
      "epoch": 5.357462019660411,
      "grad_norm": 7.15625,
      "learning_rate": 4.5829516641974676e-05,
      "loss": 1.2292,
      "num_input_tokens_seen": 20870944,
      "step": 35970
    },
    {
      "epoch": 5.358206732201371,
      "grad_norm": 5.0625,
      "learning_rate": 4.582771953309393e-05,
      "loss": 0.796,
      "num_input_tokens_seen": 20873760,
      "step": 35975
    },
    {
      "epoch": 5.358951444742329,
      "grad_norm": 17.25,
      "learning_rate": 4.582592207235002e-05,
      "loss": 1.2435,
      "num_input_tokens_seen": 20876352,
      "step": 35980
    },
    {
      "epoch": 5.359696157283288,
      "grad_norm": 12.75,
      "learning_rate": 4.5824124259773336e-05,
      "loss": 1.1652,
      "num_input_tokens_seen": 20879104,
      "step": 35985
    },
    {
      "epoch": 5.360440869824248,
      "grad_norm": 10.125,
      "learning_rate": 4.582232609539423e-05,
      "loss": 1.1876,
      "num_input_tokens_seen": 20881920,
      "step": 35990
    },
    {
      "epoch": 5.3611855823652075,
      "grad_norm": 8.6875,
      "learning_rate": 4.582052757924309e-05,
      "loss": 0.9565,
      "num_input_tokens_seen": 20884928,
      "step": 35995
    },
    {
      "epoch": 5.361930294906166,
      "grad_norm": 7.46875,
      "learning_rate": 4.5818728711350296e-05,
      "loss": 0.885,
      "num_input_tokens_seen": 20887616,
      "step": 36000
    },
    {
      "epoch": 5.362675007447125,
      "grad_norm": 15.1875,
      "learning_rate": 4.581692949174624e-05,
      "loss": 1.1481,
      "num_input_tokens_seen": 20890624,
      "step": 36005
    },
    {
      "epoch": 5.363419719988085,
      "grad_norm": 6.9375,
      "learning_rate": 4.581512992046132e-05,
      "loss": 0.9966,
      "num_input_tokens_seen": 20893568,
      "step": 36010
    },
    {
      "epoch": 5.364164432529043,
      "grad_norm": 7.875,
      "learning_rate": 4.5813329997525925e-05,
      "loss": 1.0331,
      "num_input_tokens_seen": 20896384,
      "step": 36015
    },
    {
      "epoch": 5.364909145070003,
      "grad_norm": 13.9375,
      "learning_rate": 4.5811529722970484e-05,
      "loss": 1.302,
      "num_input_tokens_seen": 20899232,
      "step": 36020
    },
    {
      "epoch": 5.365653857610962,
      "grad_norm": 8.0,
      "learning_rate": 4.5809729096825396e-05,
      "loss": 1.1307,
      "num_input_tokens_seen": 20902272,
      "step": 36025
    },
    {
      "epoch": 5.3663985701519215,
      "grad_norm": 10.5,
      "learning_rate": 4.580792811912109e-05,
      "loss": 0.6969,
      "num_input_tokens_seen": 20904864,
      "step": 36030
    },
    {
      "epoch": 5.36714328269288,
      "grad_norm": 10.375,
      "learning_rate": 4.5806126789887984e-05,
      "loss": 0.7497,
      "num_input_tokens_seen": 20907424,
      "step": 36035
    },
    {
      "epoch": 5.36788799523384,
      "grad_norm": 13.4375,
      "learning_rate": 4.580432510915651e-05,
      "loss": 1.3794,
      "num_input_tokens_seen": 20910112,
      "step": 36040
    },
    {
      "epoch": 5.368632707774799,
      "grad_norm": 6.9375,
      "learning_rate": 4.580252307695711e-05,
      "loss": 1.1567,
      "num_input_tokens_seen": 20913184,
      "step": 36045
    },
    {
      "epoch": 5.369377420315758,
      "grad_norm": 12.625,
      "learning_rate": 4.580072069332022e-05,
      "loss": 1.2955,
      "num_input_tokens_seen": 20916000,
      "step": 36050
    },
    {
      "epoch": 5.370122132856717,
      "grad_norm": 10.75,
      "learning_rate": 4.57989179582763e-05,
      "loss": 1.1481,
      "num_input_tokens_seen": 20918784,
      "step": 36055
    },
    {
      "epoch": 5.370866845397677,
      "grad_norm": 8.1875,
      "learning_rate": 4.57971148718558e-05,
      "loss": 0.8002,
      "num_input_tokens_seen": 20921696,
      "step": 36060
    },
    {
      "epoch": 5.371611557938635,
      "grad_norm": 18.75,
      "learning_rate": 4.579531143408918e-05,
      "loss": 0.9538,
      "num_input_tokens_seen": 20924736,
      "step": 36065
    },
    {
      "epoch": 5.372356270479595,
      "grad_norm": 10.4375,
      "learning_rate": 4.579350764500691e-05,
      "loss": 0.7865,
      "num_input_tokens_seen": 20927808,
      "step": 36070
    },
    {
      "epoch": 5.373100983020554,
      "grad_norm": 7.40625,
      "learning_rate": 4.579170350463946e-05,
      "loss": 1.2839,
      "num_input_tokens_seen": 20930656,
      "step": 36075
    },
    {
      "epoch": 5.3738456955615135,
      "grad_norm": 9.0,
      "learning_rate": 4.5789899013017315e-05,
      "loss": 0.9988,
      "num_input_tokens_seen": 20933312,
      "step": 36080
    },
    {
      "epoch": 5.374590408102472,
      "grad_norm": 14.125,
      "learning_rate": 4.578809417017095e-05,
      "loss": 1.2381,
      "num_input_tokens_seen": 20936608,
      "step": 36085
    },
    {
      "epoch": 5.375335120643432,
      "grad_norm": 10.6875,
      "learning_rate": 4.578628897613087e-05,
      "loss": 1.2015,
      "num_input_tokens_seen": 20939616,
      "step": 36090
    },
    {
      "epoch": 5.376079833184391,
      "grad_norm": 8.625,
      "learning_rate": 4.578448343092756e-05,
      "loss": 0.8376,
      "num_input_tokens_seen": 20942560,
      "step": 36095
    },
    {
      "epoch": 5.37682454572535,
      "grad_norm": 4.09375,
      "learning_rate": 4.5782677534591524e-05,
      "loss": 1.2316,
      "num_input_tokens_seen": 20945504,
      "step": 36100
    },
    {
      "epoch": 5.377569258266309,
      "grad_norm": 7.125,
      "learning_rate": 4.578087128715328e-05,
      "loss": 1.2351,
      "num_input_tokens_seen": 20948640,
      "step": 36105
    },
    {
      "epoch": 5.378313970807269,
      "grad_norm": 8.0,
      "learning_rate": 4.577906468864333e-05,
      "loss": 0.6684,
      "num_input_tokens_seen": 20951552,
      "step": 36110
    },
    {
      "epoch": 5.3790586833482275,
      "grad_norm": 8.3125,
      "learning_rate": 4.577725773909221e-05,
      "loss": 1.2394,
      "num_input_tokens_seen": 20954112,
      "step": 36115
    },
    {
      "epoch": 5.379803395889187,
      "grad_norm": 7.59375,
      "learning_rate": 4.577545043853042e-05,
      "loss": 0.7986,
      "num_input_tokens_seen": 20956864,
      "step": 36120
    },
    {
      "epoch": 5.380548108430146,
      "grad_norm": 11.8125,
      "learning_rate": 4.577364278698852e-05,
      "loss": 1.2576,
      "num_input_tokens_seen": 20959840,
      "step": 36125
    },
    {
      "epoch": 5.3812928209711055,
      "grad_norm": 12.875,
      "learning_rate": 4.577183478449705e-05,
      "loss": 0.9497,
      "num_input_tokens_seen": 20962560,
      "step": 36130
    },
    {
      "epoch": 5.382037533512064,
      "grad_norm": 9.375,
      "learning_rate": 4.5770026431086524e-05,
      "loss": 1.3857,
      "num_input_tokens_seen": 20965760,
      "step": 36135
    },
    {
      "epoch": 5.382782246053024,
      "grad_norm": 12.4375,
      "learning_rate": 4.576821772678752e-05,
      "loss": 1.1964,
      "num_input_tokens_seen": 20968768,
      "step": 36140
    },
    {
      "epoch": 5.383526958593983,
      "grad_norm": 7.40625,
      "learning_rate": 4.576640867163059e-05,
      "loss": 1.4047,
      "num_input_tokens_seen": 20971872,
      "step": 36145
    },
    {
      "epoch": 5.384271671134942,
      "grad_norm": 9.6875,
      "learning_rate": 4.5764599265646286e-05,
      "loss": 1.095,
      "num_input_tokens_seen": 20974560,
      "step": 36150
    },
    {
      "epoch": 5.385016383675901,
      "grad_norm": 3.859375,
      "learning_rate": 4.576278950886518e-05,
      "loss": 0.989,
      "num_input_tokens_seen": 20977536,
      "step": 36155
    },
    {
      "epoch": 5.385761096216861,
      "grad_norm": 7.875,
      "learning_rate": 4.576097940131785e-05,
      "loss": 1.1472,
      "num_input_tokens_seen": 20980160,
      "step": 36160
    },
    {
      "epoch": 5.3865058087578195,
      "grad_norm": 8.9375,
      "learning_rate": 4.5759168943034875e-05,
      "loss": 0.7547,
      "num_input_tokens_seen": 20983200,
      "step": 36165
    },
    {
      "epoch": 5.387250521298778,
      "grad_norm": 8.0,
      "learning_rate": 4.5757358134046835e-05,
      "loss": 1.4984,
      "num_input_tokens_seen": 20986016,
      "step": 36170
    },
    {
      "epoch": 5.387995233839738,
      "grad_norm": 8.5625,
      "learning_rate": 4.5755546974384336e-05,
      "loss": 0.9967,
      "num_input_tokens_seen": 20988768,
      "step": 36175
    },
    {
      "epoch": 5.388739946380697,
      "grad_norm": 5.875,
      "learning_rate": 4.575373546407795e-05,
      "loss": 1.2051,
      "num_input_tokens_seen": 20991744,
      "step": 36180
    },
    {
      "epoch": 5.389484658921656,
      "grad_norm": 18.25,
      "learning_rate": 4.5751923603158305e-05,
      "loss": 1.0811,
      "num_input_tokens_seen": 20995168,
      "step": 36185
    },
    {
      "epoch": 5.390229371462615,
      "grad_norm": 15.4375,
      "learning_rate": 4.5750111391656005e-05,
      "loss": 1.3061,
      "num_input_tokens_seen": 20998080,
      "step": 36190
    },
    {
      "epoch": 5.390974084003575,
      "grad_norm": 9.0625,
      "learning_rate": 4.574829882960166e-05,
      "loss": 0.9212,
      "num_input_tokens_seen": 21000672,
      "step": 36195
    },
    {
      "epoch": 5.3917187965445335,
      "grad_norm": 6.40625,
      "learning_rate": 4.5746485917025894e-05,
      "loss": 0.7954,
      "num_input_tokens_seen": 21003424,
      "step": 36200
    },
    {
      "epoch": 5.392463509085493,
      "grad_norm": 7.40625,
      "learning_rate": 4.574467265395933e-05,
      "loss": 1.0021,
      "num_input_tokens_seen": 21006240,
      "step": 36205
    },
    {
      "epoch": 5.393208221626452,
      "grad_norm": 7.8125,
      "learning_rate": 4.574285904043261e-05,
      "loss": 1.1331,
      "num_input_tokens_seen": 21009312,
      "step": 36210
    },
    {
      "epoch": 5.3939529341674115,
      "grad_norm": 12.3125,
      "learning_rate": 4.574104507647637e-05,
      "loss": 1.5126,
      "num_input_tokens_seen": 21012256,
      "step": 36215
    },
    {
      "epoch": 5.39469764670837,
      "grad_norm": 13.4375,
      "learning_rate": 4.5739230762121255e-05,
      "loss": 1.3038,
      "num_input_tokens_seen": 21015232,
      "step": 36220
    },
    {
      "epoch": 5.39544235924933,
      "grad_norm": 6.90625,
      "learning_rate": 4.573741609739791e-05,
      "loss": 0.9297,
      "num_input_tokens_seen": 21017984,
      "step": 36225
    },
    {
      "epoch": 5.396187071790289,
      "grad_norm": 10.5,
      "learning_rate": 4.5735601082336995e-05,
      "loss": 1.0026,
      "num_input_tokens_seen": 21020768,
      "step": 36230
    },
    {
      "epoch": 5.396931784331248,
      "grad_norm": 10.8125,
      "learning_rate": 4.573378571696918e-05,
      "loss": 0.8716,
      "num_input_tokens_seen": 21023840,
      "step": 36235
    },
    {
      "epoch": 5.397676496872207,
      "grad_norm": 11.0625,
      "learning_rate": 4.573197000132512e-05,
      "loss": 1.2645,
      "num_input_tokens_seen": 21026528,
      "step": 36240
    },
    {
      "epoch": 5.398421209413167,
      "grad_norm": 14.375,
      "learning_rate": 4.57301539354355e-05,
      "loss": 1.2605,
      "num_input_tokens_seen": 21029280,
      "step": 36245
    },
    {
      "epoch": 5.3991659219541255,
      "grad_norm": 13.0,
      "learning_rate": 4.572833751933101e-05,
      "loss": 1.0424,
      "num_input_tokens_seen": 21032128,
      "step": 36250
    },
    {
      "epoch": 5.399910634495085,
      "grad_norm": 8.25,
      "learning_rate": 4.5726520753042314e-05,
      "loss": 1.0276,
      "num_input_tokens_seen": 21035200,
      "step": 36255
    },
    {
      "epoch": 5.400655347036044,
      "grad_norm": 9.6875,
      "learning_rate": 4.572470363660012e-05,
      "loss": 1.1055,
      "num_input_tokens_seen": 21038208,
      "step": 36260
    },
    {
      "epoch": 5.4014000595770035,
      "grad_norm": 11.0625,
      "learning_rate": 4.572288617003512e-05,
      "loss": 0.7588,
      "num_input_tokens_seen": 21041056,
      "step": 36265
    },
    {
      "epoch": 5.402144772117962,
      "grad_norm": 10.0,
      "learning_rate": 4.5721068353378016e-05,
      "loss": 1.2242,
      "num_input_tokens_seen": 21044160,
      "step": 36270
    },
    {
      "epoch": 5.402889484658922,
      "grad_norm": 7.34375,
      "learning_rate": 4.571925018665953e-05,
      "loss": 1.3207,
      "num_input_tokens_seen": 21047072,
      "step": 36275
    },
    {
      "epoch": 5.403634197199881,
      "grad_norm": 7.59375,
      "learning_rate": 4.5717431669910364e-05,
      "loss": 0.8251,
      "num_input_tokens_seen": 21049888,
      "step": 36280
    },
    {
      "epoch": 5.40437890974084,
      "grad_norm": 5.21875,
      "learning_rate": 4.571561280316125e-05,
      "loss": 1.0719,
      "num_input_tokens_seen": 21052448,
      "step": 36285
    },
    {
      "epoch": 5.405123622281799,
      "grad_norm": 8.125,
      "learning_rate": 4.571379358644291e-05,
      "loss": 0.8922,
      "num_input_tokens_seen": 21055264,
      "step": 36290
    },
    {
      "epoch": 5.405868334822759,
      "grad_norm": 7.65625,
      "learning_rate": 4.571197401978608e-05,
      "loss": 1.1384,
      "num_input_tokens_seen": 21058048,
      "step": 36295
    },
    {
      "epoch": 5.4066130473637175,
      "grad_norm": 7.875,
      "learning_rate": 4.5710154103221504e-05,
      "loss": 0.8215,
      "num_input_tokens_seen": 21061120,
      "step": 36300
    },
    {
      "epoch": 5.407357759904677,
      "grad_norm": 14.0,
      "learning_rate": 4.570833383677991e-05,
      "loss": 0.7622,
      "num_input_tokens_seen": 21064128,
      "step": 36305
    },
    {
      "epoch": 5.408102472445636,
      "grad_norm": 11.625,
      "learning_rate": 4.570651322049208e-05,
      "loss": 1.0892,
      "num_input_tokens_seen": 21066752,
      "step": 36310
    },
    {
      "epoch": 5.408847184986596,
      "grad_norm": 12.0,
      "learning_rate": 4.570469225438875e-05,
      "loss": 1.2167,
      "num_input_tokens_seen": 21069728,
      "step": 36315
    },
    {
      "epoch": 5.409591897527554,
      "grad_norm": 10.75,
      "learning_rate": 4.570287093850068e-05,
      "loss": 1.1147,
      "num_input_tokens_seen": 21072384,
      "step": 36320
    },
    {
      "epoch": 5.410336610068514,
      "grad_norm": 9.5,
      "learning_rate": 4.570104927285865e-05,
      "loss": 1.1186,
      "num_input_tokens_seen": 21075456,
      "step": 36325
    },
    {
      "epoch": 5.411081322609473,
      "grad_norm": 6.90625,
      "learning_rate": 4.5699227257493434e-05,
      "loss": 1.0545,
      "num_input_tokens_seen": 21078208,
      "step": 36330
    },
    {
      "epoch": 5.4118260351504315,
      "grad_norm": 5.375,
      "learning_rate": 4.5697404892435816e-05,
      "loss": 0.9216,
      "num_input_tokens_seen": 21081280,
      "step": 36335
    },
    {
      "epoch": 5.412570747691391,
      "grad_norm": 17.5,
      "learning_rate": 4.5695582177716566e-05,
      "loss": 1.1273,
      "num_input_tokens_seen": 21084192,
      "step": 36340
    },
    {
      "epoch": 5.413315460232351,
      "grad_norm": 16.25,
      "learning_rate": 4.56937591133665e-05,
      "loss": 1.2763,
      "num_input_tokens_seen": 21087264,
      "step": 36345
    },
    {
      "epoch": 5.4140601727733095,
      "grad_norm": 11.9375,
      "learning_rate": 4.56919356994164e-05,
      "loss": 1.2094,
      "num_input_tokens_seen": 21090176,
      "step": 36350
    },
    {
      "epoch": 5.414804885314268,
      "grad_norm": 18.125,
      "learning_rate": 4.569011193589707e-05,
      "loss": 1.2269,
      "num_input_tokens_seen": 21092960,
      "step": 36355
    },
    {
      "epoch": 5.415549597855228,
      "grad_norm": 15.9375,
      "learning_rate": 4.568828782283934e-05,
      "loss": 0.8103,
      "num_input_tokens_seen": 21095712,
      "step": 36360
    },
    {
      "epoch": 5.416294310396187,
      "grad_norm": 11.9375,
      "learning_rate": 4.5686463360274015e-05,
      "loss": 1.0187,
      "num_input_tokens_seen": 21098656,
      "step": 36365
    },
    {
      "epoch": 5.417039022937146,
      "grad_norm": 21.0,
      "learning_rate": 4.568463854823191e-05,
      "loss": 1.1376,
      "num_input_tokens_seen": 21101344,
      "step": 36370
    },
    {
      "epoch": 5.417783735478105,
      "grad_norm": 9.0625,
      "learning_rate": 4.5682813386743864e-05,
      "loss": 0.8642,
      "num_input_tokens_seen": 21104384,
      "step": 36375
    },
    {
      "epoch": 5.418528448019065,
      "grad_norm": 14.8125,
      "learning_rate": 4.56809878758407e-05,
      "loss": 0.9323,
      "num_input_tokens_seen": 21107264,
      "step": 36380
    },
    {
      "epoch": 5.4192731605600235,
      "grad_norm": 13.125,
      "learning_rate": 4.567916201555327e-05,
      "loss": 1.0721,
      "num_input_tokens_seen": 21110208,
      "step": 36385
    },
    {
      "epoch": 5.420017873100983,
      "grad_norm": 10.0,
      "learning_rate": 4.567733580591241e-05,
      "loss": 1.1763,
      "num_input_tokens_seen": 21113184,
      "step": 36390
    },
    {
      "epoch": 5.420762585641942,
      "grad_norm": 9.9375,
      "learning_rate": 4.567550924694898e-05,
      "loss": 0.9597,
      "num_input_tokens_seen": 21116160,
      "step": 36395
    },
    {
      "epoch": 5.421507298182902,
      "grad_norm": 12.6875,
      "learning_rate": 4.5673682338693836e-05,
      "loss": 1.221,
      "num_input_tokens_seen": 21119040,
      "step": 36400
    },
    {
      "epoch": 5.42225201072386,
      "grad_norm": 3.640625,
      "learning_rate": 4.567185508117784e-05,
      "loss": 0.9255,
      "num_input_tokens_seen": 21122400,
      "step": 36405
    },
    {
      "epoch": 5.42299672326482,
      "grad_norm": 10.6875,
      "learning_rate": 4.567002747443186e-05,
      "loss": 1.1197,
      "num_input_tokens_seen": 21125216,
      "step": 36410
    },
    {
      "epoch": 5.423741435805779,
      "grad_norm": 8.4375,
      "learning_rate": 4.5668199518486785e-05,
      "loss": 1.085,
      "num_input_tokens_seen": 21127808,
      "step": 36415
    },
    {
      "epoch": 5.424486148346738,
      "grad_norm": 11.9375,
      "learning_rate": 4.566637121337347e-05,
      "loss": 1.2934,
      "num_input_tokens_seen": 21130784,
      "step": 36420
    },
    {
      "epoch": 5.425230860887697,
      "grad_norm": 10.25,
      "learning_rate": 4.566454255912283e-05,
      "loss": 1.0332,
      "num_input_tokens_seen": 21133728,
      "step": 36425
    },
    {
      "epoch": 5.425975573428657,
      "grad_norm": 15.3125,
      "learning_rate": 4.5662713555765735e-05,
      "loss": 1.1087,
      "num_input_tokens_seen": 21136736,
      "step": 36430
    },
    {
      "epoch": 5.4267202859696155,
      "grad_norm": 17.25,
      "learning_rate": 4.56608842033331e-05,
      "loss": 1.02,
      "num_input_tokens_seen": 21139392,
      "step": 36435
    },
    {
      "epoch": 5.427464998510575,
      "grad_norm": 11.25,
      "learning_rate": 4.565905450185583e-05,
      "loss": 1.6172,
      "num_input_tokens_seen": 21142432,
      "step": 36440
    },
    {
      "epoch": 5.428209711051534,
      "grad_norm": 6.5,
      "learning_rate": 4.565722445136483e-05,
      "loss": 0.9277,
      "num_input_tokens_seen": 21145280,
      "step": 36445
    },
    {
      "epoch": 5.428954423592494,
      "grad_norm": 9.5625,
      "learning_rate": 4.565539405189101e-05,
      "loss": 1.0483,
      "num_input_tokens_seen": 21148384,
      "step": 36450
    },
    {
      "epoch": 5.429699136133452,
      "grad_norm": 12.25,
      "learning_rate": 4.5653563303465306e-05,
      "loss": 1.037,
      "num_input_tokens_seen": 21151456,
      "step": 36455
    },
    {
      "epoch": 5.430443848674412,
      "grad_norm": 10.3125,
      "learning_rate": 4.565173220611864e-05,
      "loss": 1.4035,
      "num_input_tokens_seen": 21154304,
      "step": 36460
    },
    {
      "epoch": 5.431188561215371,
      "grad_norm": 16.0,
      "learning_rate": 4.5649900759881956e-05,
      "loss": 0.7815,
      "num_input_tokens_seen": 21157408,
      "step": 36465
    },
    {
      "epoch": 5.43193327375633,
      "grad_norm": 17.125,
      "learning_rate": 4.564806896478617e-05,
      "loss": 0.9917,
      "num_input_tokens_seen": 21160448,
      "step": 36470
    },
    {
      "epoch": 5.432677986297289,
      "grad_norm": 12.0,
      "learning_rate": 4.564623682086226e-05,
      "loss": 1.2927,
      "num_input_tokens_seen": 21163392,
      "step": 36475
    },
    {
      "epoch": 5.433422698838249,
      "grad_norm": 7.09375,
      "learning_rate": 4.564440432814116e-05,
      "loss": 1.1351,
      "num_input_tokens_seen": 21166336,
      "step": 36480
    },
    {
      "epoch": 5.434167411379208,
      "grad_norm": 15.375,
      "learning_rate": 4.5642571486653825e-05,
      "loss": 1.1532,
      "num_input_tokens_seen": 21169056,
      "step": 36485
    },
    {
      "epoch": 5.434912123920167,
      "grad_norm": 6.5625,
      "learning_rate": 4.5640738296431224e-05,
      "loss": 0.8216,
      "num_input_tokens_seen": 21171936,
      "step": 36490
    },
    {
      "epoch": 5.435656836461126,
      "grad_norm": 9.4375,
      "learning_rate": 4.563890475750433e-05,
      "loss": 1.2228,
      "num_input_tokens_seen": 21174976,
      "step": 36495
    },
    {
      "epoch": 5.436401549002086,
      "grad_norm": 10.9375,
      "learning_rate": 4.563707086990412e-05,
      "loss": 1.067,
      "num_input_tokens_seen": 21177888,
      "step": 36500
    },
    {
      "epoch": 5.437146261543044,
      "grad_norm": 8.4375,
      "learning_rate": 4.563523663366157e-05,
      "loss": 1.0614,
      "num_input_tokens_seen": 21180768,
      "step": 36505
    },
    {
      "epoch": 5.437890974084004,
      "grad_norm": 8.125,
      "learning_rate": 4.563340204880767e-05,
      "loss": 0.7644,
      "num_input_tokens_seen": 21183872,
      "step": 36510
    },
    {
      "epoch": 5.438635686624963,
      "grad_norm": 8.375,
      "learning_rate": 4.563156711537341e-05,
      "loss": 1.0913,
      "num_input_tokens_seen": 21186816,
      "step": 36515
    },
    {
      "epoch": 5.4393803991659215,
      "grad_norm": 22.0,
      "learning_rate": 4.56297318333898e-05,
      "loss": 1.2525,
      "num_input_tokens_seen": 21189472,
      "step": 36520
    },
    {
      "epoch": 5.440125111706881,
      "grad_norm": 8.5625,
      "learning_rate": 4.562789620288783e-05,
      "loss": 1.0214,
      "num_input_tokens_seen": 21192224,
      "step": 36525
    },
    {
      "epoch": 5.44086982424784,
      "grad_norm": 10.25,
      "learning_rate": 4.562606022389853e-05,
      "loss": 0.8055,
      "num_input_tokens_seen": 21194912,
      "step": 36530
    },
    {
      "epoch": 5.4416145367888,
      "grad_norm": 14.0,
      "learning_rate": 4.5624223896452894e-05,
      "loss": 1.2688,
      "num_input_tokens_seen": 21198080,
      "step": 36535
    },
    {
      "epoch": 5.442359249329758,
      "grad_norm": 22.75,
      "learning_rate": 4.5622387220581965e-05,
      "loss": 0.899,
      "num_input_tokens_seen": 21200896,
      "step": 36540
    },
    {
      "epoch": 5.443103961870718,
      "grad_norm": 10.75,
      "learning_rate": 4.5620550196316757e-05,
      "loss": 0.8933,
      "num_input_tokens_seen": 21203776,
      "step": 36545
    },
    {
      "epoch": 5.443848674411677,
      "grad_norm": 18.125,
      "learning_rate": 4.5618712823688316e-05,
      "loss": 1.1172,
      "num_input_tokens_seen": 21206496,
      "step": 36550
    },
    {
      "epoch": 5.444593386952636,
      "grad_norm": 9.6875,
      "learning_rate": 4.561687510272767e-05,
      "loss": 1.1348,
      "num_input_tokens_seen": 21209440,
      "step": 36555
    },
    {
      "epoch": 5.445338099493595,
      "grad_norm": 18.75,
      "learning_rate": 4.5615037033465876e-05,
      "loss": 1.078,
      "num_input_tokens_seen": 21212480,
      "step": 36560
    },
    {
      "epoch": 5.446082812034555,
      "grad_norm": 14.5625,
      "learning_rate": 4.5613198615933994e-05,
      "loss": 1.3275,
      "num_input_tokens_seen": 21215328,
      "step": 36565
    },
    {
      "epoch": 5.446827524575514,
      "grad_norm": 7.125,
      "learning_rate": 4.561135985016306e-05,
      "loss": 1.2756,
      "num_input_tokens_seen": 21218240,
      "step": 36570
    },
    {
      "epoch": 5.447572237116473,
      "grad_norm": 8.4375,
      "learning_rate": 4.560952073618415e-05,
      "loss": 1.075,
      "num_input_tokens_seen": 21221408,
      "step": 36575
    },
    {
      "epoch": 5.448316949657432,
      "grad_norm": 9.8125,
      "learning_rate": 4.560768127402834e-05,
      "loss": 1.2035,
      "num_input_tokens_seen": 21224352,
      "step": 36580
    },
    {
      "epoch": 5.449061662198392,
      "grad_norm": 9.8125,
      "learning_rate": 4.5605841463726695e-05,
      "loss": 1.1086,
      "num_input_tokens_seen": 21227360,
      "step": 36585
    },
    {
      "epoch": 5.44980637473935,
      "grad_norm": 10.375,
      "learning_rate": 4.5604001305310304e-05,
      "loss": 0.9966,
      "num_input_tokens_seen": 21229952,
      "step": 36590
    },
    {
      "epoch": 5.45055108728031,
      "grad_norm": 10.0625,
      "learning_rate": 4.5602160798810256e-05,
      "loss": 1.4356,
      "num_input_tokens_seen": 21232800,
      "step": 36595
    },
    {
      "epoch": 5.451295799821269,
      "grad_norm": 17.75,
      "learning_rate": 4.5600319944257635e-05,
      "loss": 1.5348,
      "num_input_tokens_seen": 21235584,
      "step": 36600
    },
    {
      "epoch": 5.452040512362228,
      "grad_norm": 13.6875,
      "learning_rate": 4.559847874168355e-05,
      "loss": 1.0264,
      "num_input_tokens_seen": 21238656,
      "step": 36605
    },
    {
      "epoch": 5.452785224903187,
      "grad_norm": 9.4375,
      "learning_rate": 4.55966371911191e-05,
      "loss": 1.2246,
      "num_input_tokens_seen": 21241408,
      "step": 36610
    },
    {
      "epoch": 5.453529937444147,
      "grad_norm": 11.875,
      "learning_rate": 4.5594795292595394e-05,
      "loss": 1.1367,
      "num_input_tokens_seen": 21244416,
      "step": 36615
    },
    {
      "epoch": 5.454274649985106,
      "grad_norm": 10.0625,
      "learning_rate": 4.559295304614355e-05,
      "loss": 1.3409,
      "num_input_tokens_seen": 21247296,
      "step": 36620
    },
    {
      "epoch": 5.455019362526065,
      "grad_norm": 13.625,
      "learning_rate": 4.559111045179471e-05,
      "loss": 1.1077,
      "num_input_tokens_seen": 21250432,
      "step": 36625
    },
    {
      "epoch": 5.455764075067024,
      "grad_norm": 10.1875,
      "learning_rate": 4.558926750957997e-05,
      "loss": 1.1726,
      "num_input_tokens_seen": 21253216,
      "step": 36630
    },
    {
      "epoch": 5.456508787607984,
      "grad_norm": 12.5625,
      "learning_rate": 4.558742421953049e-05,
      "loss": 1.0482,
      "num_input_tokens_seen": 21256096,
      "step": 36635
    },
    {
      "epoch": 5.457253500148942,
      "grad_norm": 8.375,
      "learning_rate": 4.55855805816774e-05,
      "loss": 1.0557,
      "num_input_tokens_seen": 21258720,
      "step": 36640
    },
    {
      "epoch": 5.457998212689902,
      "grad_norm": 11.75,
      "learning_rate": 4.558373659605185e-05,
      "loss": 1.1492,
      "num_input_tokens_seen": 21261536,
      "step": 36645
    },
    {
      "epoch": 5.458742925230861,
      "grad_norm": 15.125,
      "learning_rate": 4.5581892262684984e-05,
      "loss": 1.3676,
      "num_input_tokens_seen": 21264864,
      "step": 36650
    },
    {
      "epoch": 5.4594876377718204,
      "grad_norm": 11.0,
      "learning_rate": 4.558004758160798e-05,
      "loss": 1.2137,
      "num_input_tokens_seen": 21267712,
      "step": 36655
    },
    {
      "epoch": 5.460232350312779,
      "grad_norm": 13.75,
      "learning_rate": 4.5578202552851976e-05,
      "loss": 1.1287,
      "num_input_tokens_seen": 21270688,
      "step": 36660
    },
    {
      "epoch": 5.460977062853739,
      "grad_norm": 8.4375,
      "learning_rate": 4.557635717644816e-05,
      "loss": 1.0558,
      "num_input_tokens_seen": 21273376,
      "step": 36665
    },
    {
      "epoch": 5.461721775394698,
      "grad_norm": 4.53125,
      "learning_rate": 4.557451145242769e-05,
      "loss": 0.8101,
      "num_input_tokens_seen": 21276128,
      "step": 36670
    },
    {
      "epoch": 5.462466487935657,
      "grad_norm": 6.96875,
      "learning_rate": 4.557266538082178e-05,
      "loss": 0.8618,
      "num_input_tokens_seen": 21278848,
      "step": 36675
    },
    {
      "epoch": 5.463211200476616,
      "grad_norm": 6.34375,
      "learning_rate": 4.557081896166159e-05,
      "loss": 0.8631,
      "num_input_tokens_seen": 21281728,
      "step": 36680
    },
    {
      "epoch": 5.463955913017575,
      "grad_norm": 9.9375,
      "learning_rate": 4.556897219497832e-05,
      "loss": 1.1021,
      "num_input_tokens_seen": 21284480,
      "step": 36685
    },
    {
      "epoch": 5.464700625558534,
      "grad_norm": 6.0625,
      "learning_rate": 4.556712508080316e-05,
      "loss": 0.7736,
      "num_input_tokens_seen": 21287584,
      "step": 36690
    },
    {
      "epoch": 5.465445338099494,
      "grad_norm": 8.0625,
      "learning_rate": 4.556527761916735e-05,
      "loss": 1.1758,
      "num_input_tokens_seen": 21290560,
      "step": 36695
    },
    {
      "epoch": 5.466190050640453,
      "grad_norm": 11.6875,
      "learning_rate": 4.556342981010205e-05,
      "loss": 1.2634,
      "num_input_tokens_seen": 21293824,
      "step": 36700
    },
    {
      "epoch": 5.466934763181412,
      "grad_norm": 7.125,
      "learning_rate": 4.5561581653638516e-05,
      "loss": 0.7156,
      "num_input_tokens_seen": 21296768,
      "step": 36705
    },
    {
      "epoch": 5.467679475722371,
      "grad_norm": 15.625,
      "learning_rate": 4.555973314980796e-05,
      "loss": 1.0839,
      "num_input_tokens_seen": 21300064,
      "step": 36710
    },
    {
      "epoch": 5.46842418826333,
      "grad_norm": 8.6875,
      "learning_rate": 4.555788429864161e-05,
      "loss": 1.1318,
      "num_input_tokens_seen": 21303104,
      "step": 36715
    },
    {
      "epoch": 5.46916890080429,
      "grad_norm": 11.875,
      "learning_rate": 4.5556035100170683e-05,
      "loss": 1.3109,
      "num_input_tokens_seen": 21305920,
      "step": 36720
    },
    {
      "epoch": 5.469913613345248,
      "grad_norm": 6.09375,
      "learning_rate": 4.555418555442645e-05,
      "loss": 1.2297,
      "num_input_tokens_seen": 21308640,
      "step": 36725
    },
    {
      "epoch": 5.470658325886208,
      "grad_norm": 14.1875,
      "learning_rate": 4.555233566144014e-05,
      "loss": 1.1166,
      "num_input_tokens_seen": 21311552,
      "step": 36730
    },
    {
      "epoch": 5.471403038427167,
      "grad_norm": 9.75,
      "learning_rate": 4.5550485421243006e-05,
      "loss": 0.6965,
      "num_input_tokens_seen": 21314464,
      "step": 36735
    },
    {
      "epoch": 5.4721477509681264,
      "grad_norm": 8.5625,
      "learning_rate": 4.554863483386631e-05,
      "loss": 0.9922,
      "num_input_tokens_seen": 21317280,
      "step": 36740
    },
    {
      "epoch": 5.472892463509085,
      "grad_norm": 6.8125,
      "learning_rate": 4.554678389934131e-05,
      "loss": 0.8747,
      "num_input_tokens_seen": 21320608,
      "step": 36745
    },
    {
      "epoch": 5.473637176050045,
      "grad_norm": 8.0625,
      "learning_rate": 4.554493261769928e-05,
      "loss": 0.9559,
      "num_input_tokens_seen": 21323488,
      "step": 36750
    },
    {
      "epoch": 5.474381888591004,
      "grad_norm": 7.9375,
      "learning_rate": 4.5543080988971484e-05,
      "loss": 1.4213,
      "num_input_tokens_seen": 21326176,
      "step": 36755
    },
    {
      "epoch": 5.475126601131963,
      "grad_norm": 15.5625,
      "learning_rate": 4.554122901318922e-05,
      "loss": 1.1121,
      "num_input_tokens_seen": 21328928,
      "step": 36760
    },
    {
      "epoch": 5.475871313672922,
      "grad_norm": 10.6875,
      "learning_rate": 4.553937669038378e-05,
      "loss": 1.0539,
      "num_input_tokens_seen": 21331744,
      "step": 36765
    },
    {
      "epoch": 5.476616026213882,
      "grad_norm": 5.53125,
      "learning_rate": 4.553752402058644e-05,
      "loss": 0.9434,
      "num_input_tokens_seen": 21334496,
      "step": 36770
    },
    {
      "epoch": 5.47736073875484,
      "grad_norm": 14.875,
      "learning_rate": 4.55356710038285e-05,
      "loss": 0.972,
      "num_input_tokens_seen": 21337440,
      "step": 36775
    },
    {
      "epoch": 5.4781054512958,
      "grad_norm": 15.375,
      "learning_rate": 4.5533817640141275e-05,
      "loss": 1.1945,
      "num_input_tokens_seen": 21340320,
      "step": 36780
    },
    {
      "epoch": 5.478850163836759,
      "grad_norm": 5.5,
      "learning_rate": 4.553196392955606e-05,
      "loss": 0.8281,
      "num_input_tokens_seen": 21342944,
      "step": 36785
    },
    {
      "epoch": 5.4795948763777185,
      "grad_norm": 8.625,
      "learning_rate": 4.55301098721042e-05,
      "loss": 0.8673,
      "num_input_tokens_seen": 21345856,
      "step": 36790
    },
    {
      "epoch": 5.480339588918677,
      "grad_norm": 11.5625,
      "learning_rate": 4.5528255467816994e-05,
      "loss": 1.2888,
      "num_input_tokens_seen": 21348992,
      "step": 36795
    },
    {
      "epoch": 5.481084301459637,
      "grad_norm": 22.5,
      "learning_rate": 4.552640071672577e-05,
      "loss": 1.3421,
      "num_input_tokens_seen": 21351648,
      "step": 36800
    },
    {
      "epoch": 5.481829014000596,
      "grad_norm": 4.875,
      "learning_rate": 4.552454561886187e-05,
      "loss": 0.8272,
      "num_input_tokens_seen": 21354272,
      "step": 36805
    },
    {
      "epoch": 5.482573726541555,
      "grad_norm": 19.0,
      "learning_rate": 4.5522690174256635e-05,
      "loss": 1.2647,
      "num_input_tokens_seen": 21356960,
      "step": 36810
    },
    {
      "epoch": 5.483318439082514,
      "grad_norm": 16.375,
      "learning_rate": 4.55208343829414e-05,
      "loss": 1.0771,
      "num_input_tokens_seen": 21360096,
      "step": 36815
    },
    {
      "epoch": 5.484063151623474,
      "grad_norm": 14.1875,
      "learning_rate": 4.551897824494753e-05,
      "loss": 0.9717,
      "num_input_tokens_seen": 21363584,
      "step": 36820
    },
    {
      "epoch": 5.4848078641644324,
      "grad_norm": 32.75,
      "learning_rate": 4.551712176030638e-05,
      "loss": 1.5132,
      "num_input_tokens_seen": 21366592,
      "step": 36825
    },
    {
      "epoch": 5.485552576705392,
      "grad_norm": 8.875,
      "learning_rate": 4.551526492904931e-05,
      "loss": 0.9498,
      "num_input_tokens_seen": 21369440,
      "step": 36830
    },
    {
      "epoch": 5.486297289246351,
      "grad_norm": 17.375,
      "learning_rate": 4.551340775120768e-05,
      "loss": 0.9286,
      "num_input_tokens_seen": 21372256,
      "step": 36835
    },
    {
      "epoch": 5.4870420017873105,
      "grad_norm": 10.3125,
      "learning_rate": 4.551155022681288e-05,
      "loss": 0.8761,
      "num_input_tokens_seen": 21374848,
      "step": 36840
    },
    {
      "epoch": 5.487786714328269,
      "grad_norm": 6.5,
      "learning_rate": 4.5509692355896296e-05,
      "loss": 1.2306,
      "num_input_tokens_seen": 21377728,
      "step": 36845
    },
    {
      "epoch": 5.488531426869228,
      "grad_norm": 13.875,
      "learning_rate": 4.550783413848929e-05,
      "loss": 0.989,
      "num_input_tokens_seen": 21380384,
      "step": 36850
    },
    {
      "epoch": 5.489276139410188,
      "grad_norm": 6.3125,
      "learning_rate": 4.550597557462328e-05,
      "loss": 1.0157,
      "num_input_tokens_seen": 21383136,
      "step": 36855
    },
    {
      "epoch": 5.490020851951147,
      "grad_norm": 7.09375,
      "learning_rate": 4.5504116664329656e-05,
      "loss": 0.7546,
      "num_input_tokens_seen": 21385920,
      "step": 36860
    },
    {
      "epoch": 5.490765564492106,
      "grad_norm": 5.5625,
      "learning_rate": 4.550225740763981e-05,
      "loss": 0.8865,
      "num_input_tokens_seen": 21388576,
      "step": 36865
    },
    {
      "epoch": 5.491510277033065,
      "grad_norm": 10.1875,
      "learning_rate": 4.5500397804585166e-05,
      "loss": 0.9773,
      "num_input_tokens_seen": 21391456,
      "step": 36870
    },
    {
      "epoch": 5.4922549895740245,
      "grad_norm": 9.125,
      "learning_rate": 4.5498537855197145e-05,
      "loss": 0.9391,
      "num_input_tokens_seen": 21394304,
      "step": 36875
    },
    {
      "epoch": 5.492999702114983,
      "grad_norm": 11.375,
      "learning_rate": 4.549667755950715e-05,
      "loss": 1.2711,
      "num_input_tokens_seen": 21397024,
      "step": 36880
    },
    {
      "epoch": 5.493744414655943,
      "grad_norm": 10.5,
      "learning_rate": 4.5494816917546625e-05,
      "loss": 1.1764,
      "num_input_tokens_seen": 21399872,
      "step": 36885
    },
    {
      "epoch": 5.494489127196902,
      "grad_norm": 12.25,
      "learning_rate": 4.549295592934699e-05,
      "loss": 1.0536,
      "num_input_tokens_seen": 21402784,
      "step": 36890
    },
    {
      "epoch": 5.495233839737861,
      "grad_norm": 7.78125,
      "learning_rate": 4.5491094594939705e-05,
      "loss": 0.8224,
      "num_input_tokens_seen": 21405984,
      "step": 36895
    },
    {
      "epoch": 5.49597855227882,
      "grad_norm": 23.0,
      "learning_rate": 4.5489232914356196e-05,
      "loss": 1.1424,
      "num_input_tokens_seen": 21409120,
      "step": 36900
    },
    {
      "epoch": 5.49672326481978,
      "grad_norm": 4.125,
      "learning_rate": 4.548737088762792e-05,
      "loss": 0.6411,
      "num_input_tokens_seen": 21411680,
      "step": 36905
    },
    {
      "epoch": 5.4974679773607384,
      "grad_norm": 6.8125,
      "learning_rate": 4.548550851478634e-05,
      "loss": 1.0775,
      "num_input_tokens_seen": 21414752,
      "step": 36910
    },
    {
      "epoch": 5.498212689901698,
      "grad_norm": 18.25,
      "learning_rate": 4.548364579586291e-05,
      "loss": 0.981,
      "num_input_tokens_seen": 21417504,
      "step": 36915
    },
    {
      "epoch": 5.498957402442657,
      "grad_norm": 15.6875,
      "learning_rate": 4.548178273088911e-05,
      "loss": 0.7379,
      "num_input_tokens_seen": 21420736,
      "step": 36920
    },
    {
      "epoch": 5.4997021149836165,
      "grad_norm": 14.5625,
      "learning_rate": 4.54799193198964e-05,
      "loss": 0.9003,
      "num_input_tokens_seen": 21423680,
      "step": 36925
    },
    {
      "epoch": 5.500446827524575,
      "grad_norm": 12.625,
      "learning_rate": 4.547805556291627e-05,
      "loss": 1.3053,
      "num_input_tokens_seen": 21426464,
      "step": 36930
    },
    {
      "epoch": 5.501191540065535,
      "grad_norm": 4.25,
      "learning_rate": 4.54761914599802e-05,
      "loss": 0.8465,
      "num_input_tokens_seen": 21429632,
      "step": 36935
    },
    {
      "epoch": 5.501936252606494,
      "grad_norm": 14.4375,
      "learning_rate": 4.54743270111197e-05,
      "loss": 1.0376,
      "num_input_tokens_seen": 21432928,
      "step": 36940
    },
    {
      "epoch": 5.502680965147453,
      "grad_norm": 13.125,
      "learning_rate": 4.547246221636624e-05,
      "loss": 1.2124,
      "num_input_tokens_seen": 21436032,
      "step": 36945
    },
    {
      "epoch": 5.503425677688412,
      "grad_norm": 17.75,
      "learning_rate": 4.5470597075751345e-05,
      "loss": 1.1963,
      "num_input_tokens_seen": 21438976,
      "step": 36950
    },
    {
      "epoch": 5.504170390229372,
      "grad_norm": 10.625,
      "learning_rate": 4.5468731589306516e-05,
      "loss": 1.2429,
      "num_input_tokens_seen": 21442016,
      "step": 36955
    },
    {
      "epoch": 5.5049151027703305,
      "grad_norm": 5.90625,
      "learning_rate": 4.546686575706327e-05,
      "loss": 0.6372,
      "num_input_tokens_seen": 21444768,
      "step": 36960
    },
    {
      "epoch": 5.50565981531129,
      "grad_norm": 7.9375,
      "learning_rate": 4.546499957905313e-05,
      "loss": 1.0165,
      "num_input_tokens_seen": 21447456,
      "step": 36965
    },
    {
      "epoch": 5.506404527852249,
      "grad_norm": 9.5625,
      "learning_rate": 4.546313305530762e-05,
      "loss": 0.808,
      "num_input_tokens_seen": 21450656,
      "step": 36970
    },
    {
      "epoch": 5.5071492403932085,
      "grad_norm": 14.3125,
      "learning_rate": 4.546126618585828e-05,
      "loss": 1.0874,
      "num_input_tokens_seen": 21453600,
      "step": 36975
    },
    {
      "epoch": 5.507893952934167,
      "grad_norm": 6.59375,
      "learning_rate": 4.5459398970736636e-05,
      "loss": 1.0308,
      "num_input_tokens_seen": 21456640,
      "step": 36980
    },
    {
      "epoch": 5.508638665475127,
      "grad_norm": 6.90625,
      "learning_rate": 4.545753140997424e-05,
      "loss": 1.019,
      "num_input_tokens_seen": 21459392,
      "step": 36985
    },
    {
      "epoch": 5.509383378016086,
      "grad_norm": 8.9375,
      "learning_rate": 4.545566350360265e-05,
      "loss": 1.2676,
      "num_input_tokens_seen": 21462656,
      "step": 36990
    },
    {
      "epoch": 5.510128090557045,
      "grad_norm": 9.3125,
      "learning_rate": 4.5453795251653416e-05,
      "loss": 1.0193,
      "num_input_tokens_seen": 21465376,
      "step": 36995
    },
    {
      "epoch": 5.510872803098004,
      "grad_norm": 10.6875,
      "learning_rate": 4.545192665415809e-05,
      "loss": 1.1633,
      "num_input_tokens_seen": 21468224,
      "step": 37000
    },
    {
      "epoch": 5.511617515638964,
      "grad_norm": 10.3125,
      "learning_rate": 4.545005771114826e-05,
      "loss": 1.3118,
      "num_input_tokens_seen": 21470848,
      "step": 37005
    },
    {
      "epoch": 5.5123622281799225,
      "grad_norm": 12.3125,
      "learning_rate": 4.544818842265548e-05,
      "loss": 1.1661,
      "num_input_tokens_seen": 21473504,
      "step": 37010
    },
    {
      "epoch": 5.513106940720881,
      "grad_norm": 5.53125,
      "learning_rate": 4.544631878871135e-05,
      "loss": 0.8921,
      "num_input_tokens_seen": 21476384,
      "step": 37015
    },
    {
      "epoch": 5.513851653261841,
      "grad_norm": 19.375,
      "learning_rate": 4.544444880934744e-05,
      "loss": 1.4248,
      "num_input_tokens_seen": 21478912,
      "step": 37020
    },
    {
      "epoch": 5.5145963658028005,
      "grad_norm": 8.625,
      "learning_rate": 4.5442578484595346e-05,
      "loss": 1.19,
      "num_input_tokens_seen": 21481664,
      "step": 37025
    },
    {
      "epoch": 5.515341078343759,
      "grad_norm": 12.5,
      "learning_rate": 4.544070781448666e-05,
      "loss": 1.1389,
      "num_input_tokens_seen": 21484576,
      "step": 37030
    },
    {
      "epoch": 5.516085790884718,
      "grad_norm": 9.0,
      "learning_rate": 4.5438836799053e-05,
      "loss": 1.2157,
      "num_input_tokens_seen": 21487552,
      "step": 37035
    },
    {
      "epoch": 5.516830503425678,
      "grad_norm": 3.828125,
      "learning_rate": 4.5436965438325953e-05,
      "loss": 0.8069,
      "num_input_tokens_seen": 21490432,
      "step": 37040
    },
    {
      "epoch": 5.517575215966637,
      "grad_norm": 12.1875,
      "learning_rate": 4.543509373233715e-05,
      "loss": 1.1768,
      "num_input_tokens_seen": 21493088,
      "step": 37045
    },
    {
      "epoch": 5.518319928507596,
      "grad_norm": 12.375,
      "learning_rate": 4.5433221681118215e-05,
      "loss": 0.9587,
      "num_input_tokens_seen": 21496064,
      "step": 37050
    },
    {
      "epoch": 5.519064641048555,
      "grad_norm": 6.65625,
      "learning_rate": 4.5431349284700764e-05,
      "loss": 0.8747,
      "num_input_tokens_seen": 21498784,
      "step": 37055
    },
    {
      "epoch": 5.5198093535895145,
      "grad_norm": 6.28125,
      "learning_rate": 4.542947654311643e-05,
      "loss": 1.1093,
      "num_input_tokens_seen": 21501824,
      "step": 37060
    },
    {
      "epoch": 5.520554066130473,
      "grad_norm": 21.625,
      "learning_rate": 4.542760345639686e-05,
      "loss": 1.2117,
      "num_input_tokens_seen": 21505056,
      "step": 37065
    },
    {
      "epoch": 5.521298778671433,
      "grad_norm": 14.25,
      "learning_rate": 4.542573002457368e-05,
      "loss": 1.1412,
      "num_input_tokens_seen": 21507680,
      "step": 37070
    },
    {
      "epoch": 5.522043491212392,
      "grad_norm": 11.75,
      "learning_rate": 4.5423856247678556e-05,
      "loss": 0.8819,
      "num_input_tokens_seen": 21510400,
      "step": 37075
    },
    {
      "epoch": 5.522788203753351,
      "grad_norm": 8.0,
      "learning_rate": 4.542198212574314e-05,
      "loss": 1.0118,
      "num_input_tokens_seen": 21513152,
      "step": 37080
    },
    {
      "epoch": 5.52353291629431,
      "grad_norm": 14.25,
      "learning_rate": 4.5420107658799094e-05,
      "loss": 1.2441,
      "num_input_tokens_seen": 21516160,
      "step": 37085
    },
    {
      "epoch": 5.52427762883527,
      "grad_norm": 8.125,
      "learning_rate": 4.541823284687808e-05,
      "loss": 1.1954,
      "num_input_tokens_seen": 21519008,
      "step": 37090
    },
    {
      "epoch": 5.5250223413762285,
      "grad_norm": 7.875,
      "learning_rate": 4.541635769001178e-05,
      "loss": 0.9457,
      "num_input_tokens_seen": 21521824,
      "step": 37095
    },
    {
      "epoch": 5.525767053917188,
      "grad_norm": 17.0,
      "learning_rate": 4.5414482188231864e-05,
      "loss": 0.9735,
      "num_input_tokens_seen": 21524672,
      "step": 37100
    },
    {
      "epoch": 5.526511766458147,
      "grad_norm": 9.1875,
      "learning_rate": 4.5412606341570016e-05,
      "loss": 1.0779,
      "num_input_tokens_seen": 21527488,
      "step": 37105
    },
    {
      "epoch": 5.5272564789991065,
      "grad_norm": 17.0,
      "learning_rate": 4.5410730150057935e-05,
      "loss": 1.4439,
      "num_input_tokens_seen": 21530528,
      "step": 37110
    },
    {
      "epoch": 5.528001191540065,
      "grad_norm": 6.84375,
      "learning_rate": 4.5408853613727307e-05,
      "loss": 0.9618,
      "num_input_tokens_seen": 21533280,
      "step": 37115
    },
    {
      "epoch": 5.528745904081025,
      "grad_norm": 19.75,
      "learning_rate": 4.540697673260984e-05,
      "loss": 1.2806,
      "num_input_tokens_seen": 21535936,
      "step": 37120
    },
    {
      "epoch": 5.529490616621984,
      "grad_norm": 5.78125,
      "learning_rate": 4.5405099506737244e-05,
      "loss": 1.0406,
      "num_input_tokens_seen": 21538720,
      "step": 37125
    },
    {
      "epoch": 5.530235329162943,
      "grad_norm": 8.25,
      "learning_rate": 4.540322193614123e-05,
      "loss": 1.0977,
      "num_input_tokens_seen": 21541728,
      "step": 37130
    },
    {
      "epoch": 5.530980041703902,
      "grad_norm": 8.375,
      "learning_rate": 4.540134402085352e-05,
      "loss": 1.128,
      "num_input_tokens_seen": 21544768,
      "step": 37135
    },
    {
      "epoch": 5.531724754244862,
      "grad_norm": 3.859375,
      "learning_rate": 4.539946576090584e-05,
      "loss": 0.8107,
      "num_input_tokens_seen": 21547840,
      "step": 37140
    },
    {
      "epoch": 5.5324694667858205,
      "grad_norm": 7.34375,
      "learning_rate": 4.539758715632992e-05,
      "loss": 1.0975,
      "num_input_tokens_seen": 21551872,
      "step": 37145
    },
    {
      "epoch": 5.53321417932678,
      "grad_norm": 12.125,
      "learning_rate": 4.539570820715749e-05,
      "loss": 1.2731,
      "num_input_tokens_seen": 21554880,
      "step": 37150
    },
    {
      "epoch": 5.533958891867739,
      "grad_norm": 9.625,
      "learning_rate": 4.53938289134203e-05,
      "loss": 1.0957,
      "num_input_tokens_seen": 21557632,
      "step": 37155
    },
    {
      "epoch": 5.534703604408699,
      "grad_norm": 12.9375,
      "learning_rate": 4.5391949275150104e-05,
      "loss": 0.9897,
      "num_input_tokens_seen": 21560352,
      "step": 37160
    },
    {
      "epoch": 5.535448316949657,
      "grad_norm": 6.5,
      "learning_rate": 4.539006929237864e-05,
      "loss": 0.947,
      "num_input_tokens_seen": 21563104,
      "step": 37165
    },
    {
      "epoch": 5.536193029490617,
      "grad_norm": 5.5625,
      "learning_rate": 4.538818896513769e-05,
      "loss": 0.7657,
      "num_input_tokens_seen": 21565952,
      "step": 37170
    },
    {
      "epoch": 5.536937742031576,
      "grad_norm": 14.75,
      "learning_rate": 4.5386308293459e-05,
      "loss": 1.0509,
      "num_input_tokens_seen": 21568704,
      "step": 37175
    },
    {
      "epoch": 5.537682454572535,
      "grad_norm": 11.5625,
      "learning_rate": 4.5384427277374355e-05,
      "loss": 1.0697,
      "num_input_tokens_seen": 21571584,
      "step": 37180
    },
    {
      "epoch": 5.538427167113494,
      "grad_norm": 4.65625,
      "learning_rate": 4.538254591691553e-05,
      "loss": 1.2179,
      "num_input_tokens_seen": 21574560,
      "step": 37185
    },
    {
      "epoch": 5.539171879654454,
      "grad_norm": 13.1875,
      "learning_rate": 4.538066421211431e-05,
      "loss": 1.2879,
      "num_input_tokens_seen": 21577504,
      "step": 37190
    },
    {
      "epoch": 5.5399165921954125,
      "grad_norm": 9.375,
      "learning_rate": 4.5378782163002476e-05,
      "loss": 0.9278,
      "num_input_tokens_seen": 21580288,
      "step": 37195
    },
    {
      "epoch": 5.540661304736371,
      "grad_norm": 9.375,
      "learning_rate": 4.537689976961184e-05,
      "loss": 1.0256,
      "num_input_tokens_seen": 21583264,
      "step": 37200
    },
    {
      "epoch": 5.541406017277331,
      "grad_norm": 8.625,
      "learning_rate": 4.537501703197418e-05,
      "loss": 0.9285,
      "num_input_tokens_seen": 21586080,
      "step": 37205
    },
    {
      "epoch": 5.542150729818291,
      "grad_norm": 7.1875,
      "learning_rate": 4.5373133950121314e-05,
      "loss": 0.6813,
      "num_input_tokens_seen": 21588768,
      "step": 37210
    },
    {
      "epoch": 5.542895442359249,
      "grad_norm": 8.0,
      "learning_rate": 4.537125052408506e-05,
      "loss": 0.8899,
      "num_input_tokens_seen": 21591616,
      "step": 37215
    },
    {
      "epoch": 5.543640154900208,
      "grad_norm": 8.9375,
      "learning_rate": 4.536936675389724e-05,
      "loss": 0.6868,
      "num_input_tokens_seen": 21594720,
      "step": 37220
    },
    {
      "epoch": 5.544384867441168,
      "grad_norm": 9.0,
      "learning_rate": 4.5367482639589665e-05,
      "loss": 0.9576,
      "num_input_tokens_seen": 21597408,
      "step": 37225
    },
    {
      "epoch": 5.5451295799821265,
      "grad_norm": 7.46875,
      "learning_rate": 4.536559818119418e-05,
      "loss": 0.9154,
      "num_input_tokens_seen": 21600096,
      "step": 37230
    },
    {
      "epoch": 5.545874292523086,
      "grad_norm": 7.75,
      "learning_rate": 4.53637133787426e-05,
      "loss": 1.3058,
      "num_input_tokens_seen": 21603072,
      "step": 37235
    },
    {
      "epoch": 5.546619005064045,
      "grad_norm": 5.84375,
      "learning_rate": 4.536182823226678e-05,
      "loss": 0.8384,
      "num_input_tokens_seen": 21606112,
      "step": 37240
    },
    {
      "epoch": 5.547363717605005,
      "grad_norm": 13.625,
      "learning_rate": 4.535994274179858e-05,
      "loss": 0.9328,
      "num_input_tokens_seen": 21609600,
      "step": 37245
    },
    {
      "epoch": 5.548108430145963,
      "grad_norm": 7.0,
      "learning_rate": 4.535805690736983e-05,
      "loss": 0.9847,
      "num_input_tokens_seen": 21612576,
      "step": 37250
    },
    {
      "epoch": 5.548853142686923,
      "grad_norm": 11.5,
      "learning_rate": 4.535617072901239e-05,
      "loss": 1.0618,
      "num_input_tokens_seen": 21615264,
      "step": 37255
    },
    {
      "epoch": 5.549597855227882,
      "grad_norm": 4.875,
      "learning_rate": 4.535428420675816e-05,
      "loss": 1.0348,
      "num_input_tokens_seen": 21617984,
      "step": 37260
    },
    {
      "epoch": 5.550342567768841,
      "grad_norm": 9.625,
      "learning_rate": 4.535239734063896e-05,
      "loss": 1.341,
      "num_input_tokens_seen": 21620896,
      "step": 37265
    },
    {
      "epoch": 5.5510872803098,
      "grad_norm": 8.375,
      "learning_rate": 4.535051013068671e-05,
      "loss": 0.9589,
      "num_input_tokens_seen": 21623872,
      "step": 37270
    },
    {
      "epoch": 5.55183199285076,
      "grad_norm": 5.5625,
      "learning_rate": 4.5348622576933265e-05,
      "loss": 0.9865,
      "num_input_tokens_seen": 21627232,
      "step": 37275
    },
    {
      "epoch": 5.5525767053917185,
      "grad_norm": 16.25,
      "learning_rate": 4.534673467941053e-05,
      "loss": 1.3154,
      "num_input_tokens_seen": 21630176,
      "step": 37280
    },
    {
      "epoch": 5.553321417932678,
      "grad_norm": 10.375,
      "learning_rate": 4.534484643815038e-05,
      "loss": 1.1297,
      "num_input_tokens_seen": 21632832,
      "step": 37285
    },
    {
      "epoch": 5.554066130473637,
      "grad_norm": 6.75,
      "learning_rate": 4.534295785318474e-05,
      "loss": 1.1196,
      "num_input_tokens_seen": 21635744,
      "step": 37290
    },
    {
      "epoch": 5.554810843014597,
      "grad_norm": 12.0,
      "learning_rate": 4.53410689245455e-05,
      "loss": 1.0902,
      "num_input_tokens_seen": 21638464,
      "step": 37295
    },
    {
      "epoch": 5.555555555555555,
      "grad_norm": 9.25,
      "learning_rate": 4.5339179652264576e-05,
      "loss": 0.7731,
      "num_input_tokens_seen": 21641344,
      "step": 37300
    },
    {
      "epoch": 5.556300268096515,
      "grad_norm": 12.0,
      "learning_rate": 4.5337290036373875e-05,
      "loss": 1.322,
      "num_input_tokens_seen": 21644224,
      "step": 37305
    },
    {
      "epoch": 5.557044980637474,
      "grad_norm": 10.5,
      "learning_rate": 4.533540007690533e-05,
      "loss": 1.1064,
      "num_input_tokens_seen": 21647232,
      "step": 37310
    },
    {
      "epoch": 5.557789693178433,
      "grad_norm": 7.46875,
      "learning_rate": 4.533350977389087e-05,
      "loss": 1.3565,
      "num_input_tokens_seen": 21650336,
      "step": 37315
    },
    {
      "epoch": 5.558534405719392,
      "grad_norm": 12.25,
      "learning_rate": 4.533161912736243e-05,
      "loss": 1.1872,
      "num_input_tokens_seen": 21653024,
      "step": 37320
    },
    {
      "epoch": 5.559279118260352,
      "grad_norm": 6.125,
      "learning_rate": 4.532972813735196e-05,
      "loss": 1.5272,
      "num_input_tokens_seen": 21656000,
      "step": 37325
    },
    {
      "epoch": 5.560023830801311,
      "grad_norm": 20.5,
      "learning_rate": 4.532783680389138e-05,
      "loss": 0.7364,
      "num_input_tokens_seen": 21658848,
      "step": 37330
    },
    {
      "epoch": 5.56076854334227,
      "grad_norm": 15.75,
      "learning_rate": 4.532594512701266e-05,
      "loss": 1.0622,
      "num_input_tokens_seen": 21661952,
      "step": 37335
    },
    {
      "epoch": 5.561513255883229,
      "grad_norm": 12.0625,
      "learning_rate": 4.532405310674776e-05,
      "loss": 1.5312,
      "num_input_tokens_seen": 21665248,
      "step": 37340
    },
    {
      "epoch": 5.562257968424189,
      "grad_norm": 6.875,
      "learning_rate": 4.532216074312864e-05,
      "loss": 1.1574,
      "num_input_tokens_seen": 21668416,
      "step": 37345
    },
    {
      "epoch": 5.563002680965147,
      "grad_norm": 14.8125,
      "learning_rate": 4.5320268036187266e-05,
      "loss": 1.4745,
      "num_input_tokens_seen": 21671104,
      "step": 37350
    },
    {
      "epoch": 5.563747393506107,
      "grad_norm": 10.75,
      "learning_rate": 4.531837498595561e-05,
      "loss": 1.1249,
      "num_input_tokens_seen": 21674208,
      "step": 37355
    },
    {
      "epoch": 5.564492106047066,
      "grad_norm": 6.1875,
      "learning_rate": 4.531648159246567e-05,
      "loss": 1.2675,
      "num_input_tokens_seen": 21677440,
      "step": 37360
    },
    {
      "epoch": 5.5652368185880245,
      "grad_norm": 11.9375,
      "learning_rate": 4.531458785574941e-05,
      "loss": 1.1543,
      "num_input_tokens_seen": 21680576,
      "step": 37365
    },
    {
      "epoch": 5.565981531128984,
      "grad_norm": 13.5,
      "learning_rate": 4.531269377583885e-05,
      "loss": 1.0646,
      "num_input_tokens_seen": 21683392,
      "step": 37370
    },
    {
      "epoch": 5.566726243669944,
      "grad_norm": 7.9375,
      "learning_rate": 4.5310799352765964e-05,
      "loss": 0.9337,
      "num_input_tokens_seen": 21686208,
      "step": 37375
    },
    {
      "epoch": 5.567470956210903,
      "grad_norm": 12.0625,
      "learning_rate": 4.5308904586562774e-05,
      "loss": 1.1715,
      "num_input_tokens_seen": 21689184,
      "step": 37380
    },
    {
      "epoch": 5.568215668751861,
      "grad_norm": 12.625,
      "learning_rate": 4.530700947726127e-05,
      "loss": 1.2483,
      "num_input_tokens_seen": 21691904,
      "step": 37385
    },
    {
      "epoch": 5.568960381292821,
      "grad_norm": 11.3125,
      "learning_rate": 4.530511402489349e-05,
      "loss": 1.126,
      "num_input_tokens_seen": 21694688,
      "step": 37390
    },
    {
      "epoch": 5.569705093833781,
      "grad_norm": 6.8125,
      "learning_rate": 4.530321822949144e-05,
      "loss": 1.3842,
      "num_input_tokens_seen": 21697216,
      "step": 37395
    },
    {
      "epoch": 5.570449806374739,
      "grad_norm": 10.5625,
      "learning_rate": 4.530132209108715e-05,
      "loss": 1.0219,
      "num_input_tokens_seen": 21700000,
      "step": 37400
    },
    {
      "epoch": 5.571194518915698,
      "grad_norm": 17.0,
      "learning_rate": 4.529942560971266e-05,
      "loss": 1.1136,
      "num_input_tokens_seen": 21702656,
      "step": 37405
    },
    {
      "epoch": 5.571939231456658,
      "grad_norm": 9.125,
      "learning_rate": 4.529752878540001e-05,
      "loss": 0.9405,
      "num_input_tokens_seen": 21705664,
      "step": 37410
    },
    {
      "epoch": 5.572683943997617,
      "grad_norm": 8.1875,
      "learning_rate": 4.529563161818124e-05,
      "loss": 0.7985,
      "num_input_tokens_seen": 21708384,
      "step": 37415
    },
    {
      "epoch": 5.573428656538576,
      "grad_norm": 15.375,
      "learning_rate": 4.529373410808841e-05,
      "loss": 1.3048,
      "num_input_tokens_seen": 21711424,
      "step": 37420
    },
    {
      "epoch": 5.574173369079535,
      "grad_norm": 7.96875,
      "learning_rate": 4.5291836255153555e-05,
      "loss": 1.4375,
      "num_input_tokens_seen": 21714336,
      "step": 37425
    },
    {
      "epoch": 5.574918081620495,
      "grad_norm": 11.25,
      "learning_rate": 4.528993805940874e-05,
      "loss": 1.1705,
      "num_input_tokens_seen": 21717184,
      "step": 37430
    },
    {
      "epoch": 5.575662794161453,
      "grad_norm": 6.5,
      "learning_rate": 4.528803952088606e-05,
      "loss": 0.9181,
      "num_input_tokens_seen": 21720224,
      "step": 37435
    },
    {
      "epoch": 5.576407506702413,
      "grad_norm": 7.96875,
      "learning_rate": 4.5286140639617566e-05,
      "loss": 1.2356,
      "num_input_tokens_seen": 21722912,
      "step": 37440
    },
    {
      "epoch": 5.577152219243372,
      "grad_norm": 8.875,
      "learning_rate": 4.528424141563535e-05,
      "loss": 0.9622,
      "num_input_tokens_seen": 21725760,
      "step": 37445
    },
    {
      "epoch": 5.577896931784331,
      "grad_norm": 10.8125,
      "learning_rate": 4.528234184897149e-05,
      "loss": 0.9574,
      "num_input_tokens_seen": 21728704,
      "step": 37450
    },
    {
      "epoch": 5.57864164432529,
      "grad_norm": 9.5,
      "learning_rate": 4.528044193965807e-05,
      "loss": 0.7962,
      "num_input_tokens_seen": 21731616,
      "step": 37455
    },
    {
      "epoch": 5.57938635686625,
      "grad_norm": 7.8125,
      "learning_rate": 4.527854168772721e-05,
      "loss": 0.9353,
      "num_input_tokens_seen": 21734560,
      "step": 37460
    },
    {
      "epoch": 5.580131069407209,
      "grad_norm": 12.0625,
      "learning_rate": 4.527664109321098e-05,
      "loss": 1.0652,
      "num_input_tokens_seen": 21737280,
      "step": 37465
    },
    {
      "epoch": 5.580875781948168,
      "grad_norm": 13.0,
      "learning_rate": 4.5274740156141516e-05,
      "loss": 1.0916,
      "num_input_tokens_seen": 21740032,
      "step": 37470
    },
    {
      "epoch": 5.581620494489127,
      "grad_norm": 9.4375,
      "learning_rate": 4.527283887655093e-05,
      "loss": 1.0519,
      "num_input_tokens_seen": 21742848,
      "step": 37475
    },
    {
      "epoch": 5.582365207030087,
      "grad_norm": 9.125,
      "learning_rate": 4.5270937254471325e-05,
      "loss": 1.1732,
      "num_input_tokens_seen": 21745600,
      "step": 37480
    },
    {
      "epoch": 5.583109919571045,
      "grad_norm": 8.0,
      "learning_rate": 4.526903528993484e-05,
      "loss": 0.9681,
      "num_input_tokens_seen": 21748416,
      "step": 37485
    },
    {
      "epoch": 5.583854632112005,
      "grad_norm": 11.25,
      "learning_rate": 4.526713298297361e-05,
      "loss": 1.2561,
      "num_input_tokens_seen": 21751328,
      "step": 37490
    },
    {
      "epoch": 5.584599344652964,
      "grad_norm": 12.3125,
      "learning_rate": 4.526523033361976e-05,
      "loss": 1.3053,
      "num_input_tokens_seen": 21754144,
      "step": 37495
    },
    {
      "epoch": 5.5853440571939235,
      "grad_norm": 7.59375,
      "learning_rate": 4.5263327341905443e-05,
      "loss": 1.0431,
      "num_input_tokens_seen": 21756704,
      "step": 37500
    },
    {
      "epoch": 5.586088769734882,
      "grad_norm": 21.0,
      "learning_rate": 4.526142400786281e-05,
      "loss": 1.1241,
      "num_input_tokens_seen": 21759872,
      "step": 37505
    },
    {
      "epoch": 5.586833482275842,
      "grad_norm": 9.25,
      "learning_rate": 4.5259520331524004e-05,
      "loss": 1.0283,
      "num_input_tokens_seen": 21762656,
      "step": 37510
    },
    {
      "epoch": 5.587578194816801,
      "grad_norm": 4.34375,
      "learning_rate": 4.525761631292119e-05,
      "loss": 0.7471,
      "num_input_tokens_seen": 21765408,
      "step": 37515
    },
    {
      "epoch": 5.58832290735776,
      "grad_norm": 12.75,
      "learning_rate": 4.5255711952086545e-05,
      "loss": 0.8598,
      "num_input_tokens_seen": 21768064,
      "step": 37520
    },
    {
      "epoch": 5.589067619898719,
      "grad_norm": 8.4375,
      "learning_rate": 4.525380724905224e-05,
      "loss": 1.1357,
      "num_input_tokens_seen": 21771136,
      "step": 37525
    },
    {
      "epoch": 5.589812332439678,
      "grad_norm": 6.34375,
      "learning_rate": 4.525190220385043e-05,
      "loss": 1.0218,
      "num_input_tokens_seen": 21773920,
      "step": 37530
    },
    {
      "epoch": 5.590557044980637,
      "grad_norm": 7.25,
      "learning_rate": 4.5249996816513325e-05,
      "loss": 1.0627,
      "num_input_tokens_seen": 21776800,
      "step": 37535
    },
    {
      "epoch": 5.591301757521597,
      "grad_norm": 15.5625,
      "learning_rate": 4.52480910870731e-05,
      "loss": 1.1209,
      "num_input_tokens_seen": 21780000,
      "step": 37540
    },
    {
      "epoch": 5.592046470062556,
      "grad_norm": 6.8125,
      "learning_rate": 4.524618501556196e-05,
      "loss": 1.2026,
      "num_input_tokens_seen": 21782944,
      "step": 37545
    },
    {
      "epoch": 5.592791182603515,
      "grad_norm": 10.4375,
      "learning_rate": 4.52442786020121e-05,
      "loss": 0.8181,
      "num_input_tokens_seen": 21785824,
      "step": 37550
    },
    {
      "epoch": 5.593535895144474,
      "grad_norm": 10.875,
      "learning_rate": 4.524237184645573e-05,
      "loss": 0.9918,
      "num_input_tokens_seen": 21788672,
      "step": 37555
    },
    {
      "epoch": 5.594280607685434,
      "grad_norm": 8.625,
      "learning_rate": 4.524046474892506e-05,
      "loss": 1.1314,
      "num_input_tokens_seen": 21791552,
      "step": 37560
    },
    {
      "epoch": 5.595025320226393,
      "grad_norm": 14.25,
      "learning_rate": 4.523855730945231e-05,
      "loss": 1.0109,
      "num_input_tokens_seen": 21794624,
      "step": 37565
    },
    {
      "epoch": 5.595770032767351,
      "grad_norm": 14.4375,
      "learning_rate": 4.52366495280697e-05,
      "loss": 1.1852,
      "num_input_tokens_seen": 21797664,
      "step": 37570
    },
    {
      "epoch": 5.596514745308311,
      "grad_norm": 4.625,
      "learning_rate": 4.523474140480947e-05,
      "loss": 1.0737,
      "num_input_tokens_seen": 21800576,
      "step": 37575
    },
    {
      "epoch": 5.59725945784927,
      "grad_norm": 4.8125,
      "learning_rate": 4.5232832939703846e-05,
      "loss": 1.1118,
      "num_input_tokens_seen": 21803648,
      "step": 37580
    },
    {
      "epoch": 5.5980041703902295,
      "grad_norm": 7.84375,
      "learning_rate": 4.5230924132785066e-05,
      "loss": 0.9861,
      "num_input_tokens_seen": 21806624,
      "step": 37585
    },
    {
      "epoch": 5.598748882931188,
      "grad_norm": 10.75,
      "learning_rate": 4.52290149840854e-05,
      "loss": 1.1353,
      "num_input_tokens_seen": 21809728,
      "step": 37590
    },
    {
      "epoch": 5.599493595472148,
      "grad_norm": 15.5,
      "learning_rate": 4.522710549363708e-05,
      "loss": 1.1908,
      "num_input_tokens_seen": 21812576,
      "step": 37595
    },
    {
      "epoch": 5.600238308013107,
      "grad_norm": 12.5,
      "learning_rate": 4.5225195661472364e-05,
      "loss": 0.9124,
      "num_input_tokens_seen": 21815552,
      "step": 37600
    },
    {
      "epoch": 5.600983020554066,
      "grad_norm": 5.25,
      "learning_rate": 4.522328548762353e-05,
      "loss": 0.9999,
      "num_input_tokens_seen": 21818592,
      "step": 37605
    },
    {
      "epoch": 5.601727733095025,
      "grad_norm": 4.71875,
      "learning_rate": 4.5221374972122837e-05,
      "loss": 1.1625,
      "num_input_tokens_seen": 21821376,
      "step": 37610
    },
    {
      "epoch": 5.602472445635985,
      "grad_norm": 10.75,
      "learning_rate": 4.521946411500257e-05,
      "loss": 1.0839,
      "num_input_tokens_seen": 21824160,
      "step": 37615
    },
    {
      "epoch": 5.603217158176943,
      "grad_norm": 10.4375,
      "learning_rate": 4.5217552916295e-05,
      "loss": 1.0459,
      "num_input_tokens_seen": 21827072,
      "step": 37620
    },
    {
      "epoch": 5.603961870717903,
      "grad_norm": 18.25,
      "learning_rate": 4.521564137603244e-05,
      "loss": 1.0754,
      "num_input_tokens_seen": 21829728,
      "step": 37625
    },
    {
      "epoch": 5.604706583258862,
      "grad_norm": 7.21875,
      "learning_rate": 4.521372949424715e-05,
      "loss": 0.8962,
      "num_input_tokens_seen": 21832416,
      "step": 37630
    },
    {
      "epoch": 5.6054512957998215,
      "grad_norm": 6.84375,
      "learning_rate": 4.521181727097144e-05,
      "loss": 0.8193,
      "num_input_tokens_seen": 21835072,
      "step": 37635
    },
    {
      "epoch": 5.60619600834078,
      "grad_norm": 13.375,
      "learning_rate": 4.5209904706237626e-05,
      "loss": 1.1519,
      "num_input_tokens_seen": 21837952,
      "step": 37640
    },
    {
      "epoch": 5.60694072088174,
      "grad_norm": 16.375,
      "learning_rate": 4.5207991800078015e-05,
      "loss": 0.8807,
      "num_input_tokens_seen": 21841056,
      "step": 37645
    },
    {
      "epoch": 5.607685433422699,
      "grad_norm": 11.8125,
      "learning_rate": 4.520607855252492e-05,
      "loss": 1.4138,
      "num_input_tokens_seen": 21843936,
      "step": 37650
    },
    {
      "epoch": 5.608430145963658,
      "grad_norm": 4.5625,
      "learning_rate": 4.520416496361066e-05,
      "loss": 0.6307,
      "num_input_tokens_seen": 21846624,
      "step": 37655
    },
    {
      "epoch": 5.609174858504617,
      "grad_norm": 11.875,
      "learning_rate": 4.5202251033367574e-05,
      "loss": 1.0628,
      "num_input_tokens_seen": 21849600,
      "step": 37660
    },
    {
      "epoch": 5.609919571045577,
      "grad_norm": 12.75,
      "learning_rate": 4.5200336761827985e-05,
      "loss": 1.0691,
      "num_input_tokens_seen": 21852544,
      "step": 37665
    },
    {
      "epoch": 5.6106642835865355,
      "grad_norm": 14.8125,
      "learning_rate": 4.519842214902423e-05,
      "loss": 1.1903,
      "num_input_tokens_seen": 21855456,
      "step": 37670
    },
    {
      "epoch": 5.611408996127495,
      "grad_norm": 12.9375,
      "learning_rate": 4.519650719498868e-05,
      "loss": 0.7718,
      "num_input_tokens_seen": 21858464,
      "step": 37675
    },
    {
      "epoch": 5.612153708668454,
      "grad_norm": 5.09375,
      "learning_rate": 4.519459189975365e-05,
      "loss": 0.9597,
      "num_input_tokens_seen": 21861536,
      "step": 37680
    },
    {
      "epoch": 5.6128984212094135,
      "grad_norm": 9.625,
      "learning_rate": 4.519267626335153e-05,
      "loss": 1.2778,
      "num_input_tokens_seen": 21864384,
      "step": 37685
    },
    {
      "epoch": 5.613643133750372,
      "grad_norm": 10.1875,
      "learning_rate": 4.519076028581466e-05,
      "loss": 0.7392,
      "num_input_tokens_seen": 21867232,
      "step": 37690
    },
    {
      "epoch": 5.614387846291332,
      "grad_norm": 10.5625,
      "learning_rate": 4.518884396717541e-05,
      "loss": 1.2361,
      "num_input_tokens_seen": 21870240,
      "step": 37695
    },
    {
      "epoch": 5.615132558832291,
      "grad_norm": 7.65625,
      "learning_rate": 4.518692730746616e-05,
      "loss": 0.9235,
      "num_input_tokens_seen": 21873184,
      "step": 37700
    },
    {
      "epoch": 5.61587727137325,
      "grad_norm": 9.875,
      "learning_rate": 4.51850103067193e-05,
      "loss": 1.2575,
      "num_input_tokens_seen": 21876000,
      "step": 37705
    },
    {
      "epoch": 5.616621983914209,
      "grad_norm": 10.25,
      "learning_rate": 4.5183092964967204e-05,
      "loss": 1.0774,
      "num_input_tokens_seen": 21879168,
      "step": 37710
    },
    {
      "epoch": 5.617366696455168,
      "grad_norm": 9.3125,
      "learning_rate": 4.518117528224226e-05,
      "loss": 1.1865,
      "num_input_tokens_seen": 21882080,
      "step": 37715
    },
    {
      "epoch": 5.6181114089961275,
      "grad_norm": 15.25,
      "learning_rate": 4.517925725857688e-05,
      "loss": 1.2286,
      "num_input_tokens_seen": 21884800,
      "step": 37720
    },
    {
      "epoch": 5.618856121537087,
      "grad_norm": 8.6875,
      "learning_rate": 4.5177338894003454e-05,
      "loss": 1.1465,
      "num_input_tokens_seen": 21887904,
      "step": 37725
    },
    {
      "epoch": 5.619600834078046,
      "grad_norm": 13.5625,
      "learning_rate": 4.517542018855439e-05,
      "loss": 0.9956,
      "num_input_tokens_seen": 21890752,
      "step": 37730
    },
    {
      "epoch": 5.620345546619005,
      "grad_norm": 11.9375,
      "learning_rate": 4.517350114226211e-05,
      "loss": 0.8002,
      "num_input_tokens_seen": 21893504,
      "step": 37735
    },
    {
      "epoch": 5.621090259159964,
      "grad_norm": 7.75,
      "learning_rate": 4.517158175515903e-05,
      "loss": 0.9667,
      "num_input_tokens_seen": 21896416,
      "step": 37740
    },
    {
      "epoch": 5.621834971700923,
      "grad_norm": 7.96875,
      "learning_rate": 4.516966202727758e-05,
      "loss": 1.1784,
      "num_input_tokens_seen": 21899360,
      "step": 37745
    },
    {
      "epoch": 5.622579684241883,
      "grad_norm": 12.1875,
      "learning_rate": 4.516774195865019e-05,
      "loss": 1.1673,
      "num_input_tokens_seen": 21902304,
      "step": 37750
    },
    {
      "epoch": 5.6233243967828415,
      "grad_norm": 13.125,
      "learning_rate": 4.5165821549309294e-05,
      "loss": 1.1595,
      "num_input_tokens_seen": 21905216,
      "step": 37755
    },
    {
      "epoch": 5.624069109323801,
      "grad_norm": 8.6875,
      "learning_rate": 4.516390079928734e-05,
      "loss": 1.2517,
      "num_input_tokens_seen": 21908160,
      "step": 37760
    },
    {
      "epoch": 5.62481382186476,
      "grad_norm": 7.4375,
      "learning_rate": 4.516197970861679e-05,
      "loss": 1.0981,
      "num_input_tokens_seen": 21910912,
      "step": 37765
    },
    {
      "epoch": 5.6255585344057195,
      "grad_norm": 8.8125,
      "learning_rate": 4.5160058277330066e-05,
      "loss": 0.8849,
      "num_input_tokens_seen": 21913696,
      "step": 37770
    },
    {
      "epoch": 5.626303246946678,
      "grad_norm": 7.84375,
      "learning_rate": 4.515813650545965e-05,
      "loss": 1.0966,
      "num_input_tokens_seen": 21916448,
      "step": 37775
    },
    {
      "epoch": 5.627047959487638,
      "grad_norm": 13.125,
      "learning_rate": 4.515621439303801e-05,
      "loss": 1.1682,
      "num_input_tokens_seen": 21919424,
      "step": 37780
    },
    {
      "epoch": 5.627792672028597,
      "grad_norm": 6.71875,
      "learning_rate": 4.515429194009761e-05,
      "loss": 0.9825,
      "num_input_tokens_seen": 21922272,
      "step": 37785
    },
    {
      "epoch": 5.628537384569556,
      "grad_norm": 7.875,
      "learning_rate": 4.515236914667094e-05,
      "loss": 1.1983,
      "num_input_tokens_seen": 21925280,
      "step": 37790
    },
    {
      "epoch": 5.629282097110515,
      "grad_norm": 9.5625,
      "learning_rate": 4.515044601279046e-05,
      "loss": 0.803,
      "num_input_tokens_seen": 21928192,
      "step": 37795
    },
    {
      "epoch": 5.630026809651475,
      "grad_norm": 19.5,
      "learning_rate": 4.514852253848868e-05,
      "loss": 1.0884,
      "num_input_tokens_seen": 21931168,
      "step": 37800
    },
    {
      "epoch": 5.6307715221924335,
      "grad_norm": 4.9375,
      "learning_rate": 4.51465987237981e-05,
      "loss": 0.9446,
      "num_input_tokens_seen": 21933824,
      "step": 37805
    },
    {
      "epoch": 5.631516234733393,
      "grad_norm": 9.9375,
      "learning_rate": 4.51446745687512e-05,
      "loss": 1.0123,
      "num_input_tokens_seen": 21936672,
      "step": 37810
    },
    {
      "epoch": 5.632260947274352,
      "grad_norm": 8.75,
      "learning_rate": 4.5142750073380505e-05,
      "loss": 1.1383,
      "num_input_tokens_seen": 21939584,
      "step": 37815
    },
    {
      "epoch": 5.6330056598153115,
      "grad_norm": 9.75,
      "learning_rate": 4.514082523771851e-05,
      "loss": 0.981,
      "num_input_tokens_seen": 21942592,
      "step": 37820
    },
    {
      "epoch": 5.63375037235627,
      "grad_norm": 9.125,
      "learning_rate": 4.513890006179775e-05,
      "loss": 1.0518,
      "num_input_tokens_seen": 21945600,
      "step": 37825
    },
    {
      "epoch": 5.63449508489723,
      "grad_norm": 6.46875,
      "learning_rate": 4.513697454565074e-05,
      "loss": 0.9047,
      "num_input_tokens_seen": 21948352,
      "step": 37830
    },
    {
      "epoch": 5.635239797438189,
      "grad_norm": 8.8125,
      "learning_rate": 4.513504868931001e-05,
      "loss": 0.7611,
      "num_input_tokens_seen": 21950976,
      "step": 37835
    },
    {
      "epoch": 5.635984509979148,
      "grad_norm": 13.1875,
      "learning_rate": 4.51331224928081e-05,
      "loss": 1.111,
      "num_input_tokens_seen": 21953728,
      "step": 37840
    },
    {
      "epoch": 5.636729222520107,
      "grad_norm": 9.875,
      "learning_rate": 4.5131195956177546e-05,
      "loss": 1.2452,
      "num_input_tokens_seen": 21956704,
      "step": 37845
    },
    {
      "epoch": 5.637473935061067,
      "grad_norm": 8.25,
      "learning_rate": 4.5129269079450894e-05,
      "loss": 1.3137,
      "num_input_tokens_seen": 21959712,
      "step": 37850
    },
    {
      "epoch": 5.6382186476020255,
      "grad_norm": 13.0625,
      "learning_rate": 4.512734186266071e-05,
      "loss": 0.9494,
      "num_input_tokens_seen": 21962688,
      "step": 37855
    },
    {
      "epoch": 5.638963360142985,
      "grad_norm": 7.5625,
      "learning_rate": 4.512541430583953e-05,
      "loss": 0.9399,
      "num_input_tokens_seen": 21965632,
      "step": 37860
    },
    {
      "epoch": 5.639708072683944,
      "grad_norm": 9.3125,
      "learning_rate": 4.5123486409019936e-05,
      "loss": 0.9187,
      "num_input_tokens_seen": 21968448,
      "step": 37865
    },
    {
      "epoch": 5.640452785224904,
      "grad_norm": 5.8125,
      "learning_rate": 4.5121558172234484e-05,
      "loss": 1.056,
      "num_input_tokens_seen": 21971392,
      "step": 37870
    },
    {
      "epoch": 5.641197497765862,
      "grad_norm": 7.84375,
      "learning_rate": 4.511962959551576e-05,
      "loss": 1.0662,
      "num_input_tokens_seen": 21974304,
      "step": 37875
    },
    {
      "epoch": 5.641942210306821,
      "grad_norm": 2.890625,
      "learning_rate": 4.511770067889635e-05,
      "loss": 1.0187,
      "num_input_tokens_seen": 21977344,
      "step": 37880
    },
    {
      "epoch": 5.642686922847781,
      "grad_norm": 7.75,
      "learning_rate": 4.5115771422408826e-05,
      "loss": 0.9241,
      "num_input_tokens_seen": 21979776,
      "step": 37885
    },
    {
      "epoch": 5.64343163538874,
      "grad_norm": 19.375,
      "learning_rate": 4.5113841826085796e-05,
      "loss": 1.223,
      "num_input_tokens_seen": 21982432,
      "step": 37890
    },
    {
      "epoch": 5.644176347929699,
      "grad_norm": 7.15625,
      "learning_rate": 4.5111911889959846e-05,
      "loss": 0.7557,
      "num_input_tokens_seen": 21985088,
      "step": 37895
    },
    {
      "epoch": 5.644921060470658,
      "grad_norm": 5.46875,
      "learning_rate": 4.5109981614063584e-05,
      "loss": 1.2135,
      "num_input_tokens_seen": 21988256,
      "step": 37900
    },
    {
      "epoch": 5.6456657730116175,
      "grad_norm": 5.625,
      "learning_rate": 4.510805099842963e-05,
      "loss": 0.9183,
      "num_input_tokens_seen": 21991104,
      "step": 37905
    },
    {
      "epoch": 5.646410485552577,
      "grad_norm": 14.4375,
      "learning_rate": 4.5106120043090585e-05,
      "loss": 1.1025,
      "num_input_tokens_seen": 21993696,
      "step": 37910
    },
    {
      "epoch": 5.647155198093536,
      "grad_norm": 14.375,
      "learning_rate": 4.510418874807907e-05,
      "loss": 1.2406,
      "num_input_tokens_seen": 21996640,
      "step": 37915
    },
    {
      "epoch": 5.647899910634495,
      "grad_norm": 7.5,
      "learning_rate": 4.5102257113427726e-05,
      "loss": 1.4453,
      "num_input_tokens_seen": 21999360,
      "step": 37920
    },
    {
      "epoch": 5.648644623175454,
      "grad_norm": 5.53125,
      "learning_rate": 4.510032513916919e-05,
      "loss": 1.3682,
      "num_input_tokens_seen": 22002176,
      "step": 37925
    },
    {
      "epoch": 5.649389335716413,
      "grad_norm": 9.1875,
      "learning_rate": 4.509839282533607e-05,
      "loss": 1.3949,
      "num_input_tokens_seen": 22004800,
      "step": 37930
    },
    {
      "epoch": 5.650134048257373,
      "grad_norm": 6.8125,
      "learning_rate": 4.509646017196104e-05,
      "loss": 1.4268,
      "num_input_tokens_seen": 22007808,
      "step": 37935
    },
    {
      "epoch": 5.6508787607983315,
      "grad_norm": 8.375,
      "learning_rate": 4.509452717907674e-05,
      "loss": 1.2181,
      "num_input_tokens_seen": 22010784,
      "step": 37940
    },
    {
      "epoch": 5.651623473339291,
      "grad_norm": 16.75,
      "learning_rate": 4.509259384671582e-05,
      "loss": 1.1361,
      "num_input_tokens_seen": 22013696,
      "step": 37945
    },
    {
      "epoch": 5.65236818588025,
      "grad_norm": 12.875,
      "learning_rate": 4.509066017491096e-05,
      "loss": 1.3633,
      "num_input_tokens_seen": 22016480,
      "step": 37950
    },
    {
      "epoch": 5.65311289842121,
      "grad_norm": 4.71875,
      "learning_rate": 4.508872616369481e-05,
      "loss": 0.9873,
      "num_input_tokens_seen": 22019616,
      "step": 37955
    },
    {
      "epoch": 5.653857610962168,
      "grad_norm": 6.90625,
      "learning_rate": 4.508679181310005e-05,
      "loss": 1.2514,
      "num_input_tokens_seen": 22022592,
      "step": 37960
    },
    {
      "epoch": 5.654602323503128,
      "grad_norm": 8.5625,
      "learning_rate": 4.508485712315935e-05,
      "loss": 1.0634,
      "num_input_tokens_seen": 22025440,
      "step": 37965
    },
    {
      "epoch": 5.655347036044087,
      "grad_norm": 15.375,
      "learning_rate": 4.508292209390541e-05,
      "loss": 1.3041,
      "num_input_tokens_seen": 22028128,
      "step": 37970
    },
    {
      "epoch": 5.656091748585046,
      "grad_norm": 9.4375,
      "learning_rate": 4.5080986725370914e-05,
      "loss": 0.9615,
      "num_input_tokens_seen": 22031040,
      "step": 37975
    },
    {
      "epoch": 5.656836461126005,
      "grad_norm": 8.3125,
      "learning_rate": 4.507905101758855e-05,
      "loss": 0.9237,
      "num_input_tokens_seen": 22034016,
      "step": 37980
    },
    {
      "epoch": 5.657581173666965,
      "grad_norm": 6.1875,
      "learning_rate": 4.507711497059104e-05,
      "loss": 0.9925,
      "num_input_tokens_seen": 22037152,
      "step": 37985
    },
    {
      "epoch": 5.6583258862079235,
      "grad_norm": 16.25,
      "learning_rate": 4.5075178584411064e-05,
      "loss": 0.8964,
      "num_input_tokens_seen": 22039936,
      "step": 37990
    },
    {
      "epoch": 5.659070598748883,
      "grad_norm": 5.53125,
      "learning_rate": 4.507324185908135e-05,
      "loss": 0.9982,
      "num_input_tokens_seen": 22043168,
      "step": 37995
    },
    {
      "epoch": 5.659815311289842,
      "grad_norm": 7.5,
      "learning_rate": 4.507130479463462e-05,
      "loss": 0.965,
      "num_input_tokens_seen": 22046080,
      "step": 38000
    },
    {
      "epoch": 5.660560023830802,
      "grad_norm": 17.75,
      "learning_rate": 4.50693673911036e-05,
      "loss": 1.3298,
      "num_input_tokens_seen": 22048928,
      "step": 38005
    },
    {
      "epoch": 5.66130473637176,
      "grad_norm": 7.375,
      "learning_rate": 4.506742964852101e-05,
      "loss": 0.9684,
      "num_input_tokens_seen": 22051808,
      "step": 38010
    },
    {
      "epoch": 5.66204944891272,
      "grad_norm": 4.75,
      "learning_rate": 4.506549156691959e-05,
      "loss": 0.8739,
      "num_input_tokens_seen": 22054912,
      "step": 38015
    },
    {
      "epoch": 5.662794161453679,
      "grad_norm": 5.65625,
      "learning_rate": 4.506355314633209e-05,
      "loss": 1.0753,
      "num_input_tokens_seen": 22057920,
      "step": 38020
    },
    {
      "epoch": 5.663538873994638,
      "grad_norm": 21.5,
      "learning_rate": 4.506161438679125e-05,
      "loss": 1.2278,
      "num_input_tokens_seen": 22060896,
      "step": 38025
    },
    {
      "epoch": 5.664283586535597,
      "grad_norm": 6.9375,
      "learning_rate": 4.5059675288329815e-05,
      "loss": 0.8069,
      "num_input_tokens_seen": 22063744,
      "step": 38030
    },
    {
      "epoch": 5.665028299076557,
      "grad_norm": 9.0,
      "learning_rate": 4.5057735850980564e-05,
      "loss": 1.2043,
      "num_input_tokens_seen": 22066624,
      "step": 38035
    },
    {
      "epoch": 5.665773011617516,
      "grad_norm": 11.4375,
      "learning_rate": 4.5055796074776244e-05,
      "loss": 0.845,
      "num_input_tokens_seen": 22069504,
      "step": 38040
    },
    {
      "epoch": 5.666517724158475,
      "grad_norm": 16.625,
      "learning_rate": 4.505385595974964e-05,
      "loss": 1.3131,
      "num_input_tokens_seen": 22072384,
      "step": 38045
    },
    {
      "epoch": 5.667262436699434,
      "grad_norm": 13.75,
      "learning_rate": 4.505191550593352e-05,
      "loss": 1.3377,
      "num_input_tokens_seen": 22075264,
      "step": 38050
    },
    {
      "epoch": 5.668007149240394,
      "grad_norm": 12.8125,
      "learning_rate": 4.5049974713360665e-05,
      "loss": 1.1688,
      "num_input_tokens_seen": 22078176,
      "step": 38055
    },
    {
      "epoch": 5.668751861781352,
      "grad_norm": 18.75,
      "learning_rate": 4.504803358206387e-05,
      "loss": 1.0045,
      "num_input_tokens_seen": 22080800,
      "step": 38060
    },
    {
      "epoch": 5.669496574322311,
      "grad_norm": 9.25,
      "learning_rate": 4.504609211207591e-05,
      "loss": 1.0787,
      "num_input_tokens_seen": 22083648,
      "step": 38065
    },
    {
      "epoch": 5.670241286863271,
      "grad_norm": 14.9375,
      "learning_rate": 4.504415030342961e-05,
      "loss": 1.1803,
      "num_input_tokens_seen": 22086592,
      "step": 38070
    },
    {
      "epoch": 5.67098599940423,
      "grad_norm": 11.3125,
      "learning_rate": 4.504220815615776e-05,
      "loss": 1.2448,
      "num_input_tokens_seen": 22089472,
      "step": 38075
    },
    {
      "epoch": 5.671730711945189,
      "grad_norm": 9.5,
      "learning_rate": 4.5040265670293174e-05,
      "loss": 1.1982,
      "num_input_tokens_seen": 22092576,
      "step": 38080
    },
    {
      "epoch": 5.672475424486148,
      "grad_norm": 8.625,
      "learning_rate": 4.503832284586867e-05,
      "loss": 1.0222,
      "num_input_tokens_seen": 22095584,
      "step": 38085
    },
    {
      "epoch": 5.673220137027108,
      "grad_norm": 10.375,
      "learning_rate": 4.5036379682917065e-05,
      "loss": 1.1723,
      "num_input_tokens_seen": 22098400,
      "step": 38090
    },
    {
      "epoch": 5.673964849568066,
      "grad_norm": 10.75,
      "learning_rate": 4.503443618147119e-05,
      "loss": 1.2596,
      "num_input_tokens_seen": 22100960,
      "step": 38095
    },
    {
      "epoch": 5.674709562109026,
      "grad_norm": 11.25,
      "learning_rate": 4.503249234156387e-05,
      "loss": 1.1313,
      "num_input_tokens_seen": 22103808,
      "step": 38100
    },
    {
      "epoch": 5.675454274649985,
      "grad_norm": 6.6875,
      "learning_rate": 4.503054816322796e-05,
      "loss": 0.938,
      "num_input_tokens_seen": 22106624,
      "step": 38105
    },
    {
      "epoch": 5.676198987190944,
      "grad_norm": 24.375,
      "learning_rate": 4.50286036464963e-05,
      "loss": 1.1689,
      "num_input_tokens_seen": 22109376,
      "step": 38110
    },
    {
      "epoch": 5.676943699731903,
      "grad_norm": 9.8125,
      "learning_rate": 4.502665879140173e-05,
      "loss": 0.991,
      "num_input_tokens_seen": 22112128,
      "step": 38115
    },
    {
      "epoch": 5.677688412272863,
      "grad_norm": 13.8125,
      "learning_rate": 4.502471359797712e-05,
      "loss": 0.8719,
      "num_input_tokens_seen": 22115136,
      "step": 38120
    },
    {
      "epoch": 5.678433124813822,
      "grad_norm": 10.1875,
      "learning_rate": 4.5022768066255315e-05,
      "loss": 0.8119,
      "num_input_tokens_seen": 22118144,
      "step": 38125
    },
    {
      "epoch": 5.679177837354781,
      "grad_norm": 10.5625,
      "learning_rate": 4.502082219626921e-05,
      "loss": 1.1961,
      "num_input_tokens_seen": 22121408,
      "step": 38130
    },
    {
      "epoch": 5.67992254989574,
      "grad_norm": 10.8125,
      "learning_rate": 4.501887598805165e-05,
      "loss": 0.876,
      "num_input_tokens_seen": 22124512,
      "step": 38135
    },
    {
      "epoch": 5.6806672624367,
      "grad_norm": 12.625,
      "learning_rate": 4.501692944163553e-05,
      "loss": 1.1719,
      "num_input_tokens_seen": 22127296,
      "step": 38140
    },
    {
      "epoch": 5.681411974977658,
      "grad_norm": 9.75,
      "learning_rate": 4.501498255705373e-05,
      "loss": 1.17,
      "num_input_tokens_seen": 22130432,
      "step": 38145
    },
    {
      "epoch": 5.682156687518618,
      "grad_norm": 9.5625,
      "learning_rate": 4.501303533433915e-05,
      "loss": 0.9559,
      "num_input_tokens_seen": 22133568,
      "step": 38150
    },
    {
      "epoch": 5.682901400059577,
      "grad_norm": 9.1875,
      "learning_rate": 4.501108777352467e-05,
      "loss": 1.0574,
      "num_input_tokens_seen": 22136672,
      "step": 38155
    },
    {
      "epoch": 5.683646112600536,
      "grad_norm": 10.75,
      "learning_rate": 4.50091398746432e-05,
      "loss": 0.9584,
      "num_input_tokens_seen": 22139968,
      "step": 38160
    },
    {
      "epoch": 5.684390825141495,
      "grad_norm": 11.875,
      "learning_rate": 4.500719163772765e-05,
      "loss": 1.1378,
      "num_input_tokens_seen": 22142432,
      "step": 38165
    },
    {
      "epoch": 5.685135537682455,
      "grad_norm": 8.3125,
      "learning_rate": 4.5005243062810934e-05,
      "loss": 1.229,
      "num_input_tokens_seen": 22145216,
      "step": 38170
    },
    {
      "epoch": 5.685880250223414,
      "grad_norm": 18.5,
      "learning_rate": 4.500329414992597e-05,
      "loss": 1.2964,
      "num_input_tokens_seen": 22148320,
      "step": 38175
    },
    {
      "epoch": 5.686624962764373,
      "grad_norm": 9.3125,
      "learning_rate": 4.500134489910567e-05,
      "loss": 1.2036,
      "num_input_tokens_seen": 22151008,
      "step": 38180
    },
    {
      "epoch": 5.687369675305332,
      "grad_norm": 7.21875,
      "learning_rate": 4.4999395310382994e-05,
      "loss": 1.0519,
      "num_input_tokens_seen": 22153632,
      "step": 38185
    },
    {
      "epoch": 5.688114387846292,
      "grad_norm": 12.4375,
      "learning_rate": 4.4997445383790846e-05,
      "loss": 0.8715,
      "num_input_tokens_seen": 22156608,
      "step": 38190
    },
    {
      "epoch": 5.68885910038725,
      "grad_norm": 18.0,
      "learning_rate": 4.499549511936219e-05,
      "loss": 0.8793,
      "num_input_tokens_seen": 22159584,
      "step": 38195
    },
    {
      "epoch": 5.68960381292821,
      "grad_norm": 9.125,
      "learning_rate": 4.499354451712997e-05,
      "loss": 0.9437,
      "num_input_tokens_seen": 22162496,
      "step": 38200
    },
    {
      "epoch": 5.690348525469169,
      "grad_norm": 7.59375,
      "learning_rate": 4.499159357712713e-05,
      "loss": 1.251,
      "num_input_tokens_seen": 22165376,
      "step": 38205
    },
    {
      "epoch": 5.6910932380101285,
      "grad_norm": 7.40625,
      "learning_rate": 4.4989642299386636e-05,
      "loss": 1.3771,
      "num_input_tokens_seen": 22168160,
      "step": 38210
    },
    {
      "epoch": 5.691837950551087,
      "grad_norm": 18.375,
      "learning_rate": 4.498769068394145e-05,
      "loss": 1.196,
      "num_input_tokens_seen": 22171360,
      "step": 38215
    },
    {
      "epoch": 5.692582663092047,
      "grad_norm": 6.6875,
      "learning_rate": 4.498573873082454e-05,
      "loss": 0.9203,
      "num_input_tokens_seen": 22174368,
      "step": 38220
    },
    {
      "epoch": 5.693327375633006,
      "grad_norm": 12.5625,
      "learning_rate": 4.4983786440068896e-05,
      "loss": 0.8055,
      "num_input_tokens_seen": 22177440,
      "step": 38225
    },
    {
      "epoch": 5.694072088173964,
      "grad_norm": 6.875,
      "learning_rate": 4.498183381170749e-05,
      "loss": 0.8994,
      "num_input_tokens_seen": 22180192,
      "step": 38230
    },
    {
      "epoch": 5.694816800714924,
      "grad_norm": 8.5,
      "learning_rate": 4.497988084577331e-05,
      "loss": 0.8855,
      "num_input_tokens_seen": 22183264,
      "step": 38235
    },
    {
      "epoch": 5.695561513255884,
      "grad_norm": 7.78125,
      "learning_rate": 4.497792754229935e-05,
      "loss": 1.1654,
      "num_input_tokens_seen": 22186336,
      "step": 38240
    },
    {
      "epoch": 5.696306225796842,
      "grad_norm": 5.40625,
      "learning_rate": 4.49759739013186e-05,
      "loss": 1.0618,
      "num_input_tokens_seen": 22189184,
      "step": 38245
    },
    {
      "epoch": 5.697050938337801,
      "grad_norm": 8.6875,
      "learning_rate": 4.4974019922864086e-05,
      "loss": 0.9436,
      "num_input_tokens_seen": 22192064,
      "step": 38250
    },
    {
      "epoch": 5.697795650878761,
      "grad_norm": 9.3125,
      "learning_rate": 4.497206560696881e-05,
      "loss": 0.9977,
      "num_input_tokens_seen": 22194880,
      "step": 38255
    },
    {
      "epoch": 5.6985403634197205,
      "grad_norm": 11.125,
      "learning_rate": 4.497011095366577e-05,
      "loss": 1.2371,
      "num_input_tokens_seen": 22197824,
      "step": 38260
    },
    {
      "epoch": 5.699285075960679,
      "grad_norm": 10.3125,
      "learning_rate": 4.4968155962988e-05,
      "loss": 0.8422,
      "num_input_tokens_seen": 22200640,
      "step": 38265
    },
    {
      "epoch": 5.700029788501638,
      "grad_norm": 12.3125,
      "learning_rate": 4.496620063496854e-05,
      "loss": 1.025,
      "num_input_tokens_seen": 22203520,
      "step": 38270
    },
    {
      "epoch": 5.700774501042598,
      "grad_norm": 15.25,
      "learning_rate": 4.496424496964041e-05,
      "loss": 1.2628,
      "num_input_tokens_seen": 22206368,
      "step": 38275
    },
    {
      "epoch": 5.701519213583556,
      "grad_norm": 11.5625,
      "learning_rate": 4.496228896703665e-05,
      "loss": 1.1559,
      "num_input_tokens_seen": 22209088,
      "step": 38280
    },
    {
      "epoch": 5.702263926124516,
      "grad_norm": 3.078125,
      "learning_rate": 4.496033262719031e-05,
      "loss": 1.296,
      "num_input_tokens_seen": 22211776,
      "step": 38285
    },
    {
      "epoch": 5.703008638665475,
      "grad_norm": 12.875,
      "learning_rate": 4.495837595013443e-05,
      "loss": 1.3271,
      "num_input_tokens_seen": 22214752,
      "step": 38290
    },
    {
      "epoch": 5.7037533512064345,
      "grad_norm": 7.0,
      "learning_rate": 4.495641893590209e-05,
      "loss": 1.5988,
      "num_input_tokens_seen": 22217728,
      "step": 38295
    },
    {
      "epoch": 5.704498063747393,
      "grad_norm": 16.75,
      "learning_rate": 4.495446158452632e-05,
      "loss": 1.2918,
      "num_input_tokens_seen": 22220384,
      "step": 38300
    },
    {
      "epoch": 5.705242776288353,
      "grad_norm": 8.4375,
      "learning_rate": 4.49525038960402e-05,
      "loss": 1.1741,
      "num_input_tokens_seen": 22223488,
      "step": 38305
    },
    {
      "epoch": 5.705987488829312,
      "grad_norm": 10.3125,
      "learning_rate": 4.495054587047682e-05,
      "loss": 0.7318,
      "num_input_tokens_seen": 22226240,
      "step": 38310
    },
    {
      "epoch": 5.706732201370271,
      "grad_norm": 9.5,
      "learning_rate": 4.4948587507869235e-05,
      "loss": 1.2687,
      "num_input_tokens_seen": 22228736,
      "step": 38315
    },
    {
      "epoch": 5.70747691391123,
      "grad_norm": 10.6875,
      "learning_rate": 4.494662880825053e-05,
      "loss": 0.9192,
      "num_input_tokens_seen": 22231680,
      "step": 38320
    },
    {
      "epoch": 5.70822162645219,
      "grad_norm": 8.0,
      "learning_rate": 4.494466977165382e-05,
      "loss": 0.6902,
      "num_input_tokens_seen": 22234656,
      "step": 38325
    },
    {
      "epoch": 5.708966338993148,
      "grad_norm": 13.625,
      "learning_rate": 4.494271039811217e-05,
      "loss": 1.2059,
      "num_input_tokens_seen": 22237280,
      "step": 38330
    },
    {
      "epoch": 5.709711051534108,
      "grad_norm": 11.6875,
      "learning_rate": 4.4940750687658716e-05,
      "loss": 0.7241,
      "num_input_tokens_seen": 22240480,
      "step": 38335
    },
    {
      "epoch": 5.710455764075067,
      "grad_norm": 13.8125,
      "learning_rate": 4.4938790640326534e-05,
      "loss": 0.8576,
      "num_input_tokens_seen": 22243488,
      "step": 38340
    },
    {
      "epoch": 5.7112004766160265,
      "grad_norm": 10.4375,
      "learning_rate": 4.4936830256148755e-05,
      "loss": 1.3397,
      "num_input_tokens_seen": 22246464,
      "step": 38345
    },
    {
      "epoch": 5.711945189156985,
      "grad_norm": 10.375,
      "learning_rate": 4.493486953515848e-05,
      "loss": 1.4441,
      "num_input_tokens_seen": 22249408,
      "step": 38350
    },
    {
      "epoch": 5.712689901697945,
      "grad_norm": 5.25,
      "learning_rate": 4.493290847738886e-05,
      "loss": 0.936,
      "num_input_tokens_seen": 22252320,
      "step": 38355
    },
    {
      "epoch": 5.713434614238904,
      "grad_norm": 7.59375,
      "learning_rate": 4.4930947082873e-05,
      "loss": 0.9559,
      "num_input_tokens_seen": 22255264,
      "step": 38360
    },
    {
      "epoch": 5.714179326779863,
      "grad_norm": 15.5625,
      "learning_rate": 4.492898535164405e-05,
      "loss": 1.4075,
      "num_input_tokens_seen": 22258304,
      "step": 38365
    },
    {
      "epoch": 5.714924039320822,
      "grad_norm": 13.9375,
      "learning_rate": 4.492702328373515e-05,
      "loss": 1.0839,
      "num_input_tokens_seen": 22261120,
      "step": 38370
    },
    {
      "epoch": 5.715668751861782,
      "grad_norm": 10.4375,
      "learning_rate": 4.492506087917944e-05,
      "loss": 1.0983,
      "num_input_tokens_seen": 22263936,
      "step": 38375
    },
    {
      "epoch": 5.7164134644027405,
      "grad_norm": 23.25,
      "learning_rate": 4.4923098138010064e-05,
      "loss": 1.1283,
      "num_input_tokens_seen": 22266528,
      "step": 38380
    },
    {
      "epoch": 5.7171581769437,
      "grad_norm": 11.5625,
      "learning_rate": 4.492113506026021e-05,
      "loss": 1.1449,
      "num_input_tokens_seen": 22269120,
      "step": 38385
    },
    {
      "epoch": 5.717902889484659,
      "grad_norm": 10.875,
      "learning_rate": 4.491917164596303e-05,
      "loss": 1.3126,
      "num_input_tokens_seen": 22271840,
      "step": 38390
    },
    {
      "epoch": 5.718647602025618,
      "grad_norm": 14.875,
      "learning_rate": 4.491720789515168e-05,
      "loss": 0.8968,
      "num_input_tokens_seen": 22274656,
      "step": 38395
    },
    {
      "epoch": 5.719392314566577,
      "grad_norm": 8.875,
      "learning_rate": 4.491524380785935e-05,
      "loss": 0.9255,
      "num_input_tokens_seen": 22277344,
      "step": 38400
    },
    {
      "epoch": 5.720137027107537,
      "grad_norm": 5.96875,
      "learning_rate": 4.4913279384119214e-05,
      "loss": 1.0699,
      "num_input_tokens_seen": 22280160,
      "step": 38405
    },
    {
      "epoch": 5.720881739648496,
      "grad_norm": 10.9375,
      "learning_rate": 4.4911314623964466e-05,
      "loss": 1.231,
      "num_input_tokens_seen": 22283392,
      "step": 38410
    },
    {
      "epoch": 5.721626452189454,
      "grad_norm": 9.1875,
      "learning_rate": 4.490934952742829e-05,
      "loss": 0.7793,
      "num_input_tokens_seen": 22286080,
      "step": 38415
    },
    {
      "epoch": 5.722371164730414,
      "grad_norm": 10.5,
      "learning_rate": 4.490738409454389e-05,
      "loss": 1.2168,
      "num_input_tokens_seen": 22288896,
      "step": 38420
    },
    {
      "epoch": 5.723115877271374,
      "grad_norm": 9.5625,
      "learning_rate": 4.4905418325344475e-05,
      "loss": 1.2536,
      "num_input_tokens_seen": 22291424,
      "step": 38425
    },
    {
      "epoch": 5.7238605898123325,
      "grad_norm": 11.125,
      "learning_rate": 4.490345221986324e-05,
      "loss": 0.9142,
      "num_input_tokens_seen": 22294336,
      "step": 38430
    },
    {
      "epoch": 5.724605302353291,
      "grad_norm": 20.125,
      "learning_rate": 4.490148577813341e-05,
      "loss": 0.9948,
      "num_input_tokens_seen": 22297376,
      "step": 38435
    },
    {
      "epoch": 5.725350014894251,
      "grad_norm": 11.375,
      "learning_rate": 4.489951900018821e-05,
      "loss": 1.0163,
      "num_input_tokens_seen": 22300096,
      "step": 38440
    },
    {
      "epoch": 5.72609472743521,
      "grad_norm": 5.71875,
      "learning_rate": 4.4897551886060866e-05,
      "loss": 0.9144,
      "num_input_tokens_seen": 22302976,
      "step": 38445
    },
    {
      "epoch": 5.726839439976169,
      "grad_norm": 7.53125,
      "learning_rate": 4.489558443578459e-05,
      "loss": 1.3884,
      "num_input_tokens_seen": 22306080,
      "step": 38450
    },
    {
      "epoch": 5.727584152517128,
      "grad_norm": 15.0,
      "learning_rate": 4.4893616649392646e-05,
      "loss": 1.3097,
      "num_input_tokens_seen": 22308832,
      "step": 38455
    },
    {
      "epoch": 5.728328865058088,
      "grad_norm": 16.25,
      "learning_rate": 4.4891648526918265e-05,
      "loss": 1.0928,
      "num_input_tokens_seen": 22311936,
      "step": 38460
    },
    {
      "epoch": 5.7290735775990465,
      "grad_norm": 8.75,
      "learning_rate": 4.48896800683947e-05,
      "loss": 0.9075,
      "num_input_tokens_seen": 22314560,
      "step": 38465
    },
    {
      "epoch": 5.729818290140006,
      "grad_norm": 10.1875,
      "learning_rate": 4.48877112738552e-05,
      "loss": 0.9259,
      "num_input_tokens_seen": 22317536,
      "step": 38470
    },
    {
      "epoch": 5.730563002680965,
      "grad_norm": 11.1875,
      "learning_rate": 4.488574214333304e-05,
      "loss": 0.782,
      "num_input_tokens_seen": 22320608,
      "step": 38475
    },
    {
      "epoch": 5.7313077152219245,
      "grad_norm": 16.875,
      "learning_rate": 4.488377267686147e-05,
      "loss": 0.9733,
      "num_input_tokens_seen": 22323456,
      "step": 38480
    },
    {
      "epoch": 5.732052427762883,
      "grad_norm": 7.5,
      "learning_rate": 4.488180287447378e-05,
      "loss": 1.2089,
      "num_input_tokens_seen": 22326304,
      "step": 38485
    },
    {
      "epoch": 5.732797140303843,
      "grad_norm": 15.4375,
      "learning_rate": 4.4879832736203224e-05,
      "loss": 0.9594,
      "num_input_tokens_seen": 22329216,
      "step": 38490
    },
    {
      "epoch": 5.733541852844802,
      "grad_norm": 13.375,
      "learning_rate": 4.48778622620831e-05,
      "loss": 1.1058,
      "num_input_tokens_seen": 22331936,
      "step": 38495
    },
    {
      "epoch": 5.734286565385761,
      "grad_norm": 5.71875,
      "learning_rate": 4.487589145214671e-05,
      "loss": 0.8206,
      "num_input_tokens_seen": 22334656,
      "step": 38500
    },
    {
      "epoch": 5.73503127792672,
      "grad_norm": 9.25,
      "learning_rate": 4.487392030642733e-05,
      "loss": 1.2736,
      "num_input_tokens_seen": 22337504,
      "step": 38505
    },
    {
      "epoch": 5.73577599046768,
      "grad_norm": 10.375,
      "learning_rate": 4.487194882495826e-05,
      "loss": 1.3039,
      "num_input_tokens_seen": 22340224,
      "step": 38510
    },
    {
      "epoch": 5.7365207030086385,
      "grad_norm": 17.375,
      "learning_rate": 4.486997700777281e-05,
      "loss": 1.1578,
      "num_input_tokens_seen": 22343360,
      "step": 38515
    },
    {
      "epoch": 5.737265415549598,
      "grad_norm": 9.375,
      "learning_rate": 4.486800485490429e-05,
      "loss": 0.9623,
      "num_input_tokens_seen": 22346016,
      "step": 38520
    },
    {
      "epoch": 5.738010128090557,
      "grad_norm": 19.375,
      "learning_rate": 4.4866032366386034e-05,
      "loss": 0.9276,
      "num_input_tokens_seen": 22348928,
      "step": 38525
    },
    {
      "epoch": 5.7387548406315165,
      "grad_norm": 15.625,
      "learning_rate": 4.486405954225135e-05,
      "loss": 1.2951,
      "num_input_tokens_seen": 22351712,
      "step": 38530
    },
    {
      "epoch": 5.739499553172475,
      "grad_norm": 15.75,
      "learning_rate": 4.486208638253356e-05,
      "loss": 1.1648,
      "num_input_tokens_seen": 22354528,
      "step": 38535
    },
    {
      "epoch": 5.740244265713435,
      "grad_norm": 11.3125,
      "learning_rate": 4.486011288726601e-05,
      "loss": 1.1773,
      "num_input_tokens_seen": 22357376,
      "step": 38540
    },
    {
      "epoch": 5.740988978254394,
      "grad_norm": 13.8125,
      "learning_rate": 4.485813905648204e-05,
      "loss": 0.9738,
      "num_input_tokens_seen": 22360192,
      "step": 38545
    },
    {
      "epoch": 5.741733690795353,
      "grad_norm": 7.875,
      "learning_rate": 4.485616489021499e-05,
      "loss": 0.9237,
      "num_input_tokens_seen": 22363296,
      "step": 38550
    },
    {
      "epoch": 5.742478403336312,
      "grad_norm": 12.375,
      "learning_rate": 4.485419038849822e-05,
      "loss": 1.1536,
      "num_input_tokens_seen": 22366176,
      "step": 38555
    },
    {
      "epoch": 5.743223115877272,
      "grad_norm": 12.5625,
      "learning_rate": 4.485221555136508e-05,
      "loss": 1.3917,
      "num_input_tokens_seen": 22369088,
      "step": 38560
    },
    {
      "epoch": 5.7439678284182305,
      "grad_norm": 7.4375,
      "learning_rate": 4.485024037884894e-05,
      "loss": 1.1508,
      "num_input_tokens_seen": 22371840,
      "step": 38565
    },
    {
      "epoch": 5.74471254095919,
      "grad_norm": 5.21875,
      "learning_rate": 4.484826487098316e-05,
      "loss": 0.8992,
      "num_input_tokens_seen": 22374656,
      "step": 38570
    },
    {
      "epoch": 5.745457253500149,
      "grad_norm": 10.8125,
      "learning_rate": 4.484628902780111e-05,
      "loss": 0.678,
      "num_input_tokens_seen": 22377504,
      "step": 38575
    },
    {
      "epoch": 5.746201966041108,
      "grad_norm": 7.6875,
      "learning_rate": 4.484431284933619e-05,
      "loss": 1.1386,
      "num_input_tokens_seen": 22380736,
      "step": 38580
    },
    {
      "epoch": 5.746946678582067,
      "grad_norm": 14.6875,
      "learning_rate": 4.484233633562176e-05,
      "loss": 1.1934,
      "num_input_tokens_seen": 22383584,
      "step": 38585
    },
    {
      "epoch": 5.747691391123027,
      "grad_norm": 15.4375,
      "learning_rate": 4.484035948669124e-05,
      "loss": 1.2975,
      "num_input_tokens_seen": 22386688,
      "step": 38590
    },
    {
      "epoch": 5.748436103663986,
      "grad_norm": 7.15625,
      "learning_rate": 4.4838382302577995e-05,
      "loss": 0.9232,
      "num_input_tokens_seen": 22389728,
      "step": 38595
    },
    {
      "epoch": 5.7491808162049445,
      "grad_norm": 14.25,
      "learning_rate": 4.483640478331546e-05,
      "loss": 1.1779,
      "num_input_tokens_seen": 22393312,
      "step": 38600
    },
    {
      "epoch": 5.749925528745904,
      "grad_norm": 7.21875,
      "learning_rate": 4.483442692893702e-05,
      "loss": 1.0352,
      "num_input_tokens_seen": 22396224,
      "step": 38605
    },
    {
      "epoch": 5.750670241286863,
      "grad_norm": 10.6875,
      "learning_rate": 4.483244873947609e-05,
      "loss": 1.5,
      "num_input_tokens_seen": 22399232,
      "step": 38610
    },
    {
      "epoch": 5.7514149538278225,
      "grad_norm": 9.3125,
      "learning_rate": 4.48304702149661e-05,
      "loss": 1.1452,
      "num_input_tokens_seen": 22402144,
      "step": 38615
    },
    {
      "epoch": 5.752159666368781,
      "grad_norm": 12.4375,
      "learning_rate": 4.482849135544048e-05,
      "loss": 1.1509,
      "num_input_tokens_seen": 22405120,
      "step": 38620
    },
    {
      "epoch": 5.752904378909741,
      "grad_norm": 6.46875,
      "learning_rate": 4.4826512160932636e-05,
      "loss": 1.0213,
      "num_input_tokens_seen": 22408000,
      "step": 38625
    },
    {
      "epoch": 5.7536490914507,
      "grad_norm": 10.5625,
      "learning_rate": 4.482453263147603e-05,
      "loss": 1.0235,
      "num_input_tokens_seen": 22410976,
      "step": 38630
    },
    {
      "epoch": 5.754393803991659,
      "grad_norm": 5.03125,
      "learning_rate": 4.4822552767104095e-05,
      "loss": 0.9001,
      "num_input_tokens_seen": 22413856,
      "step": 38635
    },
    {
      "epoch": 5.755138516532618,
      "grad_norm": 7.21875,
      "learning_rate": 4.482057256785027e-05,
      "loss": 1.1753,
      "num_input_tokens_seen": 22416704,
      "step": 38640
    },
    {
      "epoch": 5.755883229073578,
      "grad_norm": 9.625,
      "learning_rate": 4.481859203374802e-05,
      "loss": 0.8513,
      "num_input_tokens_seen": 22419552,
      "step": 38645
    },
    {
      "epoch": 5.7566279416145365,
      "grad_norm": 8.875,
      "learning_rate": 4.48166111648308e-05,
      "loss": 0.8381,
      "num_input_tokens_seen": 22422272,
      "step": 38650
    },
    {
      "epoch": 5.757372654155496,
      "grad_norm": 5.96875,
      "learning_rate": 4.481462996113207e-05,
      "loss": 1.142,
      "num_input_tokens_seen": 22425120,
      "step": 38655
    },
    {
      "epoch": 5.758117366696455,
      "grad_norm": 7.71875,
      "learning_rate": 4.481264842268531e-05,
      "loss": 1.0198,
      "num_input_tokens_seen": 22427712,
      "step": 38660
    },
    {
      "epoch": 5.7588620792374146,
      "grad_norm": 7.15625,
      "learning_rate": 4.4810666549523997e-05,
      "loss": 1.1109,
      "num_input_tokens_seen": 22430592,
      "step": 38665
    },
    {
      "epoch": 5.759606791778373,
      "grad_norm": 6.4375,
      "learning_rate": 4.48086843416816e-05,
      "loss": 1.2798,
      "num_input_tokens_seen": 22433568,
      "step": 38670
    },
    {
      "epoch": 5.760351504319333,
      "grad_norm": 7.96875,
      "learning_rate": 4.480670179919162e-05,
      "loss": 1.1604,
      "num_input_tokens_seen": 22436704,
      "step": 38675
    },
    {
      "epoch": 5.761096216860292,
      "grad_norm": 11.3125,
      "learning_rate": 4.480471892208754e-05,
      "loss": 1.1646,
      "num_input_tokens_seen": 22439776,
      "step": 38680
    },
    {
      "epoch": 5.761840929401251,
      "grad_norm": 11.8125,
      "learning_rate": 4.480273571040285e-05,
      "loss": 1.1983,
      "num_input_tokens_seen": 22442848,
      "step": 38685
    },
    {
      "epoch": 5.76258564194221,
      "grad_norm": 12.875,
      "learning_rate": 4.480075216417109e-05,
      "loss": 1.4582,
      "num_input_tokens_seen": 22445440,
      "step": 38690
    },
    {
      "epoch": 5.76333035448317,
      "grad_norm": 12.625,
      "learning_rate": 4.479876828342573e-05,
      "loss": 1.135,
      "num_input_tokens_seen": 22448032,
      "step": 38695
    },
    {
      "epoch": 5.7640750670241285,
      "grad_norm": 23.375,
      "learning_rate": 4.479678406820031e-05,
      "loss": 1.3061,
      "num_input_tokens_seen": 22451040,
      "step": 38700
    },
    {
      "epoch": 5.764819779565088,
      "grad_norm": 14.1875,
      "learning_rate": 4.479479951852834e-05,
      "loss": 1.1377,
      "num_input_tokens_seen": 22454304,
      "step": 38705
    },
    {
      "epoch": 5.765564492106047,
      "grad_norm": 8.3125,
      "learning_rate": 4.479281463444335e-05,
      "loss": 0.9756,
      "num_input_tokens_seen": 22457408,
      "step": 38710
    },
    {
      "epoch": 5.766309204647007,
      "grad_norm": 5.59375,
      "learning_rate": 4.479082941597888e-05,
      "loss": 1.1962,
      "num_input_tokens_seen": 22460192,
      "step": 38715
    },
    {
      "epoch": 5.767053917187965,
      "grad_norm": 13.25,
      "learning_rate": 4.4788843863168455e-05,
      "loss": 1.0081,
      "num_input_tokens_seen": 22463456,
      "step": 38720
    },
    {
      "epoch": 5.767798629728925,
      "grad_norm": 5.59375,
      "learning_rate": 4.4786857976045625e-05,
      "loss": 0.8227,
      "num_input_tokens_seen": 22466400,
      "step": 38725
    },
    {
      "epoch": 5.768543342269884,
      "grad_norm": 13.625,
      "learning_rate": 4.4784871754643946e-05,
      "loss": 1.4117,
      "num_input_tokens_seen": 22469120,
      "step": 38730
    },
    {
      "epoch": 5.769288054810843,
      "grad_norm": 6.9375,
      "learning_rate": 4.478288519899697e-05,
      "loss": 1.1933,
      "num_input_tokens_seen": 22472064,
      "step": 38735
    },
    {
      "epoch": 5.770032767351802,
      "grad_norm": 11.5,
      "learning_rate": 4.4780898309138245e-05,
      "loss": 0.9604,
      "num_input_tokens_seen": 22474976,
      "step": 38740
    },
    {
      "epoch": 5.770777479892761,
      "grad_norm": 6.625,
      "learning_rate": 4.477891108510135e-05,
      "loss": 0.9551,
      "num_input_tokens_seen": 22478016,
      "step": 38745
    },
    {
      "epoch": 5.7715221924337206,
      "grad_norm": 22.75,
      "learning_rate": 4.4776923526919855e-05,
      "loss": 1.1605,
      "num_input_tokens_seen": 22480832,
      "step": 38750
    },
    {
      "epoch": 5.77226690497468,
      "grad_norm": 9.3125,
      "learning_rate": 4.477493563462733e-05,
      "loss": 1.0353,
      "num_input_tokens_seen": 22483552,
      "step": 38755
    },
    {
      "epoch": 5.773011617515639,
      "grad_norm": 4.625,
      "learning_rate": 4.477294740825738e-05,
      "loss": 0.8502,
      "num_input_tokens_seen": 22486176,
      "step": 38760
    },
    {
      "epoch": 5.773756330056598,
      "grad_norm": 6.28125,
      "learning_rate": 4.477095884784358e-05,
      "loss": 1.1547,
      "num_input_tokens_seen": 22489216,
      "step": 38765
    },
    {
      "epoch": 5.774501042597557,
      "grad_norm": 16.5,
      "learning_rate": 4.476896995341951e-05,
      "loss": 1.217,
      "num_input_tokens_seen": 22492064,
      "step": 38770
    },
    {
      "epoch": 5.775245755138517,
      "grad_norm": 8.875,
      "learning_rate": 4.47669807250188e-05,
      "loss": 1.131,
      "num_input_tokens_seen": 22494976,
      "step": 38775
    },
    {
      "epoch": 5.775990467679476,
      "grad_norm": 8.3125,
      "learning_rate": 4.476499116267503e-05,
      "loss": 1.0768,
      "num_input_tokens_seen": 22498144,
      "step": 38780
    },
    {
      "epoch": 5.7767351802204345,
      "grad_norm": 9.25,
      "learning_rate": 4.476300126642183e-05,
      "loss": 1.4508,
      "num_input_tokens_seen": 22501088,
      "step": 38785
    },
    {
      "epoch": 5.777479892761394,
      "grad_norm": 6.5625,
      "learning_rate": 4.4761011036292804e-05,
      "loss": 1.0546,
      "num_input_tokens_seen": 22504000,
      "step": 38790
    },
    {
      "epoch": 5.778224605302353,
      "grad_norm": 6.90625,
      "learning_rate": 4.475902047232159e-05,
      "loss": 1.0984,
      "num_input_tokens_seen": 22507008,
      "step": 38795
    },
    {
      "epoch": 5.778969317843313,
      "grad_norm": 9.9375,
      "learning_rate": 4.4757029574541795e-05,
      "loss": 1.0724,
      "num_input_tokens_seen": 22509984,
      "step": 38800
    },
    {
      "epoch": 5.779714030384271,
      "grad_norm": 8.3125,
      "learning_rate": 4.475503834298707e-05,
      "loss": 0.877,
      "num_input_tokens_seen": 22512832,
      "step": 38805
    },
    {
      "epoch": 5.780458742925231,
      "grad_norm": 8.875,
      "learning_rate": 4.475304677769105e-05,
      "loss": 0.8328,
      "num_input_tokens_seen": 22515648,
      "step": 38810
    },
    {
      "epoch": 5.78120345546619,
      "grad_norm": 10.5625,
      "learning_rate": 4.475105487868739e-05,
      "loss": 0.7842,
      "num_input_tokens_seen": 22518528,
      "step": 38815
    },
    {
      "epoch": 5.781948168007149,
      "grad_norm": 6.71875,
      "learning_rate": 4.474906264600972e-05,
      "loss": 0.8911,
      "num_input_tokens_seen": 22521408,
      "step": 38820
    },
    {
      "epoch": 5.782692880548108,
      "grad_norm": 20.875,
      "learning_rate": 4.474707007969171e-05,
      "loss": 1.0639,
      "num_input_tokens_seen": 22524192,
      "step": 38825
    },
    {
      "epoch": 5.783437593089068,
      "grad_norm": 11.0625,
      "learning_rate": 4.4745077179767026e-05,
      "loss": 1.4388,
      "num_input_tokens_seen": 22527072,
      "step": 38830
    },
    {
      "epoch": 5.7841823056300266,
      "grad_norm": 4.15625,
      "learning_rate": 4.4743083946269324e-05,
      "loss": 1.0719,
      "num_input_tokens_seen": 22530368,
      "step": 38835
    },
    {
      "epoch": 5.784927018170986,
      "grad_norm": 6.40625,
      "learning_rate": 4.47410903792323e-05,
      "loss": 0.9999,
      "num_input_tokens_seen": 22533408,
      "step": 38840
    },
    {
      "epoch": 5.785671730711945,
      "grad_norm": 7.25,
      "learning_rate": 4.47390964786896e-05,
      "loss": 1.01,
      "num_input_tokens_seen": 22536544,
      "step": 38845
    },
    {
      "epoch": 5.786416443252905,
      "grad_norm": 12.3125,
      "learning_rate": 4.4737102244674934e-05,
      "loss": 1.0128,
      "num_input_tokens_seen": 22539520,
      "step": 38850
    },
    {
      "epoch": 5.787161155793863,
      "grad_norm": 10.5625,
      "learning_rate": 4.473510767722199e-05,
      "loss": 1.4416,
      "num_input_tokens_seen": 22542528,
      "step": 38855
    },
    {
      "epoch": 5.787905868334823,
      "grad_norm": 8.8125,
      "learning_rate": 4.473311277636445e-05,
      "loss": 1.0308,
      "num_input_tokens_seen": 22545504,
      "step": 38860
    },
    {
      "epoch": 5.788650580875782,
      "grad_norm": 5.90625,
      "learning_rate": 4.4731117542136034e-05,
      "loss": 0.7168,
      "num_input_tokens_seen": 22548288,
      "step": 38865
    },
    {
      "epoch": 5.789395293416741,
      "grad_norm": 16.0,
      "learning_rate": 4.472912197457044e-05,
      "loss": 0.9498,
      "num_input_tokens_seen": 22551296,
      "step": 38870
    },
    {
      "epoch": 5.7901400059577,
      "grad_norm": 10.5625,
      "learning_rate": 4.472712607370137e-05,
      "loss": 0.6842,
      "num_input_tokens_seen": 22554336,
      "step": 38875
    },
    {
      "epoch": 5.79088471849866,
      "grad_norm": 12.25,
      "learning_rate": 4.472512983956257e-05,
      "loss": 0.964,
      "num_input_tokens_seen": 22557120,
      "step": 38880
    },
    {
      "epoch": 5.791629431039619,
      "grad_norm": 10.5,
      "learning_rate": 4.4723133272187745e-05,
      "loss": 1.1141,
      "num_input_tokens_seen": 22560192,
      "step": 38885
    },
    {
      "epoch": 5.792374143580578,
      "grad_norm": 13.0,
      "learning_rate": 4.4721136371610626e-05,
      "loss": 1.171,
      "num_input_tokens_seen": 22563168,
      "step": 38890
    },
    {
      "epoch": 5.793118856121537,
      "grad_norm": 11.375,
      "learning_rate": 4.4719139137864956e-05,
      "loss": 0.8292,
      "num_input_tokens_seen": 22565888,
      "step": 38895
    },
    {
      "epoch": 5.793863568662497,
      "grad_norm": 7.53125,
      "learning_rate": 4.4717141570984474e-05,
      "loss": 0.9922,
      "num_input_tokens_seen": 22568864,
      "step": 38900
    },
    {
      "epoch": 5.794608281203455,
      "grad_norm": 13.5625,
      "learning_rate": 4.471514367100292e-05,
      "loss": 0.9825,
      "num_input_tokens_seen": 22571808,
      "step": 38905
    },
    {
      "epoch": 5.795352993744415,
      "grad_norm": 7.25,
      "learning_rate": 4.471314543795405e-05,
      "loss": 1.102,
      "num_input_tokens_seen": 22575008,
      "step": 38910
    },
    {
      "epoch": 5.796097706285374,
      "grad_norm": 11.625,
      "learning_rate": 4.4711146871871625e-05,
      "loss": 1.2207,
      "num_input_tokens_seen": 22577984,
      "step": 38915
    },
    {
      "epoch": 5.796842418826333,
      "grad_norm": 10.625,
      "learning_rate": 4.4709147972789405e-05,
      "loss": 1.2003,
      "num_input_tokens_seen": 22580992,
      "step": 38920
    },
    {
      "epoch": 5.797587131367292,
      "grad_norm": 6.03125,
      "learning_rate": 4.470714874074117e-05,
      "loss": 1.2399,
      "num_input_tokens_seen": 22584000,
      "step": 38925
    },
    {
      "epoch": 5.798331843908251,
      "grad_norm": 12.875,
      "learning_rate": 4.470514917576067e-05,
      "loss": 0.8929,
      "num_input_tokens_seen": 22586816,
      "step": 38930
    },
    {
      "epoch": 5.799076556449211,
      "grad_norm": 8.25,
      "learning_rate": 4.470314927788172e-05,
      "loss": 1.1093,
      "num_input_tokens_seen": 22589440,
      "step": 38935
    },
    {
      "epoch": 5.79982126899017,
      "grad_norm": 8.0625,
      "learning_rate": 4.470114904713808e-05,
      "loss": 1.1373,
      "num_input_tokens_seen": 22592352,
      "step": 38940
    },
    {
      "epoch": 5.800565981531129,
      "grad_norm": 8.625,
      "learning_rate": 4.4699148483563546e-05,
      "loss": 1.1436,
      "num_input_tokens_seen": 22595456,
      "step": 38945
    },
    {
      "epoch": 5.801310694072088,
      "grad_norm": 7.875,
      "learning_rate": 4.469714758719192e-05,
      "loss": 0.8531,
      "num_input_tokens_seen": 22598560,
      "step": 38950
    },
    {
      "epoch": 5.802055406613047,
      "grad_norm": 8.6875,
      "learning_rate": 4.469514635805702e-05,
      "loss": 1.0615,
      "num_input_tokens_seen": 22601632,
      "step": 38955
    },
    {
      "epoch": 5.802800119154006,
      "grad_norm": 9.1875,
      "learning_rate": 4.469314479619262e-05,
      "loss": 1.1364,
      "num_input_tokens_seen": 22604448,
      "step": 38960
    },
    {
      "epoch": 5.803544831694966,
      "grad_norm": 5.53125,
      "learning_rate": 4.469114290163257e-05,
      "loss": 0.987,
      "num_input_tokens_seen": 22607520,
      "step": 38965
    },
    {
      "epoch": 5.804289544235925,
      "grad_norm": 13.25,
      "learning_rate": 4.468914067441066e-05,
      "loss": 0.9435,
      "num_input_tokens_seen": 22610528,
      "step": 38970
    },
    {
      "epoch": 5.805034256776884,
      "grad_norm": 10.5625,
      "learning_rate": 4.468713811456074e-05,
      "loss": 1.0407,
      "num_input_tokens_seen": 22613344,
      "step": 38975
    },
    {
      "epoch": 5.805778969317843,
      "grad_norm": 10.1875,
      "learning_rate": 4.468513522211662e-05,
      "loss": 0.7881,
      "num_input_tokens_seen": 22616192,
      "step": 38980
    },
    {
      "epoch": 5.806523681858803,
      "grad_norm": 10.1875,
      "learning_rate": 4.468313199711216e-05,
      "loss": 1.0196,
      "num_input_tokens_seen": 22619168,
      "step": 38985
    },
    {
      "epoch": 5.807268394399761,
      "grad_norm": 5.53125,
      "learning_rate": 4.468112843958118e-05,
      "loss": 1.1736,
      "num_input_tokens_seen": 22622048,
      "step": 38990
    },
    {
      "epoch": 5.808013106940721,
      "grad_norm": 7.46875,
      "learning_rate": 4.467912454955755e-05,
      "loss": 0.7275,
      "num_input_tokens_seen": 22625056,
      "step": 38995
    },
    {
      "epoch": 5.80875781948168,
      "grad_norm": 12.8125,
      "learning_rate": 4.46771203270751e-05,
      "loss": 1.4584,
      "num_input_tokens_seen": 22627936,
      "step": 39000
    },
    {
      "epoch": 5.809502532022639,
      "grad_norm": 5.75,
      "learning_rate": 4.4675115772167706e-05,
      "loss": 1.0615,
      "num_input_tokens_seen": 22630976,
      "step": 39005
    },
    {
      "epoch": 5.810247244563598,
      "grad_norm": 5.6875,
      "learning_rate": 4.467311088486922e-05,
      "loss": 1.0197,
      "num_input_tokens_seen": 22633920,
      "step": 39010
    },
    {
      "epoch": 5.810991957104558,
      "grad_norm": 13.1875,
      "learning_rate": 4.467110566521353e-05,
      "loss": 0.8309,
      "num_input_tokens_seen": 22636704,
      "step": 39015
    },
    {
      "epoch": 5.811736669645517,
      "grad_norm": 19.5,
      "learning_rate": 4.4669100113234504e-05,
      "loss": 1.0815,
      "num_input_tokens_seen": 22639872,
      "step": 39020
    },
    {
      "epoch": 5.812481382186476,
      "grad_norm": 12.375,
      "learning_rate": 4.466709422896601e-05,
      "loss": 1.4319,
      "num_input_tokens_seen": 22642688,
      "step": 39025
    },
    {
      "epoch": 5.813226094727435,
      "grad_norm": 21.125,
      "learning_rate": 4.466508801244196e-05,
      "loss": 1.4015,
      "num_input_tokens_seen": 22645440,
      "step": 39030
    },
    {
      "epoch": 5.813970807268395,
      "grad_norm": 13.375,
      "learning_rate": 4.466308146369623e-05,
      "loss": 1.0835,
      "num_input_tokens_seen": 22648192,
      "step": 39035
    },
    {
      "epoch": 5.814715519809353,
      "grad_norm": 12.125,
      "learning_rate": 4.466107458276273e-05,
      "loss": 0.9976,
      "num_input_tokens_seen": 22651232,
      "step": 39040
    },
    {
      "epoch": 5.815460232350313,
      "grad_norm": 22.5,
      "learning_rate": 4.465906736967534e-05,
      "loss": 1.3885,
      "num_input_tokens_seen": 22654048,
      "step": 39045
    },
    {
      "epoch": 5.816204944891272,
      "grad_norm": 7.59375,
      "learning_rate": 4.465705982446801e-05,
      "loss": 0.9783,
      "num_input_tokens_seen": 22656800,
      "step": 39050
    },
    {
      "epoch": 5.8169496574322315,
      "grad_norm": 6.25,
      "learning_rate": 4.465505194717462e-05,
      "loss": 1.2531,
      "num_input_tokens_seen": 22659968,
      "step": 39055
    },
    {
      "epoch": 5.81769436997319,
      "grad_norm": 14.5,
      "learning_rate": 4.46530437378291e-05,
      "loss": 0.921,
      "num_input_tokens_seen": 22662720,
      "step": 39060
    },
    {
      "epoch": 5.81843908251415,
      "grad_norm": 10.75,
      "learning_rate": 4.465103519646539e-05,
      "loss": 1.0775,
      "num_input_tokens_seen": 22665312,
      "step": 39065
    },
    {
      "epoch": 5.819183795055109,
      "grad_norm": 8.1875,
      "learning_rate": 4.4649026323117404e-05,
      "loss": 0.8371,
      "num_input_tokens_seen": 22668320,
      "step": 39070
    },
    {
      "epoch": 5.819928507596068,
      "grad_norm": 5.53125,
      "learning_rate": 4.464701711781909e-05,
      "loss": 1.2652,
      "num_input_tokens_seen": 22671264,
      "step": 39075
    },
    {
      "epoch": 5.820673220137027,
      "grad_norm": 7.53125,
      "learning_rate": 4.46450075806044e-05,
      "loss": 1.0076,
      "num_input_tokens_seen": 22674240,
      "step": 39080
    },
    {
      "epoch": 5.821417932677987,
      "grad_norm": 17.75,
      "learning_rate": 4.464299771150727e-05,
      "loss": 1.2295,
      "num_input_tokens_seen": 22677504,
      "step": 39085
    },
    {
      "epoch": 5.822162645218945,
      "grad_norm": 10.875,
      "learning_rate": 4.464098751056165e-05,
      "loss": 1.4347,
      "num_input_tokens_seen": 22680192,
      "step": 39090
    },
    {
      "epoch": 5.822907357759904,
      "grad_norm": 7.15625,
      "learning_rate": 4.463897697780152e-05,
      "loss": 1.2519,
      "num_input_tokens_seen": 22682976,
      "step": 39095
    },
    {
      "epoch": 5.823652070300864,
      "grad_norm": 14.8125,
      "learning_rate": 4.463696611326082e-05,
      "loss": 0.9658,
      "num_input_tokens_seen": 22685984,
      "step": 39100
    },
    {
      "epoch": 5.8243967828418235,
      "grad_norm": 17.0,
      "learning_rate": 4.4634954916973545e-05,
      "loss": 1.2255,
      "num_input_tokens_seen": 22689024,
      "step": 39105
    },
    {
      "epoch": 5.825141495382782,
      "grad_norm": 6.5625,
      "learning_rate": 4.463294338897366e-05,
      "loss": 1.2255,
      "num_input_tokens_seen": 22691840,
      "step": 39110
    },
    {
      "epoch": 5.825886207923741,
      "grad_norm": 7.34375,
      "learning_rate": 4.463093152929515e-05,
      "loss": 0.9647,
      "num_input_tokens_seen": 22694656,
      "step": 39115
    },
    {
      "epoch": 5.826630920464701,
      "grad_norm": 10.625,
      "learning_rate": 4.4628919337972e-05,
      "loss": 1.1527,
      "num_input_tokens_seen": 22697408,
      "step": 39120
    },
    {
      "epoch": 5.82737563300566,
      "grad_norm": 11.6875,
      "learning_rate": 4.462690681503822e-05,
      "loss": 1.3673,
      "num_input_tokens_seen": 22700064,
      "step": 39125
    },
    {
      "epoch": 5.828120345546619,
      "grad_norm": 15.875,
      "learning_rate": 4.462489396052779e-05,
      "loss": 0.9147,
      "num_input_tokens_seen": 22702688,
      "step": 39130
    },
    {
      "epoch": 5.828865058087578,
      "grad_norm": 14.5,
      "learning_rate": 4.462288077447472e-05,
      "loss": 1.1669,
      "num_input_tokens_seen": 22705664,
      "step": 39135
    },
    {
      "epoch": 5.8296097706285375,
      "grad_norm": 14.125,
      "learning_rate": 4.462086725691302e-05,
      "loss": 0.9744,
      "num_input_tokens_seen": 22708352,
      "step": 39140
    },
    {
      "epoch": 5.830354483169496,
      "grad_norm": 11.625,
      "learning_rate": 4.4618853407876714e-05,
      "loss": 0.9236,
      "num_input_tokens_seen": 22711104,
      "step": 39145
    },
    {
      "epoch": 5.831099195710456,
      "grad_norm": 21.75,
      "learning_rate": 4.461683922739982e-05,
      "loss": 0.9657,
      "num_input_tokens_seen": 22713696,
      "step": 39150
    },
    {
      "epoch": 5.831843908251415,
      "grad_norm": 5.15625,
      "learning_rate": 4.461482471551637e-05,
      "loss": 0.9171,
      "num_input_tokens_seen": 22716448,
      "step": 39155
    },
    {
      "epoch": 5.832588620792374,
      "grad_norm": 8.375,
      "learning_rate": 4.4612809872260386e-05,
      "loss": 1.2861,
      "num_input_tokens_seen": 22719168,
      "step": 39160
    },
    {
      "epoch": 5.833333333333333,
      "grad_norm": 7.125,
      "learning_rate": 4.461079469766592e-05,
      "loss": 0.971,
      "num_input_tokens_seen": 22722240,
      "step": 39165
    },
    {
      "epoch": 5.834078045874293,
      "grad_norm": 7.3125,
      "learning_rate": 4.4608779191766994e-05,
      "loss": 0.8612,
      "num_input_tokens_seen": 22725120,
      "step": 39170
    },
    {
      "epoch": 5.834822758415251,
      "grad_norm": 13.0,
      "learning_rate": 4.460676335459768e-05,
      "loss": 0.9693,
      "num_input_tokens_seen": 22727968,
      "step": 39175
    },
    {
      "epoch": 5.835567470956211,
      "grad_norm": 7.90625,
      "learning_rate": 4.460474718619203e-05,
      "loss": 0.8472,
      "num_input_tokens_seen": 22730688,
      "step": 39180
    },
    {
      "epoch": 5.83631218349717,
      "grad_norm": 4.15625,
      "learning_rate": 4.4602730686584105e-05,
      "loss": 1.1487,
      "num_input_tokens_seen": 22733568,
      "step": 39185
    },
    {
      "epoch": 5.8370568960381295,
      "grad_norm": 11.3125,
      "learning_rate": 4.460071385580796e-05,
      "loss": 1.1356,
      "num_input_tokens_seen": 22736288,
      "step": 39190
    },
    {
      "epoch": 5.837801608579088,
      "grad_norm": 11.8125,
      "learning_rate": 4.459869669389768e-05,
      "loss": 1.1868,
      "num_input_tokens_seen": 22739104,
      "step": 39195
    },
    {
      "epoch": 5.838546321120048,
      "grad_norm": 15.3125,
      "learning_rate": 4.459667920088734e-05,
      "loss": 1.1918,
      "num_input_tokens_seen": 22741920,
      "step": 39200
    },
    {
      "epoch": 5.839291033661007,
      "grad_norm": 9.8125,
      "learning_rate": 4.459466137681102e-05,
      "loss": 1.2244,
      "num_input_tokens_seen": 22744896,
      "step": 39205
    },
    {
      "epoch": 5.840035746201966,
      "grad_norm": 14.125,
      "learning_rate": 4.4592643221702805e-05,
      "loss": 1.184,
      "num_input_tokens_seen": 22747808,
      "step": 39210
    },
    {
      "epoch": 5.840780458742925,
      "grad_norm": 10.75,
      "learning_rate": 4.459062473559681e-05,
      "loss": 1.2961,
      "num_input_tokens_seen": 22750432,
      "step": 39215
    },
    {
      "epoch": 5.841525171283885,
      "grad_norm": 11.4375,
      "learning_rate": 4.4588605918527104e-05,
      "loss": 1.1076,
      "num_input_tokens_seen": 22753504,
      "step": 39220
    },
    {
      "epoch": 5.8422698838248435,
      "grad_norm": 8.125,
      "learning_rate": 4.458658677052782e-05,
      "loss": 0.9066,
      "num_input_tokens_seen": 22756096,
      "step": 39225
    },
    {
      "epoch": 5.843014596365803,
      "grad_norm": 10.875,
      "learning_rate": 4.458456729163306e-05,
      "loss": 1.1729,
      "num_input_tokens_seen": 22758624,
      "step": 39230
    },
    {
      "epoch": 5.843759308906762,
      "grad_norm": 4.53125,
      "learning_rate": 4.458254748187693e-05,
      "loss": 1.0008,
      "num_input_tokens_seen": 22761600,
      "step": 39235
    },
    {
      "epoch": 5.8445040214477215,
      "grad_norm": 11.3125,
      "learning_rate": 4.458052734129358e-05,
      "loss": 1.3536,
      "num_input_tokens_seen": 22764480,
      "step": 39240
    },
    {
      "epoch": 5.84524873398868,
      "grad_norm": 11.125,
      "learning_rate": 4.457850686991711e-05,
      "loss": 0.8101,
      "num_input_tokens_seen": 22767520,
      "step": 39245
    },
    {
      "epoch": 5.84599344652964,
      "grad_norm": 12.8125,
      "learning_rate": 4.4576486067781675e-05,
      "loss": 1.1067,
      "num_input_tokens_seen": 22770688,
      "step": 39250
    },
    {
      "epoch": 5.846738159070599,
      "grad_norm": 10.6875,
      "learning_rate": 4.45744649349214e-05,
      "loss": 1.0802,
      "num_input_tokens_seen": 22773312,
      "step": 39255
    },
    {
      "epoch": 5.847482871611557,
      "grad_norm": 11.25,
      "learning_rate": 4.457244347137043e-05,
      "loss": 1.7595,
      "num_input_tokens_seen": 22776160,
      "step": 39260
    },
    {
      "epoch": 5.848227584152517,
      "grad_norm": 9.1875,
      "learning_rate": 4.457042167716292e-05,
      "loss": 1.2777,
      "num_input_tokens_seen": 22779136,
      "step": 39265
    },
    {
      "epoch": 5.848972296693477,
      "grad_norm": 5.5625,
      "learning_rate": 4.456839955233303e-05,
      "loss": 0.9178,
      "num_input_tokens_seen": 22782080,
      "step": 39270
    },
    {
      "epoch": 5.8497170092344355,
      "grad_norm": 17.25,
      "learning_rate": 4.456637709691491e-05,
      "loss": 1.6371,
      "num_input_tokens_seen": 22784704,
      "step": 39275
    },
    {
      "epoch": 5.850461721775394,
      "grad_norm": 8.6875,
      "learning_rate": 4.456435431094275e-05,
      "loss": 1.0345,
      "num_input_tokens_seen": 22787552,
      "step": 39280
    },
    {
      "epoch": 5.851206434316354,
      "grad_norm": 12.9375,
      "learning_rate": 4.45623311944507e-05,
      "loss": 1.0882,
      "num_input_tokens_seen": 22790208,
      "step": 39285
    },
    {
      "epoch": 5.8519511468573135,
      "grad_norm": 12.5625,
      "learning_rate": 4.4560307747472945e-05,
      "loss": 1.0777,
      "num_input_tokens_seen": 22792864,
      "step": 39290
    },
    {
      "epoch": 5.852695859398272,
      "grad_norm": 10.4375,
      "learning_rate": 4.4558283970043676e-05,
      "loss": 1.3816,
      "num_input_tokens_seen": 22796000,
      "step": 39295
    },
    {
      "epoch": 5.853440571939231,
      "grad_norm": 9.4375,
      "learning_rate": 4.4556259862197067e-05,
      "loss": 1.0021,
      "num_input_tokens_seen": 22798976,
      "step": 39300
    },
    {
      "epoch": 5.854185284480191,
      "grad_norm": 13.0,
      "learning_rate": 4.4554235423967336e-05,
      "loss": 0.894,
      "num_input_tokens_seen": 22801952,
      "step": 39305
    },
    {
      "epoch": 5.8549299970211495,
      "grad_norm": 6.3125,
      "learning_rate": 4.4552210655388664e-05,
      "loss": 1.2242,
      "num_input_tokens_seen": 22804960,
      "step": 39310
    },
    {
      "epoch": 5.855674709562109,
      "grad_norm": 9.8125,
      "learning_rate": 4.455018555649527e-05,
      "loss": 1.3054,
      "num_input_tokens_seen": 22807776,
      "step": 39315
    },
    {
      "epoch": 5.856419422103068,
      "grad_norm": 14.4375,
      "learning_rate": 4.4548160127321356e-05,
      "loss": 1.0424,
      "num_input_tokens_seen": 22810848,
      "step": 39320
    },
    {
      "epoch": 5.8571641346440275,
      "grad_norm": 13.0,
      "learning_rate": 4.454613436790115e-05,
      "loss": 1.1113,
      "num_input_tokens_seen": 22813664,
      "step": 39325
    },
    {
      "epoch": 5.857908847184986,
      "grad_norm": 15.375,
      "learning_rate": 4.454410827826887e-05,
      "loss": 1.2303,
      "num_input_tokens_seen": 22816544,
      "step": 39330
    },
    {
      "epoch": 5.858653559725946,
      "grad_norm": 13.3125,
      "learning_rate": 4.454208185845874e-05,
      "loss": 1.1874,
      "num_input_tokens_seen": 22819552,
      "step": 39335
    },
    {
      "epoch": 5.859398272266905,
      "grad_norm": 6.96875,
      "learning_rate": 4.4540055108504996e-05,
      "loss": 1.2583,
      "num_input_tokens_seen": 22822688,
      "step": 39340
    },
    {
      "epoch": 5.860142984807864,
      "grad_norm": 16.75,
      "learning_rate": 4.4538028028441885e-05,
      "loss": 0.8346,
      "num_input_tokens_seen": 22825344,
      "step": 39345
    },
    {
      "epoch": 5.860887697348823,
      "grad_norm": 10.375,
      "learning_rate": 4.453600061830365e-05,
      "loss": 1.1505,
      "num_input_tokens_seen": 22828256,
      "step": 39350
    },
    {
      "epoch": 5.861632409889783,
      "grad_norm": 11.6875,
      "learning_rate": 4.453397287812453e-05,
      "loss": 1.1349,
      "num_input_tokens_seen": 22831040,
      "step": 39355
    },
    {
      "epoch": 5.8623771224307415,
      "grad_norm": 15.875,
      "learning_rate": 4.4531944807938806e-05,
      "loss": 1.1935,
      "num_input_tokens_seen": 22833888,
      "step": 39360
    },
    {
      "epoch": 5.863121834971701,
      "grad_norm": 6.40625,
      "learning_rate": 4.4529916407780715e-05,
      "loss": 0.9819,
      "num_input_tokens_seen": 22836736,
      "step": 39365
    },
    {
      "epoch": 5.86386654751266,
      "grad_norm": 14.9375,
      "learning_rate": 4.452788767768454e-05,
      "loss": 1.1709,
      "num_input_tokens_seen": 22839616,
      "step": 39370
    },
    {
      "epoch": 5.8646112600536195,
      "grad_norm": 11.0625,
      "learning_rate": 4.4525858617684545e-05,
      "loss": 1.2816,
      "num_input_tokens_seen": 22842752,
      "step": 39375
    },
    {
      "epoch": 5.865355972594578,
      "grad_norm": 16.75,
      "learning_rate": 4.452382922781503e-05,
      "loss": 1.0674,
      "num_input_tokens_seen": 22845760,
      "step": 39380
    },
    {
      "epoch": 5.866100685135538,
      "grad_norm": 12.8125,
      "learning_rate": 4.4521799508110245e-05,
      "loss": 1.1786,
      "num_input_tokens_seen": 22848544,
      "step": 39385
    },
    {
      "epoch": 5.866845397676497,
      "grad_norm": 14.375,
      "learning_rate": 4.4519769458604504e-05,
      "loss": 1.1696,
      "num_input_tokens_seen": 22851392,
      "step": 39390
    },
    {
      "epoch": 5.867590110217456,
      "grad_norm": 10.8125,
      "learning_rate": 4.45177390793321e-05,
      "loss": 1.3003,
      "num_input_tokens_seen": 22854656,
      "step": 39395
    },
    {
      "epoch": 5.868334822758415,
      "grad_norm": 11.875,
      "learning_rate": 4.451570837032733e-05,
      "loss": 0.9777,
      "num_input_tokens_seen": 22857568,
      "step": 39400
    },
    {
      "epoch": 5.869079535299375,
      "grad_norm": 14.0625,
      "learning_rate": 4.45136773316245e-05,
      "loss": 1.3947,
      "num_input_tokens_seen": 22860864,
      "step": 39405
    },
    {
      "epoch": 5.8698242478403335,
      "grad_norm": 16.875,
      "learning_rate": 4.451164596325793e-05,
      "loss": 1.0923,
      "num_input_tokens_seen": 22863712,
      "step": 39410
    },
    {
      "epoch": 5.870568960381293,
      "grad_norm": 11.3125,
      "learning_rate": 4.450961426526192e-05,
      "loss": 1.081,
      "num_input_tokens_seen": 22866656,
      "step": 39415
    },
    {
      "epoch": 5.871313672922252,
      "grad_norm": 8.9375,
      "learning_rate": 4.450758223767082e-05,
      "loss": 1.2757,
      "num_input_tokens_seen": 22869344,
      "step": 39420
    },
    {
      "epoch": 5.872058385463212,
      "grad_norm": 13.625,
      "learning_rate": 4.4505549880518935e-05,
      "loss": 1.1672,
      "num_input_tokens_seen": 22872192,
      "step": 39425
    },
    {
      "epoch": 5.87280309800417,
      "grad_norm": 8.0625,
      "learning_rate": 4.4503517193840615e-05,
      "loss": 1.1416,
      "num_input_tokens_seen": 22875136,
      "step": 39430
    },
    {
      "epoch": 5.87354781054513,
      "grad_norm": 17.125,
      "learning_rate": 4.4501484177670186e-05,
      "loss": 1.126,
      "num_input_tokens_seen": 22878176,
      "step": 39435
    },
    {
      "epoch": 5.874292523086089,
      "grad_norm": 10.4375,
      "learning_rate": 4.449945083204201e-05,
      "loss": 0.9431,
      "num_input_tokens_seen": 22881344,
      "step": 39440
    },
    {
      "epoch": 5.8750372356270475,
      "grad_norm": 7.5625,
      "learning_rate": 4.4497417156990427e-05,
      "loss": 1.4147,
      "num_input_tokens_seen": 22884128,
      "step": 39445
    },
    {
      "epoch": 5.875781948168007,
      "grad_norm": 6.28125,
      "learning_rate": 4.44953831525498e-05,
      "loss": 1.0515,
      "num_input_tokens_seen": 22886944,
      "step": 39450
    },
    {
      "epoch": 5.876526660708967,
      "grad_norm": 10.875,
      "learning_rate": 4.449334881875449e-05,
      "loss": 0.986,
      "num_input_tokens_seen": 22889760,
      "step": 39455
    },
    {
      "epoch": 5.8772713732499255,
      "grad_norm": 13.5625,
      "learning_rate": 4.4491314155638865e-05,
      "loss": 1.1278,
      "num_input_tokens_seen": 22892800,
      "step": 39460
    },
    {
      "epoch": 5.878016085790884,
      "grad_norm": 18.125,
      "learning_rate": 4.448927916323729e-05,
      "loss": 1.2769,
      "num_input_tokens_seen": 22895680,
      "step": 39465
    },
    {
      "epoch": 5.878760798331844,
      "grad_norm": 10.375,
      "learning_rate": 4.448724384158416e-05,
      "loss": 1.2814,
      "num_input_tokens_seen": 22898912,
      "step": 39470
    },
    {
      "epoch": 5.879505510872804,
      "grad_norm": 10.4375,
      "learning_rate": 4.4485208190713846e-05,
      "loss": 0.9324,
      "num_input_tokens_seen": 22901408,
      "step": 39475
    },
    {
      "epoch": 5.880250223413762,
      "grad_norm": 6.53125,
      "learning_rate": 4.448317221066074e-05,
      "loss": 0.9732,
      "num_input_tokens_seen": 22904256,
      "step": 39480
    },
    {
      "epoch": 5.880994935954721,
      "grad_norm": 10.9375,
      "learning_rate": 4.4481135901459245e-05,
      "loss": 1.2218,
      "num_input_tokens_seen": 22907072,
      "step": 39485
    },
    {
      "epoch": 5.881739648495681,
      "grad_norm": 9.5625,
      "learning_rate": 4.4479099263143765e-05,
      "loss": 1.0667,
      "num_input_tokens_seen": 22909792,
      "step": 39490
    },
    {
      "epoch": 5.8824843610366395,
      "grad_norm": 18.125,
      "learning_rate": 4.447706229574869e-05,
      "loss": 0.9071,
      "num_input_tokens_seen": 22912928,
      "step": 39495
    },
    {
      "epoch": 5.883229073577599,
      "grad_norm": 10.75,
      "learning_rate": 4.4475024999308454e-05,
      "loss": 1.1282,
      "num_input_tokens_seen": 22915648,
      "step": 39500
    },
    {
      "epoch": 5.883973786118558,
      "grad_norm": 17.5,
      "learning_rate": 4.4472987373857456e-05,
      "loss": 1.0539,
      "num_input_tokens_seen": 22918368,
      "step": 39505
    },
    {
      "epoch": 5.884718498659518,
      "grad_norm": 15.0625,
      "learning_rate": 4.447094941943013e-05,
      "loss": 1.144,
      "num_input_tokens_seen": 22921568,
      "step": 39510
    },
    {
      "epoch": 5.885463211200476,
      "grad_norm": 15.9375,
      "learning_rate": 4.44689111360609e-05,
      "loss": 1.0092,
      "num_input_tokens_seen": 22924320,
      "step": 39515
    },
    {
      "epoch": 5.886207923741436,
      "grad_norm": 7.03125,
      "learning_rate": 4.446687252378421e-05,
      "loss": 0.8779,
      "num_input_tokens_seen": 22927456,
      "step": 39520
    },
    {
      "epoch": 5.886952636282395,
      "grad_norm": 8.5,
      "learning_rate": 4.446483358263449e-05,
      "loss": 1.0673,
      "num_input_tokens_seen": 22930208,
      "step": 39525
    },
    {
      "epoch": 5.887697348823354,
      "grad_norm": 12.1875,
      "learning_rate": 4.44627943126462e-05,
      "loss": 0.7393,
      "num_input_tokens_seen": 22932896,
      "step": 39530
    },
    {
      "epoch": 5.888442061364313,
      "grad_norm": 24.625,
      "learning_rate": 4.446075471385376e-05,
      "loss": 1.2704,
      "num_input_tokens_seen": 22935872,
      "step": 39535
    },
    {
      "epoch": 5.889186773905273,
      "grad_norm": 13.375,
      "learning_rate": 4.4458714786291666e-05,
      "loss": 0.9933,
      "num_input_tokens_seen": 22938848,
      "step": 39540
    },
    {
      "epoch": 5.8899314864462315,
      "grad_norm": 7.84375,
      "learning_rate": 4.4456674529994356e-05,
      "loss": 0.9401,
      "num_input_tokens_seen": 22941664,
      "step": 39545
    },
    {
      "epoch": 5.890676198987191,
      "grad_norm": 8.625,
      "learning_rate": 4.44546339449963e-05,
      "loss": 1.1665,
      "num_input_tokens_seen": 22944480,
      "step": 39550
    },
    {
      "epoch": 5.89142091152815,
      "grad_norm": 5.0625,
      "learning_rate": 4.445259303133198e-05,
      "loss": 1.1584,
      "num_input_tokens_seen": 22947520,
      "step": 39555
    },
    {
      "epoch": 5.89216562406911,
      "grad_norm": 10.125,
      "learning_rate": 4.445055178903588e-05,
      "loss": 1.1012,
      "num_input_tokens_seen": 22950464,
      "step": 39560
    },
    {
      "epoch": 5.892910336610068,
      "grad_norm": 13.25,
      "learning_rate": 4.444851021814247e-05,
      "loss": 1.3921,
      "num_input_tokens_seen": 22953184,
      "step": 39565
    },
    {
      "epoch": 5.893655049151028,
      "grad_norm": 10.4375,
      "learning_rate": 4.444646831868624e-05,
      "loss": 0.9498,
      "num_input_tokens_seen": 22956032,
      "step": 39570
    },
    {
      "epoch": 5.894399761691987,
      "grad_norm": 12.75,
      "learning_rate": 4.44444260907017e-05,
      "loss": 1.3322,
      "num_input_tokens_seen": 22958784,
      "step": 39575
    },
    {
      "epoch": 5.895144474232946,
      "grad_norm": 9.875,
      "learning_rate": 4.444238353422334e-05,
      "loss": 1.0486,
      "num_input_tokens_seen": 22961568,
      "step": 39580
    },
    {
      "epoch": 5.895889186773905,
      "grad_norm": 4.625,
      "learning_rate": 4.444034064928567e-05,
      "loss": 1.2178,
      "num_input_tokens_seen": 22964480,
      "step": 39585
    },
    {
      "epoch": 5.896633899314865,
      "grad_norm": 11.75,
      "learning_rate": 4.443829743592321e-05,
      "loss": 0.9515,
      "num_input_tokens_seen": 22967200,
      "step": 39590
    },
    {
      "epoch": 5.897378611855824,
      "grad_norm": 11.625,
      "learning_rate": 4.4436253894170464e-05,
      "loss": 0.8782,
      "num_input_tokens_seen": 22969792,
      "step": 39595
    },
    {
      "epoch": 5.898123324396783,
      "grad_norm": 14.0625,
      "learning_rate": 4.4434210024061966e-05,
      "loss": 1.1226,
      "num_input_tokens_seen": 22972960,
      "step": 39600
    },
    {
      "epoch": 5.898868036937742,
      "grad_norm": 11.125,
      "learning_rate": 4.443216582563224e-05,
      "loss": 0.9822,
      "num_input_tokens_seen": 22975776,
      "step": 39605
    },
    {
      "epoch": 5.899612749478701,
      "grad_norm": 9.9375,
      "learning_rate": 4.443012129891583e-05,
      "loss": 0.9594,
      "num_input_tokens_seen": 22978336,
      "step": 39610
    },
    {
      "epoch": 5.90035746201966,
      "grad_norm": 9.3125,
      "learning_rate": 4.442807644394725e-05,
      "loss": 1.1673,
      "num_input_tokens_seen": 22981248,
      "step": 39615
    },
    {
      "epoch": 5.90110217456062,
      "grad_norm": 11.6875,
      "learning_rate": 4.442603126076108e-05,
      "loss": 1.0801,
      "num_input_tokens_seen": 22983968,
      "step": 39620
    },
    {
      "epoch": 5.901846887101579,
      "grad_norm": 8.8125,
      "learning_rate": 4.442398574939185e-05,
      "loss": 1.4136,
      "num_input_tokens_seen": 22986848,
      "step": 39625
    },
    {
      "epoch": 5.9025915996425375,
      "grad_norm": 14.125,
      "learning_rate": 4.442193990987412e-05,
      "loss": 1.1324,
      "num_input_tokens_seen": 22989728,
      "step": 39630
    },
    {
      "epoch": 5.903336312183497,
      "grad_norm": 6.625,
      "learning_rate": 4.441989374224246e-05,
      "loss": 0.7329,
      "num_input_tokens_seen": 22992416,
      "step": 39635
    },
    {
      "epoch": 5.904081024724457,
      "grad_norm": 10.125,
      "learning_rate": 4.4417847246531435e-05,
      "loss": 0.935,
      "num_input_tokens_seen": 22995040,
      "step": 39640
    },
    {
      "epoch": 5.904825737265416,
      "grad_norm": 5.15625,
      "learning_rate": 4.4415800422775614e-05,
      "loss": 1.1541,
      "num_input_tokens_seen": 22997920,
      "step": 39645
    },
    {
      "epoch": 5.905570449806374,
      "grad_norm": 11.875,
      "learning_rate": 4.441375327100957e-05,
      "loss": 1.1426,
      "num_input_tokens_seen": 23000960,
      "step": 39650
    },
    {
      "epoch": 5.906315162347334,
      "grad_norm": 12.375,
      "learning_rate": 4.4411705791267904e-05,
      "loss": 1.2207,
      "num_input_tokens_seen": 23003872,
      "step": 39655
    },
    {
      "epoch": 5.907059874888293,
      "grad_norm": 11.9375,
      "learning_rate": 4.44096579835852e-05,
      "loss": 1.4206,
      "num_input_tokens_seen": 23006656,
      "step": 39660
    },
    {
      "epoch": 5.907804587429252,
      "grad_norm": 7.75,
      "learning_rate": 4.440760984799605e-05,
      "loss": 1.0712,
      "num_input_tokens_seen": 23009408,
      "step": 39665
    },
    {
      "epoch": 5.908549299970211,
      "grad_norm": 11.25,
      "learning_rate": 4.440556138453505e-05,
      "loss": 1.0343,
      "num_input_tokens_seen": 23012352,
      "step": 39670
    },
    {
      "epoch": 5.909294012511171,
      "grad_norm": 4.4375,
      "learning_rate": 4.440351259323682e-05,
      "loss": 0.9391,
      "num_input_tokens_seen": 23015040,
      "step": 39675
    },
    {
      "epoch": 5.91003872505213,
      "grad_norm": 14.5,
      "learning_rate": 4.440146347413596e-05,
      "loss": 1.2085,
      "num_input_tokens_seen": 23017920,
      "step": 39680
    },
    {
      "epoch": 5.910783437593089,
      "grad_norm": 10.375,
      "learning_rate": 4.4399414027267094e-05,
      "loss": 1.0792,
      "num_input_tokens_seen": 23020672,
      "step": 39685
    },
    {
      "epoch": 5.911528150134048,
      "grad_norm": 7.6875,
      "learning_rate": 4.439736425266485e-05,
      "loss": 0.9404,
      "num_input_tokens_seen": 23023552,
      "step": 39690
    },
    {
      "epoch": 5.912272862675008,
      "grad_norm": 9.1875,
      "learning_rate": 4.4395314150363856e-05,
      "loss": 0.9381,
      "num_input_tokens_seen": 23026400,
      "step": 39695
    },
    {
      "epoch": 5.913017575215966,
      "grad_norm": 6.0,
      "learning_rate": 4.439326372039872e-05,
      "loss": 1.2781,
      "num_input_tokens_seen": 23029504,
      "step": 39700
    },
    {
      "epoch": 5.913762287756926,
      "grad_norm": 10.5,
      "learning_rate": 4.439121296280413e-05,
      "loss": 1.0268,
      "num_input_tokens_seen": 23032224,
      "step": 39705
    },
    {
      "epoch": 5.914507000297885,
      "grad_norm": 7.6875,
      "learning_rate": 4.438916187761469e-05,
      "loss": 0.9553,
      "num_input_tokens_seen": 23034912,
      "step": 39710
    },
    {
      "epoch": 5.915251712838844,
      "grad_norm": 11.25,
      "learning_rate": 4.4387110464865066e-05,
      "loss": 0.9129,
      "num_input_tokens_seen": 23037600,
      "step": 39715
    },
    {
      "epoch": 5.915996425379803,
      "grad_norm": 8.3125,
      "learning_rate": 4.4385058724589925e-05,
      "loss": 1.1251,
      "num_input_tokens_seen": 23040544,
      "step": 39720
    },
    {
      "epoch": 5.916741137920763,
      "grad_norm": 10.0625,
      "learning_rate": 4.438300665682391e-05,
      "loss": 1.0708,
      "num_input_tokens_seen": 23043584,
      "step": 39725
    },
    {
      "epoch": 5.917485850461722,
      "grad_norm": 10.625,
      "learning_rate": 4.43809542616017e-05,
      "loss": 0.8598,
      "num_input_tokens_seen": 23046560,
      "step": 39730
    },
    {
      "epoch": 5.918230563002681,
      "grad_norm": 8.5,
      "learning_rate": 4.437890153895797e-05,
      "loss": 1.1619,
      "num_input_tokens_seen": 23049344,
      "step": 39735
    },
    {
      "epoch": 5.91897527554364,
      "grad_norm": 10.9375,
      "learning_rate": 4.437684848892739e-05,
      "loss": 0.9399,
      "num_input_tokens_seen": 23052224,
      "step": 39740
    },
    {
      "epoch": 5.9197199880846,
      "grad_norm": 7.6875,
      "learning_rate": 4.437479511154465e-05,
      "loss": 1.2722,
      "num_input_tokens_seen": 23054880,
      "step": 39745
    },
    {
      "epoch": 5.920464700625558,
      "grad_norm": 16.375,
      "learning_rate": 4.4372741406844434e-05,
      "loss": 1.352,
      "num_input_tokens_seen": 23057728,
      "step": 39750
    },
    {
      "epoch": 5.921209413166518,
      "grad_norm": 12.75,
      "learning_rate": 4.437068737486145e-05,
      "loss": 1.4704,
      "num_input_tokens_seen": 23060736,
      "step": 39755
    },
    {
      "epoch": 5.921954125707477,
      "grad_norm": 3.125,
      "learning_rate": 4.4368633015630385e-05,
      "loss": 1.0423,
      "num_input_tokens_seen": 23063744,
      "step": 39760
    },
    {
      "epoch": 5.9226988382484365,
      "grad_norm": 12.125,
      "learning_rate": 4.436657832918595e-05,
      "loss": 1.0775,
      "num_input_tokens_seen": 23066656,
      "step": 39765
    },
    {
      "epoch": 5.923443550789395,
      "grad_norm": 7.125,
      "learning_rate": 4.436452331556286e-05,
      "loss": 0.9242,
      "num_input_tokens_seen": 23069888,
      "step": 39770
    },
    {
      "epoch": 5.924188263330355,
      "grad_norm": 12.125,
      "learning_rate": 4.436246797479582e-05,
      "loss": 1.1605,
      "num_input_tokens_seen": 23072448,
      "step": 39775
    },
    {
      "epoch": 5.924932975871314,
      "grad_norm": 15.25,
      "learning_rate": 4.436041230691957e-05,
      "loss": 1.249,
      "num_input_tokens_seen": 23075392,
      "step": 39780
    },
    {
      "epoch": 5.925677688412273,
      "grad_norm": 6.25,
      "learning_rate": 4.435835631196884e-05,
      "loss": 1.2421,
      "num_input_tokens_seen": 23078400,
      "step": 39785
    },
    {
      "epoch": 5.926422400953232,
      "grad_norm": 10.3125,
      "learning_rate": 4.435629998997835e-05,
      "loss": 0.8066,
      "num_input_tokens_seen": 23081280,
      "step": 39790
    },
    {
      "epoch": 5.927167113494191,
      "grad_norm": 15.6875,
      "learning_rate": 4.435424334098284e-05,
      "loss": 1.0415,
      "num_input_tokens_seen": 23084000,
      "step": 39795
    },
    {
      "epoch": 5.92791182603515,
      "grad_norm": 4.28125,
      "learning_rate": 4.435218636501706e-05,
      "loss": 0.8297,
      "num_input_tokens_seen": 23086944,
      "step": 39800
    },
    {
      "epoch": 5.92865653857611,
      "grad_norm": 6.25,
      "learning_rate": 4.435012906211576e-05,
      "loss": 0.9517,
      "num_input_tokens_seen": 23089600,
      "step": 39805
    },
    {
      "epoch": 5.929401251117069,
      "grad_norm": 10.625,
      "learning_rate": 4.43480714323137e-05,
      "loss": 0.9159,
      "num_input_tokens_seen": 23092096,
      "step": 39810
    },
    {
      "epoch": 5.930145963658028,
      "grad_norm": 13.875,
      "learning_rate": 4.434601347564563e-05,
      "loss": 1.0568,
      "num_input_tokens_seen": 23094816,
      "step": 39815
    },
    {
      "epoch": 5.930890676198987,
      "grad_norm": 11.0,
      "learning_rate": 4.434395519214633e-05,
      "loss": 0.8677,
      "num_input_tokens_seen": 23097632,
      "step": 39820
    },
    {
      "epoch": 5.931635388739946,
      "grad_norm": 14.0,
      "learning_rate": 4.4341896581850566e-05,
      "loss": 1.3566,
      "num_input_tokens_seen": 23100448,
      "step": 39825
    },
    {
      "epoch": 5.932380101280906,
      "grad_norm": 5.53125,
      "learning_rate": 4.433983764479312e-05,
      "loss": 0.8778,
      "num_input_tokens_seen": 23103424,
      "step": 39830
    },
    {
      "epoch": 5.933124813821864,
      "grad_norm": 13.75,
      "learning_rate": 4.433777838100876e-05,
      "loss": 1.2559,
      "num_input_tokens_seen": 23106144,
      "step": 39835
    },
    {
      "epoch": 5.933869526362824,
      "grad_norm": 11.375,
      "learning_rate": 4.433571879053231e-05,
      "loss": 1.0162,
      "num_input_tokens_seen": 23108864,
      "step": 39840
    },
    {
      "epoch": 5.934614238903783,
      "grad_norm": 10.25,
      "learning_rate": 4.433365887339853e-05,
      "loss": 0.8551,
      "num_input_tokens_seen": 23111904,
      "step": 39845
    },
    {
      "epoch": 5.9353589514447425,
      "grad_norm": 11.875,
      "learning_rate": 4.4331598629642235e-05,
      "loss": 1.4415,
      "num_input_tokens_seen": 23114944,
      "step": 39850
    },
    {
      "epoch": 5.936103663985701,
      "grad_norm": 4.5,
      "learning_rate": 4.432953805929823e-05,
      "loss": 0.9081,
      "num_input_tokens_seen": 23117760,
      "step": 39855
    },
    {
      "epoch": 5.936848376526661,
      "grad_norm": 15.1875,
      "learning_rate": 4.432747716240133e-05,
      "loss": 1.432,
      "num_input_tokens_seen": 23120800,
      "step": 39860
    },
    {
      "epoch": 5.93759308906762,
      "grad_norm": 12.0,
      "learning_rate": 4.432541593898634e-05,
      "loss": 1.0442,
      "num_input_tokens_seen": 23123584,
      "step": 39865
    },
    {
      "epoch": 5.938337801608579,
      "grad_norm": 7.46875,
      "learning_rate": 4.432335438908809e-05,
      "loss": 1.1864,
      "num_input_tokens_seen": 23126400,
      "step": 39870
    },
    {
      "epoch": 5.939082514149538,
      "grad_norm": 10.9375,
      "learning_rate": 4.432129251274141e-05,
      "loss": 1.0924,
      "num_input_tokens_seen": 23129408,
      "step": 39875
    },
    {
      "epoch": 5.939827226690498,
      "grad_norm": 20.25,
      "learning_rate": 4.431923030998113e-05,
      "loss": 1.1399,
      "num_input_tokens_seen": 23132384,
      "step": 39880
    },
    {
      "epoch": 5.940571939231456,
      "grad_norm": 4.96875,
      "learning_rate": 4.4317167780842086e-05,
      "loss": 0.8746,
      "num_input_tokens_seen": 23135360,
      "step": 39885
    },
    {
      "epoch": 5.941316651772416,
      "grad_norm": 10.5625,
      "learning_rate": 4.4315104925359124e-05,
      "loss": 1.212,
      "num_input_tokens_seen": 23138240,
      "step": 39890
    },
    {
      "epoch": 5.942061364313375,
      "grad_norm": 8.0625,
      "learning_rate": 4.431304174356709e-05,
      "loss": 1.2498,
      "num_input_tokens_seen": 23141024,
      "step": 39895
    },
    {
      "epoch": 5.9428060768543345,
      "grad_norm": 5.9375,
      "learning_rate": 4.431097823550086e-05,
      "loss": 1.1288,
      "num_input_tokens_seen": 23143840,
      "step": 39900
    },
    {
      "epoch": 5.943550789395293,
      "grad_norm": 15.25,
      "learning_rate": 4.4308914401195275e-05,
      "loss": 1.1584,
      "num_input_tokens_seen": 23146816,
      "step": 39905
    },
    {
      "epoch": 5.944295501936253,
      "grad_norm": 9.6875,
      "learning_rate": 4.43068502406852e-05,
      "loss": 1.4795,
      "num_input_tokens_seen": 23149632,
      "step": 39910
    },
    {
      "epoch": 5.945040214477212,
      "grad_norm": 10.5625,
      "learning_rate": 4.4304785754005516e-05,
      "loss": 1.1534,
      "num_input_tokens_seen": 23152768,
      "step": 39915
    },
    {
      "epoch": 5.945784927018171,
      "grad_norm": 14.75,
      "learning_rate": 4.43027209411911e-05,
      "loss": 1.1525,
      "num_input_tokens_seen": 23155552,
      "step": 39920
    },
    {
      "epoch": 5.94652963955913,
      "grad_norm": 15.875,
      "learning_rate": 4.430065580227683e-05,
      "loss": 1.0294,
      "num_input_tokens_seen": 23158496,
      "step": 39925
    },
    {
      "epoch": 5.94727435210009,
      "grad_norm": 9.375,
      "learning_rate": 4.4298590337297595e-05,
      "loss": 1.0396,
      "num_input_tokens_seen": 23161376,
      "step": 39930
    },
    {
      "epoch": 5.9480190646410485,
      "grad_norm": 12.6875,
      "learning_rate": 4.4296524546288286e-05,
      "loss": 1.2354,
      "num_input_tokens_seen": 23164192,
      "step": 39935
    },
    {
      "epoch": 5.948763777182008,
      "grad_norm": 12.5,
      "learning_rate": 4.429445842928382e-05,
      "loss": 1.0273,
      "num_input_tokens_seen": 23166848,
      "step": 39940
    },
    {
      "epoch": 5.949508489722967,
      "grad_norm": 23.625,
      "learning_rate": 4.4292391986319084e-05,
      "loss": 1.0747,
      "num_input_tokens_seen": 23169856,
      "step": 39945
    },
    {
      "epoch": 5.9502532022639265,
      "grad_norm": 11.25,
      "learning_rate": 4.429032521742899e-05,
      "loss": 1.0709,
      "num_input_tokens_seen": 23172544,
      "step": 39950
    },
    {
      "epoch": 5.950997914804885,
      "grad_norm": 6.90625,
      "learning_rate": 4.428825812264845e-05,
      "loss": 0.9952,
      "num_input_tokens_seen": 23175840,
      "step": 39955
    },
    {
      "epoch": 5.951742627345844,
      "grad_norm": 9.625,
      "learning_rate": 4.4286190702012405e-05,
      "loss": 0.8613,
      "num_input_tokens_seen": 23178976,
      "step": 39960
    },
    {
      "epoch": 5.952487339886804,
      "grad_norm": 7.90625,
      "learning_rate": 4.428412295555576e-05,
      "loss": 1.1854,
      "num_input_tokens_seen": 23182112,
      "step": 39965
    },
    {
      "epoch": 5.953232052427763,
      "grad_norm": 7.625,
      "learning_rate": 4.4282054883313464e-05,
      "loss": 0.8426,
      "num_input_tokens_seen": 23184864,
      "step": 39970
    },
    {
      "epoch": 5.953976764968722,
      "grad_norm": 12.3125,
      "learning_rate": 4.427998648532045e-05,
      "loss": 0.8322,
      "num_input_tokens_seen": 23188128,
      "step": 39975
    },
    {
      "epoch": 5.954721477509681,
      "grad_norm": 10.75,
      "learning_rate": 4.427791776161165e-05,
      "loss": 1.1706,
      "num_input_tokens_seen": 23190880,
      "step": 39980
    },
    {
      "epoch": 5.9554661900506405,
      "grad_norm": 10.0,
      "learning_rate": 4.4275848712222035e-05,
      "loss": 0.858,
      "num_input_tokens_seen": 23193760,
      "step": 39985
    },
    {
      "epoch": 5.9562109025916,
      "grad_norm": 6.0625,
      "learning_rate": 4.4273779337186536e-05,
      "loss": 1.0466,
      "num_input_tokens_seen": 23196576,
      "step": 39990
    },
    {
      "epoch": 5.956955615132559,
      "grad_norm": 12.375,
      "learning_rate": 4.427170963654013e-05,
      "loss": 1.2286,
      "num_input_tokens_seen": 23199424,
      "step": 39995
    },
    {
      "epoch": 5.957700327673518,
      "grad_norm": 13.5625,
      "learning_rate": 4.426963961031777e-05,
      "loss": 1.2584,
      "num_input_tokens_seen": 23202272,
      "step": 40000
    },
    {
      "epoch": 5.958445040214477,
      "grad_norm": 12.25,
      "learning_rate": 4.426756925855444e-05,
      "loss": 1.1324,
      "num_input_tokens_seen": 23205248,
      "step": 40005
    },
    {
      "epoch": 5.959189752755436,
      "grad_norm": 12.875,
      "learning_rate": 4.4265498581285114e-05,
      "loss": 1.5623,
      "num_input_tokens_seen": 23208448,
      "step": 40010
    },
    {
      "epoch": 5.959934465296396,
      "grad_norm": 5.84375,
      "learning_rate": 4.426342757854476e-05,
      "loss": 1.0024,
      "num_input_tokens_seen": 23211488,
      "step": 40015
    },
    {
      "epoch": 5.9606791778373545,
      "grad_norm": 10.1875,
      "learning_rate": 4.4261356250368386e-05,
      "loss": 1.0964,
      "num_input_tokens_seen": 23214176,
      "step": 40020
    },
    {
      "epoch": 5.961423890378314,
      "grad_norm": 11.8125,
      "learning_rate": 4.4259284596790976e-05,
      "loss": 1.0954,
      "num_input_tokens_seen": 23216864,
      "step": 40025
    },
    {
      "epoch": 5.962168602919273,
      "grad_norm": 6.125,
      "learning_rate": 4.425721261784751e-05,
      "loss": 1.3346,
      "num_input_tokens_seen": 23219392,
      "step": 40030
    },
    {
      "epoch": 5.9629133154602325,
      "grad_norm": 21.875,
      "learning_rate": 4.425514031357302e-05,
      "loss": 1.2786,
      "num_input_tokens_seen": 23222272,
      "step": 40035
    },
    {
      "epoch": 5.963658028001191,
      "grad_norm": 7.5625,
      "learning_rate": 4.4253067684002505e-05,
      "loss": 1.1472,
      "num_input_tokens_seen": 23225088,
      "step": 40040
    },
    {
      "epoch": 5.964402740542151,
      "grad_norm": 8.125,
      "learning_rate": 4.425099472917098e-05,
      "loss": 1.1274,
      "num_input_tokens_seen": 23227840,
      "step": 40045
    },
    {
      "epoch": 5.96514745308311,
      "grad_norm": 13.375,
      "learning_rate": 4.4248921449113464e-05,
      "loss": 1.2879,
      "num_input_tokens_seen": 23230592,
      "step": 40050
    },
    {
      "epoch": 5.965892165624069,
      "grad_norm": 9.75,
      "learning_rate": 4.424684784386498e-05,
      "loss": 1.2801,
      "num_input_tokens_seen": 23233504,
      "step": 40055
    },
    {
      "epoch": 5.966636878165028,
      "grad_norm": 9.1875,
      "learning_rate": 4.424477391346057e-05,
      "loss": 0.9742,
      "num_input_tokens_seen": 23236480,
      "step": 40060
    },
    {
      "epoch": 5.967381590705988,
      "grad_norm": 5.96875,
      "learning_rate": 4.424269965793526e-05,
      "loss": 1.1195,
      "num_input_tokens_seen": 23239200,
      "step": 40065
    },
    {
      "epoch": 5.9681263032469465,
      "grad_norm": 13.625,
      "learning_rate": 4.424062507732409e-05,
      "loss": 1.1131,
      "num_input_tokens_seen": 23242080,
      "step": 40070
    },
    {
      "epoch": 5.968871015787906,
      "grad_norm": 4.34375,
      "learning_rate": 4.4238550171662127e-05,
      "loss": 0.9599,
      "num_input_tokens_seen": 23245088,
      "step": 40075
    },
    {
      "epoch": 5.969615728328865,
      "grad_norm": 9.5,
      "learning_rate": 4.423647494098441e-05,
      "loss": 0.7959,
      "num_input_tokens_seen": 23247904,
      "step": 40080
    },
    {
      "epoch": 5.9703604408698245,
      "grad_norm": 10.4375,
      "learning_rate": 4.423439938532599e-05,
      "loss": 1.0027,
      "num_input_tokens_seen": 23251104,
      "step": 40085
    },
    {
      "epoch": 5.971105153410783,
      "grad_norm": 8.1875,
      "learning_rate": 4.423232350472195e-05,
      "loss": 1.1806,
      "num_input_tokens_seen": 23254080,
      "step": 40090
    },
    {
      "epoch": 5.971849865951743,
      "grad_norm": 9.5625,
      "learning_rate": 4.423024729920735e-05,
      "loss": 1.1443,
      "num_input_tokens_seen": 23257312,
      "step": 40095
    },
    {
      "epoch": 5.972594578492702,
      "grad_norm": 10.375,
      "learning_rate": 4.422817076881726e-05,
      "loss": 1.1577,
      "num_input_tokens_seen": 23260064,
      "step": 40100
    },
    {
      "epoch": 5.973339291033661,
      "grad_norm": 5.5625,
      "learning_rate": 4.422609391358677e-05,
      "loss": 1.3156,
      "num_input_tokens_seen": 23263040,
      "step": 40105
    },
    {
      "epoch": 5.97408400357462,
      "grad_norm": 14.8125,
      "learning_rate": 4.4224016733550975e-05,
      "loss": 1.0746,
      "num_input_tokens_seen": 23265984,
      "step": 40110
    },
    {
      "epoch": 5.97482871611558,
      "grad_norm": 11.0,
      "learning_rate": 4.4221939228744945e-05,
      "loss": 1.2353,
      "num_input_tokens_seen": 23268768,
      "step": 40115
    },
    {
      "epoch": 5.9755734286565385,
      "grad_norm": 12.3125,
      "learning_rate": 4.421986139920379e-05,
      "loss": 1.0702,
      "num_input_tokens_seen": 23271456,
      "step": 40120
    },
    {
      "epoch": 5.976318141197497,
      "grad_norm": 12.5,
      "learning_rate": 4.4217783244962615e-05,
      "loss": 1.017,
      "num_input_tokens_seen": 23274496,
      "step": 40125
    },
    {
      "epoch": 5.977062853738457,
      "grad_norm": 12.75,
      "learning_rate": 4.421570476605652e-05,
      "loss": 0.8545,
      "num_input_tokens_seen": 23277504,
      "step": 40130
    },
    {
      "epoch": 5.977807566279417,
      "grad_norm": 6.21875,
      "learning_rate": 4.421362596252062e-05,
      "loss": 1.0069,
      "num_input_tokens_seen": 23280448,
      "step": 40135
    },
    {
      "epoch": 5.978552278820375,
      "grad_norm": 10.4375,
      "learning_rate": 4.4211546834390046e-05,
      "loss": 1.1138,
      "num_input_tokens_seen": 23283104,
      "step": 40140
    },
    {
      "epoch": 5.979296991361334,
      "grad_norm": 11.0,
      "learning_rate": 4.420946738169991e-05,
      "loss": 1.1626,
      "num_input_tokens_seen": 23286176,
      "step": 40145
    },
    {
      "epoch": 5.980041703902294,
      "grad_norm": 5.9375,
      "learning_rate": 4.4207387604485345e-05,
      "loss": 1.2281,
      "num_input_tokens_seen": 23289184,
      "step": 40150
    },
    {
      "epoch": 5.980786416443253,
      "grad_norm": 12.8125,
      "learning_rate": 4.420530750278149e-05,
      "loss": 1.2867,
      "num_input_tokens_seen": 23292288,
      "step": 40155
    },
    {
      "epoch": 5.981531128984212,
      "grad_norm": 4.53125,
      "learning_rate": 4.420322707662348e-05,
      "loss": 1.2394,
      "num_input_tokens_seen": 23296192,
      "step": 40160
    },
    {
      "epoch": 5.982275841525171,
      "grad_norm": 9.0,
      "learning_rate": 4.420114632604647e-05,
      "loss": 0.9975,
      "num_input_tokens_seen": 23299200,
      "step": 40165
    },
    {
      "epoch": 5.9830205540661305,
      "grad_norm": 11.25,
      "learning_rate": 4.41990652510856e-05,
      "loss": 1.1216,
      "num_input_tokens_seen": 23302208,
      "step": 40170
    },
    {
      "epoch": 5.983765266607089,
      "grad_norm": 11.875,
      "learning_rate": 4.4196983851776044e-05,
      "loss": 0.9786,
      "num_input_tokens_seen": 23304960,
      "step": 40175
    },
    {
      "epoch": 5.984509979148049,
      "grad_norm": 6.90625,
      "learning_rate": 4.419490212815296e-05,
      "loss": 0.7926,
      "num_input_tokens_seen": 23308032,
      "step": 40180
    },
    {
      "epoch": 5.985254691689008,
      "grad_norm": 8.1875,
      "learning_rate": 4.419282008025151e-05,
      "loss": 0.8623,
      "num_input_tokens_seen": 23311040,
      "step": 40185
    },
    {
      "epoch": 5.985999404229967,
      "grad_norm": 12.9375,
      "learning_rate": 4.4190737708106864e-05,
      "loss": 1.189,
      "num_input_tokens_seen": 23314080,
      "step": 40190
    },
    {
      "epoch": 5.986744116770926,
      "grad_norm": 8.75,
      "learning_rate": 4.418865501175422e-05,
      "loss": 0.8247,
      "num_input_tokens_seen": 23316992,
      "step": 40195
    },
    {
      "epoch": 5.987488829311886,
      "grad_norm": 12.125,
      "learning_rate": 4.418657199122874e-05,
      "loss": 1.1686,
      "num_input_tokens_seen": 23319904,
      "step": 40200
    },
    {
      "epoch": 5.9882335418528445,
      "grad_norm": 9.4375,
      "learning_rate": 4.418448864656564e-05,
      "loss": 1.2809,
      "num_input_tokens_seen": 23322720,
      "step": 40205
    },
    {
      "epoch": 5.988978254393804,
      "grad_norm": 10.625,
      "learning_rate": 4.418240497780009e-05,
      "loss": 1.4147,
      "num_input_tokens_seen": 23325632,
      "step": 40210
    },
    {
      "epoch": 5.989722966934763,
      "grad_norm": 7.6875,
      "learning_rate": 4.4180320984967305e-05,
      "loss": 0.8578,
      "num_input_tokens_seen": 23328608,
      "step": 40215
    },
    {
      "epoch": 5.990467679475723,
      "grad_norm": 6.15625,
      "learning_rate": 4.4178236668102504e-05,
      "loss": 1.235,
      "num_input_tokens_seen": 23331648,
      "step": 40220
    },
    {
      "epoch": 5.991212392016681,
      "grad_norm": 9.4375,
      "learning_rate": 4.417615202724087e-05,
      "loss": 1.1242,
      "num_input_tokens_seen": 23334560,
      "step": 40225
    },
    {
      "epoch": 5.991957104557641,
      "grad_norm": 12.1875,
      "learning_rate": 4.4174067062417645e-05,
      "loss": 1.1997,
      "num_input_tokens_seen": 23337376,
      "step": 40230
    },
    {
      "epoch": 5.9927018170986,
      "grad_norm": 13.125,
      "learning_rate": 4.417198177366805e-05,
      "loss": 1.1301,
      "num_input_tokens_seen": 23340352,
      "step": 40235
    },
    {
      "epoch": 5.993446529639559,
      "grad_norm": 8.875,
      "learning_rate": 4.41698961610273e-05,
      "loss": 0.8821,
      "num_input_tokens_seen": 23343328,
      "step": 40240
    },
    {
      "epoch": 5.994191242180518,
      "grad_norm": 10.75,
      "learning_rate": 4.416781022453064e-05,
      "loss": 1.0151,
      "num_input_tokens_seen": 23345984,
      "step": 40245
    },
    {
      "epoch": 5.994935954721478,
      "grad_norm": 20.875,
      "learning_rate": 4.4165723964213314e-05,
      "loss": 1.3505,
      "num_input_tokens_seen": 23348896,
      "step": 40250
    },
    {
      "epoch": 5.9956806672624365,
      "grad_norm": 7.125,
      "learning_rate": 4.4163637380110555e-05,
      "loss": 1.0663,
      "num_input_tokens_seen": 23352000,
      "step": 40255
    },
    {
      "epoch": 5.996425379803396,
      "grad_norm": 11.875,
      "learning_rate": 4.416155047225762e-05,
      "loss": 0.8918,
      "num_input_tokens_seen": 23354880,
      "step": 40260
    },
    {
      "epoch": 5.997170092344355,
      "grad_norm": 18.125,
      "learning_rate": 4.415946324068976e-05,
      "loss": 0.9623,
      "num_input_tokens_seen": 23357792,
      "step": 40265
    },
    {
      "epoch": 5.997914804885315,
      "grad_norm": 11.625,
      "learning_rate": 4.4157375685442246e-05,
      "loss": 1.1232,
      "num_input_tokens_seen": 23360736,
      "step": 40270
    },
    {
      "epoch": 5.998659517426273,
      "grad_norm": 9.9375,
      "learning_rate": 4.415528780655034e-05,
      "loss": 0.9765,
      "num_input_tokens_seen": 23363744,
      "step": 40275
    },
    {
      "epoch": 5.999404229967233,
      "grad_norm": 15.0625,
      "learning_rate": 4.4153199604049315e-05,
      "loss": 1.0457,
      "num_input_tokens_seen": 23366464,
      "step": 40280
    },
    {
      "epoch": 6.0,
      "eval_loss": 1.0865566730499268,
      "eval_runtime": 43.7189,
      "eval_samples_per_second": 68.254,
      "eval_steps_per_second": 17.064,
      "num_input_tokens_seen": 23368392,
      "step": 40284
    },
    {
      "epoch": 6.000148942508192,
      "grad_norm": 20.625,
      "learning_rate": 4.415111107797445e-05,
      "loss": 1.4127,
      "num_input_tokens_seen": 23368904,
      "step": 40285
    },
    {
      "epoch": 6.000893655049151,
      "grad_norm": 3.3125,
      "learning_rate": 4.414902222836103e-05,
      "loss": 0.5746,
      "num_input_tokens_seen": 23371816,
      "step": 40290
    },
    {
      "epoch": 6.00163836759011,
      "grad_norm": 15.1875,
      "learning_rate": 4.414693305524434e-05,
      "loss": 1.1393,
      "num_input_tokens_seen": 23374728,
      "step": 40295
    },
    {
      "epoch": 6.00238308013107,
      "grad_norm": 6.09375,
      "learning_rate": 4.4144843558659675e-05,
      "loss": 0.9872,
      "num_input_tokens_seen": 23377544,
      "step": 40300
    },
    {
      "epoch": 6.003127792672029,
      "grad_norm": 8.25,
      "learning_rate": 4.414275373864234e-05,
      "loss": 0.9994,
      "num_input_tokens_seen": 23380488,
      "step": 40305
    },
    {
      "epoch": 6.003872505212988,
      "grad_norm": 6.46875,
      "learning_rate": 4.4140663595227624e-05,
      "loss": 0.646,
      "num_input_tokens_seen": 23383176,
      "step": 40310
    },
    {
      "epoch": 6.004617217753947,
      "grad_norm": 12.375,
      "learning_rate": 4.413857312845086e-05,
      "loss": 1.0483,
      "num_input_tokens_seen": 23386152,
      "step": 40315
    },
    {
      "epoch": 6.005361930294906,
      "grad_norm": 5.84375,
      "learning_rate": 4.4136482338347356e-05,
      "loss": 0.9949,
      "num_input_tokens_seen": 23389448,
      "step": 40320
    },
    {
      "epoch": 6.006106642835865,
      "grad_norm": 9.0625,
      "learning_rate": 4.413439122495243e-05,
      "loss": 1.0876,
      "num_input_tokens_seen": 23392328,
      "step": 40325
    },
    {
      "epoch": 6.006851355376824,
      "grad_norm": 10.125,
      "learning_rate": 4.413229978830141e-05,
      "loss": 0.9851,
      "num_input_tokens_seen": 23395272,
      "step": 40330
    },
    {
      "epoch": 6.007596067917784,
      "grad_norm": 13.5,
      "learning_rate": 4.413020802842963e-05,
      "loss": 0.9192,
      "num_input_tokens_seen": 23398184,
      "step": 40335
    },
    {
      "epoch": 6.0083407804587425,
      "grad_norm": 12.8125,
      "learning_rate": 4.412811594537243e-05,
      "loss": 1.3288,
      "num_input_tokens_seen": 23401160,
      "step": 40340
    },
    {
      "epoch": 6.009085492999702,
      "grad_norm": 11.3125,
      "learning_rate": 4.4126023539165155e-05,
      "loss": 0.9307,
      "num_input_tokens_seen": 23404232,
      "step": 40345
    },
    {
      "epoch": 6.009830205540661,
      "grad_norm": 11.75,
      "learning_rate": 4.412393080984315e-05,
      "loss": 1.1776,
      "num_input_tokens_seen": 23407176,
      "step": 40350
    },
    {
      "epoch": 6.010574918081621,
      "grad_norm": 13.875,
      "learning_rate": 4.412183775744177e-05,
      "loss": 0.7904,
      "num_input_tokens_seen": 23410024,
      "step": 40355
    },
    {
      "epoch": 6.011319630622579,
      "grad_norm": 11.0625,
      "learning_rate": 4.411974438199637e-05,
      "loss": 1.0508,
      "num_input_tokens_seen": 23413384,
      "step": 40360
    },
    {
      "epoch": 6.012064343163539,
      "grad_norm": 12.5625,
      "learning_rate": 4.411765068354233e-05,
      "loss": 1.3083,
      "num_input_tokens_seen": 23416360,
      "step": 40365
    },
    {
      "epoch": 6.012809055704498,
      "grad_norm": 9.375,
      "learning_rate": 4.4115556662115004e-05,
      "loss": 1.0279,
      "num_input_tokens_seen": 23419304,
      "step": 40370
    },
    {
      "epoch": 6.013553768245457,
      "grad_norm": 9.5,
      "learning_rate": 4.411346231774978e-05,
      "loss": 0.8618,
      "num_input_tokens_seen": 23422088,
      "step": 40375
    },
    {
      "epoch": 6.014298480786416,
      "grad_norm": 11.625,
      "learning_rate": 4.411136765048204e-05,
      "loss": 1.0792,
      "num_input_tokens_seen": 23424968,
      "step": 40380
    },
    {
      "epoch": 6.015043193327376,
      "grad_norm": 15.875,
      "learning_rate": 4.410927266034716e-05,
      "loss": 0.9634,
      "num_input_tokens_seen": 23428168,
      "step": 40385
    },
    {
      "epoch": 6.015787905868335,
      "grad_norm": 17.0,
      "learning_rate": 4.4107177347380545e-05,
      "loss": 1.2604,
      "num_input_tokens_seen": 23431112,
      "step": 40390
    },
    {
      "epoch": 6.016532618409294,
      "grad_norm": 11.9375,
      "learning_rate": 4.4105081711617594e-05,
      "loss": 1.2495,
      "num_input_tokens_seen": 23433640,
      "step": 40395
    },
    {
      "epoch": 6.017277330950253,
      "grad_norm": 8.0625,
      "learning_rate": 4.410298575309369e-05,
      "loss": 0.9932,
      "num_input_tokens_seen": 23436552,
      "step": 40400
    },
    {
      "epoch": 6.018022043491213,
      "grad_norm": 7.8125,
      "learning_rate": 4.4100889471844263e-05,
      "loss": 1.1937,
      "num_input_tokens_seen": 23439336,
      "step": 40405
    },
    {
      "epoch": 6.018766756032171,
      "grad_norm": 8.4375,
      "learning_rate": 4.4098792867904724e-05,
      "loss": 0.957,
      "num_input_tokens_seen": 23442120,
      "step": 40410
    },
    {
      "epoch": 6.019511468573131,
      "grad_norm": 7.03125,
      "learning_rate": 4.409669594131049e-05,
      "loss": 0.7427,
      "num_input_tokens_seen": 23444936,
      "step": 40415
    },
    {
      "epoch": 6.02025618111409,
      "grad_norm": 13.875,
      "learning_rate": 4.409459869209699e-05,
      "loss": 1.2734,
      "num_input_tokens_seen": 23447720,
      "step": 40420
    },
    {
      "epoch": 6.021000893655049,
      "grad_norm": 8.0,
      "learning_rate": 4.409250112029965e-05,
      "loss": 1.0004,
      "num_input_tokens_seen": 23450568,
      "step": 40425
    },
    {
      "epoch": 6.021745606196008,
      "grad_norm": 9.6875,
      "learning_rate": 4.4090403225953915e-05,
      "loss": 1.042,
      "num_input_tokens_seen": 23453320,
      "step": 40430
    },
    {
      "epoch": 6.022490318736968,
      "grad_norm": 15.0625,
      "learning_rate": 4.408830500909521e-05,
      "loss": 1.0424,
      "num_input_tokens_seen": 23456136,
      "step": 40435
    },
    {
      "epoch": 6.023235031277927,
      "grad_norm": 7.90625,
      "learning_rate": 4.408620646975899e-05,
      "loss": 1.1876,
      "num_input_tokens_seen": 23459272,
      "step": 40440
    },
    {
      "epoch": 6.023979743818886,
      "grad_norm": 7.78125,
      "learning_rate": 4.408410760798072e-05,
      "loss": 1.1974,
      "num_input_tokens_seen": 23462056,
      "step": 40445
    },
    {
      "epoch": 6.024724456359845,
      "grad_norm": 12.1875,
      "learning_rate": 4.408200842379584e-05,
      "loss": 0.8492,
      "num_input_tokens_seen": 23465224,
      "step": 40450
    },
    {
      "epoch": 6.025469168900805,
      "grad_norm": 10.75,
      "learning_rate": 4.407990891723983e-05,
      "loss": 1.1211,
      "num_input_tokens_seen": 23467848,
      "step": 40455
    },
    {
      "epoch": 6.026213881441763,
      "grad_norm": 17.75,
      "learning_rate": 4.407780908834814e-05,
      "loss": 1.3514,
      "num_input_tokens_seen": 23470472,
      "step": 40460
    },
    {
      "epoch": 6.026958593982723,
      "grad_norm": 8.5,
      "learning_rate": 4.407570893715627e-05,
      "loss": 1.0025,
      "num_input_tokens_seen": 23472936,
      "step": 40465
    },
    {
      "epoch": 6.027703306523682,
      "grad_norm": 8.25,
      "learning_rate": 4.4073608463699676e-05,
      "loss": 0.8877,
      "num_input_tokens_seen": 23475816,
      "step": 40470
    },
    {
      "epoch": 6.0284480190646414,
      "grad_norm": 16.25,
      "learning_rate": 4.4071507668013854e-05,
      "loss": 1.2184,
      "num_input_tokens_seen": 23478632,
      "step": 40475
    },
    {
      "epoch": 6.0291927316056,
      "grad_norm": 4.625,
      "learning_rate": 4.406940655013429e-05,
      "loss": 1.11,
      "num_input_tokens_seen": 23481448,
      "step": 40480
    },
    {
      "epoch": 6.02993744414656,
      "grad_norm": 7.0625,
      "learning_rate": 4.406730511009649e-05,
      "loss": 0.8992,
      "num_input_tokens_seen": 23484552,
      "step": 40485
    },
    {
      "epoch": 6.030682156687519,
      "grad_norm": 5.40625,
      "learning_rate": 4.406520334793595e-05,
      "loss": 0.8503,
      "num_input_tokens_seen": 23487336,
      "step": 40490
    },
    {
      "epoch": 6.031426869228477,
      "grad_norm": 9.25,
      "learning_rate": 4.4063101263688164e-05,
      "loss": 0.977,
      "num_input_tokens_seen": 23490088,
      "step": 40495
    },
    {
      "epoch": 6.032171581769437,
      "grad_norm": 5.0625,
      "learning_rate": 4.406099885738866e-05,
      "loss": 1.0846,
      "num_input_tokens_seen": 23493096,
      "step": 40500
    },
    {
      "epoch": 6.032916294310396,
      "grad_norm": 12.125,
      "learning_rate": 4.405889612907296e-05,
      "loss": 1.3389,
      "num_input_tokens_seen": 23496392,
      "step": 40505
    },
    {
      "epoch": 6.033661006851355,
      "grad_norm": 14.25,
      "learning_rate": 4.405679307877658e-05,
      "loss": 1.1211,
      "num_input_tokens_seen": 23499432,
      "step": 40510
    },
    {
      "epoch": 6.034405719392314,
      "grad_norm": 12.4375,
      "learning_rate": 4.4054689706535044e-05,
      "loss": 1.3984,
      "num_input_tokens_seen": 23502184,
      "step": 40515
    },
    {
      "epoch": 6.035150431933274,
      "grad_norm": 16.125,
      "learning_rate": 4.40525860123839e-05,
      "loss": 1.2547,
      "num_input_tokens_seen": 23504840,
      "step": 40520
    },
    {
      "epoch": 6.035895144474233,
      "grad_norm": 4.75,
      "learning_rate": 4.405048199635868e-05,
      "loss": 0.893,
      "num_input_tokens_seen": 23507528,
      "step": 40525
    },
    {
      "epoch": 6.036639857015192,
      "grad_norm": 9.8125,
      "learning_rate": 4.404837765849492e-05,
      "loss": 1.0566,
      "num_input_tokens_seen": 23510184,
      "step": 40530
    },
    {
      "epoch": 6.037384569556151,
      "grad_norm": 10.625,
      "learning_rate": 4.4046272998828186e-05,
      "loss": 1.0287,
      "num_input_tokens_seen": 23512744,
      "step": 40535
    },
    {
      "epoch": 6.038129282097111,
      "grad_norm": 14.875,
      "learning_rate": 4.4044168017394025e-05,
      "loss": 1.0359,
      "num_input_tokens_seen": 23515400,
      "step": 40540
    },
    {
      "epoch": 6.038873994638069,
      "grad_norm": 12.0625,
      "learning_rate": 4.4042062714228e-05,
      "loss": 1.1128,
      "num_input_tokens_seen": 23518440,
      "step": 40545
    },
    {
      "epoch": 6.039618707179029,
      "grad_norm": 9.3125,
      "learning_rate": 4.403995708936568e-05,
      "loss": 0.7244,
      "num_input_tokens_seen": 23521320,
      "step": 40550
    },
    {
      "epoch": 6.040363419719988,
      "grad_norm": 7.4375,
      "learning_rate": 4.403785114284263e-05,
      "loss": 1.0852,
      "num_input_tokens_seen": 23524200,
      "step": 40555
    },
    {
      "epoch": 6.0411081322609474,
      "grad_norm": 10.5,
      "learning_rate": 4.4035744874694444e-05,
      "loss": 1.2907,
      "num_input_tokens_seen": 23526888,
      "step": 40560
    },
    {
      "epoch": 6.041852844801906,
      "grad_norm": 13.75,
      "learning_rate": 4.403363828495669e-05,
      "loss": 1.1235,
      "num_input_tokens_seen": 23529800,
      "step": 40565
    },
    {
      "epoch": 6.042597557342866,
      "grad_norm": 6.625,
      "learning_rate": 4.403153137366497e-05,
      "loss": 1.4043,
      "num_input_tokens_seen": 23532584,
      "step": 40570
    },
    {
      "epoch": 6.043342269883825,
      "grad_norm": 9.625,
      "learning_rate": 4.402942414085486e-05,
      "loss": 0.8103,
      "num_input_tokens_seen": 23535368,
      "step": 40575
    },
    {
      "epoch": 6.044086982424784,
      "grad_norm": 12.625,
      "learning_rate": 4.4027316586561976e-05,
      "loss": 1.1968,
      "num_input_tokens_seen": 23538344,
      "step": 40580
    },
    {
      "epoch": 6.044831694965743,
      "grad_norm": 9.9375,
      "learning_rate": 4.402520871082191e-05,
      "loss": 1.2558,
      "num_input_tokens_seen": 23541352,
      "step": 40585
    },
    {
      "epoch": 6.045576407506703,
      "grad_norm": 8.8125,
      "learning_rate": 4.402310051367029e-05,
      "loss": 1.1665,
      "num_input_tokens_seen": 23544392,
      "step": 40590
    },
    {
      "epoch": 6.046321120047661,
      "grad_norm": 8.1875,
      "learning_rate": 4.4020991995142716e-05,
      "loss": 1.2824,
      "num_input_tokens_seen": 23547336,
      "step": 40595
    },
    {
      "epoch": 6.047065832588621,
      "grad_norm": 11.625,
      "learning_rate": 4.401888315527481e-05,
      "loss": 1.2,
      "num_input_tokens_seen": 23549896,
      "step": 40600
    },
    {
      "epoch": 6.04781054512958,
      "grad_norm": 9.3125,
      "learning_rate": 4.40167739941022e-05,
      "loss": 0.9297,
      "num_input_tokens_seen": 23552680,
      "step": 40605
    },
    {
      "epoch": 6.0485552576705395,
      "grad_norm": 8.3125,
      "learning_rate": 4.401466451166053e-05,
      "loss": 1.0775,
      "num_input_tokens_seen": 23555368,
      "step": 40610
    },
    {
      "epoch": 6.049299970211498,
      "grad_norm": 6.40625,
      "learning_rate": 4.401255470798543e-05,
      "loss": 1.0207,
      "num_input_tokens_seen": 23558344,
      "step": 40615
    },
    {
      "epoch": 6.050044682752458,
      "grad_norm": 13.3125,
      "learning_rate": 4.401044458311254e-05,
      "loss": 1.1734,
      "num_input_tokens_seen": 23561640,
      "step": 40620
    },
    {
      "epoch": 6.050789395293417,
      "grad_norm": 5.21875,
      "learning_rate": 4.40083341370775e-05,
      "loss": 1.0391,
      "num_input_tokens_seen": 23564456,
      "step": 40625
    },
    {
      "epoch": 6.051534107834376,
      "grad_norm": 8.75,
      "learning_rate": 4.400622336991599e-05,
      "loss": 1.4066,
      "num_input_tokens_seen": 23567112,
      "step": 40630
    },
    {
      "epoch": 6.052278820375335,
      "grad_norm": 6.0,
      "learning_rate": 4.400411228166364e-05,
      "loss": 1.4922,
      "num_input_tokens_seen": 23570312,
      "step": 40635
    },
    {
      "epoch": 6.053023532916295,
      "grad_norm": 10.0,
      "learning_rate": 4.400200087235613e-05,
      "loss": 1.2387,
      "num_input_tokens_seen": 23573128,
      "step": 40640
    },
    {
      "epoch": 6.0537682454572534,
      "grad_norm": 10.125,
      "learning_rate": 4.399988914202913e-05,
      "loss": 0.8983,
      "num_input_tokens_seen": 23575816,
      "step": 40645
    },
    {
      "epoch": 6.054512957998213,
      "grad_norm": 12.625,
      "learning_rate": 4.399777709071832e-05,
      "loss": 0.8439,
      "num_input_tokens_seen": 23578792,
      "step": 40650
    },
    {
      "epoch": 6.055257670539172,
      "grad_norm": 16.5,
      "learning_rate": 4.399566471845937e-05,
      "loss": 1.6127,
      "num_input_tokens_seen": 23581480,
      "step": 40655
    },
    {
      "epoch": 6.0560023830801315,
      "grad_norm": 9.6875,
      "learning_rate": 4.3993552025287966e-05,
      "loss": 0.8316,
      "num_input_tokens_seen": 23584520,
      "step": 40660
    },
    {
      "epoch": 6.05674709562109,
      "grad_norm": 11.3125,
      "learning_rate": 4.399143901123981e-05,
      "loss": 1.7963,
      "num_input_tokens_seen": 23587464,
      "step": 40665
    },
    {
      "epoch": 6.057491808162049,
      "grad_norm": 9.0,
      "learning_rate": 4.398932567635059e-05,
      "loss": 1.1862,
      "num_input_tokens_seen": 23590440,
      "step": 40670
    },
    {
      "epoch": 6.058236520703009,
      "grad_norm": 12.625,
      "learning_rate": 4.398721202065602e-05,
      "loss": 1.4726,
      "num_input_tokens_seen": 23593032,
      "step": 40675
    },
    {
      "epoch": 6.058981233243967,
      "grad_norm": 12.625,
      "learning_rate": 4.398509804419179e-05,
      "loss": 0.9301,
      "num_input_tokens_seen": 23595752,
      "step": 40680
    },
    {
      "epoch": 6.059725945784927,
      "grad_norm": 13.625,
      "learning_rate": 4.3982983746993636e-05,
      "loss": 0.9471,
      "num_input_tokens_seen": 23598440,
      "step": 40685
    },
    {
      "epoch": 6.060470658325886,
      "grad_norm": 5.625,
      "learning_rate": 4.398086912909726e-05,
      "loss": 0.7949,
      "num_input_tokens_seen": 23601224,
      "step": 40690
    },
    {
      "epoch": 6.0612153708668455,
      "grad_norm": 19.375,
      "learning_rate": 4.397875419053838e-05,
      "loss": 1.3242,
      "num_input_tokens_seen": 23604264,
      "step": 40695
    },
    {
      "epoch": 6.061960083407804,
      "grad_norm": 9.375,
      "learning_rate": 4.397663893135275e-05,
      "loss": 1.0517,
      "num_input_tokens_seen": 23607144,
      "step": 40700
    },
    {
      "epoch": 6.062704795948764,
      "grad_norm": 15.5,
      "learning_rate": 4.397452335157609e-05,
      "loss": 1.0778,
      "num_input_tokens_seen": 23610024,
      "step": 40705
    },
    {
      "epoch": 6.063449508489723,
      "grad_norm": 15.75,
      "learning_rate": 4.397240745124414e-05,
      "loss": 0.9347,
      "num_input_tokens_seen": 23612840,
      "step": 40710
    },
    {
      "epoch": 6.064194221030682,
      "grad_norm": 10.8125,
      "learning_rate": 4.397029123039266e-05,
      "loss": 1.2938,
      "num_input_tokens_seen": 23615656,
      "step": 40715
    },
    {
      "epoch": 6.064938933571641,
      "grad_norm": 13.5625,
      "learning_rate": 4.396817468905738e-05,
      "loss": 1.0416,
      "num_input_tokens_seen": 23618536,
      "step": 40720
    },
    {
      "epoch": 6.065683646112601,
      "grad_norm": 22.375,
      "learning_rate": 4.396605782727406e-05,
      "loss": 0.8116,
      "num_input_tokens_seen": 23621512,
      "step": 40725
    },
    {
      "epoch": 6.0664283586535594,
      "grad_norm": 10.75,
      "learning_rate": 4.3963940645078484e-05,
      "loss": 1.1331,
      "num_input_tokens_seen": 23624136,
      "step": 40730
    },
    {
      "epoch": 6.067173071194519,
      "grad_norm": 6.59375,
      "learning_rate": 4.3961823142506395e-05,
      "loss": 1.1906,
      "num_input_tokens_seen": 23627624,
      "step": 40735
    },
    {
      "epoch": 6.067917783735478,
      "grad_norm": 9.5625,
      "learning_rate": 4.395970531959358e-05,
      "loss": 1.3065,
      "num_input_tokens_seen": 23630888,
      "step": 40740
    },
    {
      "epoch": 6.0686624962764375,
      "grad_norm": 12.5625,
      "learning_rate": 4.395758717637581e-05,
      "loss": 0.9625,
      "num_input_tokens_seen": 23633992,
      "step": 40745
    },
    {
      "epoch": 6.069407208817396,
      "grad_norm": 5.8125,
      "learning_rate": 4.3955468712888884e-05,
      "loss": 1.3972,
      "num_input_tokens_seen": 23636872,
      "step": 40750
    },
    {
      "epoch": 6.070151921358356,
      "grad_norm": 9.0625,
      "learning_rate": 4.395334992916857e-05,
      "loss": 0.8976,
      "num_input_tokens_seen": 23639656,
      "step": 40755
    },
    {
      "epoch": 6.070896633899315,
      "grad_norm": 9.1875,
      "learning_rate": 4.395123082525067e-05,
      "loss": 1.2823,
      "num_input_tokens_seen": 23642696,
      "step": 40760
    },
    {
      "epoch": 6.071641346440274,
      "grad_norm": 9.5625,
      "learning_rate": 4.394911140117099e-05,
      "loss": 1.1422,
      "num_input_tokens_seen": 23645480,
      "step": 40765
    },
    {
      "epoch": 6.072386058981233,
      "grad_norm": 10.8125,
      "learning_rate": 4.3946991656965334e-05,
      "loss": 1.392,
      "num_input_tokens_seen": 23648456,
      "step": 40770
    },
    {
      "epoch": 6.073130771522193,
      "grad_norm": 16.875,
      "learning_rate": 4.394487159266951e-05,
      "loss": 1.0436,
      "num_input_tokens_seen": 23651208,
      "step": 40775
    },
    {
      "epoch": 6.0738754840631515,
      "grad_norm": 16.625,
      "learning_rate": 4.394275120831933e-05,
      "loss": 1.2762,
      "num_input_tokens_seen": 23654344,
      "step": 40780
    },
    {
      "epoch": 6.074620196604111,
      "grad_norm": 8.4375,
      "learning_rate": 4.394063050395063e-05,
      "loss": 0.9752,
      "num_input_tokens_seen": 23657096,
      "step": 40785
    },
    {
      "epoch": 6.07536490914507,
      "grad_norm": 15.5625,
      "learning_rate": 4.393850947959922e-05,
      "loss": 1.3103,
      "num_input_tokens_seen": 23659976,
      "step": 40790
    },
    {
      "epoch": 6.0761096216860295,
      "grad_norm": 18.75,
      "learning_rate": 4.3936388135300946e-05,
      "loss": 1.023,
      "num_input_tokens_seen": 23663016,
      "step": 40795
    },
    {
      "epoch": 6.076854334226988,
      "grad_norm": 8.8125,
      "learning_rate": 4.3934266471091635e-05,
      "loss": 0.924,
      "num_input_tokens_seen": 23666184,
      "step": 40800
    },
    {
      "epoch": 6.077599046767948,
      "grad_norm": 11.875,
      "learning_rate": 4.393214448700713e-05,
      "loss": 0.8601,
      "num_input_tokens_seen": 23669032,
      "step": 40805
    },
    {
      "epoch": 6.078343759308907,
      "grad_norm": 7.375,
      "learning_rate": 4.39300221830833e-05,
      "loss": 0.9752,
      "num_input_tokens_seen": 23671976,
      "step": 40810
    },
    {
      "epoch": 6.079088471849866,
      "grad_norm": 6.84375,
      "learning_rate": 4.392789955935598e-05,
      "loss": 0.9958,
      "num_input_tokens_seen": 23674952,
      "step": 40815
    },
    {
      "epoch": 6.079833184390825,
      "grad_norm": 10.25,
      "learning_rate": 4.3925776615861034e-05,
      "loss": 1.0117,
      "num_input_tokens_seen": 23677928,
      "step": 40820
    },
    {
      "epoch": 6.080577896931785,
      "grad_norm": 4.5625,
      "learning_rate": 4.392365335263432e-05,
      "loss": 1.0554,
      "num_input_tokens_seen": 23680808,
      "step": 40825
    },
    {
      "epoch": 6.0813226094727435,
      "grad_norm": 17.875,
      "learning_rate": 4.392152976971173e-05,
      "loss": 1.2872,
      "num_input_tokens_seen": 23683688,
      "step": 40830
    },
    {
      "epoch": 6.082067322013703,
      "grad_norm": 10.375,
      "learning_rate": 4.3919405867129114e-05,
      "loss": 1.0906,
      "num_input_tokens_seen": 23686376,
      "step": 40835
    },
    {
      "epoch": 6.082812034554662,
      "grad_norm": 11.1875,
      "learning_rate": 4.391728164492237e-05,
      "loss": 1.5077,
      "num_input_tokens_seen": 23689416,
      "step": 40840
    },
    {
      "epoch": 6.083556747095621,
      "grad_norm": 10.9375,
      "learning_rate": 4.391515710312738e-05,
      "loss": 1.0795,
      "num_input_tokens_seen": 23692296,
      "step": 40845
    },
    {
      "epoch": 6.08430145963658,
      "grad_norm": 5.09375,
      "learning_rate": 4.391303224178003e-05,
      "loss": 1.0678,
      "num_input_tokens_seen": 23695080,
      "step": 40850
    },
    {
      "epoch": 6.085046172177539,
      "grad_norm": 7.8125,
      "learning_rate": 4.391090706091623e-05,
      "loss": 1.0239,
      "num_input_tokens_seen": 23697992,
      "step": 40855
    },
    {
      "epoch": 6.085790884718499,
      "grad_norm": 12.0,
      "learning_rate": 4.390878156057186e-05,
      "loss": 1.3249,
      "num_input_tokens_seen": 23700616,
      "step": 40860
    },
    {
      "epoch": 6.0865355972594575,
      "grad_norm": 9.3125,
      "learning_rate": 4.390665574078286e-05,
      "loss": 0.996,
      "num_input_tokens_seen": 23703464,
      "step": 40865
    },
    {
      "epoch": 6.087280309800417,
      "grad_norm": 11.75,
      "learning_rate": 4.390452960158512e-05,
      "loss": 1.0484,
      "num_input_tokens_seen": 23706792,
      "step": 40870
    },
    {
      "epoch": 6.088025022341376,
      "grad_norm": 8.1875,
      "learning_rate": 4.390240314301457e-05,
      "loss": 0.9356,
      "num_input_tokens_seen": 23709896,
      "step": 40875
    },
    {
      "epoch": 6.0887697348823355,
      "grad_norm": 10.25,
      "learning_rate": 4.3900276365107126e-05,
      "loss": 1.0017,
      "num_input_tokens_seen": 23712872,
      "step": 40880
    },
    {
      "epoch": 6.089514447423294,
      "grad_norm": 10.5,
      "learning_rate": 4.3898149267898727e-05,
      "loss": 1.1278,
      "num_input_tokens_seen": 23715784,
      "step": 40885
    },
    {
      "epoch": 6.090259159964254,
      "grad_norm": 14.0,
      "learning_rate": 4.3896021851425306e-05,
      "loss": 1.1956,
      "num_input_tokens_seen": 23718792,
      "step": 40890
    },
    {
      "epoch": 6.091003872505213,
      "grad_norm": 11.8125,
      "learning_rate": 4.389389411572279e-05,
      "loss": 1.0043,
      "num_input_tokens_seen": 23721416,
      "step": 40895
    },
    {
      "epoch": 6.091748585046172,
      "grad_norm": 15.25,
      "learning_rate": 4.389176606082714e-05,
      "loss": 1.052,
      "num_input_tokens_seen": 23725416,
      "step": 40900
    },
    {
      "epoch": 6.092493297587131,
      "grad_norm": 9.625,
      "learning_rate": 4.388963768677431e-05,
      "loss": 0.9934,
      "num_input_tokens_seen": 23728232,
      "step": 40905
    },
    {
      "epoch": 6.093238010128091,
      "grad_norm": 10.3125,
      "learning_rate": 4.388750899360025e-05,
      "loss": 0.9641,
      "num_input_tokens_seen": 23731272,
      "step": 40910
    },
    {
      "epoch": 6.0939827226690495,
      "grad_norm": 6.375,
      "learning_rate": 4.3885379981340905e-05,
      "loss": 0.994,
      "num_input_tokens_seen": 23734248,
      "step": 40915
    },
    {
      "epoch": 6.094727435210009,
      "grad_norm": 8.4375,
      "learning_rate": 4.388325065003228e-05,
      "loss": 0.9157,
      "num_input_tokens_seen": 23737448,
      "step": 40920
    },
    {
      "epoch": 6.095472147750968,
      "grad_norm": 11.75,
      "learning_rate": 4.3881120999710315e-05,
      "loss": 1.0766,
      "num_input_tokens_seen": 23740456,
      "step": 40925
    },
    {
      "epoch": 6.0962168602919276,
      "grad_norm": 6.1875,
      "learning_rate": 4.3878991030411e-05,
      "loss": 1.2713,
      "num_input_tokens_seen": 23743208,
      "step": 40930
    },
    {
      "epoch": 6.096961572832886,
      "grad_norm": 12.8125,
      "learning_rate": 4.387686074217032e-05,
      "loss": 0.9967,
      "num_input_tokens_seen": 23745864,
      "step": 40935
    },
    {
      "epoch": 6.097706285373846,
      "grad_norm": 16.0,
      "learning_rate": 4.387473013502427e-05,
      "loss": 1.1991,
      "num_input_tokens_seen": 23748808,
      "step": 40940
    },
    {
      "epoch": 6.098450997914805,
      "grad_norm": 10.0,
      "learning_rate": 4.387259920900884e-05,
      "loss": 1.0568,
      "num_input_tokens_seen": 23751592,
      "step": 40945
    },
    {
      "epoch": 6.099195710455764,
      "grad_norm": 11.3125,
      "learning_rate": 4.3870467964160015e-05,
      "loss": 1.1542,
      "num_input_tokens_seen": 23754440,
      "step": 40950
    },
    {
      "epoch": 6.099940422996723,
      "grad_norm": 17.0,
      "learning_rate": 4.3868336400513823e-05,
      "loss": 0.9188,
      "num_input_tokens_seen": 23757480,
      "step": 40955
    },
    {
      "epoch": 6.100685135537683,
      "grad_norm": 9.75,
      "learning_rate": 4.386620451810626e-05,
      "loss": 0.8607,
      "num_input_tokens_seen": 23760392,
      "step": 40960
    },
    {
      "epoch": 6.1014298480786415,
      "grad_norm": 14.375,
      "learning_rate": 4.3864072316973345e-05,
      "loss": 1.192,
      "num_input_tokens_seen": 23763496,
      "step": 40965
    },
    {
      "epoch": 6.102174560619601,
      "grad_norm": 11.625,
      "learning_rate": 4.386193979715111e-05,
      "loss": 0.9501,
      "num_input_tokens_seen": 23766088,
      "step": 40970
    },
    {
      "epoch": 6.10291927316056,
      "grad_norm": 9.5625,
      "learning_rate": 4.385980695867556e-05,
      "loss": 1.3624,
      "num_input_tokens_seen": 23769064,
      "step": 40975
    },
    {
      "epoch": 6.10366398570152,
      "grad_norm": 9.9375,
      "learning_rate": 4.385767380158275e-05,
      "loss": 0.8078,
      "num_input_tokens_seen": 23771848,
      "step": 40980
    },
    {
      "epoch": 6.104408698242478,
      "grad_norm": 5.15625,
      "learning_rate": 4.38555403259087e-05,
      "loss": 0.8927,
      "num_input_tokens_seen": 23774952,
      "step": 40985
    },
    {
      "epoch": 6.105153410783438,
      "grad_norm": 7.5625,
      "learning_rate": 4.3853406531689465e-05,
      "loss": 1.0558,
      "num_input_tokens_seen": 23777640,
      "step": 40990
    },
    {
      "epoch": 6.105898123324397,
      "grad_norm": 9.6875,
      "learning_rate": 4.3851272418961085e-05,
      "loss": 1.0637,
      "num_input_tokens_seen": 23780296,
      "step": 40995
    },
    {
      "epoch": 6.106642835865356,
      "grad_norm": 12.1875,
      "learning_rate": 4.384913798775962e-05,
      "loss": 1.0967,
      "num_input_tokens_seen": 23783176,
      "step": 41000
    },
    {
      "epoch": 6.107387548406315,
      "grad_norm": 14.1875,
      "learning_rate": 4.384700323812112e-05,
      "loss": 1.1659,
      "num_input_tokens_seen": 23785672,
      "step": 41005
    },
    {
      "epoch": 6.108132260947274,
      "grad_norm": 13.875,
      "learning_rate": 4.3844868170081665e-05,
      "loss": 0.7502,
      "num_input_tokens_seen": 23788552,
      "step": 41010
    },
    {
      "epoch": 6.1088769734882336,
      "grad_norm": 14.0625,
      "learning_rate": 4.384273278367731e-05,
      "loss": 1.4681,
      "num_input_tokens_seen": 23791400,
      "step": 41015
    },
    {
      "epoch": 6.109621686029192,
      "grad_norm": 9.0625,
      "learning_rate": 4.3840597078944135e-05,
      "loss": 0.8947,
      "num_input_tokens_seen": 23794088,
      "step": 41020
    },
    {
      "epoch": 6.110366398570152,
      "grad_norm": 9.8125,
      "learning_rate": 4.3838461055918226e-05,
      "loss": 1.257,
      "num_input_tokens_seen": 23796808,
      "step": 41025
    },
    {
      "epoch": 6.111111111111111,
      "grad_norm": 21.0,
      "learning_rate": 4.383632471463566e-05,
      "loss": 1.5967,
      "num_input_tokens_seen": 23800168,
      "step": 41030
    },
    {
      "epoch": 6.11185582365207,
      "grad_norm": 6.3125,
      "learning_rate": 4.383418805513253e-05,
      "loss": 1.0925,
      "num_input_tokens_seen": 23803080,
      "step": 41035
    },
    {
      "epoch": 6.112600536193029,
      "grad_norm": 15.3125,
      "learning_rate": 4.3832051077444937e-05,
      "loss": 0.9798,
      "num_input_tokens_seen": 23805960,
      "step": 41040
    },
    {
      "epoch": 6.113345248733989,
      "grad_norm": 15.6875,
      "learning_rate": 4.382991378160898e-05,
      "loss": 1.047,
      "num_input_tokens_seen": 23808904,
      "step": 41045
    },
    {
      "epoch": 6.1140899612749475,
      "grad_norm": 8.6875,
      "learning_rate": 4.3827776167660775e-05,
      "loss": 0.5582,
      "num_input_tokens_seen": 23811656,
      "step": 41050
    },
    {
      "epoch": 6.114834673815907,
      "grad_norm": 19.75,
      "learning_rate": 4.382563823563642e-05,
      "loss": 1.3093,
      "num_input_tokens_seen": 23814632,
      "step": 41055
    },
    {
      "epoch": 6.115579386356866,
      "grad_norm": 7.46875,
      "learning_rate": 4.382349998557204e-05,
      "loss": 0.901,
      "num_input_tokens_seen": 23817672,
      "step": 41060
    },
    {
      "epoch": 6.116324098897826,
      "grad_norm": 11.75,
      "learning_rate": 4.382136141750376e-05,
      "loss": 1.0832,
      "num_input_tokens_seen": 23820680,
      "step": 41065
    },
    {
      "epoch": 6.117068811438784,
      "grad_norm": 7.15625,
      "learning_rate": 4.381922253146771e-05,
      "loss": 1.0334,
      "num_input_tokens_seen": 23823816,
      "step": 41070
    },
    {
      "epoch": 6.117813523979744,
      "grad_norm": 18.0,
      "learning_rate": 4.381708332750002e-05,
      "loss": 1.0549,
      "num_input_tokens_seen": 23826792,
      "step": 41075
    },
    {
      "epoch": 6.118558236520703,
      "grad_norm": 10.0,
      "learning_rate": 4.381494380563683e-05,
      "loss": 1.2368,
      "num_input_tokens_seen": 23829352,
      "step": 41080
    },
    {
      "epoch": 6.119302949061662,
      "grad_norm": 6.40625,
      "learning_rate": 4.3812803965914296e-05,
      "loss": 1.309,
      "num_input_tokens_seen": 23832200,
      "step": 41085
    },
    {
      "epoch": 6.120047661602621,
      "grad_norm": 10.1875,
      "learning_rate": 4.381066380836855e-05,
      "loss": 0.9828,
      "num_input_tokens_seen": 23834824,
      "step": 41090
    },
    {
      "epoch": 6.120792374143581,
      "grad_norm": 6.46875,
      "learning_rate": 4.380852333303576e-05,
      "loss": 0.9071,
      "num_input_tokens_seen": 23837640,
      "step": 41095
    },
    {
      "epoch": 6.1215370866845396,
      "grad_norm": 14.6875,
      "learning_rate": 4.380638253995209e-05,
      "loss": 0.9674,
      "num_input_tokens_seen": 23840616,
      "step": 41100
    },
    {
      "epoch": 6.122281799225499,
      "grad_norm": 17.375,
      "learning_rate": 4.380424142915369e-05,
      "loss": 1.217,
      "num_input_tokens_seen": 23843560,
      "step": 41105
    },
    {
      "epoch": 6.123026511766458,
      "grad_norm": 7.6875,
      "learning_rate": 4.380210000067675e-05,
      "loss": 1.3032,
      "num_input_tokens_seen": 23846792,
      "step": 41110
    },
    {
      "epoch": 6.123771224307418,
      "grad_norm": 5.59375,
      "learning_rate": 4.379995825455744e-05,
      "loss": 0.8708,
      "num_input_tokens_seen": 23849576,
      "step": 41115
    },
    {
      "epoch": 6.124515936848376,
      "grad_norm": 20.25,
      "learning_rate": 4.379781619083195e-05,
      "loss": 0.843,
      "num_input_tokens_seen": 23852520,
      "step": 41120
    },
    {
      "epoch": 6.125260649389336,
      "grad_norm": 8.5625,
      "learning_rate": 4.379567380953645e-05,
      "loss": 1.0203,
      "num_input_tokens_seen": 23855464,
      "step": 41125
    },
    {
      "epoch": 6.126005361930295,
      "grad_norm": 16.875,
      "learning_rate": 4.3793531110707143e-05,
      "loss": 0.9201,
      "num_input_tokens_seen": 23858408,
      "step": 41130
    },
    {
      "epoch": 6.126750074471254,
      "grad_norm": 8.5625,
      "learning_rate": 4.3791388094380236e-05,
      "loss": 1.0433,
      "num_input_tokens_seen": 23861864,
      "step": 41135
    },
    {
      "epoch": 6.127494787012213,
      "grad_norm": 7.84375,
      "learning_rate": 4.378924476059192e-05,
      "loss": 0.8008,
      "num_input_tokens_seen": 23864616,
      "step": 41140
    },
    {
      "epoch": 6.128239499553173,
      "grad_norm": 7.21875,
      "learning_rate": 4.378710110937842e-05,
      "loss": 1.0451,
      "num_input_tokens_seen": 23867688,
      "step": 41145
    },
    {
      "epoch": 6.128984212094132,
      "grad_norm": 7.34375,
      "learning_rate": 4.378495714077593e-05,
      "loss": 1.046,
      "num_input_tokens_seen": 23870600,
      "step": 41150
    },
    {
      "epoch": 6.129728924635091,
      "grad_norm": 6.46875,
      "learning_rate": 4.3782812854820687e-05,
      "loss": 0.9322,
      "num_input_tokens_seen": 23873416,
      "step": 41155
    },
    {
      "epoch": 6.13047363717605,
      "grad_norm": 12.0,
      "learning_rate": 4.378066825154891e-05,
      "loss": 1.3792,
      "num_input_tokens_seen": 23876456,
      "step": 41160
    },
    {
      "epoch": 6.13121834971701,
      "grad_norm": 11.375,
      "learning_rate": 4.3778523330996824e-05,
      "loss": 0.8919,
      "num_input_tokens_seen": 23879176,
      "step": 41165
    },
    {
      "epoch": 6.131963062257968,
      "grad_norm": 5.84375,
      "learning_rate": 4.377637809320068e-05,
      "loss": 1.0785,
      "num_input_tokens_seen": 23882280,
      "step": 41170
    },
    {
      "epoch": 6.132707774798928,
      "grad_norm": 10.875,
      "learning_rate": 4.377423253819671e-05,
      "loss": 1.1037,
      "num_input_tokens_seen": 23884968,
      "step": 41175
    },
    {
      "epoch": 6.133452487339887,
      "grad_norm": 4.96875,
      "learning_rate": 4.377208666602116e-05,
      "loss": 1.0442,
      "num_input_tokens_seen": 23887880,
      "step": 41180
    },
    {
      "epoch": 6.134197199880846,
      "grad_norm": 16.875,
      "learning_rate": 4.3769940476710284e-05,
      "loss": 1.2368,
      "num_input_tokens_seen": 23890600,
      "step": 41185
    },
    {
      "epoch": 6.134941912421805,
      "grad_norm": 16.25,
      "learning_rate": 4.376779397030034e-05,
      "loss": 1.2349,
      "num_input_tokens_seen": 23893608,
      "step": 41190
    },
    {
      "epoch": 6.135686624962764,
      "grad_norm": 12.125,
      "learning_rate": 4.376564714682761e-05,
      "loss": 0.8393,
      "num_input_tokens_seen": 23896616,
      "step": 41195
    },
    {
      "epoch": 6.136431337503724,
      "grad_norm": 13.3125,
      "learning_rate": 4.376350000632832e-05,
      "loss": 1.0872,
      "num_input_tokens_seen": 23899848,
      "step": 41200
    },
    {
      "epoch": 6.137176050044682,
      "grad_norm": 6.6875,
      "learning_rate": 4.376135254883877e-05,
      "loss": 1.1212,
      "num_input_tokens_seen": 23902760,
      "step": 41205
    },
    {
      "epoch": 6.137920762585642,
      "grad_norm": 11.1875,
      "learning_rate": 4.375920477439525e-05,
      "loss": 0.9512,
      "num_input_tokens_seen": 23905256,
      "step": 41210
    },
    {
      "epoch": 6.138665475126601,
      "grad_norm": 4.15625,
      "learning_rate": 4.375705668303403e-05,
      "loss": 1.0911,
      "num_input_tokens_seen": 23907976,
      "step": 41215
    },
    {
      "epoch": 6.13941018766756,
      "grad_norm": 9.5625,
      "learning_rate": 4.37549082747914e-05,
      "loss": 1.0585,
      "num_input_tokens_seen": 23910856,
      "step": 41220
    },
    {
      "epoch": 6.140154900208519,
      "grad_norm": 9.6875,
      "learning_rate": 4.375275954970364e-05,
      "loss": 0.9149,
      "num_input_tokens_seen": 23913544,
      "step": 41225
    },
    {
      "epoch": 6.140899612749479,
      "grad_norm": 5.6875,
      "learning_rate": 4.3750610507807075e-05,
      "loss": 0.777,
      "num_input_tokens_seen": 23916456,
      "step": 41230
    },
    {
      "epoch": 6.141644325290438,
      "grad_norm": 14.4375,
      "learning_rate": 4.3748461149138016e-05,
      "loss": 1.0895,
      "num_input_tokens_seen": 23919208,
      "step": 41235
    },
    {
      "epoch": 6.142389037831397,
      "grad_norm": 12.0625,
      "learning_rate": 4.374631147373275e-05,
      "loss": 0.7376,
      "num_input_tokens_seen": 23922152,
      "step": 41240
    },
    {
      "epoch": 6.143133750372356,
      "grad_norm": 16.75,
      "learning_rate": 4.374416148162761e-05,
      "loss": 0.8768,
      "num_input_tokens_seen": 23925256,
      "step": 41245
    },
    {
      "epoch": 6.143878462913316,
      "grad_norm": 9.0,
      "learning_rate": 4.374201117285891e-05,
      "loss": 1.127,
      "num_input_tokens_seen": 23928072,
      "step": 41250
    },
    {
      "epoch": 6.144623175454274,
      "grad_norm": 10.625,
      "learning_rate": 4.3739860547462976e-05,
      "loss": 1.2621,
      "num_input_tokens_seen": 23930952,
      "step": 41255
    },
    {
      "epoch": 6.145367887995234,
      "grad_norm": 4.5,
      "learning_rate": 4.373770960547614e-05,
      "loss": 1.3368,
      "num_input_tokens_seen": 23933992,
      "step": 41260
    },
    {
      "epoch": 6.146112600536193,
      "grad_norm": 9.9375,
      "learning_rate": 4.3735558346934755e-05,
      "loss": 1.0956,
      "num_input_tokens_seen": 23936968,
      "step": 41265
    },
    {
      "epoch": 6.146857313077152,
      "grad_norm": 13.125,
      "learning_rate": 4.373340677187515e-05,
      "loss": 1.2006,
      "num_input_tokens_seen": 23940008,
      "step": 41270
    },
    {
      "epoch": 6.147602025618111,
      "grad_norm": 11.125,
      "learning_rate": 4.373125488033368e-05,
      "loss": 1.0737,
      "num_input_tokens_seen": 23943336,
      "step": 41275
    },
    {
      "epoch": 6.148346738159071,
      "grad_norm": 16.25,
      "learning_rate": 4.372910267234669e-05,
      "loss": 1.5368,
      "num_input_tokens_seen": 23946216,
      "step": 41280
    },
    {
      "epoch": 6.14909145070003,
      "grad_norm": 7.15625,
      "learning_rate": 4.3726950147950554e-05,
      "loss": 0.8723,
      "num_input_tokens_seen": 23949320,
      "step": 41285
    },
    {
      "epoch": 6.149836163240989,
      "grad_norm": 8.8125,
      "learning_rate": 4.372479730718162e-05,
      "loss": 1.1952,
      "num_input_tokens_seen": 23952136,
      "step": 41290
    },
    {
      "epoch": 6.150580875781948,
      "grad_norm": 16.375,
      "learning_rate": 4.3722644150076275e-05,
      "loss": 1.4183,
      "num_input_tokens_seen": 23954984,
      "step": 41295
    },
    {
      "epoch": 6.151325588322908,
      "grad_norm": 17.125,
      "learning_rate": 4.3720490676670886e-05,
      "loss": 1.2229,
      "num_input_tokens_seen": 23957736,
      "step": 41300
    },
    {
      "epoch": 6.152070300863866,
      "grad_norm": 9.875,
      "learning_rate": 4.371833688700182e-05,
      "loss": 0.8669,
      "num_input_tokens_seen": 23960488,
      "step": 41305
    },
    {
      "epoch": 6.152815013404826,
      "grad_norm": 17.0,
      "learning_rate": 4.3716182781105484e-05,
      "loss": 1.1017,
      "num_input_tokens_seen": 23963112,
      "step": 41310
    },
    {
      "epoch": 6.153559725945785,
      "grad_norm": 9.5625,
      "learning_rate": 4.3714028359018274e-05,
      "loss": 1.3499,
      "num_input_tokens_seen": 23965768,
      "step": 41315
    },
    {
      "epoch": 6.1543044384867445,
      "grad_norm": 13.1875,
      "learning_rate": 4.3711873620776566e-05,
      "loss": 1.2386,
      "num_input_tokens_seen": 23968840,
      "step": 41320
    },
    {
      "epoch": 6.155049151027703,
      "grad_norm": 10.875,
      "learning_rate": 4.370971856641677e-05,
      "loss": 1.0787,
      "num_input_tokens_seen": 23971624,
      "step": 41325
    },
    {
      "epoch": 6.155793863568663,
      "grad_norm": 4.46875,
      "learning_rate": 4.3707563195975296e-05,
      "loss": 1.0712,
      "num_input_tokens_seen": 23974472,
      "step": 41330
    },
    {
      "epoch": 6.156538576109622,
      "grad_norm": 16.0,
      "learning_rate": 4.370540750948855e-05,
      "loss": 1.1403,
      "num_input_tokens_seen": 23977320,
      "step": 41335
    },
    {
      "epoch": 6.157283288650581,
      "grad_norm": 13.8125,
      "learning_rate": 4.370325150699296e-05,
      "loss": 1.2172,
      "num_input_tokens_seen": 23980328,
      "step": 41340
    },
    {
      "epoch": 6.15802800119154,
      "grad_norm": 14.625,
      "learning_rate": 4.3701095188524943e-05,
      "loss": 1.1757,
      "num_input_tokens_seen": 23983240,
      "step": 41345
    },
    {
      "epoch": 6.1587727137325,
      "grad_norm": 5.96875,
      "learning_rate": 4.369893855412093e-05,
      "loss": 1.0634,
      "num_input_tokens_seen": 23986216,
      "step": 41350
    },
    {
      "epoch": 6.159517426273458,
      "grad_norm": 5.78125,
      "learning_rate": 4.369678160381736e-05,
      "loss": 1.1161,
      "num_input_tokens_seen": 23989544,
      "step": 41355
    },
    {
      "epoch": 6.160262138814417,
      "grad_norm": 10.3125,
      "learning_rate": 4.3694624337650656e-05,
      "loss": 1.2171,
      "num_input_tokens_seen": 23992392,
      "step": 41360
    },
    {
      "epoch": 6.161006851355377,
      "grad_norm": 7.125,
      "learning_rate": 4.369246675565729e-05,
      "loss": 0.8222,
      "num_input_tokens_seen": 23995400,
      "step": 41365
    },
    {
      "epoch": 6.161751563896336,
      "grad_norm": 11.3125,
      "learning_rate": 4.369030885787369e-05,
      "loss": 1.1286,
      "num_input_tokens_seen": 23998184,
      "step": 41370
    },
    {
      "epoch": 6.162496276437295,
      "grad_norm": 12.0,
      "learning_rate": 4.368815064433631e-05,
      "loss": 0.7934,
      "num_input_tokens_seen": 24001096,
      "step": 41375
    },
    {
      "epoch": 6.163240988978254,
      "grad_norm": 7.21875,
      "learning_rate": 4.368599211508162e-05,
      "loss": 0.8858,
      "num_input_tokens_seen": 24004168,
      "step": 41380
    },
    {
      "epoch": 6.163985701519214,
      "grad_norm": 4.8125,
      "learning_rate": 4.3683833270146095e-05,
      "loss": 1.1645,
      "num_input_tokens_seen": 24007144,
      "step": 41385
    },
    {
      "epoch": 6.164730414060172,
      "grad_norm": 7.15625,
      "learning_rate": 4.368167410956619e-05,
      "loss": 1.0698,
      "num_input_tokens_seen": 24010216,
      "step": 41390
    },
    {
      "epoch": 6.165475126601132,
      "grad_norm": 7.28125,
      "learning_rate": 4.367951463337839e-05,
      "loss": 0.8675,
      "num_input_tokens_seen": 24013032,
      "step": 41395
    },
    {
      "epoch": 6.166219839142091,
      "grad_norm": 10.5,
      "learning_rate": 4.367735484161918e-05,
      "loss": 0.9976,
      "num_input_tokens_seen": 24015816,
      "step": 41400
    },
    {
      "epoch": 6.1669645516830505,
      "grad_norm": 8.625,
      "learning_rate": 4.367519473432503e-05,
      "loss": 1.0451,
      "num_input_tokens_seen": 24018568,
      "step": 41405
    },
    {
      "epoch": 6.167709264224009,
      "grad_norm": 9.125,
      "learning_rate": 4.367303431153245e-05,
      "loss": 1.0781,
      "num_input_tokens_seen": 24021480,
      "step": 41410
    },
    {
      "epoch": 6.168453976764969,
      "grad_norm": 8.125,
      "learning_rate": 4.367087357327794e-05,
      "loss": 0.921,
      "num_input_tokens_seen": 24023944,
      "step": 41415
    },
    {
      "epoch": 6.169198689305928,
      "grad_norm": 24.375,
      "learning_rate": 4.366871251959799e-05,
      "loss": 0.9687,
      "num_input_tokens_seen": 24026920,
      "step": 41420
    },
    {
      "epoch": 6.169943401846887,
      "grad_norm": 6.59375,
      "learning_rate": 4.3666551150529124e-05,
      "loss": 0.8085,
      "num_input_tokens_seen": 24029992,
      "step": 41425
    },
    {
      "epoch": 6.170688114387846,
      "grad_norm": 21.375,
      "learning_rate": 4.366438946610784e-05,
      "loss": 1.1428,
      "num_input_tokens_seen": 24032840,
      "step": 41430
    },
    {
      "epoch": 6.171432826928806,
      "grad_norm": 6.65625,
      "learning_rate": 4.366222746637067e-05,
      "loss": 0.7874,
      "num_input_tokens_seen": 24035880,
      "step": 41435
    },
    {
      "epoch": 6.172177539469764,
      "grad_norm": 10.25,
      "learning_rate": 4.366006515135413e-05,
      "loss": 0.8698,
      "num_input_tokens_seen": 24039464,
      "step": 41440
    },
    {
      "epoch": 6.172922252010724,
      "grad_norm": 8.375,
      "learning_rate": 4.3657902521094764e-05,
      "loss": 0.8342,
      "num_input_tokens_seen": 24042536,
      "step": 41445
    },
    {
      "epoch": 6.173666964551683,
      "grad_norm": 19.125,
      "learning_rate": 4.365573957562909e-05,
      "loss": 1.0793,
      "num_input_tokens_seen": 24045512,
      "step": 41450
    },
    {
      "epoch": 6.1744116770926425,
      "grad_norm": 31.125,
      "learning_rate": 4.365357631499366e-05,
      "loss": 1.0491,
      "num_input_tokens_seen": 24048392,
      "step": 41455
    },
    {
      "epoch": 6.175156389633601,
      "grad_norm": 9.5,
      "learning_rate": 4.365141273922502e-05,
      "loss": 1.0541,
      "num_input_tokens_seen": 24051272,
      "step": 41460
    },
    {
      "epoch": 6.175901102174561,
      "grad_norm": 9.25,
      "learning_rate": 4.3649248848359706e-05,
      "loss": 0.8993,
      "num_input_tokens_seen": 24053928,
      "step": 41465
    },
    {
      "epoch": 6.17664581471552,
      "grad_norm": 6.46875,
      "learning_rate": 4.36470846424343e-05,
      "loss": 1.5974,
      "num_input_tokens_seen": 24056648,
      "step": 41470
    },
    {
      "epoch": 6.177390527256479,
      "grad_norm": 6.15625,
      "learning_rate": 4.364492012148534e-05,
      "loss": 0.9845,
      "num_input_tokens_seen": 24059464,
      "step": 41475
    },
    {
      "epoch": 6.178135239797438,
      "grad_norm": 5.9375,
      "learning_rate": 4.364275528554941e-05,
      "loss": 0.8999,
      "num_input_tokens_seen": 24062376,
      "step": 41480
    },
    {
      "epoch": 6.178879952338398,
      "grad_norm": 7.84375,
      "learning_rate": 4.3640590134663076e-05,
      "loss": 1.1343,
      "num_input_tokens_seen": 24065256,
      "step": 41485
    },
    {
      "epoch": 6.1796246648793565,
      "grad_norm": 8.6875,
      "learning_rate": 4.363842466886292e-05,
      "loss": 0.9851,
      "num_input_tokens_seen": 24067976,
      "step": 41490
    },
    {
      "epoch": 6.180369377420316,
      "grad_norm": 7.9375,
      "learning_rate": 4.363625888818552e-05,
      "loss": 1.0524,
      "num_input_tokens_seen": 24071144,
      "step": 41495
    },
    {
      "epoch": 6.181114089961275,
      "grad_norm": 13.5625,
      "learning_rate": 4.363409279266747e-05,
      "loss": 0.9848,
      "num_input_tokens_seen": 24074248,
      "step": 41500
    },
    {
      "epoch": 6.1818588025022345,
      "grad_norm": 10.6875,
      "learning_rate": 4.3631926382345356e-05,
      "loss": 0.8543,
      "num_input_tokens_seen": 24077160,
      "step": 41505
    },
    {
      "epoch": 6.182603515043193,
      "grad_norm": 6.59375,
      "learning_rate": 4.3629759657255786e-05,
      "loss": 1.0624,
      "num_input_tokens_seen": 24079880,
      "step": 41510
    },
    {
      "epoch": 6.183348227584153,
      "grad_norm": 19.5,
      "learning_rate": 4.3627592617435363e-05,
      "loss": 1.2518,
      "num_input_tokens_seen": 24082856,
      "step": 41515
    },
    {
      "epoch": 6.184092940125112,
      "grad_norm": 6.21875,
      "learning_rate": 4.362542526292069e-05,
      "loss": 0.8768,
      "num_input_tokens_seen": 24085704,
      "step": 41520
    },
    {
      "epoch": 6.18483765266607,
      "grad_norm": 4.1875,
      "learning_rate": 4.362325759374839e-05,
      "loss": 1.4181,
      "num_input_tokens_seen": 24088616,
      "step": 41525
    },
    {
      "epoch": 6.18558236520703,
      "grad_norm": 7.34375,
      "learning_rate": 4.3621089609955084e-05,
      "loss": 1.4382,
      "num_input_tokens_seen": 24091592,
      "step": 41530
    },
    {
      "epoch": 6.18632707774799,
      "grad_norm": 10.875,
      "learning_rate": 4.3618921311577384e-05,
      "loss": 0.9282,
      "num_input_tokens_seen": 24094504,
      "step": 41535
    },
    {
      "epoch": 6.1870717902889485,
      "grad_norm": 7.9375,
      "learning_rate": 4.361675269865194e-05,
      "loss": 1.1139,
      "num_input_tokens_seen": 24097448,
      "step": 41540
    },
    {
      "epoch": 6.187816502829907,
      "grad_norm": 9.3125,
      "learning_rate": 4.361458377121538e-05,
      "loss": 0.9505,
      "num_input_tokens_seen": 24100296,
      "step": 41545
    },
    {
      "epoch": 6.188561215370867,
      "grad_norm": 8.9375,
      "learning_rate": 4.3612414529304344e-05,
      "loss": 1.0986,
      "num_input_tokens_seen": 24103240,
      "step": 41550
    },
    {
      "epoch": 6.189305927911826,
      "grad_norm": 15.4375,
      "learning_rate": 4.3610244972955486e-05,
      "loss": 1.2924,
      "num_input_tokens_seen": 24106248,
      "step": 41555
    },
    {
      "epoch": 6.190050640452785,
      "grad_norm": 12.1875,
      "learning_rate": 4.3608075102205454e-05,
      "loss": 1.2286,
      "num_input_tokens_seen": 24109064,
      "step": 41560
    },
    {
      "epoch": 6.190795352993744,
      "grad_norm": 5.40625,
      "learning_rate": 4.36059049170909e-05,
      "loss": 0.9553,
      "num_input_tokens_seen": 24111560,
      "step": 41565
    },
    {
      "epoch": 6.191540065534704,
      "grad_norm": 11.5625,
      "learning_rate": 4.36037344176485e-05,
      "loss": 1.3978,
      "num_input_tokens_seen": 24114312,
      "step": 41570
    },
    {
      "epoch": 6.1922847780756625,
      "grad_norm": 9.3125,
      "learning_rate": 4.3601563603914906e-05,
      "loss": 1.0215,
      "num_input_tokens_seen": 24117096,
      "step": 41575
    },
    {
      "epoch": 6.193029490616622,
      "grad_norm": 13.875,
      "learning_rate": 4.3599392475926806e-05,
      "loss": 0.9154,
      "num_input_tokens_seen": 24120104,
      "step": 41580
    },
    {
      "epoch": 6.193774203157581,
      "grad_norm": 10.5,
      "learning_rate": 4.359722103372087e-05,
      "loss": 1.0253,
      "num_input_tokens_seen": 24122920,
      "step": 41585
    },
    {
      "epoch": 6.1945189156985405,
      "grad_norm": 24.875,
      "learning_rate": 4.3595049277333785e-05,
      "loss": 1.2805,
      "num_input_tokens_seen": 24125800,
      "step": 41590
    },
    {
      "epoch": 6.195263628239499,
      "grad_norm": 11.1875,
      "learning_rate": 4.359287720680225e-05,
      "loss": 0.9338,
      "num_input_tokens_seen": 24128488,
      "step": 41595
    },
    {
      "epoch": 6.196008340780459,
      "grad_norm": 8.75,
      "learning_rate": 4.359070482216295e-05,
      "loss": 1.121,
      "num_input_tokens_seen": 24131528,
      "step": 41600
    },
    {
      "epoch": 6.196753053321418,
      "grad_norm": 14.75,
      "learning_rate": 4.358853212345258e-05,
      "loss": 1.107,
      "num_input_tokens_seen": 24134344,
      "step": 41605
    },
    {
      "epoch": 6.197497765862377,
      "grad_norm": 5.5,
      "learning_rate": 4.358635911070785e-05,
      "loss": 0.7208,
      "num_input_tokens_seen": 24137256,
      "step": 41610
    },
    {
      "epoch": 6.198242478403336,
      "grad_norm": 18.75,
      "learning_rate": 4.3584185783965484e-05,
      "loss": 1.3025,
      "num_input_tokens_seen": 24140328,
      "step": 41615
    },
    {
      "epoch": 6.198987190944296,
      "grad_norm": 13.0,
      "learning_rate": 4.358201214326218e-05,
      "loss": 1.2466,
      "num_input_tokens_seen": 24143080,
      "step": 41620
    },
    {
      "epoch": 6.1997319034852545,
      "grad_norm": 18.5,
      "learning_rate": 4.357983818863467e-05,
      "loss": 1.5268,
      "num_input_tokens_seen": 24145992,
      "step": 41625
    },
    {
      "epoch": 6.200476616026214,
      "grad_norm": 6.28125,
      "learning_rate": 4.357766392011968e-05,
      "loss": 1.1996,
      "num_input_tokens_seen": 24149096,
      "step": 41630
    },
    {
      "epoch": 6.201221328567173,
      "grad_norm": 11.25,
      "learning_rate": 4.357548933775393e-05,
      "loss": 1.1284,
      "num_input_tokens_seen": 24152008,
      "step": 41635
    },
    {
      "epoch": 6.2019660411081325,
      "grad_norm": 4.8125,
      "learning_rate": 4.3573314441574176e-05,
      "loss": 0.7925,
      "num_input_tokens_seen": 24154920,
      "step": 41640
    },
    {
      "epoch": 6.202710753649091,
      "grad_norm": 9.5,
      "learning_rate": 4.357113923161715e-05,
      "loss": 1.1204,
      "num_input_tokens_seen": 24157832,
      "step": 41645
    },
    {
      "epoch": 6.203455466190051,
      "grad_norm": 12.6875,
      "learning_rate": 4.35689637079196e-05,
      "loss": 1.2351,
      "num_input_tokens_seen": 24160776,
      "step": 41650
    },
    {
      "epoch": 6.20420017873101,
      "grad_norm": 11.0625,
      "learning_rate": 4.356678787051828e-05,
      "loss": 1.5612,
      "num_input_tokens_seen": 24163976,
      "step": 41655
    },
    {
      "epoch": 6.204944891271969,
      "grad_norm": 9.9375,
      "learning_rate": 4.356461171944994e-05,
      "loss": 1.1582,
      "num_input_tokens_seen": 24167080,
      "step": 41660
    },
    {
      "epoch": 6.205689603812928,
      "grad_norm": 12.4375,
      "learning_rate": 4.356243525475137e-05,
      "loss": 1.0928,
      "num_input_tokens_seen": 24169704,
      "step": 41665
    },
    {
      "epoch": 6.206434316353888,
      "grad_norm": 7.40625,
      "learning_rate": 4.3560258476459315e-05,
      "loss": 1.0724,
      "num_input_tokens_seen": 24172712,
      "step": 41670
    },
    {
      "epoch": 6.2071790288948465,
      "grad_norm": 12.375,
      "learning_rate": 4.355808138461056e-05,
      "loss": 1.2138,
      "num_input_tokens_seen": 24175464,
      "step": 41675
    },
    {
      "epoch": 6.207923741435806,
      "grad_norm": 7.84375,
      "learning_rate": 4.355590397924188e-05,
      "loss": 1.2964,
      "num_input_tokens_seen": 24178408,
      "step": 41680
    },
    {
      "epoch": 6.208668453976765,
      "grad_norm": 6.71875,
      "learning_rate": 4.355372626039006e-05,
      "loss": 1.0193,
      "num_input_tokens_seen": 24181160,
      "step": 41685
    },
    {
      "epoch": 6.209413166517725,
      "grad_norm": 15.5,
      "learning_rate": 4.355154822809189e-05,
      "loss": 1.718,
      "num_input_tokens_seen": 24184200,
      "step": 41690
    },
    {
      "epoch": 6.210157879058683,
      "grad_norm": 10.9375,
      "learning_rate": 4.3549369882384174e-05,
      "loss": 1.2537,
      "num_input_tokens_seen": 24186952,
      "step": 41695
    },
    {
      "epoch": 6.210902591599643,
      "grad_norm": 10.1875,
      "learning_rate": 4.35471912233037e-05,
      "loss": 0.9212,
      "num_input_tokens_seen": 24190024,
      "step": 41700
    },
    {
      "epoch": 6.211647304140602,
      "grad_norm": 10.8125,
      "learning_rate": 4.3545012250887286e-05,
      "loss": 0.8759,
      "num_input_tokens_seen": 24193480,
      "step": 41705
    },
    {
      "epoch": 6.2123920166815605,
      "grad_norm": 10.9375,
      "learning_rate": 4.354283296517173e-05,
      "loss": 1.4262,
      "num_input_tokens_seen": 24196200,
      "step": 41710
    },
    {
      "epoch": 6.21313672922252,
      "grad_norm": 9.875,
      "learning_rate": 4.354065336619387e-05,
      "loss": 0.6989,
      "num_input_tokens_seen": 24199432,
      "step": 41715
    },
    {
      "epoch": 6.213881441763479,
      "grad_norm": 11.6875,
      "learning_rate": 4.3538473453990506e-05,
      "loss": 0.8952,
      "num_input_tokens_seen": 24202056,
      "step": 41720
    },
    {
      "epoch": 6.2146261543044385,
      "grad_norm": 10.9375,
      "learning_rate": 4.353629322859848e-05,
      "loss": 1.1197,
      "num_input_tokens_seen": 24205128,
      "step": 41725
    },
    {
      "epoch": 6.215370866845397,
      "grad_norm": 9.125,
      "learning_rate": 4.353411269005462e-05,
      "loss": 1.3789,
      "num_input_tokens_seen": 24208072,
      "step": 41730
    },
    {
      "epoch": 6.216115579386357,
      "grad_norm": 4.9375,
      "learning_rate": 4.353193183839576e-05,
      "loss": 0.7853,
      "num_input_tokens_seen": 24211016,
      "step": 41735
    },
    {
      "epoch": 6.216860291927316,
      "grad_norm": 5.375,
      "learning_rate": 4.352975067365874e-05,
      "loss": 1.0763,
      "num_input_tokens_seen": 24213800,
      "step": 41740
    },
    {
      "epoch": 6.217605004468275,
      "grad_norm": 7.25,
      "learning_rate": 4.352756919588042e-05,
      "loss": 0.7834,
      "num_input_tokens_seen": 24216808,
      "step": 41745
    },
    {
      "epoch": 6.218349717009234,
      "grad_norm": 15.125,
      "learning_rate": 4.3525387405097654e-05,
      "loss": 1.0544,
      "num_input_tokens_seen": 24219624,
      "step": 41750
    },
    {
      "epoch": 6.219094429550194,
      "grad_norm": 10.3125,
      "learning_rate": 4.352320530134729e-05,
      "loss": 0.5318,
      "num_input_tokens_seen": 24222632,
      "step": 41755
    },
    {
      "epoch": 6.2198391420911525,
      "grad_norm": 6.625,
      "learning_rate": 4.35210228846662e-05,
      "loss": 0.8331,
      "num_input_tokens_seen": 24225192,
      "step": 41760
    },
    {
      "epoch": 6.220583854632112,
      "grad_norm": 11.6875,
      "learning_rate": 4.3518840155091255e-05,
      "loss": 1.1961,
      "num_input_tokens_seen": 24228008,
      "step": 41765
    },
    {
      "epoch": 6.221328567173071,
      "grad_norm": 9.3125,
      "learning_rate": 4.351665711265933e-05,
      "loss": 1.0035,
      "num_input_tokens_seen": 24231176,
      "step": 41770
    },
    {
      "epoch": 6.222073279714031,
      "grad_norm": 13.8125,
      "learning_rate": 4.351447375740729e-05,
      "loss": 1.3479,
      "num_input_tokens_seen": 24233832,
      "step": 41775
    },
    {
      "epoch": 6.222817992254989,
      "grad_norm": 15.5625,
      "learning_rate": 4.351229008937205e-05,
      "loss": 1.0163,
      "num_input_tokens_seen": 24236456,
      "step": 41780
    },
    {
      "epoch": 6.223562704795949,
      "grad_norm": 8.5,
      "learning_rate": 4.3510106108590476e-05,
      "loss": 1.0288,
      "num_input_tokens_seen": 24239208,
      "step": 41785
    },
    {
      "epoch": 6.224307417336908,
      "grad_norm": 7.75,
      "learning_rate": 4.350792181509947e-05,
      "loss": 0.6907,
      "num_input_tokens_seen": 24241960,
      "step": 41790
    },
    {
      "epoch": 6.225052129877867,
      "grad_norm": 16.625,
      "learning_rate": 4.350573720893594e-05,
      "loss": 1.2719,
      "num_input_tokens_seen": 24244840,
      "step": 41795
    },
    {
      "epoch": 6.225796842418826,
      "grad_norm": 20.75,
      "learning_rate": 4.350355229013679e-05,
      "loss": 1.1052,
      "num_input_tokens_seen": 24247848,
      "step": 41800
    },
    {
      "epoch": 6.226541554959786,
      "grad_norm": 9.125,
      "learning_rate": 4.3501367058738916e-05,
      "loss": 0.9483,
      "num_input_tokens_seen": 24250856,
      "step": 41805
    },
    {
      "epoch": 6.2272862675007445,
      "grad_norm": 16.625,
      "learning_rate": 4.3499181514779266e-05,
      "loss": 1.2078,
      "num_input_tokens_seen": 24253928,
      "step": 41810
    },
    {
      "epoch": 6.228030980041704,
      "grad_norm": 10.625,
      "learning_rate": 4.3496995658294735e-05,
      "loss": 1.0297,
      "num_input_tokens_seen": 24256744,
      "step": 41815
    },
    {
      "epoch": 6.228775692582663,
      "grad_norm": 18.625,
      "learning_rate": 4.349480948932226e-05,
      "loss": 1.1419,
      "num_input_tokens_seen": 24259976,
      "step": 41820
    },
    {
      "epoch": 6.229520405123623,
      "grad_norm": 7.65625,
      "learning_rate": 4.3492623007898786e-05,
      "loss": 1.0961,
      "num_input_tokens_seen": 24262728,
      "step": 41825
    },
    {
      "epoch": 6.230265117664581,
      "grad_norm": 6.625,
      "learning_rate": 4.3490436214061236e-05,
      "loss": 1.0498,
      "num_input_tokens_seen": 24265640,
      "step": 41830
    },
    {
      "epoch": 6.231009830205541,
      "grad_norm": 8.8125,
      "learning_rate": 4.348824910784656e-05,
      "loss": 1.2791,
      "num_input_tokens_seen": 24268456,
      "step": 41835
    },
    {
      "epoch": 6.2317545427465,
      "grad_norm": 16.75,
      "learning_rate": 4.348606168929171e-05,
      "loss": 1.3217,
      "num_input_tokens_seen": 24271464,
      "step": 41840
    },
    {
      "epoch": 6.232499255287459,
      "grad_norm": 7.78125,
      "learning_rate": 4.348387395843363e-05,
      "loss": 1.3668,
      "num_input_tokens_seen": 24274280,
      "step": 41845
    },
    {
      "epoch": 6.233243967828418,
      "grad_norm": 6.8125,
      "learning_rate": 4.348168591530929e-05,
      "loss": 0.9582,
      "num_input_tokens_seen": 24277384,
      "step": 41850
    },
    {
      "epoch": 6.233988680369378,
      "grad_norm": 6.4375,
      "learning_rate": 4.3479497559955654e-05,
      "loss": 0.6245,
      "num_input_tokens_seen": 24280392,
      "step": 41855
    },
    {
      "epoch": 6.234733392910337,
      "grad_norm": 8.0625,
      "learning_rate": 4.347730889240968e-05,
      "loss": 1.0552,
      "num_input_tokens_seen": 24283144,
      "step": 41860
    },
    {
      "epoch": 6.235478105451296,
      "grad_norm": 11.0,
      "learning_rate": 4.347511991270835e-05,
      "loss": 1.0632,
      "num_input_tokens_seen": 24285928,
      "step": 41865
    },
    {
      "epoch": 6.236222817992255,
      "grad_norm": 11.875,
      "learning_rate": 4.347293062088865e-05,
      "loss": 1.0345,
      "num_input_tokens_seen": 24288712,
      "step": 41870
    },
    {
      "epoch": 6.236967530533214,
      "grad_norm": 18.75,
      "learning_rate": 4.3470741016987574e-05,
      "loss": 1.1927,
      "num_input_tokens_seen": 24291368,
      "step": 41875
    },
    {
      "epoch": 6.237712243074173,
      "grad_norm": 5.9375,
      "learning_rate": 4.3468551101042084e-05,
      "loss": 1.2041,
      "num_input_tokens_seen": 24294440,
      "step": 41880
    },
    {
      "epoch": 6.238456955615132,
      "grad_norm": 17.375,
      "learning_rate": 4.3466360873089204e-05,
      "loss": 1.3182,
      "num_input_tokens_seen": 24297096,
      "step": 41885
    },
    {
      "epoch": 6.239201668156092,
      "grad_norm": 10.25,
      "learning_rate": 4.346417033316592e-05,
      "loss": 0.9923,
      "num_input_tokens_seen": 24300136,
      "step": 41890
    },
    {
      "epoch": 6.2399463806970505,
      "grad_norm": 14.0,
      "learning_rate": 4.346197948130925e-05,
      "loss": 1.1322,
      "num_input_tokens_seen": 24302856,
      "step": 41895
    },
    {
      "epoch": 6.24069109323801,
      "grad_norm": 19.125,
      "learning_rate": 4.34597883175562e-05,
      "loss": 0.8956,
      "num_input_tokens_seen": 24305832,
      "step": 41900
    },
    {
      "epoch": 6.241435805778969,
      "grad_norm": 9.375,
      "learning_rate": 4.3457596841943775e-05,
      "loss": 0.9149,
      "num_input_tokens_seen": 24308520,
      "step": 41905
    },
    {
      "epoch": 6.242180518319929,
      "grad_norm": 10.0625,
      "learning_rate": 4.345540505450902e-05,
      "loss": 1.0614,
      "num_input_tokens_seen": 24311624,
      "step": 41910
    },
    {
      "epoch": 6.242925230860887,
      "grad_norm": 16.375,
      "learning_rate": 4.345321295528896e-05,
      "loss": 1.7056,
      "num_input_tokens_seen": 24314472,
      "step": 41915
    },
    {
      "epoch": 6.243669943401847,
      "grad_norm": 11.9375,
      "learning_rate": 4.345102054432061e-05,
      "loss": 1.1147,
      "num_input_tokens_seen": 24317256,
      "step": 41920
    },
    {
      "epoch": 6.244414655942806,
      "grad_norm": 10.75,
      "learning_rate": 4.344882782164103e-05,
      "loss": 0.826,
      "num_input_tokens_seen": 24320040,
      "step": 41925
    },
    {
      "epoch": 6.245159368483765,
      "grad_norm": 8.125,
      "learning_rate": 4.344663478728725e-05,
      "loss": 1.5981,
      "num_input_tokens_seen": 24323144,
      "step": 41930
    },
    {
      "epoch": 6.245904081024724,
      "grad_norm": 7.84375,
      "learning_rate": 4.3444441441296324e-05,
      "loss": 1.1601,
      "num_input_tokens_seen": 24325768,
      "step": 41935
    },
    {
      "epoch": 6.246648793565684,
      "grad_norm": 12.125,
      "learning_rate": 4.344224778370531e-05,
      "loss": 1.0964,
      "num_input_tokens_seen": 24328808,
      "step": 41940
    },
    {
      "epoch": 6.247393506106643,
      "grad_norm": 6.1875,
      "learning_rate": 4.344005381455126e-05,
      "loss": 0.8557,
      "num_input_tokens_seen": 24331720,
      "step": 41945
    },
    {
      "epoch": 6.248138218647602,
      "grad_norm": 10.0,
      "learning_rate": 4.343785953387125e-05,
      "loss": 1.2598,
      "num_input_tokens_seen": 24334664,
      "step": 41950
    },
    {
      "epoch": 6.248882931188561,
      "grad_norm": 9.5,
      "learning_rate": 4.343566494170233e-05,
      "loss": 1.1395,
      "num_input_tokens_seen": 24337576,
      "step": 41955
    },
    {
      "epoch": 6.249627643729521,
      "grad_norm": 9.3125,
      "learning_rate": 4.34334700380816e-05,
      "loss": 0.8884,
      "num_input_tokens_seen": 24340360,
      "step": 41960
    },
    {
      "epoch": 6.250372356270479,
      "grad_norm": 12.4375,
      "learning_rate": 4.343127482304612e-05,
      "loss": 0.9716,
      "num_input_tokens_seen": 24343112,
      "step": 41965
    },
    {
      "epoch": 6.251117068811439,
      "grad_norm": 12.1875,
      "learning_rate": 4.342907929663299e-05,
      "loss": 0.8119,
      "num_input_tokens_seen": 24345832,
      "step": 41970
    },
    {
      "epoch": 6.251861781352398,
      "grad_norm": 8.0,
      "learning_rate": 4.342688345887929e-05,
      "loss": 1.082,
      "num_input_tokens_seen": 24348904,
      "step": 41975
    },
    {
      "epoch": 6.252606493893357,
      "grad_norm": 13.9375,
      "learning_rate": 4.342468730982212e-05,
      "loss": 1.168,
      "num_input_tokens_seen": 24351752,
      "step": 41980
    },
    {
      "epoch": 6.253351206434316,
      "grad_norm": 14.6875,
      "learning_rate": 4.342249084949859e-05,
      "loss": 1.0988,
      "num_input_tokens_seen": 24354472,
      "step": 41985
    },
    {
      "epoch": 6.254095918975276,
      "grad_norm": 12.5,
      "learning_rate": 4.34202940779458e-05,
      "loss": 1.1792,
      "num_input_tokens_seen": 24357224,
      "step": 41990
    },
    {
      "epoch": 6.254840631516235,
      "grad_norm": 11.75,
      "learning_rate": 4.341809699520086e-05,
      "loss": 1.1438,
      "num_input_tokens_seen": 24359912,
      "step": 41995
    },
    {
      "epoch": 6.255585344057194,
      "grad_norm": 15.125,
      "learning_rate": 4.34158996013009e-05,
      "loss": 1.3258,
      "num_input_tokens_seen": 24362728,
      "step": 42000
    },
    {
      "epoch": 6.256330056598153,
      "grad_norm": 12.8125,
      "learning_rate": 4.3413701896283024e-05,
      "loss": 1.2732,
      "num_input_tokens_seen": 24365736,
      "step": 42005
    },
    {
      "epoch": 6.257074769139113,
      "grad_norm": 5.21875,
      "learning_rate": 4.341150388018437e-05,
      "loss": 0.728,
      "num_input_tokens_seen": 24368488,
      "step": 42010
    },
    {
      "epoch": 6.257819481680071,
      "grad_norm": 8.3125,
      "learning_rate": 4.340930555304208e-05,
      "loss": 1.1568,
      "num_input_tokens_seen": 24371240,
      "step": 42015
    },
    {
      "epoch": 6.258564194221031,
      "grad_norm": 15.6875,
      "learning_rate": 4.340710691489327e-05,
      "loss": 1.0376,
      "num_input_tokens_seen": 24374216,
      "step": 42020
    },
    {
      "epoch": 6.25930890676199,
      "grad_norm": 6.78125,
      "learning_rate": 4.340490796577511e-05,
      "loss": 1.1888,
      "num_input_tokens_seen": 24377128,
      "step": 42025
    },
    {
      "epoch": 6.2600536193029495,
      "grad_norm": 9.9375,
      "learning_rate": 4.340270870572472e-05,
      "loss": 0.8595,
      "num_input_tokens_seen": 24380360,
      "step": 42030
    },
    {
      "epoch": 6.260798331843908,
      "grad_norm": 10.5625,
      "learning_rate": 4.340050913477928e-05,
      "loss": 0.8715,
      "num_input_tokens_seen": 24383368,
      "step": 42035
    },
    {
      "epoch": 6.261543044384867,
      "grad_norm": 10.125,
      "learning_rate": 4.339830925297594e-05,
      "loss": 1.1579,
      "num_input_tokens_seen": 24386472,
      "step": 42040
    },
    {
      "epoch": 6.262287756925827,
      "grad_norm": 19.125,
      "learning_rate": 4.3396109060351864e-05,
      "loss": 0.8475,
      "num_input_tokens_seen": 24389416,
      "step": 42045
    },
    {
      "epoch": 6.263032469466786,
      "grad_norm": 11.5,
      "learning_rate": 4.339390855694422e-05,
      "loss": 1.099,
      "num_input_tokens_seen": 24392392,
      "step": 42050
    },
    {
      "epoch": 6.263777182007745,
      "grad_norm": 12.9375,
      "learning_rate": 4.339170774279019e-05,
      "loss": 0.9751,
      "num_input_tokens_seen": 24395112,
      "step": 42055
    },
    {
      "epoch": 6.264521894548704,
      "grad_norm": 5.84375,
      "learning_rate": 4.3389506617926945e-05,
      "loss": 0.8481,
      "num_input_tokens_seen": 24397896,
      "step": 42060
    },
    {
      "epoch": 6.265266607089663,
      "grad_norm": 10.4375,
      "learning_rate": 4.3387305182391677e-05,
      "loss": 0.9585,
      "num_input_tokens_seen": 24400936,
      "step": 42065
    },
    {
      "epoch": 6.266011319630622,
      "grad_norm": 6.59375,
      "learning_rate": 4.3385103436221575e-05,
      "loss": 1.2304,
      "num_input_tokens_seen": 24403784,
      "step": 42070
    },
    {
      "epoch": 6.266756032171582,
      "grad_norm": 10.6875,
      "learning_rate": 4.338290137945384e-05,
      "loss": 1.2556,
      "num_input_tokens_seen": 24406472,
      "step": 42075
    },
    {
      "epoch": 6.267500744712541,
      "grad_norm": 17.25,
      "learning_rate": 4.338069901212567e-05,
      "loss": 1.2347,
      "num_input_tokens_seen": 24409128,
      "step": 42080
    },
    {
      "epoch": 6.2682454572535,
      "grad_norm": 14.6875,
      "learning_rate": 4.337849633427427e-05,
      "loss": 1.258,
      "num_input_tokens_seen": 24411880,
      "step": 42085
    },
    {
      "epoch": 6.268990169794459,
      "grad_norm": 12.25,
      "learning_rate": 4.337629334593685e-05,
      "loss": 1.1817,
      "num_input_tokens_seen": 24414824,
      "step": 42090
    },
    {
      "epoch": 6.269734882335419,
      "grad_norm": 8.1875,
      "learning_rate": 4.337409004715063e-05,
      "loss": 1.103,
      "num_input_tokens_seen": 24417576,
      "step": 42095
    },
    {
      "epoch": 6.270479594876377,
      "grad_norm": 8.3125,
      "learning_rate": 4.337188643795284e-05,
      "loss": 0.9529,
      "num_input_tokens_seen": 24420296,
      "step": 42100
    },
    {
      "epoch": 6.271224307417337,
      "grad_norm": 15.75,
      "learning_rate": 4.33696825183807e-05,
      "loss": 0.9577,
      "num_input_tokens_seen": 24423112,
      "step": 42105
    },
    {
      "epoch": 6.271969019958296,
      "grad_norm": 13.0,
      "learning_rate": 4.3367478288471444e-05,
      "loss": 1.2562,
      "num_input_tokens_seen": 24425800,
      "step": 42110
    },
    {
      "epoch": 6.2727137324992555,
      "grad_norm": 4.90625,
      "learning_rate": 4.33652737482623e-05,
      "loss": 1.139,
      "num_input_tokens_seen": 24428808,
      "step": 42115
    },
    {
      "epoch": 6.273458445040214,
      "grad_norm": 12.625,
      "learning_rate": 4.336306889779054e-05,
      "loss": 1.3252,
      "num_input_tokens_seen": 24432008,
      "step": 42120
    },
    {
      "epoch": 6.274203157581174,
      "grad_norm": 19.5,
      "learning_rate": 4.3360863737093375e-05,
      "loss": 1.1768,
      "num_input_tokens_seen": 24434760,
      "step": 42125
    },
    {
      "epoch": 6.274947870122133,
      "grad_norm": 9.5,
      "learning_rate": 4.335865826620809e-05,
      "loss": 0.992,
      "num_input_tokens_seen": 24437672,
      "step": 42130
    },
    {
      "epoch": 6.275692582663092,
      "grad_norm": 9.3125,
      "learning_rate": 4.335645248517193e-05,
      "loss": 0.9605,
      "num_input_tokens_seen": 24440712,
      "step": 42135
    },
    {
      "epoch": 6.276437295204051,
      "grad_norm": 10.4375,
      "learning_rate": 4.335424639402216e-05,
      "loss": 1.1103,
      "num_input_tokens_seen": 24444008,
      "step": 42140
    },
    {
      "epoch": 6.277182007745011,
      "grad_norm": 17.75,
      "learning_rate": 4.3352039992796056e-05,
      "loss": 0.9646,
      "num_input_tokens_seen": 24446696,
      "step": 42145
    },
    {
      "epoch": 6.277926720285969,
      "grad_norm": 10.1875,
      "learning_rate": 4.334983328153088e-05,
      "loss": 1.5028,
      "num_input_tokens_seen": 24449864,
      "step": 42150
    },
    {
      "epoch": 6.278671432826929,
      "grad_norm": 8.25,
      "learning_rate": 4.334762626026393e-05,
      "loss": 0.8038,
      "num_input_tokens_seen": 24452520,
      "step": 42155
    },
    {
      "epoch": 6.279416145367888,
      "grad_norm": 13.6875,
      "learning_rate": 4.3345418929032475e-05,
      "loss": 1.2263,
      "num_input_tokens_seen": 24455528,
      "step": 42160
    },
    {
      "epoch": 6.2801608579088475,
      "grad_norm": 8.9375,
      "learning_rate": 4.334321128787382e-05,
      "loss": 1.2789,
      "num_input_tokens_seen": 24458536,
      "step": 42165
    },
    {
      "epoch": 6.280905570449806,
      "grad_norm": 8.6875,
      "learning_rate": 4.3341003336825246e-05,
      "loss": 1.0822,
      "num_input_tokens_seen": 24461512,
      "step": 42170
    },
    {
      "epoch": 6.281650282990766,
      "grad_norm": 10.3125,
      "learning_rate": 4.333879507592407e-05,
      "loss": 0.9853,
      "num_input_tokens_seen": 24464232,
      "step": 42175
    },
    {
      "epoch": 6.282394995531725,
      "grad_norm": 6.5,
      "learning_rate": 4.3336586505207587e-05,
      "loss": 1.0882,
      "num_input_tokens_seen": 24467208,
      "step": 42180
    },
    {
      "epoch": 6.283139708072684,
      "grad_norm": 12.875,
      "learning_rate": 4.3334377624713104e-05,
      "loss": 1.0805,
      "num_input_tokens_seen": 24470376,
      "step": 42185
    },
    {
      "epoch": 6.283884420613643,
      "grad_norm": 4.6875,
      "learning_rate": 4.333216843447795e-05,
      "loss": 1.1184,
      "num_input_tokens_seen": 24473256,
      "step": 42190
    },
    {
      "epoch": 6.284629133154603,
      "grad_norm": 9.0,
      "learning_rate": 4.332995893453945e-05,
      "loss": 0.6612,
      "num_input_tokens_seen": 24475944,
      "step": 42195
    },
    {
      "epoch": 6.2853738456955615,
      "grad_norm": 8.4375,
      "learning_rate": 4.3327749124934916e-05,
      "loss": 1.1756,
      "num_input_tokens_seen": 24478696,
      "step": 42200
    },
    {
      "epoch": 6.286118558236521,
      "grad_norm": 14.875,
      "learning_rate": 4.332553900570169e-05,
      "loss": 0.9529,
      "num_input_tokens_seen": 24481576,
      "step": 42205
    },
    {
      "epoch": 6.28686327077748,
      "grad_norm": 3.6875,
      "learning_rate": 4.3323328576877104e-05,
      "loss": 0.9615,
      "num_input_tokens_seen": 24484680,
      "step": 42210
    },
    {
      "epoch": 6.2876079833184395,
      "grad_norm": 7.3125,
      "learning_rate": 4.33211178384985e-05,
      "loss": 1.2924,
      "num_input_tokens_seen": 24488104,
      "step": 42215
    },
    {
      "epoch": 6.288352695859398,
      "grad_norm": 8.5,
      "learning_rate": 4.331890679060324e-05,
      "loss": 0.9808,
      "num_input_tokens_seen": 24491080,
      "step": 42220
    },
    {
      "epoch": 6.289097408400357,
      "grad_norm": 21.0,
      "learning_rate": 4.331669543322867e-05,
      "loss": 1.3112,
      "num_input_tokens_seen": 24493768,
      "step": 42225
    },
    {
      "epoch": 6.289842120941317,
      "grad_norm": 7.46875,
      "learning_rate": 4.331448376641214e-05,
      "loss": 1.0515,
      "num_input_tokens_seen": 24496424,
      "step": 42230
    },
    {
      "epoch": 6.290586833482275,
      "grad_norm": 12.4375,
      "learning_rate": 4.331227179019103e-05,
      "loss": 0.936,
      "num_input_tokens_seen": 24499240,
      "step": 42235
    },
    {
      "epoch": 6.291331546023235,
      "grad_norm": 7.8125,
      "learning_rate": 4.3310059504602685e-05,
      "loss": 0.9388,
      "num_input_tokens_seen": 24502312,
      "step": 42240
    },
    {
      "epoch": 6.292076258564194,
      "grad_norm": 12.125,
      "learning_rate": 4.330784690968451e-05,
      "loss": 0.9321,
      "num_input_tokens_seen": 24505448,
      "step": 42245
    },
    {
      "epoch": 6.2928209711051535,
      "grad_norm": 12.0,
      "learning_rate": 4.330563400547386e-05,
      "loss": 0.8334,
      "num_input_tokens_seen": 24508360,
      "step": 42250
    },
    {
      "epoch": 6.293565683646112,
      "grad_norm": 14.75,
      "learning_rate": 4.330342079200813e-05,
      "loss": 0.9697,
      "num_input_tokens_seen": 24511304,
      "step": 42255
    },
    {
      "epoch": 6.294310396187072,
      "grad_norm": 7.28125,
      "learning_rate": 4.330120726932471e-05,
      "loss": 1.0407,
      "num_input_tokens_seen": 24514280,
      "step": 42260
    },
    {
      "epoch": 6.295055108728031,
      "grad_norm": 9.5,
      "learning_rate": 4.329899343746099e-05,
      "loss": 0.9094,
      "num_input_tokens_seen": 24517224,
      "step": 42265
    },
    {
      "epoch": 6.29579982126899,
      "grad_norm": 11.9375,
      "learning_rate": 4.3296779296454374e-05,
      "loss": 1.0026,
      "num_input_tokens_seen": 24520104,
      "step": 42270
    },
    {
      "epoch": 6.296544533809949,
      "grad_norm": 9.9375,
      "learning_rate": 4.3294564846342275e-05,
      "loss": 1.0439,
      "num_input_tokens_seen": 24523048,
      "step": 42275
    },
    {
      "epoch": 6.297289246350909,
      "grad_norm": 9.9375,
      "learning_rate": 4.329235008716209e-05,
      "loss": 1.4037,
      "num_input_tokens_seen": 24525896,
      "step": 42280
    },
    {
      "epoch": 6.2980339588918675,
      "grad_norm": 12.75,
      "learning_rate": 4.329013501895125e-05,
      "loss": 1.465,
      "num_input_tokens_seen": 24528648,
      "step": 42285
    },
    {
      "epoch": 6.298778671432827,
      "grad_norm": 6.71875,
      "learning_rate": 4.3287919641747155e-05,
      "loss": 1.0301,
      "num_input_tokens_seen": 24531592,
      "step": 42290
    },
    {
      "epoch": 6.299523383973786,
      "grad_norm": 6.6875,
      "learning_rate": 4.328570395558725e-05,
      "loss": 1.3359,
      "num_input_tokens_seen": 24534184,
      "step": 42295
    },
    {
      "epoch": 6.3002680965147455,
      "grad_norm": 14.9375,
      "learning_rate": 4.328348796050896e-05,
      "loss": 1.2093,
      "num_input_tokens_seen": 24537224,
      "step": 42300
    },
    {
      "epoch": 6.301012809055704,
      "grad_norm": 3.71875,
      "learning_rate": 4.3281271656549734e-05,
      "loss": 0.8321,
      "num_input_tokens_seen": 24540136,
      "step": 42305
    },
    {
      "epoch": 6.301757521596664,
      "grad_norm": 4.21875,
      "learning_rate": 4.3279055043746996e-05,
      "loss": 0.8752,
      "num_input_tokens_seen": 24542952,
      "step": 42310
    },
    {
      "epoch": 6.302502234137623,
      "grad_norm": 10.875,
      "learning_rate": 4.3276838122138196e-05,
      "loss": 1.1666,
      "num_input_tokens_seen": 24545736,
      "step": 42315
    },
    {
      "epoch": 6.303246946678582,
      "grad_norm": 6.09375,
      "learning_rate": 4.3274620891760795e-05,
      "loss": 0.9764,
      "num_input_tokens_seen": 24548584,
      "step": 42320
    },
    {
      "epoch": 6.303991659219541,
      "grad_norm": 8.4375,
      "learning_rate": 4.327240335265226e-05,
      "loss": 0.8753,
      "num_input_tokens_seen": 24551496,
      "step": 42325
    },
    {
      "epoch": 6.304736371760501,
      "grad_norm": 15.5625,
      "learning_rate": 4.3270185504850024e-05,
      "loss": 1.2039,
      "num_input_tokens_seen": 24554248,
      "step": 42330
    },
    {
      "epoch": 6.3054810843014595,
      "grad_norm": 15.125,
      "learning_rate": 4.326796734839158e-05,
      "loss": 1.0484,
      "num_input_tokens_seen": 24557096,
      "step": 42335
    },
    {
      "epoch": 6.306225796842419,
      "grad_norm": 8.125,
      "learning_rate": 4.32657488833144e-05,
      "loss": 0.8783,
      "num_input_tokens_seen": 24559784,
      "step": 42340
    },
    {
      "epoch": 6.306970509383378,
      "grad_norm": 10.1875,
      "learning_rate": 4.326353010965595e-05,
      "loss": 0.8402,
      "num_input_tokens_seen": 24562728,
      "step": 42345
    },
    {
      "epoch": 6.3077152219243375,
      "grad_norm": 13.375,
      "learning_rate": 4.326131102745372e-05,
      "loss": 0.9879,
      "num_input_tokens_seen": 24565704,
      "step": 42350
    },
    {
      "epoch": 6.308459934465296,
      "grad_norm": 7.53125,
      "learning_rate": 4.3259091636745196e-05,
      "loss": 0.9989,
      "num_input_tokens_seen": 24568584,
      "step": 42355
    },
    {
      "epoch": 6.309204647006256,
      "grad_norm": 17.0,
      "learning_rate": 4.325687193756789e-05,
      "loss": 1.3656,
      "num_input_tokens_seen": 24571272,
      "step": 42360
    },
    {
      "epoch": 6.309949359547215,
      "grad_norm": 16.25,
      "learning_rate": 4.325465192995928e-05,
      "loss": 1.0969,
      "num_input_tokens_seen": 24573896,
      "step": 42365
    },
    {
      "epoch": 6.310694072088174,
      "grad_norm": 9.125,
      "learning_rate": 4.325243161395688e-05,
      "loss": 0.8528,
      "num_input_tokens_seen": 24576680,
      "step": 42370
    },
    {
      "epoch": 6.311438784629133,
      "grad_norm": 4.0625,
      "learning_rate": 4.3250210989598196e-05,
      "loss": 1.0023,
      "num_input_tokens_seen": 24579464,
      "step": 42375
    },
    {
      "epoch": 6.312183497170093,
      "grad_norm": 6.1875,
      "learning_rate": 4.324799005692075e-05,
      "loss": 0.9893,
      "num_input_tokens_seen": 24582184,
      "step": 42380
    },
    {
      "epoch": 6.3129282097110515,
      "grad_norm": 13.0,
      "learning_rate": 4.3245768815962055e-05,
      "loss": 1.5133,
      "num_input_tokens_seen": 24585416,
      "step": 42385
    },
    {
      "epoch": 6.31367292225201,
      "grad_norm": 9.125,
      "learning_rate": 4.3243547266759646e-05,
      "loss": 1.1772,
      "num_input_tokens_seen": 24588360,
      "step": 42390
    },
    {
      "epoch": 6.31441763479297,
      "grad_norm": 13.8125,
      "learning_rate": 4.3241325409351044e-05,
      "loss": 1.1823,
      "num_input_tokens_seen": 24591336,
      "step": 42395
    },
    {
      "epoch": 6.31516234733393,
      "grad_norm": 9.4375,
      "learning_rate": 4.323910324377379e-05,
      "loss": 1.0168,
      "num_input_tokens_seen": 24594152,
      "step": 42400
    },
    {
      "epoch": 6.315907059874888,
      "grad_norm": 11.3125,
      "learning_rate": 4.3236880770065426e-05,
      "loss": 0.9592,
      "num_input_tokens_seen": 24597224,
      "step": 42405
    },
    {
      "epoch": 6.316651772415847,
      "grad_norm": 14.25,
      "learning_rate": 4.323465798826349e-05,
      "loss": 0.9871,
      "num_input_tokens_seen": 24600072,
      "step": 42410
    },
    {
      "epoch": 6.317396484956807,
      "grad_norm": 10.75,
      "learning_rate": 4.323243489840554e-05,
      "loss": 1.0456,
      "num_input_tokens_seen": 24603080,
      "step": 42415
    },
    {
      "epoch": 6.3181411974977655,
      "grad_norm": 8.125,
      "learning_rate": 4.323021150052914e-05,
      "loss": 1.072,
      "num_input_tokens_seen": 24605928,
      "step": 42420
    },
    {
      "epoch": 6.318885910038725,
      "grad_norm": 13.25,
      "learning_rate": 4.322798779467184e-05,
      "loss": 1.2395,
      "num_input_tokens_seen": 24608936,
      "step": 42425
    },
    {
      "epoch": 6.319630622579684,
      "grad_norm": 6.1875,
      "learning_rate": 4.322576378087121e-05,
      "loss": 0.9667,
      "num_input_tokens_seen": 24611752,
      "step": 42430
    },
    {
      "epoch": 6.3203753351206435,
      "grad_norm": 6.0625,
      "learning_rate": 4.322353945916483e-05,
      "loss": 0.9597,
      "num_input_tokens_seen": 24614280,
      "step": 42435
    },
    {
      "epoch": 6.321120047661602,
      "grad_norm": 16.5,
      "learning_rate": 4.322131482959027e-05,
      "loss": 1.1677,
      "num_input_tokens_seen": 24617544,
      "step": 42440
    },
    {
      "epoch": 6.321864760202562,
      "grad_norm": 14.375,
      "learning_rate": 4.321908989218512e-05,
      "loss": 1.0725,
      "num_input_tokens_seen": 24620232,
      "step": 42445
    },
    {
      "epoch": 6.322609472743521,
      "grad_norm": 6.5,
      "learning_rate": 4.321686464698696e-05,
      "loss": 1.0754,
      "num_input_tokens_seen": 24622984,
      "step": 42450
    },
    {
      "epoch": 6.32335418528448,
      "grad_norm": 7.8125,
      "learning_rate": 4.321463909403338e-05,
      "loss": 1.1841,
      "num_input_tokens_seen": 24625640,
      "step": 42455
    },
    {
      "epoch": 6.324098897825439,
      "grad_norm": 10.625,
      "learning_rate": 4.3212413233362e-05,
      "loss": 1.1746,
      "num_input_tokens_seen": 24628552,
      "step": 42460
    },
    {
      "epoch": 6.324843610366399,
      "grad_norm": 13.875,
      "learning_rate": 4.32101870650104e-05,
      "loss": 0.9139,
      "num_input_tokens_seen": 24631368,
      "step": 42465
    },
    {
      "epoch": 6.3255883229073575,
      "grad_norm": 15.5625,
      "learning_rate": 4.3207960589016196e-05,
      "loss": 1.1279,
      "num_input_tokens_seen": 24634184,
      "step": 42470
    },
    {
      "epoch": 6.326333035448317,
      "grad_norm": 13.9375,
      "learning_rate": 4.3205733805417e-05,
      "loss": 0.7968,
      "num_input_tokens_seen": 24637096,
      "step": 42475
    },
    {
      "epoch": 6.327077747989276,
      "grad_norm": 6.15625,
      "learning_rate": 4.320350671425044e-05,
      "loss": 1.4157,
      "num_input_tokens_seen": 24640072,
      "step": 42480
    },
    {
      "epoch": 6.327822460530236,
      "grad_norm": 12.375,
      "learning_rate": 4.320127931555415e-05,
      "loss": 1.4113,
      "num_input_tokens_seen": 24643112,
      "step": 42485
    },
    {
      "epoch": 6.328567173071194,
      "grad_norm": 6.5625,
      "learning_rate": 4.319905160936572e-05,
      "loss": 1.1208,
      "num_input_tokens_seen": 24645864,
      "step": 42490
    },
    {
      "epoch": 6.329311885612154,
      "grad_norm": 9.3125,
      "learning_rate": 4.319682359572282e-05,
      "loss": 1.2318,
      "num_input_tokens_seen": 24648552,
      "step": 42495
    },
    {
      "epoch": 6.330056598153113,
      "grad_norm": 9.3125,
      "learning_rate": 4.319459527466308e-05,
      "loss": 1.129,
      "num_input_tokens_seen": 24651400,
      "step": 42500
    },
    {
      "epoch": 6.330801310694072,
      "grad_norm": 6.78125,
      "learning_rate": 4.3192366646224146e-05,
      "loss": 1.149,
      "num_input_tokens_seen": 24654376,
      "step": 42505
    },
    {
      "epoch": 6.331546023235031,
      "grad_norm": 18.125,
      "learning_rate": 4.3190137710443666e-05,
      "loss": 1.1893,
      "num_input_tokens_seen": 24657448,
      "step": 42510
    },
    {
      "epoch": 6.332290735775991,
      "grad_norm": 6.3125,
      "learning_rate": 4.3187908467359294e-05,
      "loss": 1.1963,
      "num_input_tokens_seen": 24660264,
      "step": 42515
    },
    {
      "epoch": 6.3330354483169495,
      "grad_norm": 16.25,
      "learning_rate": 4.31856789170087e-05,
      "loss": 1.1318,
      "num_input_tokens_seen": 24662984,
      "step": 42520
    },
    {
      "epoch": 6.333780160857909,
      "grad_norm": 13.0,
      "learning_rate": 4.318344905942954e-05,
      "loss": 1.3646,
      "num_input_tokens_seen": 24665928,
      "step": 42525
    },
    {
      "epoch": 6.334524873398868,
      "grad_norm": 9.0625,
      "learning_rate": 4.318121889465949e-05,
      "loss": 1.3897,
      "num_input_tokens_seen": 24669064,
      "step": 42530
    },
    {
      "epoch": 6.335269585939828,
      "grad_norm": 6.0,
      "learning_rate": 4.317898842273622e-05,
      "loss": 1.0775,
      "num_input_tokens_seen": 24671720,
      "step": 42535
    },
    {
      "epoch": 6.336014298480786,
      "grad_norm": 10.125,
      "learning_rate": 4.317675764369743e-05,
      "loss": 1.1163,
      "num_input_tokens_seen": 24674536,
      "step": 42540
    },
    {
      "epoch": 6.336759011021746,
      "grad_norm": 10.0,
      "learning_rate": 4.3174526557580785e-05,
      "loss": 1.3337,
      "num_input_tokens_seen": 24677704,
      "step": 42545
    },
    {
      "epoch": 6.337503723562705,
      "grad_norm": 12.0625,
      "learning_rate": 4.317229516442398e-05,
      "loss": 1.2329,
      "num_input_tokens_seen": 24680808,
      "step": 42550
    },
    {
      "epoch": 6.338248436103664,
      "grad_norm": 7.125,
      "learning_rate": 4.317006346426473e-05,
      "loss": 1.0344,
      "num_input_tokens_seen": 24683528,
      "step": 42555
    },
    {
      "epoch": 6.338993148644623,
      "grad_norm": 8.125,
      "learning_rate": 4.3167831457140715e-05,
      "loss": 1.0732,
      "num_input_tokens_seen": 24686792,
      "step": 42560
    },
    {
      "epoch": 6.339737861185583,
      "grad_norm": 20.125,
      "learning_rate": 4.316559914308966e-05,
      "loss": 1.3989,
      "num_input_tokens_seen": 24689768,
      "step": 42565
    },
    {
      "epoch": 6.340482573726542,
      "grad_norm": 7.625,
      "learning_rate": 4.316336652214926e-05,
      "loss": 1.0683,
      "num_input_tokens_seen": 24692840,
      "step": 42570
    },
    {
      "epoch": 6.3412272862675,
      "grad_norm": 6.46875,
      "learning_rate": 4.316113359435725e-05,
      "loss": 0.9819,
      "num_input_tokens_seen": 24695784,
      "step": 42575
    },
    {
      "epoch": 6.34197199880846,
      "grad_norm": 12.5,
      "learning_rate": 4.315890035975135e-05,
      "loss": 0.895,
      "num_input_tokens_seen": 24698344,
      "step": 42580
    },
    {
      "epoch": 6.342716711349419,
      "grad_norm": 12.3125,
      "learning_rate": 4.315666681836928e-05,
      "loss": 1.4982,
      "num_input_tokens_seen": 24700904,
      "step": 42585
    },
    {
      "epoch": 6.343461423890378,
      "grad_norm": 11.5,
      "learning_rate": 4.315443297024878e-05,
      "loss": 1.3498,
      "num_input_tokens_seen": 24703400,
      "step": 42590
    },
    {
      "epoch": 6.344206136431337,
      "grad_norm": 4.4375,
      "learning_rate": 4.315219881542758e-05,
      "loss": 0.8787,
      "num_input_tokens_seen": 24706408,
      "step": 42595
    },
    {
      "epoch": 6.344950848972297,
      "grad_norm": 12.9375,
      "learning_rate": 4.314996435394344e-05,
      "loss": 1.1939,
      "num_input_tokens_seen": 24709320,
      "step": 42600
    },
    {
      "epoch": 6.3456955615132555,
      "grad_norm": 22.75,
      "learning_rate": 4.314772958583408e-05,
      "loss": 1.2514,
      "num_input_tokens_seen": 24712456,
      "step": 42605
    },
    {
      "epoch": 6.346440274054215,
      "grad_norm": 11.4375,
      "learning_rate": 4.3145494511137294e-05,
      "loss": 1.2592,
      "num_input_tokens_seen": 24715592,
      "step": 42610
    },
    {
      "epoch": 6.347184986595174,
      "grad_norm": 8.0,
      "learning_rate": 4.3143259129890814e-05,
      "loss": 0.6988,
      "num_input_tokens_seen": 24718728,
      "step": 42615
    },
    {
      "epoch": 6.347929699136134,
      "grad_norm": 8.5,
      "learning_rate": 4.314102344213241e-05,
      "loss": 1.0898,
      "num_input_tokens_seen": 24721640,
      "step": 42620
    },
    {
      "epoch": 6.348674411677092,
      "grad_norm": 12.0,
      "learning_rate": 4.3138787447899854e-05,
      "loss": 1.3111,
      "num_input_tokens_seen": 24724616,
      "step": 42625
    },
    {
      "epoch": 6.349419124218052,
      "grad_norm": 5.6875,
      "learning_rate": 4.313655114723092e-05,
      "loss": 0.7427,
      "num_input_tokens_seen": 24727336,
      "step": 42630
    },
    {
      "epoch": 6.350163836759011,
      "grad_norm": 11.0,
      "learning_rate": 4.3134314540163376e-05,
      "loss": 0.9372,
      "num_input_tokens_seen": 24730440,
      "step": 42635
    },
    {
      "epoch": 6.35090854929997,
      "grad_norm": 8.375,
      "learning_rate": 4.3132077626735036e-05,
      "loss": 0.9774,
      "num_input_tokens_seen": 24733352,
      "step": 42640
    },
    {
      "epoch": 6.351653261840929,
      "grad_norm": 16.25,
      "learning_rate": 4.312984040698366e-05,
      "loss": 1.2097,
      "num_input_tokens_seen": 24736296,
      "step": 42645
    },
    {
      "epoch": 6.352397974381889,
      "grad_norm": 8.0625,
      "learning_rate": 4.3127602880947065e-05,
      "loss": 0.8822,
      "num_input_tokens_seen": 24739496,
      "step": 42650
    },
    {
      "epoch": 6.353142686922848,
      "grad_norm": 9.4375,
      "learning_rate": 4.3125365048663035e-05,
      "loss": 1.0519,
      "num_input_tokens_seen": 24742312,
      "step": 42655
    },
    {
      "epoch": 6.353887399463807,
      "grad_norm": 17.625,
      "learning_rate": 4.31231269101694e-05,
      "loss": 1.3329,
      "num_input_tokens_seen": 24745000,
      "step": 42660
    },
    {
      "epoch": 6.354632112004766,
      "grad_norm": 7.9375,
      "learning_rate": 4.312088846550394e-05,
      "loss": 0.9112,
      "num_input_tokens_seen": 24747912,
      "step": 42665
    },
    {
      "epoch": 6.355376824545726,
      "grad_norm": 9.0,
      "learning_rate": 4.311864971470449e-05,
      "loss": 1.1684,
      "num_input_tokens_seen": 24750824,
      "step": 42670
    },
    {
      "epoch": 6.356121537086684,
      "grad_norm": 13.3125,
      "learning_rate": 4.311641065780887e-05,
      "loss": 1.2922,
      "num_input_tokens_seen": 24753704,
      "step": 42675
    },
    {
      "epoch": 6.356866249627644,
      "grad_norm": 6.6875,
      "learning_rate": 4.31141712948549e-05,
      "loss": 0.9771,
      "num_input_tokens_seen": 24756296,
      "step": 42680
    },
    {
      "epoch": 6.357610962168603,
      "grad_norm": 7.375,
      "learning_rate": 4.311193162588043e-05,
      "loss": 1.2603,
      "num_input_tokens_seen": 24759176,
      "step": 42685
    },
    {
      "epoch": 6.358355674709562,
      "grad_norm": 20.125,
      "learning_rate": 4.3109691650923265e-05,
      "loss": 0.9341,
      "num_input_tokens_seen": 24762120,
      "step": 42690
    },
    {
      "epoch": 6.359100387250521,
      "grad_norm": 14.4375,
      "learning_rate": 4.310745137002128e-05,
      "loss": 1.3558,
      "num_input_tokens_seen": 24764936,
      "step": 42695
    },
    {
      "epoch": 6.359845099791481,
      "grad_norm": 7.75,
      "learning_rate": 4.3105210783212304e-05,
      "loss": 0.7594,
      "num_input_tokens_seen": 24767592,
      "step": 42700
    },
    {
      "epoch": 6.36058981233244,
      "grad_norm": 11.6875,
      "learning_rate": 4.310296989053419e-05,
      "loss": 0.9514,
      "num_input_tokens_seen": 24770472,
      "step": 42705
    },
    {
      "epoch": 6.361334524873399,
      "grad_norm": 11.9375,
      "learning_rate": 4.31007286920248e-05,
      "loss": 0.7957,
      "num_input_tokens_seen": 24773352,
      "step": 42710
    },
    {
      "epoch": 6.362079237414358,
      "grad_norm": 4.71875,
      "learning_rate": 4.3098487187721995e-05,
      "loss": 0.9946,
      "num_input_tokens_seen": 24776456,
      "step": 42715
    },
    {
      "epoch": 6.362823949955318,
      "grad_norm": 9.25,
      "learning_rate": 4.3096245377663645e-05,
      "loss": 0.7605,
      "num_input_tokens_seen": 24779336,
      "step": 42720
    },
    {
      "epoch": 6.363568662496276,
      "grad_norm": 9.625,
      "learning_rate": 4.3094003261887625e-05,
      "loss": 1.25,
      "num_input_tokens_seen": 24782440,
      "step": 42725
    },
    {
      "epoch": 6.364313375037236,
      "grad_norm": 5.59375,
      "learning_rate": 4.30917608404318e-05,
      "loss": 1.4774,
      "num_input_tokens_seen": 24785448,
      "step": 42730
    },
    {
      "epoch": 6.365058087578195,
      "grad_norm": 8.5625,
      "learning_rate": 4.308951811333407e-05,
      "loss": 1.0618,
      "num_input_tokens_seen": 24788840,
      "step": 42735
    },
    {
      "epoch": 6.365802800119154,
      "grad_norm": 10.6875,
      "learning_rate": 4.3087275080632314e-05,
      "loss": 0.8394,
      "num_input_tokens_seen": 24791720,
      "step": 42740
    },
    {
      "epoch": 6.366547512660113,
      "grad_norm": 9.25,
      "learning_rate": 4.308503174236443e-05,
      "loss": 1.0185,
      "num_input_tokens_seen": 24794728,
      "step": 42745
    },
    {
      "epoch": 6.367292225201073,
      "grad_norm": 10.6875,
      "learning_rate": 4.308278809856832e-05,
      "loss": 0.9166,
      "num_input_tokens_seen": 24797352,
      "step": 42750
    },
    {
      "epoch": 6.368036937742032,
      "grad_norm": 9.1875,
      "learning_rate": 4.3080544149281875e-05,
      "loss": 1.4593,
      "num_input_tokens_seen": 24800264,
      "step": 42755
    },
    {
      "epoch": 6.36878165028299,
      "grad_norm": 13.25,
      "learning_rate": 4.307829989454302e-05,
      "loss": 1.3383,
      "num_input_tokens_seen": 24803272,
      "step": 42760
    },
    {
      "epoch": 6.36952636282395,
      "grad_norm": 8.875,
      "learning_rate": 4.307605533438965e-05,
      "loss": 0.8832,
      "num_input_tokens_seen": 24806344,
      "step": 42765
    },
    {
      "epoch": 6.370271075364909,
      "grad_norm": 15.5,
      "learning_rate": 4.307381046885971e-05,
      "loss": 1.2402,
      "num_input_tokens_seen": 24809256,
      "step": 42770
    },
    {
      "epoch": 6.371015787905868,
      "grad_norm": 10.4375,
      "learning_rate": 4.307156529799111e-05,
      "loss": 0.7934,
      "num_input_tokens_seen": 24812040,
      "step": 42775
    },
    {
      "epoch": 6.371760500446827,
      "grad_norm": 11.6875,
      "learning_rate": 4.306931982182178e-05,
      "loss": 1.3234,
      "num_input_tokens_seen": 24814920,
      "step": 42780
    },
    {
      "epoch": 6.372505212987787,
      "grad_norm": 10.4375,
      "learning_rate": 4.306707404038966e-05,
      "loss": 1.2443,
      "num_input_tokens_seen": 24817672,
      "step": 42785
    },
    {
      "epoch": 6.373249925528746,
      "grad_norm": 3.578125,
      "learning_rate": 4.306482795373268e-05,
      "loss": 0.9372,
      "num_input_tokens_seen": 24820616,
      "step": 42790
    },
    {
      "epoch": 6.373994638069705,
      "grad_norm": 5.0625,
      "learning_rate": 4.306258156188879e-05,
      "loss": 0.9199,
      "num_input_tokens_seen": 24823528,
      "step": 42795
    },
    {
      "epoch": 6.374739350610664,
      "grad_norm": 11.0625,
      "learning_rate": 4.306033486489595e-05,
      "loss": 0.9015,
      "num_input_tokens_seen": 24826280,
      "step": 42800
    },
    {
      "epoch": 6.375484063151624,
      "grad_norm": 5.71875,
      "learning_rate": 4.30580878627921e-05,
      "loss": 0.927,
      "num_input_tokens_seen": 24829448,
      "step": 42805
    },
    {
      "epoch": 6.376228775692582,
      "grad_norm": 18.25,
      "learning_rate": 4.305584055561522e-05,
      "loss": 1.4542,
      "num_input_tokens_seen": 24832232,
      "step": 42810
    },
    {
      "epoch": 6.376973488233542,
      "grad_norm": 17.5,
      "learning_rate": 4.3053592943403256e-05,
      "loss": 1.0914,
      "num_input_tokens_seen": 24835048,
      "step": 42815
    },
    {
      "epoch": 6.377718200774501,
      "grad_norm": 12.75,
      "learning_rate": 4.305134502619419e-05,
      "loss": 1.1312,
      "num_input_tokens_seen": 24837800,
      "step": 42820
    },
    {
      "epoch": 6.3784629133154604,
      "grad_norm": 8.625,
      "learning_rate": 4.3049096804026e-05,
      "loss": 1.0382,
      "num_input_tokens_seen": 24840872,
      "step": 42825
    },
    {
      "epoch": 6.379207625856419,
      "grad_norm": 7.0625,
      "learning_rate": 4.304684827693666e-05,
      "loss": 1.0194,
      "num_input_tokens_seen": 24843656,
      "step": 42830
    },
    {
      "epoch": 6.379952338397379,
      "grad_norm": 9.0625,
      "learning_rate": 4.304459944496416e-05,
      "loss": 0.9651,
      "num_input_tokens_seen": 24846504,
      "step": 42835
    },
    {
      "epoch": 6.380697050938338,
      "grad_norm": 3.859375,
      "learning_rate": 4.3042350308146496e-05,
      "loss": 1.2101,
      "num_input_tokens_seen": 24849672,
      "step": 42840
    },
    {
      "epoch": 6.381441763479297,
      "grad_norm": 7.46875,
      "learning_rate": 4.304010086652165e-05,
      "loss": 1.1681,
      "num_input_tokens_seen": 24852488,
      "step": 42845
    },
    {
      "epoch": 6.382186476020256,
      "grad_norm": 14.625,
      "learning_rate": 4.3037851120127645e-05,
      "loss": 1.2001,
      "num_input_tokens_seen": 24855368,
      "step": 42850
    },
    {
      "epoch": 6.382931188561216,
      "grad_norm": 20.0,
      "learning_rate": 4.3035601069002476e-05,
      "loss": 1.2244,
      "num_input_tokens_seen": 24858088,
      "step": 42855
    },
    {
      "epoch": 6.383675901102174,
      "grad_norm": 10.0,
      "learning_rate": 4.303335071318416e-05,
      "loss": 1.1237,
      "num_input_tokens_seen": 24861320,
      "step": 42860
    },
    {
      "epoch": 6.384420613643134,
      "grad_norm": 8.5625,
      "learning_rate": 4.303110005271071e-05,
      "loss": 1.0224,
      "num_input_tokens_seen": 24864328,
      "step": 42865
    },
    {
      "epoch": 6.385165326184093,
      "grad_norm": 5.25,
      "learning_rate": 4.302884908762015e-05,
      "loss": 0.9895,
      "num_input_tokens_seen": 24867176,
      "step": 42870
    },
    {
      "epoch": 6.3859100387250525,
      "grad_norm": 13.375,
      "learning_rate": 4.302659781795051e-05,
      "loss": 1.222,
      "num_input_tokens_seen": 24870152,
      "step": 42875
    },
    {
      "epoch": 6.386654751266011,
      "grad_norm": 17.25,
      "learning_rate": 4.302434624373982e-05,
      "loss": 1.2678,
      "num_input_tokens_seen": 24872968,
      "step": 42880
    },
    {
      "epoch": 6.387399463806971,
      "grad_norm": 5.4375,
      "learning_rate": 4.3022094365026124e-05,
      "loss": 0.7298,
      "num_input_tokens_seen": 24875912,
      "step": 42885
    },
    {
      "epoch": 6.38814417634793,
      "grad_norm": 9.375,
      "learning_rate": 4.3019842181847456e-05,
      "loss": 0.8999,
      "num_input_tokens_seen": 24878984,
      "step": 42890
    },
    {
      "epoch": 6.388888888888889,
      "grad_norm": 7.96875,
      "learning_rate": 4.301758969424187e-05,
      "loss": 1.107,
      "num_input_tokens_seen": 24882440,
      "step": 42895
    },
    {
      "epoch": 6.389633601429848,
      "grad_norm": 18.125,
      "learning_rate": 4.301533690224741e-05,
      "loss": 1.0113,
      "num_input_tokens_seen": 24885416,
      "step": 42900
    },
    {
      "epoch": 6.390378313970807,
      "grad_norm": 10.4375,
      "learning_rate": 4.3013083805902156e-05,
      "loss": 1.0026,
      "num_input_tokens_seen": 24888392,
      "step": 42905
    },
    {
      "epoch": 6.3911230265117664,
      "grad_norm": 9.0625,
      "learning_rate": 4.301083040524415e-05,
      "loss": 1.3524,
      "num_input_tokens_seen": 24891368,
      "step": 42910
    },
    {
      "epoch": 6.391867739052726,
      "grad_norm": 7.28125,
      "learning_rate": 4.3008576700311473e-05,
      "loss": 0.6215,
      "num_input_tokens_seen": 24894312,
      "step": 42915
    },
    {
      "epoch": 6.392612451593685,
      "grad_norm": 13.9375,
      "learning_rate": 4.30063226911422e-05,
      "loss": 1.1825,
      "num_input_tokens_seen": 24897000,
      "step": 42920
    },
    {
      "epoch": 6.393357164134644,
      "grad_norm": 10.6875,
      "learning_rate": 4.30040683777744e-05,
      "loss": 1.1778,
      "num_input_tokens_seen": 24899528,
      "step": 42925
    },
    {
      "epoch": 6.394101876675603,
      "grad_norm": 13.5625,
      "learning_rate": 4.300181376024616e-05,
      "loss": 1.1659,
      "num_input_tokens_seen": 24902632,
      "step": 42930
    },
    {
      "epoch": 6.394846589216562,
      "grad_norm": 9.875,
      "learning_rate": 4.299955883859558e-05,
      "loss": 1.0479,
      "num_input_tokens_seen": 24905352,
      "step": 42935
    },
    {
      "epoch": 6.395591301757522,
      "grad_norm": 4.84375,
      "learning_rate": 4.2997303612860746e-05,
      "loss": 1.1276,
      "num_input_tokens_seen": 24908520,
      "step": 42940
    },
    {
      "epoch": 6.39633601429848,
      "grad_norm": 5.71875,
      "learning_rate": 4.299504808307976e-05,
      "loss": 0.991,
      "num_input_tokens_seen": 24911240,
      "step": 42945
    },
    {
      "epoch": 6.39708072683944,
      "grad_norm": 8.875,
      "learning_rate": 4.299279224929072e-05,
      "loss": 0.8232,
      "num_input_tokens_seen": 24913960,
      "step": 42950
    },
    {
      "epoch": 6.397825439380399,
      "grad_norm": 9.5,
      "learning_rate": 4.299053611153175e-05,
      "loss": 1.2796,
      "num_input_tokens_seen": 24917096,
      "step": 42955
    },
    {
      "epoch": 6.3985701519213585,
      "grad_norm": 8.625,
      "learning_rate": 4.2988279669840945e-05,
      "loss": 1.0306,
      "num_input_tokens_seen": 24919944,
      "step": 42960
    },
    {
      "epoch": 6.399314864462317,
      "grad_norm": 10.0,
      "learning_rate": 4.298602292425645e-05,
      "loss": 1.2248,
      "num_input_tokens_seen": 24922696,
      "step": 42965
    },
    {
      "epoch": 6.400059577003277,
      "grad_norm": 11.0625,
      "learning_rate": 4.298376587481637e-05,
      "loss": 1.2336,
      "num_input_tokens_seen": 24925448,
      "step": 42970
    },
    {
      "epoch": 6.400804289544236,
      "grad_norm": 6.9375,
      "learning_rate": 4.2981508521558854e-05,
      "loss": 1.0066,
      "num_input_tokens_seen": 24927976,
      "step": 42975
    },
    {
      "epoch": 6.401549002085195,
      "grad_norm": 10.125,
      "learning_rate": 4.2979250864522016e-05,
      "loss": 1.0248,
      "num_input_tokens_seen": 24930824,
      "step": 42980
    },
    {
      "epoch": 6.402293714626154,
      "grad_norm": 15.875,
      "learning_rate": 4.297699290374401e-05,
      "loss": 0.979,
      "num_input_tokens_seen": 24933448,
      "step": 42985
    },
    {
      "epoch": 6.403038427167114,
      "grad_norm": 15.375,
      "learning_rate": 4.297473463926299e-05,
      "loss": 0.8546,
      "num_input_tokens_seen": 24936616,
      "step": 42990
    },
    {
      "epoch": 6.4037831397080724,
      "grad_norm": 5.5625,
      "learning_rate": 4.2972476071117086e-05,
      "loss": 0.7413,
      "num_input_tokens_seen": 24939464,
      "step": 42995
    },
    {
      "epoch": 6.404527852249032,
      "grad_norm": 5.3125,
      "learning_rate": 4.2970217199344465e-05,
      "loss": 1.0878,
      "num_input_tokens_seen": 24942792,
      "step": 43000
    },
    {
      "epoch": 6.405272564789991,
      "grad_norm": 4.03125,
      "learning_rate": 4.296795802398329e-05,
      "loss": 0.8341,
      "num_input_tokens_seen": 24945512,
      "step": 43005
    },
    {
      "epoch": 6.4060172773309505,
      "grad_norm": 9.6875,
      "learning_rate": 4.296569854507173e-05,
      "loss": 1.0565,
      "num_input_tokens_seen": 24948456,
      "step": 43010
    },
    {
      "epoch": 6.406761989871909,
      "grad_norm": 11.375,
      "learning_rate": 4.2963438762647954e-05,
      "loss": 1.048,
      "num_input_tokens_seen": 24951528,
      "step": 43015
    },
    {
      "epoch": 6.407506702412869,
      "grad_norm": 14.9375,
      "learning_rate": 4.2961178676750124e-05,
      "loss": 1.1076,
      "num_input_tokens_seen": 24954536,
      "step": 43020
    },
    {
      "epoch": 6.408251414953828,
      "grad_norm": 9.3125,
      "learning_rate": 4.295891828741645e-05,
      "loss": 0.8225,
      "num_input_tokens_seen": 24957576,
      "step": 43025
    },
    {
      "epoch": 6.408996127494787,
      "grad_norm": 14.5,
      "learning_rate": 4.29566575946851e-05,
      "loss": 1.0695,
      "num_input_tokens_seen": 24960264,
      "step": 43030
    },
    {
      "epoch": 6.409740840035746,
      "grad_norm": 6.71875,
      "learning_rate": 4.295439659859427e-05,
      "loss": 0.9767,
      "num_input_tokens_seen": 24963144,
      "step": 43035
    },
    {
      "epoch": 6.410485552576706,
      "grad_norm": 15.3125,
      "learning_rate": 4.2952135299182155e-05,
      "loss": 1.6705,
      "num_input_tokens_seen": 24965960,
      "step": 43040
    },
    {
      "epoch": 6.4112302651176645,
      "grad_norm": 8.0625,
      "learning_rate": 4.294987369648696e-05,
      "loss": 1.0204,
      "num_input_tokens_seen": 24968872,
      "step": 43045
    },
    {
      "epoch": 6.411974977658624,
      "grad_norm": 13.8125,
      "learning_rate": 4.2947611790546894e-05,
      "loss": 1.0006,
      "num_input_tokens_seen": 24971688,
      "step": 43050
    },
    {
      "epoch": 6.412719690199583,
      "grad_norm": 8.625,
      "learning_rate": 4.2945349581400174e-05,
      "loss": 1.1069,
      "num_input_tokens_seen": 24974632,
      "step": 43055
    },
    {
      "epoch": 6.4134644027405425,
      "grad_norm": 8.6875,
      "learning_rate": 4.2943087069085e-05,
      "loss": 0.9801,
      "num_input_tokens_seen": 24977480,
      "step": 43060
    },
    {
      "epoch": 6.414209115281501,
      "grad_norm": 20.375,
      "learning_rate": 4.294082425363961e-05,
      "loss": 1.1003,
      "num_input_tokens_seen": 24980648,
      "step": 43065
    },
    {
      "epoch": 6.414953827822461,
      "grad_norm": 17.625,
      "learning_rate": 4.293856113510223e-05,
      "loss": 0.8856,
      "num_input_tokens_seen": 24983624,
      "step": 43070
    },
    {
      "epoch": 6.41569854036342,
      "grad_norm": 9.875,
      "learning_rate": 4.29362977135111e-05,
      "loss": 1.1263,
      "num_input_tokens_seen": 24986568,
      "step": 43075
    },
    {
      "epoch": 6.416443252904379,
      "grad_norm": 14.9375,
      "learning_rate": 4.2934033988904437e-05,
      "loss": 0.8913,
      "num_input_tokens_seen": 24989128,
      "step": 43080
    },
    {
      "epoch": 6.417187965445338,
      "grad_norm": 10.75,
      "learning_rate": 4.2931769961320504e-05,
      "loss": 0.9203,
      "num_input_tokens_seen": 24991816,
      "step": 43085
    },
    {
      "epoch": 6.417932677986297,
      "grad_norm": 7.78125,
      "learning_rate": 4.292950563079754e-05,
      "loss": 0.9731,
      "num_input_tokens_seen": 24994792,
      "step": 43090
    },
    {
      "epoch": 6.4186773905272565,
      "grad_norm": 10.4375,
      "learning_rate": 4.2927240997373795e-05,
      "loss": 1.0326,
      "num_input_tokens_seen": 24997512,
      "step": 43095
    },
    {
      "epoch": 6.419422103068215,
      "grad_norm": 15.375,
      "learning_rate": 4.292497606108754e-05,
      "loss": 0.9225,
      "num_input_tokens_seen": 25000264,
      "step": 43100
    },
    {
      "epoch": 6.420166815609175,
      "grad_norm": 13.5,
      "learning_rate": 4.2922710821977044e-05,
      "loss": 0.8802,
      "num_input_tokens_seen": 25003208,
      "step": 43105
    },
    {
      "epoch": 6.420911528150134,
      "grad_norm": 10.1875,
      "learning_rate": 4.2920445280080544e-05,
      "loss": 1.3358,
      "num_input_tokens_seen": 25006088,
      "step": 43110
    },
    {
      "epoch": 6.421656240691093,
      "grad_norm": 10.5,
      "learning_rate": 4.291817943543634e-05,
      "loss": 0.7719,
      "num_input_tokens_seen": 25008712,
      "step": 43115
    },
    {
      "epoch": 6.422400953232052,
      "grad_norm": 14.1875,
      "learning_rate": 4.291591328808272e-05,
      "loss": 1.0413,
      "num_input_tokens_seen": 25011592,
      "step": 43120
    },
    {
      "epoch": 6.423145665773012,
      "grad_norm": 12.625,
      "learning_rate": 4.291364683805794e-05,
      "loss": 0.9989,
      "num_input_tokens_seen": 25014376,
      "step": 43125
    },
    {
      "epoch": 6.4238903783139705,
      "grad_norm": 19.875,
      "learning_rate": 4.291138008540031e-05,
      "loss": 1.011,
      "num_input_tokens_seen": 25017320,
      "step": 43130
    },
    {
      "epoch": 6.42463509085493,
      "grad_norm": 4.90625,
      "learning_rate": 4.2909113030148106e-05,
      "loss": 1.1408,
      "num_input_tokens_seen": 25020264,
      "step": 43135
    },
    {
      "epoch": 6.425379803395889,
      "grad_norm": 8.375,
      "learning_rate": 4.290684567233965e-05,
      "loss": 1.1654,
      "num_input_tokens_seen": 25023144,
      "step": 43140
    },
    {
      "epoch": 6.4261245159368485,
      "grad_norm": 16.125,
      "learning_rate": 4.2904578012013233e-05,
      "loss": 0.8946,
      "num_input_tokens_seen": 25026216,
      "step": 43145
    },
    {
      "epoch": 6.426869228477807,
      "grad_norm": 16.875,
      "learning_rate": 4.290231004920717e-05,
      "loss": 1.3166,
      "num_input_tokens_seen": 25029128,
      "step": 43150
    },
    {
      "epoch": 6.427613941018767,
      "grad_norm": 17.75,
      "learning_rate": 4.2900041783959775e-05,
      "loss": 1.0993,
      "num_input_tokens_seen": 25032104,
      "step": 43155
    },
    {
      "epoch": 6.428358653559726,
      "grad_norm": 24.0,
      "learning_rate": 4.2897773216309366e-05,
      "loss": 0.9084,
      "num_input_tokens_seen": 25034856,
      "step": 43160
    },
    {
      "epoch": 6.429103366100685,
      "grad_norm": 9.0625,
      "learning_rate": 4.289550434629426e-05,
      "loss": 1.1688,
      "num_input_tokens_seen": 25037768,
      "step": 43165
    },
    {
      "epoch": 6.429848078641644,
      "grad_norm": 10.375,
      "learning_rate": 4.2893235173952805e-05,
      "loss": 0.93,
      "num_input_tokens_seen": 25040616,
      "step": 43170
    },
    {
      "epoch": 6.430592791182604,
      "grad_norm": 6.53125,
      "learning_rate": 4.2890965699323335e-05,
      "loss": 0.9221,
      "num_input_tokens_seen": 25043464,
      "step": 43175
    },
    {
      "epoch": 6.4313375037235625,
      "grad_norm": 11.0,
      "learning_rate": 4.288869592244417e-05,
      "loss": 1.4129,
      "num_input_tokens_seen": 25046472,
      "step": 43180
    },
    {
      "epoch": 6.432082216264522,
      "grad_norm": 7.65625,
      "learning_rate": 4.288642584335367e-05,
      "loss": 1.0495,
      "num_input_tokens_seen": 25049512,
      "step": 43185
    },
    {
      "epoch": 6.432826928805481,
      "grad_norm": 12.375,
      "learning_rate": 4.2884155462090194e-05,
      "loss": 0.8656,
      "num_input_tokens_seen": 25052296,
      "step": 43190
    },
    {
      "epoch": 6.4335716413464406,
      "grad_norm": 12.8125,
      "learning_rate": 4.2881884778692076e-05,
      "loss": 1.516,
      "num_input_tokens_seen": 25055304,
      "step": 43195
    },
    {
      "epoch": 6.434316353887399,
      "grad_norm": 15.3125,
      "learning_rate": 4.287961379319769e-05,
      "loss": 1.004,
      "num_input_tokens_seen": 25058024,
      "step": 43200
    },
    {
      "epoch": 6.435061066428359,
      "grad_norm": 10.0,
      "learning_rate": 4.287734250564541e-05,
      "loss": 1.2298,
      "num_input_tokens_seen": 25060968,
      "step": 43205
    },
    {
      "epoch": 6.435805778969318,
      "grad_norm": 5.03125,
      "learning_rate": 4.28750709160736e-05,
      "loss": 1.1066,
      "num_input_tokens_seen": 25063752,
      "step": 43210
    },
    {
      "epoch": 6.436550491510277,
      "grad_norm": 10.6875,
      "learning_rate": 4.2872799024520626e-05,
      "loss": 1.3367,
      "num_input_tokens_seen": 25066728,
      "step": 43215
    },
    {
      "epoch": 6.437295204051236,
      "grad_norm": 6.375,
      "learning_rate": 4.287052683102488e-05,
      "loss": 0.8807,
      "num_input_tokens_seen": 25069512,
      "step": 43220
    },
    {
      "epoch": 6.438039916592196,
      "grad_norm": 15.4375,
      "learning_rate": 4.286825433562474e-05,
      "loss": 1.4369,
      "num_input_tokens_seen": 25072520,
      "step": 43225
    },
    {
      "epoch": 6.4387846291331545,
      "grad_norm": 8.4375,
      "learning_rate": 4.286598153835861e-05,
      "loss": 1.1754,
      "num_input_tokens_seen": 25075336,
      "step": 43230
    },
    {
      "epoch": 6.439529341674114,
      "grad_norm": 11.875,
      "learning_rate": 4.2863708439264886e-05,
      "loss": 1.1663,
      "num_input_tokens_seen": 25078152,
      "step": 43235
    },
    {
      "epoch": 6.440274054215073,
      "grad_norm": 18.25,
      "learning_rate": 4.286143503838195e-05,
      "loss": 1.2531,
      "num_input_tokens_seen": 25080840,
      "step": 43240
    },
    {
      "epoch": 6.441018766756033,
      "grad_norm": 14.4375,
      "learning_rate": 4.285916133574823e-05,
      "loss": 1.1992,
      "num_input_tokens_seen": 25083624,
      "step": 43245
    },
    {
      "epoch": 6.441763479296991,
      "grad_norm": 6.78125,
      "learning_rate": 4.2856887331402126e-05,
      "loss": 1.6288,
      "num_input_tokens_seen": 25086504,
      "step": 43250
    },
    {
      "epoch": 6.44250819183795,
      "grad_norm": 13.5,
      "learning_rate": 4.285461302538207e-05,
      "loss": 1.1707,
      "num_input_tokens_seen": 25089384,
      "step": 43255
    },
    {
      "epoch": 6.44325290437891,
      "grad_norm": 10.8125,
      "learning_rate": 4.285233841772647e-05,
      "loss": 1.0656,
      "num_input_tokens_seen": 25092232,
      "step": 43260
    },
    {
      "epoch": 6.443997616919869,
      "grad_norm": 11.375,
      "learning_rate": 4.2850063508473746e-05,
      "loss": 1.3269,
      "num_input_tokens_seen": 25095176,
      "step": 43265
    },
    {
      "epoch": 6.444742329460828,
      "grad_norm": 8.5,
      "learning_rate": 4.284778829766235e-05,
      "loss": 0.8598,
      "num_input_tokens_seen": 25098184,
      "step": 43270
    },
    {
      "epoch": 6.445487042001787,
      "grad_norm": 16.375,
      "learning_rate": 4.284551278533071e-05,
      "loss": 1.2485,
      "num_input_tokens_seen": 25101032,
      "step": 43275
    },
    {
      "epoch": 6.4462317545427466,
      "grad_norm": 12.75,
      "learning_rate": 4.284323697151726e-05,
      "loss": 1.1272,
      "num_input_tokens_seen": 25103752,
      "step": 43280
    },
    {
      "epoch": 6.446976467083705,
      "grad_norm": 7.5,
      "learning_rate": 4.284096085626047e-05,
      "loss": 1.0272,
      "num_input_tokens_seen": 25106536,
      "step": 43285
    },
    {
      "epoch": 6.447721179624665,
      "grad_norm": 10.5,
      "learning_rate": 4.283868443959877e-05,
      "loss": 0.8392,
      "num_input_tokens_seen": 25109224,
      "step": 43290
    },
    {
      "epoch": 6.448465892165624,
      "grad_norm": 4.28125,
      "learning_rate": 4.283640772157064e-05,
      "loss": 1.1214,
      "num_input_tokens_seen": 25112392,
      "step": 43295
    },
    {
      "epoch": 6.449210604706583,
      "grad_norm": 11.25,
      "learning_rate": 4.283413070221452e-05,
      "loss": 1.1249,
      "num_input_tokens_seen": 25115624,
      "step": 43300
    },
    {
      "epoch": 6.449955317247542,
      "grad_norm": 9.8125,
      "learning_rate": 4.283185338156888e-05,
      "loss": 1.1794,
      "num_input_tokens_seen": 25118408,
      "step": 43305
    },
    {
      "epoch": 6.450700029788502,
      "grad_norm": 11.3125,
      "learning_rate": 4.282957575967221e-05,
      "loss": 1.231,
      "num_input_tokens_seen": 25121448,
      "step": 43310
    },
    {
      "epoch": 6.4514447423294605,
      "grad_norm": 11.1875,
      "learning_rate": 4.282729783656298e-05,
      "loss": 1.2115,
      "num_input_tokens_seen": 25124424,
      "step": 43315
    },
    {
      "epoch": 6.45218945487042,
      "grad_norm": 11.9375,
      "learning_rate": 4.2825019612279666e-05,
      "loss": 1.3351,
      "num_input_tokens_seen": 25127112,
      "step": 43320
    },
    {
      "epoch": 6.452934167411379,
      "grad_norm": 11.375,
      "learning_rate": 4.282274108686076e-05,
      "loss": 1.2745,
      "num_input_tokens_seen": 25129672,
      "step": 43325
    },
    {
      "epoch": 6.453678879952339,
      "grad_norm": 17.0,
      "learning_rate": 4.282046226034476e-05,
      "loss": 0.9409,
      "num_input_tokens_seen": 25132648,
      "step": 43330
    },
    {
      "epoch": 6.454423592493297,
      "grad_norm": 8.125,
      "learning_rate": 4.2818183132770175e-05,
      "loss": 1.111,
      "num_input_tokens_seen": 25135752,
      "step": 43335
    },
    {
      "epoch": 6.455168305034257,
      "grad_norm": 8.125,
      "learning_rate": 4.281590370417548e-05,
      "loss": 0.8565,
      "num_input_tokens_seen": 25138760,
      "step": 43340
    },
    {
      "epoch": 6.455913017575216,
      "grad_norm": 9.5625,
      "learning_rate": 4.28136239745992e-05,
      "loss": 1.2242,
      "num_input_tokens_seen": 25141832,
      "step": 43345
    },
    {
      "epoch": 6.456657730116175,
      "grad_norm": 5.59375,
      "learning_rate": 4.2811343944079855e-05,
      "loss": 0.9997,
      "num_input_tokens_seen": 25144744,
      "step": 43350
    },
    {
      "epoch": 6.457402442657134,
      "grad_norm": 9.5,
      "learning_rate": 4.280906361265595e-05,
      "loss": 1.031,
      "num_input_tokens_seen": 25147688,
      "step": 43355
    },
    {
      "epoch": 6.458147155198094,
      "grad_norm": 10.75,
      "learning_rate": 4.2806782980366025e-05,
      "loss": 1.2938,
      "num_input_tokens_seen": 25150664,
      "step": 43360
    },
    {
      "epoch": 6.4588918677390526,
      "grad_norm": 26.5,
      "learning_rate": 4.2804502047248594e-05,
      "loss": 1.2507,
      "num_input_tokens_seen": 25153544,
      "step": 43365
    },
    {
      "epoch": 6.459636580280012,
      "grad_norm": 12.625,
      "learning_rate": 4.2802220813342194e-05,
      "loss": 1.3082,
      "num_input_tokens_seen": 25156712,
      "step": 43370
    },
    {
      "epoch": 6.460381292820971,
      "grad_norm": 14.125,
      "learning_rate": 4.2799939278685376e-05,
      "loss": 1.1906,
      "num_input_tokens_seen": 25159720,
      "step": 43375
    },
    {
      "epoch": 6.461126005361931,
      "grad_norm": 23.25,
      "learning_rate": 4.279765744331666e-05,
      "loss": 1.2012,
      "num_input_tokens_seen": 25162600,
      "step": 43380
    },
    {
      "epoch": 6.461870717902889,
      "grad_norm": 19.375,
      "learning_rate": 4.2795375307274624e-05,
      "loss": 1.3026,
      "num_input_tokens_seen": 25165896,
      "step": 43385
    },
    {
      "epoch": 6.462615430443849,
      "grad_norm": 8.875,
      "learning_rate": 4.2793092870597804e-05,
      "loss": 0.7545,
      "num_input_tokens_seen": 25168680,
      "step": 43390
    },
    {
      "epoch": 6.463360142984808,
      "grad_norm": 6.46875,
      "learning_rate": 4.279081013332476e-05,
      "loss": 0.7807,
      "num_input_tokens_seen": 25171912,
      "step": 43395
    },
    {
      "epoch": 6.464104855525767,
      "grad_norm": 4.625,
      "learning_rate": 4.278852709549406e-05,
      "loss": 1.0567,
      "num_input_tokens_seen": 25174952,
      "step": 43400
    },
    {
      "epoch": 6.464849568066726,
      "grad_norm": 10.125,
      "learning_rate": 4.2786243757144284e-05,
      "loss": 1.0559,
      "num_input_tokens_seen": 25177640,
      "step": 43405
    },
    {
      "epoch": 6.465594280607686,
      "grad_norm": 7.8125,
      "learning_rate": 4.278396011831399e-05,
      "loss": 0.732,
      "num_input_tokens_seen": 25180584,
      "step": 43410
    },
    {
      "epoch": 6.466338993148645,
      "grad_norm": 14.1875,
      "learning_rate": 4.2781676179041764e-05,
      "loss": 1.2393,
      "num_input_tokens_seen": 25183432,
      "step": 43415
    },
    {
      "epoch": 6.467083705689604,
      "grad_norm": 10.875,
      "learning_rate": 4.2779391939366194e-05,
      "loss": 1.1137,
      "num_input_tokens_seen": 25186504,
      "step": 43420
    },
    {
      "epoch": 6.467828418230563,
      "grad_norm": 12.9375,
      "learning_rate": 4.277710739932586e-05,
      "loss": 0.8439,
      "num_input_tokens_seen": 25188904,
      "step": 43425
    },
    {
      "epoch": 6.468573130771523,
      "grad_norm": 19.0,
      "learning_rate": 4.277482255895937e-05,
      "loss": 1.5792,
      "num_input_tokens_seen": 25191688,
      "step": 43430
    },
    {
      "epoch": 6.469317843312481,
      "grad_norm": 32.75,
      "learning_rate": 4.277253741830532e-05,
      "loss": 1.2064,
      "num_input_tokens_seen": 25194600,
      "step": 43435
    },
    {
      "epoch": 6.47006255585344,
      "grad_norm": 7.65625,
      "learning_rate": 4.2770251977402314e-05,
      "loss": 1.2855,
      "num_input_tokens_seen": 25197288,
      "step": 43440
    },
    {
      "epoch": 6.4708072683944,
      "grad_norm": 8.6875,
      "learning_rate": 4.2767966236288956e-05,
      "loss": 0.8126,
      "num_input_tokens_seen": 25200456,
      "step": 43445
    },
    {
      "epoch": 6.4715519809353586,
      "grad_norm": 16.625,
      "learning_rate": 4.276568019500388e-05,
      "loss": 1.0894,
      "num_input_tokens_seen": 25203304,
      "step": 43450
    },
    {
      "epoch": 6.472296693476318,
      "grad_norm": 9.5625,
      "learning_rate": 4.276339385358568e-05,
      "loss": 0.969,
      "num_input_tokens_seen": 25206024,
      "step": 43455
    },
    {
      "epoch": 6.473041406017277,
      "grad_norm": 16.125,
      "learning_rate": 4.2761107212073e-05,
      "loss": 1.3474,
      "num_input_tokens_seen": 25208680,
      "step": 43460
    },
    {
      "epoch": 6.473786118558237,
      "grad_norm": 8.875,
      "learning_rate": 4.275882027050446e-05,
      "loss": 1.0509,
      "num_input_tokens_seen": 25211688,
      "step": 43465
    },
    {
      "epoch": 6.474530831099195,
      "grad_norm": 14.1875,
      "learning_rate": 4.275653302891871e-05,
      "loss": 1.166,
      "num_input_tokens_seen": 25214696,
      "step": 43470
    },
    {
      "epoch": 6.475275543640155,
      "grad_norm": 11.8125,
      "learning_rate": 4.275424548735437e-05,
      "loss": 1.0965,
      "num_input_tokens_seen": 25217480,
      "step": 43475
    },
    {
      "epoch": 6.476020256181114,
      "grad_norm": 13.0625,
      "learning_rate": 4.27519576458501e-05,
      "loss": 1.0265,
      "num_input_tokens_seen": 25220360,
      "step": 43480
    },
    {
      "epoch": 6.476764968722073,
      "grad_norm": 19.0,
      "learning_rate": 4.274966950444456e-05,
      "loss": 1.1846,
      "num_input_tokens_seen": 25223112,
      "step": 43485
    },
    {
      "epoch": 6.477509681263032,
      "grad_norm": 6.65625,
      "learning_rate": 4.2747381063176384e-05,
      "loss": 0.8474,
      "num_input_tokens_seen": 25225960,
      "step": 43490
    },
    {
      "epoch": 6.478254393803992,
      "grad_norm": 11.3125,
      "learning_rate": 4.274509232208425e-05,
      "loss": 1.1497,
      "num_input_tokens_seen": 25228616,
      "step": 43495
    },
    {
      "epoch": 6.478999106344951,
      "grad_norm": 5.15625,
      "learning_rate": 4.274280328120681e-05,
      "loss": 1.0152,
      "num_input_tokens_seen": 25231560,
      "step": 43500
    },
    {
      "epoch": 6.47974381888591,
      "grad_norm": 12.75,
      "learning_rate": 4.274051394058274e-05,
      "loss": 1.2405,
      "num_input_tokens_seen": 25234376,
      "step": 43505
    },
    {
      "epoch": 6.480488531426869,
      "grad_norm": 10.0,
      "learning_rate": 4.273822430025072e-05,
      "loss": 1.1699,
      "num_input_tokens_seen": 25237288,
      "step": 43510
    },
    {
      "epoch": 6.481233243967829,
      "grad_norm": 7.28125,
      "learning_rate": 4.2735934360249426e-05,
      "loss": 0.7522,
      "num_input_tokens_seen": 25239816,
      "step": 43515
    },
    {
      "epoch": 6.481977956508787,
      "grad_norm": 5.40625,
      "learning_rate": 4.2733644120617547e-05,
      "loss": 1.17,
      "num_input_tokens_seen": 25242632,
      "step": 43520
    },
    {
      "epoch": 6.482722669049747,
      "grad_norm": 5.34375,
      "learning_rate": 4.273135358139377e-05,
      "loss": 1.1729,
      "num_input_tokens_seen": 25245480,
      "step": 43525
    },
    {
      "epoch": 6.483467381590706,
      "grad_norm": 10.0,
      "learning_rate": 4.272906274261681e-05,
      "loss": 1.0633,
      "num_input_tokens_seen": 25248360,
      "step": 43530
    },
    {
      "epoch": 6.484212094131665,
      "grad_norm": 6.84375,
      "learning_rate": 4.2726771604325346e-05,
      "loss": 0.9371,
      "num_input_tokens_seen": 25251208,
      "step": 43535
    },
    {
      "epoch": 6.484956806672624,
      "grad_norm": 6.875,
      "learning_rate": 4.272448016655809e-05,
      "loss": 0.9404,
      "num_input_tokens_seen": 25254056,
      "step": 43540
    },
    {
      "epoch": 6.485701519213584,
      "grad_norm": 13.25,
      "learning_rate": 4.272218842935376e-05,
      "loss": 1.1097,
      "num_input_tokens_seen": 25256808,
      "step": 43545
    },
    {
      "epoch": 6.486446231754543,
      "grad_norm": 20.75,
      "learning_rate": 4.271989639275107e-05,
      "loss": 1.1303,
      "num_input_tokens_seen": 25259752,
      "step": 43550
    },
    {
      "epoch": 6.487190944295502,
      "grad_norm": 7.59375,
      "learning_rate": 4.271760405678874e-05,
      "loss": 1.1857,
      "num_input_tokens_seen": 25262984,
      "step": 43555
    },
    {
      "epoch": 6.487935656836461,
      "grad_norm": 7.34375,
      "learning_rate": 4.2715311421505486e-05,
      "loss": 1.0457,
      "num_input_tokens_seen": 25265608,
      "step": 43560
    },
    {
      "epoch": 6.488680369377421,
      "grad_norm": 6.96875,
      "learning_rate": 4.271301848694006e-05,
      "loss": 1.0937,
      "num_input_tokens_seen": 25268680,
      "step": 43565
    },
    {
      "epoch": 6.489425081918379,
      "grad_norm": 8.8125,
      "learning_rate": 4.271072525313119e-05,
      "loss": 1.1835,
      "num_input_tokens_seen": 25271720,
      "step": 43570
    },
    {
      "epoch": 6.490169794459339,
      "grad_norm": 7.59375,
      "learning_rate": 4.2708431720117614e-05,
      "loss": 1.4646,
      "num_input_tokens_seen": 25274504,
      "step": 43575
    },
    {
      "epoch": 6.490914507000298,
      "grad_norm": 10.6875,
      "learning_rate": 4.270613788793808e-05,
      "loss": 1.1114,
      "num_input_tokens_seen": 25277416,
      "step": 43580
    },
    {
      "epoch": 6.4916592195412575,
      "grad_norm": 10.5625,
      "learning_rate": 4.2703843756631344e-05,
      "loss": 1.107,
      "num_input_tokens_seen": 25280328,
      "step": 43585
    },
    {
      "epoch": 6.492403932082216,
      "grad_norm": 9.4375,
      "learning_rate": 4.270154932623617e-05,
      "loss": 1.1725,
      "num_input_tokens_seen": 25283240,
      "step": 43590
    },
    {
      "epoch": 6.493148644623176,
      "grad_norm": 12.0,
      "learning_rate": 4.26992545967913e-05,
      "loss": 1.249,
      "num_input_tokens_seen": 25286376,
      "step": 43595
    },
    {
      "epoch": 6.493893357164135,
      "grad_norm": 20.625,
      "learning_rate": 4.2696959568335515e-05,
      "loss": 1.1706,
      "num_input_tokens_seen": 25289768,
      "step": 43600
    },
    {
      "epoch": 6.494638069705093,
      "grad_norm": 5.6875,
      "learning_rate": 4.2694664240907586e-05,
      "loss": 1.1391,
      "num_input_tokens_seen": 25292584,
      "step": 43605
    },
    {
      "epoch": 6.495382782246053,
      "grad_norm": 13.625,
      "learning_rate": 4.269236861454629e-05,
      "loss": 1.0805,
      "num_input_tokens_seen": 25295560,
      "step": 43610
    },
    {
      "epoch": 6.496127494787013,
      "grad_norm": 8.5,
      "learning_rate": 4.2690072689290405e-05,
      "loss": 0.9148,
      "num_input_tokens_seen": 25298408,
      "step": 43615
    },
    {
      "epoch": 6.496872207327971,
      "grad_norm": 9.1875,
      "learning_rate": 4.268777646517872e-05,
      "loss": 0.7318,
      "num_input_tokens_seen": 25301480,
      "step": 43620
    },
    {
      "epoch": 6.49761691986893,
      "grad_norm": 5.65625,
      "learning_rate": 4.268547994225003e-05,
      "loss": 0.9391,
      "num_input_tokens_seen": 25304264,
      "step": 43625
    },
    {
      "epoch": 6.49836163240989,
      "grad_norm": 9.25,
      "learning_rate": 4.2683183120543134e-05,
      "loss": 1.061,
      "num_input_tokens_seen": 25307016,
      "step": 43630
    },
    {
      "epoch": 6.499106344950849,
      "grad_norm": 6.21875,
      "learning_rate": 4.2680886000096834e-05,
      "loss": 1.0578,
      "num_input_tokens_seen": 25309864,
      "step": 43635
    },
    {
      "epoch": 6.499851057491808,
      "grad_norm": 10.625,
      "learning_rate": 4.267858858094993e-05,
      "loss": 0.9498,
      "num_input_tokens_seen": 25312648,
      "step": 43640
    },
    {
      "epoch": 6.500595770032767,
      "grad_norm": 7.625,
      "learning_rate": 4.267629086314123e-05,
      "loss": 1.0396,
      "num_input_tokens_seen": 25315592,
      "step": 43645
    },
    {
      "epoch": 6.501340482573727,
      "grad_norm": 18.875,
      "learning_rate": 4.2673992846709574e-05,
      "loss": 1.3515,
      "num_input_tokens_seen": 25318568,
      "step": 43650
    },
    {
      "epoch": 6.502085195114685,
      "grad_norm": 16.5,
      "learning_rate": 4.267169453169377e-05,
      "loss": 1.088,
      "num_input_tokens_seen": 25321384,
      "step": 43655
    },
    {
      "epoch": 6.502829907655645,
      "grad_norm": 13.25,
      "learning_rate": 4.266939591813265e-05,
      "loss": 1.0208,
      "num_input_tokens_seen": 25324232,
      "step": 43660
    },
    {
      "epoch": 6.503574620196604,
      "grad_norm": 4.96875,
      "learning_rate": 4.266709700606504e-05,
      "loss": 0.9794,
      "num_input_tokens_seen": 25327208,
      "step": 43665
    },
    {
      "epoch": 6.5043193327375635,
      "grad_norm": 7.21875,
      "learning_rate": 4.266479779552979e-05,
      "loss": 1.3148,
      "num_input_tokens_seen": 25330024,
      "step": 43670
    },
    {
      "epoch": 6.505064045278522,
      "grad_norm": 6.1875,
      "learning_rate": 4.266249828656572e-05,
      "loss": 0.9525,
      "num_input_tokens_seen": 25332936,
      "step": 43675
    },
    {
      "epoch": 6.505808757819482,
      "grad_norm": 12.3125,
      "learning_rate": 4.2660198479211705e-05,
      "loss": 1.1548,
      "num_input_tokens_seen": 25335368,
      "step": 43680
    },
    {
      "epoch": 6.506553470360441,
      "grad_norm": 8.875,
      "learning_rate": 4.265789837350658e-05,
      "loss": 0.8136,
      "num_input_tokens_seen": 25338280,
      "step": 43685
    },
    {
      "epoch": 6.5072981829014,
      "grad_norm": 13.6875,
      "learning_rate": 4.2655597969489216e-05,
      "loss": 1.0169,
      "num_input_tokens_seen": 25341320,
      "step": 43690
    },
    {
      "epoch": 6.508042895442359,
      "grad_norm": 8.9375,
      "learning_rate": 4.265329726719845e-05,
      "loss": 1.1735,
      "num_input_tokens_seen": 25344168,
      "step": 43695
    },
    {
      "epoch": 6.508787607983319,
      "grad_norm": 7.1875,
      "learning_rate": 4.2650996266673197e-05,
      "loss": 0.9673,
      "num_input_tokens_seen": 25347048,
      "step": 43700
    },
    {
      "epoch": 6.509532320524277,
      "grad_norm": 11.625,
      "learning_rate": 4.264869496795229e-05,
      "loss": 0.9868,
      "num_input_tokens_seen": 25349960,
      "step": 43705
    },
    {
      "epoch": 6.510277033065237,
      "grad_norm": 9.4375,
      "learning_rate": 4.264639337107461e-05,
      "loss": 0.8393,
      "num_input_tokens_seen": 25352552,
      "step": 43710
    },
    {
      "epoch": 6.511021745606196,
      "grad_norm": 6.65625,
      "learning_rate": 4.264409147607905e-05,
      "loss": 1.0469,
      "num_input_tokens_seen": 25355624,
      "step": 43715
    },
    {
      "epoch": 6.5117664581471555,
      "grad_norm": 6.75,
      "learning_rate": 4.264178928300451e-05,
      "loss": 1.1145,
      "num_input_tokens_seen": 25358344,
      "step": 43720
    },
    {
      "epoch": 6.512511170688114,
      "grad_norm": 15.4375,
      "learning_rate": 4.263948679188986e-05,
      "loss": 1.1602,
      "num_input_tokens_seen": 25361032,
      "step": 43725
    },
    {
      "epoch": 6.513255883229074,
      "grad_norm": 15.125,
      "learning_rate": 4.263718400277401e-05,
      "loss": 1.4811,
      "num_input_tokens_seen": 25363944,
      "step": 43730
    },
    {
      "epoch": 6.514000595770033,
      "grad_norm": 14.75,
      "learning_rate": 4.263488091569586e-05,
      "loss": 0.8192,
      "num_input_tokens_seen": 25366760,
      "step": 43735
    },
    {
      "epoch": 6.514745308310992,
      "grad_norm": 14.0625,
      "learning_rate": 4.263257753069432e-05,
      "loss": 0.9189,
      "num_input_tokens_seen": 25369672,
      "step": 43740
    },
    {
      "epoch": 6.515490020851951,
      "grad_norm": 10.8125,
      "learning_rate": 4.263027384780831e-05,
      "loss": 0.9039,
      "num_input_tokens_seen": 25372520,
      "step": 43745
    },
    {
      "epoch": 6.516234733392911,
      "grad_norm": 11.875,
      "learning_rate": 4.2627969867076736e-05,
      "loss": 0.9475,
      "num_input_tokens_seen": 25375176,
      "step": 43750
    },
    {
      "epoch": 6.5169794459338695,
      "grad_norm": 7.125,
      "learning_rate": 4.2625665588538534e-05,
      "loss": 0.8942,
      "num_input_tokens_seen": 25378216,
      "step": 43755
    },
    {
      "epoch": 6.517724158474829,
      "grad_norm": 13.75,
      "learning_rate": 4.262336101223262e-05,
      "loss": 0.7884,
      "num_input_tokens_seen": 25381160,
      "step": 43760
    },
    {
      "epoch": 6.518468871015788,
      "grad_norm": 4.9375,
      "learning_rate": 4.2621056138197936e-05,
      "loss": 0.6689,
      "num_input_tokens_seen": 25383784,
      "step": 43765
    },
    {
      "epoch": 6.519213583556747,
      "grad_norm": 12.0,
      "learning_rate": 4.261875096647341e-05,
      "loss": 0.8892,
      "num_input_tokens_seen": 25386344,
      "step": 43770
    },
    {
      "epoch": 6.519958296097706,
      "grad_norm": 8.6875,
      "learning_rate": 4.2616445497098e-05,
      "loss": 1.1583,
      "num_input_tokens_seen": 25389448,
      "step": 43775
    },
    {
      "epoch": 6.520703008638666,
      "grad_norm": 10.9375,
      "learning_rate": 4.261413973011065e-05,
      "loss": 1.2815,
      "num_input_tokens_seen": 25392552,
      "step": 43780
    },
    {
      "epoch": 6.521447721179625,
      "grad_norm": 7.4375,
      "learning_rate": 4.261183366555032e-05,
      "loss": 1.2505,
      "num_input_tokens_seen": 25395400,
      "step": 43785
    },
    {
      "epoch": 6.522192433720583,
      "grad_norm": 10.4375,
      "learning_rate": 4.260952730345594e-05,
      "loss": 0.797,
      "num_input_tokens_seen": 25398216,
      "step": 43790
    },
    {
      "epoch": 6.522937146261543,
      "grad_norm": 9.625,
      "learning_rate": 4.260722064386651e-05,
      "loss": 1.0645,
      "num_input_tokens_seen": 25401256,
      "step": 43795
    },
    {
      "epoch": 6.523681858802503,
      "grad_norm": 10.8125,
      "learning_rate": 4.2604913686820966e-05,
      "loss": 0.8652,
      "num_input_tokens_seen": 25404008,
      "step": 43800
    },
    {
      "epoch": 6.5244265713434615,
      "grad_norm": 8.8125,
      "learning_rate": 4.260260643235831e-05,
      "loss": 0.9109,
      "num_input_tokens_seen": 25406824,
      "step": 43805
    },
    {
      "epoch": 6.52517128388442,
      "grad_norm": 9.8125,
      "learning_rate": 4.260029888051751e-05,
      "loss": 1.0339,
      "num_input_tokens_seen": 25409960,
      "step": 43810
    },
    {
      "epoch": 6.52591599642538,
      "grad_norm": 18.25,
      "learning_rate": 4.259799103133754e-05,
      "loss": 0.9748,
      "num_input_tokens_seen": 25412520,
      "step": 43815
    },
    {
      "epoch": 6.526660708966339,
      "grad_norm": 5.4375,
      "learning_rate": 4.25956828848574e-05,
      "loss": 1.0217,
      "num_input_tokens_seen": 25415528,
      "step": 43820
    },
    {
      "epoch": 6.527405421507298,
      "grad_norm": 11.875,
      "learning_rate": 4.259337444111609e-05,
      "loss": 1.019,
      "num_input_tokens_seen": 25418248,
      "step": 43825
    },
    {
      "epoch": 6.528150134048257,
      "grad_norm": 8.4375,
      "learning_rate": 4.259106570015259e-05,
      "loss": 1.0127,
      "num_input_tokens_seen": 25421256,
      "step": 43830
    },
    {
      "epoch": 6.528894846589217,
      "grad_norm": 8.125,
      "learning_rate": 4.2588756662005926e-05,
      "loss": 1.5496,
      "num_input_tokens_seen": 25423976,
      "step": 43835
    },
    {
      "epoch": 6.5296395591301755,
      "grad_norm": 9.125,
      "learning_rate": 4.258644732671508e-05,
      "loss": 1.0705,
      "num_input_tokens_seen": 25426568,
      "step": 43840
    },
    {
      "epoch": 6.530384271671135,
      "grad_norm": 9.75,
      "learning_rate": 4.258413769431908e-05,
      "loss": 1.3735,
      "num_input_tokens_seen": 25429416,
      "step": 43845
    },
    {
      "epoch": 6.531128984212094,
      "grad_norm": 15.375,
      "learning_rate": 4.2581827764856955e-05,
      "loss": 1.512,
      "num_input_tokens_seen": 25432232,
      "step": 43850
    },
    {
      "epoch": 6.5318736967530535,
      "grad_norm": 9.3125,
      "learning_rate": 4.257951753836772e-05,
      "loss": 0.8551,
      "num_input_tokens_seen": 25434888,
      "step": 43855
    },
    {
      "epoch": 6.532618409294012,
      "grad_norm": 16.5,
      "learning_rate": 4.2577207014890394e-05,
      "loss": 1.2875,
      "num_input_tokens_seen": 25437672,
      "step": 43860
    },
    {
      "epoch": 6.533363121834972,
      "grad_norm": 6.4375,
      "learning_rate": 4.2574896194464033e-05,
      "loss": 1.3085,
      "num_input_tokens_seen": 25440424,
      "step": 43865
    },
    {
      "epoch": 6.534107834375931,
      "grad_norm": 14.25,
      "learning_rate": 4.2572585077127654e-05,
      "loss": 1.1321,
      "num_input_tokens_seen": 25443752,
      "step": 43870
    },
    {
      "epoch": 6.53485254691689,
      "grad_norm": 6.25,
      "learning_rate": 4.2570273662920315e-05,
      "loss": 1.0593,
      "num_input_tokens_seen": 25446504,
      "step": 43875
    },
    {
      "epoch": 6.535597259457849,
      "grad_norm": 18.375,
      "learning_rate": 4.2567961951881053e-05,
      "loss": 1.1426,
      "num_input_tokens_seen": 25449448,
      "step": 43880
    },
    {
      "epoch": 6.536341971998809,
      "grad_norm": 7.75,
      "learning_rate": 4.256564994404893e-05,
      "loss": 1.2624,
      "num_input_tokens_seen": 25452488,
      "step": 43885
    },
    {
      "epoch": 6.5370866845397675,
      "grad_norm": 11.375,
      "learning_rate": 4.2563337639463005e-05,
      "loss": 1.0964,
      "num_input_tokens_seen": 25455528,
      "step": 43890
    },
    {
      "epoch": 6.537831397080727,
      "grad_norm": 10.0625,
      "learning_rate": 4.256102503816234e-05,
      "loss": 1.0394,
      "num_input_tokens_seen": 25458408,
      "step": 43895
    },
    {
      "epoch": 6.538576109621686,
      "grad_norm": 6.3125,
      "learning_rate": 4.255871214018601e-05,
      "loss": 1.1089,
      "num_input_tokens_seen": 25461064,
      "step": 43900
    },
    {
      "epoch": 6.5393208221626455,
      "grad_norm": 6.21875,
      "learning_rate": 4.255639894557309e-05,
      "loss": 1.3592,
      "num_input_tokens_seen": 25463944,
      "step": 43905
    },
    {
      "epoch": 6.540065534703604,
      "grad_norm": 16.625,
      "learning_rate": 4.255408545436264e-05,
      "loss": 1.4439,
      "num_input_tokens_seen": 25466696,
      "step": 43910
    },
    {
      "epoch": 6.540810247244564,
      "grad_norm": 9.1875,
      "learning_rate": 4.255177166659376e-05,
      "loss": 0.7359,
      "num_input_tokens_seen": 25469576,
      "step": 43915
    },
    {
      "epoch": 6.541554959785523,
      "grad_norm": 11.875,
      "learning_rate": 4.254945758230554e-05,
      "loss": 1.1454,
      "num_input_tokens_seen": 25472648,
      "step": 43920
    },
    {
      "epoch": 6.542299672326482,
      "grad_norm": 10.1875,
      "learning_rate": 4.254714320153708e-05,
      "loss": 1.1011,
      "num_input_tokens_seen": 25475752,
      "step": 43925
    },
    {
      "epoch": 6.543044384867441,
      "grad_norm": 18.75,
      "learning_rate": 4.254482852432745e-05,
      "loss": 1.3276,
      "num_input_tokens_seen": 25479016,
      "step": 43930
    },
    {
      "epoch": 6.5437890974084,
      "grad_norm": 14.4375,
      "learning_rate": 4.254251355071579e-05,
      "loss": 0.997,
      "num_input_tokens_seen": 25482216,
      "step": 43935
    },
    {
      "epoch": 6.5445338099493595,
      "grad_norm": 11.5625,
      "learning_rate": 4.254019828074118e-05,
      "loss": 1.1661,
      "num_input_tokens_seen": 25484968,
      "step": 43940
    },
    {
      "epoch": 6.545278522490319,
      "grad_norm": 6.3125,
      "learning_rate": 4.2537882714442756e-05,
      "loss": 0.9629,
      "num_input_tokens_seen": 25487880,
      "step": 43945
    },
    {
      "epoch": 6.546023235031278,
      "grad_norm": 10.8125,
      "learning_rate": 4.253556685185963e-05,
      "loss": 1.0932,
      "num_input_tokens_seen": 25490760,
      "step": 43950
    },
    {
      "epoch": 6.546767947572237,
      "grad_norm": 7.25,
      "learning_rate": 4.2533250693030924e-05,
      "loss": 1.1213,
      "num_input_tokens_seen": 25493512,
      "step": 43955
    },
    {
      "epoch": 6.547512660113196,
      "grad_norm": 5.03125,
      "learning_rate": 4.2530934237995756e-05,
      "loss": 0.8055,
      "num_input_tokens_seen": 25496680,
      "step": 43960
    },
    {
      "epoch": 6.548257372654156,
      "grad_norm": 13.1875,
      "learning_rate": 4.252861748679329e-05,
      "loss": 1.0587,
      "num_input_tokens_seen": 25499592,
      "step": 43965
    },
    {
      "epoch": 6.549002085195115,
      "grad_norm": 16.25,
      "learning_rate": 4.252630043946263e-05,
      "loss": 1.4592,
      "num_input_tokens_seen": 25502536,
      "step": 43970
    },
    {
      "epoch": 6.5497467977360735,
      "grad_norm": 11.8125,
      "learning_rate": 4.252398309604294e-05,
      "loss": 0.9271,
      "num_input_tokens_seen": 25505544,
      "step": 43975
    },
    {
      "epoch": 6.550491510277033,
      "grad_norm": 12.5,
      "learning_rate": 4.252166545657337e-05,
      "loss": 1.2582,
      "num_input_tokens_seen": 25508648,
      "step": 43980
    },
    {
      "epoch": 6.551236222817992,
      "grad_norm": 9.3125,
      "learning_rate": 4.2519347521093077e-05,
      "loss": 0.9699,
      "num_input_tokens_seen": 25511304,
      "step": 43985
    },
    {
      "epoch": 6.5519809353589515,
      "grad_norm": 10.0,
      "learning_rate": 4.251702928964121e-05,
      "loss": 0.994,
      "num_input_tokens_seen": 25514248,
      "step": 43990
    },
    {
      "epoch": 6.55272564789991,
      "grad_norm": 7.125,
      "learning_rate": 4.2514710762256925e-05,
      "loss": 1.0812,
      "num_input_tokens_seen": 25517288,
      "step": 43995
    },
    {
      "epoch": 6.55347036044087,
      "grad_norm": 5.625,
      "learning_rate": 4.2512391938979416e-05,
      "loss": 0.9561,
      "num_input_tokens_seen": 25520392,
      "step": 44000
    },
    {
      "epoch": 6.554215072981829,
      "grad_norm": 7.25,
      "learning_rate": 4.251007281984783e-05,
      "loss": 1.0942,
      "num_input_tokens_seen": 25523208,
      "step": 44005
    },
    {
      "epoch": 6.554959785522788,
      "grad_norm": 10.0,
      "learning_rate": 4.250775340490137e-05,
      "loss": 0.8923,
      "num_input_tokens_seen": 25525832,
      "step": 44010
    },
    {
      "epoch": 6.555704498063747,
      "grad_norm": 12.9375,
      "learning_rate": 4.2505433694179216e-05,
      "loss": 1.3272,
      "num_input_tokens_seen": 25528680,
      "step": 44015
    },
    {
      "epoch": 6.556449210604707,
      "grad_norm": 19.75,
      "learning_rate": 4.250311368772054e-05,
      "loss": 1.1679,
      "num_input_tokens_seen": 25531720,
      "step": 44020
    },
    {
      "epoch": 6.5571939231456655,
      "grad_norm": 8.4375,
      "learning_rate": 4.250079338556455e-05,
      "loss": 1.484,
      "num_input_tokens_seen": 25534504,
      "step": 44025
    },
    {
      "epoch": 6.557938635686625,
      "grad_norm": 8.875,
      "learning_rate": 4.2498472787750456e-05,
      "loss": 0.9103,
      "num_input_tokens_seen": 25537512,
      "step": 44030
    },
    {
      "epoch": 6.558683348227584,
      "grad_norm": 13.5,
      "learning_rate": 4.249615189431744e-05,
      "loss": 1.2383,
      "num_input_tokens_seen": 25540808,
      "step": 44035
    },
    {
      "epoch": 6.559428060768544,
      "grad_norm": 5.0,
      "learning_rate": 4.2493830705304716e-05,
      "loss": 1.0664,
      "num_input_tokens_seen": 25543880,
      "step": 44040
    },
    {
      "epoch": 6.560172773309502,
      "grad_norm": 7.625,
      "learning_rate": 4.24915092207515e-05,
      "loss": 1.1852,
      "num_input_tokens_seen": 25546568,
      "step": 44045
    },
    {
      "epoch": 6.560917485850462,
      "grad_norm": 13.75,
      "learning_rate": 4.248918744069702e-05,
      "loss": 1.1751,
      "num_input_tokens_seen": 25549384,
      "step": 44050
    },
    {
      "epoch": 6.561662198391421,
      "grad_norm": 15.375,
      "learning_rate": 4.2486865365180494e-05,
      "loss": 1.3243,
      "num_input_tokens_seen": 25552296,
      "step": 44055
    },
    {
      "epoch": 6.56240691093238,
      "grad_norm": 7.9375,
      "learning_rate": 4.2484542994241145e-05,
      "loss": 1.1065,
      "num_input_tokens_seen": 25555240,
      "step": 44060
    },
    {
      "epoch": 6.563151623473339,
      "grad_norm": 5.84375,
      "learning_rate": 4.2482220327918214e-05,
      "loss": 0.5726,
      "num_input_tokens_seen": 25558216,
      "step": 44065
    },
    {
      "epoch": 6.563896336014299,
      "grad_norm": 5.375,
      "learning_rate": 4.2479897366250946e-05,
      "loss": 0.9765,
      "num_input_tokens_seen": 25561192,
      "step": 44070
    },
    {
      "epoch": 6.5646410485552575,
      "grad_norm": 6.40625,
      "learning_rate": 4.247757410927857e-05,
      "loss": 0.8478,
      "num_input_tokens_seen": 25564296,
      "step": 44075
    },
    {
      "epoch": 6.565385761096217,
      "grad_norm": 8.9375,
      "learning_rate": 4.247525055704034e-05,
      "loss": 0.9112,
      "num_input_tokens_seen": 25567304,
      "step": 44080
    },
    {
      "epoch": 6.566130473637176,
      "grad_norm": 12.25,
      "learning_rate": 4.247292670957552e-05,
      "loss": 1.2057,
      "num_input_tokens_seen": 25570248,
      "step": 44085
    },
    {
      "epoch": 6.566875186178136,
      "grad_norm": 5.21875,
      "learning_rate": 4.247060256692336e-05,
      "loss": 1.0075,
      "num_input_tokens_seen": 25573384,
      "step": 44090
    },
    {
      "epoch": 6.567619898719094,
      "grad_norm": 7.96875,
      "learning_rate": 4.246827812912313e-05,
      "loss": 1.0406,
      "num_input_tokens_seen": 25576232,
      "step": 44095
    },
    {
      "epoch": 6.568364611260054,
      "grad_norm": 7.59375,
      "learning_rate": 4.246595339621409e-05,
      "loss": 0.7511,
      "num_input_tokens_seen": 25579112,
      "step": 44100
    },
    {
      "epoch": 6.569109323801013,
      "grad_norm": 11.75,
      "learning_rate": 4.246362836823551e-05,
      "loss": 1.4,
      "num_input_tokens_seen": 25581992,
      "step": 44105
    },
    {
      "epoch": 6.569854036341972,
      "grad_norm": 13.8125,
      "learning_rate": 4.2461303045226695e-05,
      "loss": 1.3713,
      "num_input_tokens_seen": 25584776,
      "step": 44110
    },
    {
      "epoch": 6.570598748882931,
      "grad_norm": 6.90625,
      "learning_rate": 4.24589774272269e-05,
      "loss": 0.8342,
      "num_input_tokens_seen": 25587720,
      "step": 44115
    },
    {
      "epoch": 6.57134346142389,
      "grad_norm": 20.625,
      "learning_rate": 4.245665151427544e-05,
      "loss": 1.2259,
      "num_input_tokens_seen": 25590600,
      "step": 44120
    },
    {
      "epoch": 6.57208817396485,
      "grad_norm": 7.03125,
      "learning_rate": 4.245432530641158e-05,
      "loss": 1.1,
      "num_input_tokens_seen": 25593608,
      "step": 44125
    },
    {
      "epoch": 6.572832886505809,
      "grad_norm": 12.375,
      "learning_rate": 4.245199880367464e-05,
      "loss": 1.2756,
      "num_input_tokens_seen": 25596488,
      "step": 44130
    },
    {
      "epoch": 6.573577599046768,
      "grad_norm": 7.1875,
      "learning_rate": 4.2449672006103914e-05,
      "loss": 1.0836,
      "num_input_tokens_seen": 25599464,
      "step": 44135
    },
    {
      "epoch": 6.574322311587727,
      "grad_norm": 8.8125,
      "learning_rate": 4.244734491373872e-05,
      "loss": 0.6194,
      "num_input_tokens_seen": 25602344,
      "step": 44140
    },
    {
      "epoch": 6.575067024128686,
      "grad_norm": 12.4375,
      "learning_rate": 4.244501752661836e-05,
      "loss": 0.843,
      "num_input_tokens_seen": 25605256,
      "step": 44145
    },
    {
      "epoch": 6.575811736669645,
      "grad_norm": 15.3125,
      "learning_rate": 4.244268984478216e-05,
      "loss": 1.3225,
      "num_input_tokens_seen": 25608136,
      "step": 44150
    },
    {
      "epoch": 6.576556449210605,
      "grad_norm": 8.3125,
      "learning_rate": 4.2440361868269453e-05,
      "loss": 1.0387,
      "num_input_tokens_seen": 25611016,
      "step": 44155
    },
    {
      "epoch": 6.5773011617515635,
      "grad_norm": 8.4375,
      "learning_rate": 4.243803359711954e-05,
      "loss": 1.2926,
      "num_input_tokens_seen": 25613768,
      "step": 44160
    },
    {
      "epoch": 6.578045874292523,
      "grad_norm": 8.5,
      "learning_rate": 4.243570503137179e-05,
      "loss": 0.78,
      "num_input_tokens_seen": 25616616,
      "step": 44165
    },
    {
      "epoch": 6.578790586833482,
      "grad_norm": 7.78125,
      "learning_rate": 4.2433376171065514e-05,
      "loss": 0.9336,
      "num_input_tokens_seen": 25619336,
      "step": 44170
    },
    {
      "epoch": 6.579535299374442,
      "grad_norm": 5.84375,
      "learning_rate": 4.2431047016240064e-05,
      "loss": 0.8139,
      "num_input_tokens_seen": 25622312,
      "step": 44175
    },
    {
      "epoch": 6.5802800119154,
      "grad_norm": 13.75,
      "learning_rate": 4.242871756693481e-05,
      "loss": 1.3438,
      "num_input_tokens_seen": 25625192,
      "step": 44180
    },
    {
      "epoch": 6.58102472445636,
      "grad_norm": 6.0,
      "learning_rate": 4.242638782318906e-05,
      "loss": 1.1185,
      "num_input_tokens_seen": 25627944,
      "step": 44185
    },
    {
      "epoch": 6.581769436997319,
      "grad_norm": 13.1875,
      "learning_rate": 4.242405778504221e-05,
      "loss": 1.468,
      "num_input_tokens_seen": 25630696,
      "step": 44190
    },
    {
      "epoch": 6.582514149538278,
      "grad_norm": 14.0625,
      "learning_rate": 4.242172745253362e-05,
      "loss": 1.2001,
      "num_input_tokens_seen": 25633736,
      "step": 44195
    },
    {
      "epoch": 6.583258862079237,
      "grad_norm": 10.1875,
      "learning_rate": 4.241939682570265e-05,
      "loss": 0.9861,
      "num_input_tokens_seen": 25636744,
      "step": 44200
    },
    {
      "epoch": 6.584003574620197,
      "grad_norm": 10.3125,
      "learning_rate": 4.241706590458867e-05,
      "loss": 1.142,
      "num_input_tokens_seen": 25639848,
      "step": 44205
    },
    {
      "epoch": 6.584748287161156,
      "grad_norm": 9.25,
      "learning_rate": 4.241473468923106e-05,
      "loss": 1.0997,
      "num_input_tokens_seen": 25642792,
      "step": 44210
    },
    {
      "epoch": 6.585492999702115,
      "grad_norm": 10.8125,
      "learning_rate": 4.2412403179669216e-05,
      "loss": 0.9906,
      "num_input_tokens_seen": 25646056,
      "step": 44215
    },
    {
      "epoch": 6.586237712243074,
      "grad_norm": 17.875,
      "learning_rate": 4.2410071375942505e-05,
      "loss": 0.7415,
      "num_input_tokens_seen": 25649128,
      "step": 44220
    },
    {
      "epoch": 6.586982424784034,
      "grad_norm": 4.40625,
      "learning_rate": 4.240773927809034e-05,
      "loss": 0.9942,
      "num_input_tokens_seen": 25651944,
      "step": 44225
    },
    {
      "epoch": 6.587727137324992,
      "grad_norm": 6.65625,
      "learning_rate": 4.240540688615212e-05,
      "loss": 0.9362,
      "num_input_tokens_seen": 25654568,
      "step": 44230
    },
    {
      "epoch": 6.588471849865952,
      "grad_norm": 6.125,
      "learning_rate": 4.240307420016724e-05,
      "loss": 1.2148,
      "num_input_tokens_seen": 25657576,
      "step": 44235
    },
    {
      "epoch": 6.589216562406911,
      "grad_norm": 11.1875,
      "learning_rate": 4.24007412201751e-05,
      "loss": 1.2724,
      "num_input_tokens_seen": 25660904,
      "step": 44240
    },
    {
      "epoch": 6.58996127494787,
      "grad_norm": 12.4375,
      "learning_rate": 4.239840794621512e-05,
      "loss": 1.5994,
      "num_input_tokens_seen": 25663816,
      "step": 44245
    },
    {
      "epoch": 6.590705987488829,
      "grad_norm": 8.6875,
      "learning_rate": 4.2396074378326725e-05,
      "loss": 0.7448,
      "num_input_tokens_seen": 25666664,
      "step": 44250
    },
    {
      "epoch": 6.591450700029789,
      "grad_norm": 10.125,
      "learning_rate": 4.239374051654934e-05,
      "loss": 1.042,
      "num_input_tokens_seen": 25669992,
      "step": 44255
    },
    {
      "epoch": 6.592195412570748,
      "grad_norm": 16.625,
      "learning_rate": 4.239140636092238e-05,
      "loss": 1.1823,
      "num_input_tokens_seen": 25672904,
      "step": 44260
    },
    {
      "epoch": 6.592940125111707,
      "grad_norm": 11.25,
      "learning_rate": 4.238907191148528e-05,
      "loss": 1.2706,
      "num_input_tokens_seen": 25675720,
      "step": 44265
    },
    {
      "epoch": 6.593684837652666,
      "grad_norm": 19.25,
      "learning_rate": 4.23867371682775e-05,
      "loss": 1.0887,
      "num_input_tokens_seen": 25678376,
      "step": 44270
    },
    {
      "epoch": 6.594429550193626,
      "grad_norm": 9.375,
      "learning_rate": 4.2384402131338455e-05,
      "loss": 1.1519,
      "num_input_tokens_seen": 25681256,
      "step": 44275
    },
    {
      "epoch": 6.595174262734584,
      "grad_norm": 13.75,
      "learning_rate": 4.2382066800707606e-05,
      "loss": 0.9661,
      "num_input_tokens_seen": 25683944,
      "step": 44280
    },
    {
      "epoch": 6.595918975275543,
      "grad_norm": 8.5,
      "learning_rate": 4.237973117642441e-05,
      "loss": 0.7847,
      "num_input_tokens_seen": 25686888,
      "step": 44285
    },
    {
      "epoch": 6.596663687816503,
      "grad_norm": 8.125,
      "learning_rate": 4.237739525852831e-05,
      "loss": 1.1457,
      "num_input_tokens_seen": 25689704,
      "step": 44290
    },
    {
      "epoch": 6.5974084003574625,
      "grad_norm": 10.5,
      "learning_rate": 4.237505904705879e-05,
      "loss": 0.8134,
      "num_input_tokens_seen": 25692424,
      "step": 44295
    },
    {
      "epoch": 6.598153112898421,
      "grad_norm": 6.53125,
      "learning_rate": 4.23727225420553e-05,
      "loss": 0.8949,
      "num_input_tokens_seen": 25695368,
      "step": 44300
    },
    {
      "epoch": 6.59889782543938,
      "grad_norm": 5.5,
      "learning_rate": 4.237038574355732e-05,
      "loss": 1.3214,
      "num_input_tokens_seen": 25698440,
      "step": 44305
    },
    {
      "epoch": 6.59964253798034,
      "grad_norm": 5.9375,
      "learning_rate": 4.236804865160433e-05,
      "loss": 1.0586,
      "num_input_tokens_seen": 25701160,
      "step": 44310
    },
    {
      "epoch": 6.600387250521299,
      "grad_norm": 8.875,
      "learning_rate": 4.236571126623581e-05,
      "loss": 0.8251,
      "num_input_tokens_seen": 25704200,
      "step": 44315
    },
    {
      "epoch": 6.601131963062258,
      "grad_norm": 14.0,
      "learning_rate": 4.236337358749124e-05,
      "loss": 1.4751,
      "num_input_tokens_seen": 25707016,
      "step": 44320
    },
    {
      "epoch": 6.601876675603217,
      "grad_norm": 8.6875,
      "learning_rate": 4.2361035615410127e-05,
      "loss": 1.1582,
      "num_input_tokens_seen": 25709832,
      "step": 44325
    },
    {
      "epoch": 6.602621388144176,
      "grad_norm": 8.5625,
      "learning_rate": 4.2358697350031964e-05,
      "loss": 1.3309,
      "num_input_tokens_seen": 25712520,
      "step": 44330
    },
    {
      "epoch": 6.603366100685135,
      "grad_norm": 7.4375,
      "learning_rate": 4.2356358791396244e-05,
      "loss": 1.0518,
      "num_input_tokens_seen": 25715144,
      "step": 44335
    },
    {
      "epoch": 6.604110813226095,
      "grad_norm": 5.15625,
      "learning_rate": 4.235401993954249e-05,
      "loss": 1.1203,
      "num_input_tokens_seen": 25717736,
      "step": 44340
    },
    {
      "epoch": 6.604855525767054,
      "grad_norm": 9.3125,
      "learning_rate": 4.2351680794510205e-05,
      "loss": 1.0923,
      "num_input_tokens_seen": 25720744,
      "step": 44345
    },
    {
      "epoch": 6.605600238308013,
      "grad_norm": 7.6875,
      "learning_rate": 4.234934135633891e-05,
      "loss": 1.2356,
      "num_input_tokens_seen": 25723560,
      "step": 44350
    },
    {
      "epoch": 6.606344950848972,
      "grad_norm": 6.65625,
      "learning_rate": 4.234700162506813e-05,
      "loss": 1.3107,
      "num_input_tokens_seen": 25726120,
      "step": 44355
    },
    {
      "epoch": 6.607089663389932,
      "grad_norm": 9.75,
      "learning_rate": 4.234466160073738e-05,
      "loss": 0.6718,
      "num_input_tokens_seen": 25728936,
      "step": 44360
    },
    {
      "epoch": 6.60783437593089,
      "grad_norm": 14.5625,
      "learning_rate": 4.23423212833862e-05,
      "loss": 1.0687,
      "num_input_tokens_seen": 25731912,
      "step": 44365
    },
    {
      "epoch": 6.60857908847185,
      "grad_norm": 12.1875,
      "learning_rate": 4.233998067305413e-05,
      "loss": 0.9385,
      "num_input_tokens_seen": 25734600,
      "step": 44370
    },
    {
      "epoch": 6.609323801012809,
      "grad_norm": 16.25,
      "learning_rate": 4.233763976978071e-05,
      "loss": 0.8554,
      "num_input_tokens_seen": 25737544,
      "step": 44375
    },
    {
      "epoch": 6.6100685135537685,
      "grad_norm": 10.125,
      "learning_rate": 4.233529857360549e-05,
      "loss": 1.0715,
      "num_input_tokens_seen": 25740328,
      "step": 44380
    },
    {
      "epoch": 6.610813226094727,
      "grad_norm": 7.6875,
      "learning_rate": 4.233295708456801e-05,
      "loss": 1.1827,
      "num_input_tokens_seen": 25743304,
      "step": 44385
    },
    {
      "epoch": 6.611557938635687,
      "grad_norm": 16.125,
      "learning_rate": 4.2330615302707856e-05,
      "loss": 1.2209,
      "num_input_tokens_seen": 25746088,
      "step": 44390
    },
    {
      "epoch": 6.612302651176646,
      "grad_norm": 8.75,
      "learning_rate": 4.2328273228064555e-05,
      "loss": 1.1538,
      "num_input_tokens_seen": 25749032,
      "step": 44395
    },
    {
      "epoch": 6.613047363717605,
      "grad_norm": 8.5,
      "learning_rate": 4.2325930860677695e-05,
      "loss": 1.0965,
      "num_input_tokens_seen": 25752072,
      "step": 44400
    },
    {
      "epoch": 6.613792076258564,
      "grad_norm": 8.5625,
      "learning_rate": 4.232358820058684e-05,
      "loss": 0.85,
      "num_input_tokens_seen": 25754664,
      "step": 44405
    },
    {
      "epoch": 6.614536788799524,
      "grad_norm": 10.3125,
      "learning_rate": 4.232124524783157e-05,
      "loss": 1.1952,
      "num_input_tokens_seen": 25757384,
      "step": 44410
    },
    {
      "epoch": 6.615281501340482,
      "grad_norm": 11.75,
      "learning_rate": 4.231890200245147e-05,
      "loss": 1.1072,
      "num_input_tokens_seen": 25760072,
      "step": 44415
    },
    {
      "epoch": 6.616026213881442,
      "grad_norm": 11.625,
      "learning_rate": 4.231655846448611e-05,
      "loss": 1.256,
      "num_input_tokens_seen": 25762728,
      "step": 44420
    },
    {
      "epoch": 6.616770926422401,
      "grad_norm": 14.0625,
      "learning_rate": 4.2314214633975105e-05,
      "loss": 1.1372,
      "num_input_tokens_seen": 25765768,
      "step": 44425
    },
    {
      "epoch": 6.6175156389633605,
      "grad_norm": 5.03125,
      "learning_rate": 4.231187051095804e-05,
      "loss": 1.0957,
      "num_input_tokens_seen": 25768424,
      "step": 44430
    },
    {
      "epoch": 6.618260351504319,
      "grad_norm": 10.5,
      "learning_rate": 4.2309526095474514e-05,
      "loss": 1.2848,
      "num_input_tokens_seen": 25771304,
      "step": 44435
    },
    {
      "epoch": 6.619005064045279,
      "grad_norm": 9.4375,
      "learning_rate": 4.230718138756414e-05,
      "loss": 0.9646,
      "num_input_tokens_seen": 25774056,
      "step": 44440
    },
    {
      "epoch": 6.619749776586238,
      "grad_norm": 11.25,
      "learning_rate": 4.2304836387266534e-05,
      "loss": 1.1479,
      "num_input_tokens_seen": 25777224,
      "step": 44445
    },
    {
      "epoch": 6.620494489127196,
      "grad_norm": 5.84375,
      "learning_rate": 4.230249109462129e-05,
      "loss": 1.1716,
      "num_input_tokens_seen": 25780072,
      "step": 44450
    },
    {
      "epoch": 6.621239201668156,
      "grad_norm": 13.3125,
      "learning_rate": 4.2300145509668054e-05,
      "loss": 1.4082,
      "num_input_tokens_seen": 25783080,
      "step": 44455
    },
    {
      "epoch": 6.621983914209116,
      "grad_norm": 15.75,
      "learning_rate": 4.2297799632446444e-05,
      "loss": 1.4662,
      "num_input_tokens_seen": 25786120,
      "step": 44460
    },
    {
      "epoch": 6.6227286267500745,
      "grad_norm": 6.65625,
      "learning_rate": 4.229545346299609e-05,
      "loss": 1.0386,
      "num_input_tokens_seen": 25789096,
      "step": 44465
    },
    {
      "epoch": 6.623473339291033,
      "grad_norm": 6.0,
      "learning_rate": 4.2293107001356624e-05,
      "loss": 1.2941,
      "num_input_tokens_seen": 25792008,
      "step": 44470
    },
    {
      "epoch": 6.624218051831993,
      "grad_norm": 7.46875,
      "learning_rate": 4.2290760247567695e-05,
      "loss": 1.1201,
      "num_input_tokens_seen": 25794632,
      "step": 44475
    },
    {
      "epoch": 6.6249627643729525,
      "grad_norm": 12.4375,
      "learning_rate": 4.2288413201668945e-05,
      "loss": 1.1202,
      "num_input_tokens_seen": 25797384,
      "step": 44480
    },
    {
      "epoch": 6.625707476913911,
      "grad_norm": 7.09375,
      "learning_rate": 4.228606586370002e-05,
      "loss": 1.21,
      "num_input_tokens_seen": 25800200,
      "step": 44485
    },
    {
      "epoch": 6.62645218945487,
      "grad_norm": 8.0,
      "learning_rate": 4.228371823370058e-05,
      "loss": 1.1267,
      "num_input_tokens_seen": 25802760,
      "step": 44490
    },
    {
      "epoch": 6.62719690199583,
      "grad_norm": 9.5,
      "learning_rate": 4.228137031171029e-05,
      "loss": 1.0331,
      "num_input_tokens_seen": 25805512,
      "step": 44495
    },
    {
      "epoch": 6.627941614536788,
      "grad_norm": 14.75,
      "learning_rate": 4.227902209776881e-05,
      "loss": 1.2458,
      "num_input_tokens_seen": 25808584,
      "step": 44500
    },
    {
      "epoch": 6.628686327077748,
      "grad_norm": 11.0625,
      "learning_rate": 4.227667359191582e-05,
      "loss": 0.8802,
      "num_input_tokens_seen": 25811592,
      "step": 44505
    },
    {
      "epoch": 6.629431039618707,
      "grad_norm": 11.5625,
      "learning_rate": 4.227432479419099e-05,
      "loss": 1.11,
      "num_input_tokens_seen": 25814280,
      "step": 44510
    },
    {
      "epoch": 6.6301757521596665,
      "grad_norm": 6.1875,
      "learning_rate": 4.227197570463399e-05,
      "loss": 1.0495,
      "num_input_tokens_seen": 25817352,
      "step": 44515
    },
    {
      "epoch": 6.630920464700625,
      "grad_norm": 13.0625,
      "learning_rate": 4.2269626323284514e-05,
      "loss": 1.337,
      "num_input_tokens_seen": 25820328,
      "step": 44520
    },
    {
      "epoch": 6.631665177241585,
      "grad_norm": 13.625,
      "learning_rate": 4.226727665018226e-05,
      "loss": 0.9961,
      "num_input_tokens_seen": 25823304,
      "step": 44525
    },
    {
      "epoch": 6.632409889782544,
      "grad_norm": 9.1875,
      "learning_rate": 4.2264926685366925e-05,
      "loss": 1.1377,
      "num_input_tokens_seen": 25826440,
      "step": 44530
    },
    {
      "epoch": 6.633154602323503,
      "grad_norm": 12.25,
      "learning_rate": 4.2262576428878184e-05,
      "loss": 1.0792,
      "num_input_tokens_seen": 25829096,
      "step": 44535
    },
    {
      "epoch": 6.633899314864462,
      "grad_norm": 18.5,
      "learning_rate": 4.226022588075577e-05,
      "loss": 1.0927,
      "num_input_tokens_seen": 25832040,
      "step": 44540
    },
    {
      "epoch": 6.634644027405422,
      "grad_norm": 9.875,
      "learning_rate": 4.2257875041039375e-05,
      "loss": 1.1691,
      "num_input_tokens_seen": 25835080,
      "step": 44545
    },
    {
      "epoch": 6.6353887399463805,
      "grad_norm": 10.0,
      "learning_rate": 4.225552390976873e-05,
      "loss": 0.9329,
      "num_input_tokens_seen": 25838408,
      "step": 44550
    },
    {
      "epoch": 6.63613345248734,
      "grad_norm": 8.5625,
      "learning_rate": 4.225317248698354e-05,
      "loss": 0.9624,
      "num_input_tokens_seen": 25841352,
      "step": 44555
    },
    {
      "epoch": 6.636878165028299,
      "grad_norm": 14.1875,
      "learning_rate": 4.225082077272354e-05,
      "loss": 1.6489,
      "num_input_tokens_seen": 25844232,
      "step": 44560
    },
    {
      "epoch": 6.6376228775692585,
      "grad_norm": 7.65625,
      "learning_rate": 4.224846876702845e-05,
      "loss": 1.1848,
      "num_input_tokens_seen": 25847112,
      "step": 44565
    },
    {
      "epoch": 6.638367590110217,
      "grad_norm": 5.9375,
      "learning_rate": 4.224611646993801e-05,
      "loss": 0.8773,
      "num_input_tokens_seen": 25850184,
      "step": 44570
    },
    {
      "epoch": 6.639112302651177,
      "grad_norm": 13.0625,
      "learning_rate": 4.224376388149197e-05,
      "loss": 0.9745,
      "num_input_tokens_seen": 25852968,
      "step": 44575
    },
    {
      "epoch": 6.639857015192136,
      "grad_norm": 8.0625,
      "learning_rate": 4.2241411001730057e-05,
      "loss": 0.9745,
      "num_input_tokens_seen": 25856136,
      "step": 44580
    },
    {
      "epoch": 6.640601727733095,
      "grad_norm": 10.625,
      "learning_rate": 4.223905783069203e-05,
      "loss": 0.8458,
      "num_input_tokens_seen": 25859208,
      "step": 44585
    },
    {
      "epoch": 6.641346440274054,
      "grad_norm": 14.25,
      "learning_rate": 4.2236704368417644e-05,
      "loss": 1.5131,
      "num_input_tokens_seen": 25862408,
      "step": 44590
    },
    {
      "epoch": 6.642091152815014,
      "grad_norm": 13.8125,
      "learning_rate": 4.223435061494666e-05,
      "loss": 1.0964,
      "num_input_tokens_seen": 25865416,
      "step": 44595
    },
    {
      "epoch": 6.6428358653559725,
      "grad_norm": 12.3125,
      "learning_rate": 4.223199657031883e-05,
      "loss": 1.0763,
      "num_input_tokens_seen": 25868456,
      "step": 44600
    },
    {
      "epoch": 6.643580577896932,
      "grad_norm": 13.75,
      "learning_rate": 4.222964223457394e-05,
      "loss": 1.2294,
      "num_input_tokens_seen": 25871144,
      "step": 44605
    },
    {
      "epoch": 6.644325290437891,
      "grad_norm": 10.875,
      "learning_rate": 4.2227287607751756e-05,
      "loss": 1.5021,
      "num_input_tokens_seen": 25874024,
      "step": 44610
    },
    {
      "epoch": 6.6450700029788505,
      "grad_norm": 7.375,
      "learning_rate": 4.222493268989205e-05,
      "loss": 1.2344,
      "num_input_tokens_seen": 25877000,
      "step": 44615
    },
    {
      "epoch": 6.645814715519809,
      "grad_norm": 6.625,
      "learning_rate": 4.222257748103461e-05,
      "loss": 1.1196,
      "num_input_tokens_seen": 25879624,
      "step": 44620
    },
    {
      "epoch": 6.646559428060769,
      "grad_norm": 10.5,
      "learning_rate": 4.222022198121923e-05,
      "loss": 1.3425,
      "num_input_tokens_seen": 25882376,
      "step": 44625
    },
    {
      "epoch": 6.647304140601728,
      "grad_norm": 9.8125,
      "learning_rate": 4.221786619048571e-05,
      "loss": 0.986,
      "num_input_tokens_seen": 25885128,
      "step": 44630
    },
    {
      "epoch": 6.6480488531426865,
      "grad_norm": 9.4375,
      "learning_rate": 4.221551010887384e-05,
      "loss": 1.4323,
      "num_input_tokens_seen": 25888200,
      "step": 44635
    },
    {
      "epoch": 6.648793565683646,
      "grad_norm": 8.625,
      "learning_rate": 4.2213153736423417e-05,
      "loss": 0.9353,
      "num_input_tokens_seen": 25890888,
      "step": 44640
    },
    {
      "epoch": 6.649538278224606,
      "grad_norm": 11.8125,
      "learning_rate": 4.221079707317426e-05,
      "loss": 0.8975,
      "num_input_tokens_seen": 25893640,
      "step": 44645
    },
    {
      "epoch": 6.6502829907655645,
      "grad_norm": 13.5,
      "learning_rate": 4.220844011916617e-05,
      "loss": 1.2595,
      "num_input_tokens_seen": 25896392,
      "step": 44650
    },
    {
      "epoch": 6.651027703306523,
      "grad_norm": 15.0,
      "learning_rate": 4.2206082874438976e-05,
      "loss": 0.8081,
      "num_input_tokens_seen": 25899400,
      "step": 44655
    },
    {
      "epoch": 6.651772415847483,
      "grad_norm": 13.0625,
      "learning_rate": 4.2203725339032505e-05,
      "loss": 1.2538,
      "num_input_tokens_seen": 25902216,
      "step": 44660
    },
    {
      "epoch": 6.652517128388443,
      "grad_norm": 10.3125,
      "learning_rate": 4.220136751298659e-05,
      "loss": 1.1574,
      "num_input_tokens_seen": 25905256,
      "step": 44665
    },
    {
      "epoch": 6.653261840929401,
      "grad_norm": 16.375,
      "learning_rate": 4.219900939634103e-05,
      "loss": 1.1919,
      "num_input_tokens_seen": 25908168,
      "step": 44670
    },
    {
      "epoch": 6.65400655347036,
      "grad_norm": 6.375,
      "learning_rate": 4.2196650989135706e-05,
      "loss": 0.9915,
      "num_input_tokens_seen": 25911080,
      "step": 44675
    },
    {
      "epoch": 6.65475126601132,
      "grad_norm": 17.375,
      "learning_rate": 4.219429229141043e-05,
      "loss": 0.9498,
      "num_input_tokens_seen": 25913928,
      "step": 44680
    },
    {
      "epoch": 6.6554959785522785,
      "grad_norm": 7.375,
      "learning_rate": 4.219193330320507e-05,
      "loss": 1.0033,
      "num_input_tokens_seen": 25916712,
      "step": 44685
    },
    {
      "epoch": 6.656240691093238,
      "grad_norm": 11.75,
      "learning_rate": 4.2189574024559465e-05,
      "loss": 1.243,
      "num_input_tokens_seen": 25919656,
      "step": 44690
    },
    {
      "epoch": 6.656985403634197,
      "grad_norm": 6.75,
      "learning_rate": 4.218721445551348e-05,
      "loss": 1.0566,
      "num_input_tokens_seen": 25922664,
      "step": 44695
    },
    {
      "epoch": 6.6577301161751565,
      "grad_norm": 7.71875,
      "learning_rate": 4.218485459610697e-05,
      "loss": 0.9864,
      "num_input_tokens_seen": 25925512,
      "step": 44700
    },
    {
      "epoch": 6.658474828716115,
      "grad_norm": 7.96875,
      "learning_rate": 4.2182494446379805e-05,
      "loss": 1.0988,
      "num_input_tokens_seen": 25928712,
      "step": 44705
    },
    {
      "epoch": 6.659219541257075,
      "grad_norm": 11.375,
      "learning_rate": 4.218013400637187e-05,
      "loss": 1.1329,
      "num_input_tokens_seen": 25931688,
      "step": 44710
    },
    {
      "epoch": 6.659964253798034,
      "grad_norm": 6.125,
      "learning_rate": 4.217777327612303e-05,
      "loss": 0.9818,
      "num_input_tokens_seen": 25934504,
      "step": 44715
    },
    {
      "epoch": 6.660708966338993,
      "grad_norm": 7.0,
      "learning_rate": 4.2175412255673164e-05,
      "loss": 1.2616,
      "num_input_tokens_seen": 25937224,
      "step": 44720
    },
    {
      "epoch": 6.661453678879952,
      "grad_norm": 7.21875,
      "learning_rate": 4.2173050945062165e-05,
      "loss": 0.9817,
      "num_input_tokens_seen": 25939976,
      "step": 44725
    },
    {
      "epoch": 6.662198391420912,
      "grad_norm": 8.5,
      "learning_rate": 4.217068934432993e-05,
      "loss": 0.8138,
      "num_input_tokens_seen": 25943144,
      "step": 44730
    },
    {
      "epoch": 6.6629431039618705,
      "grad_norm": 5.9375,
      "learning_rate": 4.216832745351634e-05,
      "loss": 0.8379,
      "num_input_tokens_seen": 25946152,
      "step": 44735
    },
    {
      "epoch": 6.66368781650283,
      "grad_norm": 8.8125,
      "learning_rate": 4.2165965272661315e-05,
      "loss": 1.0056,
      "num_input_tokens_seen": 25949000,
      "step": 44740
    },
    {
      "epoch": 6.664432529043789,
      "grad_norm": 7.34375,
      "learning_rate": 4.2163602801804745e-05,
      "loss": 0.9024,
      "num_input_tokens_seen": 25951624,
      "step": 44745
    },
    {
      "epoch": 6.665177241584749,
      "grad_norm": 8.8125,
      "learning_rate": 4.216124004098656e-05,
      "loss": 1.0018,
      "num_input_tokens_seen": 25954600,
      "step": 44750
    },
    {
      "epoch": 6.665921954125707,
      "grad_norm": 5.1875,
      "learning_rate": 4.2158876990246664e-05,
      "loss": 0.7642,
      "num_input_tokens_seen": 25957544,
      "step": 44755
    },
    {
      "epoch": 6.666666666666667,
      "grad_norm": 9.1875,
      "learning_rate": 4.215651364962498e-05,
      "loss": 1.0513,
      "num_input_tokens_seen": 25960328,
      "step": 44760
    },
    {
      "epoch": 6.667411379207626,
      "grad_norm": 9.0625,
      "learning_rate": 4.215415001916144e-05,
      "loss": 0.9889,
      "num_input_tokens_seen": 25963272,
      "step": 44765
    },
    {
      "epoch": 6.668156091748585,
      "grad_norm": 7.5,
      "learning_rate": 4.215178609889596e-05,
      "loss": 1.086,
      "num_input_tokens_seen": 25966216,
      "step": 44770
    },
    {
      "epoch": 6.668900804289544,
      "grad_norm": 8.9375,
      "learning_rate": 4.214942188886849e-05,
      "loss": 1.3175,
      "num_input_tokens_seen": 25968904,
      "step": 44775
    },
    {
      "epoch": 6.669645516830504,
      "grad_norm": 8.5,
      "learning_rate": 4.2147057389118964e-05,
      "loss": 0.743,
      "num_input_tokens_seen": 25971944,
      "step": 44780
    },
    {
      "epoch": 6.6703902293714625,
      "grad_norm": 11.6875,
      "learning_rate": 4.2144692599687334e-05,
      "loss": 0.901,
      "num_input_tokens_seen": 25974856,
      "step": 44785
    },
    {
      "epoch": 6.671134941912422,
      "grad_norm": 15.625,
      "learning_rate": 4.214232752061355e-05,
      "loss": 1.3666,
      "num_input_tokens_seen": 25977768,
      "step": 44790
    },
    {
      "epoch": 6.671879654453381,
      "grad_norm": 9.75,
      "learning_rate": 4.213996215193756e-05,
      "loss": 1.0238,
      "num_input_tokens_seen": 25980808,
      "step": 44795
    },
    {
      "epoch": 6.67262436699434,
      "grad_norm": 4.75,
      "learning_rate": 4.213759649369934e-05,
      "loss": 1.2077,
      "num_input_tokens_seen": 25983624,
      "step": 44800
    },
    {
      "epoch": 6.673369079535299,
      "grad_norm": 11.0625,
      "learning_rate": 4.2135230545938835e-05,
      "loss": 1.2673,
      "num_input_tokens_seen": 25986568,
      "step": 44805
    },
    {
      "epoch": 6.674113792076259,
      "grad_norm": 9.3125,
      "learning_rate": 4.213286430869603e-05,
      "loss": 0.8321,
      "num_input_tokens_seen": 25989576,
      "step": 44810
    },
    {
      "epoch": 6.674858504617218,
      "grad_norm": 7.9375,
      "learning_rate": 4.2130497782010894e-05,
      "loss": 1.2323,
      "num_input_tokens_seen": 25992040,
      "step": 44815
    },
    {
      "epoch": 6.6756032171581765,
      "grad_norm": 10.25,
      "learning_rate": 4.212813096592341e-05,
      "loss": 1.3196,
      "num_input_tokens_seen": 25994824,
      "step": 44820
    },
    {
      "epoch": 6.676347929699136,
      "grad_norm": 11.125,
      "learning_rate": 4.212576386047356e-05,
      "loss": 1.4251,
      "num_input_tokens_seen": 25997800,
      "step": 44825
    },
    {
      "epoch": 6.677092642240096,
      "grad_norm": 10.0625,
      "learning_rate": 4.2123396465701336e-05,
      "loss": 1.1188,
      "num_input_tokens_seen": 26000904,
      "step": 44830
    },
    {
      "epoch": 6.677837354781055,
      "grad_norm": 9.0,
      "learning_rate": 4.212102878164673e-05,
      "loss": 0.9913,
      "num_input_tokens_seen": 26003720,
      "step": 44835
    },
    {
      "epoch": 6.678582067322013,
      "grad_norm": 15.125,
      "learning_rate": 4.211866080834975e-05,
      "loss": 0.9398,
      "num_input_tokens_seen": 26006376,
      "step": 44840
    },
    {
      "epoch": 6.679326779862973,
      "grad_norm": 13.0625,
      "learning_rate": 4.2116292545850386e-05,
      "loss": 1.1388,
      "num_input_tokens_seen": 26009224,
      "step": 44845
    },
    {
      "epoch": 6.680071492403932,
      "grad_norm": 21.25,
      "learning_rate": 4.2113923994188665e-05,
      "loss": 1.3038,
      "num_input_tokens_seen": 26012360,
      "step": 44850
    },
    {
      "epoch": 6.680816204944891,
      "grad_norm": 10.75,
      "learning_rate": 4.211155515340458e-05,
      "loss": 1.2399,
      "num_input_tokens_seen": 26015432,
      "step": 44855
    },
    {
      "epoch": 6.68156091748585,
      "grad_norm": 5.4375,
      "learning_rate": 4.210918602353817e-05,
      "loss": 0.9337,
      "num_input_tokens_seen": 26018408,
      "step": 44860
    },
    {
      "epoch": 6.68230563002681,
      "grad_norm": 13.625,
      "learning_rate": 4.210681660462945e-05,
      "loss": 1.3579,
      "num_input_tokens_seen": 26021288,
      "step": 44865
    },
    {
      "epoch": 6.6830503425677685,
      "grad_norm": 23.375,
      "learning_rate": 4.210444689671845e-05,
      "loss": 1.36,
      "num_input_tokens_seen": 26024680,
      "step": 44870
    },
    {
      "epoch": 6.683795055108728,
      "grad_norm": 9.1875,
      "learning_rate": 4.2102076899845207e-05,
      "loss": 0.9788,
      "num_input_tokens_seen": 26027400,
      "step": 44875
    },
    {
      "epoch": 6.684539767649687,
      "grad_norm": 7.65625,
      "learning_rate": 4.209970661404975e-05,
      "loss": 0.7002,
      "num_input_tokens_seen": 26030632,
      "step": 44880
    },
    {
      "epoch": 6.685284480190647,
      "grad_norm": 11.875,
      "learning_rate": 4.209733603937214e-05,
      "loss": 1.0316,
      "num_input_tokens_seen": 26033320,
      "step": 44885
    },
    {
      "epoch": 6.686029192731605,
      "grad_norm": 8.0625,
      "learning_rate": 4.2094965175852395e-05,
      "loss": 1.079,
      "num_input_tokens_seen": 26036328,
      "step": 44890
    },
    {
      "epoch": 6.686773905272565,
      "grad_norm": 7.90625,
      "learning_rate": 4.209259402353061e-05,
      "loss": 1.0516,
      "num_input_tokens_seen": 26039080,
      "step": 44895
    },
    {
      "epoch": 6.687518617813524,
      "grad_norm": 9.9375,
      "learning_rate": 4.20902225824468e-05,
      "loss": 0.9889,
      "num_input_tokens_seen": 26042184,
      "step": 44900
    },
    {
      "epoch": 6.688263330354483,
      "grad_norm": 17.625,
      "learning_rate": 4.208785085264106e-05,
      "loss": 1.1252,
      "num_input_tokens_seen": 26044872,
      "step": 44905
    },
    {
      "epoch": 6.689008042895442,
      "grad_norm": 10.6875,
      "learning_rate": 4.2085478834153454e-05,
      "loss": 0.9036,
      "num_input_tokens_seen": 26047880,
      "step": 44910
    },
    {
      "epoch": 6.689752755436402,
      "grad_norm": 10.1875,
      "learning_rate": 4.208310652702404e-05,
      "loss": 0.9955,
      "num_input_tokens_seen": 26050728,
      "step": 44915
    },
    {
      "epoch": 6.690497467977361,
      "grad_norm": 11.25,
      "learning_rate": 4.208073393129291e-05,
      "loss": 1.0418,
      "num_input_tokens_seen": 26053576,
      "step": 44920
    },
    {
      "epoch": 6.69124218051832,
      "grad_norm": 10.0625,
      "learning_rate": 4.207836104700013e-05,
      "loss": 0.9615,
      "num_input_tokens_seen": 26056488,
      "step": 44925
    },
    {
      "epoch": 6.691986893059279,
      "grad_norm": 11.0,
      "learning_rate": 4.207598787418581e-05,
      "loss": 1.0756,
      "num_input_tokens_seen": 26059816,
      "step": 44930
    },
    {
      "epoch": 6.692731605600239,
      "grad_norm": 7.03125,
      "learning_rate": 4.207361441289002e-05,
      "loss": 1.0968,
      "num_input_tokens_seen": 26062920,
      "step": 44935
    },
    {
      "epoch": 6.693476318141197,
      "grad_norm": 4.96875,
      "learning_rate": 4.207124066315287e-05,
      "loss": 0.8009,
      "num_input_tokens_seen": 26066056,
      "step": 44940
    },
    {
      "epoch": 6.694221030682157,
      "grad_norm": 17.125,
      "learning_rate": 4.206886662501446e-05,
      "loss": 0.962,
      "num_input_tokens_seen": 26068776,
      "step": 44945
    },
    {
      "epoch": 6.694965743223116,
      "grad_norm": 14.1875,
      "learning_rate": 4.2066492298514895e-05,
      "loss": 1.4493,
      "num_input_tokens_seen": 26071688,
      "step": 44950
    },
    {
      "epoch": 6.695710455764075,
      "grad_norm": 10.5,
      "learning_rate": 4.2064117683694294e-05,
      "loss": 1.189,
      "num_input_tokens_seen": 26074536,
      "step": 44955
    },
    {
      "epoch": 6.696455168305034,
      "grad_norm": 9.75,
      "learning_rate": 4.206174278059276e-05,
      "loss": 1.2723,
      "num_input_tokens_seen": 26077384,
      "step": 44960
    },
    {
      "epoch": 6.697199880845994,
      "grad_norm": 12.375,
      "learning_rate": 4.205936758925043e-05,
      "loss": 1.0952,
      "num_input_tokens_seen": 26079944,
      "step": 44965
    },
    {
      "epoch": 6.697944593386953,
      "grad_norm": 12.8125,
      "learning_rate": 4.2056992109707415e-05,
      "loss": 1.4811,
      "num_input_tokens_seen": 26082824,
      "step": 44970
    },
    {
      "epoch": 6.698689305927912,
      "grad_norm": 7.5,
      "learning_rate": 4.205461634200386e-05,
      "loss": 1.106,
      "num_input_tokens_seen": 26085416,
      "step": 44975
    },
    {
      "epoch": 6.699434018468871,
      "grad_norm": 8.75,
      "learning_rate": 4.2052240286179886e-05,
      "loss": 1.1196,
      "num_input_tokens_seen": 26088392,
      "step": 44980
    },
    {
      "epoch": 6.70017873100983,
      "grad_norm": 13.5625,
      "learning_rate": 4.204986394227566e-05,
      "loss": 0.9167,
      "num_input_tokens_seen": 26091496,
      "step": 44985
    },
    {
      "epoch": 6.700923443550789,
      "grad_norm": 9.8125,
      "learning_rate": 4.20474873103313e-05,
      "loss": 0.7888,
      "num_input_tokens_seen": 26094440,
      "step": 44990
    },
    {
      "epoch": 6.701668156091749,
      "grad_norm": 23.375,
      "learning_rate": 4.204511039038697e-05,
      "loss": 1.3512,
      "num_input_tokens_seen": 26097256,
      "step": 44995
    },
    {
      "epoch": 6.702412868632708,
      "grad_norm": 9.5,
      "learning_rate": 4.204273318248283e-05,
      "loss": 0.5239,
      "num_input_tokens_seen": 26100168,
      "step": 45000
    },
    {
      "epoch": 6.703157581173667,
      "grad_norm": 11.25,
      "learning_rate": 4.204035568665903e-05,
      "loss": 1.0111,
      "num_input_tokens_seen": 26102824,
      "step": 45005
    },
    {
      "epoch": 6.703902293714626,
      "grad_norm": 6.84375,
      "learning_rate": 4.203797790295574e-05,
      "loss": 0.792,
      "num_input_tokens_seen": 26105928,
      "step": 45010
    },
    {
      "epoch": 6.704647006255585,
      "grad_norm": 8.25,
      "learning_rate": 4.203559983141312e-05,
      "loss": 1.2111,
      "num_input_tokens_seen": 26108776,
      "step": 45015
    },
    {
      "epoch": 6.705391718796545,
      "grad_norm": 19.875,
      "learning_rate": 4.2033221472071364e-05,
      "loss": 1.4101,
      "num_input_tokens_seen": 26111560,
      "step": 45020
    },
    {
      "epoch": 6.706136431337503,
      "grad_norm": 3.984375,
      "learning_rate": 4.2030842824970645e-05,
      "loss": 0.8421,
      "num_input_tokens_seen": 26114792,
      "step": 45025
    },
    {
      "epoch": 6.706881143878463,
      "grad_norm": 15.25,
      "learning_rate": 4.2028463890151144e-05,
      "loss": 1.1243,
      "num_input_tokens_seen": 26117832,
      "step": 45030
    },
    {
      "epoch": 6.707625856419422,
      "grad_norm": 6.6875,
      "learning_rate": 4.202608466765306e-05,
      "loss": 1.0862,
      "num_input_tokens_seen": 26120808,
      "step": 45035
    },
    {
      "epoch": 6.708370568960381,
      "grad_norm": 6.96875,
      "learning_rate": 4.202370515751657e-05,
      "loss": 1.1294,
      "num_input_tokens_seen": 26123656,
      "step": 45040
    },
    {
      "epoch": 6.70911528150134,
      "grad_norm": 6.90625,
      "learning_rate": 4.2021325359781885e-05,
      "loss": 1.1957,
      "num_input_tokens_seen": 26126856,
      "step": 45045
    },
    {
      "epoch": 6.7098599940423,
      "grad_norm": 10.9375,
      "learning_rate": 4.201894527448921e-05,
      "loss": 1.1404,
      "num_input_tokens_seen": 26129512,
      "step": 45050
    },
    {
      "epoch": 6.710604706583259,
      "grad_norm": 13.375,
      "learning_rate": 4.2016564901678744e-05,
      "loss": 1.0874,
      "num_input_tokens_seen": 26132360,
      "step": 45055
    },
    {
      "epoch": 6.711349419124218,
      "grad_norm": 7.25,
      "learning_rate": 4.201418424139072e-05,
      "loss": 1.2239,
      "num_input_tokens_seen": 26135400,
      "step": 45060
    },
    {
      "epoch": 6.712094131665177,
      "grad_norm": 6.78125,
      "learning_rate": 4.201180329366534e-05,
      "loss": 1.0723,
      "num_input_tokens_seen": 26138536,
      "step": 45065
    },
    {
      "epoch": 6.712838844206137,
      "grad_norm": 7.8125,
      "learning_rate": 4.200942205854282e-05,
      "loss": 1.0449,
      "num_input_tokens_seen": 26141544,
      "step": 45070
    },
    {
      "epoch": 6.713583556747095,
      "grad_norm": 8.375,
      "learning_rate": 4.2007040536063424e-05,
      "loss": 1.2612,
      "num_input_tokens_seen": 26144616,
      "step": 45075
    },
    {
      "epoch": 6.714328269288055,
      "grad_norm": 14.0625,
      "learning_rate": 4.200465872626736e-05,
      "loss": 0.9196,
      "num_input_tokens_seen": 26147496,
      "step": 45080
    },
    {
      "epoch": 6.715072981829014,
      "grad_norm": 9.5,
      "learning_rate": 4.200227662919487e-05,
      "loss": 0.8298,
      "num_input_tokens_seen": 26150440,
      "step": 45085
    },
    {
      "epoch": 6.7158176943699734,
      "grad_norm": 10.375,
      "learning_rate": 4.1999894244886184e-05,
      "loss": 1.0848,
      "num_input_tokens_seen": 26153160,
      "step": 45090
    },
    {
      "epoch": 6.716562406910932,
      "grad_norm": 8.8125,
      "learning_rate": 4.1997511573381575e-05,
      "loss": 1.3274,
      "num_input_tokens_seen": 26155816,
      "step": 45095
    },
    {
      "epoch": 6.717307119451892,
      "grad_norm": 6.40625,
      "learning_rate": 4.199512861472128e-05,
      "loss": 1.2666,
      "num_input_tokens_seen": 26159016,
      "step": 45100
    },
    {
      "epoch": 6.718051831992851,
      "grad_norm": 10.0,
      "learning_rate": 4.1992745368945554e-05,
      "loss": 1.1236,
      "num_input_tokens_seen": 26161960,
      "step": 45105
    },
    {
      "epoch": 6.71879654453381,
      "grad_norm": 13.8125,
      "learning_rate": 4.199036183609467e-05,
      "loss": 1.0778,
      "num_input_tokens_seen": 26164936,
      "step": 45110
    },
    {
      "epoch": 6.719541257074769,
      "grad_norm": 18.125,
      "learning_rate": 4.1987978016208895e-05,
      "loss": 1.0102,
      "num_input_tokens_seen": 26167688,
      "step": 45115
    },
    {
      "epoch": 6.720285969615729,
      "grad_norm": 4.6875,
      "learning_rate": 4.1985593909328494e-05,
      "loss": 0.9174,
      "num_input_tokens_seen": 26170248,
      "step": 45120
    },
    {
      "epoch": 6.721030682156687,
      "grad_norm": 5.21875,
      "learning_rate": 4.198320951549375e-05,
      "loss": 0.9916,
      "num_input_tokens_seen": 26173032,
      "step": 45125
    },
    {
      "epoch": 6.721775394697647,
      "grad_norm": 8.6875,
      "learning_rate": 4.1980824834744934e-05,
      "loss": 1.0671,
      "num_input_tokens_seen": 26175752,
      "step": 45130
    },
    {
      "epoch": 6.722520107238606,
      "grad_norm": 6.8125,
      "learning_rate": 4.1978439867122344e-05,
      "loss": 1.23,
      "num_input_tokens_seen": 26178760,
      "step": 45135
    },
    {
      "epoch": 6.7232648197795655,
      "grad_norm": 11.75,
      "learning_rate": 4.197605461266627e-05,
      "loss": 0.9521,
      "num_input_tokens_seen": 26181736,
      "step": 45140
    },
    {
      "epoch": 6.724009532320524,
      "grad_norm": 7.46875,
      "learning_rate": 4.197366907141701e-05,
      "loss": 0.9258,
      "num_input_tokens_seen": 26184456,
      "step": 45145
    },
    {
      "epoch": 6.724754244861483,
      "grad_norm": 10.875,
      "learning_rate": 4.197128324341486e-05,
      "loss": 1.0176,
      "num_input_tokens_seen": 26187144,
      "step": 45150
    },
    {
      "epoch": 6.725498957402443,
      "grad_norm": 16.25,
      "learning_rate": 4.196889712870013e-05,
      "loss": 1.3822,
      "num_input_tokens_seen": 26190056,
      "step": 45155
    },
    {
      "epoch": 6.726243669943402,
      "grad_norm": 10.375,
      "learning_rate": 4.196651072731313e-05,
      "loss": 1.2126,
      "num_input_tokens_seen": 26192872,
      "step": 45160
    },
    {
      "epoch": 6.726988382484361,
      "grad_norm": 7.96875,
      "learning_rate": 4.196412403929417e-05,
      "loss": 1.1686,
      "num_input_tokens_seen": 26195592,
      "step": 45165
    },
    {
      "epoch": 6.72773309502532,
      "grad_norm": 12.4375,
      "learning_rate": 4.196173706468358e-05,
      "loss": 1.162,
      "num_input_tokens_seen": 26198344,
      "step": 45170
    },
    {
      "epoch": 6.7284778075662794,
      "grad_norm": 12.5625,
      "learning_rate": 4.195934980352169e-05,
      "loss": 1.2037,
      "num_input_tokens_seen": 26201224,
      "step": 45175
    },
    {
      "epoch": 6.729222520107239,
      "grad_norm": 13.1875,
      "learning_rate": 4.195696225584881e-05,
      "loss": 0.7211,
      "num_input_tokens_seen": 26204104,
      "step": 45180
    },
    {
      "epoch": 6.729967232648198,
      "grad_norm": 10.1875,
      "learning_rate": 4.195457442170528e-05,
      "loss": 1.1154,
      "num_input_tokens_seen": 26207240,
      "step": 45185
    },
    {
      "epoch": 6.730711945189157,
      "grad_norm": 9.5,
      "learning_rate": 4.195218630113146e-05,
      "loss": 1.0098,
      "num_input_tokens_seen": 26210120,
      "step": 45190
    },
    {
      "epoch": 6.731456657730116,
      "grad_norm": 8.8125,
      "learning_rate": 4.1949797894167676e-05,
      "loss": 1.1548,
      "num_input_tokens_seen": 26213160,
      "step": 45195
    },
    {
      "epoch": 6.732201370271075,
      "grad_norm": 10.4375,
      "learning_rate": 4.1947409200854296e-05,
      "loss": 0.9664,
      "num_input_tokens_seen": 26216232,
      "step": 45200
    },
    {
      "epoch": 6.732946082812035,
      "grad_norm": 16.75,
      "learning_rate": 4.1945020221231643e-05,
      "loss": 1.074,
      "num_input_tokens_seen": 26219016,
      "step": 45205
    },
    {
      "epoch": 6.733690795352993,
      "grad_norm": 12.3125,
      "learning_rate": 4.194263095534011e-05,
      "loss": 1.2055,
      "num_input_tokens_seen": 26221672,
      "step": 45210
    },
    {
      "epoch": 6.734435507893953,
      "grad_norm": 12.3125,
      "learning_rate": 4.194024140322004e-05,
      "loss": 1.0471,
      "num_input_tokens_seen": 26224424,
      "step": 45215
    },
    {
      "epoch": 6.735180220434912,
      "grad_norm": 15.5625,
      "learning_rate": 4.193785156491181e-05,
      "loss": 1.3181,
      "num_input_tokens_seen": 26227304,
      "step": 45220
    },
    {
      "epoch": 6.7359249329758715,
      "grad_norm": 5.1875,
      "learning_rate": 4.193546144045579e-05,
      "loss": 1.184,
      "num_input_tokens_seen": 26230408,
      "step": 45225
    },
    {
      "epoch": 6.73666964551683,
      "grad_norm": 18.375,
      "learning_rate": 4.193307102989237e-05,
      "loss": 1.3822,
      "num_input_tokens_seen": 26233448,
      "step": 45230
    },
    {
      "epoch": 6.73741435805779,
      "grad_norm": 6.625,
      "learning_rate": 4.1930680333261915e-05,
      "loss": 1.0712,
      "num_input_tokens_seen": 26236456,
      "step": 45235
    },
    {
      "epoch": 6.738159070598749,
      "grad_norm": 15.9375,
      "learning_rate": 4.1928289350604826e-05,
      "loss": 1.2459,
      "num_input_tokens_seen": 26239688,
      "step": 45240
    },
    {
      "epoch": 6.738903783139708,
      "grad_norm": 9.0,
      "learning_rate": 4.19258980819615e-05,
      "loss": 1.0972,
      "num_input_tokens_seen": 26242824,
      "step": 45245
    },
    {
      "epoch": 6.739648495680667,
      "grad_norm": 9.375,
      "learning_rate": 4.192350652737232e-05,
      "loss": 0.7356,
      "num_input_tokens_seen": 26245672,
      "step": 45250
    },
    {
      "epoch": 6.740393208221627,
      "grad_norm": 4.1875,
      "learning_rate": 4.19211146868777e-05,
      "loss": 0.9577,
      "num_input_tokens_seen": 26248392,
      "step": 45255
    },
    {
      "epoch": 6.7411379207625854,
      "grad_norm": 9.625,
      "learning_rate": 4.1918722560518045e-05,
      "loss": 1.1612,
      "num_input_tokens_seen": 26251336,
      "step": 45260
    },
    {
      "epoch": 6.741882633303545,
      "grad_norm": 5.125,
      "learning_rate": 4.191633014833377e-05,
      "loss": 1.3335,
      "num_input_tokens_seen": 26254472,
      "step": 45265
    },
    {
      "epoch": 6.742627345844504,
      "grad_norm": 9.5625,
      "learning_rate": 4.191393745036529e-05,
      "loss": 0.8546,
      "num_input_tokens_seen": 26257192,
      "step": 45270
    },
    {
      "epoch": 6.7433720583854635,
      "grad_norm": 11.0,
      "learning_rate": 4.191154446665303e-05,
      "loss": 0.7606,
      "num_input_tokens_seen": 26260008,
      "step": 45275
    },
    {
      "epoch": 6.744116770926422,
      "grad_norm": 13.625,
      "learning_rate": 4.19091511972374e-05,
      "loss": 0.9042,
      "num_input_tokens_seen": 26262920,
      "step": 45280
    },
    {
      "epoch": 6.744861483467382,
      "grad_norm": 5.96875,
      "learning_rate": 4.1906757642158865e-05,
      "loss": 0.9729,
      "num_input_tokens_seen": 26265736,
      "step": 45285
    },
    {
      "epoch": 6.745606196008341,
      "grad_norm": 7.375,
      "learning_rate": 4.1904363801457835e-05,
      "loss": 1.5001,
      "num_input_tokens_seen": 26268904,
      "step": 45290
    },
    {
      "epoch": 6.7463509085493,
      "grad_norm": 22.125,
      "learning_rate": 4.1901969675174755e-05,
      "loss": 1.3078,
      "num_input_tokens_seen": 26272104,
      "step": 45295
    },
    {
      "epoch": 6.747095621090259,
      "grad_norm": 12.4375,
      "learning_rate": 4.189957526335009e-05,
      "loss": 1.1351,
      "num_input_tokens_seen": 26274984,
      "step": 45300
    },
    {
      "epoch": 6.747840333631219,
      "grad_norm": 12.9375,
      "learning_rate": 4.1897180566024266e-05,
      "loss": 0.8883,
      "num_input_tokens_seen": 26277512,
      "step": 45305
    },
    {
      "epoch": 6.7485850461721775,
      "grad_norm": 10.125,
      "learning_rate": 4.189478558323775e-05,
      "loss": 1.6059,
      "num_input_tokens_seen": 26280328,
      "step": 45310
    },
    {
      "epoch": 6.749329758713137,
      "grad_norm": 14.1875,
      "learning_rate": 4.1892390315031e-05,
      "loss": 1.4195,
      "num_input_tokens_seen": 26283720,
      "step": 45315
    },
    {
      "epoch": 6.750074471254096,
      "grad_norm": 10.0625,
      "learning_rate": 4.188999476144449e-05,
      "loss": 1.2748,
      "num_input_tokens_seen": 26286632,
      "step": 45320
    },
    {
      "epoch": 6.7508191837950555,
      "grad_norm": 10.75,
      "learning_rate": 4.188759892251868e-05,
      "loss": 0.8065,
      "num_input_tokens_seen": 26289704,
      "step": 45325
    },
    {
      "epoch": 6.751563896336014,
      "grad_norm": 12.75,
      "learning_rate": 4.188520279829406e-05,
      "loss": 1.2767,
      "num_input_tokens_seen": 26292808,
      "step": 45330
    },
    {
      "epoch": 6.752308608876973,
      "grad_norm": 9.0625,
      "learning_rate": 4.188280638881109e-05,
      "loss": 0.955,
      "num_input_tokens_seen": 26295688,
      "step": 45335
    },
    {
      "epoch": 6.753053321417933,
      "grad_norm": 9.8125,
      "learning_rate": 4.188040969411027e-05,
      "loss": 0.8583,
      "num_input_tokens_seen": 26298600,
      "step": 45340
    },
    {
      "epoch": 6.753798033958892,
      "grad_norm": 9.5,
      "learning_rate": 4.187801271423207e-05,
      "loss": 1.265,
      "num_input_tokens_seen": 26301800,
      "step": 45345
    },
    {
      "epoch": 6.754542746499851,
      "grad_norm": 9.75,
      "learning_rate": 4.187561544921702e-05,
      "loss": 0.7996,
      "num_input_tokens_seen": 26304680,
      "step": 45350
    },
    {
      "epoch": 6.75528745904081,
      "grad_norm": 6.3125,
      "learning_rate": 4.187321789910559e-05,
      "loss": 0.8158,
      "num_input_tokens_seen": 26307400,
      "step": 45355
    },
    {
      "epoch": 6.7560321715817695,
      "grad_norm": 13.6875,
      "learning_rate": 4.1870820063938296e-05,
      "loss": 1.4849,
      "num_input_tokens_seen": 26310376,
      "step": 45360
    },
    {
      "epoch": 6.756776884122728,
      "grad_norm": 9.5,
      "learning_rate": 4.186842194375564e-05,
      "loss": 1.147,
      "num_input_tokens_seen": 26313224,
      "step": 45365
    },
    {
      "epoch": 6.757521596663688,
      "grad_norm": 7.75,
      "learning_rate": 4.1866023538598136e-05,
      "loss": 0.8052,
      "num_input_tokens_seen": 26316200,
      "step": 45370
    },
    {
      "epoch": 6.758266309204647,
      "grad_norm": 12.125,
      "learning_rate": 4.186362484850631e-05,
      "loss": 0.9841,
      "num_input_tokens_seen": 26319432,
      "step": 45375
    },
    {
      "epoch": 6.759011021745606,
      "grad_norm": 9.4375,
      "learning_rate": 4.1861225873520684e-05,
      "loss": 1.2231,
      "num_input_tokens_seen": 26322376,
      "step": 45380
    },
    {
      "epoch": 6.759755734286565,
      "grad_norm": 8.375,
      "learning_rate": 4.185882661368178e-05,
      "loss": 0.9518,
      "num_input_tokens_seen": 26325128,
      "step": 45385
    },
    {
      "epoch": 6.760500446827525,
      "grad_norm": 20.5,
      "learning_rate": 4.185642706903014e-05,
      "loss": 1.2753,
      "num_input_tokens_seen": 26327848,
      "step": 45390
    },
    {
      "epoch": 6.7612451593684835,
      "grad_norm": 11.4375,
      "learning_rate": 4.185402723960629e-05,
      "loss": 1.6754,
      "num_input_tokens_seen": 26330824,
      "step": 45395
    },
    {
      "epoch": 6.761989871909443,
      "grad_norm": 12.125,
      "learning_rate": 4.185162712545079e-05,
      "loss": 0.9433,
      "num_input_tokens_seen": 26333736,
      "step": 45400
    },
    {
      "epoch": 6.762734584450402,
      "grad_norm": 10.0625,
      "learning_rate": 4.1849226726604165e-05,
      "loss": 0.9219,
      "num_input_tokens_seen": 26336808,
      "step": 45405
    },
    {
      "epoch": 6.7634792969913615,
      "grad_norm": 9.0,
      "learning_rate": 4.184682604310698e-05,
      "loss": 1.1222,
      "num_input_tokens_seen": 26339528,
      "step": 45410
    },
    {
      "epoch": 6.76422400953232,
      "grad_norm": 14.875,
      "learning_rate": 4.18444250749998e-05,
      "loss": 1.0631,
      "num_input_tokens_seen": 26342728,
      "step": 45415
    },
    {
      "epoch": 6.76496872207328,
      "grad_norm": 6.1875,
      "learning_rate": 4.184202382232317e-05,
      "loss": 1.0222,
      "num_input_tokens_seen": 26345800,
      "step": 45420
    },
    {
      "epoch": 6.765713434614239,
      "grad_norm": 5.90625,
      "learning_rate": 4.183962228511767e-05,
      "loss": 0.9339,
      "num_input_tokens_seen": 26348808,
      "step": 45425
    },
    {
      "epoch": 6.766458147155198,
      "grad_norm": 17.75,
      "learning_rate": 4.183722046342386e-05,
      "loss": 1.6259,
      "num_input_tokens_seen": 26351624,
      "step": 45430
    },
    {
      "epoch": 6.767202859696157,
      "grad_norm": 12.0,
      "learning_rate": 4.1834818357282336e-05,
      "loss": 1.1205,
      "num_input_tokens_seen": 26354472,
      "step": 45435
    },
    {
      "epoch": 6.767947572237117,
      "grad_norm": 11.25,
      "learning_rate": 4.183241596673366e-05,
      "loss": 0.9867,
      "num_input_tokens_seen": 26357480,
      "step": 45440
    },
    {
      "epoch": 6.7686922847780755,
      "grad_norm": 10.25,
      "learning_rate": 4.183001329181843e-05,
      "loss": 0.8696,
      "num_input_tokens_seen": 26360360,
      "step": 45445
    },
    {
      "epoch": 6.769436997319035,
      "grad_norm": 9.5625,
      "learning_rate": 4.1827610332577214e-05,
      "loss": 1.5272,
      "num_input_tokens_seen": 26363080,
      "step": 45450
    },
    {
      "epoch": 6.770181709859994,
      "grad_norm": 10.3125,
      "learning_rate": 4.1825207089050634e-05,
      "loss": 1.1563,
      "num_input_tokens_seen": 26366024,
      "step": 45455
    },
    {
      "epoch": 6.7709264224009535,
      "grad_norm": 10.1875,
      "learning_rate": 4.182280356127928e-05,
      "loss": 1.3885,
      "num_input_tokens_seen": 26369128,
      "step": 45460
    },
    {
      "epoch": 6.771671134941912,
      "grad_norm": 9.1875,
      "learning_rate": 4.182039974930376e-05,
      "loss": 1.0736,
      "num_input_tokens_seen": 26372136,
      "step": 45465
    },
    {
      "epoch": 6.772415847482872,
      "grad_norm": 8.1875,
      "learning_rate": 4.1817995653164675e-05,
      "loss": 1.098,
      "num_input_tokens_seen": 26375240,
      "step": 45470
    },
    {
      "epoch": 6.773160560023831,
      "grad_norm": 10.6875,
      "learning_rate": 4.1815591272902654e-05,
      "loss": 1.1457,
      "num_input_tokens_seen": 26377992,
      "step": 45475
    },
    {
      "epoch": 6.77390527256479,
      "grad_norm": 19.875,
      "learning_rate": 4.1813186608558305e-05,
      "loss": 1.3785,
      "num_input_tokens_seen": 26381064,
      "step": 45480
    },
    {
      "epoch": 6.774649985105749,
      "grad_norm": 9.25,
      "learning_rate": 4.181078166017226e-05,
      "loss": 0.9493,
      "num_input_tokens_seen": 26384040,
      "step": 45485
    },
    {
      "epoch": 6.775394697646709,
      "grad_norm": 15.6875,
      "learning_rate": 4.180837642778513e-05,
      "loss": 1.2226,
      "num_input_tokens_seen": 26387112,
      "step": 45490
    },
    {
      "epoch": 6.7761394101876675,
      "grad_norm": 5.875,
      "learning_rate": 4.180597091143759e-05,
      "loss": 0.9882,
      "num_input_tokens_seen": 26389992,
      "step": 45495
    },
    {
      "epoch": 6.776884122728626,
      "grad_norm": 7.65625,
      "learning_rate": 4.1803565111170227e-05,
      "loss": 0.8997,
      "num_input_tokens_seen": 26392456,
      "step": 45500
    },
    {
      "epoch": 6.777628835269586,
      "grad_norm": 14.25,
      "learning_rate": 4.180115902702372e-05,
      "loss": 0.9842,
      "num_input_tokens_seen": 26395592,
      "step": 45505
    },
    {
      "epoch": 6.778373547810546,
      "grad_norm": 12.5625,
      "learning_rate": 4.179875265903871e-05,
      "loss": 1.3302,
      "num_input_tokens_seen": 26398280,
      "step": 45510
    },
    {
      "epoch": 6.779118260351504,
      "grad_norm": 21.75,
      "learning_rate": 4.1796346007255844e-05,
      "loss": 1.4811,
      "num_input_tokens_seen": 26401064,
      "step": 45515
    },
    {
      "epoch": 6.779862972892463,
      "grad_norm": 11.25,
      "learning_rate": 4.1793939071715786e-05,
      "loss": 1.2332,
      "num_input_tokens_seen": 26403752,
      "step": 45520
    },
    {
      "epoch": 6.780607685433423,
      "grad_norm": 6.6875,
      "learning_rate": 4.1791531852459196e-05,
      "loss": 1.1338,
      "num_input_tokens_seen": 26406600,
      "step": 45525
    },
    {
      "epoch": 6.781352397974382,
      "grad_norm": 8.125,
      "learning_rate": 4.1789124349526745e-05,
      "loss": 1.3054,
      "num_input_tokens_seen": 26409512,
      "step": 45530
    },
    {
      "epoch": 6.782097110515341,
      "grad_norm": 12.8125,
      "learning_rate": 4.178671656295909e-05,
      "loss": 1.0504,
      "num_input_tokens_seen": 26412584,
      "step": 45535
    },
    {
      "epoch": 6.7828418230563,
      "grad_norm": 7.65625,
      "learning_rate": 4.1784308492796926e-05,
      "loss": 0.944,
      "num_input_tokens_seen": 26415528,
      "step": 45540
    },
    {
      "epoch": 6.7835865355972595,
      "grad_norm": 8.875,
      "learning_rate": 4.1781900139080933e-05,
      "loss": 1.0123,
      "num_input_tokens_seen": 26418344,
      "step": 45545
    },
    {
      "epoch": 6.784331248138218,
      "grad_norm": 15.25,
      "learning_rate": 4.1779491501851786e-05,
      "loss": 0.9616,
      "num_input_tokens_seen": 26420872,
      "step": 45550
    },
    {
      "epoch": 6.785075960679178,
      "grad_norm": 8.625,
      "learning_rate": 4.177708258115019e-05,
      "loss": 1.124,
      "num_input_tokens_seen": 26423752,
      "step": 45555
    },
    {
      "epoch": 6.785820673220137,
      "grad_norm": 10.3125,
      "learning_rate": 4.177467337701683e-05,
      "loss": 1.2599,
      "num_input_tokens_seen": 26426568,
      "step": 45560
    },
    {
      "epoch": 6.786565385761096,
      "grad_norm": 6.71875,
      "learning_rate": 4.177226388949241e-05,
      "loss": 1.0892,
      "num_input_tokens_seen": 26429608,
      "step": 45565
    },
    {
      "epoch": 6.787310098302055,
      "grad_norm": 9.25,
      "learning_rate": 4.176985411861765e-05,
      "loss": 1.0986,
      "num_input_tokens_seen": 26432776,
      "step": 45570
    },
    {
      "epoch": 6.788054810843015,
      "grad_norm": 7.59375,
      "learning_rate": 4.1767444064433244e-05,
      "loss": 1.2514,
      "num_input_tokens_seen": 26435784,
      "step": 45575
    },
    {
      "epoch": 6.7887995233839735,
      "grad_norm": 8.875,
      "learning_rate": 4.1765033726979906e-05,
      "loss": 0.9794,
      "num_input_tokens_seen": 26438408,
      "step": 45580
    },
    {
      "epoch": 6.789544235924933,
      "grad_norm": 7.5625,
      "learning_rate": 4.176262310629837e-05,
      "loss": 1.0095,
      "num_input_tokens_seen": 26441544,
      "step": 45585
    },
    {
      "epoch": 6.790288948465892,
      "grad_norm": 16.125,
      "learning_rate": 4.176021220242935e-05,
      "loss": 1.3239,
      "num_input_tokens_seen": 26444136,
      "step": 45590
    },
    {
      "epoch": 6.791033661006852,
      "grad_norm": 11.125,
      "learning_rate": 4.175780101541358e-05,
      "loss": 1.087,
      "num_input_tokens_seen": 26446952,
      "step": 45595
    },
    {
      "epoch": 6.79177837354781,
      "grad_norm": 10.375,
      "learning_rate": 4.175538954529179e-05,
      "loss": 0.8663,
      "num_input_tokens_seen": 26449800,
      "step": 45600
    },
    {
      "epoch": 6.79252308608877,
      "grad_norm": 14.8125,
      "learning_rate": 4.175297779210473e-05,
      "loss": 1.079,
      "num_input_tokens_seen": 26452776,
      "step": 45605
    },
    {
      "epoch": 6.793267798629729,
      "grad_norm": 9.25,
      "learning_rate": 4.1750565755893134e-05,
      "loss": 1.1797,
      "num_input_tokens_seen": 26455528,
      "step": 45610
    },
    {
      "epoch": 6.794012511170688,
      "grad_norm": 8.125,
      "learning_rate": 4.174815343669775e-05,
      "loss": 1.215,
      "num_input_tokens_seen": 26458632,
      "step": 45615
    },
    {
      "epoch": 6.794757223711647,
      "grad_norm": 11.625,
      "learning_rate": 4.1745740834559335e-05,
      "loss": 0.8065,
      "num_input_tokens_seen": 26461768,
      "step": 45620
    },
    {
      "epoch": 6.795501936252607,
      "grad_norm": 14.875,
      "learning_rate": 4.174332794951866e-05,
      "loss": 1.0996,
      "num_input_tokens_seen": 26464552,
      "step": 45625
    },
    {
      "epoch": 6.7962466487935655,
      "grad_norm": 10.75,
      "learning_rate": 4.174091478161646e-05,
      "loss": 1.1344,
      "num_input_tokens_seen": 26467496,
      "step": 45630
    },
    {
      "epoch": 6.796991361334525,
      "grad_norm": 13.25,
      "learning_rate": 4.173850133089353e-05,
      "loss": 1.1326,
      "num_input_tokens_seen": 26470024,
      "step": 45635
    },
    {
      "epoch": 6.797736073875484,
      "grad_norm": 8.75,
      "learning_rate": 4.173608759739063e-05,
      "loss": 1.0416,
      "num_input_tokens_seen": 26473096,
      "step": 45640
    },
    {
      "epoch": 6.798480786416444,
      "grad_norm": 6.34375,
      "learning_rate": 4.173367358114855e-05,
      "loss": 1.2195,
      "num_input_tokens_seen": 26476424,
      "step": 45645
    },
    {
      "epoch": 6.799225498957402,
      "grad_norm": 15.0625,
      "learning_rate": 4.1731259282208047e-05,
      "loss": 1.083,
      "num_input_tokens_seen": 26479496,
      "step": 45650
    },
    {
      "epoch": 6.799970211498362,
      "grad_norm": 13.75,
      "learning_rate": 4.1728844700609926e-05,
      "loss": 1.1373,
      "num_input_tokens_seen": 26482184,
      "step": 45655
    },
    {
      "epoch": 6.800714924039321,
      "grad_norm": 12.3125,
      "learning_rate": 4.172642983639498e-05,
      "loss": 1.1088,
      "num_input_tokens_seen": 26485128,
      "step": 45660
    },
    {
      "epoch": 6.8014596365802795,
      "grad_norm": 11.3125,
      "learning_rate": 4.1724014689604e-05,
      "loss": 0.8961,
      "num_input_tokens_seen": 26487880,
      "step": 45665
    },
    {
      "epoch": 6.802204349121239,
      "grad_norm": 6.53125,
      "learning_rate": 4.1721599260277796e-05,
      "loss": 1.0271,
      "num_input_tokens_seen": 26490632,
      "step": 45670
    },
    {
      "epoch": 6.802949061662199,
      "grad_norm": 12.6875,
      "learning_rate": 4.171918354845716e-05,
      "loss": 0.6984,
      "num_input_tokens_seen": 26493416,
      "step": 45675
    },
    {
      "epoch": 6.803693774203158,
      "grad_norm": 16.75,
      "learning_rate": 4.171676755418291e-05,
      "loss": 1.1365,
      "num_input_tokens_seen": 26496424,
      "step": 45680
    },
    {
      "epoch": 6.804438486744116,
      "grad_norm": 13.75,
      "learning_rate": 4.171435127749587e-05,
      "loss": 1.0724,
      "num_input_tokens_seen": 26499496,
      "step": 45685
    },
    {
      "epoch": 6.805183199285076,
      "grad_norm": 6.5625,
      "learning_rate": 4.171193471843685e-05,
      "loss": 0.9269,
      "num_input_tokens_seen": 26502408,
      "step": 45690
    },
    {
      "epoch": 6.805927911826036,
      "grad_norm": 19.625,
      "learning_rate": 4.170951787704667e-05,
      "loss": 1.6269,
      "num_input_tokens_seen": 26505288,
      "step": 45695
    },
    {
      "epoch": 6.806672624366994,
      "grad_norm": 10.6875,
      "learning_rate": 4.170710075336617e-05,
      "loss": 0.7073,
      "num_input_tokens_seen": 26508104,
      "step": 45700
    },
    {
      "epoch": 6.807417336907953,
      "grad_norm": 7.6875,
      "learning_rate": 4.170468334743619e-05,
      "loss": 0.9887,
      "num_input_tokens_seen": 26511080,
      "step": 45705
    },
    {
      "epoch": 6.808162049448913,
      "grad_norm": 11.1875,
      "learning_rate": 4.1702265659297554e-05,
      "loss": 1.2503,
      "num_input_tokens_seen": 26514216,
      "step": 45710
    },
    {
      "epoch": 6.8089067619898715,
      "grad_norm": 7.9375,
      "learning_rate": 4.169984768899112e-05,
      "loss": 0.8997,
      "num_input_tokens_seen": 26517384,
      "step": 45715
    },
    {
      "epoch": 6.809651474530831,
      "grad_norm": 19.875,
      "learning_rate": 4.169742943655774e-05,
      "loss": 1.1851,
      "num_input_tokens_seen": 26520456,
      "step": 45720
    },
    {
      "epoch": 6.81039618707179,
      "grad_norm": 17.625,
      "learning_rate": 4.169501090203826e-05,
      "loss": 1.0302,
      "num_input_tokens_seen": 26523624,
      "step": 45725
    },
    {
      "epoch": 6.81114089961275,
      "grad_norm": 7.21875,
      "learning_rate": 4.1692592085473525e-05,
      "loss": 1.2182,
      "num_input_tokens_seen": 26526504,
      "step": 45730
    },
    {
      "epoch": 6.811885612153708,
      "grad_norm": 5.625,
      "learning_rate": 4.169017298690442e-05,
      "loss": 0.9742,
      "num_input_tokens_seen": 26529192,
      "step": 45735
    },
    {
      "epoch": 6.812630324694668,
      "grad_norm": 12.5,
      "learning_rate": 4.168775360637181e-05,
      "loss": 1.1404,
      "num_input_tokens_seen": 26532168,
      "step": 45740
    },
    {
      "epoch": 6.813375037235627,
      "grad_norm": 9.125,
      "learning_rate": 4.168533394391656e-05,
      "loss": 0.9875,
      "num_input_tokens_seen": 26534888,
      "step": 45745
    },
    {
      "epoch": 6.814119749776586,
      "grad_norm": 7.71875,
      "learning_rate": 4.1682913999579545e-05,
      "loss": 1.4291,
      "num_input_tokens_seen": 26537800,
      "step": 45750
    },
    {
      "epoch": 6.814864462317545,
      "grad_norm": 9.5625,
      "learning_rate": 4.1680493773401657e-05,
      "loss": 1.3558,
      "num_input_tokens_seen": 26540680,
      "step": 45755
    },
    {
      "epoch": 6.815609174858505,
      "grad_norm": 10.0625,
      "learning_rate": 4.167807326542379e-05,
      "loss": 0.7879,
      "num_input_tokens_seen": 26543432,
      "step": 45760
    },
    {
      "epoch": 6.816353887399464,
      "grad_norm": 9.0625,
      "learning_rate": 4.167565247568681e-05,
      "loss": 1.2209,
      "num_input_tokens_seen": 26546120,
      "step": 45765
    },
    {
      "epoch": 6.817098599940423,
      "grad_norm": 4.46875,
      "learning_rate": 4.167323140423164e-05,
      "loss": 0.8452,
      "num_input_tokens_seen": 26548936,
      "step": 45770
    },
    {
      "epoch": 6.817843312481382,
      "grad_norm": 11.6875,
      "learning_rate": 4.167081005109917e-05,
      "loss": 1.258,
      "num_input_tokens_seen": 26551816,
      "step": 45775
    },
    {
      "epoch": 6.818588025022342,
      "grad_norm": 7.40625,
      "learning_rate": 4.16683884163303e-05,
      "loss": 0.8082,
      "num_input_tokens_seen": 26554632,
      "step": 45780
    },
    {
      "epoch": 6.8193327375633,
      "grad_norm": 11.5,
      "learning_rate": 4.166596649996596e-05,
      "loss": 1.0036,
      "num_input_tokens_seen": 26557576,
      "step": 45785
    },
    {
      "epoch": 6.82007745010426,
      "grad_norm": 9.25,
      "learning_rate": 4.166354430204705e-05,
      "loss": 1.1931,
      "num_input_tokens_seen": 26560264,
      "step": 45790
    },
    {
      "epoch": 6.820822162645219,
      "grad_norm": 10.25,
      "learning_rate": 4.166112182261449e-05,
      "loss": 1.0679,
      "num_input_tokens_seen": 26563048,
      "step": 45795
    },
    {
      "epoch": 6.821566875186178,
      "grad_norm": 11.6875,
      "learning_rate": 4.1658699061709215e-05,
      "loss": 0.8581,
      "num_input_tokens_seen": 26565960,
      "step": 45800
    },
    {
      "epoch": 6.822311587727137,
      "grad_norm": 11.0,
      "learning_rate": 4.1656276019372156e-05,
      "loss": 1.0583,
      "num_input_tokens_seen": 26568872,
      "step": 45805
    },
    {
      "epoch": 6.823056300268097,
      "grad_norm": 7.03125,
      "learning_rate": 4.165385269564423e-05,
      "loss": 1.0102,
      "num_input_tokens_seen": 26571624,
      "step": 45810
    },
    {
      "epoch": 6.823801012809056,
      "grad_norm": 6.40625,
      "learning_rate": 4.16514290905664e-05,
      "loss": 0.8451,
      "num_input_tokens_seen": 26574440,
      "step": 45815
    },
    {
      "epoch": 6.824545725350015,
      "grad_norm": 10.875,
      "learning_rate": 4.164900520417959e-05,
      "loss": 0.9156,
      "num_input_tokens_seen": 26577256,
      "step": 45820
    },
    {
      "epoch": 6.825290437890974,
      "grad_norm": 14.375,
      "learning_rate": 4.164658103652477e-05,
      "loss": 1.293,
      "num_input_tokens_seen": 26580328,
      "step": 45825
    },
    {
      "epoch": 6.826035150431934,
      "grad_norm": 13.125,
      "learning_rate": 4.164415658764287e-05,
      "loss": 0.9655,
      "num_input_tokens_seen": 26583048,
      "step": 45830
    },
    {
      "epoch": 6.826779862972892,
      "grad_norm": 7.25,
      "learning_rate": 4.164173185757487e-05,
      "loss": 1.1435,
      "num_input_tokens_seen": 26585864,
      "step": 45835
    },
    {
      "epoch": 6.827524575513852,
      "grad_norm": 8.5,
      "learning_rate": 4.163930684636173e-05,
      "loss": 1.0694,
      "num_input_tokens_seen": 26588808,
      "step": 45840
    },
    {
      "epoch": 6.828269288054811,
      "grad_norm": 9.8125,
      "learning_rate": 4.16368815540444e-05,
      "loss": 1.0475,
      "num_input_tokens_seen": 26591528,
      "step": 45845
    },
    {
      "epoch": 6.82901400059577,
      "grad_norm": 15.0625,
      "learning_rate": 4.1634455980663866e-05,
      "loss": 1.4113,
      "num_input_tokens_seen": 26594312,
      "step": 45850
    },
    {
      "epoch": 6.829758713136729,
      "grad_norm": 10.875,
      "learning_rate": 4.163203012626111e-05,
      "loss": 1.0182,
      "num_input_tokens_seen": 26597224,
      "step": 45855
    },
    {
      "epoch": 6.830503425677689,
      "grad_norm": 15.3125,
      "learning_rate": 4.16296039908771e-05,
      "loss": 0.9951,
      "num_input_tokens_seen": 26600168,
      "step": 45860
    },
    {
      "epoch": 6.831248138218648,
      "grad_norm": 4.40625,
      "learning_rate": 4.162717757455284e-05,
      "loss": 0.8214,
      "num_input_tokens_seen": 26603016,
      "step": 45865
    },
    {
      "epoch": 6.831992850759606,
      "grad_norm": 8.6875,
      "learning_rate": 4.162475087732931e-05,
      "loss": 1.1359,
      "num_input_tokens_seen": 26606088,
      "step": 45870
    },
    {
      "epoch": 6.832737563300566,
      "grad_norm": 13.0,
      "learning_rate": 4.162232389924751e-05,
      "loss": 0.9283,
      "num_input_tokens_seen": 26609064,
      "step": 45875
    },
    {
      "epoch": 6.833482275841525,
      "grad_norm": 8.5625,
      "learning_rate": 4.1619896640348445e-05,
      "loss": 1.2355,
      "num_input_tokens_seen": 26611912,
      "step": 45880
    },
    {
      "epoch": 6.834226988382484,
      "grad_norm": 16.375,
      "learning_rate": 4.1617469100673126e-05,
      "loss": 1.42,
      "num_input_tokens_seen": 26614728,
      "step": 45885
    },
    {
      "epoch": 6.834971700923443,
      "grad_norm": 6.9375,
      "learning_rate": 4.161504128026255e-05,
      "loss": 0.9229,
      "num_input_tokens_seen": 26617896,
      "step": 45890
    },
    {
      "epoch": 6.835716413464403,
      "grad_norm": 14.625,
      "learning_rate": 4.1612613179157725e-05,
      "loss": 1.3468,
      "num_input_tokens_seen": 26620936,
      "step": 45895
    },
    {
      "epoch": 6.836461126005362,
      "grad_norm": 10.125,
      "learning_rate": 4.1610184797399696e-05,
      "loss": 1.073,
      "num_input_tokens_seen": 26624104,
      "step": 45900
    },
    {
      "epoch": 6.837205838546321,
      "grad_norm": 7.5625,
      "learning_rate": 4.160775613502948e-05,
      "loss": 1.0278,
      "num_input_tokens_seen": 26626760,
      "step": 45905
    },
    {
      "epoch": 6.83795055108728,
      "grad_norm": 8.1875,
      "learning_rate": 4.160532719208809e-05,
      "loss": 1.0697,
      "num_input_tokens_seen": 26629736,
      "step": 45910
    },
    {
      "epoch": 6.83869526362824,
      "grad_norm": 9.5,
      "learning_rate": 4.160289796861659e-05,
      "loss": 0.8759,
      "num_input_tokens_seen": 26632680,
      "step": 45915
    },
    {
      "epoch": 6.839439976169198,
      "grad_norm": 5.5,
      "learning_rate": 4.1600468464656e-05,
      "loss": 1.2843,
      "num_input_tokens_seen": 26635560,
      "step": 45920
    },
    {
      "epoch": 6.840184688710158,
      "grad_norm": 13.3125,
      "learning_rate": 4.1598038680247363e-05,
      "loss": 1.1781,
      "num_input_tokens_seen": 26638568,
      "step": 45925
    },
    {
      "epoch": 6.840929401251117,
      "grad_norm": 10.375,
      "learning_rate": 4.159560861543174e-05,
      "loss": 1.1416,
      "num_input_tokens_seen": 26641448,
      "step": 45930
    },
    {
      "epoch": 6.8416741137920765,
      "grad_norm": 9.6875,
      "learning_rate": 4.159317827025016e-05,
      "loss": 1.1809,
      "num_input_tokens_seen": 26644520,
      "step": 45935
    },
    {
      "epoch": 6.842418826333035,
      "grad_norm": 13.4375,
      "learning_rate": 4.159074764474371e-05,
      "loss": 0.878,
      "num_input_tokens_seen": 26647208,
      "step": 45940
    },
    {
      "epoch": 6.843163538873995,
      "grad_norm": 8.125,
      "learning_rate": 4.1588316738953434e-05,
      "loss": 1.0218,
      "num_input_tokens_seen": 26650312,
      "step": 45945
    },
    {
      "epoch": 6.843908251414954,
      "grad_norm": 9.875,
      "learning_rate": 4.158588555292041e-05,
      "loss": 1.3142,
      "num_input_tokens_seen": 26653160,
      "step": 45950
    },
    {
      "epoch": 6.844652963955913,
      "grad_norm": 9.25,
      "learning_rate": 4.158345408668571e-05,
      "loss": 1.1647,
      "num_input_tokens_seen": 26656456,
      "step": 45955
    },
    {
      "epoch": 6.845397676496872,
      "grad_norm": 8.4375,
      "learning_rate": 4.15810223402904e-05,
      "loss": 0.7923,
      "num_input_tokens_seen": 26659400,
      "step": 45960
    },
    {
      "epoch": 6.846142389037832,
      "grad_norm": 18.25,
      "learning_rate": 4.157859031377558e-05,
      "loss": 1.3678,
      "num_input_tokens_seen": 26662248,
      "step": 45965
    },
    {
      "epoch": 6.84688710157879,
      "grad_norm": 11.9375,
      "learning_rate": 4.157615800718232e-05,
      "loss": 0.8708,
      "num_input_tokens_seen": 26664904,
      "step": 45970
    },
    {
      "epoch": 6.84763181411975,
      "grad_norm": 19.25,
      "learning_rate": 4.1573725420551716e-05,
      "loss": 1.0212,
      "num_input_tokens_seen": 26667848,
      "step": 45975
    },
    {
      "epoch": 6.848376526660709,
      "grad_norm": 10.0625,
      "learning_rate": 4.157129255392487e-05,
      "loss": 1.1026,
      "num_input_tokens_seen": 26670952,
      "step": 45980
    },
    {
      "epoch": 6.8491212392016685,
      "grad_norm": 16.125,
      "learning_rate": 4.1568859407342876e-05,
      "loss": 0.9938,
      "num_input_tokens_seen": 26673384,
      "step": 45985
    },
    {
      "epoch": 6.849865951742627,
      "grad_norm": 10.625,
      "learning_rate": 4.1566425980846844e-05,
      "loss": 1.1063,
      "num_input_tokens_seen": 26676552,
      "step": 45990
    },
    {
      "epoch": 6.850610664283587,
      "grad_norm": 13.6875,
      "learning_rate": 4.156399227447788e-05,
      "loss": 1.2164,
      "num_input_tokens_seen": 26679304,
      "step": 45995
    },
    {
      "epoch": 6.851355376824546,
      "grad_norm": 10.625,
      "learning_rate": 4.15615582882771e-05,
      "loss": 1.1572,
      "num_input_tokens_seen": 26682088,
      "step": 46000
    },
    {
      "epoch": 6.852100089365505,
      "grad_norm": 12.8125,
      "learning_rate": 4.155912402228563e-05,
      "loss": 1.0823,
      "num_input_tokens_seen": 26684904,
      "step": 46005
    },
    {
      "epoch": 6.852844801906464,
      "grad_norm": 10.6875,
      "learning_rate": 4.155668947654458e-05,
      "loss": 0.8601,
      "num_input_tokens_seen": 26687976,
      "step": 46010
    },
    {
      "epoch": 6.853589514447423,
      "grad_norm": 12.3125,
      "learning_rate": 4.15542546510951e-05,
      "loss": 0.9102,
      "num_input_tokens_seen": 26690792,
      "step": 46015
    },
    {
      "epoch": 6.8543342269883825,
      "grad_norm": 8.0625,
      "learning_rate": 4.155181954597832e-05,
      "loss": 1.0394,
      "num_input_tokens_seen": 26693704,
      "step": 46020
    },
    {
      "epoch": 6.855078939529342,
      "grad_norm": 14.75,
      "learning_rate": 4.154938416123535e-05,
      "loss": 1.2815,
      "num_input_tokens_seen": 26696552,
      "step": 46025
    },
    {
      "epoch": 6.855823652070301,
      "grad_norm": 8.8125,
      "learning_rate": 4.154694849690737e-05,
      "loss": 0.9596,
      "num_input_tokens_seen": 26699784,
      "step": 46030
    },
    {
      "epoch": 6.85656836461126,
      "grad_norm": 10.5625,
      "learning_rate": 4.154451255303551e-05,
      "loss": 1.009,
      "num_input_tokens_seen": 26702856,
      "step": 46035
    },
    {
      "epoch": 6.857313077152219,
      "grad_norm": 9.625,
      "learning_rate": 4.154207632966092e-05,
      "loss": 1.2303,
      "num_input_tokens_seen": 26705832,
      "step": 46040
    },
    {
      "epoch": 6.858057789693179,
      "grad_norm": 16.625,
      "learning_rate": 4.1539639826824765e-05,
      "loss": 1.077,
      "num_input_tokens_seen": 26708616,
      "step": 46045
    },
    {
      "epoch": 6.858802502234138,
      "grad_norm": 8.75,
      "learning_rate": 4.1537203044568205e-05,
      "loss": 0.9596,
      "num_input_tokens_seen": 26711080,
      "step": 46050
    },
    {
      "epoch": 6.859547214775096,
      "grad_norm": 13.75,
      "learning_rate": 4.153476598293241e-05,
      "loss": 0.9175,
      "num_input_tokens_seen": 26713768,
      "step": 46055
    },
    {
      "epoch": 6.860291927316056,
      "grad_norm": 16.5,
      "learning_rate": 4.153232864195855e-05,
      "loss": 0.9681,
      "num_input_tokens_seen": 26716680,
      "step": 46060
    },
    {
      "epoch": 6.861036639857015,
      "grad_norm": 6.40625,
      "learning_rate": 4.1529891021687796e-05,
      "loss": 1.0259,
      "num_input_tokens_seen": 26719624,
      "step": 46065
    },
    {
      "epoch": 6.8617813523979745,
      "grad_norm": 10.3125,
      "learning_rate": 4.152745312216134e-05,
      "loss": 1.3106,
      "num_input_tokens_seen": 26722728,
      "step": 46070
    },
    {
      "epoch": 6.862526064938933,
      "grad_norm": 7.71875,
      "learning_rate": 4.152501494342035e-05,
      "loss": 0.848,
      "num_input_tokens_seen": 26725896,
      "step": 46075
    },
    {
      "epoch": 6.863270777479893,
      "grad_norm": 10.5625,
      "learning_rate": 4.152257648550604e-05,
      "loss": 1.1781,
      "num_input_tokens_seen": 26729128,
      "step": 46080
    },
    {
      "epoch": 6.864015490020852,
      "grad_norm": 7.59375,
      "learning_rate": 4.1520137748459587e-05,
      "loss": 1.3319,
      "num_input_tokens_seen": 26732008,
      "step": 46085
    },
    {
      "epoch": 6.864760202561811,
      "grad_norm": 10.6875,
      "learning_rate": 4.1517698732322194e-05,
      "loss": 1.1258,
      "num_input_tokens_seen": 26734824,
      "step": 46090
    },
    {
      "epoch": 6.86550491510277,
      "grad_norm": 13.9375,
      "learning_rate": 4.1515259437135076e-05,
      "loss": 1.156,
      "num_input_tokens_seen": 26737896,
      "step": 46095
    },
    {
      "epoch": 6.86624962764373,
      "grad_norm": 5.15625,
      "learning_rate": 4.1512819862939425e-05,
      "loss": 0.9168,
      "num_input_tokens_seen": 26740488,
      "step": 46100
    },
    {
      "epoch": 6.8669943401846885,
      "grad_norm": 8.9375,
      "learning_rate": 4.1510380009776475e-05,
      "loss": 1.1044,
      "num_input_tokens_seen": 26743336,
      "step": 46105
    },
    {
      "epoch": 6.867739052725648,
      "grad_norm": 10.8125,
      "learning_rate": 4.150793987768743e-05,
      "loss": 1.0604,
      "num_input_tokens_seen": 26746088,
      "step": 46110
    },
    {
      "epoch": 6.868483765266607,
      "grad_norm": 8.4375,
      "learning_rate": 4.1505499466713516e-05,
      "loss": 0.9242,
      "num_input_tokens_seen": 26749032,
      "step": 46115
    },
    {
      "epoch": 6.8692284778075665,
      "grad_norm": 11.8125,
      "learning_rate": 4.1503058776895974e-05,
      "loss": 1.0725,
      "num_input_tokens_seen": 26752072,
      "step": 46120
    },
    {
      "epoch": 6.869973190348525,
      "grad_norm": 12.0625,
      "learning_rate": 4.150061780827602e-05,
      "loss": 0.8895,
      "num_input_tokens_seen": 26754856,
      "step": 46125
    },
    {
      "epoch": 6.870717902889485,
      "grad_norm": 6.375,
      "learning_rate": 4.14981765608949e-05,
      "loss": 1.3779,
      "num_input_tokens_seen": 26757864,
      "step": 46130
    },
    {
      "epoch": 6.871462615430444,
      "grad_norm": 12.0,
      "learning_rate": 4.1495735034793856e-05,
      "loss": 0.685,
      "num_input_tokens_seen": 26760872,
      "step": 46135
    },
    {
      "epoch": 6.872207327971403,
      "grad_norm": 14.5,
      "learning_rate": 4.149329323001413e-05,
      "loss": 1.2083,
      "num_input_tokens_seen": 26763688,
      "step": 46140
    },
    {
      "epoch": 6.872952040512362,
      "grad_norm": 17.5,
      "learning_rate": 4.149085114659699e-05,
      "loss": 1.3534,
      "num_input_tokens_seen": 26766472,
      "step": 46145
    },
    {
      "epoch": 6.873696753053322,
      "grad_norm": 9.3125,
      "learning_rate": 4.1488408784583664e-05,
      "loss": 1.3915,
      "num_input_tokens_seen": 26769608,
      "step": 46150
    },
    {
      "epoch": 6.8744414655942805,
      "grad_norm": 6.65625,
      "learning_rate": 4.148596614401544e-05,
      "loss": 0.9953,
      "num_input_tokens_seen": 26772520,
      "step": 46155
    },
    {
      "epoch": 6.87518617813524,
      "grad_norm": 7.71875,
      "learning_rate": 4.148352322493357e-05,
      "loss": 1.1885,
      "num_input_tokens_seen": 26775496,
      "step": 46160
    },
    {
      "epoch": 6.875930890676199,
      "grad_norm": 9.375,
      "learning_rate": 4.148108002737933e-05,
      "loss": 1.3103,
      "num_input_tokens_seen": 26778568,
      "step": 46165
    },
    {
      "epoch": 6.8766756032171585,
      "grad_norm": 17.25,
      "learning_rate": 4.147863655139399e-05,
      "loss": 1.1897,
      "num_input_tokens_seen": 26781320,
      "step": 46170
    },
    {
      "epoch": 6.877420315758117,
      "grad_norm": 10.4375,
      "learning_rate": 4.1476192797018836e-05,
      "loss": 0.7978,
      "num_input_tokens_seen": 26784008,
      "step": 46175
    },
    {
      "epoch": 6.878165028299077,
      "grad_norm": 8.5,
      "learning_rate": 4.147374876429515e-05,
      "loss": 1.0652,
      "num_input_tokens_seen": 26786824,
      "step": 46180
    },
    {
      "epoch": 6.878909740840036,
      "grad_norm": 9.125,
      "learning_rate": 4.1471304453264225e-05,
      "loss": 0.8245,
      "num_input_tokens_seen": 26789800,
      "step": 46185
    },
    {
      "epoch": 6.879654453380995,
      "grad_norm": 8.9375,
      "learning_rate": 4.1468859863967345e-05,
      "loss": 1.0291,
      "num_input_tokens_seen": 26792712,
      "step": 46190
    },
    {
      "epoch": 6.880399165921954,
      "grad_norm": 7.40625,
      "learning_rate": 4.1466414996445824e-05,
      "loss": 1.3872,
      "num_input_tokens_seen": 26795592,
      "step": 46195
    },
    {
      "epoch": 6.881143878462913,
      "grad_norm": 14.25,
      "learning_rate": 4.146396985074095e-05,
      "loss": 1.2667,
      "num_input_tokens_seen": 26798440,
      "step": 46200
    },
    {
      "epoch": 6.8818885910038725,
      "grad_norm": 13.5,
      "learning_rate": 4.146152442689405e-05,
      "loss": 1.0197,
      "num_input_tokens_seen": 26801224,
      "step": 46205
    },
    {
      "epoch": 6.882633303544832,
      "grad_norm": 17.125,
      "learning_rate": 4.1459078724946406e-05,
      "loss": 1.1008,
      "num_input_tokens_seen": 26803976,
      "step": 46210
    },
    {
      "epoch": 6.883378016085791,
      "grad_norm": 12.25,
      "learning_rate": 4.1456632744939375e-05,
      "loss": 1.0439,
      "num_input_tokens_seen": 26807112,
      "step": 46215
    },
    {
      "epoch": 6.88412272862675,
      "grad_norm": 5.125,
      "learning_rate": 4.145418648691425e-05,
      "loss": 0.8253,
      "num_input_tokens_seen": 26809864,
      "step": 46220
    },
    {
      "epoch": 6.884867441167709,
      "grad_norm": 7.53125,
      "learning_rate": 4.1451739950912365e-05,
      "loss": 0.8289,
      "num_input_tokens_seen": 26812552,
      "step": 46225
    },
    {
      "epoch": 6.885612153708668,
      "grad_norm": 9.0,
      "learning_rate": 4.144929313697506e-05,
      "loss": 1.3922,
      "num_input_tokens_seen": 26815304,
      "step": 46230
    },
    {
      "epoch": 6.886356866249628,
      "grad_norm": 15.4375,
      "learning_rate": 4.144684604514366e-05,
      "loss": 1.68,
      "num_input_tokens_seen": 26818184,
      "step": 46235
    },
    {
      "epoch": 6.8871015787905865,
      "grad_norm": 13.4375,
      "learning_rate": 4.144439867545952e-05,
      "loss": 0.9219,
      "num_input_tokens_seen": 26820808,
      "step": 46240
    },
    {
      "epoch": 6.887846291331546,
      "grad_norm": 10.3125,
      "learning_rate": 4.144195102796398e-05,
      "loss": 1.305,
      "num_input_tokens_seen": 26823784,
      "step": 46245
    },
    {
      "epoch": 6.888591003872505,
      "grad_norm": 10.1875,
      "learning_rate": 4.143950310269837e-05,
      "loss": 1.0134,
      "num_input_tokens_seen": 26826632,
      "step": 46250
    },
    {
      "epoch": 6.8893357164134645,
      "grad_norm": 13.8125,
      "learning_rate": 4.143705489970408e-05,
      "loss": 1.2075,
      "num_input_tokens_seen": 26829768,
      "step": 46255
    },
    {
      "epoch": 6.890080428954423,
      "grad_norm": 11.0,
      "learning_rate": 4.143460641902245e-05,
      "loss": 1.2249,
      "num_input_tokens_seen": 26832744,
      "step": 46260
    },
    {
      "epoch": 6.890825141495383,
      "grad_norm": 5.59375,
      "learning_rate": 4.143215766069484e-05,
      "loss": 0.9151,
      "num_input_tokens_seen": 26836040,
      "step": 46265
    },
    {
      "epoch": 6.891569854036342,
      "grad_norm": 7.3125,
      "learning_rate": 4.142970862476264e-05,
      "loss": 1.1157,
      "num_input_tokens_seen": 26838952,
      "step": 46270
    },
    {
      "epoch": 6.892314566577301,
      "grad_norm": 7.4375,
      "learning_rate": 4.14272593112672e-05,
      "loss": 1.2224,
      "num_input_tokens_seen": 26841928,
      "step": 46275
    },
    {
      "epoch": 6.89305927911826,
      "grad_norm": 7.40625,
      "learning_rate": 4.142480972024991e-05,
      "loss": 0.9942,
      "num_input_tokens_seen": 26844776,
      "step": 46280
    },
    {
      "epoch": 6.89380399165922,
      "grad_norm": 10.4375,
      "learning_rate": 4.142235985175217e-05,
      "loss": 1.0838,
      "num_input_tokens_seen": 26847720,
      "step": 46285
    },
    {
      "epoch": 6.8945487042001785,
      "grad_norm": 12.6875,
      "learning_rate": 4.141990970581534e-05,
      "loss": 1.0493,
      "num_input_tokens_seen": 26850408,
      "step": 46290
    },
    {
      "epoch": 6.895293416741138,
      "grad_norm": 10.625,
      "learning_rate": 4.141745928248082e-05,
      "loss": 1.0106,
      "num_input_tokens_seen": 26853256,
      "step": 46295
    },
    {
      "epoch": 6.896038129282097,
      "grad_norm": 9.125,
      "learning_rate": 4.141500858179002e-05,
      "loss": 1.2557,
      "num_input_tokens_seen": 26856296,
      "step": 46300
    },
    {
      "epoch": 6.896782841823057,
      "grad_norm": 25.125,
      "learning_rate": 4.141255760378432e-05,
      "loss": 1.401,
      "num_input_tokens_seen": 26859016,
      "step": 46305
    },
    {
      "epoch": 6.897527554364015,
      "grad_norm": 14.4375,
      "learning_rate": 4.141010634850515e-05,
      "loss": 1.2043,
      "num_input_tokens_seen": 26861992,
      "step": 46310
    },
    {
      "epoch": 6.898272266904975,
      "grad_norm": 4.75,
      "learning_rate": 4.140765481599391e-05,
      "loss": 1.1826,
      "num_input_tokens_seen": 26864808,
      "step": 46315
    },
    {
      "epoch": 6.899016979445934,
      "grad_norm": 6.59375,
      "learning_rate": 4.1405203006292014e-05,
      "loss": 0.8421,
      "num_input_tokens_seen": 26867592,
      "step": 46320
    },
    {
      "epoch": 6.899761691986893,
      "grad_norm": 9.625,
      "learning_rate": 4.1402750919440894e-05,
      "loss": 0.8293,
      "num_input_tokens_seen": 26870568,
      "step": 46325
    },
    {
      "epoch": 6.900506404527852,
      "grad_norm": 6.65625,
      "learning_rate": 4.140029855548196e-05,
      "loss": 1.3988,
      "num_input_tokens_seen": 26873416,
      "step": 46330
    },
    {
      "epoch": 6.901251117068812,
      "grad_norm": 9.375,
      "learning_rate": 4.1397845914456656e-05,
      "loss": 1.2131,
      "num_input_tokens_seen": 26876168,
      "step": 46335
    },
    {
      "epoch": 6.9019958296097705,
      "grad_norm": 6.8125,
      "learning_rate": 4.139539299640641e-05,
      "loss": 1.08,
      "num_input_tokens_seen": 26879336,
      "step": 46340
    },
    {
      "epoch": 6.90274054215073,
      "grad_norm": 8.4375,
      "learning_rate": 4.139293980137267e-05,
      "loss": 0.9332,
      "num_input_tokens_seen": 26882728,
      "step": 46345
    },
    {
      "epoch": 6.903485254691689,
      "grad_norm": 7.75,
      "learning_rate": 4.139048632939686e-05,
      "loss": 1.0466,
      "num_input_tokens_seen": 26886088,
      "step": 46350
    },
    {
      "epoch": 6.904229967232649,
      "grad_norm": 8.6875,
      "learning_rate": 4.138803258052045e-05,
      "loss": 1.0662,
      "num_input_tokens_seen": 26889064,
      "step": 46355
    },
    {
      "epoch": 6.904974679773607,
      "grad_norm": 7.25,
      "learning_rate": 4.138557855478489e-05,
      "loss": 1.0361,
      "num_input_tokens_seen": 26891976,
      "step": 46360
    },
    {
      "epoch": 6.905719392314566,
      "grad_norm": 5.84375,
      "learning_rate": 4.1383124252231625e-05,
      "loss": 1.2178,
      "num_input_tokens_seen": 26894984,
      "step": 46365
    },
    {
      "epoch": 6.906464104855526,
      "grad_norm": 8.1875,
      "learning_rate": 4.138066967290213e-05,
      "loss": 1.1643,
      "num_input_tokens_seen": 26897768,
      "step": 46370
    },
    {
      "epoch": 6.907208817396485,
      "grad_norm": 10.625,
      "learning_rate": 4.137821481683787e-05,
      "loss": 1.108,
      "num_input_tokens_seen": 26900776,
      "step": 46375
    },
    {
      "epoch": 6.907953529937444,
      "grad_norm": 11.375,
      "learning_rate": 4.1375759684080314e-05,
      "loss": 1.0827,
      "num_input_tokens_seen": 26903624,
      "step": 46380
    },
    {
      "epoch": 6.908698242478403,
      "grad_norm": 6.90625,
      "learning_rate": 4.137330427467094e-05,
      "loss": 1.0776,
      "num_input_tokens_seen": 26906472,
      "step": 46385
    },
    {
      "epoch": 6.909442955019363,
      "grad_norm": 5.84375,
      "learning_rate": 4.137084858865124e-05,
      "loss": 1.0713,
      "num_input_tokens_seen": 26909448,
      "step": 46390
    },
    {
      "epoch": 6.910187667560322,
      "grad_norm": 10.625,
      "learning_rate": 4.1368392626062685e-05,
      "loss": 0.9735,
      "num_input_tokens_seen": 26912328,
      "step": 46395
    },
    {
      "epoch": 6.910932380101281,
      "grad_norm": 16.25,
      "learning_rate": 4.1365936386946776e-05,
      "loss": 1.3897,
      "num_input_tokens_seen": 26915016,
      "step": 46400
    },
    {
      "epoch": 6.91167709264224,
      "grad_norm": 16.625,
      "learning_rate": 4.1363479871345e-05,
      "loss": 1.0248,
      "num_input_tokens_seen": 26918440,
      "step": 46405
    },
    {
      "epoch": 6.912421805183199,
      "grad_norm": 10.5,
      "learning_rate": 4.1361023079298874e-05,
      "loss": 0.8714,
      "num_input_tokens_seen": 26921064,
      "step": 46410
    },
    {
      "epoch": 6.913166517724158,
      "grad_norm": 13.4375,
      "learning_rate": 4.135856601084988e-05,
      "loss": 1.6199,
      "num_input_tokens_seen": 26923752,
      "step": 46415
    },
    {
      "epoch": 6.913911230265118,
      "grad_norm": 12.5625,
      "learning_rate": 4.135610866603955e-05,
      "loss": 1.3511,
      "num_input_tokens_seen": 26926600,
      "step": 46420
    },
    {
      "epoch": 6.9146559428060765,
      "grad_norm": 13.25,
      "learning_rate": 4.135365104490938e-05,
      "loss": 0.764,
      "num_input_tokens_seen": 26929064,
      "step": 46425
    },
    {
      "epoch": 6.915400655347036,
      "grad_norm": 17.25,
      "learning_rate": 4.1351193147500887e-05,
      "loss": 1.1291,
      "num_input_tokens_seen": 26931592,
      "step": 46430
    },
    {
      "epoch": 6.916145367887995,
      "grad_norm": 8.375,
      "learning_rate": 4.134873497385562e-05,
      "loss": 1.064,
      "num_input_tokens_seen": 26934472,
      "step": 46435
    },
    {
      "epoch": 6.916890080428955,
      "grad_norm": 12.75,
      "learning_rate": 4.1346276524015085e-05,
      "loss": 1.1611,
      "num_input_tokens_seen": 26937512,
      "step": 46440
    },
    {
      "epoch": 6.917634792969913,
      "grad_norm": 11.8125,
      "learning_rate": 4.1343817798020824e-05,
      "loss": 0.9562,
      "num_input_tokens_seen": 26941064,
      "step": 46445
    },
    {
      "epoch": 6.918379505510873,
      "grad_norm": 12.25,
      "learning_rate": 4.1341358795914375e-05,
      "loss": 0.9748,
      "num_input_tokens_seen": 26943976,
      "step": 46450
    },
    {
      "epoch": 6.919124218051832,
      "grad_norm": 12.4375,
      "learning_rate": 4.133889951773727e-05,
      "loss": 0.8498,
      "num_input_tokens_seen": 26947048,
      "step": 46455
    },
    {
      "epoch": 6.919868930592791,
      "grad_norm": 8.25,
      "learning_rate": 4.133643996353107e-05,
      "loss": 1.4968,
      "num_input_tokens_seen": 26950088,
      "step": 46460
    },
    {
      "epoch": 6.92061364313375,
      "grad_norm": 7.78125,
      "learning_rate": 4.1333980133337324e-05,
      "loss": 1.2835,
      "num_input_tokens_seen": 26952648,
      "step": 46465
    },
    {
      "epoch": 6.92135835567471,
      "grad_norm": 12.375,
      "learning_rate": 4.133152002719758e-05,
      "loss": 0.8934,
      "num_input_tokens_seen": 26955432,
      "step": 46470
    },
    {
      "epoch": 6.922103068215669,
      "grad_norm": 6.5,
      "learning_rate": 4.13290596451534e-05,
      "loss": 1.3785,
      "num_input_tokens_seen": 26958408,
      "step": 46475
    },
    {
      "epoch": 6.922847780756628,
      "grad_norm": 7.90625,
      "learning_rate": 4.1326598987246356e-05,
      "loss": 1.004,
      "num_input_tokens_seen": 26961448,
      "step": 46480
    },
    {
      "epoch": 6.923592493297587,
      "grad_norm": 7.59375,
      "learning_rate": 4.132413805351802e-05,
      "loss": 1.3349,
      "num_input_tokens_seen": 26964616,
      "step": 46485
    },
    {
      "epoch": 6.924337205838547,
      "grad_norm": 17.375,
      "learning_rate": 4.1321676844009957e-05,
      "loss": 1.2501,
      "num_input_tokens_seen": 26967944,
      "step": 46490
    },
    {
      "epoch": 6.925081918379505,
      "grad_norm": 8.8125,
      "learning_rate": 4.1319215358763756e-05,
      "loss": 1.1098,
      "num_input_tokens_seen": 26970920,
      "step": 46495
    },
    {
      "epoch": 6.925826630920465,
      "grad_norm": 12.5625,
      "learning_rate": 4.1316753597821e-05,
      "loss": 1.1871,
      "num_input_tokens_seen": 26973704,
      "step": 46500
    },
    {
      "epoch": 6.926571343461424,
      "grad_norm": 8.5625,
      "learning_rate": 4.1314291561223276e-05,
      "loss": 0.877,
      "num_input_tokens_seen": 26976520,
      "step": 46505
    },
    {
      "epoch": 6.927316056002383,
      "grad_norm": 7.875,
      "learning_rate": 4.131182924901217e-05,
      "loss": 1.5921,
      "num_input_tokens_seen": 26979464,
      "step": 46510
    },
    {
      "epoch": 6.928060768543342,
      "grad_norm": 28.0,
      "learning_rate": 4.1309366661229286e-05,
      "loss": 1.2109,
      "num_input_tokens_seen": 26982376,
      "step": 46515
    },
    {
      "epoch": 6.928805481084302,
      "grad_norm": 13.75,
      "learning_rate": 4.130690379791623e-05,
      "loss": 1.2012,
      "num_input_tokens_seen": 26984968,
      "step": 46520
    },
    {
      "epoch": 6.929550193625261,
      "grad_norm": 8.75,
      "learning_rate": 4.1304440659114615e-05,
      "loss": 0.9851,
      "num_input_tokens_seen": 26987976,
      "step": 46525
    },
    {
      "epoch": 6.930294906166219,
      "grad_norm": 12.9375,
      "learning_rate": 4.130197724486604e-05,
      "loss": 1.1843,
      "num_input_tokens_seen": 26990984,
      "step": 46530
    },
    {
      "epoch": 6.931039618707179,
      "grad_norm": 11.9375,
      "learning_rate": 4.1299513555212135e-05,
      "loss": 1.25,
      "num_input_tokens_seen": 26993608,
      "step": 46535
    },
    {
      "epoch": 6.931784331248139,
      "grad_norm": 12.3125,
      "learning_rate": 4.129704959019451e-05,
      "loss": 1.1385,
      "num_input_tokens_seen": 26996392,
      "step": 46540
    },
    {
      "epoch": 6.932529043789097,
      "grad_norm": 7.65625,
      "learning_rate": 4.129458534985479e-05,
      "loss": 0.9253,
      "num_input_tokens_seen": 26999368,
      "step": 46545
    },
    {
      "epoch": 6.933273756330056,
      "grad_norm": 12.4375,
      "learning_rate": 4.1292120834234624e-05,
      "loss": 0.8566,
      "num_input_tokens_seen": 27002216,
      "step": 46550
    },
    {
      "epoch": 6.934018468871016,
      "grad_norm": 8.5625,
      "learning_rate": 4.128965604337563e-05,
      "loss": 1.052,
      "num_input_tokens_seen": 27004936,
      "step": 46555
    },
    {
      "epoch": 6.9347631814119755,
      "grad_norm": 7.1875,
      "learning_rate": 4.128719097731945e-05,
      "loss": 1.0876,
      "num_input_tokens_seen": 27007784,
      "step": 46560
    },
    {
      "epoch": 6.935507893952934,
      "grad_norm": 12.625,
      "learning_rate": 4.1284725636107726e-05,
      "loss": 1.2676,
      "num_input_tokens_seen": 27010696,
      "step": 46565
    },
    {
      "epoch": 6.936252606493893,
      "grad_norm": 9.625,
      "learning_rate": 4.128226001978213e-05,
      "loss": 0.8355,
      "num_input_tokens_seen": 27013608,
      "step": 46570
    },
    {
      "epoch": 6.936997319034853,
      "grad_norm": 7.78125,
      "learning_rate": 4.127979412838428e-05,
      "loss": 1.1297,
      "num_input_tokens_seen": 27016264,
      "step": 46575
    },
    {
      "epoch": 6.937742031575811,
      "grad_norm": 9.0,
      "learning_rate": 4.127732796195587e-05,
      "loss": 1.198,
      "num_input_tokens_seen": 27019080,
      "step": 46580
    },
    {
      "epoch": 6.938486744116771,
      "grad_norm": 6.78125,
      "learning_rate": 4.127486152053854e-05,
      "loss": 0.9005,
      "num_input_tokens_seen": 27022024,
      "step": 46585
    },
    {
      "epoch": 6.93923145665773,
      "grad_norm": 6.03125,
      "learning_rate": 4.1272394804173966e-05,
      "loss": 0.8491,
      "num_input_tokens_seen": 27025000,
      "step": 46590
    },
    {
      "epoch": 6.939976169198689,
      "grad_norm": 5.90625,
      "learning_rate": 4.126992781290382e-05,
      "loss": 1.3919,
      "num_input_tokens_seen": 27027720,
      "step": 46595
    },
    {
      "epoch": 6.940720881739648,
      "grad_norm": 20.625,
      "learning_rate": 4.1267460546769784e-05,
      "loss": 1.102,
      "num_input_tokens_seen": 27030888,
      "step": 46600
    },
    {
      "epoch": 6.941465594280608,
      "grad_norm": 6.8125,
      "learning_rate": 4.126499300581353e-05,
      "loss": 1.0138,
      "num_input_tokens_seen": 27033928,
      "step": 46605
    },
    {
      "epoch": 6.942210306821567,
      "grad_norm": 11.8125,
      "learning_rate": 4.1262525190076763e-05,
      "loss": 0.6807,
      "num_input_tokens_seen": 27036456,
      "step": 46610
    },
    {
      "epoch": 6.942955019362526,
      "grad_norm": 10.625,
      "learning_rate": 4.1260057099601145e-05,
      "loss": 1.429,
      "num_input_tokens_seen": 27039624,
      "step": 46615
    },
    {
      "epoch": 6.943699731903485,
      "grad_norm": 8.375,
      "learning_rate": 4.12575887344284e-05,
      "loss": 1.1365,
      "num_input_tokens_seen": 27042472,
      "step": 46620
    },
    {
      "epoch": 6.944444444444445,
      "grad_norm": 11.4375,
      "learning_rate": 4.125512009460021e-05,
      "loss": 0.7214,
      "num_input_tokens_seen": 27045672,
      "step": 46625
    },
    {
      "epoch": 6.945189156985403,
      "grad_norm": 9.625,
      "learning_rate": 4.125265118015829e-05,
      "loss": 1.094,
      "num_input_tokens_seen": 27048776,
      "step": 46630
    },
    {
      "epoch": 6.945933869526363,
      "grad_norm": 11.8125,
      "learning_rate": 4.125018199114434e-05,
      "loss": 1.018,
      "num_input_tokens_seen": 27051816,
      "step": 46635
    },
    {
      "epoch": 6.946678582067322,
      "grad_norm": 6.90625,
      "learning_rate": 4.124771252760009e-05,
      "loss": 1.0045,
      "num_input_tokens_seen": 27054824,
      "step": 46640
    },
    {
      "epoch": 6.9474232946082815,
      "grad_norm": 9.1875,
      "learning_rate": 4.124524278956725e-05,
      "loss": 0.8024,
      "num_input_tokens_seen": 27057800,
      "step": 46645
    },
    {
      "epoch": 6.94816800714924,
      "grad_norm": 10.4375,
      "learning_rate": 4.1242772777087536e-05,
      "loss": 1.3222,
      "num_input_tokens_seen": 27060520,
      "step": 46650
    },
    {
      "epoch": 6.9489127196902,
      "grad_norm": 6.46875,
      "learning_rate": 4.124030249020269e-05,
      "loss": 1.5897,
      "num_input_tokens_seen": 27063144,
      "step": 46655
    },
    {
      "epoch": 6.949657432231159,
      "grad_norm": 11.375,
      "learning_rate": 4.123783192895444e-05,
      "loss": 1.1016,
      "num_input_tokens_seen": 27066120,
      "step": 46660
    },
    {
      "epoch": 6.950402144772118,
      "grad_norm": 6.15625,
      "learning_rate": 4.1235361093384523e-05,
      "loss": 1.3394,
      "num_input_tokens_seen": 27068904,
      "step": 46665
    },
    {
      "epoch": 6.951146857313077,
      "grad_norm": 5.8125,
      "learning_rate": 4.123288998353468e-05,
      "loss": 0.8419,
      "num_input_tokens_seen": 27071688,
      "step": 46670
    },
    {
      "epoch": 6.951891569854037,
      "grad_norm": 9.0625,
      "learning_rate": 4.123041859944666e-05,
      "loss": 0.8097,
      "num_input_tokens_seen": 27074664,
      "step": 46675
    },
    {
      "epoch": 6.952636282394995,
      "grad_norm": 10.9375,
      "learning_rate": 4.122794694116221e-05,
      "loss": 1.2509,
      "num_input_tokens_seen": 27077480,
      "step": 46680
    },
    {
      "epoch": 6.953380994935955,
      "grad_norm": 10.25,
      "learning_rate": 4.122547500872309e-05,
      "loss": 0.9412,
      "num_input_tokens_seen": 27080488,
      "step": 46685
    },
    {
      "epoch": 6.954125707476914,
      "grad_norm": 5.46875,
      "learning_rate": 4.122300280217107e-05,
      "loss": 0.9048,
      "num_input_tokens_seen": 27083176,
      "step": 46690
    },
    {
      "epoch": 6.9548704200178735,
      "grad_norm": 8.5625,
      "learning_rate": 4.1220530321547894e-05,
      "loss": 0.8346,
      "num_input_tokens_seen": 27085896,
      "step": 46695
    },
    {
      "epoch": 6.955615132558832,
      "grad_norm": 12.9375,
      "learning_rate": 4.121805756689535e-05,
      "loss": 0.9875,
      "num_input_tokens_seen": 27088936,
      "step": 46700
    },
    {
      "epoch": 6.956359845099792,
      "grad_norm": 11.75,
      "learning_rate": 4.1215584538255206e-05,
      "loss": 1.1288,
      "num_input_tokens_seen": 27091752,
      "step": 46705
    },
    {
      "epoch": 6.957104557640751,
      "grad_norm": 18.125,
      "learning_rate": 4.121311123566924e-05,
      "loss": 1.2328,
      "num_input_tokens_seen": 27095048,
      "step": 46710
    },
    {
      "epoch": 6.957849270181709,
      "grad_norm": 8.5625,
      "learning_rate": 4.121063765917924e-05,
      "loss": 1.775,
      "num_input_tokens_seen": 27097800,
      "step": 46715
    },
    {
      "epoch": 6.958593982722669,
      "grad_norm": 9.25,
      "learning_rate": 4.120816380882699e-05,
      "loss": 1.2071,
      "num_input_tokens_seen": 27100680,
      "step": 46720
    },
    {
      "epoch": 6.959338695263629,
      "grad_norm": 8.125,
      "learning_rate": 4.120568968465429e-05,
      "loss": 0.9467,
      "num_input_tokens_seen": 27103400,
      "step": 46725
    },
    {
      "epoch": 6.9600834078045875,
      "grad_norm": 12.1875,
      "learning_rate": 4.120321528670293e-05,
      "loss": 0.9883,
      "num_input_tokens_seen": 27106312,
      "step": 46730
    },
    {
      "epoch": 6.960828120345546,
      "grad_norm": 6.15625,
      "learning_rate": 4.120074061501472e-05,
      "loss": 1.0288,
      "num_input_tokens_seen": 27109160,
      "step": 46735
    },
    {
      "epoch": 6.961572832886506,
      "grad_norm": 11.5,
      "learning_rate": 4.1198265669631464e-05,
      "loss": 0.7474,
      "num_input_tokens_seen": 27112040,
      "step": 46740
    },
    {
      "epoch": 6.962317545427465,
      "grad_norm": 9.5,
      "learning_rate": 4.119579045059496e-05,
      "loss": 1.1275,
      "num_input_tokens_seen": 27115144,
      "step": 46745
    },
    {
      "epoch": 6.963062257968424,
      "grad_norm": 11.0625,
      "learning_rate": 4.119331495794705e-05,
      "loss": 0.8471,
      "num_input_tokens_seen": 27118440,
      "step": 46750
    },
    {
      "epoch": 6.963806970509383,
      "grad_norm": 12.0625,
      "learning_rate": 4.119083919172954e-05,
      "loss": 1.2987,
      "num_input_tokens_seen": 27121224,
      "step": 46755
    },
    {
      "epoch": 6.964551683050343,
      "grad_norm": 13.875,
      "learning_rate": 4.118836315198425e-05,
      "loss": 1.1974,
      "num_input_tokens_seen": 27123880,
      "step": 46760
    },
    {
      "epoch": 6.965296395591301,
      "grad_norm": 7.46875,
      "learning_rate": 4.118588683875303e-05,
      "loss": 1.2902,
      "num_input_tokens_seen": 27126728,
      "step": 46765
    },
    {
      "epoch": 6.966041108132261,
      "grad_norm": 12.5625,
      "learning_rate": 4.11834102520777e-05,
      "loss": 0.778,
      "num_input_tokens_seen": 27129832,
      "step": 46770
    },
    {
      "epoch": 6.96678582067322,
      "grad_norm": 8.9375,
      "learning_rate": 4.118093339200009e-05,
      "loss": 1.0899,
      "num_input_tokens_seen": 27133320,
      "step": 46775
    },
    {
      "epoch": 6.9675305332141795,
      "grad_norm": 16.75,
      "learning_rate": 4.1178456258562064e-05,
      "loss": 0.9131,
      "num_input_tokens_seen": 27137224,
      "step": 46780
    },
    {
      "epoch": 6.968275245755138,
      "grad_norm": 13.0625,
      "learning_rate": 4.117597885180546e-05,
      "loss": 0.9058,
      "num_input_tokens_seen": 27140360,
      "step": 46785
    },
    {
      "epoch": 6.969019958296098,
      "grad_norm": 7.65625,
      "learning_rate": 4.117350117177214e-05,
      "loss": 1.1717,
      "num_input_tokens_seen": 27143048,
      "step": 46790
    },
    {
      "epoch": 6.969764670837057,
      "grad_norm": 15.25,
      "learning_rate": 4.1171023218503945e-05,
      "loss": 0.9676,
      "num_input_tokens_seen": 27145672,
      "step": 46795
    },
    {
      "epoch": 6.970509383378016,
      "grad_norm": 16.25,
      "learning_rate": 4.1168544992042756e-05,
      "loss": 1.3275,
      "num_input_tokens_seen": 27148584,
      "step": 46800
    },
    {
      "epoch": 6.971254095918975,
      "grad_norm": 13.1875,
      "learning_rate": 4.116606649243043e-05,
      "loss": 1.1867,
      "num_input_tokens_seen": 27151400,
      "step": 46805
    },
    {
      "epoch": 6.971998808459935,
      "grad_norm": 7.65625,
      "learning_rate": 4.116358771970885e-05,
      "loss": 0.9466,
      "num_input_tokens_seen": 27154088,
      "step": 46810
    },
    {
      "epoch": 6.9727435210008935,
      "grad_norm": 19.375,
      "learning_rate": 4.1161108673919874e-05,
      "loss": 0.9483,
      "num_input_tokens_seen": 27156968,
      "step": 46815
    },
    {
      "epoch": 6.973488233541853,
      "grad_norm": 8.4375,
      "learning_rate": 4.115862935510539e-05,
      "loss": 1.1129,
      "num_input_tokens_seen": 27159912,
      "step": 46820
    },
    {
      "epoch": 6.974232946082812,
      "grad_norm": 4.65625,
      "learning_rate": 4.1156149763307296e-05,
      "loss": 0.9807,
      "num_input_tokens_seen": 27162760,
      "step": 46825
    },
    {
      "epoch": 6.9749776586237715,
      "grad_norm": 7.625,
      "learning_rate": 4.115366989856746e-05,
      "loss": 0.9297,
      "num_input_tokens_seen": 27166152,
      "step": 46830
    },
    {
      "epoch": 6.97572237116473,
      "grad_norm": 7.25,
      "learning_rate": 4.11511897609278e-05,
      "loss": 0.9836,
      "num_input_tokens_seen": 27169000,
      "step": 46835
    },
    {
      "epoch": 6.97646708370569,
      "grad_norm": 6.78125,
      "learning_rate": 4.1148709350430194e-05,
      "loss": 0.8666,
      "num_input_tokens_seen": 27172104,
      "step": 46840
    },
    {
      "epoch": 6.977211796246649,
      "grad_norm": 9.0625,
      "learning_rate": 4.114622866711657e-05,
      "loss": 0.9736,
      "num_input_tokens_seen": 27175144,
      "step": 46845
    },
    {
      "epoch": 6.977956508787608,
      "grad_norm": 10.375,
      "learning_rate": 4.114374771102881e-05,
      "loss": 1.1487,
      "num_input_tokens_seen": 27177992,
      "step": 46850
    },
    {
      "epoch": 6.978701221328567,
      "grad_norm": 8.5,
      "learning_rate": 4.114126648220884e-05,
      "loss": 0.9214,
      "num_input_tokens_seen": 27180840,
      "step": 46855
    },
    {
      "epoch": 6.979445933869527,
      "grad_norm": 6.28125,
      "learning_rate": 4.1138784980698585e-05,
      "loss": 1.2963,
      "num_input_tokens_seen": 27183624,
      "step": 46860
    },
    {
      "epoch": 6.9801906464104855,
      "grad_norm": 6.15625,
      "learning_rate": 4.113630320653996e-05,
      "loss": 1.3221,
      "num_input_tokens_seen": 27186408,
      "step": 46865
    },
    {
      "epoch": 6.980935358951445,
      "grad_norm": 11.3125,
      "learning_rate": 4.113382115977489e-05,
      "loss": 0.8821,
      "num_input_tokens_seen": 27189352,
      "step": 46870
    },
    {
      "epoch": 6.981680071492404,
      "grad_norm": 7.84375,
      "learning_rate": 4.113133884044531e-05,
      "loss": 0.8413,
      "num_input_tokens_seen": 27192136,
      "step": 46875
    },
    {
      "epoch": 6.982424784033363,
      "grad_norm": 10.1875,
      "learning_rate": 4.112885624859316e-05,
      "loss": 0.8675,
      "num_input_tokens_seen": 27195304,
      "step": 46880
    },
    {
      "epoch": 6.983169496574322,
      "grad_norm": 13.1875,
      "learning_rate": 4.1126373384260365e-05,
      "loss": 0.9991,
      "num_input_tokens_seen": 27198344,
      "step": 46885
    },
    {
      "epoch": 6.983914209115282,
      "grad_norm": 6.625,
      "learning_rate": 4.112389024748889e-05,
      "loss": 1.1553,
      "num_input_tokens_seen": 27201096,
      "step": 46890
    },
    {
      "epoch": 6.984658921656241,
      "grad_norm": 14.5625,
      "learning_rate": 4.112140683832068e-05,
      "loss": 1.3006,
      "num_input_tokens_seen": 27204136,
      "step": 46895
    },
    {
      "epoch": 6.9854036341971995,
      "grad_norm": 13.6875,
      "learning_rate": 4.1118923156797684e-05,
      "loss": 1.3742,
      "num_input_tokens_seen": 27206792,
      "step": 46900
    },
    {
      "epoch": 6.986148346738159,
      "grad_norm": 9.3125,
      "learning_rate": 4.111643920296185e-05,
      "loss": 0.9082,
      "num_input_tokens_seen": 27209512,
      "step": 46905
    },
    {
      "epoch": 6.986893059279119,
      "grad_norm": 20.5,
      "learning_rate": 4.1113954976855174e-05,
      "loss": 1.3466,
      "num_input_tokens_seen": 27212680,
      "step": 46910
    },
    {
      "epoch": 6.9876377718200775,
      "grad_norm": 7.375,
      "learning_rate": 4.11114704785196e-05,
      "loss": 0.9638,
      "num_input_tokens_seen": 27215432,
      "step": 46915
    },
    {
      "epoch": 6.988382484361036,
      "grad_norm": 4.03125,
      "learning_rate": 4.1108985707997105e-05,
      "loss": 0.9338,
      "num_input_tokens_seen": 27218216,
      "step": 46920
    },
    {
      "epoch": 6.989127196901996,
      "grad_norm": 7.59375,
      "learning_rate": 4.110650066532967e-05,
      "loss": 0.9332,
      "num_input_tokens_seen": 27221288,
      "step": 46925
    },
    {
      "epoch": 6.989871909442955,
      "grad_norm": 7.5625,
      "learning_rate": 4.1104015350559275e-05,
      "loss": 1.0706,
      "num_input_tokens_seen": 27224200,
      "step": 46930
    },
    {
      "epoch": 6.990616621983914,
      "grad_norm": 13.3125,
      "learning_rate": 4.110152976372791e-05,
      "loss": 1.2227,
      "num_input_tokens_seen": 27227400,
      "step": 46935
    },
    {
      "epoch": 6.991361334524873,
      "grad_norm": 12.8125,
      "learning_rate": 4.1099043904877564e-05,
      "loss": 1.3663,
      "num_input_tokens_seen": 27230344,
      "step": 46940
    },
    {
      "epoch": 6.992106047065833,
      "grad_norm": 5.65625,
      "learning_rate": 4.1096557774050235e-05,
      "loss": 1.127,
      "num_input_tokens_seen": 27233384,
      "step": 46945
    },
    {
      "epoch": 6.9928507596067915,
      "grad_norm": 18.25,
      "learning_rate": 4.1094071371287925e-05,
      "loss": 1.2373,
      "num_input_tokens_seen": 27236296,
      "step": 46950
    },
    {
      "epoch": 6.993595472147751,
      "grad_norm": 6.75,
      "learning_rate": 4.109158469663263e-05,
      "loss": 0.8978,
      "num_input_tokens_seen": 27239048,
      "step": 46955
    },
    {
      "epoch": 6.99434018468871,
      "grad_norm": 6.96875,
      "learning_rate": 4.108909775012637e-05,
      "loss": 1.1666,
      "num_input_tokens_seen": 27241768,
      "step": 46960
    },
    {
      "epoch": 6.9950848972296695,
      "grad_norm": 9.5625,
      "learning_rate": 4.1086610531811155e-05,
      "loss": 1.0463,
      "num_input_tokens_seen": 27244680,
      "step": 46965
    },
    {
      "epoch": 6.995829609770628,
      "grad_norm": 13.375,
      "learning_rate": 4.108412304172901e-05,
      "loss": 1.2091,
      "num_input_tokens_seen": 27247528,
      "step": 46970
    },
    {
      "epoch": 6.996574322311588,
      "grad_norm": 14.6875,
      "learning_rate": 4.1081635279921945e-05,
      "loss": 0.8711,
      "num_input_tokens_seen": 27250600,
      "step": 46975
    },
    {
      "epoch": 6.997319034852547,
      "grad_norm": 10.0,
      "learning_rate": 4.107914724643199e-05,
      "loss": 1.2484,
      "num_input_tokens_seen": 27253544,
      "step": 46980
    },
    {
      "epoch": 6.998063747393506,
      "grad_norm": 12.125,
      "learning_rate": 4.107665894130121e-05,
      "loss": 1.0055,
      "num_input_tokens_seen": 27256360,
      "step": 46985
    },
    {
      "epoch": 6.998808459934465,
      "grad_norm": 11.6875,
      "learning_rate": 4.107417036457159e-05,
      "loss": 0.9795,
      "num_input_tokens_seen": 27259592,
      "step": 46990
    },
    {
      "epoch": 6.999553172475425,
      "grad_norm": 8.1875,
      "learning_rate": 4.107168151628521e-05,
      "loss": 1.499,
      "num_input_tokens_seen": 27262472,
      "step": 46995
    },
    {
      "epoch": 7.0,
      "eval_loss": 1.0881797075271606,
      "eval_runtime": 43.6883,
      "eval_samples_per_second": 68.302,
      "eval_steps_per_second": 17.076,
      "num_input_tokens_seen": 27263880,
      "step": 46998
    },
    {
      "epoch": 7.0002978850163835,
      "grad_norm": 7.4375,
      "learning_rate": 4.10691923964841e-05,
      "loss": 0.8694,
      "num_input_tokens_seen": 27265128,
      "step": 47000
    },
    {
      "epoch": 7.001042597557343,
      "grad_norm": 11.25,
      "learning_rate": 4.106670300521033e-05,
      "loss": 0.9721,
      "num_input_tokens_seen": 27267944,
      "step": 47005
    },
    {
      "epoch": 7.001787310098302,
      "grad_norm": 8.625,
      "learning_rate": 4.106421334250593e-05,
      "loss": 1.0211,
      "num_input_tokens_seen": 27270632,
      "step": 47010
    },
    {
      "epoch": 7.0025320226392616,
      "grad_norm": 7.46875,
      "learning_rate": 4.106172340841298e-05,
      "loss": 0.9589,
      "num_input_tokens_seen": 27273832,
      "step": 47015
    },
    {
      "epoch": 7.00327673518022,
      "grad_norm": 5.4375,
      "learning_rate": 4.105923320297353e-05,
      "loss": 1.0436,
      "num_input_tokens_seen": 27276616,
      "step": 47020
    },
    {
      "epoch": 7.00402144772118,
      "grad_norm": 16.875,
      "learning_rate": 4.1056742726229655e-05,
      "loss": 1.2186,
      "num_input_tokens_seen": 27279464,
      "step": 47025
    },
    {
      "epoch": 7.004766160262139,
      "grad_norm": 9.1875,
      "learning_rate": 4.105425197822344e-05,
      "loss": 1.1408,
      "num_input_tokens_seen": 27282344,
      "step": 47030
    },
    {
      "epoch": 7.005510872803098,
      "grad_norm": 9.875,
      "learning_rate": 4.105176095899696e-05,
      "loss": 0.9639,
      "num_input_tokens_seen": 27285352,
      "step": 47035
    },
    {
      "epoch": 7.006255585344057,
      "grad_norm": 9.5625,
      "learning_rate": 4.104926966859227e-05,
      "loss": 0.8627,
      "num_input_tokens_seen": 27288392,
      "step": 47040
    },
    {
      "epoch": 7.007000297885017,
      "grad_norm": 6.84375,
      "learning_rate": 4.1046778107051495e-05,
      "loss": 1.2483,
      "num_input_tokens_seen": 27291144,
      "step": 47045
    },
    {
      "epoch": 7.0077450104259755,
      "grad_norm": 8.25,
      "learning_rate": 4.104428627441672e-05,
      "loss": 0.9322,
      "num_input_tokens_seen": 27294024,
      "step": 47050
    },
    {
      "epoch": 7.008489722966935,
      "grad_norm": 12.4375,
      "learning_rate": 4.104179417073002e-05,
      "loss": 0.8271,
      "num_input_tokens_seen": 27296808,
      "step": 47055
    },
    {
      "epoch": 7.009234435507894,
      "grad_norm": 8.875,
      "learning_rate": 4.103930179603352e-05,
      "loss": 1.1214,
      "num_input_tokens_seen": 27300264,
      "step": 47060
    },
    {
      "epoch": 7.009979148048854,
      "grad_norm": 10.1875,
      "learning_rate": 4.103680915036932e-05,
      "loss": 1.453,
      "num_input_tokens_seen": 27303432,
      "step": 47065
    },
    {
      "epoch": 7.010723860589812,
      "grad_norm": 12.1875,
      "learning_rate": 4.1034316233779526e-05,
      "loss": 1.4197,
      "num_input_tokens_seen": 27306408,
      "step": 47070
    },
    {
      "epoch": 7.011468573130771,
      "grad_norm": 10.0,
      "learning_rate": 4.103182304630625e-05,
      "loss": 1.0102,
      "num_input_tokens_seen": 27309096,
      "step": 47075
    },
    {
      "epoch": 7.012213285671731,
      "grad_norm": 7.25,
      "learning_rate": 4.102932958799163e-05,
      "loss": 1.1589,
      "num_input_tokens_seen": 27311880,
      "step": 47080
    },
    {
      "epoch": 7.0129579982126895,
      "grad_norm": 7.75,
      "learning_rate": 4.102683585887777e-05,
      "loss": 0.7696,
      "num_input_tokens_seen": 27314440,
      "step": 47085
    },
    {
      "epoch": 7.013702710753649,
      "grad_norm": 9.1875,
      "learning_rate": 4.102434185900681e-05,
      "loss": 0.852,
      "num_input_tokens_seen": 27317192,
      "step": 47090
    },
    {
      "epoch": 7.014447423294608,
      "grad_norm": 18.5,
      "learning_rate": 4.1021847588420876e-05,
      "loss": 1.1872,
      "num_input_tokens_seen": 27320264,
      "step": 47095
    },
    {
      "epoch": 7.0151921358355676,
      "grad_norm": 9.75,
      "learning_rate": 4.101935304716211e-05,
      "loss": 1.3252,
      "num_input_tokens_seen": 27322952,
      "step": 47100
    },
    {
      "epoch": 7.015936848376526,
      "grad_norm": 6.90625,
      "learning_rate": 4.101685823527266e-05,
      "loss": 0.9095,
      "num_input_tokens_seen": 27325672,
      "step": 47105
    },
    {
      "epoch": 7.016681560917486,
      "grad_norm": 7.09375,
      "learning_rate": 4.1014363152794664e-05,
      "loss": 1.2194,
      "num_input_tokens_seen": 27328616,
      "step": 47110
    },
    {
      "epoch": 7.017426273458445,
      "grad_norm": 7.0625,
      "learning_rate": 4.101186779977029e-05,
      "loss": 0.8241,
      "num_input_tokens_seen": 27331656,
      "step": 47115
    },
    {
      "epoch": 7.018170985999404,
      "grad_norm": 13.25,
      "learning_rate": 4.1009372176241675e-05,
      "loss": 1.534,
      "num_input_tokens_seen": 27334600,
      "step": 47120
    },
    {
      "epoch": 7.018915698540363,
      "grad_norm": 14.3125,
      "learning_rate": 4.100687628225099e-05,
      "loss": 1.1557,
      "num_input_tokens_seen": 27337384,
      "step": 47125
    },
    {
      "epoch": 7.019660411081323,
      "grad_norm": 8.125,
      "learning_rate": 4.10043801178404e-05,
      "loss": 1.0104,
      "num_input_tokens_seen": 27340328,
      "step": 47130
    },
    {
      "epoch": 7.0204051236222815,
      "grad_norm": 18.125,
      "learning_rate": 4.100188368305207e-05,
      "loss": 1.1972,
      "num_input_tokens_seen": 27343208,
      "step": 47135
    },
    {
      "epoch": 7.021149836163241,
      "grad_norm": 6.0,
      "learning_rate": 4.099938697792818e-05,
      "loss": 1.0696,
      "num_input_tokens_seen": 27345992,
      "step": 47140
    },
    {
      "epoch": 7.0218945487042,
      "grad_norm": 10.125,
      "learning_rate": 4.099689000251091e-05,
      "loss": 1.0143,
      "num_input_tokens_seen": 27348680,
      "step": 47145
    },
    {
      "epoch": 7.02263926124516,
      "grad_norm": 3.703125,
      "learning_rate": 4.0994392756842444e-05,
      "loss": 1.2716,
      "num_input_tokens_seen": 27351560,
      "step": 47150
    },
    {
      "epoch": 7.023383973786118,
      "grad_norm": 12.875,
      "learning_rate": 4.099189524096496e-05,
      "loss": 1.295,
      "num_input_tokens_seen": 27354824,
      "step": 47155
    },
    {
      "epoch": 7.024128686327078,
      "grad_norm": 10.625,
      "learning_rate": 4.098939745492066e-05,
      "loss": 1.218,
      "num_input_tokens_seen": 27357704,
      "step": 47160
    },
    {
      "epoch": 7.024873398868037,
      "grad_norm": 8.75,
      "learning_rate": 4.0986899398751754e-05,
      "loss": 1.0783,
      "num_input_tokens_seen": 27360616,
      "step": 47165
    },
    {
      "epoch": 7.025618111408996,
      "grad_norm": 11.875,
      "learning_rate": 4.098440107250042e-05,
      "loss": 1.1351,
      "num_input_tokens_seen": 27363688,
      "step": 47170
    },
    {
      "epoch": 7.026362823949955,
      "grad_norm": 7.1875,
      "learning_rate": 4.098190247620888e-05,
      "loss": 1.131,
      "num_input_tokens_seen": 27366504,
      "step": 47175
    },
    {
      "epoch": 7.027107536490915,
      "grad_norm": 8.375,
      "learning_rate": 4.097940360991934e-05,
      "loss": 0.9172,
      "num_input_tokens_seen": 27369128,
      "step": 47180
    },
    {
      "epoch": 7.0278522490318736,
      "grad_norm": 5.0,
      "learning_rate": 4.097690447367402e-05,
      "loss": 0.9557,
      "num_input_tokens_seen": 27372072,
      "step": 47185
    },
    {
      "epoch": 7.028596961572833,
      "grad_norm": 8.6875,
      "learning_rate": 4.097440506751513e-05,
      "loss": 0.7697,
      "num_input_tokens_seen": 27374984,
      "step": 47190
    },
    {
      "epoch": 7.029341674113792,
      "grad_norm": 14.0625,
      "learning_rate": 4.097190539148491e-05,
      "loss": 0.9673,
      "num_input_tokens_seen": 27378152,
      "step": 47195
    },
    {
      "epoch": 7.030086386654752,
      "grad_norm": 28.125,
      "learning_rate": 4.096940544562557e-05,
      "loss": 1.1034,
      "num_input_tokens_seen": 27382760,
      "step": 47200
    },
    {
      "epoch": 7.03083109919571,
      "grad_norm": 14.5,
      "learning_rate": 4.096690522997936e-05,
      "loss": 1.088,
      "num_input_tokens_seen": 27385672,
      "step": 47205
    },
    {
      "epoch": 7.03157581173667,
      "grad_norm": 8.0,
      "learning_rate": 4.096440474458852e-05,
      "loss": 1.1346,
      "num_input_tokens_seen": 27388520,
      "step": 47210
    },
    {
      "epoch": 7.032320524277629,
      "grad_norm": 20.0,
      "learning_rate": 4.096190398949529e-05,
      "loss": 1.1626,
      "num_input_tokens_seen": 27391368,
      "step": 47215
    },
    {
      "epoch": 7.033065236818588,
      "grad_norm": 12.625,
      "learning_rate": 4.09594029647419e-05,
      "loss": 1.0736,
      "num_input_tokens_seen": 27394248,
      "step": 47220
    },
    {
      "epoch": 7.033809949359547,
      "grad_norm": 15.5625,
      "learning_rate": 4.095690167037063e-05,
      "loss": 1.3695,
      "num_input_tokens_seen": 27397480,
      "step": 47225
    },
    {
      "epoch": 7.034554661900507,
      "grad_norm": 8.4375,
      "learning_rate": 4.095440010642372e-05,
      "loss": 1.1381,
      "num_input_tokens_seen": 27400360,
      "step": 47230
    },
    {
      "epoch": 7.035299374441466,
      "grad_norm": 13.5625,
      "learning_rate": 4.0951898272943436e-05,
      "loss": 0.7942,
      "num_input_tokens_seen": 27403240,
      "step": 47235
    },
    {
      "epoch": 7.036044086982425,
      "grad_norm": 11.8125,
      "learning_rate": 4.094939616997204e-05,
      "loss": 1.342,
      "num_input_tokens_seen": 27406344,
      "step": 47240
    },
    {
      "epoch": 7.036788799523384,
      "grad_norm": 12.1875,
      "learning_rate": 4.094689379755181e-05,
      "loss": 1.2356,
      "num_input_tokens_seen": 27409224,
      "step": 47245
    },
    {
      "epoch": 7.037533512064343,
      "grad_norm": 15.9375,
      "learning_rate": 4.094439115572502e-05,
      "loss": 1.4717,
      "num_input_tokens_seen": 27412008,
      "step": 47250
    },
    {
      "epoch": 7.038278224605302,
      "grad_norm": 10.5625,
      "learning_rate": 4.094188824453394e-05,
      "loss": 1.0123,
      "num_input_tokens_seen": 27414856,
      "step": 47255
    },
    {
      "epoch": 7.039022937146261,
      "grad_norm": 17.75,
      "learning_rate": 4.0939385064020866e-05,
      "loss": 1.1461,
      "num_input_tokens_seen": 27417800,
      "step": 47260
    },
    {
      "epoch": 7.039767649687221,
      "grad_norm": 12.5625,
      "learning_rate": 4.093688161422808e-05,
      "loss": 0.7262,
      "num_input_tokens_seen": 27420712,
      "step": 47265
    },
    {
      "epoch": 7.0405123622281796,
      "grad_norm": 12.25,
      "learning_rate": 4.093437789519787e-05,
      "loss": 1.1028,
      "num_input_tokens_seen": 27423688,
      "step": 47270
    },
    {
      "epoch": 7.041257074769139,
      "grad_norm": 4.875,
      "learning_rate": 4.093187390697255e-05,
      "loss": 0.7293,
      "num_input_tokens_seen": 27426952,
      "step": 47275
    },
    {
      "epoch": 7.042001787310098,
      "grad_norm": 11.875,
      "learning_rate": 4.0929369649594416e-05,
      "loss": 0.9676,
      "num_input_tokens_seen": 27429544,
      "step": 47280
    },
    {
      "epoch": 7.042746499851058,
      "grad_norm": 13.875,
      "learning_rate": 4.092686512310576e-05,
      "loss": 1.1751,
      "num_input_tokens_seen": 27432456,
      "step": 47285
    },
    {
      "epoch": 7.043491212392016,
      "grad_norm": 12.875,
      "learning_rate": 4.09243603275489e-05,
      "loss": 1.2462,
      "num_input_tokens_seen": 27435336,
      "step": 47290
    },
    {
      "epoch": 7.044235924932976,
      "grad_norm": 6.0,
      "learning_rate": 4.092185526296618e-05,
      "loss": 0.9065,
      "num_input_tokens_seen": 27438760,
      "step": 47295
    },
    {
      "epoch": 7.044980637473935,
      "grad_norm": 22.125,
      "learning_rate": 4.091934992939989e-05,
      "loss": 1.4855,
      "num_input_tokens_seen": 27441608,
      "step": 47300
    },
    {
      "epoch": 7.045725350014894,
      "grad_norm": 20.125,
      "learning_rate": 4.0916844326892344e-05,
      "loss": 1.3045,
      "num_input_tokens_seen": 27444168,
      "step": 47305
    },
    {
      "epoch": 7.046470062555853,
      "grad_norm": 8.1875,
      "learning_rate": 4.091433845548591e-05,
      "loss": 1.0858,
      "num_input_tokens_seen": 27447080,
      "step": 47310
    },
    {
      "epoch": 7.047214775096813,
      "grad_norm": 15.4375,
      "learning_rate": 4.0911832315222896e-05,
      "loss": 1.0221,
      "num_input_tokens_seen": 27450248,
      "step": 47315
    },
    {
      "epoch": 7.047959487637772,
      "grad_norm": 8.125,
      "learning_rate": 4.090932590614565e-05,
      "loss": 1.2822,
      "num_input_tokens_seen": 27453512,
      "step": 47320
    },
    {
      "epoch": 7.048704200178731,
      "grad_norm": 7.875,
      "learning_rate": 4.09068192282965e-05,
      "loss": 1.1354,
      "num_input_tokens_seen": 27456904,
      "step": 47325
    },
    {
      "epoch": 7.04944891271969,
      "grad_norm": 10.5625,
      "learning_rate": 4.090431228171782e-05,
      "loss": 1.0667,
      "num_input_tokens_seen": 27459624,
      "step": 47330
    },
    {
      "epoch": 7.05019362526065,
      "grad_norm": 8.5,
      "learning_rate": 4.0901805066451946e-05,
      "loss": 1.0062,
      "num_input_tokens_seen": 27462472,
      "step": 47335
    },
    {
      "epoch": 7.050938337801608,
      "grad_norm": 7.40625,
      "learning_rate": 4.089929758254123e-05,
      "loss": 0.9531,
      "num_input_tokens_seen": 27465224,
      "step": 47340
    },
    {
      "epoch": 7.051683050342568,
      "grad_norm": 5.15625,
      "learning_rate": 4.089678983002805e-05,
      "loss": 0.8374,
      "num_input_tokens_seen": 27468008,
      "step": 47345
    },
    {
      "epoch": 7.052427762883527,
      "grad_norm": 5.78125,
      "learning_rate": 4.089428180895476e-05,
      "loss": 1.0405,
      "num_input_tokens_seen": 27470824,
      "step": 47350
    },
    {
      "epoch": 7.053172475424486,
      "grad_norm": 8.875,
      "learning_rate": 4.089177351936373e-05,
      "loss": 1.1696,
      "num_input_tokens_seen": 27473800,
      "step": 47355
    },
    {
      "epoch": 7.053917187965445,
      "grad_norm": 15.25,
      "learning_rate": 4.0889264961297336e-05,
      "loss": 1.2682,
      "num_input_tokens_seen": 27477000,
      "step": 47360
    },
    {
      "epoch": 7.054661900506405,
      "grad_norm": 13.3125,
      "learning_rate": 4.0886756134797964e-05,
      "loss": 0.9181,
      "num_input_tokens_seen": 27479688,
      "step": 47365
    },
    {
      "epoch": 7.055406613047364,
      "grad_norm": 5.75,
      "learning_rate": 4.0884247039907984e-05,
      "loss": 0.9006,
      "num_input_tokens_seen": 27482632,
      "step": 47370
    },
    {
      "epoch": 7.056151325588323,
      "grad_norm": 8.1875,
      "learning_rate": 4.0881737676669813e-05,
      "loss": 0.9064,
      "num_input_tokens_seen": 27485512,
      "step": 47375
    },
    {
      "epoch": 7.056896038129282,
      "grad_norm": 6.40625,
      "learning_rate": 4.087922804512582e-05,
      "loss": 0.804,
      "num_input_tokens_seen": 27488360,
      "step": 47380
    },
    {
      "epoch": 7.057640750670242,
      "grad_norm": 15.4375,
      "learning_rate": 4.087671814531839e-05,
      "loss": 1.4718,
      "num_input_tokens_seen": 27491656,
      "step": 47385
    },
    {
      "epoch": 7.0583854632112,
      "grad_norm": 9.375,
      "learning_rate": 4.087420797728996e-05,
      "loss": 1.1633,
      "num_input_tokens_seen": 27494728,
      "step": 47390
    },
    {
      "epoch": 7.05913017575216,
      "grad_norm": 13.1875,
      "learning_rate": 4.087169754108292e-05,
      "loss": 1.1222,
      "num_input_tokens_seen": 27497768,
      "step": 47395
    },
    {
      "epoch": 7.059874888293119,
      "grad_norm": 17.0,
      "learning_rate": 4.0869186836739674e-05,
      "loss": 0.9069,
      "num_input_tokens_seen": 27500648,
      "step": 47400
    },
    {
      "epoch": 7.0606196008340785,
      "grad_norm": 7.8125,
      "learning_rate": 4.086667586430265e-05,
      "loss": 0.9035,
      "num_input_tokens_seen": 27503624,
      "step": 47405
    },
    {
      "epoch": 7.061364313375037,
      "grad_norm": 18.625,
      "learning_rate": 4.086416462381426e-05,
      "loss": 1.2421,
      "num_input_tokens_seen": 27506504,
      "step": 47410
    },
    {
      "epoch": 7.062109025915996,
      "grad_norm": 7.71875,
      "learning_rate": 4.086165311531694e-05,
      "loss": 1.0295,
      "num_input_tokens_seen": 27509224,
      "step": 47415
    },
    {
      "epoch": 7.062853738456956,
      "grad_norm": 9.25,
      "learning_rate": 4.085914133885311e-05,
      "loss": 0.927,
      "num_input_tokens_seen": 27512040,
      "step": 47420
    },
    {
      "epoch": 7.063598450997914,
      "grad_norm": 9.0625,
      "learning_rate": 4.08566292944652e-05,
      "loss": 1.362,
      "num_input_tokens_seen": 27515272,
      "step": 47425
    },
    {
      "epoch": 7.064343163538874,
      "grad_norm": 7.0625,
      "learning_rate": 4.085411698219566e-05,
      "loss": 1.0131,
      "num_input_tokens_seen": 27518376,
      "step": 47430
    },
    {
      "epoch": 7.065087876079833,
      "grad_norm": 8.8125,
      "learning_rate": 4.085160440208692e-05,
      "loss": 1.0035,
      "num_input_tokens_seen": 27521192,
      "step": 47435
    },
    {
      "epoch": 7.065832588620792,
      "grad_norm": 11.5625,
      "learning_rate": 4.084909155418143e-05,
      "loss": 1.0677,
      "num_input_tokens_seen": 27524104,
      "step": 47440
    },
    {
      "epoch": 7.066577301161751,
      "grad_norm": 10.25,
      "learning_rate": 4.084657843852166e-05,
      "loss": 0.9204,
      "num_input_tokens_seen": 27526984,
      "step": 47445
    },
    {
      "epoch": 7.067322013702711,
      "grad_norm": 8.125,
      "learning_rate": 4.0844065055150046e-05,
      "loss": 1.3044,
      "num_input_tokens_seen": 27529960,
      "step": 47450
    },
    {
      "epoch": 7.06806672624367,
      "grad_norm": 5.40625,
      "learning_rate": 4.0841551404109056e-05,
      "loss": 0.8342,
      "num_input_tokens_seen": 27533032,
      "step": 47455
    },
    {
      "epoch": 7.068811438784629,
      "grad_norm": 9.4375,
      "learning_rate": 4.083903748544116e-05,
      "loss": 0.7984,
      "num_input_tokens_seen": 27535848,
      "step": 47460
    },
    {
      "epoch": 7.069556151325588,
      "grad_norm": 7.78125,
      "learning_rate": 4.0836523299188826e-05,
      "loss": 0.8659,
      "num_input_tokens_seen": 27538760,
      "step": 47465
    },
    {
      "epoch": 7.070300863866548,
      "grad_norm": 7.03125,
      "learning_rate": 4.083400884539452e-05,
      "loss": 1.0048,
      "num_input_tokens_seen": 27541544,
      "step": 47470
    },
    {
      "epoch": 7.071045576407506,
      "grad_norm": 10.5625,
      "learning_rate": 4.083149412410072e-05,
      "loss": 1.3776,
      "num_input_tokens_seen": 27544616,
      "step": 47475
    },
    {
      "epoch": 7.071790288948466,
      "grad_norm": 8.5625,
      "learning_rate": 4.082897913534993e-05,
      "loss": 1.1342,
      "num_input_tokens_seen": 27547432,
      "step": 47480
    },
    {
      "epoch": 7.072535001489425,
      "grad_norm": 6.4375,
      "learning_rate": 4.0826463879184615e-05,
      "loss": 1.1392,
      "num_input_tokens_seen": 27550216,
      "step": 47485
    },
    {
      "epoch": 7.0732797140303845,
      "grad_norm": 15.0,
      "learning_rate": 4.082394835564729e-05,
      "loss": 1.2265,
      "num_input_tokens_seen": 27553544,
      "step": 47490
    },
    {
      "epoch": 7.074024426571343,
      "grad_norm": 6.59375,
      "learning_rate": 4.082143256478044e-05,
      "loss": 1.2614,
      "num_input_tokens_seen": 27556168,
      "step": 47495
    },
    {
      "epoch": 7.074769139112303,
      "grad_norm": 13.5,
      "learning_rate": 4.081891650662656e-05,
      "loss": 1.2981,
      "num_input_tokens_seen": 27559272,
      "step": 47500
    },
    {
      "epoch": 7.075513851653262,
      "grad_norm": 13.9375,
      "learning_rate": 4.0816400181228165e-05,
      "loss": 1.224,
      "num_input_tokens_seen": 27562024,
      "step": 47505
    },
    {
      "epoch": 7.076258564194221,
      "grad_norm": 13.4375,
      "learning_rate": 4.081388358862776e-05,
      "loss": 1.098,
      "num_input_tokens_seen": 27564904,
      "step": 47510
    },
    {
      "epoch": 7.07700327673518,
      "grad_norm": 4.90625,
      "learning_rate": 4.0811366728867874e-05,
      "loss": 1.2559,
      "num_input_tokens_seen": 27567688,
      "step": 47515
    },
    {
      "epoch": 7.07774798927614,
      "grad_norm": 5.1875,
      "learning_rate": 4.080884960199101e-05,
      "loss": 0.7692,
      "num_input_tokens_seen": 27570664,
      "step": 47520
    },
    {
      "epoch": 7.078492701817098,
      "grad_norm": 7.5,
      "learning_rate": 4.08063322080397e-05,
      "loss": 0.9923,
      "num_input_tokens_seen": 27573864,
      "step": 47525
    },
    {
      "epoch": 7.079237414358058,
      "grad_norm": 5.03125,
      "learning_rate": 4.080381454705647e-05,
      "loss": 0.8613,
      "num_input_tokens_seen": 27576680,
      "step": 47530
    },
    {
      "epoch": 7.079982126899017,
      "grad_norm": 5.6875,
      "learning_rate": 4.080129661908386e-05,
      "loss": 1.0104,
      "num_input_tokens_seen": 27579688,
      "step": 47535
    },
    {
      "epoch": 7.0807268394399765,
      "grad_norm": 14.0,
      "learning_rate": 4.07987784241644e-05,
      "loss": 0.9791,
      "num_input_tokens_seen": 27582472,
      "step": 47540
    },
    {
      "epoch": 7.081471551980935,
      "grad_norm": 11.25,
      "learning_rate": 4.0796259962340636e-05,
      "loss": 1.1972,
      "num_input_tokens_seen": 27585160,
      "step": 47545
    },
    {
      "epoch": 7.082216264521895,
      "grad_norm": 8.5625,
      "learning_rate": 4.079374123365512e-05,
      "loss": 0.8144,
      "num_input_tokens_seen": 27588072,
      "step": 47550
    },
    {
      "epoch": 7.082960977062854,
      "grad_norm": 12.9375,
      "learning_rate": 4.079122223815039e-05,
      "loss": 1.2909,
      "num_input_tokens_seen": 27590952,
      "step": 47555
    },
    {
      "epoch": 7.083705689603813,
      "grad_norm": 13.5625,
      "learning_rate": 4.0788702975869013e-05,
      "loss": 1.4785,
      "num_input_tokens_seen": 27594120,
      "step": 47560
    },
    {
      "epoch": 7.084450402144772,
      "grad_norm": 3.765625,
      "learning_rate": 4.0786183446853545e-05,
      "loss": 0.9201,
      "num_input_tokens_seen": 27597128,
      "step": 47565
    },
    {
      "epoch": 7.085195114685732,
      "grad_norm": 10.8125,
      "learning_rate": 4.0783663651146555e-05,
      "loss": 1.1532,
      "num_input_tokens_seen": 27600264,
      "step": 47570
    },
    {
      "epoch": 7.0859398272266905,
      "grad_norm": 12.375,
      "learning_rate": 4.078114358879061e-05,
      "loss": 1.1358,
      "num_input_tokens_seen": 27603464,
      "step": 47575
    },
    {
      "epoch": 7.08668453976765,
      "grad_norm": 7.65625,
      "learning_rate": 4.077862325982828e-05,
      "loss": 0.9243,
      "num_input_tokens_seen": 27606152,
      "step": 47580
    },
    {
      "epoch": 7.087429252308609,
      "grad_norm": 10.75,
      "learning_rate": 4.077610266430215e-05,
      "loss": 1.2036,
      "num_input_tokens_seen": 27608776,
      "step": 47585
    },
    {
      "epoch": 7.088173964849568,
      "grad_norm": 10.75,
      "learning_rate": 4.0773581802254795e-05,
      "loss": 1.2296,
      "num_input_tokens_seen": 27611496,
      "step": 47590
    },
    {
      "epoch": 7.088918677390527,
      "grad_norm": 11.1875,
      "learning_rate": 4.077106067372881e-05,
      "loss": 0.9731,
      "num_input_tokens_seen": 27614472,
      "step": 47595
    },
    {
      "epoch": 7.089663389931486,
      "grad_norm": 8.9375,
      "learning_rate": 4.0768539278766784e-05,
      "loss": 0.7936,
      "num_input_tokens_seen": 27617448,
      "step": 47600
    },
    {
      "epoch": 7.090408102472446,
      "grad_norm": 14.25,
      "learning_rate": 4.076601761741131e-05,
      "loss": 1.364,
      "num_input_tokens_seen": 27620456,
      "step": 47605
    },
    {
      "epoch": 7.091152815013404,
      "grad_norm": 13.5625,
      "learning_rate": 4.0763495689705004e-05,
      "loss": 1.0679,
      "num_input_tokens_seen": 27623368,
      "step": 47610
    },
    {
      "epoch": 7.091897527554364,
      "grad_norm": 9.4375,
      "learning_rate": 4.076097349569044e-05,
      "loss": 1.1068,
      "num_input_tokens_seen": 27626184,
      "step": 47615
    },
    {
      "epoch": 7.092642240095323,
      "grad_norm": 7.84375,
      "learning_rate": 4.075845103541026e-05,
      "loss": 1.0894,
      "num_input_tokens_seen": 27629064,
      "step": 47620
    },
    {
      "epoch": 7.0933869526362825,
      "grad_norm": 8.5,
      "learning_rate": 4.0755928308907065e-05,
      "loss": 0.9679,
      "num_input_tokens_seen": 27631752,
      "step": 47625
    },
    {
      "epoch": 7.094131665177241,
      "grad_norm": 11.3125,
      "learning_rate": 4.0753405316223476e-05,
      "loss": 1.1394,
      "num_input_tokens_seen": 27634664,
      "step": 47630
    },
    {
      "epoch": 7.094876377718201,
      "grad_norm": 7.59375,
      "learning_rate": 4.0750882057402116e-05,
      "loss": 1.0233,
      "num_input_tokens_seen": 27637640,
      "step": 47635
    },
    {
      "epoch": 7.09562109025916,
      "grad_norm": 16.125,
      "learning_rate": 4.074835853248561e-05,
      "loss": 1.2004,
      "num_input_tokens_seen": 27640488,
      "step": 47640
    },
    {
      "epoch": 7.096365802800119,
      "grad_norm": 14.0625,
      "learning_rate": 4.074583474151659e-05,
      "loss": 1.2545,
      "num_input_tokens_seen": 27643368,
      "step": 47645
    },
    {
      "epoch": 7.097110515341078,
      "grad_norm": 19.625,
      "learning_rate": 4.074331068453769e-05,
      "loss": 1.1093,
      "num_input_tokens_seen": 27646472,
      "step": 47650
    },
    {
      "epoch": 7.097855227882038,
      "grad_norm": 6.15625,
      "learning_rate": 4.0740786361591565e-05,
      "loss": 1.1009,
      "num_input_tokens_seen": 27649288,
      "step": 47655
    },
    {
      "epoch": 7.0985999404229965,
      "grad_norm": 7.375,
      "learning_rate": 4.073826177272085e-05,
      "loss": 0.7307,
      "num_input_tokens_seen": 27652360,
      "step": 47660
    },
    {
      "epoch": 7.099344652963956,
      "grad_norm": 7.375,
      "learning_rate": 4.0735736917968205e-05,
      "loss": 0.8933,
      "num_input_tokens_seen": 27655304,
      "step": 47665
    },
    {
      "epoch": 7.100089365504915,
      "grad_norm": 21.5,
      "learning_rate": 4.073321179737627e-05,
      "loss": 1.0018,
      "num_input_tokens_seen": 27658184,
      "step": 47670
    },
    {
      "epoch": 7.1008340780458745,
      "grad_norm": 12.375,
      "learning_rate": 4.073068641098772e-05,
      "loss": 1.1255,
      "num_input_tokens_seen": 27661064,
      "step": 47675
    },
    {
      "epoch": 7.101578790586833,
      "grad_norm": 6.65625,
      "learning_rate": 4.07281607588452e-05,
      "loss": 1.2495,
      "num_input_tokens_seen": 27663816,
      "step": 47680
    },
    {
      "epoch": 7.102323503127793,
      "grad_norm": 14.125,
      "learning_rate": 4.07256348409914e-05,
      "loss": 1.2693,
      "num_input_tokens_seen": 27666760,
      "step": 47685
    },
    {
      "epoch": 7.103068215668752,
      "grad_norm": 9.125,
      "learning_rate": 4.072310865746898e-05,
      "loss": 1.4699,
      "num_input_tokens_seen": 27669576,
      "step": 47690
    },
    {
      "epoch": 7.103812928209711,
      "grad_norm": 11.0,
      "learning_rate": 4.072058220832061e-05,
      "loss": 0.9864,
      "num_input_tokens_seen": 27672360,
      "step": 47695
    },
    {
      "epoch": 7.10455764075067,
      "grad_norm": 10.0,
      "learning_rate": 4.071805549358899e-05,
      "loss": 1.2576,
      "num_input_tokens_seen": 27675048,
      "step": 47700
    },
    {
      "epoch": 7.10530235329163,
      "grad_norm": 16.25,
      "learning_rate": 4.0715528513316796e-05,
      "loss": 1.0354,
      "num_input_tokens_seen": 27678024,
      "step": 47705
    },
    {
      "epoch": 7.1060470658325885,
      "grad_norm": 15.25,
      "learning_rate": 4.0713001267546724e-05,
      "loss": 1.0494,
      "num_input_tokens_seen": 27681192,
      "step": 47710
    },
    {
      "epoch": 7.106791778373548,
      "grad_norm": 12.75,
      "learning_rate": 4.0710473756321453e-05,
      "loss": 1.0003,
      "num_input_tokens_seen": 27684072,
      "step": 47715
    },
    {
      "epoch": 7.107536490914507,
      "grad_norm": 6.46875,
      "learning_rate": 4.07079459796837e-05,
      "loss": 1.254,
      "num_input_tokens_seen": 27687080,
      "step": 47720
    },
    {
      "epoch": 7.1082812034554665,
      "grad_norm": 5.875,
      "learning_rate": 4.070541793767618e-05,
      "loss": 0.9089,
      "num_input_tokens_seen": 27689800,
      "step": 47725
    },
    {
      "epoch": 7.109025915996425,
      "grad_norm": 7.25,
      "learning_rate": 4.0702889630341566e-05,
      "loss": 0.7905,
      "num_input_tokens_seen": 27692776,
      "step": 47730
    },
    {
      "epoch": 7.109770628537385,
      "grad_norm": 18.0,
      "learning_rate": 4.07003610577226e-05,
      "loss": 1.2671,
      "num_input_tokens_seen": 27695592,
      "step": 47735
    },
    {
      "epoch": 7.110515341078344,
      "grad_norm": 7.5,
      "learning_rate": 4.0697832219862e-05,
      "loss": 0.9702,
      "num_input_tokens_seen": 27698312,
      "step": 47740
    },
    {
      "epoch": 7.111260053619303,
      "grad_norm": 18.875,
      "learning_rate": 4.069530311680247e-05,
      "loss": 1.2189,
      "num_input_tokens_seen": 27701480,
      "step": 47745
    },
    {
      "epoch": 7.112004766160262,
      "grad_norm": 7.4375,
      "learning_rate": 4.0692773748586743e-05,
      "loss": 0.9596,
      "num_input_tokens_seen": 27704424,
      "step": 47750
    },
    {
      "epoch": 7.112749478701222,
      "grad_norm": 14.5625,
      "learning_rate": 4.069024411525756e-05,
      "loss": 1.2569,
      "num_input_tokens_seen": 27707464,
      "step": 47755
    },
    {
      "epoch": 7.1134941912421805,
      "grad_norm": 8.6875,
      "learning_rate": 4.0687714216857645e-05,
      "loss": 1.0256,
      "num_input_tokens_seen": 27710152,
      "step": 47760
    },
    {
      "epoch": 7.114238903783139,
      "grad_norm": 17.625,
      "learning_rate": 4.068518405342974e-05,
      "loss": 1.1421,
      "num_input_tokens_seen": 27713000,
      "step": 47765
    },
    {
      "epoch": 7.114983616324099,
      "grad_norm": 18.0,
      "learning_rate": 4.068265362501659e-05,
      "loss": 0.8925,
      "num_input_tokens_seen": 27715944,
      "step": 47770
    },
    {
      "epoch": 7.115728328865058,
      "grad_norm": 10.0,
      "learning_rate": 4.0680122931660955e-05,
      "loss": 0.7328,
      "num_input_tokens_seen": 27718824,
      "step": 47775
    },
    {
      "epoch": 7.116473041406017,
      "grad_norm": 12.5,
      "learning_rate": 4.067759197340558e-05,
      "loss": 1.241,
      "num_input_tokens_seen": 27721640,
      "step": 47780
    },
    {
      "epoch": 7.117217753946976,
      "grad_norm": 7.21875,
      "learning_rate": 4.0675060750293216e-05,
      "loss": 0.9957,
      "num_input_tokens_seen": 27724456,
      "step": 47785
    },
    {
      "epoch": 7.117962466487936,
      "grad_norm": 20.875,
      "learning_rate": 4.067252926236663e-05,
      "loss": 0.8563,
      "num_input_tokens_seen": 27727144,
      "step": 47790
    },
    {
      "epoch": 7.1187071790288945,
      "grad_norm": 8.4375,
      "learning_rate": 4.06699975096686e-05,
      "loss": 1.4194,
      "num_input_tokens_seen": 27729864,
      "step": 47795
    },
    {
      "epoch": 7.119451891569854,
      "grad_norm": 17.125,
      "learning_rate": 4.066746549224189e-05,
      "loss": 1.3643,
      "num_input_tokens_seen": 27732936,
      "step": 47800
    },
    {
      "epoch": 7.120196604110813,
      "grad_norm": 14.5625,
      "learning_rate": 4.0664933210129265e-05,
      "loss": 1.0314,
      "num_input_tokens_seen": 27735848,
      "step": 47805
    },
    {
      "epoch": 7.1209413166517725,
      "grad_norm": 5.625,
      "learning_rate": 4.066240066337351e-05,
      "loss": 1.2736,
      "num_input_tokens_seen": 27738792,
      "step": 47810
    },
    {
      "epoch": 7.121686029192731,
      "grad_norm": 13.9375,
      "learning_rate": 4.065986785201743e-05,
      "loss": 1.0594,
      "num_input_tokens_seen": 27741736,
      "step": 47815
    },
    {
      "epoch": 7.122430741733691,
      "grad_norm": 7.59375,
      "learning_rate": 4.065733477610379e-05,
      "loss": 0.9095,
      "num_input_tokens_seen": 27744648,
      "step": 47820
    },
    {
      "epoch": 7.12317545427465,
      "grad_norm": 14.5,
      "learning_rate": 4.065480143567539e-05,
      "loss": 1.1386,
      "num_input_tokens_seen": 27747560,
      "step": 47825
    },
    {
      "epoch": 7.123920166815609,
      "grad_norm": 10.625,
      "learning_rate": 4.065226783077504e-05,
      "loss": 1.1687,
      "num_input_tokens_seen": 27750408,
      "step": 47830
    },
    {
      "epoch": 7.124664879356568,
      "grad_norm": 8.3125,
      "learning_rate": 4.0649733961445525e-05,
      "loss": 1.0435,
      "num_input_tokens_seen": 27753384,
      "step": 47835
    },
    {
      "epoch": 7.125409591897528,
      "grad_norm": 7.5,
      "learning_rate": 4.064719982772965e-05,
      "loss": 1.1137,
      "num_input_tokens_seen": 27756136,
      "step": 47840
    },
    {
      "epoch": 7.1261543044384865,
      "grad_norm": 9.0625,
      "learning_rate": 4.064466542967026e-05,
      "loss": 0.8635,
      "num_input_tokens_seen": 27759080,
      "step": 47845
    },
    {
      "epoch": 7.126899016979446,
      "grad_norm": 10.3125,
      "learning_rate": 4.0642130767310136e-05,
      "loss": 0.7809,
      "num_input_tokens_seen": 27761864,
      "step": 47850
    },
    {
      "epoch": 7.127643729520405,
      "grad_norm": 7.84375,
      "learning_rate": 4.0639595840692116e-05,
      "loss": 0.9006,
      "num_input_tokens_seen": 27764776,
      "step": 47855
    },
    {
      "epoch": 7.128388442061365,
      "grad_norm": 7.21875,
      "learning_rate": 4.063706064985901e-05,
      "loss": 1.0985,
      "num_input_tokens_seen": 27767560,
      "step": 47860
    },
    {
      "epoch": 7.129133154602323,
      "grad_norm": 13.625,
      "learning_rate": 4.063452519485367e-05,
      "loss": 1.0485,
      "num_input_tokens_seen": 27770728,
      "step": 47865
    },
    {
      "epoch": 7.129877867143283,
      "grad_norm": 13.4375,
      "learning_rate": 4.06319894757189e-05,
      "loss": 0.7816,
      "num_input_tokens_seen": 27773480,
      "step": 47870
    },
    {
      "epoch": 7.130622579684242,
      "grad_norm": 19.125,
      "learning_rate": 4.062945349249757e-05,
      "loss": 1.063,
      "num_input_tokens_seen": 27776424,
      "step": 47875
    },
    {
      "epoch": 7.131367292225201,
      "grad_norm": 10.5,
      "learning_rate": 4.06269172452325e-05,
      "loss": 1.1441,
      "num_input_tokens_seen": 27779560,
      "step": 47880
    },
    {
      "epoch": 7.13211200476616,
      "grad_norm": 9.4375,
      "learning_rate": 4.0624380733966546e-05,
      "loss": 0.7265,
      "num_input_tokens_seen": 27782472,
      "step": 47885
    },
    {
      "epoch": 7.13285671730712,
      "grad_norm": 8.8125,
      "learning_rate": 4.062184395874257e-05,
      "loss": 0.8259,
      "num_input_tokens_seen": 27785064,
      "step": 47890
    },
    {
      "epoch": 7.1336014298480785,
      "grad_norm": 8.3125,
      "learning_rate": 4.0619306919603405e-05,
      "loss": 0.9887,
      "num_input_tokens_seen": 27787880,
      "step": 47895
    },
    {
      "epoch": 7.134346142389038,
      "grad_norm": 11.25,
      "learning_rate": 4.061676961659193e-05,
      "loss": 0.7613,
      "num_input_tokens_seen": 27790920,
      "step": 47900
    },
    {
      "epoch": 7.135090854929997,
      "grad_norm": 12.0,
      "learning_rate": 4.061423204975101e-05,
      "loss": 1.3676,
      "num_input_tokens_seen": 27793896,
      "step": 47905
    },
    {
      "epoch": 7.135835567470957,
      "grad_norm": 4.90625,
      "learning_rate": 4.06116942191235e-05,
      "loss": 0.8495,
      "num_input_tokens_seen": 27797064,
      "step": 47910
    },
    {
      "epoch": 7.136580280011915,
      "grad_norm": 10.75,
      "learning_rate": 4.060915612475229e-05,
      "loss": 1.0699,
      "num_input_tokens_seen": 27799752,
      "step": 47915
    },
    {
      "epoch": 7.137324992552875,
      "grad_norm": 10.5625,
      "learning_rate": 4.060661776668024e-05,
      "loss": 0.8737,
      "num_input_tokens_seen": 27802472,
      "step": 47920
    },
    {
      "epoch": 7.138069705093834,
      "grad_norm": 8.0625,
      "learning_rate": 4.060407914495026e-05,
      "loss": 0.7499,
      "num_input_tokens_seen": 27805288,
      "step": 47925
    },
    {
      "epoch": 7.1388144176347925,
      "grad_norm": 3.609375,
      "learning_rate": 4.060154025960521e-05,
      "loss": 1.429,
      "num_input_tokens_seen": 27808040,
      "step": 47930
    },
    {
      "epoch": 7.139559130175752,
      "grad_norm": 13.875,
      "learning_rate": 4.0599001110688e-05,
      "loss": 0.8165,
      "num_input_tokens_seen": 27811080,
      "step": 47935
    },
    {
      "epoch": 7.140303842716711,
      "grad_norm": 17.625,
      "learning_rate": 4.0596461698241524e-05,
      "loss": 1.2135,
      "num_input_tokens_seen": 27813640,
      "step": 47940
    },
    {
      "epoch": 7.141048555257671,
      "grad_norm": 13.625,
      "learning_rate": 4.059392202230867e-05,
      "loss": 1.1282,
      "num_input_tokens_seen": 27816392,
      "step": 47945
    },
    {
      "epoch": 7.141793267798629,
      "grad_norm": 10.25,
      "learning_rate": 4.059138208293236e-05,
      "loss": 1.2446,
      "num_input_tokens_seen": 27819272,
      "step": 47950
    },
    {
      "epoch": 7.142537980339589,
      "grad_norm": 19.0,
      "learning_rate": 4.058884188015549e-05,
      "loss": 1.4181,
      "num_input_tokens_seen": 27822184,
      "step": 47955
    },
    {
      "epoch": 7.143282692880548,
      "grad_norm": 9.0625,
      "learning_rate": 4.058630141402099e-05,
      "loss": 1.1318,
      "num_input_tokens_seen": 27824936,
      "step": 47960
    },
    {
      "epoch": 7.144027405421507,
      "grad_norm": 10.125,
      "learning_rate": 4.058376068457176e-05,
      "loss": 0.7407,
      "num_input_tokens_seen": 27827720,
      "step": 47965
    },
    {
      "epoch": 7.144772117962466,
      "grad_norm": 12.8125,
      "learning_rate": 4.058121969185073e-05,
      "loss": 1.0467,
      "num_input_tokens_seen": 27830824,
      "step": 47970
    },
    {
      "epoch": 7.145516830503426,
      "grad_norm": 5.5,
      "learning_rate": 4.057867843590083e-05,
      "loss": 1.035,
      "num_input_tokens_seen": 27833512,
      "step": 47975
    },
    {
      "epoch": 7.1462615430443845,
      "grad_norm": 10.75,
      "learning_rate": 4.0576136916765e-05,
      "loss": 1.0656,
      "num_input_tokens_seen": 27836552,
      "step": 47980
    },
    {
      "epoch": 7.147006255585344,
      "grad_norm": 13.375,
      "learning_rate": 4.0573595134486166e-05,
      "loss": 1.0548,
      "num_input_tokens_seen": 27839400,
      "step": 47985
    },
    {
      "epoch": 7.147750968126303,
      "grad_norm": 12.875,
      "learning_rate": 4.0571053089107256e-05,
      "loss": 1.1249,
      "num_input_tokens_seen": 27842472,
      "step": 47990
    },
    {
      "epoch": 7.148495680667263,
      "grad_norm": 12.75,
      "learning_rate": 4.056851078067124e-05,
      "loss": 0.9632,
      "num_input_tokens_seen": 27845288,
      "step": 47995
    },
    {
      "epoch": 7.149240393208221,
      "grad_norm": 6.6875,
      "learning_rate": 4.056596820922106e-05,
      "loss": 1.2627,
      "num_input_tokens_seen": 27848520,
      "step": 48000
    },
    {
      "epoch": 7.149985105749181,
      "grad_norm": 10.0,
      "learning_rate": 4.0563425374799665e-05,
      "loss": 1.4292,
      "num_input_tokens_seen": 27851496,
      "step": 48005
    },
    {
      "epoch": 7.15072981829014,
      "grad_norm": 15.5,
      "learning_rate": 4.0560882277450017e-05,
      "loss": 0.9005,
      "num_input_tokens_seen": 27854248,
      "step": 48010
    },
    {
      "epoch": 7.151474530831099,
      "grad_norm": 10.75,
      "learning_rate": 4.055833891721508e-05,
      "loss": 0.9271,
      "num_input_tokens_seen": 27857000,
      "step": 48015
    },
    {
      "epoch": 7.152219243372058,
      "grad_norm": 9.8125,
      "learning_rate": 4.0555795294137824e-05,
      "loss": 1.0908,
      "num_input_tokens_seen": 27859848,
      "step": 48020
    },
    {
      "epoch": 7.152963955913018,
      "grad_norm": 16.625,
      "learning_rate": 4.05532514082612e-05,
      "loss": 1.4961,
      "num_input_tokens_seen": 27862888,
      "step": 48025
    },
    {
      "epoch": 7.153708668453977,
      "grad_norm": 11.875,
      "learning_rate": 4.055070725962822e-05,
      "loss": 0.9796,
      "num_input_tokens_seen": 27865992,
      "step": 48030
    },
    {
      "epoch": 7.154453380994936,
      "grad_norm": 11.625,
      "learning_rate": 4.0548162848281835e-05,
      "loss": 1.1131,
      "num_input_tokens_seen": 27868936,
      "step": 48035
    },
    {
      "epoch": 7.155198093535895,
      "grad_norm": 9.9375,
      "learning_rate": 4.0545618174265045e-05,
      "loss": 1.0315,
      "num_input_tokens_seen": 27871816,
      "step": 48040
    },
    {
      "epoch": 7.155942806076855,
      "grad_norm": 11.375,
      "learning_rate": 4.054307323762083e-05,
      "loss": 1.1087,
      "num_input_tokens_seen": 27874504,
      "step": 48045
    },
    {
      "epoch": 7.156687518617813,
      "grad_norm": 6.625,
      "learning_rate": 4.05405280383922e-05,
      "loss": 0.9509,
      "num_input_tokens_seen": 27877672,
      "step": 48050
    },
    {
      "epoch": 7.157432231158773,
      "grad_norm": 9.75,
      "learning_rate": 4.053798257662213e-05,
      "loss": 1.2688,
      "num_input_tokens_seen": 27880392,
      "step": 48055
    },
    {
      "epoch": 7.158176943699732,
      "grad_norm": 9.25,
      "learning_rate": 4.053543685235365e-05,
      "loss": 0.6582,
      "num_input_tokens_seen": 27883016,
      "step": 48060
    },
    {
      "epoch": 7.158921656240691,
      "grad_norm": 9.0625,
      "learning_rate": 4.0532890865629744e-05,
      "loss": 1.1359,
      "num_input_tokens_seen": 27885832,
      "step": 48065
    },
    {
      "epoch": 7.15966636878165,
      "grad_norm": 6.9375,
      "learning_rate": 4.053034461649344e-05,
      "loss": 0.8894,
      "num_input_tokens_seen": 27888520,
      "step": 48070
    },
    {
      "epoch": 7.16041108132261,
      "grad_norm": 9.375,
      "learning_rate": 4.0527798104987745e-05,
      "loss": 1.1508,
      "num_input_tokens_seen": 27891400,
      "step": 48075
    },
    {
      "epoch": 7.161155793863569,
      "grad_norm": 9.6875,
      "learning_rate": 4.052525133115569e-05,
      "loss": 0.9637,
      "num_input_tokens_seen": 27894440,
      "step": 48080
    },
    {
      "epoch": 7.161900506404528,
      "grad_norm": 12.0,
      "learning_rate": 4.052270429504028e-05,
      "loss": 1.2171,
      "num_input_tokens_seen": 27897480,
      "step": 48085
    },
    {
      "epoch": 7.162645218945487,
      "grad_norm": 8.5625,
      "learning_rate": 4.0520156996684565e-05,
      "loss": 1.0263,
      "num_input_tokens_seen": 27900648,
      "step": 48090
    },
    {
      "epoch": 7.163389931486447,
      "grad_norm": 19.625,
      "learning_rate": 4.051760943613158e-05,
      "loss": 1.2083,
      "num_input_tokens_seen": 27903720,
      "step": 48095
    },
    {
      "epoch": 7.164134644027405,
      "grad_norm": 9.3125,
      "learning_rate": 4.0515061613424345e-05,
      "loss": 1.0046,
      "num_input_tokens_seen": 27906472,
      "step": 48100
    },
    {
      "epoch": 7.164879356568365,
      "grad_norm": 9.4375,
      "learning_rate": 4.051251352860591e-05,
      "loss": 0.9784,
      "num_input_tokens_seen": 27909384,
      "step": 48105
    },
    {
      "epoch": 7.165624069109324,
      "grad_norm": 11.0625,
      "learning_rate": 4.0509965181719326e-05,
      "loss": 1.293,
      "num_input_tokens_seen": 27912584,
      "step": 48110
    },
    {
      "epoch": 7.166368781650283,
      "grad_norm": 7.6875,
      "learning_rate": 4.050741657280765e-05,
      "loss": 1.104,
      "num_input_tokens_seen": 27915720,
      "step": 48115
    },
    {
      "epoch": 7.167113494191242,
      "grad_norm": 10.625,
      "learning_rate": 4.050486770191393e-05,
      "loss": 1.1105,
      "num_input_tokens_seen": 27918344,
      "step": 48120
    },
    {
      "epoch": 7.167858206732201,
      "grad_norm": 8.125,
      "learning_rate": 4.050231856908122e-05,
      "loss": 0.9633,
      "num_input_tokens_seen": 27921320,
      "step": 48125
    },
    {
      "epoch": 7.168602919273161,
      "grad_norm": 10.5625,
      "learning_rate": 4.04997691743526e-05,
      "loss": 0.9308,
      "num_input_tokens_seen": 27924520,
      "step": 48130
    },
    {
      "epoch": 7.169347631814119,
      "grad_norm": 7.625,
      "learning_rate": 4.0497219517771137e-05,
      "loss": 1.3213,
      "num_input_tokens_seen": 27927496,
      "step": 48135
    },
    {
      "epoch": 7.170092344355079,
      "grad_norm": 13.75,
      "learning_rate": 4.04946695993799e-05,
      "loss": 0.954,
      "num_input_tokens_seen": 27930504,
      "step": 48140
    },
    {
      "epoch": 7.170837056896038,
      "grad_norm": 4.46875,
      "learning_rate": 4.0492119419221966e-05,
      "loss": 0.9692,
      "num_input_tokens_seen": 27933608,
      "step": 48145
    },
    {
      "epoch": 7.171581769436997,
      "grad_norm": 13.6875,
      "learning_rate": 4.048956897734042e-05,
      "loss": 1.1963,
      "num_input_tokens_seen": 27936424,
      "step": 48150
    },
    {
      "epoch": 7.172326481977956,
      "grad_norm": 9.9375,
      "learning_rate": 4.048701827377835e-05,
      "loss": 1.0416,
      "num_input_tokens_seen": 27939208,
      "step": 48155
    },
    {
      "epoch": 7.173071194518916,
      "grad_norm": 11.625,
      "learning_rate": 4.0484467308578844e-05,
      "loss": 0.9511,
      "num_input_tokens_seen": 27942056,
      "step": 48160
    },
    {
      "epoch": 7.173815907059875,
      "grad_norm": 7.4375,
      "learning_rate": 4.0481916081785e-05,
      "loss": 1.3875,
      "num_input_tokens_seen": 27945416,
      "step": 48165
    },
    {
      "epoch": 7.174560619600834,
      "grad_norm": 13.0625,
      "learning_rate": 4.047936459343992e-05,
      "loss": 0.9554,
      "num_input_tokens_seen": 27948072,
      "step": 48170
    },
    {
      "epoch": 7.175305332141793,
      "grad_norm": 12.25,
      "learning_rate": 4.047681284358671e-05,
      "loss": 1.3577,
      "num_input_tokens_seen": 27950856,
      "step": 48175
    },
    {
      "epoch": 7.176050044682753,
      "grad_norm": 15.5,
      "learning_rate": 4.0474260832268476e-05,
      "loss": 1.0185,
      "num_input_tokens_seen": 27953736,
      "step": 48180
    },
    {
      "epoch": 7.176794757223711,
      "grad_norm": 9.1875,
      "learning_rate": 4.047170855952833e-05,
      "loss": 0.8327,
      "num_input_tokens_seen": 27956488,
      "step": 48185
    },
    {
      "epoch": 7.177539469764671,
      "grad_norm": 8.0,
      "learning_rate": 4.04691560254094e-05,
      "loss": 0.8192,
      "num_input_tokens_seen": 27959272,
      "step": 48190
    },
    {
      "epoch": 7.17828418230563,
      "grad_norm": 5.8125,
      "learning_rate": 4.046660322995479e-05,
      "loss": 0.5942,
      "num_input_tokens_seen": 27962024,
      "step": 48195
    },
    {
      "epoch": 7.1790288948465895,
      "grad_norm": 11.4375,
      "learning_rate": 4.046405017320765e-05,
      "loss": 0.9835,
      "num_input_tokens_seen": 27964840,
      "step": 48200
    },
    {
      "epoch": 7.179773607387548,
      "grad_norm": 8.875,
      "learning_rate": 4.046149685521109e-05,
      "loss": 0.6952,
      "num_input_tokens_seen": 27967688,
      "step": 48205
    },
    {
      "epoch": 7.180518319928508,
      "grad_norm": 10.25,
      "learning_rate": 4.045894327600826e-05,
      "loss": 0.9941,
      "num_input_tokens_seen": 27970632,
      "step": 48210
    },
    {
      "epoch": 7.181263032469467,
      "grad_norm": 8.8125,
      "learning_rate": 4.04563894356423e-05,
      "loss": 1.2251,
      "num_input_tokens_seen": 27973800,
      "step": 48215
    },
    {
      "epoch": 7.182007745010426,
      "grad_norm": 14.9375,
      "learning_rate": 4.045383533415634e-05,
      "loss": 0.9669,
      "num_input_tokens_seen": 27976488,
      "step": 48220
    },
    {
      "epoch": 7.182752457551385,
      "grad_norm": 9.5,
      "learning_rate": 4.045128097159354e-05,
      "loss": 1.2923,
      "num_input_tokens_seen": 27979592,
      "step": 48225
    },
    {
      "epoch": 7.183497170092345,
      "grad_norm": 12.625,
      "learning_rate": 4.044872634799706e-05,
      "loss": 1.3293,
      "num_input_tokens_seen": 27982344,
      "step": 48230
    },
    {
      "epoch": 7.184241882633303,
      "grad_norm": 11.0,
      "learning_rate": 4.044617146341003e-05,
      "loss": 1.2164,
      "num_input_tokens_seen": 27985480,
      "step": 48235
    },
    {
      "epoch": 7.184986595174263,
      "grad_norm": 8.0625,
      "learning_rate": 4.044361631787565e-05,
      "loss": 1.0046,
      "num_input_tokens_seen": 27988328,
      "step": 48240
    },
    {
      "epoch": 7.185731307715222,
      "grad_norm": 11.125,
      "learning_rate": 4.044106091143707e-05,
      "loss": 1.1379,
      "num_input_tokens_seen": 27991112,
      "step": 48245
    },
    {
      "epoch": 7.1864760202561815,
      "grad_norm": 13.5,
      "learning_rate": 4.043850524413745e-05,
      "loss": 0.7401,
      "num_input_tokens_seen": 27993672,
      "step": 48250
    },
    {
      "epoch": 7.18722073279714,
      "grad_norm": 11.9375,
      "learning_rate": 4.0435949316019974e-05,
      "loss": 1.1599,
      "num_input_tokens_seen": 27996616,
      "step": 48255
    },
    {
      "epoch": 7.1879654453381,
      "grad_norm": 11.25,
      "learning_rate": 4.0433393127127827e-05,
      "loss": 1.2833,
      "num_input_tokens_seen": 27999496,
      "step": 48260
    },
    {
      "epoch": 7.188710157879059,
      "grad_norm": 14.375,
      "learning_rate": 4.043083667750419e-05,
      "loss": 1.1858,
      "num_input_tokens_seen": 28002312,
      "step": 48265
    },
    {
      "epoch": 7.189454870420018,
      "grad_norm": 13.3125,
      "learning_rate": 4.042827996719225e-05,
      "loss": 1.0064,
      "num_input_tokens_seen": 28005352,
      "step": 48270
    },
    {
      "epoch": 7.190199582960977,
      "grad_norm": 12.6875,
      "learning_rate": 4.04257229962352e-05,
      "loss": 1.039,
      "num_input_tokens_seen": 28008296,
      "step": 48275
    },
    {
      "epoch": 7.190944295501936,
      "grad_norm": 12.5,
      "learning_rate": 4.042316576467624e-05,
      "loss": 1.4065,
      "num_input_tokens_seen": 28011400,
      "step": 48280
    },
    {
      "epoch": 7.1916890080428955,
      "grad_norm": 4.65625,
      "learning_rate": 4.0420608272558566e-05,
      "loss": 0.7986,
      "num_input_tokens_seen": 28014376,
      "step": 48285
    },
    {
      "epoch": 7.192433720583854,
      "grad_norm": 17.75,
      "learning_rate": 4.0418050519925386e-05,
      "loss": 1.0189,
      "num_input_tokens_seen": 28017480,
      "step": 48290
    },
    {
      "epoch": 7.193178433124814,
      "grad_norm": 11.8125,
      "learning_rate": 4.041549250681992e-05,
      "loss": 1.2354,
      "num_input_tokens_seen": 28020328,
      "step": 48295
    },
    {
      "epoch": 7.193923145665773,
      "grad_norm": 6.3125,
      "learning_rate": 4.041293423328537e-05,
      "loss": 1.1113,
      "num_input_tokens_seen": 28023176,
      "step": 48300
    },
    {
      "epoch": 7.194667858206732,
      "grad_norm": 7.53125,
      "learning_rate": 4.0410375699364964e-05,
      "loss": 1.4609,
      "num_input_tokens_seen": 28026312,
      "step": 48305
    },
    {
      "epoch": 7.195412570747691,
      "grad_norm": 9.5,
      "learning_rate": 4.040781690510193e-05,
      "loss": 1.156,
      "num_input_tokens_seen": 28029416,
      "step": 48310
    },
    {
      "epoch": 7.196157283288651,
      "grad_norm": 7.09375,
      "learning_rate": 4.0405257850539474e-05,
      "loss": 1.1973,
      "num_input_tokens_seen": 28032232,
      "step": 48315
    },
    {
      "epoch": 7.196901995829609,
      "grad_norm": 10.875,
      "learning_rate": 4.040269853572085e-05,
      "loss": 1.0457,
      "num_input_tokens_seen": 28035080,
      "step": 48320
    },
    {
      "epoch": 7.197646708370569,
      "grad_norm": 5.875,
      "learning_rate": 4.0400138960689295e-05,
      "loss": 1.0649,
      "num_input_tokens_seen": 28037864,
      "step": 48325
    },
    {
      "epoch": 7.198391420911528,
      "grad_norm": 10.75,
      "learning_rate": 4.0397579125488036e-05,
      "loss": 1.423,
      "num_input_tokens_seen": 28040904,
      "step": 48330
    },
    {
      "epoch": 7.1991361334524875,
      "grad_norm": 6.84375,
      "learning_rate": 4.039501903016033e-05,
      "loss": 0.6988,
      "num_input_tokens_seen": 28043560,
      "step": 48335
    },
    {
      "epoch": 7.199880845993446,
      "grad_norm": 16.0,
      "learning_rate": 4.039245867474942e-05,
      "loss": 1.0938,
      "num_input_tokens_seen": 28046344,
      "step": 48340
    },
    {
      "epoch": 7.200625558534406,
      "grad_norm": 16.0,
      "learning_rate": 4.038989805929857e-05,
      "loss": 1.0249,
      "num_input_tokens_seen": 28049352,
      "step": 48345
    },
    {
      "epoch": 7.201370271075365,
      "grad_norm": 12.0625,
      "learning_rate": 4.0387337183851035e-05,
      "loss": 1.0281,
      "num_input_tokens_seen": 28052360,
      "step": 48350
    },
    {
      "epoch": 7.202114983616324,
      "grad_norm": 8.8125,
      "learning_rate": 4.038477604845008e-05,
      "loss": 1.2271,
      "num_input_tokens_seen": 28055048,
      "step": 48355
    },
    {
      "epoch": 7.202859696157283,
      "grad_norm": 10.8125,
      "learning_rate": 4.038221465313896e-05,
      "loss": 1.2013,
      "num_input_tokens_seen": 28057960,
      "step": 48360
    },
    {
      "epoch": 7.203604408698243,
      "grad_norm": 8.125,
      "learning_rate": 4.037965299796096e-05,
      "loss": 0.7941,
      "num_input_tokens_seen": 28061032,
      "step": 48365
    },
    {
      "epoch": 7.2043491212392015,
      "grad_norm": 17.25,
      "learning_rate": 4.037709108295936e-05,
      "loss": 1.4285,
      "num_input_tokens_seen": 28064072,
      "step": 48370
    },
    {
      "epoch": 7.205093833780161,
      "grad_norm": 9.0,
      "learning_rate": 4.037452890817743e-05,
      "loss": 0.8902,
      "num_input_tokens_seen": 28067592,
      "step": 48375
    },
    {
      "epoch": 7.20583854632112,
      "grad_norm": 4.5,
      "learning_rate": 4.0371966473658465e-05,
      "loss": 0.972,
      "num_input_tokens_seen": 28070632,
      "step": 48380
    },
    {
      "epoch": 7.2065832588620795,
      "grad_norm": 12.375,
      "learning_rate": 4.0369403779445744e-05,
      "loss": 0.9919,
      "num_input_tokens_seen": 28073704,
      "step": 48385
    },
    {
      "epoch": 7.207327971403038,
      "grad_norm": 16.625,
      "learning_rate": 4.0366840825582574e-05,
      "loss": 1.1297,
      "num_input_tokens_seen": 28076520,
      "step": 48390
    },
    {
      "epoch": 7.208072683943998,
      "grad_norm": 10.375,
      "learning_rate": 4.036427761211224e-05,
      "loss": 0.9564,
      "num_input_tokens_seen": 28079688,
      "step": 48395
    },
    {
      "epoch": 7.208817396484957,
      "grad_norm": 12.1875,
      "learning_rate": 4.0361714139078055e-05,
      "loss": 1.0148,
      "num_input_tokens_seen": 28082184,
      "step": 48400
    },
    {
      "epoch": 7.209562109025916,
      "grad_norm": 6.03125,
      "learning_rate": 4.0359150406523314e-05,
      "loss": 1.2134,
      "num_input_tokens_seen": 28085128,
      "step": 48405
    },
    {
      "epoch": 7.210306821566875,
      "grad_norm": 12.6875,
      "learning_rate": 4.0356586414491345e-05,
      "loss": 0.8907,
      "num_input_tokens_seen": 28087912,
      "step": 48410
    },
    {
      "epoch": 7.211051534107835,
      "grad_norm": 9.9375,
      "learning_rate": 4.035402216302546e-05,
      "loss": 1.0303,
      "num_input_tokens_seen": 28091176,
      "step": 48415
    },
    {
      "epoch": 7.2117962466487935,
      "grad_norm": 17.5,
      "learning_rate": 4.035145765216897e-05,
      "loss": 1.3495,
      "num_input_tokens_seen": 28093928,
      "step": 48420
    },
    {
      "epoch": 7.212540959189753,
      "grad_norm": 8.75,
      "learning_rate": 4.03488928819652e-05,
      "loss": 1.1057,
      "num_input_tokens_seen": 28096744,
      "step": 48425
    },
    {
      "epoch": 7.213285671730712,
      "grad_norm": 9.6875,
      "learning_rate": 4.03463278524575e-05,
      "loss": 1.4457,
      "num_input_tokens_seen": 28099528,
      "step": 48430
    },
    {
      "epoch": 7.2140303842716715,
      "grad_norm": 7.53125,
      "learning_rate": 4.034376256368917e-05,
      "loss": 0.9167,
      "num_input_tokens_seen": 28102536,
      "step": 48435
    },
    {
      "epoch": 7.21477509681263,
      "grad_norm": 7.5,
      "learning_rate": 4.034119701570358e-05,
      "loss": 0.7759,
      "num_input_tokens_seen": 28105800,
      "step": 48440
    },
    {
      "epoch": 7.21551980935359,
      "grad_norm": 7.125,
      "learning_rate": 4.033863120854405e-05,
      "loss": 0.9258,
      "num_input_tokens_seen": 28108744,
      "step": 48445
    },
    {
      "epoch": 7.216264521894549,
      "grad_norm": 5.875,
      "learning_rate": 4.0336065142253945e-05,
      "loss": 0.6828,
      "num_input_tokens_seen": 28111848,
      "step": 48450
    },
    {
      "epoch": 7.217009234435508,
      "grad_norm": 12.5,
      "learning_rate": 4.03334988168766e-05,
      "loss": 0.9625,
      "num_input_tokens_seen": 28114984,
      "step": 48455
    },
    {
      "epoch": 7.217753946976467,
      "grad_norm": 6.15625,
      "learning_rate": 4.0330932232455376e-05,
      "loss": 0.8048,
      "num_input_tokens_seen": 28117544,
      "step": 48460
    },
    {
      "epoch": 7.218498659517426,
      "grad_norm": 9.875,
      "learning_rate": 4.0328365389033636e-05,
      "loss": 1.1313,
      "num_input_tokens_seen": 28120872,
      "step": 48465
    },
    {
      "epoch": 7.2192433720583855,
      "grad_norm": 15.9375,
      "learning_rate": 4.0325798286654734e-05,
      "loss": 1.0018,
      "num_input_tokens_seen": 28123624,
      "step": 48470
    },
    {
      "epoch": 7.219988084599344,
      "grad_norm": 12.0,
      "learning_rate": 4.032323092536206e-05,
      "loss": 1.0821,
      "num_input_tokens_seen": 28126536,
      "step": 48475
    },
    {
      "epoch": 7.220732797140304,
      "grad_norm": 8.0625,
      "learning_rate": 4.032066330519896e-05,
      "loss": 0.9047,
      "num_input_tokens_seen": 28129512,
      "step": 48480
    },
    {
      "epoch": 7.221477509681263,
      "grad_norm": 8.5625,
      "learning_rate": 4.0318095426208835e-05,
      "loss": 1.0735,
      "num_input_tokens_seen": 28132648,
      "step": 48485
    },
    {
      "epoch": 7.222222222222222,
      "grad_norm": 9.5625,
      "learning_rate": 4.031552728843505e-05,
      "loss": 1.221,
      "num_input_tokens_seen": 28135336,
      "step": 48490
    },
    {
      "epoch": 7.222966934763181,
      "grad_norm": 9.875,
      "learning_rate": 4.0312958891921e-05,
      "loss": 1.1782,
      "num_input_tokens_seen": 28138536,
      "step": 48495
    },
    {
      "epoch": 7.223711647304141,
      "grad_norm": 5.875,
      "learning_rate": 4.031039023671007e-05,
      "loss": 0.8175,
      "num_input_tokens_seen": 28141096,
      "step": 48500
    },
    {
      "epoch": 7.2244563598450995,
      "grad_norm": 3.96875,
      "learning_rate": 4.0307821322845664e-05,
      "loss": 0.9418,
      "num_input_tokens_seen": 28143976,
      "step": 48505
    },
    {
      "epoch": 7.225201072386059,
      "grad_norm": 9.5625,
      "learning_rate": 4.0305252150371175e-05,
      "loss": 0.8787,
      "num_input_tokens_seen": 28147080,
      "step": 48510
    },
    {
      "epoch": 7.225945784927018,
      "grad_norm": 4.71875,
      "learning_rate": 4.030268271933e-05,
      "loss": 0.979,
      "num_input_tokens_seen": 28149960,
      "step": 48515
    },
    {
      "epoch": 7.2266904974679775,
      "grad_norm": 15.0,
      "learning_rate": 4.030011302976555e-05,
      "loss": 1.3443,
      "num_input_tokens_seen": 28152680,
      "step": 48520
    },
    {
      "epoch": 7.227435210008936,
      "grad_norm": 11.0625,
      "learning_rate": 4.0297543081721254e-05,
      "loss": 0.9734,
      "num_input_tokens_seen": 28155624,
      "step": 48525
    },
    {
      "epoch": 7.228179922549896,
      "grad_norm": 12.6875,
      "learning_rate": 4.02949728752405e-05,
      "loss": 0.8166,
      "num_input_tokens_seen": 28158728,
      "step": 48530
    },
    {
      "epoch": 7.228924635090855,
      "grad_norm": 7.03125,
      "learning_rate": 4.0292402410366734e-05,
      "loss": 1.0606,
      "num_input_tokens_seen": 28161288,
      "step": 48535
    },
    {
      "epoch": 7.229669347631814,
      "grad_norm": 6.6875,
      "learning_rate": 4.0289831687143376e-05,
      "loss": 0.9885,
      "num_input_tokens_seen": 28164296,
      "step": 48540
    },
    {
      "epoch": 7.230414060172773,
      "grad_norm": 9.3125,
      "learning_rate": 4.028726070561385e-05,
      "loss": 1.1147,
      "num_input_tokens_seen": 28167112,
      "step": 48545
    },
    {
      "epoch": 7.231158772713733,
      "grad_norm": 14.9375,
      "learning_rate": 4.028468946582158e-05,
      "loss": 1.1265,
      "num_input_tokens_seen": 28170216,
      "step": 48550
    },
    {
      "epoch": 7.2319034852546915,
      "grad_norm": 9.75,
      "learning_rate": 4.028211796781003e-05,
      "loss": 1.0151,
      "num_input_tokens_seen": 28173448,
      "step": 48555
    },
    {
      "epoch": 7.232648197795651,
      "grad_norm": 6.96875,
      "learning_rate": 4.027954621162262e-05,
      "loss": 0.8901,
      "num_input_tokens_seen": 28176264,
      "step": 48560
    },
    {
      "epoch": 7.23339291033661,
      "grad_norm": 10.5625,
      "learning_rate": 4.027697419730281e-05,
      "loss": 1.1837,
      "num_input_tokens_seen": 28179112,
      "step": 48565
    },
    {
      "epoch": 7.23413762287757,
      "grad_norm": 13.25,
      "learning_rate": 4.027440192489404e-05,
      "loss": 1.276,
      "num_input_tokens_seen": 28182408,
      "step": 48570
    },
    {
      "epoch": 7.234882335418528,
      "grad_norm": 11.625,
      "learning_rate": 4.0271829394439786e-05,
      "loss": 1.2252,
      "num_input_tokens_seen": 28185288,
      "step": 48575
    },
    {
      "epoch": 7.235627047959488,
      "grad_norm": 11.125,
      "learning_rate": 4.026925660598349e-05,
      "loss": 1.1328,
      "num_input_tokens_seen": 28188072,
      "step": 48580
    },
    {
      "epoch": 7.236371760500447,
      "grad_norm": 11.875,
      "learning_rate": 4.0266683559568625e-05,
      "loss": 1.2077,
      "num_input_tokens_seen": 28190984,
      "step": 48585
    },
    {
      "epoch": 7.237116473041406,
      "grad_norm": 16.375,
      "learning_rate": 4.0264110255238654e-05,
      "loss": 0.9446,
      "num_input_tokens_seen": 28194216,
      "step": 48590
    },
    {
      "epoch": 7.237861185582365,
      "grad_norm": 9.75,
      "learning_rate": 4.026153669303706e-05,
      "loss": 1.1056,
      "num_input_tokens_seen": 28197064,
      "step": 48595
    },
    {
      "epoch": 7.238605898123325,
      "grad_norm": 15.3125,
      "learning_rate": 4.0258962873007305e-05,
      "loss": 0.9953,
      "num_input_tokens_seen": 28200040,
      "step": 48600
    },
    {
      "epoch": 7.2393506106642835,
      "grad_norm": 8.6875,
      "learning_rate": 4.025638879519289e-05,
      "loss": 0.9729,
      "num_input_tokens_seen": 28202984,
      "step": 48605
    },
    {
      "epoch": 7.240095323205243,
      "grad_norm": 11.9375,
      "learning_rate": 4.025381445963728e-05,
      "loss": 1.0366,
      "num_input_tokens_seen": 28205800,
      "step": 48610
    },
    {
      "epoch": 7.240840035746202,
      "grad_norm": 12.8125,
      "learning_rate": 4.025123986638399e-05,
      "loss": 1.0174,
      "num_input_tokens_seen": 28208648,
      "step": 48615
    },
    {
      "epoch": 7.241584748287162,
      "grad_norm": 14.375,
      "learning_rate": 4.02486650154765e-05,
      "loss": 1.0054,
      "num_input_tokens_seen": 28211240,
      "step": 48620
    },
    {
      "epoch": 7.24232946082812,
      "grad_norm": 10.8125,
      "learning_rate": 4.0246089906958317e-05,
      "loss": 1.1504,
      "num_input_tokens_seen": 28214056,
      "step": 48625
    },
    {
      "epoch": 7.243074173369079,
      "grad_norm": 15.1875,
      "learning_rate": 4.024351454087293e-05,
      "loss": 0.7612,
      "num_input_tokens_seen": 28217064,
      "step": 48630
    },
    {
      "epoch": 7.243818885910039,
      "grad_norm": 14.5,
      "learning_rate": 4.0240938917263864e-05,
      "loss": 0.9327,
      "num_input_tokens_seen": 28219944,
      "step": 48635
    },
    {
      "epoch": 7.2445635984509975,
      "grad_norm": 13.0,
      "learning_rate": 4.0238363036174625e-05,
      "loss": 1.2144,
      "num_input_tokens_seen": 28222760,
      "step": 48640
    },
    {
      "epoch": 7.245308310991957,
      "grad_norm": 7.5625,
      "learning_rate": 4.023578689764873e-05,
      "loss": 1.2081,
      "num_input_tokens_seen": 28225608,
      "step": 48645
    },
    {
      "epoch": 7.246053023532916,
      "grad_norm": 9.0,
      "learning_rate": 4.02332105017297e-05,
      "loss": 0.8558,
      "num_input_tokens_seen": 28228296,
      "step": 48650
    },
    {
      "epoch": 7.246797736073876,
      "grad_norm": 11.375,
      "learning_rate": 4.0230633848461056e-05,
      "loss": 1.3013,
      "num_input_tokens_seen": 28231752,
      "step": 48655
    },
    {
      "epoch": 7.247542448614834,
      "grad_norm": 13.625,
      "learning_rate": 4.022805693788634e-05,
      "loss": 1.1739,
      "num_input_tokens_seen": 28234536,
      "step": 48660
    },
    {
      "epoch": 7.248287161155794,
      "grad_norm": 8.9375,
      "learning_rate": 4.0225479770049076e-05,
      "loss": 0.7288,
      "num_input_tokens_seen": 28237576,
      "step": 48665
    },
    {
      "epoch": 7.249031873696753,
      "grad_norm": 6.125,
      "learning_rate": 4.022290234499281e-05,
      "loss": 0.9004,
      "num_input_tokens_seen": 28240392,
      "step": 48670
    },
    {
      "epoch": 7.249776586237712,
      "grad_norm": 9.125,
      "learning_rate": 4.0220324662761076e-05,
      "loss": 1.0097,
      "num_input_tokens_seen": 28243368,
      "step": 48675
    },
    {
      "epoch": 7.250521298778671,
      "grad_norm": 21.5,
      "learning_rate": 4.021774672339743e-05,
      "loss": 1.0135,
      "num_input_tokens_seen": 28246184,
      "step": 48680
    },
    {
      "epoch": 7.251266011319631,
      "grad_norm": 8.875,
      "learning_rate": 4.021516852694541e-05,
      "loss": 1.3271,
      "num_input_tokens_seen": 28249192,
      "step": 48685
    },
    {
      "epoch": 7.2520107238605895,
      "grad_norm": 8.9375,
      "learning_rate": 4.021259007344859e-05,
      "loss": 1.1183,
      "num_input_tokens_seen": 28252200,
      "step": 48690
    },
    {
      "epoch": 7.252755436401549,
      "grad_norm": 5.875,
      "learning_rate": 4.021001136295052e-05,
      "loss": 0.8502,
      "num_input_tokens_seen": 28254888,
      "step": 48695
    },
    {
      "epoch": 7.253500148942508,
      "grad_norm": 9.5625,
      "learning_rate": 4.020743239549477e-05,
      "loss": 1.2018,
      "num_input_tokens_seen": 28257960,
      "step": 48700
    },
    {
      "epoch": 7.254244861483468,
      "grad_norm": 9.4375,
      "learning_rate": 4.0204853171124904e-05,
      "loss": 0.8812,
      "num_input_tokens_seen": 28260648,
      "step": 48705
    },
    {
      "epoch": 7.254989574024426,
      "grad_norm": 9.0625,
      "learning_rate": 4.0202273689884496e-05,
      "loss": 0.7983,
      "num_input_tokens_seen": 28263688,
      "step": 48710
    },
    {
      "epoch": 7.255734286565386,
      "grad_norm": 8.875,
      "learning_rate": 4.019969395181713e-05,
      "loss": 0.9539,
      "num_input_tokens_seen": 28266792,
      "step": 48715
    },
    {
      "epoch": 7.256478999106345,
      "grad_norm": 12.125,
      "learning_rate": 4.0197113956966376e-05,
      "loss": 0.9059,
      "num_input_tokens_seen": 28269832,
      "step": 48720
    },
    {
      "epoch": 7.257223711647304,
      "grad_norm": 11.375,
      "learning_rate": 4.019453370537583e-05,
      "loss": 1.1458,
      "num_input_tokens_seen": 28272296,
      "step": 48725
    },
    {
      "epoch": 7.257968424188263,
      "grad_norm": 9.375,
      "learning_rate": 4.019195319708908e-05,
      "loss": 1.0087,
      "num_input_tokens_seen": 28275144,
      "step": 48730
    },
    {
      "epoch": 7.258713136729223,
      "grad_norm": 9.875,
      "learning_rate": 4.018937243214972e-05,
      "loss": 1.3457,
      "num_input_tokens_seen": 28277768,
      "step": 48735
    },
    {
      "epoch": 7.259457849270182,
      "grad_norm": 16.625,
      "learning_rate": 4.018679141060136e-05,
      "loss": 1.309,
      "num_input_tokens_seen": 28280392,
      "step": 48740
    },
    {
      "epoch": 7.260202561811141,
      "grad_norm": 14.3125,
      "learning_rate": 4.0184210132487576e-05,
      "loss": 1.4756,
      "num_input_tokens_seen": 28283240,
      "step": 48745
    },
    {
      "epoch": 7.2609472743521,
      "grad_norm": 7.71875,
      "learning_rate": 4.018162859785201e-05,
      "loss": 1.2451,
      "num_input_tokens_seen": 28286280,
      "step": 48750
    },
    {
      "epoch": 7.26169198689306,
      "grad_norm": 15.5,
      "learning_rate": 4.017904680673825e-05,
      "loss": 1.138,
      "num_input_tokens_seen": 28289064,
      "step": 48755
    },
    {
      "epoch": 7.262436699434018,
      "grad_norm": 12.25,
      "learning_rate": 4.0176464759189924e-05,
      "loss": 1.0634,
      "num_input_tokens_seen": 28291976,
      "step": 48760
    },
    {
      "epoch": 7.263181411974978,
      "grad_norm": 10.75,
      "learning_rate": 4.017388245525065e-05,
      "loss": 1.1292,
      "num_input_tokens_seen": 28294696,
      "step": 48765
    },
    {
      "epoch": 7.263926124515937,
      "grad_norm": 7.4375,
      "learning_rate": 4.017129989496405e-05,
      "loss": 1.2764,
      "num_input_tokens_seen": 28297608,
      "step": 48770
    },
    {
      "epoch": 7.264670837056896,
      "grad_norm": 13.625,
      "learning_rate": 4.0168717078373763e-05,
      "loss": 0.9517,
      "num_input_tokens_seen": 28300360,
      "step": 48775
    },
    {
      "epoch": 7.265415549597855,
      "grad_norm": 16.5,
      "learning_rate": 4.016613400552342e-05,
      "loss": 1.2093,
      "num_input_tokens_seen": 28303144,
      "step": 48780
    },
    {
      "epoch": 7.266160262138815,
      "grad_norm": 6.8125,
      "learning_rate": 4.016355067645666e-05,
      "loss": 0.8625,
      "num_input_tokens_seen": 28306024,
      "step": 48785
    },
    {
      "epoch": 7.266904974679774,
      "grad_norm": 12.375,
      "learning_rate": 4.0160967091217114e-05,
      "loss": 1.298,
      "num_input_tokens_seen": 28308712,
      "step": 48790
    },
    {
      "epoch": 7.267649687220732,
      "grad_norm": 12.6875,
      "learning_rate": 4.015838324984844e-05,
      "loss": 1.1206,
      "num_input_tokens_seen": 28311528,
      "step": 48795
    },
    {
      "epoch": 7.268394399761692,
      "grad_norm": 18.125,
      "learning_rate": 4.015579915239429e-05,
      "loss": 1.1369,
      "num_input_tokens_seen": 28314664,
      "step": 48800
    },
    {
      "epoch": 7.269139112302652,
      "grad_norm": 14.8125,
      "learning_rate": 4.015321479889832e-05,
      "loss": 1.4747,
      "num_input_tokens_seen": 28317544,
      "step": 48805
    },
    {
      "epoch": 7.26988382484361,
      "grad_norm": 10.9375,
      "learning_rate": 4.015063018940418e-05,
      "loss": 0.7684,
      "num_input_tokens_seen": 28320424,
      "step": 48810
    },
    {
      "epoch": 7.270628537384569,
      "grad_norm": 8.5,
      "learning_rate": 4.014804532395554e-05,
      "loss": 0.883,
      "num_input_tokens_seen": 28323368,
      "step": 48815
    },
    {
      "epoch": 7.271373249925529,
      "grad_norm": 17.75,
      "learning_rate": 4.014546020259607e-05,
      "loss": 1.0268,
      "num_input_tokens_seen": 28326440,
      "step": 48820
    },
    {
      "epoch": 7.272117962466488,
      "grad_norm": 9.625,
      "learning_rate": 4.014287482536945e-05,
      "loss": 0.744,
      "num_input_tokens_seen": 28329256,
      "step": 48825
    },
    {
      "epoch": 7.272862675007447,
      "grad_norm": 6.03125,
      "learning_rate": 4.0140289192319355e-05,
      "loss": 0.9271,
      "num_input_tokens_seen": 28331944,
      "step": 48830
    },
    {
      "epoch": 7.273607387548406,
      "grad_norm": 7.59375,
      "learning_rate": 4.013770330348945e-05,
      "loss": 1.1918,
      "num_input_tokens_seen": 28334696,
      "step": 48835
    },
    {
      "epoch": 7.274352100089366,
      "grad_norm": 6.6875,
      "learning_rate": 4.013511715892344e-05,
      "loss": 0.9141,
      "num_input_tokens_seen": 28337480,
      "step": 48840
    },
    {
      "epoch": 7.275096812630324,
      "grad_norm": 19.0,
      "learning_rate": 4.0132530758665006e-05,
      "loss": 1.2986,
      "num_input_tokens_seen": 28340232,
      "step": 48845
    },
    {
      "epoch": 7.275841525171284,
      "grad_norm": 10.3125,
      "learning_rate": 4.0129944102757847e-05,
      "loss": 1.3916,
      "num_input_tokens_seen": 28342952,
      "step": 48850
    },
    {
      "epoch": 7.276586237712243,
      "grad_norm": 5.875,
      "learning_rate": 4.0127357191245654e-05,
      "loss": 1.1263,
      "num_input_tokens_seen": 28346024,
      "step": 48855
    },
    {
      "epoch": 7.277330950253202,
      "grad_norm": 10.4375,
      "learning_rate": 4.0124770024172135e-05,
      "loss": 1.1769,
      "num_input_tokens_seen": 28348936,
      "step": 48860
    },
    {
      "epoch": 7.278075662794161,
      "grad_norm": 6.75,
      "learning_rate": 4.0122182601581005e-05,
      "loss": 1.0347,
      "num_input_tokens_seen": 28351944,
      "step": 48865
    },
    {
      "epoch": 7.278820375335121,
      "grad_norm": 12.8125,
      "learning_rate": 4.011959492351597e-05,
      "loss": 1.1021,
      "num_input_tokens_seen": 28354600,
      "step": 48870
    },
    {
      "epoch": 7.27956508787608,
      "grad_norm": 14.5625,
      "learning_rate": 4.011700699002075e-05,
      "loss": 1.2543,
      "num_input_tokens_seen": 28357288,
      "step": 48875
    },
    {
      "epoch": 7.280309800417039,
      "grad_norm": 6.53125,
      "learning_rate": 4.011441880113905e-05,
      "loss": 1.1167,
      "num_input_tokens_seen": 28360264,
      "step": 48880
    },
    {
      "epoch": 7.281054512957998,
      "grad_norm": 15.5,
      "learning_rate": 4.0111830356914605e-05,
      "loss": 1.2006,
      "num_input_tokens_seen": 28363368,
      "step": 48885
    },
    {
      "epoch": 7.281799225498958,
      "grad_norm": 8.3125,
      "learning_rate": 4.010924165739115e-05,
      "loss": 0.9962,
      "num_input_tokens_seen": 28366312,
      "step": 48890
    },
    {
      "epoch": 7.282543938039916,
      "grad_norm": 10.4375,
      "learning_rate": 4.0106652702612416e-05,
      "loss": 1.1109,
      "num_input_tokens_seen": 28369288,
      "step": 48895
    },
    {
      "epoch": 7.283288650580876,
      "grad_norm": 5.90625,
      "learning_rate": 4.010406349262214e-05,
      "loss": 0.8114,
      "num_input_tokens_seen": 28372072,
      "step": 48900
    },
    {
      "epoch": 7.284033363121835,
      "grad_norm": 7.96875,
      "learning_rate": 4.010147402746405e-05,
      "loss": 1.1844,
      "num_input_tokens_seen": 28374952,
      "step": 48905
    },
    {
      "epoch": 7.2847780756627944,
      "grad_norm": 14.875,
      "learning_rate": 4.009888430718192e-05,
      "loss": 1.2626,
      "num_input_tokens_seen": 28377768,
      "step": 48910
    },
    {
      "epoch": 7.285522788203753,
      "grad_norm": 9.25,
      "learning_rate": 4.009629433181947e-05,
      "loss": 1.1138,
      "num_input_tokens_seen": 28380744,
      "step": 48915
    },
    {
      "epoch": 7.286267500744713,
      "grad_norm": 10.1875,
      "learning_rate": 4.009370410142049e-05,
      "loss": 0.8624,
      "num_input_tokens_seen": 28383720,
      "step": 48920
    },
    {
      "epoch": 7.287012213285672,
      "grad_norm": 11.0,
      "learning_rate": 4.00911136160287e-05,
      "loss": 1.4555,
      "num_input_tokens_seen": 28386536,
      "step": 48925
    },
    {
      "epoch": 7.287756925826631,
      "grad_norm": 18.375,
      "learning_rate": 4.00885228756879e-05,
      "loss": 1.0714,
      "num_input_tokens_seen": 28389640,
      "step": 48930
    },
    {
      "epoch": 7.28850163836759,
      "grad_norm": 7.71875,
      "learning_rate": 4.008593188044183e-05,
      "loss": 1.144,
      "num_input_tokens_seen": 28392520,
      "step": 48935
    },
    {
      "epoch": 7.28924635090855,
      "grad_norm": 12.9375,
      "learning_rate": 4.008334063033428e-05,
      "loss": 0.711,
      "num_input_tokens_seen": 28395368,
      "step": 48940
    },
    {
      "epoch": 7.289991063449508,
      "grad_norm": 10.75,
      "learning_rate": 4.008074912540901e-05,
      "loss": 1.0246,
      "num_input_tokens_seen": 28398280,
      "step": 48945
    },
    {
      "epoch": 7.290735775990468,
      "grad_norm": 5.5,
      "learning_rate": 4.0078157365709823e-05,
      "loss": 0.816,
      "num_input_tokens_seen": 28401288,
      "step": 48950
    },
    {
      "epoch": 7.291480488531427,
      "grad_norm": 9.25,
      "learning_rate": 4.0075565351280485e-05,
      "loss": 0.8316,
      "num_input_tokens_seen": 28403816,
      "step": 48955
    },
    {
      "epoch": 7.292225201072386,
      "grad_norm": 15.1875,
      "learning_rate": 4.00729730821648e-05,
      "loss": 1.1973,
      "num_input_tokens_seen": 28406760,
      "step": 48960
    },
    {
      "epoch": 7.292969913613345,
      "grad_norm": 11.25,
      "learning_rate": 4.007038055840654e-05,
      "loss": 1.1025,
      "num_input_tokens_seen": 28410024,
      "step": 48965
    },
    {
      "epoch": 7.293714626154305,
      "grad_norm": 8.5625,
      "learning_rate": 4.0067787780049535e-05,
      "loss": 1.1903,
      "num_input_tokens_seen": 28412744,
      "step": 48970
    },
    {
      "epoch": 7.294459338695264,
      "grad_norm": 12.5625,
      "learning_rate": 4.0065194747137555e-05,
      "loss": 1.1156,
      "num_input_tokens_seen": 28415624,
      "step": 48975
    },
    {
      "epoch": 7.295204051236222,
      "grad_norm": 12.4375,
      "learning_rate": 4.006260145971443e-05,
      "loss": 1.0201,
      "num_input_tokens_seen": 28418248,
      "step": 48980
    },
    {
      "epoch": 7.295948763777182,
      "grad_norm": 6.15625,
      "learning_rate": 4.006000791782396e-05,
      "loss": 0.937,
      "num_input_tokens_seen": 28420872,
      "step": 48985
    },
    {
      "epoch": 7.296693476318141,
      "grad_norm": 14.5,
      "learning_rate": 4.0057414121509965e-05,
      "loss": 1.139,
      "num_input_tokens_seen": 28423752,
      "step": 48990
    },
    {
      "epoch": 7.2974381888591004,
      "grad_norm": 9.5625,
      "learning_rate": 4.005482007081626e-05,
      "loss": 1.3503,
      "num_input_tokens_seen": 28426600,
      "step": 48995
    },
    {
      "epoch": 7.298182901400059,
      "grad_norm": 6.78125,
      "learning_rate": 4.005222576578667e-05,
      "loss": 1.2237,
      "num_input_tokens_seen": 28429640,
      "step": 49000
    },
    {
      "epoch": 7.298927613941019,
      "grad_norm": 7.5625,
      "learning_rate": 4.004963120646502e-05,
      "loss": 1.1934,
      "num_input_tokens_seen": 28432520,
      "step": 49005
    },
    {
      "epoch": 7.299672326481978,
      "grad_norm": 13.9375,
      "learning_rate": 4.004703639289515e-05,
      "loss": 1.3389,
      "num_input_tokens_seen": 28435624,
      "step": 49010
    },
    {
      "epoch": 7.300417039022937,
      "grad_norm": 12.1875,
      "learning_rate": 4.004444132512089e-05,
      "loss": 1.3116,
      "num_input_tokens_seen": 28438792,
      "step": 49015
    },
    {
      "epoch": 7.301161751563896,
      "grad_norm": 17.5,
      "learning_rate": 4.004184600318609e-05,
      "loss": 0.9582,
      "num_input_tokens_seen": 28441704,
      "step": 49020
    },
    {
      "epoch": 7.301906464104856,
      "grad_norm": 7.0625,
      "learning_rate": 4.003925042713459e-05,
      "loss": 1.0002,
      "num_input_tokens_seen": 28444424,
      "step": 49025
    },
    {
      "epoch": 7.302651176645814,
      "grad_norm": 11.375,
      "learning_rate": 4.003665459701024e-05,
      "loss": 1.4225,
      "num_input_tokens_seen": 28447016,
      "step": 49030
    },
    {
      "epoch": 7.303395889186774,
      "grad_norm": 7.09375,
      "learning_rate": 4.003405851285689e-05,
      "loss": 1.2649,
      "num_input_tokens_seen": 28449928,
      "step": 49035
    },
    {
      "epoch": 7.304140601727733,
      "grad_norm": 5.8125,
      "learning_rate": 4.00314621747184e-05,
      "loss": 0.9672,
      "num_input_tokens_seen": 28453288,
      "step": 49040
    },
    {
      "epoch": 7.3048853142686925,
      "grad_norm": 9.875,
      "learning_rate": 4.002886558263863e-05,
      "loss": 1.2682,
      "num_input_tokens_seen": 28456296,
      "step": 49045
    },
    {
      "epoch": 7.305630026809651,
      "grad_norm": 19.875,
      "learning_rate": 4.0026268736661457e-05,
      "loss": 1.4313,
      "num_input_tokens_seen": 28459560,
      "step": 49050
    },
    {
      "epoch": 7.306374739350611,
      "grad_norm": 13.5,
      "learning_rate": 4.002367163683075e-05,
      "loss": 1.1312,
      "num_input_tokens_seen": 28462376,
      "step": 49055
    },
    {
      "epoch": 7.30711945189157,
      "grad_norm": 3.984375,
      "learning_rate": 4.002107428319037e-05,
      "loss": 0.7767,
      "num_input_tokens_seen": 28465064,
      "step": 49060
    },
    {
      "epoch": 7.307864164432529,
      "grad_norm": 7.84375,
      "learning_rate": 4.0018476675784214e-05,
      "loss": 1.1287,
      "num_input_tokens_seen": 28468072,
      "step": 49065
    },
    {
      "epoch": 7.308608876973488,
      "grad_norm": 12.125,
      "learning_rate": 4.001587881465616e-05,
      "loss": 1.2616,
      "num_input_tokens_seen": 28471016,
      "step": 49070
    },
    {
      "epoch": 7.309353589514448,
      "grad_norm": 4.3125,
      "learning_rate": 4.001328069985009e-05,
      "loss": 1.2444,
      "num_input_tokens_seen": 28473768,
      "step": 49075
    },
    {
      "epoch": 7.3100983020554064,
      "grad_norm": 17.375,
      "learning_rate": 4.00106823314099e-05,
      "loss": 1.1596,
      "num_input_tokens_seen": 28476680,
      "step": 49080
    },
    {
      "epoch": 7.310843014596366,
      "grad_norm": 10.625,
      "learning_rate": 4.0008083709379496e-05,
      "loss": 1.1105,
      "num_input_tokens_seen": 28479400,
      "step": 49085
    },
    {
      "epoch": 7.311587727137325,
      "grad_norm": 4.53125,
      "learning_rate": 4.0005484833802765e-05,
      "loss": 1.1189,
      "num_input_tokens_seen": 28482248,
      "step": 49090
    },
    {
      "epoch": 7.3123324396782845,
      "grad_norm": 15.125,
      "learning_rate": 4.0002885704723614e-05,
      "loss": 1.1667,
      "num_input_tokens_seen": 28485064,
      "step": 49095
    },
    {
      "epoch": 7.313077152219243,
      "grad_norm": 6.15625,
      "learning_rate": 4.000028632218596e-05,
      "loss": 1.2648,
      "num_input_tokens_seen": 28487784,
      "step": 49100
    },
    {
      "epoch": 7.313821864760203,
      "grad_norm": 14.0625,
      "learning_rate": 3.9997686686233724e-05,
      "loss": 1.0979,
      "num_input_tokens_seen": 28490568,
      "step": 49105
    },
    {
      "epoch": 7.314566577301162,
      "grad_norm": 9.875,
      "learning_rate": 3.999508679691081e-05,
      "loss": 1.1436,
      "num_input_tokens_seen": 28493320,
      "step": 49110
    },
    {
      "epoch": 7.315311289842121,
      "grad_norm": 14.9375,
      "learning_rate": 3.999248665426114e-05,
      "loss": 1.0235,
      "num_input_tokens_seen": 28496392,
      "step": 49115
    },
    {
      "epoch": 7.31605600238308,
      "grad_norm": 7.875,
      "learning_rate": 3.998988625832865e-05,
      "loss": 0.8285,
      "num_input_tokens_seen": 28499336,
      "step": 49120
    },
    {
      "epoch": 7.31680071492404,
      "grad_norm": 8.0,
      "learning_rate": 3.998728560915726e-05,
      "loss": 1.1025,
      "num_input_tokens_seen": 28502408,
      "step": 49125
    },
    {
      "epoch": 7.3175454274649985,
      "grad_norm": 10.0625,
      "learning_rate": 3.9984684706790915e-05,
      "loss": 0.9579,
      "num_input_tokens_seen": 28505544,
      "step": 49130
    },
    {
      "epoch": 7.318290140005958,
      "grad_norm": 11.4375,
      "learning_rate": 3.998208355127355e-05,
      "loss": 1.0008,
      "num_input_tokens_seen": 28508392,
      "step": 49135
    },
    {
      "epoch": 7.319034852546917,
      "grad_norm": 8.625,
      "learning_rate": 3.997948214264911e-05,
      "loss": 0.9342,
      "num_input_tokens_seen": 28511464,
      "step": 49140
    },
    {
      "epoch": 7.319779565087876,
      "grad_norm": 14.125,
      "learning_rate": 3.9976880480961556e-05,
      "loss": 1.1747,
      "num_input_tokens_seen": 28514472,
      "step": 49145
    },
    {
      "epoch": 7.320524277628835,
      "grad_norm": 10.125,
      "learning_rate": 3.997427856625482e-05,
      "loss": 1.148,
      "num_input_tokens_seen": 28517576,
      "step": 49150
    },
    {
      "epoch": 7.321268990169794,
      "grad_norm": 7.21875,
      "learning_rate": 3.997167639857287e-05,
      "loss": 1.5197,
      "num_input_tokens_seen": 28520328,
      "step": 49155
    },
    {
      "epoch": 7.322013702710754,
      "grad_norm": 10.5,
      "learning_rate": 3.996907397795966e-05,
      "loss": 1.1618,
      "num_input_tokens_seen": 28523240,
      "step": 49160
    },
    {
      "epoch": 7.3227584152517124,
      "grad_norm": 9.0625,
      "learning_rate": 3.9966471304459154e-05,
      "loss": 0.9956,
      "num_input_tokens_seen": 28525992,
      "step": 49165
    },
    {
      "epoch": 7.323503127792672,
      "grad_norm": 14.3125,
      "learning_rate": 3.996386837811533e-05,
      "loss": 1.2181,
      "num_input_tokens_seen": 28528712,
      "step": 49170
    },
    {
      "epoch": 7.324247840333631,
      "grad_norm": 14.25,
      "learning_rate": 3.996126519897216e-05,
      "loss": 1.3171,
      "num_input_tokens_seen": 28531432,
      "step": 49175
    },
    {
      "epoch": 7.3249925528745905,
      "grad_norm": 11.125,
      "learning_rate": 3.995866176707363e-05,
      "loss": 1.1051,
      "num_input_tokens_seen": 28534184,
      "step": 49180
    },
    {
      "epoch": 7.325737265415549,
      "grad_norm": 14.0625,
      "learning_rate": 3.99560580824637e-05,
      "loss": 1.2168,
      "num_input_tokens_seen": 28537064,
      "step": 49185
    },
    {
      "epoch": 7.326481977956509,
      "grad_norm": 6.6875,
      "learning_rate": 3.995345414518638e-05,
      "loss": 0.9772,
      "num_input_tokens_seen": 28540008,
      "step": 49190
    },
    {
      "epoch": 7.327226690497468,
      "grad_norm": 14.0,
      "learning_rate": 3.995084995528563e-05,
      "loss": 1.4116,
      "num_input_tokens_seen": 28543368,
      "step": 49195
    },
    {
      "epoch": 7.327971403038427,
      "grad_norm": 14.4375,
      "learning_rate": 3.9948245512805484e-05,
      "loss": 1.2764,
      "num_input_tokens_seen": 28546344,
      "step": 49200
    },
    {
      "epoch": 7.328716115579386,
      "grad_norm": 12.25,
      "learning_rate": 3.994564081778992e-05,
      "loss": 1.1963,
      "num_input_tokens_seen": 28549160,
      "step": 49205
    },
    {
      "epoch": 7.329460828120346,
      "grad_norm": 12.4375,
      "learning_rate": 3.994303587028294e-05,
      "loss": 1.0349,
      "num_input_tokens_seen": 28552136,
      "step": 49210
    },
    {
      "epoch": 7.3302055406613045,
      "grad_norm": 7.125,
      "learning_rate": 3.9940430670328556e-05,
      "loss": 0.9584,
      "num_input_tokens_seen": 28555048,
      "step": 49215
    },
    {
      "epoch": 7.330950253202264,
      "grad_norm": 10.5,
      "learning_rate": 3.993782521797078e-05,
      "loss": 0.8606,
      "num_input_tokens_seen": 28557992,
      "step": 49220
    },
    {
      "epoch": 7.331694965743223,
      "grad_norm": 9.25,
      "learning_rate": 3.993521951325363e-05,
      "loss": 1.049,
      "num_input_tokens_seen": 28561128,
      "step": 49225
    },
    {
      "epoch": 7.3324396782841825,
      "grad_norm": 9.3125,
      "learning_rate": 3.993261355622113e-05,
      "loss": 1.1661,
      "num_input_tokens_seen": 28563816,
      "step": 49230
    },
    {
      "epoch": 7.333184390825141,
      "grad_norm": 12.4375,
      "learning_rate": 3.99300073469173e-05,
      "loss": 1.0441,
      "num_input_tokens_seen": 28566664,
      "step": 49235
    },
    {
      "epoch": 7.333929103366101,
      "grad_norm": 8.375,
      "learning_rate": 3.9927400885386165e-05,
      "loss": 1.1521,
      "num_input_tokens_seen": 28569448,
      "step": 49240
    },
    {
      "epoch": 7.33467381590706,
      "grad_norm": 9.75,
      "learning_rate": 3.992479417167177e-05,
      "loss": 0.8886,
      "num_input_tokens_seen": 28572360,
      "step": 49245
    },
    {
      "epoch": 7.335418528448019,
      "grad_norm": 3.609375,
      "learning_rate": 3.992218720581814e-05,
      "loss": 1.1634,
      "num_input_tokens_seen": 28575176,
      "step": 49250
    },
    {
      "epoch": 7.336163240988978,
      "grad_norm": 12.75,
      "learning_rate": 3.9919579987869324e-05,
      "loss": 1.3482,
      "num_input_tokens_seen": 28578056,
      "step": 49255
    },
    {
      "epoch": 7.336907953529938,
      "grad_norm": 12.5,
      "learning_rate": 3.991697251786938e-05,
      "loss": 1.2181,
      "num_input_tokens_seen": 28580808,
      "step": 49260
    },
    {
      "epoch": 7.3376526660708965,
      "grad_norm": 10.9375,
      "learning_rate": 3.991436479586233e-05,
      "loss": 0.9793,
      "num_input_tokens_seen": 28583464,
      "step": 49265
    },
    {
      "epoch": 7.338397378611856,
      "grad_norm": 7.3125,
      "learning_rate": 3.9911756821892256e-05,
      "loss": 1.1395,
      "num_input_tokens_seen": 28586312,
      "step": 49270
    },
    {
      "epoch": 7.339142091152815,
      "grad_norm": 10.1875,
      "learning_rate": 3.99091485960032e-05,
      "loss": 1.0606,
      "num_input_tokens_seen": 28588936,
      "step": 49275
    },
    {
      "epoch": 7.3398868036937746,
      "grad_norm": 11.3125,
      "learning_rate": 3.990654011823923e-05,
      "loss": 0.9519,
      "num_input_tokens_seen": 28591880,
      "step": 49280
    },
    {
      "epoch": 7.340631516234733,
      "grad_norm": 11.8125,
      "learning_rate": 3.990393138864442e-05,
      "loss": 1.1142,
      "num_input_tokens_seen": 28594856,
      "step": 49285
    },
    {
      "epoch": 7.341376228775693,
      "grad_norm": 8.3125,
      "learning_rate": 3.990132240726284e-05,
      "loss": 0.9265,
      "num_input_tokens_seen": 28597544,
      "step": 49290
    },
    {
      "epoch": 7.342120941316652,
      "grad_norm": 5.53125,
      "learning_rate": 3.989871317413855e-05,
      "loss": 0.8362,
      "num_input_tokens_seen": 28600296,
      "step": 49295
    },
    {
      "epoch": 7.342865653857611,
      "grad_norm": 10.6875,
      "learning_rate": 3.989610368931566e-05,
      "loss": 0.8557,
      "num_input_tokens_seen": 28603208,
      "step": 49300
    },
    {
      "epoch": 7.34361036639857,
      "grad_norm": 8.375,
      "learning_rate": 3.9893493952838226e-05,
      "loss": 1.1469,
      "num_input_tokens_seen": 28606056,
      "step": 49305
    },
    {
      "epoch": 7.344355078939529,
      "grad_norm": 11.9375,
      "learning_rate": 3.9890883964750355e-05,
      "loss": 0.9853,
      "num_input_tokens_seen": 28608872,
      "step": 49310
    },
    {
      "epoch": 7.3450997914804885,
      "grad_norm": 10.5625,
      "learning_rate": 3.9888273725096126e-05,
      "loss": 1.1449,
      "num_input_tokens_seen": 28611688,
      "step": 49315
    },
    {
      "epoch": 7.345844504021448,
      "grad_norm": 12.25,
      "learning_rate": 3.988566323391965e-05,
      "loss": 1.2545,
      "num_input_tokens_seen": 28614920,
      "step": 49320
    },
    {
      "epoch": 7.346589216562407,
      "grad_norm": 5.625,
      "learning_rate": 3.988305249126502e-05,
      "loss": 1.0387,
      "num_input_tokens_seen": 28617768,
      "step": 49325
    },
    {
      "epoch": 7.347333929103366,
      "grad_norm": 7.4375,
      "learning_rate": 3.988044149717635e-05,
      "loss": 1.0223,
      "num_input_tokens_seen": 28620712,
      "step": 49330
    },
    {
      "epoch": 7.348078641644325,
      "grad_norm": 11.5625,
      "learning_rate": 3.987783025169773e-05,
      "loss": 1.0897,
      "num_input_tokens_seen": 28623368,
      "step": 49335
    },
    {
      "epoch": 7.348823354185284,
      "grad_norm": 4.875,
      "learning_rate": 3.987521875487331e-05,
      "loss": 0.7407,
      "num_input_tokens_seen": 28626376,
      "step": 49340
    },
    {
      "epoch": 7.349568066726244,
      "grad_norm": 7.375,
      "learning_rate": 3.9872607006747174e-05,
      "loss": 1.2557,
      "num_input_tokens_seen": 28629448,
      "step": 49345
    },
    {
      "epoch": 7.3503127792672025,
      "grad_norm": 9.3125,
      "learning_rate": 3.986999500736346e-05,
      "loss": 1.0057,
      "num_input_tokens_seen": 28632360,
      "step": 49350
    },
    {
      "epoch": 7.351057491808162,
      "grad_norm": 9.0625,
      "learning_rate": 3.98673827567663e-05,
      "loss": 0.867,
      "num_input_tokens_seen": 28635400,
      "step": 49355
    },
    {
      "epoch": 7.351802204349121,
      "grad_norm": 5.40625,
      "learning_rate": 3.9864770254999814e-05,
      "loss": 1.0502,
      "num_input_tokens_seen": 28638312,
      "step": 49360
    },
    {
      "epoch": 7.3525469168900806,
      "grad_norm": 12.1875,
      "learning_rate": 3.986215750210814e-05,
      "loss": 1.1035,
      "num_input_tokens_seen": 28641480,
      "step": 49365
    },
    {
      "epoch": 7.353291629431039,
      "grad_norm": 4.1875,
      "learning_rate": 3.985954449813543e-05,
      "loss": 1.0382,
      "num_input_tokens_seen": 28644424,
      "step": 49370
    },
    {
      "epoch": 7.354036341971999,
      "grad_norm": 11.1875,
      "learning_rate": 3.9856931243125804e-05,
      "loss": 1.396,
      "num_input_tokens_seen": 28647272,
      "step": 49375
    },
    {
      "epoch": 7.354781054512958,
      "grad_norm": 13.0,
      "learning_rate": 3.985431773712344e-05,
      "loss": 0.9969,
      "num_input_tokens_seen": 28650120,
      "step": 49380
    },
    {
      "epoch": 7.355525767053917,
      "grad_norm": 14.0625,
      "learning_rate": 3.9851703980172464e-05,
      "loss": 1.0054,
      "num_input_tokens_seen": 28652808,
      "step": 49385
    },
    {
      "epoch": 7.356270479594876,
      "grad_norm": 6.34375,
      "learning_rate": 3.984908997231704e-05,
      "loss": 1.0347,
      "num_input_tokens_seen": 28656040,
      "step": 49390
    },
    {
      "epoch": 7.357015192135836,
      "grad_norm": 11.8125,
      "learning_rate": 3.984647571360135e-05,
      "loss": 0.7116,
      "num_input_tokens_seen": 28658792,
      "step": 49395
    },
    {
      "epoch": 7.3577599046767945,
      "grad_norm": 10.0,
      "learning_rate": 3.9843861204069536e-05,
      "loss": 1.088,
      "num_input_tokens_seen": 28661640,
      "step": 49400
    },
    {
      "epoch": 7.358504617217754,
      "grad_norm": 11.0,
      "learning_rate": 3.9841246443765765e-05,
      "loss": 1.1991,
      "num_input_tokens_seen": 28664488,
      "step": 49405
    },
    {
      "epoch": 7.359249329758713,
      "grad_norm": 12.0,
      "learning_rate": 3.983863143273422e-05,
      "loss": 1.3489,
      "num_input_tokens_seen": 28667368,
      "step": 49410
    },
    {
      "epoch": 7.359994042299673,
      "grad_norm": 13.4375,
      "learning_rate": 3.983601617101909e-05,
      "loss": 1.1406,
      "num_input_tokens_seen": 28670600,
      "step": 49415
    },
    {
      "epoch": 7.360738754840631,
      "grad_norm": 6.0,
      "learning_rate": 3.983340065866453e-05,
      "loss": 0.6737,
      "num_input_tokens_seen": 28673288,
      "step": 49420
    },
    {
      "epoch": 7.361483467381591,
      "grad_norm": 9.5625,
      "learning_rate": 3.9830784895714744e-05,
      "loss": 0.9663,
      "num_input_tokens_seen": 28676072,
      "step": 49425
    },
    {
      "epoch": 7.36222817992255,
      "grad_norm": 15.4375,
      "learning_rate": 3.982816888221394e-05,
      "loss": 1.0097,
      "num_input_tokens_seen": 28678920,
      "step": 49430
    },
    {
      "epoch": 7.362972892463509,
      "grad_norm": 12.25,
      "learning_rate": 3.9825552618206274e-05,
      "loss": 0.9182,
      "num_input_tokens_seen": 28681480,
      "step": 49435
    },
    {
      "epoch": 7.363717605004468,
      "grad_norm": 10.9375,
      "learning_rate": 3.982293610373597e-05,
      "loss": 1.3789,
      "num_input_tokens_seen": 28684360,
      "step": 49440
    },
    {
      "epoch": 7.364462317545428,
      "grad_norm": 9.1875,
      "learning_rate": 3.9820319338847224e-05,
      "loss": 0.6783,
      "num_input_tokens_seen": 28687048,
      "step": 49445
    },
    {
      "epoch": 7.3652070300863866,
      "grad_norm": 8.4375,
      "learning_rate": 3.981770232358425e-05,
      "loss": 1.1355,
      "num_input_tokens_seen": 28689704,
      "step": 49450
    },
    {
      "epoch": 7.365951742627346,
      "grad_norm": 10.75,
      "learning_rate": 3.9815085057991254e-05,
      "loss": 1.1972,
      "num_input_tokens_seen": 28692680,
      "step": 49455
    },
    {
      "epoch": 7.366696455168305,
      "grad_norm": 12.0,
      "learning_rate": 3.981246754211244e-05,
      "loss": 0.868,
      "num_input_tokens_seen": 28695304,
      "step": 49460
    },
    {
      "epoch": 7.367441167709265,
      "grad_norm": 7.03125,
      "learning_rate": 3.980984977599206e-05,
      "loss": 0.9552,
      "num_input_tokens_seen": 28698088,
      "step": 49465
    },
    {
      "epoch": 7.368185880250223,
      "grad_norm": 9.625,
      "learning_rate": 3.980723175967431e-05,
      "loss": 1.4248,
      "num_input_tokens_seen": 28701032,
      "step": 49470
    },
    {
      "epoch": 7.368930592791183,
      "grad_norm": 7.125,
      "learning_rate": 3.980461349320344e-05,
      "loss": 1.1993,
      "num_input_tokens_seen": 28704136,
      "step": 49475
    },
    {
      "epoch": 7.369675305332142,
      "grad_norm": 17.875,
      "learning_rate": 3.9801994976623655e-05,
      "loss": 0.9679,
      "num_input_tokens_seen": 28707240,
      "step": 49480
    },
    {
      "epoch": 7.370420017873101,
      "grad_norm": 9.875,
      "learning_rate": 3.979937620997922e-05,
      "loss": 1.0365,
      "num_input_tokens_seen": 28710376,
      "step": 49485
    },
    {
      "epoch": 7.37116473041406,
      "grad_norm": 7.71875,
      "learning_rate": 3.979675719331437e-05,
      "loss": 1.2694,
      "num_input_tokens_seen": 28713128,
      "step": 49490
    },
    {
      "epoch": 7.371909442955019,
      "grad_norm": 11.625,
      "learning_rate": 3.9794137926673337e-05,
      "loss": 0.9262,
      "num_input_tokens_seen": 28716232,
      "step": 49495
    },
    {
      "epoch": 7.372654155495979,
      "grad_norm": 17.25,
      "learning_rate": 3.979151841010038e-05,
      "loss": 1.2779,
      "num_input_tokens_seen": 28719208,
      "step": 49500
    },
    {
      "epoch": 7.373398868036937,
      "grad_norm": 9.9375,
      "learning_rate": 3.978889864363975e-05,
      "loss": 1.0334,
      "num_input_tokens_seen": 28721960,
      "step": 49505
    },
    {
      "epoch": 7.374143580577897,
      "grad_norm": 4.34375,
      "learning_rate": 3.978627862733572e-05,
      "loss": 0.8001,
      "num_input_tokens_seen": 28724712,
      "step": 49510
    },
    {
      "epoch": 7.374888293118856,
      "grad_norm": 10.0,
      "learning_rate": 3.978365836123254e-05,
      "loss": 1.1501,
      "num_input_tokens_seen": 28727528,
      "step": 49515
    },
    {
      "epoch": 7.375633005659815,
      "grad_norm": 8.75,
      "learning_rate": 3.978103784537447e-05,
      "loss": 1.1463,
      "num_input_tokens_seen": 28730728,
      "step": 49520
    },
    {
      "epoch": 7.376377718200774,
      "grad_norm": 8.8125,
      "learning_rate": 3.977841707980578e-05,
      "loss": 0.8574,
      "num_input_tokens_seen": 28733768,
      "step": 49525
    },
    {
      "epoch": 7.377122430741734,
      "grad_norm": 16.875,
      "learning_rate": 3.977579606457077e-05,
      "loss": 1.1731,
      "num_input_tokens_seen": 28736520,
      "step": 49530
    },
    {
      "epoch": 7.3778671432826926,
      "grad_norm": 9.125,
      "learning_rate": 3.97731747997137e-05,
      "loss": 1.129,
      "num_input_tokens_seen": 28739240,
      "step": 49535
    },
    {
      "epoch": 7.378611855823652,
      "grad_norm": 5.65625,
      "learning_rate": 3.9770553285278846e-05,
      "loss": 1.1647,
      "num_input_tokens_seen": 28741992,
      "step": 49540
    },
    {
      "epoch": 7.379356568364611,
      "grad_norm": 14.5625,
      "learning_rate": 3.9767931521310514e-05,
      "loss": 1.3095,
      "num_input_tokens_seen": 28744744,
      "step": 49545
    },
    {
      "epoch": 7.380101280905571,
      "grad_norm": 18.0,
      "learning_rate": 3.976530950785299e-05,
      "loss": 1.184,
      "num_input_tokens_seen": 28747432,
      "step": 49550
    },
    {
      "epoch": 7.380845993446529,
      "grad_norm": 13.5,
      "learning_rate": 3.976268724495057e-05,
      "loss": 0.9152,
      "num_input_tokens_seen": 28750184,
      "step": 49555
    },
    {
      "epoch": 7.381590705987489,
      "grad_norm": 10.5625,
      "learning_rate": 3.9760064732647545e-05,
      "loss": 1.6576,
      "num_input_tokens_seen": 28753032,
      "step": 49560
    },
    {
      "epoch": 7.382335418528448,
      "grad_norm": 10.375,
      "learning_rate": 3.975744197098823e-05,
      "loss": 1.3547,
      "num_input_tokens_seen": 28755624,
      "step": 49565
    },
    {
      "epoch": 7.383080131069407,
      "grad_norm": 16.125,
      "learning_rate": 3.9754818960016934e-05,
      "loss": 0.9424,
      "num_input_tokens_seen": 28758472,
      "step": 49570
    },
    {
      "epoch": 7.383824843610366,
      "grad_norm": 13.125,
      "learning_rate": 3.975219569977797e-05,
      "loss": 1.649,
      "num_input_tokens_seen": 28761704,
      "step": 49575
    },
    {
      "epoch": 7.384569556151326,
      "grad_norm": 12.25,
      "learning_rate": 3.974957219031565e-05,
      "loss": 1.0721,
      "num_input_tokens_seen": 28764616,
      "step": 49580
    },
    {
      "epoch": 7.385314268692285,
      "grad_norm": 9.0,
      "learning_rate": 3.9746948431674304e-05,
      "loss": 1.4812,
      "num_input_tokens_seen": 28767976,
      "step": 49585
    },
    {
      "epoch": 7.386058981233244,
      "grad_norm": 17.25,
      "learning_rate": 3.974432442389824e-05,
      "loss": 1.1795,
      "num_input_tokens_seen": 28771016,
      "step": 49590
    },
    {
      "epoch": 7.386803693774203,
      "grad_norm": 9.1875,
      "learning_rate": 3.974170016703181e-05,
      "loss": 0.7771,
      "num_input_tokens_seen": 28774376,
      "step": 49595
    },
    {
      "epoch": 7.387548406315163,
      "grad_norm": 6.28125,
      "learning_rate": 3.973907566111934e-05,
      "loss": 1.1976,
      "num_input_tokens_seen": 28777160,
      "step": 49600
    },
    {
      "epoch": 7.388293118856121,
      "grad_norm": 10.125,
      "learning_rate": 3.9736450906205156e-05,
      "loss": 0.9506,
      "num_input_tokens_seen": 28779784,
      "step": 49605
    },
    {
      "epoch": 7.389037831397081,
      "grad_norm": 6.28125,
      "learning_rate": 3.973382590233362e-05,
      "loss": 0.8931,
      "num_input_tokens_seen": 28782600,
      "step": 49610
    },
    {
      "epoch": 7.38978254393804,
      "grad_norm": 11.5625,
      "learning_rate": 3.973120064954907e-05,
      "loss": 1.0869,
      "num_input_tokens_seen": 28785416,
      "step": 49615
    },
    {
      "epoch": 7.390527256478999,
      "grad_norm": 11.25,
      "learning_rate": 3.972857514789586e-05,
      "loss": 1.1807,
      "num_input_tokens_seen": 28788264,
      "step": 49620
    },
    {
      "epoch": 7.391271969019958,
      "grad_norm": 9.125,
      "learning_rate": 3.972594939741834e-05,
      "loss": 0.8746,
      "num_input_tokens_seen": 28791208,
      "step": 49625
    },
    {
      "epoch": 7.392016681560918,
      "grad_norm": 7.375,
      "learning_rate": 3.9723323398160863e-05,
      "loss": 1.2037,
      "num_input_tokens_seen": 28793992,
      "step": 49630
    },
    {
      "epoch": 7.392761394101877,
      "grad_norm": 16.75,
      "learning_rate": 3.972069715016782e-05,
      "loss": 1.111,
      "num_input_tokens_seen": 28796840,
      "step": 49635
    },
    {
      "epoch": 7.393506106642836,
      "grad_norm": 15.0,
      "learning_rate": 3.971807065348354e-05,
      "loss": 1.3845,
      "num_input_tokens_seen": 28799496,
      "step": 49640
    },
    {
      "epoch": 7.394250819183795,
      "grad_norm": 7.65625,
      "learning_rate": 3.9715443908152426e-05,
      "loss": 0.9697,
      "num_input_tokens_seen": 28802696,
      "step": 49645
    },
    {
      "epoch": 7.394995531724755,
      "grad_norm": 11.25,
      "learning_rate": 3.971281691421884e-05,
      "loss": 1.0172,
      "num_input_tokens_seen": 28805736,
      "step": 49650
    },
    {
      "epoch": 7.395740244265713,
      "grad_norm": 10.3125,
      "learning_rate": 3.971018967172717e-05,
      "loss": 1.2224,
      "num_input_tokens_seen": 28808488,
      "step": 49655
    },
    {
      "epoch": 7.396484956806672,
      "grad_norm": 10.625,
      "learning_rate": 3.970756218072179e-05,
      "loss": 1.2107,
      "num_input_tokens_seen": 28811400,
      "step": 49660
    },
    {
      "epoch": 7.397229669347632,
      "grad_norm": 15.375,
      "learning_rate": 3.97049344412471e-05,
      "loss": 0.9314,
      "num_input_tokens_seen": 28814376,
      "step": 49665
    },
    {
      "epoch": 7.3979743818885915,
      "grad_norm": 16.25,
      "learning_rate": 3.970230645334748e-05,
      "loss": 0.9802,
      "num_input_tokens_seen": 28817480,
      "step": 49670
    },
    {
      "epoch": 7.39871909442955,
      "grad_norm": 7.75,
      "learning_rate": 3.9699678217067346e-05,
      "loss": 0.8475,
      "num_input_tokens_seen": 28820136,
      "step": 49675
    },
    {
      "epoch": 7.399463806970509,
      "grad_norm": 16.5,
      "learning_rate": 3.9697049732451084e-05,
      "loss": 1.3795,
      "num_input_tokens_seen": 28822984,
      "step": 49680
    },
    {
      "epoch": 7.400208519511469,
      "grad_norm": 10.4375,
      "learning_rate": 3.9694420999543105e-05,
      "loss": 1.3965,
      "num_input_tokens_seen": 28825512,
      "step": 49685
    },
    {
      "epoch": 7.400953232052427,
      "grad_norm": 8.25,
      "learning_rate": 3.969179201838782e-05,
      "loss": 1.01,
      "num_input_tokens_seen": 28828584,
      "step": 49690
    },
    {
      "epoch": 7.401697944593387,
      "grad_norm": 4.90625,
      "learning_rate": 3.968916278902963e-05,
      "loss": 0.8403,
      "num_input_tokens_seen": 28831464,
      "step": 49695
    },
    {
      "epoch": 7.402442657134346,
      "grad_norm": 5.0,
      "learning_rate": 3.968653331151297e-05,
      "loss": 1.3273,
      "num_input_tokens_seen": 28834248,
      "step": 49700
    },
    {
      "epoch": 7.403187369675305,
      "grad_norm": 7.53125,
      "learning_rate": 3.9683903585882264e-05,
      "loss": 0.9458,
      "num_input_tokens_seen": 28837384,
      "step": 49705
    },
    {
      "epoch": 7.403932082216264,
      "grad_norm": 8.1875,
      "learning_rate": 3.9681273612181924e-05,
      "loss": 1.5415,
      "num_input_tokens_seen": 28840488,
      "step": 49710
    },
    {
      "epoch": 7.404676794757224,
      "grad_norm": 7.8125,
      "learning_rate": 3.967864339045639e-05,
      "loss": 1.2414,
      "num_input_tokens_seen": 28843368,
      "step": 49715
    },
    {
      "epoch": 7.405421507298183,
      "grad_norm": 12.3125,
      "learning_rate": 3.967601292075009e-05,
      "loss": 1.2376,
      "num_input_tokens_seen": 28846280,
      "step": 49720
    },
    {
      "epoch": 7.406166219839142,
      "grad_norm": 13.0,
      "learning_rate": 3.967338220310748e-05,
      "loss": 1.0859,
      "num_input_tokens_seen": 28849384,
      "step": 49725
    },
    {
      "epoch": 7.406910932380101,
      "grad_norm": 16.875,
      "learning_rate": 3.967075123757298e-05,
      "loss": 1.239,
      "num_input_tokens_seen": 28852360,
      "step": 49730
    },
    {
      "epoch": 7.407655644921061,
      "grad_norm": 8.8125,
      "learning_rate": 3.9668120024191046e-05,
      "loss": 0.9764,
      "num_input_tokens_seen": 28855336,
      "step": 49735
    },
    {
      "epoch": 7.408400357462019,
      "grad_norm": 17.25,
      "learning_rate": 3.966548856300614e-05,
      "loss": 0.9341,
      "num_input_tokens_seen": 28858088,
      "step": 49740
    },
    {
      "epoch": 7.409145070002979,
      "grad_norm": 8.5625,
      "learning_rate": 3.9662856854062706e-05,
      "loss": 0.8972,
      "num_input_tokens_seen": 28861160,
      "step": 49745
    },
    {
      "epoch": 7.409889782543938,
      "grad_norm": 16.625,
      "learning_rate": 3.9660224897405206e-05,
      "loss": 1.1815,
      "num_input_tokens_seen": 28864072,
      "step": 49750
    },
    {
      "epoch": 7.4106344950848975,
      "grad_norm": 7.5625,
      "learning_rate": 3.965759269307812e-05,
      "loss": 1.0245,
      "num_input_tokens_seen": 28866984,
      "step": 49755
    },
    {
      "epoch": 7.411379207625856,
      "grad_norm": 6.53125,
      "learning_rate": 3.965496024112589e-05,
      "loss": 1.1245,
      "num_input_tokens_seen": 28869768,
      "step": 49760
    },
    {
      "epoch": 7.412123920166816,
      "grad_norm": 8.9375,
      "learning_rate": 3.9652327541593e-05,
      "loss": 0.9028,
      "num_input_tokens_seen": 28872584,
      "step": 49765
    },
    {
      "epoch": 7.412868632707775,
      "grad_norm": 10.6875,
      "learning_rate": 3.964969459452393e-05,
      "loss": 0.9333,
      "num_input_tokens_seen": 28875848,
      "step": 49770
    },
    {
      "epoch": 7.413613345248734,
      "grad_norm": 7.25,
      "learning_rate": 3.964706139996316e-05,
      "loss": 1.046,
      "num_input_tokens_seen": 28878728,
      "step": 49775
    },
    {
      "epoch": 7.414358057789693,
      "grad_norm": 14.3125,
      "learning_rate": 3.9644427957955174e-05,
      "loss": 1.1627,
      "num_input_tokens_seen": 28881544,
      "step": 49780
    },
    {
      "epoch": 7.415102770330653,
      "grad_norm": 5.59375,
      "learning_rate": 3.9641794268544465e-05,
      "loss": 1.0129,
      "num_input_tokens_seen": 28884456,
      "step": 49785
    },
    {
      "epoch": 7.415847482871611,
      "grad_norm": 11.6875,
      "learning_rate": 3.963916033177552e-05,
      "loss": 0.9988,
      "num_input_tokens_seen": 28887304,
      "step": 49790
    },
    {
      "epoch": 7.416592195412571,
      "grad_norm": 14.875,
      "learning_rate": 3.963652614769284e-05,
      "loss": 1.2033,
      "num_input_tokens_seen": 28890024,
      "step": 49795
    },
    {
      "epoch": 7.41733690795353,
      "grad_norm": 16.125,
      "learning_rate": 3.963389171634093e-05,
      "loss": 1.2322,
      "num_input_tokens_seen": 28893000,
      "step": 49800
    },
    {
      "epoch": 7.4180816204944895,
      "grad_norm": 16.375,
      "learning_rate": 3.963125703776429e-05,
      "loss": 0.8806,
      "num_input_tokens_seen": 28895656,
      "step": 49805
    },
    {
      "epoch": 7.418826333035448,
      "grad_norm": 15.8125,
      "learning_rate": 3.962862211200744e-05,
      "loss": 1.2581,
      "num_input_tokens_seen": 28898792,
      "step": 49810
    },
    {
      "epoch": 7.419571045576408,
      "grad_norm": 12.1875,
      "learning_rate": 3.962598693911488e-05,
      "loss": 1.1434,
      "num_input_tokens_seen": 28901576,
      "step": 49815
    },
    {
      "epoch": 7.420315758117367,
      "grad_norm": 11.125,
      "learning_rate": 3.962335151913113e-05,
      "loss": 0.7172,
      "num_input_tokens_seen": 28904520,
      "step": 49820
    },
    {
      "epoch": 7.421060470658326,
      "grad_norm": 14.0,
      "learning_rate": 3.962071585210072e-05,
      "loss": 1.0445,
      "num_input_tokens_seen": 28907496,
      "step": 49825
    },
    {
      "epoch": 7.421805183199285,
      "grad_norm": 5.09375,
      "learning_rate": 3.961807993806819e-05,
      "loss": 0.9305,
      "num_input_tokens_seen": 28910184,
      "step": 49830
    },
    {
      "epoch": 7.422549895740245,
      "grad_norm": 10.3125,
      "learning_rate": 3.9615443777078046e-05,
      "loss": 1.0846,
      "num_input_tokens_seen": 28913192,
      "step": 49835
    },
    {
      "epoch": 7.4232946082812035,
      "grad_norm": 11.3125,
      "learning_rate": 3.961280736917483e-05,
      "loss": 1.2472,
      "num_input_tokens_seen": 28916040,
      "step": 49840
    },
    {
      "epoch": 7.424039320822162,
      "grad_norm": 8.6875,
      "learning_rate": 3.961017071440309e-05,
      "loss": 1.147,
      "num_input_tokens_seen": 28918760,
      "step": 49845
    },
    {
      "epoch": 7.424784033363122,
      "grad_norm": 20.5,
      "learning_rate": 3.960753381280737e-05,
      "loss": 1.2782,
      "num_input_tokens_seen": 28921736,
      "step": 49850
    },
    {
      "epoch": 7.425528745904081,
      "grad_norm": 10.4375,
      "learning_rate": 3.96048966644322e-05,
      "loss": 1.1842,
      "num_input_tokens_seen": 28924616,
      "step": 49855
    },
    {
      "epoch": 7.42627345844504,
      "grad_norm": 7.625,
      "learning_rate": 3.9602259269322155e-05,
      "loss": 1.381,
      "num_input_tokens_seen": 28927528,
      "step": 49860
    },
    {
      "epoch": 7.427018170985999,
      "grad_norm": 11.0,
      "learning_rate": 3.9599621627521774e-05,
      "loss": 0.9862,
      "num_input_tokens_seen": 28930312,
      "step": 49865
    },
    {
      "epoch": 7.427762883526959,
      "grad_norm": 8.625,
      "learning_rate": 3.959698373907563e-05,
      "loss": 1.4685,
      "num_input_tokens_seen": 28933320,
      "step": 49870
    },
    {
      "epoch": 7.428507596067917,
      "grad_norm": 9.125,
      "learning_rate": 3.959434560402828e-05,
      "loss": 1.0414,
      "num_input_tokens_seen": 28936136,
      "step": 49875
    },
    {
      "epoch": 7.429252308608877,
      "grad_norm": 10.25,
      "learning_rate": 3.9591707222424294e-05,
      "loss": 1.2803,
      "num_input_tokens_seen": 28939080,
      "step": 49880
    },
    {
      "epoch": 7.429997021149836,
      "grad_norm": 13.8125,
      "learning_rate": 3.958906859430825e-05,
      "loss": 1.2424,
      "num_input_tokens_seen": 28941960,
      "step": 49885
    },
    {
      "epoch": 7.4307417336907955,
      "grad_norm": 16.375,
      "learning_rate": 3.958642971972471e-05,
      "loss": 0.9708,
      "num_input_tokens_seen": 28945032,
      "step": 49890
    },
    {
      "epoch": 7.431486446231754,
      "grad_norm": 12.9375,
      "learning_rate": 3.958379059871827e-05,
      "loss": 1.0129,
      "num_input_tokens_seen": 28947688,
      "step": 49895
    },
    {
      "epoch": 7.432231158772714,
      "grad_norm": 5.59375,
      "learning_rate": 3.9581151231333506e-05,
      "loss": 1.0423,
      "num_input_tokens_seen": 28950344,
      "step": 49900
    },
    {
      "epoch": 7.432975871313673,
      "grad_norm": 13.0,
      "learning_rate": 3.957851161761502e-05,
      "loss": 1.0483,
      "num_input_tokens_seen": 28953160,
      "step": 49905
    },
    {
      "epoch": 7.433720583854632,
      "grad_norm": 7.84375,
      "learning_rate": 3.9575871757607385e-05,
      "loss": 0.9648,
      "num_input_tokens_seen": 28956136,
      "step": 49910
    },
    {
      "epoch": 7.434465296395591,
      "grad_norm": 6.4375,
      "learning_rate": 3.9573231651355225e-05,
      "loss": 1.2047,
      "num_input_tokens_seen": 28958920,
      "step": 49915
    },
    {
      "epoch": 7.435210008936551,
      "grad_norm": 10.3125,
      "learning_rate": 3.957059129890311e-05,
      "loss": 0.9267,
      "num_input_tokens_seen": 28962152,
      "step": 49920
    },
    {
      "epoch": 7.4359547214775095,
      "grad_norm": 5.3125,
      "learning_rate": 3.956795070029568e-05,
      "loss": 1.0467,
      "num_input_tokens_seen": 28965256,
      "step": 49925
    },
    {
      "epoch": 7.436699434018469,
      "grad_norm": 9.875,
      "learning_rate": 3.956530985557753e-05,
      "loss": 1.1477,
      "num_input_tokens_seen": 28968008,
      "step": 49930
    },
    {
      "epoch": 7.437444146559428,
      "grad_norm": 14.375,
      "learning_rate": 3.9562668764793264e-05,
      "loss": 1.0222,
      "num_input_tokens_seen": 28970952,
      "step": 49935
    },
    {
      "epoch": 7.4381888591003875,
      "grad_norm": 9.6875,
      "learning_rate": 3.9560027427987515e-05,
      "loss": 0.8103,
      "num_input_tokens_seen": 28974056,
      "step": 49940
    },
    {
      "epoch": 7.438933571641346,
      "grad_norm": 9.875,
      "learning_rate": 3.9557385845204895e-05,
      "loss": 1.1836,
      "num_input_tokens_seen": 28976840,
      "step": 49945
    },
    {
      "epoch": 7.439678284182306,
      "grad_norm": 13.3125,
      "learning_rate": 3.955474401649004e-05,
      "loss": 0.9943,
      "num_input_tokens_seen": 28979560,
      "step": 49950
    },
    {
      "epoch": 7.440422996723265,
      "grad_norm": 6.03125,
      "learning_rate": 3.955210194188758e-05,
      "loss": 1.0249,
      "num_input_tokens_seen": 28982536,
      "step": 49955
    },
    {
      "epoch": 7.441167709264224,
      "grad_norm": 13.0,
      "learning_rate": 3.954945962144214e-05,
      "loss": 1.533,
      "num_input_tokens_seen": 28985608,
      "step": 49960
    },
    {
      "epoch": 7.441912421805183,
      "grad_norm": 10.125,
      "learning_rate": 3.9546817055198385e-05,
      "loss": 1.1321,
      "num_input_tokens_seen": 28988360,
      "step": 49965
    },
    {
      "epoch": 7.442657134346143,
      "grad_norm": 8.5625,
      "learning_rate": 3.954417424320092e-05,
      "loss": 1.1591,
      "num_input_tokens_seen": 28991464,
      "step": 49970
    },
    {
      "epoch": 7.4434018468871015,
      "grad_norm": 11.8125,
      "learning_rate": 3.954153118549442e-05,
      "loss": 0.7608,
      "num_input_tokens_seen": 28994472,
      "step": 49975
    },
    {
      "epoch": 7.444146559428061,
      "grad_norm": 7.0625,
      "learning_rate": 3.953888788212353e-05,
      "loss": 1.0856,
      "num_input_tokens_seen": 28997512,
      "step": 49980
    },
    {
      "epoch": 7.44489127196902,
      "grad_norm": 6.6875,
      "learning_rate": 3.953624433313291e-05,
      "loss": 0.9515,
      "num_input_tokens_seen": 29000168,
      "step": 49985
    },
    {
      "epoch": 7.4456359845099795,
      "grad_norm": 7.625,
      "learning_rate": 3.9533600538567214e-05,
      "loss": 0.8751,
      "num_input_tokens_seen": 29003080,
      "step": 49990
    },
    {
      "epoch": 7.446380697050938,
      "grad_norm": 7.59375,
      "learning_rate": 3.953095649847111e-05,
      "loss": 1.1368,
      "num_input_tokens_seen": 29006088,
      "step": 49995
    },
    {
      "epoch": 7.447125409591898,
      "grad_norm": 10.0,
      "learning_rate": 3.952831221288926e-05,
      "loss": 1.1578,
      "num_input_tokens_seen": 29009096,
      "step": 50000
    },
    {
      "epoch": 7.447870122132857,
      "grad_norm": 7.625,
      "learning_rate": 3.9525667681866344e-05,
      "loss": 1.0242,
      "num_input_tokens_seen": 29012264,
      "step": 50005
    },
    {
      "epoch": 7.4486148346738155,
      "grad_norm": 14.5,
      "learning_rate": 3.952302290544704e-05,
      "loss": 1.2264,
      "num_input_tokens_seen": 29015720,
      "step": 50010
    },
    {
      "epoch": 7.449359547214775,
      "grad_norm": 11.375,
      "learning_rate": 3.952037788367602e-05,
      "loss": 0.9931,
      "num_input_tokens_seen": 29018856,
      "step": 50015
    },
    {
      "epoch": 7.450104259755734,
      "grad_norm": 4.125,
      "learning_rate": 3.951773261659797e-05,
      "loss": 0.8623,
      "num_input_tokens_seen": 29021544,
      "step": 50020
    },
    {
      "epoch": 7.4508489722966935,
      "grad_norm": 7.0625,
      "learning_rate": 3.951508710425758e-05,
      "loss": 0.8973,
      "num_input_tokens_seen": 29024584,
      "step": 50025
    },
    {
      "epoch": 7.451593684837652,
      "grad_norm": 7.03125,
      "learning_rate": 3.9512441346699554e-05,
      "loss": 1.176,
      "num_input_tokens_seen": 29027656,
      "step": 50030
    },
    {
      "epoch": 7.452338397378612,
      "grad_norm": 14.1875,
      "learning_rate": 3.950979534396858e-05,
      "loss": 1.1664,
      "num_input_tokens_seen": 29030568,
      "step": 50035
    },
    {
      "epoch": 7.453083109919571,
      "grad_norm": 16.125,
      "learning_rate": 3.9507149096109366e-05,
      "loss": 1.0703,
      "num_input_tokens_seen": 29033288,
      "step": 50040
    },
    {
      "epoch": 7.45382782246053,
      "grad_norm": 13.625,
      "learning_rate": 3.95045026031666e-05,
      "loss": 0.9086,
      "num_input_tokens_seen": 29035976,
      "step": 50045
    },
    {
      "epoch": 7.454572535001489,
      "grad_norm": 12.625,
      "learning_rate": 3.950185586518501e-05,
      "loss": 1.2984,
      "num_input_tokens_seen": 29039080,
      "step": 50050
    },
    {
      "epoch": 7.455317247542449,
      "grad_norm": 8.25,
      "learning_rate": 3.94992088822093e-05,
      "loss": 0.9965,
      "num_input_tokens_seen": 29042024,
      "step": 50055
    },
    {
      "epoch": 7.4560619600834075,
      "grad_norm": 17.875,
      "learning_rate": 3.94965616542842e-05,
      "loss": 1.4088,
      "num_input_tokens_seen": 29044616,
      "step": 50060
    },
    {
      "epoch": 7.456806672624367,
      "grad_norm": 5.96875,
      "learning_rate": 3.949391418145442e-05,
      "loss": 1.3048,
      "num_input_tokens_seen": 29047432,
      "step": 50065
    },
    {
      "epoch": 7.457551385165326,
      "grad_norm": 6.75,
      "learning_rate": 3.9491266463764694e-05,
      "loss": 0.8996,
      "num_input_tokens_seen": 29050248,
      "step": 50070
    },
    {
      "epoch": 7.4582960977062855,
      "grad_norm": 10.8125,
      "learning_rate": 3.948861850125974e-05,
      "loss": 1.1496,
      "num_input_tokens_seen": 29053096,
      "step": 50075
    },
    {
      "epoch": 7.459040810247244,
      "grad_norm": 7.375,
      "learning_rate": 3.948597029398432e-05,
      "loss": 1.1127,
      "num_input_tokens_seen": 29056264,
      "step": 50080
    },
    {
      "epoch": 7.459785522788204,
      "grad_norm": 9.5625,
      "learning_rate": 3.9483321841983146e-05,
      "loss": 1.0158,
      "num_input_tokens_seen": 29058888,
      "step": 50085
    },
    {
      "epoch": 7.460530235329163,
      "grad_norm": 11.75,
      "learning_rate": 3.948067314530096e-05,
      "loss": 1.1023,
      "num_input_tokens_seen": 29061704,
      "step": 50090
    },
    {
      "epoch": 7.461274947870122,
      "grad_norm": 5.40625,
      "learning_rate": 3.947802420398253e-05,
      "loss": 0.8219,
      "num_input_tokens_seen": 29064584,
      "step": 50095
    },
    {
      "epoch": 7.462019660411081,
      "grad_norm": 15.5625,
      "learning_rate": 3.947537501807259e-05,
      "loss": 1.2547,
      "num_input_tokens_seen": 29067432,
      "step": 50100
    },
    {
      "epoch": 7.462764372952041,
      "grad_norm": 7.0,
      "learning_rate": 3.947272558761591e-05,
      "loss": 1.0554,
      "num_input_tokens_seen": 29070344,
      "step": 50105
    },
    {
      "epoch": 7.4635090854929995,
      "grad_norm": 11.5,
      "learning_rate": 3.947007591265723e-05,
      "loss": 0.9254,
      "num_input_tokens_seen": 29073160,
      "step": 50110
    },
    {
      "epoch": 7.464253798033959,
      "grad_norm": 11.5,
      "learning_rate": 3.9467425993241326e-05,
      "loss": 1.3754,
      "num_input_tokens_seen": 29076104,
      "step": 50115
    },
    {
      "epoch": 7.464998510574918,
      "grad_norm": 10.6875,
      "learning_rate": 3.946477582941297e-05,
      "loss": 0.9102,
      "num_input_tokens_seen": 29079080,
      "step": 50120
    },
    {
      "epoch": 7.465743223115878,
      "grad_norm": 12.5625,
      "learning_rate": 3.946212542121692e-05,
      "loss": 1.1187,
      "num_input_tokens_seen": 29081768,
      "step": 50125
    },
    {
      "epoch": 7.466487935656836,
      "grad_norm": 8.625,
      "learning_rate": 3.945947476869797e-05,
      "loss": 1.6022,
      "num_input_tokens_seen": 29084840,
      "step": 50130
    },
    {
      "epoch": 7.467232648197796,
      "grad_norm": 5.59375,
      "learning_rate": 3.945682387190088e-05,
      "loss": 1.2035,
      "num_input_tokens_seen": 29087464,
      "step": 50135
    },
    {
      "epoch": 7.467977360738755,
      "grad_norm": 22.75,
      "learning_rate": 3.9454172730870445e-05,
      "loss": 1.5145,
      "num_input_tokens_seen": 29090312,
      "step": 50140
    },
    {
      "epoch": 7.468722073279714,
      "grad_norm": 4.65625,
      "learning_rate": 3.9451521345651456e-05,
      "loss": 0.8236,
      "num_input_tokens_seen": 29093448,
      "step": 50145
    },
    {
      "epoch": 7.469466785820673,
      "grad_norm": 7.03125,
      "learning_rate": 3.94488697162887e-05,
      "loss": 1.2771,
      "num_input_tokens_seen": 29096840,
      "step": 50150
    },
    {
      "epoch": 7.470211498361633,
      "grad_norm": 11.0625,
      "learning_rate": 3.944621784282697e-05,
      "loss": 1.0987,
      "num_input_tokens_seen": 29099720,
      "step": 50155
    },
    {
      "epoch": 7.4709562109025915,
      "grad_norm": 10.8125,
      "learning_rate": 3.944356572531108e-05,
      "loss": 1.1781,
      "num_input_tokens_seen": 29103016,
      "step": 50160
    },
    {
      "epoch": 7.471700923443551,
      "grad_norm": 3.8125,
      "learning_rate": 3.944091336378583e-05,
      "loss": 0.9844,
      "num_input_tokens_seen": 29105768,
      "step": 50165
    },
    {
      "epoch": 7.47244563598451,
      "grad_norm": 11.0625,
      "learning_rate": 3.943826075829602e-05,
      "loss": 1.0326,
      "num_input_tokens_seen": 29108808,
      "step": 50170
    },
    {
      "epoch": 7.473190348525469,
      "grad_norm": 13.1875,
      "learning_rate": 3.943560790888647e-05,
      "loss": 1.1594,
      "num_input_tokens_seen": 29111336,
      "step": 50175
    },
    {
      "epoch": 7.473935061066428,
      "grad_norm": 6.78125,
      "learning_rate": 3.9432954815601995e-05,
      "loss": 1.1299,
      "num_input_tokens_seen": 29114216,
      "step": 50180
    },
    {
      "epoch": 7.474679773607388,
      "grad_norm": 15.5625,
      "learning_rate": 3.943030147848742e-05,
      "loss": 1.3415,
      "num_input_tokens_seen": 29117032,
      "step": 50185
    },
    {
      "epoch": 7.475424486148347,
      "grad_norm": 9.3125,
      "learning_rate": 3.9427647897587564e-05,
      "loss": 1.3329,
      "num_input_tokens_seen": 29120136,
      "step": 50190
    },
    {
      "epoch": 7.4761691986893055,
      "grad_norm": 11.0625,
      "learning_rate": 3.9424994072947256e-05,
      "loss": 1.014,
      "num_input_tokens_seen": 29123016,
      "step": 50195
    },
    {
      "epoch": 7.476913911230265,
      "grad_norm": 6.34375,
      "learning_rate": 3.942234000461135e-05,
      "loss": 1.1278,
      "num_input_tokens_seen": 29126024,
      "step": 50200
    },
    {
      "epoch": 7.477658623771224,
      "grad_norm": 11.0,
      "learning_rate": 3.941968569262465e-05,
      "loss": 1.0272,
      "num_input_tokens_seen": 29129000,
      "step": 50205
    },
    {
      "epoch": 7.478403336312184,
      "grad_norm": 12.125,
      "learning_rate": 3.9417031137032025e-05,
      "loss": 1.4818,
      "num_input_tokens_seen": 29132104,
      "step": 50210
    },
    {
      "epoch": 7.479148048853142,
      "grad_norm": 11.75,
      "learning_rate": 3.941437633787831e-05,
      "loss": 1.181,
      "num_input_tokens_seen": 29135368,
      "step": 50215
    },
    {
      "epoch": 7.479892761394102,
      "grad_norm": 12.3125,
      "learning_rate": 3.941172129520836e-05,
      "loss": 1.0291,
      "num_input_tokens_seen": 29138088,
      "step": 50220
    },
    {
      "epoch": 7.480637473935061,
      "grad_norm": 6.375,
      "learning_rate": 3.940906600906702e-05,
      "loss": 1.0404,
      "num_input_tokens_seen": 29141000,
      "step": 50225
    },
    {
      "epoch": 7.48138218647602,
      "grad_norm": 4.28125,
      "learning_rate": 3.9406410479499155e-05,
      "loss": 1.1803,
      "num_input_tokens_seen": 29144104,
      "step": 50230
    },
    {
      "epoch": 7.482126899016979,
      "grad_norm": 10.0625,
      "learning_rate": 3.940375470654963e-05,
      "loss": 0.9446,
      "num_input_tokens_seen": 29146792,
      "step": 50235
    },
    {
      "epoch": 7.482871611557939,
      "grad_norm": 7.59375,
      "learning_rate": 3.9401098690263316e-05,
      "loss": 0.9125,
      "num_input_tokens_seen": 29149736,
      "step": 50240
    },
    {
      "epoch": 7.4836163240988975,
      "grad_norm": 12.5625,
      "learning_rate": 3.939844243068507e-05,
      "loss": 1.034,
      "num_input_tokens_seen": 29152584,
      "step": 50245
    },
    {
      "epoch": 7.484361036639857,
      "grad_norm": 16.375,
      "learning_rate": 3.939578592785977e-05,
      "loss": 1.253,
      "num_input_tokens_seen": 29155656,
      "step": 50250
    },
    {
      "epoch": 7.485105749180816,
      "grad_norm": 11.8125,
      "learning_rate": 3.93931291818323e-05,
      "loss": 1.1626,
      "num_input_tokens_seen": 29158472,
      "step": 50255
    },
    {
      "epoch": 7.485850461721776,
      "grad_norm": 16.75,
      "learning_rate": 3.939047219264754e-05,
      "loss": 1.1808,
      "num_input_tokens_seen": 29161256,
      "step": 50260
    },
    {
      "epoch": 7.486595174262734,
      "grad_norm": 8.1875,
      "learning_rate": 3.938781496035038e-05,
      "loss": 0.9723,
      "num_input_tokens_seen": 29164104,
      "step": 50265
    },
    {
      "epoch": 7.487339886803694,
      "grad_norm": 11.5,
      "learning_rate": 3.93851574849857e-05,
      "loss": 1.344,
      "num_input_tokens_seen": 29166824,
      "step": 50270
    },
    {
      "epoch": 7.488084599344653,
      "grad_norm": 8.0,
      "learning_rate": 3.9382499766598416e-05,
      "loss": 0.8746,
      "num_input_tokens_seen": 29169864,
      "step": 50275
    },
    {
      "epoch": 7.488829311885612,
      "grad_norm": 12.5625,
      "learning_rate": 3.937984180523342e-05,
      "loss": 1.1188,
      "num_input_tokens_seen": 29172712,
      "step": 50280
    },
    {
      "epoch": 7.489574024426571,
      "grad_norm": 8.875,
      "learning_rate": 3.9377183600935595e-05,
      "loss": 1.1945,
      "num_input_tokens_seen": 29175592,
      "step": 50285
    },
    {
      "epoch": 7.490318736967531,
      "grad_norm": 13.1875,
      "learning_rate": 3.937452515374987e-05,
      "loss": 1.3438,
      "num_input_tokens_seen": 29178440,
      "step": 50290
    },
    {
      "epoch": 7.49106344950849,
      "grad_norm": 6.75,
      "learning_rate": 3.9371866463721165e-05,
      "loss": 1.2529,
      "num_input_tokens_seen": 29181352,
      "step": 50295
    },
    {
      "epoch": 7.491808162049449,
      "grad_norm": 5.5625,
      "learning_rate": 3.9369207530894374e-05,
      "loss": 0.952,
      "num_input_tokens_seen": 29183944,
      "step": 50300
    },
    {
      "epoch": 7.492552874590408,
      "grad_norm": 14.5625,
      "learning_rate": 3.9366548355314426e-05,
      "loss": 1.4086,
      "num_input_tokens_seen": 29186696,
      "step": 50305
    },
    {
      "epoch": 7.493297587131368,
      "grad_norm": 18.5,
      "learning_rate": 3.936388893702625e-05,
      "loss": 0.7738,
      "num_input_tokens_seen": 29189480,
      "step": 50310
    },
    {
      "epoch": 7.494042299672326,
      "grad_norm": 10.8125,
      "learning_rate": 3.936122927607476e-05,
      "loss": 1.3512,
      "num_input_tokens_seen": 29192520,
      "step": 50315
    },
    {
      "epoch": 7.494787012213286,
      "grad_norm": 9.5625,
      "learning_rate": 3.935856937250491e-05,
      "loss": 1.1018,
      "num_input_tokens_seen": 29195304,
      "step": 50320
    },
    {
      "epoch": 7.495531724754245,
      "grad_norm": 8.75,
      "learning_rate": 3.935590922636161e-05,
      "loss": 1.1129,
      "num_input_tokens_seen": 29198184,
      "step": 50325
    },
    {
      "epoch": 7.496276437295204,
      "grad_norm": 13.5,
      "learning_rate": 3.935324883768983e-05,
      "loss": 1.307,
      "num_input_tokens_seen": 29201096,
      "step": 50330
    },
    {
      "epoch": 7.497021149836163,
      "grad_norm": 4.5625,
      "learning_rate": 3.9350588206534486e-05,
      "loss": 0.9245,
      "num_input_tokens_seen": 29204072,
      "step": 50335
    },
    {
      "epoch": 7.497765862377123,
      "grad_norm": 7.15625,
      "learning_rate": 3.934792733294054e-05,
      "loss": 0.8303,
      "num_input_tokens_seen": 29206952,
      "step": 50340
    },
    {
      "epoch": 7.498510574918082,
      "grad_norm": 11.25,
      "learning_rate": 3.9345266216952945e-05,
      "loss": 1.5075,
      "num_input_tokens_seen": 29209800,
      "step": 50345
    },
    {
      "epoch": 7.499255287459041,
      "grad_norm": 8.5625,
      "learning_rate": 3.934260485861667e-05,
      "loss": 1.0035,
      "num_input_tokens_seen": 29212680,
      "step": 50350
    },
    {
      "epoch": 7.5,
      "grad_norm": 15.375,
      "learning_rate": 3.933994325797665e-05,
      "loss": 1.0926,
      "num_input_tokens_seen": 29215400,
      "step": 50355
    },
    {
      "epoch": 7.500744712540959,
      "grad_norm": 16.75,
      "learning_rate": 3.9337281415077866e-05,
      "loss": 1.2965,
      "num_input_tokens_seen": 29218504,
      "step": 50360
    },
    {
      "epoch": 7.501489425081918,
      "grad_norm": 8.25,
      "learning_rate": 3.933461932996528e-05,
      "loss": 1.2306,
      "num_input_tokens_seen": 29221896,
      "step": 50365
    },
    {
      "epoch": 7.502234137622878,
      "grad_norm": 12.8125,
      "learning_rate": 3.933195700268388e-05,
      "loss": 1.1679,
      "num_input_tokens_seen": 29224520,
      "step": 50370
    },
    {
      "epoch": 7.502978850163837,
      "grad_norm": 14.0625,
      "learning_rate": 3.932929443327862e-05,
      "loss": 1.3517,
      "num_input_tokens_seen": 29227528,
      "step": 50375
    },
    {
      "epoch": 7.503723562704796,
      "grad_norm": 5.625,
      "learning_rate": 3.932663162179451e-05,
      "loss": 0.9862,
      "num_input_tokens_seen": 29230632,
      "step": 50380
    },
    {
      "epoch": 7.504468275245755,
      "grad_norm": 13.4375,
      "learning_rate": 3.93239685682765e-05,
      "loss": 1.2537,
      "num_input_tokens_seen": 29233320,
      "step": 50385
    },
    {
      "epoch": 7.505212987786714,
      "grad_norm": 4.96875,
      "learning_rate": 3.932130527276961e-05,
      "loss": 0.8536,
      "num_input_tokens_seen": 29236200,
      "step": 50390
    },
    {
      "epoch": 7.505957700327674,
      "grad_norm": 5.03125,
      "learning_rate": 3.931864173531883e-05,
      "loss": 0.8861,
      "num_input_tokens_seen": 29239112,
      "step": 50395
    },
    {
      "epoch": 7.506702412868632,
      "grad_norm": 17.875,
      "learning_rate": 3.931597795596914e-05,
      "loss": 1.401,
      "num_input_tokens_seen": 29242248,
      "step": 50400
    },
    {
      "epoch": 7.507447125409592,
      "grad_norm": 11.8125,
      "learning_rate": 3.931331393476556e-05,
      "loss": 1.0699,
      "num_input_tokens_seen": 29244904,
      "step": 50405
    },
    {
      "epoch": 7.508191837950551,
      "grad_norm": 10.3125,
      "learning_rate": 3.931064967175309e-05,
      "loss": 1.1295,
      "num_input_tokens_seen": 29247752,
      "step": 50410
    },
    {
      "epoch": 7.50893655049151,
      "grad_norm": 8.75,
      "learning_rate": 3.9307985166976726e-05,
      "loss": 1.1038,
      "num_input_tokens_seen": 29250888,
      "step": 50415
    },
    {
      "epoch": 7.509681263032469,
      "grad_norm": 11.75,
      "learning_rate": 3.93053204204815e-05,
      "loss": 1.5387,
      "num_input_tokens_seen": 29253832,
      "step": 50420
    },
    {
      "epoch": 7.510425975573429,
      "grad_norm": 15.1875,
      "learning_rate": 3.930265543231243e-05,
      "loss": 1.0546,
      "num_input_tokens_seen": 29256648,
      "step": 50425
    },
    {
      "epoch": 7.511170688114388,
      "grad_norm": 11.4375,
      "learning_rate": 3.9299990202514525e-05,
      "loss": 1.2013,
      "num_input_tokens_seen": 29259400,
      "step": 50430
    },
    {
      "epoch": 7.511915400655347,
      "grad_norm": 10.5,
      "learning_rate": 3.9297324731132826e-05,
      "loss": 1.0394,
      "num_input_tokens_seen": 29262376,
      "step": 50435
    },
    {
      "epoch": 7.512660113196306,
      "grad_norm": 9.625,
      "learning_rate": 3.9294659018212356e-05,
      "loss": 0.6678,
      "num_input_tokens_seen": 29265160,
      "step": 50440
    },
    {
      "epoch": 7.513404825737266,
      "grad_norm": 13.0,
      "learning_rate": 3.929199306379815e-05,
      "loss": 1.0687,
      "num_input_tokens_seen": 29267880,
      "step": 50445
    },
    {
      "epoch": 7.514149538278224,
      "grad_norm": 11.8125,
      "learning_rate": 3.928932686793524e-05,
      "loss": 0.7945,
      "num_input_tokens_seen": 29270600,
      "step": 50450
    },
    {
      "epoch": 7.514894250819184,
      "grad_norm": 7.125,
      "learning_rate": 3.9286660430668686e-05,
      "loss": 0.9898,
      "num_input_tokens_seen": 29273480,
      "step": 50455
    },
    {
      "epoch": 7.515638963360143,
      "grad_norm": 9.5625,
      "learning_rate": 3.928399375204352e-05,
      "loss": 1.1441,
      "num_input_tokens_seen": 29276040,
      "step": 50460
    },
    {
      "epoch": 7.5163836759011025,
      "grad_norm": 13.0,
      "learning_rate": 3.9281326832104795e-05,
      "loss": 1.2081,
      "num_input_tokens_seen": 29278472,
      "step": 50465
    },
    {
      "epoch": 7.517128388442061,
      "grad_norm": 11.0625,
      "learning_rate": 3.9278659670897564e-05,
      "loss": 1.0284,
      "num_input_tokens_seen": 29281320,
      "step": 50470
    },
    {
      "epoch": 7.517873100983021,
      "grad_norm": 11.0625,
      "learning_rate": 3.92759922684669e-05,
      "loss": 1.2978,
      "num_input_tokens_seen": 29284296,
      "step": 50475
    },
    {
      "epoch": 7.51861781352398,
      "grad_norm": 12.25,
      "learning_rate": 3.927332462485785e-05,
      "loss": 1.3387,
      "num_input_tokens_seen": 29287048,
      "step": 50480
    },
    {
      "epoch": 7.519362526064939,
      "grad_norm": 11.5,
      "learning_rate": 3.92706567401155e-05,
      "loss": 0.9465,
      "num_input_tokens_seen": 29289992,
      "step": 50485
    },
    {
      "epoch": 7.520107238605898,
      "grad_norm": 8.625,
      "learning_rate": 3.9267988614284886e-05,
      "loss": 0.8766,
      "num_input_tokens_seen": 29292840,
      "step": 50490
    },
    {
      "epoch": 7.520851951146858,
      "grad_norm": 7.25,
      "learning_rate": 3.926532024741113e-05,
      "loss": 1.0887,
      "num_input_tokens_seen": 29296136,
      "step": 50495
    },
    {
      "epoch": 7.521596663687816,
      "grad_norm": 6.84375,
      "learning_rate": 3.926265163953927e-05,
      "loss": 0.8031,
      "num_input_tokens_seen": 29299368,
      "step": 50500
    },
    {
      "epoch": 7.522341376228776,
      "grad_norm": 20.625,
      "learning_rate": 3.925998279071441e-05,
      "loss": 1.7334,
      "num_input_tokens_seen": 29302088,
      "step": 50505
    },
    {
      "epoch": 7.523086088769735,
      "grad_norm": 9.375,
      "learning_rate": 3.9257313700981634e-05,
      "loss": 1.1926,
      "num_input_tokens_seen": 29305096,
      "step": 50510
    },
    {
      "epoch": 7.5238308013106945,
      "grad_norm": 10.375,
      "learning_rate": 3.9254644370386036e-05,
      "loss": 0.8978,
      "num_input_tokens_seen": 29308072,
      "step": 50515
    },
    {
      "epoch": 7.524575513851653,
      "grad_norm": 11.4375,
      "learning_rate": 3.925197479897271e-05,
      "loss": 1.1246,
      "num_input_tokens_seen": 29310952,
      "step": 50520
    },
    {
      "epoch": 7.525320226392612,
      "grad_norm": 11.75,
      "learning_rate": 3.924930498678675e-05,
      "loss": 1.1562,
      "num_input_tokens_seen": 29313832,
      "step": 50525
    },
    {
      "epoch": 7.526064938933572,
      "grad_norm": 7.5,
      "learning_rate": 3.924663493387326e-05,
      "loss": 0.9174,
      "num_input_tokens_seen": 29316488,
      "step": 50530
    },
    {
      "epoch": 7.526809651474531,
      "grad_norm": 9.4375,
      "learning_rate": 3.924396464027736e-05,
      "loss": 1.0705,
      "num_input_tokens_seen": 29319272,
      "step": 50535
    },
    {
      "epoch": 7.52755436401549,
      "grad_norm": 14.0625,
      "learning_rate": 3.924129410604416e-05,
      "loss": 1.2424,
      "num_input_tokens_seen": 29322184,
      "step": 50540
    },
    {
      "epoch": 7.528299076556449,
      "grad_norm": 14.8125,
      "learning_rate": 3.923862333121876e-05,
      "loss": 1.2179,
      "num_input_tokens_seen": 29325032,
      "step": 50545
    },
    {
      "epoch": 7.5290437890974085,
      "grad_norm": 18.5,
      "learning_rate": 3.92359523158463e-05,
      "loss": 1.5516,
      "num_input_tokens_seen": 29327880,
      "step": 50550
    },
    {
      "epoch": 7.529788501638367,
      "grad_norm": 6.46875,
      "learning_rate": 3.923328105997188e-05,
      "loss": 0.9695,
      "num_input_tokens_seen": 29330568,
      "step": 50555
    },
    {
      "epoch": 7.530533214179327,
      "grad_norm": 10.1875,
      "learning_rate": 3.923060956364066e-05,
      "loss": 1.1371,
      "num_input_tokens_seen": 29333736,
      "step": 50560
    },
    {
      "epoch": 7.531277926720286,
      "grad_norm": 7.5,
      "learning_rate": 3.922793782689774e-05,
      "loss": 1.2759,
      "num_input_tokens_seen": 29336712,
      "step": 50565
    },
    {
      "epoch": 7.532022639261245,
      "grad_norm": 18.0,
      "learning_rate": 3.922526584978829e-05,
      "loss": 1.1138,
      "num_input_tokens_seen": 29339528,
      "step": 50570
    },
    {
      "epoch": 7.532767351802204,
      "grad_norm": 16.875,
      "learning_rate": 3.922259363235741e-05,
      "loss": 1.2353,
      "num_input_tokens_seen": 29342920,
      "step": 50575
    },
    {
      "epoch": 7.533512064343164,
      "grad_norm": 10.0625,
      "learning_rate": 3.921992117465028e-05,
      "loss": 1.1593,
      "num_input_tokens_seen": 29345832,
      "step": 50580
    },
    {
      "epoch": 7.534256776884122,
      "grad_norm": 17.5,
      "learning_rate": 3.921724847671202e-05,
      "loss": 1.1203,
      "num_input_tokens_seen": 29348712,
      "step": 50585
    },
    {
      "epoch": 7.535001489425082,
      "grad_norm": 12.3125,
      "learning_rate": 3.9214575538587804e-05,
      "loss": 1.104,
      "num_input_tokens_seen": 29351720,
      "step": 50590
    },
    {
      "epoch": 7.535746201966041,
      "grad_norm": 10.0,
      "learning_rate": 3.921190236032278e-05,
      "loss": 0.918,
      "num_input_tokens_seen": 29354504,
      "step": 50595
    },
    {
      "epoch": 7.5364909145070005,
      "grad_norm": 8.5,
      "learning_rate": 3.920922894196212e-05,
      "loss": 0.9968,
      "num_input_tokens_seen": 29357736,
      "step": 50600
    },
    {
      "epoch": 7.537235627047959,
      "grad_norm": 6.65625,
      "learning_rate": 3.920655528355097e-05,
      "loss": 0.9667,
      "num_input_tokens_seen": 29360808,
      "step": 50605
    },
    {
      "epoch": 7.537980339588919,
      "grad_norm": 9.5625,
      "learning_rate": 3.920388138513451e-05,
      "loss": 1.1,
      "num_input_tokens_seen": 29363976,
      "step": 50610
    },
    {
      "epoch": 7.538725052129878,
      "grad_norm": 9.9375,
      "learning_rate": 3.920120724675791e-05,
      "loss": 1.0961,
      "num_input_tokens_seen": 29366728,
      "step": 50615
    },
    {
      "epoch": 7.539469764670837,
      "grad_norm": 18.875,
      "learning_rate": 3.9198532868466345e-05,
      "loss": 1.2194,
      "num_input_tokens_seen": 29369544,
      "step": 50620
    },
    {
      "epoch": 7.540214477211796,
      "grad_norm": 11.6875,
      "learning_rate": 3.919585825030499e-05,
      "loss": 1.1371,
      "num_input_tokens_seen": 29372552,
      "step": 50625
    },
    {
      "epoch": 7.540959189752756,
      "grad_norm": 18.5,
      "learning_rate": 3.9193183392319054e-05,
      "loss": 0.9822,
      "num_input_tokens_seen": 29375784,
      "step": 50630
    },
    {
      "epoch": 7.5417039022937145,
      "grad_norm": 7.5625,
      "learning_rate": 3.9190508294553694e-05,
      "loss": 1.1302,
      "num_input_tokens_seen": 29378792,
      "step": 50635
    },
    {
      "epoch": 7.542448614834674,
      "grad_norm": 13.5625,
      "learning_rate": 3.918783295705414e-05,
      "loss": 0.8325,
      "num_input_tokens_seen": 29381608,
      "step": 50640
    },
    {
      "epoch": 7.543193327375633,
      "grad_norm": 5.15625,
      "learning_rate": 3.9185157379865553e-05,
      "loss": 1.1888,
      "num_input_tokens_seen": 29384904,
      "step": 50645
    },
    {
      "epoch": 7.5439380399165925,
      "grad_norm": 9.125,
      "learning_rate": 3.9182481563033155e-05,
      "loss": 1.0272,
      "num_input_tokens_seen": 29387720,
      "step": 50650
    },
    {
      "epoch": 7.544682752457551,
      "grad_norm": 8.6875,
      "learning_rate": 3.917980550660214e-05,
      "loss": 1.0413,
      "num_input_tokens_seen": 29390632,
      "step": 50655
    },
    {
      "epoch": 7.545427464998511,
      "grad_norm": 5.53125,
      "learning_rate": 3.9177129210617725e-05,
      "loss": 0.9692,
      "num_input_tokens_seen": 29393672,
      "step": 50660
    },
    {
      "epoch": 7.54617217753947,
      "grad_norm": 13.75,
      "learning_rate": 3.9174452675125115e-05,
      "loss": 0.9211,
      "num_input_tokens_seen": 29396424,
      "step": 50665
    },
    {
      "epoch": 7.546916890080429,
      "grad_norm": 4.8125,
      "learning_rate": 3.917177590016954e-05,
      "loss": 1.048,
      "num_input_tokens_seen": 29399272,
      "step": 50670
    },
    {
      "epoch": 7.547661602621388,
      "grad_norm": 10.875,
      "learning_rate": 3.9169098885796216e-05,
      "loss": 1.0834,
      "num_input_tokens_seen": 29402472,
      "step": 50675
    },
    {
      "epoch": 7.548406315162348,
      "grad_norm": 14.1875,
      "learning_rate": 3.916642163205036e-05,
      "loss": 1.4713,
      "num_input_tokens_seen": 29405352,
      "step": 50680
    },
    {
      "epoch": 7.5491510277033065,
      "grad_norm": 8.9375,
      "learning_rate": 3.916374413897722e-05,
      "loss": 1.0666,
      "num_input_tokens_seen": 29408264,
      "step": 50685
    },
    {
      "epoch": 7.549895740244265,
      "grad_norm": 9.875,
      "learning_rate": 3.916106640662201e-05,
      "loss": 1.136,
      "num_input_tokens_seen": 29411144,
      "step": 50690
    },
    {
      "epoch": 7.550640452785225,
      "grad_norm": 6.625,
      "learning_rate": 3.915838843502998e-05,
      "loss": 1.1291,
      "num_input_tokens_seen": 29414152,
      "step": 50695
    },
    {
      "epoch": 7.5513851653261845,
      "grad_norm": 4.65625,
      "learning_rate": 3.9155710224246365e-05,
      "loss": 1.0007,
      "num_input_tokens_seen": 29417224,
      "step": 50700
    },
    {
      "epoch": 7.552129877867143,
      "grad_norm": 8.375,
      "learning_rate": 3.915303177431641e-05,
      "loss": 0.8489,
      "num_input_tokens_seen": 29420040,
      "step": 50705
    },
    {
      "epoch": 7.552874590408102,
      "grad_norm": 11.625,
      "learning_rate": 3.915035308528537e-05,
      "loss": 1.331,
      "num_input_tokens_seen": 29422856,
      "step": 50710
    },
    {
      "epoch": 7.553619302949062,
      "grad_norm": 7.96875,
      "learning_rate": 3.91476741571985e-05,
      "loss": 1.008,
      "num_input_tokens_seen": 29425672,
      "step": 50715
    },
    {
      "epoch": 7.554364015490021,
      "grad_norm": 11.625,
      "learning_rate": 3.914499499010105e-05,
      "loss": 1.2985,
      "num_input_tokens_seen": 29428424,
      "step": 50720
    },
    {
      "epoch": 7.55510872803098,
      "grad_norm": 6.875,
      "learning_rate": 3.9142315584038284e-05,
      "loss": 1.0128,
      "num_input_tokens_seen": 29431208,
      "step": 50725
    },
    {
      "epoch": 7.555853440571939,
      "grad_norm": 10.0,
      "learning_rate": 3.913963593905548e-05,
      "loss": 0.8111,
      "num_input_tokens_seen": 29434056,
      "step": 50730
    },
    {
      "epoch": 7.5565981531128985,
      "grad_norm": 11.25,
      "learning_rate": 3.913695605519788e-05,
      "loss": 1.0432,
      "num_input_tokens_seen": 29437000,
      "step": 50735
    },
    {
      "epoch": 7.557342865653857,
      "grad_norm": 5.75,
      "learning_rate": 3.913427593251079e-05,
      "loss": 1.0176,
      "num_input_tokens_seen": 29439752,
      "step": 50740
    },
    {
      "epoch": 7.558087578194817,
      "grad_norm": 9.6875,
      "learning_rate": 3.913159557103947e-05,
      "loss": 1.1272,
      "num_input_tokens_seen": 29442952,
      "step": 50745
    },
    {
      "epoch": 7.558832290735776,
      "grad_norm": 31.0,
      "learning_rate": 3.912891497082921e-05,
      "loss": 1.2379,
      "num_input_tokens_seen": 29445800,
      "step": 50750
    },
    {
      "epoch": 7.559577003276735,
      "grad_norm": 8.1875,
      "learning_rate": 3.9126234131925285e-05,
      "loss": 1.0004,
      "num_input_tokens_seen": 29448456,
      "step": 50755
    },
    {
      "epoch": 7.560321715817694,
      "grad_norm": 8.1875,
      "learning_rate": 3.9123553054372994e-05,
      "loss": 1.1439,
      "num_input_tokens_seen": 29451176,
      "step": 50760
    },
    {
      "epoch": 7.561066428358654,
      "grad_norm": 9.0,
      "learning_rate": 3.912087173821762e-05,
      "loss": 1.4102,
      "num_input_tokens_seen": 29454088,
      "step": 50765
    },
    {
      "epoch": 7.5618111408996125,
      "grad_norm": 3.484375,
      "learning_rate": 3.911819018350449e-05,
      "loss": 1.0955,
      "num_input_tokens_seen": 29456936,
      "step": 50770
    },
    {
      "epoch": 7.562555853440572,
      "grad_norm": 11.1875,
      "learning_rate": 3.9115508390278864e-05,
      "loss": 1.0534,
      "num_input_tokens_seen": 29459624,
      "step": 50775
    },
    {
      "epoch": 7.563300565981531,
      "grad_norm": 8.875,
      "learning_rate": 3.9112826358586086e-05,
      "loss": 1.4059,
      "num_input_tokens_seen": 29462728,
      "step": 50780
    },
    {
      "epoch": 7.5640452785224905,
      "grad_norm": 7.40625,
      "learning_rate": 3.9110144088471437e-05,
      "loss": 1.078,
      "num_input_tokens_seen": 29465576,
      "step": 50785
    },
    {
      "epoch": 7.564789991063449,
      "grad_norm": 7.96875,
      "learning_rate": 3.9107461579980255e-05,
      "loss": 0.9875,
      "num_input_tokens_seen": 29468680,
      "step": 50790
    },
    {
      "epoch": 7.565534703604409,
      "grad_norm": 13.6875,
      "learning_rate": 3.910477883315785e-05,
      "loss": 1.1138,
      "num_input_tokens_seen": 29471528,
      "step": 50795
    },
    {
      "epoch": 7.566279416145368,
      "grad_norm": 13.875,
      "learning_rate": 3.910209584804953e-05,
      "loss": 1.0835,
      "num_input_tokens_seen": 29474152,
      "step": 50800
    },
    {
      "epoch": 7.567024128686327,
      "grad_norm": 11.4375,
      "learning_rate": 3.909941262470064e-05,
      "loss": 1.1187,
      "num_input_tokens_seen": 29477288,
      "step": 50805
    },
    {
      "epoch": 7.567768841227286,
      "grad_norm": 8.125,
      "learning_rate": 3.909672916315651e-05,
      "loss": 1.0483,
      "num_input_tokens_seen": 29480488,
      "step": 50810
    },
    {
      "epoch": 7.568513553768246,
      "grad_norm": 11.625,
      "learning_rate": 3.909404546346246e-05,
      "loss": 1.1494,
      "num_input_tokens_seen": 29483144,
      "step": 50815
    },
    {
      "epoch": 7.5692582663092045,
      "grad_norm": 8.6875,
      "learning_rate": 3.909136152566384e-05,
      "loss": 1.0744,
      "num_input_tokens_seen": 29486152,
      "step": 50820
    },
    {
      "epoch": 7.570002978850164,
      "grad_norm": 8.75,
      "learning_rate": 3.908867734980599e-05,
      "loss": 0.8921,
      "num_input_tokens_seen": 29489032,
      "step": 50825
    },
    {
      "epoch": 7.570747691391123,
      "grad_norm": 10.0625,
      "learning_rate": 3.908599293593425e-05,
      "loss": 1.2961,
      "num_input_tokens_seen": 29491752,
      "step": 50830
    },
    {
      "epoch": 7.571492403932083,
      "grad_norm": 10.3125,
      "learning_rate": 3.908330828409397e-05,
      "loss": 1.2965,
      "num_input_tokens_seen": 29494408,
      "step": 50835
    },
    {
      "epoch": 7.572237116473041,
      "grad_norm": 9.25,
      "learning_rate": 3.908062339433052e-05,
      "loss": 1.0547,
      "num_input_tokens_seen": 29497448,
      "step": 50840
    },
    {
      "epoch": 7.572981829014001,
      "grad_norm": 6.125,
      "learning_rate": 3.907793826668925e-05,
      "loss": 1.0646,
      "num_input_tokens_seen": 29500200,
      "step": 50845
    },
    {
      "epoch": 7.57372654155496,
      "grad_norm": 16.25,
      "learning_rate": 3.907525290121552e-05,
      "loss": 1.3117,
      "num_input_tokens_seen": 29503208,
      "step": 50850
    },
    {
      "epoch": 7.5744712540959185,
      "grad_norm": 7.9375,
      "learning_rate": 3.9072567297954694e-05,
      "loss": 0.9291,
      "num_input_tokens_seen": 29505992,
      "step": 50855
    },
    {
      "epoch": 7.575215966636878,
      "grad_norm": 11.5,
      "learning_rate": 3.906988145695215e-05,
      "loss": 1.0701,
      "num_input_tokens_seen": 29508744,
      "step": 50860
    },
    {
      "epoch": 7.575960679177838,
      "grad_norm": 8.75,
      "learning_rate": 3.906719537825325e-05,
      "loss": 1.2778,
      "num_input_tokens_seen": 29511624,
      "step": 50865
    },
    {
      "epoch": 7.5767053917187965,
      "grad_norm": 12.6875,
      "learning_rate": 3.906450906190339e-05,
      "loss": 1.1786,
      "num_input_tokens_seen": 29514664,
      "step": 50870
    },
    {
      "epoch": 7.577450104259755,
      "grad_norm": 9.1875,
      "learning_rate": 3.9061822507947945e-05,
      "loss": 1.0791,
      "num_input_tokens_seen": 29517288,
      "step": 50875
    },
    {
      "epoch": 7.578194816800715,
      "grad_norm": 10.25,
      "learning_rate": 3.9059135716432294e-05,
      "loss": 1.1379,
      "num_input_tokens_seen": 29520328,
      "step": 50880
    },
    {
      "epoch": 7.578939529341675,
      "grad_norm": 15.6875,
      "learning_rate": 3.905644868740184e-05,
      "loss": 1.1349,
      "num_input_tokens_seen": 29523048,
      "step": 50885
    },
    {
      "epoch": 7.579684241882633,
      "grad_norm": 13.875,
      "learning_rate": 3.905376142090197e-05,
      "loss": 0.9832,
      "num_input_tokens_seen": 29525832,
      "step": 50890
    },
    {
      "epoch": 7.580428954423592,
      "grad_norm": 12.375,
      "learning_rate": 3.9051073916978084e-05,
      "loss": 1.1458,
      "num_input_tokens_seen": 29528520,
      "step": 50895
    },
    {
      "epoch": 7.581173666964552,
      "grad_norm": 21.0,
      "learning_rate": 3.904838617567558e-05,
      "loss": 1.0956,
      "num_input_tokens_seen": 29531336,
      "step": 50900
    },
    {
      "epoch": 7.5819183795055105,
      "grad_norm": 8.875,
      "learning_rate": 3.904569819703988e-05,
      "loss": 0.9865,
      "num_input_tokens_seen": 29534216,
      "step": 50905
    },
    {
      "epoch": 7.58266309204647,
      "grad_norm": 9.8125,
      "learning_rate": 3.9043009981116376e-05,
      "loss": 0.6911,
      "num_input_tokens_seen": 29537448,
      "step": 50910
    },
    {
      "epoch": 7.583407804587429,
      "grad_norm": 7.625,
      "learning_rate": 3.9040321527950497e-05,
      "loss": 1.1159,
      "num_input_tokens_seen": 29540360,
      "step": 50915
    },
    {
      "epoch": 7.584152517128389,
      "grad_norm": 8.0,
      "learning_rate": 3.903763283758765e-05,
      "loss": 0.8234,
      "num_input_tokens_seen": 29543336,
      "step": 50920
    },
    {
      "epoch": 7.584897229669347,
      "grad_norm": 5.03125,
      "learning_rate": 3.903494391007327e-05,
      "loss": 0.8267,
      "num_input_tokens_seen": 29546280,
      "step": 50925
    },
    {
      "epoch": 7.585641942210307,
      "grad_norm": 8.0625,
      "learning_rate": 3.9032254745452775e-05,
      "loss": 0.8412,
      "num_input_tokens_seen": 29549160,
      "step": 50930
    },
    {
      "epoch": 7.586386654751266,
      "grad_norm": 15.0,
      "learning_rate": 3.902956534377159e-05,
      "loss": 1.1171,
      "num_input_tokens_seen": 29552168,
      "step": 50935
    },
    {
      "epoch": 7.587131367292225,
      "grad_norm": 12.3125,
      "learning_rate": 3.902687570507517e-05,
      "loss": 0.8959,
      "num_input_tokens_seen": 29555336,
      "step": 50940
    },
    {
      "epoch": 7.587876079833184,
      "grad_norm": 8.0,
      "learning_rate": 3.902418582940893e-05,
      "loss": 0.9316,
      "num_input_tokens_seen": 29558216,
      "step": 50945
    },
    {
      "epoch": 7.588620792374144,
      "grad_norm": 4.875,
      "learning_rate": 3.902149571681833e-05,
      "loss": 1.0217,
      "num_input_tokens_seen": 29561192,
      "step": 50950
    },
    {
      "epoch": 7.5893655049151025,
      "grad_norm": 14.25,
      "learning_rate": 3.901880536734881e-05,
      "loss": 1.3763,
      "num_input_tokens_seen": 29564104,
      "step": 50955
    },
    {
      "epoch": 7.590110217456062,
      "grad_norm": 9.75,
      "learning_rate": 3.901611478104582e-05,
      "loss": 0.9015,
      "num_input_tokens_seen": 29566568,
      "step": 50960
    },
    {
      "epoch": 7.590854929997021,
      "grad_norm": 9.375,
      "learning_rate": 3.901342395795482e-05,
      "loss": 1.0114,
      "num_input_tokens_seen": 29569608,
      "step": 50965
    },
    {
      "epoch": 7.591599642537981,
      "grad_norm": 14.125,
      "learning_rate": 3.901073289812126e-05,
      "loss": 1.0773,
      "num_input_tokens_seen": 29572488,
      "step": 50970
    },
    {
      "epoch": 7.592344355078939,
      "grad_norm": 14.5,
      "learning_rate": 3.900804160159061e-05,
      "loss": 1.068,
      "num_input_tokens_seen": 29575464,
      "step": 50975
    },
    {
      "epoch": 7.593089067619899,
      "grad_norm": 6.0,
      "learning_rate": 3.900535006840833e-05,
      "loss": 0.842,
      "num_input_tokens_seen": 29578344,
      "step": 50980
    },
    {
      "epoch": 7.593833780160858,
      "grad_norm": 15.5625,
      "learning_rate": 3.90026582986199e-05,
      "loss": 0.9358,
      "num_input_tokens_seen": 29581320,
      "step": 50985
    },
    {
      "epoch": 7.594578492701817,
      "grad_norm": 11.4375,
      "learning_rate": 3.899996629227079e-05,
      "loss": 1.1264,
      "num_input_tokens_seen": 29584200,
      "step": 50990
    },
    {
      "epoch": 7.595323205242776,
      "grad_norm": 4.8125,
      "learning_rate": 3.899727404940647e-05,
      "loss": 1.0977,
      "num_input_tokens_seen": 29587144,
      "step": 50995
    },
    {
      "epoch": 7.596067917783736,
      "grad_norm": 8.0,
      "learning_rate": 3.899458157007244e-05,
      "loss": 0.9718,
      "num_input_tokens_seen": 29589704,
      "step": 51000
    },
    {
      "epoch": 7.596812630324695,
      "grad_norm": 16.5,
      "learning_rate": 3.899188885431419e-05,
      "loss": 1.0311,
      "num_input_tokens_seen": 29592808,
      "step": 51005
    },
    {
      "epoch": 7.597557342865654,
      "grad_norm": 11.5,
      "learning_rate": 3.898919590217718e-05,
      "loss": 1.2821,
      "num_input_tokens_seen": 29595784,
      "step": 51010
    },
    {
      "epoch": 7.598302055406613,
      "grad_norm": 8.9375,
      "learning_rate": 3.898650271370692e-05,
      "loss": 1.1119,
      "num_input_tokens_seen": 29598664,
      "step": 51015
    },
    {
      "epoch": 7.599046767947573,
      "grad_norm": 10.5625,
      "learning_rate": 3.898380928894892e-05,
      "loss": 1.3153,
      "num_input_tokens_seen": 29601480,
      "step": 51020
    },
    {
      "epoch": 7.599791480488531,
      "grad_norm": 13.6875,
      "learning_rate": 3.8981115627948675e-05,
      "loss": 1.2262,
      "num_input_tokens_seen": 29604360,
      "step": 51025
    },
    {
      "epoch": 7.600536193029491,
      "grad_norm": 9.625,
      "learning_rate": 3.897842173075169e-05,
      "loss": 0.775,
      "num_input_tokens_seen": 29607592,
      "step": 51030
    },
    {
      "epoch": 7.60128090557045,
      "grad_norm": 9.5625,
      "learning_rate": 3.8975727597403475e-05,
      "loss": 0.6081,
      "num_input_tokens_seen": 29610344,
      "step": 51035
    },
    {
      "epoch": 7.6020256181114085,
      "grad_norm": 10.625,
      "learning_rate": 3.8973033227949554e-05,
      "loss": 0.9422,
      "num_input_tokens_seen": 29613064,
      "step": 51040
    },
    {
      "epoch": 7.602770330652368,
      "grad_norm": 12.125,
      "learning_rate": 3.897033862243543e-05,
      "loss": 1.4846,
      "num_input_tokens_seen": 29616072,
      "step": 51045
    },
    {
      "epoch": 7.603515043193328,
      "grad_norm": 10.0625,
      "learning_rate": 3.896764378090664e-05,
      "loss": 0.8299,
      "num_input_tokens_seen": 29618792,
      "step": 51050
    },
    {
      "epoch": 7.604259755734287,
      "grad_norm": 11.0,
      "learning_rate": 3.89649487034087e-05,
      "loss": 1.07,
      "num_input_tokens_seen": 29621768,
      "step": 51055
    },
    {
      "epoch": 7.605004468275245,
      "grad_norm": 16.5,
      "learning_rate": 3.8962253389987145e-05,
      "loss": 1.2992,
      "num_input_tokens_seen": 29624776,
      "step": 51060
    },
    {
      "epoch": 7.605749180816205,
      "grad_norm": 5.53125,
      "learning_rate": 3.895955784068751e-05,
      "loss": 0.7468,
      "num_input_tokens_seen": 29627688,
      "step": 51065
    },
    {
      "epoch": 7.606493893357164,
      "grad_norm": 11.875,
      "learning_rate": 3.8956862055555335e-05,
      "loss": 1.275,
      "num_input_tokens_seen": 29630472,
      "step": 51070
    },
    {
      "epoch": 7.607238605898123,
      "grad_norm": 6.0625,
      "learning_rate": 3.895416603463616e-05,
      "loss": 0.9636,
      "num_input_tokens_seen": 29633832,
      "step": 51075
    },
    {
      "epoch": 7.607983318439082,
      "grad_norm": 10.75,
      "learning_rate": 3.895146977797553e-05,
      "loss": 1.3172,
      "num_input_tokens_seen": 29636840,
      "step": 51080
    },
    {
      "epoch": 7.608728030980042,
      "grad_norm": 10.375,
      "learning_rate": 3.8948773285619e-05,
      "loss": 1.0737,
      "num_input_tokens_seen": 29639688,
      "step": 51085
    },
    {
      "epoch": 7.609472743521001,
      "grad_norm": 10.8125,
      "learning_rate": 3.894607655761212e-05,
      "loss": 0.9256,
      "num_input_tokens_seen": 29642536,
      "step": 51090
    },
    {
      "epoch": 7.61021745606196,
      "grad_norm": 16.5,
      "learning_rate": 3.894337959400045e-05,
      "loss": 1.2537,
      "num_input_tokens_seen": 29645352,
      "step": 51095
    },
    {
      "epoch": 7.610962168602919,
      "grad_norm": 7.75,
      "learning_rate": 3.894068239482956e-05,
      "loss": 0.8079,
      "num_input_tokens_seen": 29648200,
      "step": 51100
    },
    {
      "epoch": 7.611706881143879,
      "grad_norm": 16.25,
      "learning_rate": 3.8937984960145004e-05,
      "loss": 1.1109,
      "num_input_tokens_seen": 29650952,
      "step": 51105
    },
    {
      "epoch": 7.612451593684837,
      "grad_norm": 5.40625,
      "learning_rate": 3.893528728999236e-05,
      "loss": 0.9387,
      "num_input_tokens_seen": 29653736,
      "step": 51110
    },
    {
      "epoch": 7.613196306225797,
      "grad_norm": 16.125,
      "learning_rate": 3.893258938441719e-05,
      "loss": 1.1276,
      "num_input_tokens_seen": 29656744,
      "step": 51115
    },
    {
      "epoch": 7.613941018766756,
      "grad_norm": 12.75,
      "learning_rate": 3.89298912434651e-05,
      "loss": 0.8993,
      "num_input_tokens_seen": 29659784,
      "step": 51120
    },
    {
      "epoch": 7.614685731307715,
      "grad_norm": 13.6875,
      "learning_rate": 3.892719286718165e-05,
      "loss": 0.9437,
      "num_input_tokens_seen": 29662696,
      "step": 51125
    },
    {
      "epoch": 7.615430443848674,
      "grad_norm": 6.65625,
      "learning_rate": 3.892449425561243e-05,
      "loss": 1.0135,
      "num_input_tokens_seen": 29665352,
      "step": 51130
    },
    {
      "epoch": 7.616175156389634,
      "grad_norm": 9.875,
      "learning_rate": 3.892179540880303e-05,
      "loss": 1.1866,
      "num_input_tokens_seen": 29667912,
      "step": 51135
    },
    {
      "epoch": 7.616919868930593,
      "grad_norm": 11.75,
      "learning_rate": 3.891909632679904e-05,
      "loss": 1.2118,
      "num_input_tokens_seen": 29670920,
      "step": 51140
    },
    {
      "epoch": 7.617664581471552,
      "grad_norm": 10.75,
      "learning_rate": 3.8916397009646076e-05,
      "loss": 1.0455,
      "num_input_tokens_seen": 29673896,
      "step": 51145
    },
    {
      "epoch": 7.618409294012511,
      "grad_norm": 6.90625,
      "learning_rate": 3.891369745738972e-05,
      "loss": 1.0304,
      "num_input_tokens_seen": 29676872,
      "step": 51150
    },
    {
      "epoch": 7.619154006553471,
      "grad_norm": 8.9375,
      "learning_rate": 3.8910997670075593e-05,
      "loss": 1.0525,
      "num_input_tokens_seen": 29679592,
      "step": 51155
    },
    {
      "epoch": 7.619898719094429,
      "grad_norm": 17.875,
      "learning_rate": 3.890829764774929e-05,
      "loss": 1.1774,
      "num_input_tokens_seen": 29682376,
      "step": 51160
    },
    {
      "epoch": 7.620643431635389,
      "grad_norm": 9.8125,
      "learning_rate": 3.8905597390456446e-05,
      "loss": 0.8807,
      "num_input_tokens_seen": 29685032,
      "step": 51165
    },
    {
      "epoch": 7.621388144176348,
      "grad_norm": 7.84375,
      "learning_rate": 3.890289689824266e-05,
      "loss": 0.9058,
      "num_input_tokens_seen": 29688008,
      "step": 51170
    },
    {
      "epoch": 7.6221328567173074,
      "grad_norm": 6.3125,
      "learning_rate": 3.890019617115357e-05,
      "loss": 1.1248,
      "num_input_tokens_seen": 29691112,
      "step": 51175
    },
    {
      "epoch": 7.622877569258266,
      "grad_norm": 19.5,
      "learning_rate": 3.889749520923478e-05,
      "loss": 0.8441,
      "num_input_tokens_seen": 29694120,
      "step": 51180
    },
    {
      "epoch": 7.623622281799226,
      "grad_norm": 9.0625,
      "learning_rate": 3.889479401253194e-05,
      "loss": 1.133,
      "num_input_tokens_seen": 29696872,
      "step": 51185
    },
    {
      "epoch": 7.624366994340185,
      "grad_norm": 6.875,
      "learning_rate": 3.8892092581090675e-05,
      "loss": 1.0197,
      "num_input_tokens_seen": 29699624,
      "step": 51190
    },
    {
      "epoch": 7.625111706881144,
      "grad_norm": 10.75,
      "learning_rate": 3.888939091495663e-05,
      "loss": 1.4753,
      "num_input_tokens_seen": 29702472,
      "step": 51195
    },
    {
      "epoch": 7.625856419422103,
      "grad_norm": 5.375,
      "learning_rate": 3.888668901417544e-05,
      "loss": 1.0908,
      "num_input_tokens_seen": 29705384,
      "step": 51200
    },
    {
      "epoch": 7.626601131963062,
      "grad_norm": 10.3125,
      "learning_rate": 3.888398687879274e-05,
      "loss": 1.5573,
      "num_input_tokens_seen": 29708296,
      "step": 51205
    },
    {
      "epoch": 7.627345844504021,
      "grad_norm": 11.6875,
      "learning_rate": 3.888128450885421e-05,
      "loss": 1.0414,
      "num_input_tokens_seen": 29711272,
      "step": 51210
    },
    {
      "epoch": 7.628090557044981,
      "grad_norm": 15.0,
      "learning_rate": 3.887858190440549e-05,
      "loss": 1.2785,
      "num_input_tokens_seen": 29714152,
      "step": 51215
    },
    {
      "epoch": 7.62883526958594,
      "grad_norm": 14.9375,
      "learning_rate": 3.8875879065492216e-05,
      "loss": 1.0529,
      "num_input_tokens_seen": 29716840,
      "step": 51220
    },
    {
      "epoch": 7.629579982126899,
      "grad_norm": 5.9375,
      "learning_rate": 3.887317599216008e-05,
      "loss": 1.356,
      "num_input_tokens_seen": 29720008,
      "step": 51225
    },
    {
      "epoch": 7.630324694667858,
      "grad_norm": 18.125,
      "learning_rate": 3.887047268445473e-05,
      "loss": 1.0385,
      "num_input_tokens_seen": 29722952,
      "step": 51230
    },
    {
      "epoch": 7.631069407208818,
      "grad_norm": 11.25,
      "learning_rate": 3.8867769142421844e-05,
      "loss": 1.265,
      "num_input_tokens_seen": 29725960,
      "step": 51235
    },
    {
      "epoch": 7.631814119749777,
      "grad_norm": 12.125,
      "learning_rate": 3.886506536610709e-05,
      "loss": 1.085,
      "num_input_tokens_seen": 29728904,
      "step": 51240
    },
    {
      "epoch": 7.632558832290735,
      "grad_norm": 10.375,
      "learning_rate": 3.8862361355556156e-05,
      "loss": 1.1267,
      "num_input_tokens_seen": 29731752,
      "step": 51245
    },
    {
      "epoch": 7.633303544831695,
      "grad_norm": 14.0,
      "learning_rate": 3.8859657110814704e-05,
      "loss": 1.2877,
      "num_input_tokens_seen": 29734664,
      "step": 51250
    },
    {
      "epoch": 7.634048257372654,
      "grad_norm": 11.875,
      "learning_rate": 3.885695263192844e-05,
      "loss": 1.1414,
      "num_input_tokens_seen": 29737672,
      "step": 51255
    },
    {
      "epoch": 7.6347929699136134,
      "grad_norm": 13.125,
      "learning_rate": 3.885424791894305e-05,
      "loss": 0.8261,
      "num_input_tokens_seen": 29740456,
      "step": 51260
    },
    {
      "epoch": 7.635537682454572,
      "grad_norm": 7.03125,
      "learning_rate": 3.885154297190421e-05,
      "loss": 0.9639,
      "num_input_tokens_seen": 29743560,
      "step": 51265
    },
    {
      "epoch": 7.636282394995532,
      "grad_norm": 13.1875,
      "learning_rate": 3.884883779085764e-05,
      "loss": 1.2267,
      "num_input_tokens_seen": 29746152,
      "step": 51270
    },
    {
      "epoch": 7.637027107536491,
      "grad_norm": 7.0,
      "learning_rate": 3.884613237584902e-05,
      "loss": 1.1019,
      "num_input_tokens_seen": 29749192,
      "step": 51275
    },
    {
      "epoch": 7.63777182007745,
      "grad_norm": 15.0625,
      "learning_rate": 3.884342672692407e-05,
      "loss": 1.1069,
      "num_input_tokens_seen": 29752296,
      "step": 51280
    },
    {
      "epoch": 7.638516532618409,
      "grad_norm": 14.8125,
      "learning_rate": 3.88407208441285e-05,
      "loss": 1.1203,
      "num_input_tokens_seen": 29755432,
      "step": 51285
    },
    {
      "epoch": 7.639261245159369,
      "grad_norm": 12.25,
      "learning_rate": 3.883801472750802e-05,
      "loss": 1.5555,
      "num_input_tokens_seen": 29758216,
      "step": 51290
    },
    {
      "epoch": 7.640005957700327,
      "grad_norm": 11.1875,
      "learning_rate": 3.8835308377108344e-05,
      "loss": 0.9235,
      "num_input_tokens_seen": 29760872,
      "step": 51295
    },
    {
      "epoch": 7.640750670241287,
      "grad_norm": 15.6875,
      "learning_rate": 3.883260179297519e-05,
      "loss": 1.0161,
      "num_input_tokens_seen": 29763816,
      "step": 51300
    },
    {
      "epoch": 7.641495382782246,
      "grad_norm": 9.0,
      "learning_rate": 3.882989497515429e-05,
      "loss": 0.8147,
      "num_input_tokens_seen": 29766984,
      "step": 51305
    },
    {
      "epoch": 7.6422400953232055,
      "grad_norm": 25.25,
      "learning_rate": 3.8827187923691365e-05,
      "loss": 1.1194,
      "num_input_tokens_seen": 29769896,
      "step": 51310
    },
    {
      "epoch": 7.642984807864164,
      "grad_norm": 9.5,
      "learning_rate": 3.882448063863216e-05,
      "loss": 0.8583,
      "num_input_tokens_seen": 29772808,
      "step": 51315
    },
    {
      "epoch": 7.643729520405124,
      "grad_norm": 3.59375,
      "learning_rate": 3.882177312002241e-05,
      "loss": 0.7503,
      "num_input_tokens_seen": 29775688,
      "step": 51320
    },
    {
      "epoch": 7.644474232946083,
      "grad_norm": 12.125,
      "learning_rate": 3.881906536790784e-05,
      "loss": 1.5504,
      "num_input_tokens_seen": 29778600,
      "step": 51325
    },
    {
      "epoch": 7.645218945487042,
      "grad_norm": 15.25,
      "learning_rate": 3.881635738233421e-05,
      "loss": 1.561,
      "num_input_tokens_seen": 29781512,
      "step": 51330
    },
    {
      "epoch": 7.645963658028001,
      "grad_norm": 11.625,
      "learning_rate": 3.8813649163347266e-05,
      "loss": 1.076,
      "num_input_tokens_seen": 29784104,
      "step": 51335
    },
    {
      "epoch": 7.646708370568961,
      "grad_norm": 9.5,
      "learning_rate": 3.881094071099276e-05,
      "loss": 1.0017,
      "num_input_tokens_seen": 29786824,
      "step": 51340
    },
    {
      "epoch": 7.6474530831099194,
      "grad_norm": 19.5,
      "learning_rate": 3.880823202531644e-05,
      "loss": 1.0312,
      "num_input_tokens_seen": 29789512,
      "step": 51345
    },
    {
      "epoch": 7.648197795650879,
      "grad_norm": 9.1875,
      "learning_rate": 3.880552310636408e-05,
      "loss": 0.9378,
      "num_input_tokens_seen": 29792456,
      "step": 51350
    },
    {
      "epoch": 7.648942508191838,
      "grad_norm": 6.875,
      "learning_rate": 3.880281395418144e-05,
      "loss": 1.0383,
      "num_input_tokens_seen": 29795272,
      "step": 51355
    },
    {
      "epoch": 7.6496872207327975,
      "grad_norm": 10.3125,
      "learning_rate": 3.8800104568814275e-05,
      "loss": 1.1261,
      "num_input_tokens_seen": 29797896,
      "step": 51360
    },
    {
      "epoch": 7.650431933273756,
      "grad_norm": 19.5,
      "learning_rate": 3.879739495030839e-05,
      "loss": 1.3951,
      "num_input_tokens_seen": 29800968,
      "step": 51365
    },
    {
      "epoch": 7.651176645814716,
      "grad_norm": 14.8125,
      "learning_rate": 3.879468509870953e-05,
      "loss": 1.2393,
      "num_input_tokens_seen": 29803816,
      "step": 51370
    },
    {
      "epoch": 7.651921358355675,
      "grad_norm": 7.65625,
      "learning_rate": 3.879197501406347e-05,
      "loss": 0.9475,
      "num_input_tokens_seen": 29806728,
      "step": 51375
    },
    {
      "epoch": 7.652666070896634,
      "grad_norm": 10.625,
      "learning_rate": 3.878926469641603e-05,
      "loss": 1.1191,
      "num_input_tokens_seen": 29809448,
      "step": 51380
    },
    {
      "epoch": 7.653410783437593,
      "grad_norm": 14.9375,
      "learning_rate": 3.878655414581297e-05,
      "loss": 1.0787,
      "num_input_tokens_seen": 29812136,
      "step": 51385
    },
    {
      "epoch": 7.654155495978552,
      "grad_norm": 13.3125,
      "learning_rate": 3.878384336230009e-05,
      "loss": 1.279,
      "num_input_tokens_seen": 29814984,
      "step": 51390
    },
    {
      "epoch": 7.6549002085195115,
      "grad_norm": 7.40625,
      "learning_rate": 3.878113234592319e-05,
      "loss": 1.2022,
      "num_input_tokens_seen": 29817992,
      "step": 51395
    },
    {
      "epoch": 7.655644921060471,
      "grad_norm": 5.59375,
      "learning_rate": 3.8778421096728065e-05,
      "loss": 1.1188,
      "num_input_tokens_seen": 29820616,
      "step": 51400
    },
    {
      "epoch": 7.65638963360143,
      "grad_norm": 13.5,
      "learning_rate": 3.8775709614760514e-05,
      "loss": 1.0236,
      "num_input_tokens_seen": 29823368,
      "step": 51405
    },
    {
      "epoch": 7.657134346142389,
      "grad_norm": 9.6875,
      "learning_rate": 3.877299790006635e-05,
      "loss": 1.3482,
      "num_input_tokens_seen": 29826152,
      "step": 51410
    },
    {
      "epoch": 7.657879058683348,
      "grad_norm": 6.09375,
      "learning_rate": 3.877028595269139e-05,
      "loss": 1.4389,
      "num_input_tokens_seen": 29829704,
      "step": 51415
    },
    {
      "epoch": 7.658623771224307,
      "grad_norm": 7.53125,
      "learning_rate": 3.876757377268144e-05,
      "loss": 0.9867,
      "num_input_tokens_seen": 29832520,
      "step": 51420
    },
    {
      "epoch": 7.659368483765267,
      "grad_norm": 8.9375,
      "learning_rate": 3.8764861360082324e-05,
      "loss": 1.279,
      "num_input_tokens_seen": 29835176,
      "step": 51425
    },
    {
      "epoch": 7.6601131963062254,
      "grad_norm": 11.4375,
      "learning_rate": 3.876214871493987e-05,
      "loss": 1.2204,
      "num_input_tokens_seen": 29837768,
      "step": 51430
    },
    {
      "epoch": 7.660857908847185,
      "grad_norm": 3.171875,
      "learning_rate": 3.8759435837299904e-05,
      "loss": 1.059,
      "num_input_tokens_seen": 29840840,
      "step": 51435
    },
    {
      "epoch": 7.661602621388144,
      "grad_norm": 9.5,
      "learning_rate": 3.8756722727208246e-05,
      "loss": 0.9552,
      "num_input_tokens_seen": 29843656,
      "step": 51440
    },
    {
      "epoch": 7.6623473339291035,
      "grad_norm": 13.875,
      "learning_rate": 3.8754009384710736e-05,
      "loss": 0.9111,
      "num_input_tokens_seen": 29846440,
      "step": 51445
    },
    {
      "epoch": 7.663092046470062,
      "grad_norm": 13.125,
      "learning_rate": 3.8751295809853225e-05,
      "loss": 0.9346,
      "num_input_tokens_seen": 29849480,
      "step": 51450
    },
    {
      "epoch": 7.663836759011022,
      "grad_norm": 6.09375,
      "learning_rate": 3.8748582002681545e-05,
      "loss": 0.999,
      "num_input_tokens_seen": 29852264,
      "step": 51455
    },
    {
      "epoch": 7.664581471551981,
      "grad_norm": 6.375,
      "learning_rate": 3.8745867963241545e-05,
      "loss": 1.0418,
      "num_input_tokens_seen": 29854856,
      "step": 51460
    },
    {
      "epoch": 7.66532618409294,
      "grad_norm": 11.1875,
      "learning_rate": 3.874315369157907e-05,
      "loss": 0.9643,
      "num_input_tokens_seen": 29857704,
      "step": 51465
    },
    {
      "epoch": 7.666070896633899,
      "grad_norm": 7.0,
      "learning_rate": 3.8740439187739993e-05,
      "loss": 1.2378,
      "num_input_tokens_seen": 29860552,
      "step": 51470
    },
    {
      "epoch": 7.666815609174859,
      "grad_norm": 12.375,
      "learning_rate": 3.873772445177015e-05,
      "loss": 1.1088,
      "num_input_tokens_seen": 29863208,
      "step": 51475
    },
    {
      "epoch": 7.6675603217158175,
      "grad_norm": 22.625,
      "learning_rate": 3.873500948371542e-05,
      "loss": 1.2935,
      "num_input_tokens_seen": 29865896,
      "step": 51480
    },
    {
      "epoch": 7.668305034256777,
      "grad_norm": 7.28125,
      "learning_rate": 3.873229428362167e-05,
      "loss": 0.9517,
      "num_input_tokens_seen": 29868840,
      "step": 51485
    },
    {
      "epoch": 7.669049746797736,
      "grad_norm": 5.0625,
      "learning_rate": 3.872957885153476e-05,
      "loss": 0.9755,
      "num_input_tokens_seen": 29871560,
      "step": 51490
    },
    {
      "epoch": 7.6697944593386955,
      "grad_norm": 7.5,
      "learning_rate": 3.8726863187500564e-05,
      "loss": 0.8312,
      "num_input_tokens_seen": 29874408,
      "step": 51495
    },
    {
      "epoch": 7.670539171879654,
      "grad_norm": 5.21875,
      "learning_rate": 3.872414729156497e-05,
      "loss": 1.0988,
      "num_input_tokens_seen": 29877128,
      "step": 51500
    },
    {
      "epoch": 7.671283884420614,
      "grad_norm": 4.28125,
      "learning_rate": 3.872143116377386e-05,
      "loss": 1.3293,
      "num_input_tokens_seen": 29879976,
      "step": 51505
    },
    {
      "epoch": 7.672028596961573,
      "grad_norm": 12.6875,
      "learning_rate": 3.871871480417311e-05,
      "loss": 1.2242,
      "num_input_tokens_seen": 29882792,
      "step": 51510
    },
    {
      "epoch": 7.672773309502532,
      "grad_norm": 5.84375,
      "learning_rate": 3.871599821280863e-05,
      "loss": 0.8878,
      "num_input_tokens_seen": 29886024,
      "step": 51515
    },
    {
      "epoch": 7.673518022043491,
      "grad_norm": 13.0625,
      "learning_rate": 3.8713281389726285e-05,
      "loss": 0.9583,
      "num_input_tokens_seen": 29889160,
      "step": 51520
    },
    {
      "epoch": 7.674262734584451,
      "grad_norm": 6.34375,
      "learning_rate": 3.871056433497199e-05,
      "loss": 1.1955,
      "num_input_tokens_seen": 29892296,
      "step": 51525
    },
    {
      "epoch": 7.6750074471254095,
      "grad_norm": 9.4375,
      "learning_rate": 3.870784704859165e-05,
      "loss": 1.1972,
      "num_input_tokens_seen": 29895400,
      "step": 51530
    },
    {
      "epoch": 7.675752159666369,
      "grad_norm": 10.125,
      "learning_rate": 3.8705129530631165e-05,
      "loss": 1.1547,
      "num_input_tokens_seen": 29898280,
      "step": 51535
    },
    {
      "epoch": 7.676496872207328,
      "grad_norm": 6.625,
      "learning_rate": 3.870241178113645e-05,
      "loss": 1.0765,
      "num_input_tokens_seen": 29901096,
      "step": 51540
    },
    {
      "epoch": 7.6772415847482876,
      "grad_norm": 6.34375,
      "learning_rate": 3.86996938001534e-05,
      "loss": 1.2232,
      "num_input_tokens_seen": 29904104,
      "step": 51545
    },
    {
      "epoch": 7.677986297289246,
      "grad_norm": 8.6875,
      "learning_rate": 3.869697558772796e-05,
      "loss": 1.037,
      "num_input_tokens_seen": 29907048,
      "step": 51550
    },
    {
      "epoch": 7.678731009830205,
      "grad_norm": 13.8125,
      "learning_rate": 3.8694257143906035e-05,
      "loss": 1.8707,
      "num_input_tokens_seen": 29909864,
      "step": 51555
    },
    {
      "epoch": 7.679475722371165,
      "grad_norm": 6.84375,
      "learning_rate": 3.869153846873356e-05,
      "loss": 0.8787,
      "num_input_tokens_seen": 29912776,
      "step": 51560
    },
    {
      "epoch": 7.680220434912124,
      "grad_norm": 5.375,
      "learning_rate": 3.868881956225645e-05,
      "loss": 0.9244,
      "num_input_tokens_seen": 29915752,
      "step": 51565
    },
    {
      "epoch": 7.680965147453083,
      "grad_norm": 10.5,
      "learning_rate": 3.868610042452065e-05,
      "loss": 1.2025,
      "num_input_tokens_seen": 29918504,
      "step": 51570
    },
    {
      "epoch": 7.681709859994042,
      "grad_norm": 10.875,
      "learning_rate": 3.8683381055572095e-05,
      "loss": 1.2606,
      "num_input_tokens_seen": 29921160,
      "step": 51575
    },
    {
      "epoch": 7.6824545725350015,
      "grad_norm": 10.0625,
      "learning_rate": 3.868066145545672e-05,
      "loss": 0.9365,
      "num_input_tokens_seen": 29924392,
      "step": 51580
    },
    {
      "epoch": 7.683199285075961,
      "grad_norm": 7.46875,
      "learning_rate": 3.867794162422047e-05,
      "loss": 0.9996,
      "num_input_tokens_seen": 29927272,
      "step": 51585
    },
    {
      "epoch": 7.68394399761692,
      "grad_norm": 20.375,
      "learning_rate": 3.86752215619093e-05,
      "loss": 1.0347,
      "num_input_tokens_seen": 29930184,
      "step": 51590
    },
    {
      "epoch": 7.684688710157879,
      "grad_norm": 14.3125,
      "learning_rate": 3.867250126856917e-05,
      "loss": 1.2391,
      "num_input_tokens_seen": 29933448,
      "step": 51595
    },
    {
      "epoch": 7.685433422698838,
      "grad_norm": 14.0,
      "learning_rate": 3.866978074424602e-05,
      "loss": 0.9196,
      "num_input_tokens_seen": 29936104,
      "step": 51600
    },
    {
      "epoch": 7.686178135239797,
      "grad_norm": 5.375,
      "learning_rate": 3.866705998898582e-05,
      "loss": 0.9206,
      "num_input_tokens_seen": 29939592,
      "step": 51605
    },
    {
      "epoch": 7.686922847780757,
      "grad_norm": 19.875,
      "learning_rate": 3.866433900283453e-05,
      "loss": 1.23,
      "num_input_tokens_seen": 29942280,
      "step": 51610
    },
    {
      "epoch": 7.6876675603217155,
      "grad_norm": 9.1875,
      "learning_rate": 3.866161778583812e-05,
      "loss": 1.0993,
      "num_input_tokens_seen": 29945320,
      "step": 51615
    },
    {
      "epoch": 7.688412272862675,
      "grad_norm": 18.875,
      "learning_rate": 3.865889633804257e-05,
      "loss": 1.6152,
      "num_input_tokens_seen": 29948232,
      "step": 51620
    },
    {
      "epoch": 7.689156985403634,
      "grad_norm": 12.375,
      "learning_rate": 3.8656174659493835e-05,
      "loss": 0.9318,
      "num_input_tokens_seen": 29951144,
      "step": 51625
    },
    {
      "epoch": 7.6899016979445936,
      "grad_norm": 9.625,
      "learning_rate": 3.865345275023792e-05,
      "loss": 1.0748,
      "num_input_tokens_seen": 29953896,
      "step": 51630
    },
    {
      "epoch": 7.690646410485552,
      "grad_norm": 14.8125,
      "learning_rate": 3.8650730610320796e-05,
      "loss": 1.262,
      "num_input_tokens_seen": 29956808,
      "step": 51635
    },
    {
      "epoch": 7.691391123026512,
      "grad_norm": 10.625,
      "learning_rate": 3.864800823978845e-05,
      "loss": 1.1448,
      "num_input_tokens_seen": 29959688,
      "step": 51640
    },
    {
      "epoch": 7.692135835567471,
      "grad_norm": 16.5,
      "learning_rate": 3.864528563868687e-05,
      "loss": 1.0952,
      "num_input_tokens_seen": 29962504,
      "step": 51645
    },
    {
      "epoch": 7.69288054810843,
      "grad_norm": 10.4375,
      "learning_rate": 3.864256280706206e-05,
      "loss": 1.1052,
      "num_input_tokens_seen": 29965064,
      "step": 51650
    },
    {
      "epoch": 7.693625260649389,
      "grad_norm": 13.25,
      "learning_rate": 3.8639839744960025e-05,
      "loss": 1.0729,
      "num_input_tokens_seen": 29967848,
      "step": 51655
    },
    {
      "epoch": 7.694369973190349,
      "grad_norm": 7.09375,
      "learning_rate": 3.863711645242676e-05,
      "loss": 1.1163,
      "num_input_tokens_seen": 29970696,
      "step": 51660
    },
    {
      "epoch": 7.6951146857313075,
      "grad_norm": 9.625,
      "learning_rate": 3.863439292950827e-05,
      "loss": 0.9743,
      "num_input_tokens_seen": 29973960,
      "step": 51665
    },
    {
      "epoch": 7.695859398272267,
      "grad_norm": 16.125,
      "learning_rate": 3.863166917625056e-05,
      "loss": 0.9345,
      "num_input_tokens_seen": 29976744,
      "step": 51670
    },
    {
      "epoch": 7.696604110813226,
      "grad_norm": 10.1875,
      "learning_rate": 3.862894519269966e-05,
      "loss": 0.8355,
      "num_input_tokens_seen": 29979528,
      "step": 51675
    },
    {
      "epoch": 7.697348823354186,
      "grad_norm": 7.0625,
      "learning_rate": 3.8626220978901585e-05,
      "loss": 0.9385,
      "num_input_tokens_seen": 29982504,
      "step": 51680
    },
    {
      "epoch": 7.698093535895144,
      "grad_norm": 17.625,
      "learning_rate": 3.862349653490236e-05,
      "loss": 0.8758,
      "num_input_tokens_seen": 29985352,
      "step": 51685
    },
    {
      "epoch": 7.698838248436104,
      "grad_norm": 5.8125,
      "learning_rate": 3.8620771860748005e-05,
      "loss": 1.0975,
      "num_input_tokens_seen": 29988648,
      "step": 51690
    },
    {
      "epoch": 7.699582960977063,
      "grad_norm": 21.75,
      "learning_rate": 3.861804695648455e-05,
      "loss": 0.943,
      "num_input_tokens_seen": 29991624,
      "step": 51695
    },
    {
      "epoch": 7.700327673518022,
      "grad_norm": 8.0625,
      "learning_rate": 3.861532182215802e-05,
      "loss": 0.8741,
      "num_input_tokens_seen": 29994568,
      "step": 51700
    },
    {
      "epoch": 7.701072386058981,
      "grad_norm": 8.1875,
      "learning_rate": 3.861259645781449e-05,
      "loss": 1.0597,
      "num_input_tokens_seen": 29997640,
      "step": 51705
    },
    {
      "epoch": 7.701817098599941,
      "grad_norm": 14.6875,
      "learning_rate": 3.860987086349996e-05,
      "loss": 1.1215,
      "num_input_tokens_seen": 30000776,
      "step": 51710
    },
    {
      "epoch": 7.7025618111408996,
      "grad_norm": 8.9375,
      "learning_rate": 3.86071450392605e-05,
      "loss": 0.9076,
      "num_input_tokens_seen": 30003400,
      "step": 51715
    },
    {
      "epoch": 7.703306523681858,
      "grad_norm": 11.9375,
      "learning_rate": 3.860441898514215e-05,
      "loss": 1.109,
      "num_input_tokens_seen": 30006408,
      "step": 51720
    },
    {
      "epoch": 7.704051236222818,
      "grad_norm": 8.3125,
      "learning_rate": 3.8601692701190975e-05,
      "loss": 0.8718,
      "num_input_tokens_seen": 30009192,
      "step": 51725
    },
    {
      "epoch": 7.704795948763778,
      "grad_norm": 18.0,
      "learning_rate": 3.8598966187453034e-05,
      "loss": 1.3721,
      "num_input_tokens_seen": 30012328,
      "step": 51730
    },
    {
      "epoch": 7.705540661304736,
      "grad_norm": 6.03125,
      "learning_rate": 3.859623944397437e-05,
      "loss": 1.1231,
      "num_input_tokens_seen": 30015176,
      "step": 51735
    },
    {
      "epoch": 7.706285373845695,
      "grad_norm": 15.375,
      "learning_rate": 3.859351247080106e-05,
      "loss": 0.9755,
      "num_input_tokens_seen": 30017928,
      "step": 51740
    },
    {
      "epoch": 7.707030086386655,
      "grad_norm": 9.625,
      "learning_rate": 3.859078526797917e-05,
      "loss": 1.0609,
      "num_input_tokens_seen": 30020744,
      "step": 51745
    },
    {
      "epoch": 7.707774798927614,
      "grad_norm": 18.125,
      "learning_rate": 3.8588057835554776e-05,
      "loss": 1.0282,
      "num_input_tokens_seen": 30023688,
      "step": 51750
    },
    {
      "epoch": 7.708519511468573,
      "grad_norm": 12.75,
      "learning_rate": 3.858533017357396e-05,
      "loss": 1.0374,
      "num_input_tokens_seen": 30026472,
      "step": 51755
    },
    {
      "epoch": 7.709264224009532,
      "grad_norm": 15.5625,
      "learning_rate": 3.858260228208279e-05,
      "loss": 1.0785,
      "num_input_tokens_seen": 30029000,
      "step": 51760
    },
    {
      "epoch": 7.710008936550492,
      "grad_norm": 12.5625,
      "learning_rate": 3.857987416112737e-05,
      "loss": 1.4157,
      "num_input_tokens_seen": 30031656,
      "step": 51765
    },
    {
      "epoch": 7.71075364909145,
      "grad_norm": 5.5,
      "learning_rate": 3.857714581075377e-05,
      "loss": 0.9298,
      "num_input_tokens_seen": 30034568,
      "step": 51770
    },
    {
      "epoch": 7.71149836163241,
      "grad_norm": 6.5,
      "learning_rate": 3.85744172310081e-05,
      "loss": 1.2044,
      "num_input_tokens_seen": 30037064,
      "step": 51775
    },
    {
      "epoch": 7.712243074173369,
      "grad_norm": 6.09375,
      "learning_rate": 3.8571688421936434e-05,
      "loss": 1.0143,
      "num_input_tokens_seen": 30039848,
      "step": 51780
    },
    {
      "epoch": 7.712987786714328,
      "grad_norm": 7.28125,
      "learning_rate": 3.85689593835849e-05,
      "loss": 1.1905,
      "num_input_tokens_seen": 30042728,
      "step": 51785
    },
    {
      "epoch": 7.713732499255287,
      "grad_norm": 10.1875,
      "learning_rate": 3.8566230115999575e-05,
      "loss": 1.1207,
      "num_input_tokens_seen": 30045512,
      "step": 51790
    },
    {
      "epoch": 7.714477211796247,
      "grad_norm": 13.125,
      "learning_rate": 3.856350061922659e-05,
      "loss": 1.2204,
      "num_input_tokens_seen": 30048488,
      "step": 51795
    },
    {
      "epoch": 7.7152219243372056,
      "grad_norm": 9.875,
      "learning_rate": 3.856077089331204e-05,
      "loss": 1.1078,
      "num_input_tokens_seen": 30051432,
      "step": 51800
    },
    {
      "epoch": 7.715966636878165,
      "grad_norm": 12.6875,
      "learning_rate": 3.855804093830205e-05,
      "loss": 1.2909,
      "num_input_tokens_seen": 30054440,
      "step": 51805
    },
    {
      "epoch": 7.716711349419124,
      "grad_norm": 8.4375,
      "learning_rate": 3.855531075424274e-05,
      "loss": 0.8419,
      "num_input_tokens_seen": 30057000,
      "step": 51810
    },
    {
      "epoch": 7.717456061960084,
      "grad_norm": 8.125,
      "learning_rate": 3.8552580341180236e-05,
      "loss": 1.3132,
      "num_input_tokens_seen": 30059976,
      "step": 51815
    },
    {
      "epoch": 7.718200774501042,
      "grad_norm": 12.75,
      "learning_rate": 3.8549849699160655e-05,
      "loss": 1.271,
      "num_input_tokens_seen": 30062856,
      "step": 51820
    },
    {
      "epoch": 7.718945487042002,
      "grad_norm": 10.0,
      "learning_rate": 3.8547118828230135e-05,
      "loss": 1.0367,
      "num_input_tokens_seen": 30065736,
      "step": 51825
    },
    {
      "epoch": 7.719690199582961,
      "grad_norm": 5.6875,
      "learning_rate": 3.854438772843482e-05,
      "loss": 0.7367,
      "num_input_tokens_seen": 30069032,
      "step": 51830
    },
    {
      "epoch": 7.72043491212392,
      "grad_norm": 6.53125,
      "learning_rate": 3.8541656399820825e-05,
      "loss": 0.9752,
      "num_input_tokens_seen": 30071880,
      "step": 51835
    },
    {
      "epoch": 7.721179624664879,
      "grad_norm": 5.6875,
      "learning_rate": 3.853892484243432e-05,
      "loss": 1.1691,
      "num_input_tokens_seen": 30074760,
      "step": 51840
    },
    {
      "epoch": 7.721924337205839,
      "grad_norm": 13.9375,
      "learning_rate": 3.8536193056321436e-05,
      "loss": 1.3837,
      "num_input_tokens_seen": 30077416,
      "step": 51845
    },
    {
      "epoch": 7.722669049746798,
      "grad_norm": 16.625,
      "learning_rate": 3.853346104152833e-05,
      "loss": 1.0303,
      "num_input_tokens_seen": 30080424,
      "step": 51850
    },
    {
      "epoch": 7.723413762287757,
      "grad_norm": 9.0,
      "learning_rate": 3.853072879810115e-05,
      "loss": 1.0182,
      "num_input_tokens_seen": 30083336,
      "step": 51855
    },
    {
      "epoch": 7.724158474828716,
      "grad_norm": 8.375,
      "learning_rate": 3.8527996326086065e-05,
      "loss": 1.2318,
      "num_input_tokens_seen": 30086280,
      "step": 51860
    },
    {
      "epoch": 7.724903187369676,
      "grad_norm": 5.125,
      "learning_rate": 3.852526362552923e-05,
      "loss": 0.8554,
      "num_input_tokens_seen": 30088872,
      "step": 51865
    },
    {
      "epoch": 7.725647899910634,
      "grad_norm": 9.6875,
      "learning_rate": 3.852253069647681e-05,
      "loss": 1.2133,
      "num_input_tokens_seen": 30091688,
      "step": 51870
    },
    {
      "epoch": 7.726392612451594,
      "grad_norm": 15.875,
      "learning_rate": 3.851979753897498e-05,
      "loss": 1.4831,
      "num_input_tokens_seen": 30094728,
      "step": 51875
    },
    {
      "epoch": 7.727137324992553,
      "grad_norm": 16.0,
      "learning_rate": 3.8517064153069905e-05,
      "loss": 0.9488,
      "num_input_tokens_seen": 30097448,
      "step": 51880
    },
    {
      "epoch": 7.727882037533512,
      "grad_norm": 10.25,
      "learning_rate": 3.8514330538807775e-05,
      "loss": 0.9398,
      "num_input_tokens_seen": 30100616,
      "step": 51885
    },
    {
      "epoch": 7.728626750074471,
      "grad_norm": 10.4375,
      "learning_rate": 3.8511596696234765e-05,
      "loss": 1.1009,
      "num_input_tokens_seen": 30103848,
      "step": 51890
    },
    {
      "epoch": 7.729371462615431,
      "grad_norm": 9.9375,
      "learning_rate": 3.8508862625397055e-05,
      "loss": 1.1127,
      "num_input_tokens_seen": 30106664,
      "step": 51895
    },
    {
      "epoch": 7.73011617515639,
      "grad_norm": 17.375,
      "learning_rate": 3.850612832634085e-05,
      "loss": 1.0716,
      "num_input_tokens_seen": 30109544,
      "step": 51900
    },
    {
      "epoch": 7.730860887697348,
      "grad_norm": 11.6875,
      "learning_rate": 3.850339379911233e-05,
      "loss": 1.0168,
      "num_input_tokens_seen": 30112584,
      "step": 51905
    },
    {
      "epoch": 7.731605600238308,
      "grad_norm": 5.40625,
      "learning_rate": 3.8500659043757705e-05,
      "loss": 0.8935,
      "num_input_tokens_seen": 30115176,
      "step": 51910
    },
    {
      "epoch": 7.732350312779268,
      "grad_norm": 12.0625,
      "learning_rate": 3.8497924060323154e-05,
      "loss": 1.1269,
      "num_input_tokens_seen": 30118120,
      "step": 51915
    },
    {
      "epoch": 7.733095025320226,
      "grad_norm": 10.4375,
      "learning_rate": 3.84951888488549e-05,
      "loss": 0.8707,
      "num_input_tokens_seen": 30121224,
      "step": 51920
    },
    {
      "epoch": 7.733839737861185,
      "grad_norm": 9.1875,
      "learning_rate": 3.849245340939914e-05,
      "loss": 1.2097,
      "num_input_tokens_seen": 30124392,
      "step": 51925
    },
    {
      "epoch": 7.734584450402145,
      "grad_norm": 6.28125,
      "learning_rate": 3.84897177420021e-05,
      "loss": 1.0034,
      "num_input_tokens_seen": 30127624,
      "step": 51930
    },
    {
      "epoch": 7.735329162943104,
      "grad_norm": 5.9375,
      "learning_rate": 3.848698184670999e-05,
      "loss": 1.1982,
      "num_input_tokens_seen": 30130920,
      "step": 51935
    },
    {
      "epoch": 7.736073875484063,
      "grad_norm": 9.25,
      "learning_rate": 3.848424572356902e-05,
      "loss": 0.998,
      "num_input_tokens_seen": 30134024,
      "step": 51940
    },
    {
      "epoch": 7.736818588025022,
      "grad_norm": 6.65625,
      "learning_rate": 3.848150937262544e-05,
      "loss": 0.9651,
      "num_input_tokens_seen": 30137160,
      "step": 51945
    },
    {
      "epoch": 7.737563300565982,
      "grad_norm": 8.5,
      "learning_rate": 3.847877279392546e-05,
      "loss": 0.8266,
      "num_input_tokens_seen": 30140168,
      "step": 51950
    },
    {
      "epoch": 7.73830801310694,
      "grad_norm": 6.65625,
      "learning_rate": 3.847603598751529e-05,
      "loss": 0.8301,
      "num_input_tokens_seen": 30143144,
      "step": 51955
    },
    {
      "epoch": 7.7390527256479,
      "grad_norm": 6.21875,
      "learning_rate": 3.847329895344121e-05,
      "loss": 0.8514,
      "num_input_tokens_seen": 30146248,
      "step": 51960
    },
    {
      "epoch": 7.739797438188859,
      "grad_norm": 21.125,
      "learning_rate": 3.847056169174942e-05,
      "loss": 1.4038,
      "num_input_tokens_seen": 30149064,
      "step": 51965
    },
    {
      "epoch": 7.740542150729818,
      "grad_norm": 8.375,
      "learning_rate": 3.846782420248619e-05,
      "loss": 0.8478,
      "num_input_tokens_seen": 30151880,
      "step": 51970
    },
    {
      "epoch": 7.741286863270777,
      "grad_norm": 7.0625,
      "learning_rate": 3.8465086485697766e-05,
      "loss": 1.3978,
      "num_input_tokens_seen": 30154600,
      "step": 51975
    },
    {
      "epoch": 7.742031575811737,
      "grad_norm": 11.0,
      "learning_rate": 3.8462348541430396e-05,
      "loss": 1.0938,
      "num_input_tokens_seen": 30157576,
      "step": 51980
    },
    {
      "epoch": 7.742776288352696,
      "grad_norm": 10.6875,
      "learning_rate": 3.8459610369730316e-05,
      "loss": 1.1384,
      "num_input_tokens_seen": 30160264,
      "step": 51985
    },
    {
      "epoch": 7.743521000893655,
      "grad_norm": 7.8125,
      "learning_rate": 3.8456871970643794e-05,
      "loss": 1.1554,
      "num_input_tokens_seen": 30162920,
      "step": 51990
    },
    {
      "epoch": 7.744265713434614,
      "grad_norm": 10.5625,
      "learning_rate": 3.8454133344217105e-05,
      "loss": 1.4469,
      "num_input_tokens_seen": 30165896,
      "step": 51995
    },
    {
      "epoch": 7.745010425975574,
      "grad_norm": 8.3125,
      "learning_rate": 3.8451394490496505e-05,
      "loss": 1.1245,
      "num_input_tokens_seen": 30168808,
      "step": 52000
    },
    {
      "epoch": 7.745755138516532,
      "grad_norm": 17.875,
      "learning_rate": 3.8448655409528274e-05,
      "loss": 1.1466,
      "num_input_tokens_seen": 30171432,
      "step": 52005
    },
    {
      "epoch": 7.746499851057492,
      "grad_norm": 12.3125,
      "learning_rate": 3.844591610135867e-05,
      "loss": 1.1533,
      "num_input_tokens_seen": 30174312,
      "step": 52010
    },
    {
      "epoch": 7.747244563598451,
      "grad_norm": 19.5,
      "learning_rate": 3.844317656603398e-05,
      "loss": 1.1174,
      "num_input_tokens_seen": 30177032,
      "step": 52015
    },
    {
      "epoch": 7.7479892761394105,
      "grad_norm": 6.125,
      "learning_rate": 3.844043680360049e-05,
      "loss": 0.7352,
      "num_input_tokens_seen": 30179784,
      "step": 52020
    },
    {
      "epoch": 7.748733988680369,
      "grad_norm": 12.375,
      "learning_rate": 3.8437696814104476e-05,
      "loss": 0.8689,
      "num_input_tokens_seen": 30182312,
      "step": 52025
    },
    {
      "epoch": 7.749478701221329,
      "grad_norm": 6.4375,
      "learning_rate": 3.8434956597592234e-05,
      "loss": 0.6951,
      "num_input_tokens_seen": 30185096,
      "step": 52030
    },
    {
      "epoch": 7.750223413762288,
      "grad_norm": 9.25,
      "learning_rate": 3.8432216154110053e-05,
      "loss": 0.9927,
      "num_input_tokens_seen": 30187752,
      "step": 52035
    },
    {
      "epoch": 7.750968126303247,
      "grad_norm": 17.75,
      "learning_rate": 3.8429475483704236e-05,
      "loss": 1.393,
      "num_input_tokens_seen": 30190408,
      "step": 52040
    },
    {
      "epoch": 7.751712838844206,
      "grad_norm": 8.6875,
      "learning_rate": 3.842673458642108e-05,
      "loss": 1.0748,
      "num_input_tokens_seen": 30193224,
      "step": 52045
    },
    {
      "epoch": 7.752457551385166,
      "grad_norm": 12.25,
      "learning_rate": 3.842399346230688e-05,
      "loss": 1.0748,
      "num_input_tokens_seen": 30195976,
      "step": 52050
    },
    {
      "epoch": 7.753202263926124,
      "grad_norm": 14.25,
      "learning_rate": 3.842125211140796e-05,
      "loss": 1.0928,
      "num_input_tokens_seen": 30198664,
      "step": 52055
    },
    {
      "epoch": 7.753946976467084,
      "grad_norm": 7.46875,
      "learning_rate": 3.8418510533770624e-05,
      "loss": 1.2832,
      "num_input_tokens_seen": 30201384,
      "step": 52060
    },
    {
      "epoch": 7.754691689008043,
      "grad_norm": 10.5625,
      "learning_rate": 3.841576872944119e-05,
      "loss": 1.0243,
      "num_input_tokens_seen": 30204104,
      "step": 52065
    },
    {
      "epoch": 7.755436401549002,
      "grad_norm": 15.625,
      "learning_rate": 3.841302669846599e-05,
      "loss": 1.0275,
      "num_input_tokens_seen": 30206792,
      "step": 52070
    },
    {
      "epoch": 7.756181114089961,
      "grad_norm": 21.75,
      "learning_rate": 3.841028444089133e-05,
      "loss": 1.502,
      "num_input_tokens_seen": 30209768,
      "step": 52075
    },
    {
      "epoch": 7.756925826630921,
      "grad_norm": 19.75,
      "learning_rate": 3.840754195676354e-05,
      "loss": 1.1802,
      "num_input_tokens_seen": 30212360,
      "step": 52080
    },
    {
      "epoch": 7.75767053917188,
      "grad_norm": 8.6875,
      "learning_rate": 3.8404799246128956e-05,
      "loss": 1.1758,
      "num_input_tokens_seen": 30215176,
      "step": 52085
    },
    {
      "epoch": 7.758415251712838,
      "grad_norm": 12.5625,
      "learning_rate": 3.8402056309033915e-05,
      "loss": 1.1039,
      "num_input_tokens_seen": 30218152,
      "step": 52090
    },
    {
      "epoch": 7.759159964253798,
      "grad_norm": 10.4375,
      "learning_rate": 3.839931314552475e-05,
      "loss": 1.1377,
      "num_input_tokens_seen": 30221128,
      "step": 52095
    },
    {
      "epoch": 7.759904676794758,
      "grad_norm": 11.0,
      "learning_rate": 3.8396569755647816e-05,
      "loss": 1.0765,
      "num_input_tokens_seen": 30223720,
      "step": 52100
    },
    {
      "epoch": 7.7606493893357165,
      "grad_norm": 13.9375,
      "learning_rate": 3.839382613944944e-05,
      "loss": 1.0648,
      "num_input_tokens_seen": 30226632,
      "step": 52105
    },
    {
      "epoch": 7.761394101876675,
      "grad_norm": 6.15625,
      "learning_rate": 3.8391082296976e-05,
      "loss": 1.038,
      "num_input_tokens_seen": 30229608,
      "step": 52110
    },
    {
      "epoch": 7.762138814417635,
      "grad_norm": 6.65625,
      "learning_rate": 3.8388338228273824e-05,
      "loss": 1.1127,
      "num_input_tokens_seen": 30232424,
      "step": 52115
    },
    {
      "epoch": 7.762883526958594,
      "grad_norm": 17.5,
      "learning_rate": 3.838559393338927e-05,
      "loss": 1.2264,
      "num_input_tokens_seen": 30235272,
      "step": 52120
    },
    {
      "epoch": 7.763628239499553,
      "grad_norm": 8.875,
      "learning_rate": 3.838284941236873e-05,
      "loss": 1.1378,
      "num_input_tokens_seen": 30238408,
      "step": 52125
    },
    {
      "epoch": 7.764372952040512,
      "grad_norm": 7.3125,
      "learning_rate": 3.8380104665258545e-05,
      "loss": 1.1443,
      "num_input_tokens_seen": 30241672,
      "step": 52130
    },
    {
      "epoch": 7.765117664581472,
      "grad_norm": 4.9375,
      "learning_rate": 3.837735969210509e-05,
      "loss": 0.9904,
      "num_input_tokens_seen": 30244520,
      "step": 52135
    },
    {
      "epoch": 7.76586237712243,
      "grad_norm": 12.3125,
      "learning_rate": 3.837461449295474e-05,
      "loss": 0.9183,
      "num_input_tokens_seen": 30247272,
      "step": 52140
    },
    {
      "epoch": 7.76660708966339,
      "grad_norm": 13.0625,
      "learning_rate": 3.837186906785387e-05,
      "loss": 1.0731,
      "num_input_tokens_seen": 30250280,
      "step": 52145
    },
    {
      "epoch": 7.767351802204349,
      "grad_norm": 14.25,
      "learning_rate": 3.836912341684886e-05,
      "loss": 1.2872,
      "num_input_tokens_seen": 30253224,
      "step": 52150
    },
    {
      "epoch": 7.7680965147453085,
      "grad_norm": 7.65625,
      "learning_rate": 3.83663775399861e-05,
      "loss": 0.8414,
      "num_input_tokens_seen": 30256296,
      "step": 52155
    },
    {
      "epoch": 7.768841227286267,
      "grad_norm": 7.53125,
      "learning_rate": 3.836363143731198e-05,
      "loss": 0.7991,
      "num_input_tokens_seen": 30259048,
      "step": 52160
    },
    {
      "epoch": 7.769585939827227,
      "grad_norm": 6.03125,
      "learning_rate": 3.8360885108872885e-05,
      "loss": 0.8111,
      "num_input_tokens_seen": 30262344,
      "step": 52165
    },
    {
      "epoch": 7.770330652368186,
      "grad_norm": 21.75,
      "learning_rate": 3.8358138554715215e-05,
      "loss": 1.1844,
      "num_input_tokens_seen": 30265832,
      "step": 52170
    },
    {
      "epoch": 7.771075364909145,
      "grad_norm": 9.25,
      "learning_rate": 3.8355391774885375e-05,
      "loss": 0.6544,
      "num_input_tokens_seen": 30268808,
      "step": 52175
    },
    {
      "epoch": 7.771820077450104,
      "grad_norm": 7.84375,
      "learning_rate": 3.835264476942977e-05,
      "loss": 1.0503,
      "num_input_tokens_seen": 30271944,
      "step": 52180
    },
    {
      "epoch": 7.772564789991064,
      "grad_norm": 21.25,
      "learning_rate": 3.834989753839479e-05,
      "loss": 0.7259,
      "num_input_tokens_seen": 30274952,
      "step": 52185
    },
    {
      "epoch": 7.7733095025320225,
      "grad_norm": 10.4375,
      "learning_rate": 3.834715008182687e-05,
      "loss": 0.8956,
      "num_input_tokens_seen": 30277864,
      "step": 52190
    },
    {
      "epoch": 7.774054215072982,
      "grad_norm": 7.75,
      "learning_rate": 3.83444023997724e-05,
      "loss": 0.739,
      "num_input_tokens_seen": 30280680,
      "step": 52195
    },
    {
      "epoch": 7.774798927613941,
      "grad_norm": 19.125,
      "learning_rate": 3.834165449227782e-05,
      "loss": 1.1255,
      "num_input_tokens_seen": 30283144,
      "step": 52200
    },
    {
      "epoch": 7.7755436401549005,
      "grad_norm": 12.5,
      "learning_rate": 3.833890635938956e-05,
      "loss": 0.8898,
      "num_input_tokens_seen": 30286568,
      "step": 52205
    },
    {
      "epoch": 7.776288352695859,
      "grad_norm": 9.5,
      "learning_rate": 3.8336158001154024e-05,
      "loss": 0.9705,
      "num_input_tokens_seen": 30289736,
      "step": 52210
    },
    {
      "epoch": 7.777033065236819,
      "grad_norm": 10.9375,
      "learning_rate": 3.8333409417617654e-05,
      "loss": 1.0757,
      "num_input_tokens_seen": 30292584,
      "step": 52215
    },
    {
      "epoch": 7.777777777777778,
      "grad_norm": 6.84375,
      "learning_rate": 3.8330660608826885e-05,
      "loss": 0.7186,
      "num_input_tokens_seen": 30295304,
      "step": 52220
    },
    {
      "epoch": 7.778522490318737,
      "grad_norm": 9.3125,
      "learning_rate": 3.832791157482815e-05,
      "loss": 1.0303,
      "num_input_tokens_seen": 30298184,
      "step": 52225
    },
    {
      "epoch": 7.779267202859696,
      "grad_norm": 6.0625,
      "learning_rate": 3.8325162315667895e-05,
      "loss": 0.9238,
      "num_input_tokens_seen": 30300872,
      "step": 52230
    },
    {
      "epoch": 7.780011915400656,
      "grad_norm": 8.3125,
      "learning_rate": 3.832241283139256e-05,
      "loss": 0.7888,
      "num_input_tokens_seen": 30303912,
      "step": 52235
    },
    {
      "epoch": 7.7807566279416145,
      "grad_norm": 5.84375,
      "learning_rate": 3.831966312204861e-05,
      "loss": 0.926,
      "num_input_tokens_seen": 30306376,
      "step": 52240
    },
    {
      "epoch": 7.781501340482574,
      "grad_norm": 13.6875,
      "learning_rate": 3.831691318768249e-05,
      "loss": 1.0492,
      "num_input_tokens_seen": 30309160,
      "step": 52245
    },
    {
      "epoch": 7.782246053023533,
      "grad_norm": 10.1875,
      "learning_rate": 3.831416302834065e-05,
      "loss": 1.0084,
      "num_input_tokens_seen": 30312008,
      "step": 52250
    },
    {
      "epoch": 7.782990765564492,
      "grad_norm": 5.6875,
      "learning_rate": 3.831141264406957e-05,
      "loss": 1.0905,
      "num_input_tokens_seen": 30314792,
      "step": 52255
    },
    {
      "epoch": 7.783735478105451,
      "grad_norm": 11.0,
      "learning_rate": 3.8308662034915685e-05,
      "loss": 1.0752,
      "num_input_tokens_seen": 30317960,
      "step": 52260
    },
    {
      "epoch": 7.784480190646411,
      "grad_norm": 6.5625,
      "learning_rate": 3.830591120092549e-05,
      "loss": 1.2425,
      "num_input_tokens_seen": 30320808,
      "step": 52265
    },
    {
      "epoch": 7.78522490318737,
      "grad_norm": 6.53125,
      "learning_rate": 3.8303160142145444e-05,
      "loss": 0.9162,
      "num_input_tokens_seen": 30323848,
      "step": 52270
    },
    {
      "epoch": 7.7859696157283285,
      "grad_norm": 9.0,
      "learning_rate": 3.830040885862204e-05,
      "loss": 1.2354,
      "num_input_tokens_seen": 30326888,
      "step": 52275
    },
    {
      "epoch": 7.786714328269288,
      "grad_norm": 13.625,
      "learning_rate": 3.8297657350401735e-05,
      "loss": 1.2513,
      "num_input_tokens_seen": 30330024,
      "step": 52280
    },
    {
      "epoch": 7.787459040810247,
      "grad_norm": 12.875,
      "learning_rate": 3.829490561753103e-05,
      "loss": 0.942,
      "num_input_tokens_seen": 30332936,
      "step": 52285
    },
    {
      "epoch": 7.7882037533512065,
      "grad_norm": 12.75,
      "learning_rate": 3.82921536600564e-05,
      "loss": 1.1789,
      "num_input_tokens_seen": 30335592,
      "step": 52290
    },
    {
      "epoch": 7.788948465892165,
      "grad_norm": 8.5625,
      "learning_rate": 3.828940147802435e-05,
      "loss": 1.0033,
      "num_input_tokens_seen": 30338568,
      "step": 52295
    },
    {
      "epoch": 7.789693178433125,
      "grad_norm": 22.125,
      "learning_rate": 3.828664907148137e-05,
      "loss": 1.0619,
      "num_input_tokens_seen": 30341448,
      "step": 52300
    },
    {
      "epoch": 7.790437890974084,
      "grad_norm": 12.4375,
      "learning_rate": 3.828389644047395e-05,
      "loss": 0.9885,
      "num_input_tokens_seen": 30343976,
      "step": 52305
    },
    {
      "epoch": 7.791182603515043,
      "grad_norm": 10.1875,
      "learning_rate": 3.8281143585048604e-05,
      "loss": 1.2155,
      "num_input_tokens_seen": 30347144,
      "step": 52310
    },
    {
      "epoch": 7.791927316056002,
      "grad_norm": 9.25,
      "learning_rate": 3.8278390505251835e-05,
      "loss": 0.8112,
      "num_input_tokens_seen": 30349768,
      "step": 52315
    },
    {
      "epoch": 7.792672028596962,
      "grad_norm": 9.0625,
      "learning_rate": 3.827563720113016e-05,
      "loss": 1.5952,
      "num_input_tokens_seen": 30352808,
      "step": 52320
    },
    {
      "epoch": 7.7934167411379205,
      "grad_norm": 9.875,
      "learning_rate": 3.827288367273008e-05,
      "loss": 1.1995,
      "num_input_tokens_seen": 30355336,
      "step": 52325
    },
    {
      "epoch": 7.79416145367888,
      "grad_norm": 11.6875,
      "learning_rate": 3.827012992009812e-05,
      "loss": 1.2493,
      "num_input_tokens_seen": 30358280,
      "step": 52330
    },
    {
      "epoch": 7.794906166219839,
      "grad_norm": 5.96875,
      "learning_rate": 3.826737594328082e-05,
      "loss": 1.0004,
      "num_input_tokens_seen": 30361160,
      "step": 52335
    },
    {
      "epoch": 7.7956508787607985,
      "grad_norm": 6.78125,
      "learning_rate": 3.826462174232467e-05,
      "loss": 1.2211,
      "num_input_tokens_seen": 30364232,
      "step": 52340
    },
    {
      "epoch": 7.796395591301757,
      "grad_norm": 7.59375,
      "learning_rate": 3.8261867317276225e-05,
      "loss": 1.1914,
      "num_input_tokens_seen": 30367016,
      "step": 52345
    },
    {
      "epoch": 7.797140303842717,
      "grad_norm": 7.34375,
      "learning_rate": 3.8259112668181995e-05,
      "loss": 1.157,
      "num_input_tokens_seen": 30370088,
      "step": 52350
    },
    {
      "epoch": 7.797885016383676,
      "grad_norm": 11.9375,
      "learning_rate": 3.825635779508855e-05,
      "loss": 0.9306,
      "num_input_tokens_seen": 30373000,
      "step": 52355
    },
    {
      "epoch": 7.798629728924635,
      "grad_norm": 13.0,
      "learning_rate": 3.82536026980424e-05,
      "loss": 1.2535,
      "num_input_tokens_seen": 30375848,
      "step": 52360
    },
    {
      "epoch": 7.799374441465594,
      "grad_norm": 6.34375,
      "learning_rate": 3.825084737709011e-05,
      "loss": 0.9358,
      "num_input_tokens_seen": 30378536,
      "step": 52365
    },
    {
      "epoch": 7.800119154006554,
      "grad_norm": 8.625,
      "learning_rate": 3.824809183227822e-05,
      "loss": 1.1057,
      "num_input_tokens_seen": 30381512,
      "step": 52370
    },
    {
      "epoch": 7.8008638665475125,
      "grad_norm": 13.3125,
      "learning_rate": 3.824533606365329e-05,
      "loss": 1.2629,
      "num_input_tokens_seen": 30384584,
      "step": 52375
    },
    {
      "epoch": 7.801608579088472,
      "grad_norm": 14.5625,
      "learning_rate": 3.824258007126186e-05,
      "loss": 1.2913,
      "num_input_tokens_seen": 30387560,
      "step": 52380
    },
    {
      "epoch": 7.802353291629431,
      "grad_norm": 8.75,
      "learning_rate": 3.82398238551505e-05,
      "loss": 1.129,
      "num_input_tokens_seen": 30390248,
      "step": 52385
    },
    {
      "epoch": 7.803098004170391,
      "grad_norm": 13.3125,
      "learning_rate": 3.823706741536578e-05,
      "loss": 1.2249,
      "num_input_tokens_seen": 30393032,
      "step": 52390
    },
    {
      "epoch": 7.803842716711349,
      "grad_norm": 8.375,
      "learning_rate": 3.823431075195425e-05,
      "loss": 1.0673,
      "num_input_tokens_seen": 30395784,
      "step": 52395
    },
    {
      "epoch": 7.804587429252309,
      "grad_norm": 8.0,
      "learning_rate": 3.8231553864962486e-05,
      "loss": 1.0,
      "num_input_tokens_seen": 30398600,
      "step": 52400
    },
    {
      "epoch": 7.805332141793268,
      "grad_norm": 8.5,
      "learning_rate": 3.8228796754437086e-05,
      "loss": 1.3266,
      "num_input_tokens_seen": 30401480,
      "step": 52405
    },
    {
      "epoch": 7.806076854334227,
      "grad_norm": 7.8125,
      "learning_rate": 3.8226039420424596e-05,
      "loss": 1.1948,
      "num_input_tokens_seen": 30404392,
      "step": 52410
    },
    {
      "epoch": 7.806821566875186,
      "grad_norm": 16.125,
      "learning_rate": 3.822328186297162e-05,
      "loss": 1.0644,
      "num_input_tokens_seen": 30407016,
      "step": 52415
    },
    {
      "epoch": 7.807566279416145,
      "grad_norm": 6.1875,
      "learning_rate": 3.822052408212473e-05,
      "loss": 0.9835,
      "num_input_tokens_seen": 30409992,
      "step": 52420
    },
    {
      "epoch": 7.8083109919571045,
      "grad_norm": 8.375,
      "learning_rate": 3.8217766077930527e-05,
      "loss": 1.2,
      "num_input_tokens_seen": 30412872,
      "step": 52425
    },
    {
      "epoch": 7.809055704498064,
      "grad_norm": 17.25,
      "learning_rate": 3.82150078504356e-05,
      "loss": 0.7413,
      "num_input_tokens_seen": 30415592,
      "step": 52430
    },
    {
      "epoch": 7.809800417039023,
      "grad_norm": 8.75,
      "learning_rate": 3.821224939968654e-05,
      "loss": 0.9041,
      "num_input_tokens_seen": 30418312,
      "step": 52435
    },
    {
      "epoch": 7.810545129579982,
      "grad_norm": 6.46875,
      "learning_rate": 3.820949072572996e-05,
      "loss": 0.9606,
      "num_input_tokens_seen": 30421096,
      "step": 52440
    },
    {
      "epoch": 7.811289842120941,
      "grad_norm": 7.28125,
      "learning_rate": 3.820673182861246e-05,
      "loss": 0.8231,
      "num_input_tokens_seen": 30423880,
      "step": 52445
    },
    {
      "epoch": 7.812034554661901,
      "grad_norm": 12.0,
      "learning_rate": 3.820397270838064e-05,
      "loss": 0.8703,
      "num_input_tokens_seen": 30426888,
      "step": 52450
    },
    {
      "epoch": 7.81277926720286,
      "grad_norm": 4.4375,
      "learning_rate": 3.820121336508113e-05,
      "loss": 0.9734,
      "num_input_tokens_seen": 30429864,
      "step": 52455
    },
    {
      "epoch": 7.8135239797438185,
      "grad_norm": 6.8125,
      "learning_rate": 3.819845379876054e-05,
      "loss": 0.8206,
      "num_input_tokens_seen": 30433000,
      "step": 52460
    },
    {
      "epoch": 7.814268692284778,
      "grad_norm": 3.03125,
      "learning_rate": 3.8195694009465486e-05,
      "loss": 0.6567,
      "num_input_tokens_seen": 30435784,
      "step": 52465
    },
    {
      "epoch": 7.815013404825737,
      "grad_norm": 10.875,
      "learning_rate": 3.819293399724259e-05,
      "loss": 0.8265,
      "num_input_tokens_seen": 30438568,
      "step": 52470
    },
    {
      "epoch": 7.815758117366697,
      "grad_norm": 12.0625,
      "learning_rate": 3.819017376213848e-05,
      "loss": 0.9393,
      "num_input_tokens_seen": 30441672,
      "step": 52475
    },
    {
      "epoch": 7.816502829907655,
      "grad_norm": 7.8125,
      "learning_rate": 3.8187413304199796e-05,
      "loss": 0.6086,
      "num_input_tokens_seen": 30444488,
      "step": 52480
    },
    {
      "epoch": 7.817247542448615,
      "grad_norm": 12.1875,
      "learning_rate": 3.818465262347316e-05,
      "loss": 1.0588,
      "num_input_tokens_seen": 30447240,
      "step": 52485
    },
    {
      "epoch": 7.817992254989574,
      "grad_norm": 12.5,
      "learning_rate": 3.818189172000522e-05,
      "loss": 1.2285,
      "num_input_tokens_seen": 30449992,
      "step": 52490
    },
    {
      "epoch": 7.818736967530533,
      "grad_norm": 15.875,
      "learning_rate": 3.8179130593842626e-05,
      "loss": 1.1538,
      "num_input_tokens_seen": 30453288,
      "step": 52495
    },
    {
      "epoch": 7.819481680071492,
      "grad_norm": 18.5,
      "learning_rate": 3.8176369245032006e-05,
      "loss": 1.1782,
      "num_input_tokens_seen": 30456136,
      "step": 52500
    },
    {
      "epoch": 7.820226392612452,
      "grad_norm": 13.4375,
      "learning_rate": 3.817360767362003e-05,
      "loss": 1.0489,
      "num_input_tokens_seen": 30458952,
      "step": 52505
    },
    {
      "epoch": 7.8209711051534105,
      "grad_norm": 5.46875,
      "learning_rate": 3.817084587965333e-05,
      "loss": 1.0173,
      "num_input_tokens_seen": 30461608,
      "step": 52510
    },
    {
      "epoch": 7.82171581769437,
      "grad_norm": 7.8125,
      "learning_rate": 3.8168083863178586e-05,
      "loss": 0.9602,
      "num_input_tokens_seen": 30464808,
      "step": 52515
    },
    {
      "epoch": 7.822460530235329,
      "grad_norm": 6.375,
      "learning_rate": 3.8165321624242434e-05,
      "loss": 0.9533,
      "num_input_tokens_seen": 30467944,
      "step": 52520
    },
    {
      "epoch": 7.823205242776289,
      "grad_norm": 12.0,
      "learning_rate": 3.816255916289156e-05,
      "loss": 1.3341,
      "num_input_tokens_seen": 30470856,
      "step": 52525
    },
    {
      "epoch": 7.823949955317247,
      "grad_norm": 14.6875,
      "learning_rate": 3.8159796479172626e-05,
      "loss": 0.8995,
      "num_input_tokens_seen": 30473704,
      "step": 52530
    },
    {
      "epoch": 7.824694667858207,
      "grad_norm": 10.75,
      "learning_rate": 3.815703357313231e-05,
      "loss": 1.1004,
      "num_input_tokens_seen": 30476552,
      "step": 52535
    },
    {
      "epoch": 7.825439380399166,
      "grad_norm": 9.25,
      "learning_rate": 3.8154270444817285e-05,
      "loss": 1.1022,
      "num_input_tokens_seen": 30479560,
      "step": 52540
    },
    {
      "epoch": 7.826184092940125,
      "grad_norm": 16.125,
      "learning_rate": 3.815150709427423e-05,
      "loss": 1.1752,
      "num_input_tokens_seen": 30482440,
      "step": 52545
    },
    {
      "epoch": 7.826928805481084,
      "grad_norm": 10.4375,
      "learning_rate": 3.8148743521549824e-05,
      "loss": 0.7104,
      "num_input_tokens_seen": 30485192,
      "step": 52550
    },
    {
      "epoch": 7.827673518022044,
      "grad_norm": 9.375,
      "learning_rate": 3.814597972669076e-05,
      "loss": 1.0315,
      "num_input_tokens_seen": 30488136,
      "step": 52555
    },
    {
      "epoch": 7.828418230563003,
      "grad_norm": 7.53125,
      "learning_rate": 3.814321570974373e-05,
      "loss": 1.1134,
      "num_input_tokens_seen": 30490920,
      "step": 52560
    },
    {
      "epoch": 7.829162943103962,
      "grad_norm": 7.5,
      "learning_rate": 3.814045147075543e-05,
      "loss": 0.9165,
      "num_input_tokens_seen": 30493672,
      "step": 52565
    },
    {
      "epoch": 7.829907655644921,
      "grad_norm": 7.125,
      "learning_rate": 3.813768700977256e-05,
      "loss": 1.1126,
      "num_input_tokens_seen": 30496680,
      "step": 52570
    },
    {
      "epoch": 7.830652368185881,
      "grad_norm": 9.75,
      "learning_rate": 3.813492232684182e-05,
      "loss": 0.9657,
      "num_input_tokens_seen": 30499560,
      "step": 52575
    },
    {
      "epoch": 7.831397080726839,
      "grad_norm": 13.0,
      "learning_rate": 3.813215742200992e-05,
      "loss": 1.3107,
      "num_input_tokens_seen": 30502504,
      "step": 52580
    },
    {
      "epoch": 7.832141793267798,
      "grad_norm": 15.6875,
      "learning_rate": 3.8129392295323566e-05,
      "loss": 1.0365,
      "num_input_tokens_seen": 30505480,
      "step": 52585
    },
    {
      "epoch": 7.832886505808758,
      "grad_norm": 13.0,
      "learning_rate": 3.812662694682946e-05,
      "loss": 1.1829,
      "num_input_tokens_seen": 30508264,
      "step": 52590
    },
    {
      "epoch": 7.833631218349717,
      "grad_norm": 11.0,
      "learning_rate": 3.8123861376574344e-05,
      "loss": 0.8102,
      "num_input_tokens_seen": 30511016,
      "step": 52595
    },
    {
      "epoch": 7.834375930890676,
      "grad_norm": 9.0,
      "learning_rate": 3.8121095584604925e-05,
      "loss": 0.7376,
      "num_input_tokens_seen": 30513864,
      "step": 52600
    },
    {
      "epoch": 7.835120643431635,
      "grad_norm": 10.3125,
      "learning_rate": 3.811832957096794e-05,
      "loss": 1.1041,
      "num_input_tokens_seen": 30516744,
      "step": 52605
    },
    {
      "epoch": 7.835865355972595,
      "grad_norm": 22.375,
      "learning_rate": 3.81155633357101e-05,
      "loss": 1.2353,
      "num_input_tokens_seen": 30519656,
      "step": 52610
    },
    {
      "epoch": 7.836610068513554,
      "grad_norm": 10.3125,
      "learning_rate": 3.8112796878878155e-05,
      "loss": 0.8944,
      "num_input_tokens_seen": 30522504,
      "step": 52615
    },
    {
      "epoch": 7.837354781054513,
      "grad_norm": 5.375,
      "learning_rate": 3.811003020051883e-05,
      "loss": 1.0778,
      "num_input_tokens_seen": 30525480,
      "step": 52620
    },
    {
      "epoch": 7.838099493595472,
      "grad_norm": 19.125,
      "learning_rate": 3.8107263300678874e-05,
      "loss": 1.794,
      "num_input_tokens_seen": 30528488,
      "step": 52625
    },
    {
      "epoch": 7.838844206136431,
      "grad_norm": 10.125,
      "learning_rate": 3.810449617940502e-05,
      "loss": 1.1325,
      "num_input_tokens_seen": 30531624,
      "step": 52630
    },
    {
      "epoch": 7.83958891867739,
      "grad_norm": 9.875,
      "learning_rate": 3.810172883674402e-05,
      "loss": 1.1789,
      "num_input_tokens_seen": 30534856,
      "step": 52635
    },
    {
      "epoch": 7.84033363121835,
      "grad_norm": 8.25,
      "learning_rate": 3.809896127274264e-05,
      "loss": 0.7534,
      "num_input_tokens_seen": 30537960,
      "step": 52640
    },
    {
      "epoch": 7.841078343759309,
      "grad_norm": 11.625,
      "learning_rate": 3.8096193487447604e-05,
      "loss": 1.2481,
      "num_input_tokens_seen": 30541000,
      "step": 52645
    },
    {
      "epoch": 7.841823056300268,
      "grad_norm": 9.8125,
      "learning_rate": 3.8093425480905706e-05,
      "loss": 1.3577,
      "num_input_tokens_seen": 30544072,
      "step": 52650
    },
    {
      "epoch": 7.842567768841227,
      "grad_norm": 8.5625,
      "learning_rate": 3.809065725316368e-05,
      "loss": 1.204,
      "num_input_tokens_seen": 30546920,
      "step": 52655
    },
    {
      "epoch": 7.843312481382187,
      "grad_norm": 8.3125,
      "learning_rate": 3.808788880426831e-05,
      "loss": 0.8759,
      "num_input_tokens_seen": 30550184,
      "step": 52660
    },
    {
      "epoch": 7.844057193923145,
      "grad_norm": 10.3125,
      "learning_rate": 3.8085120134266364e-05,
      "loss": 0.9198,
      "num_input_tokens_seen": 30553480,
      "step": 52665
    },
    {
      "epoch": 7.844801906464105,
      "grad_norm": 15.125,
      "learning_rate": 3.8082351243204605e-05,
      "loss": 1.287,
      "num_input_tokens_seen": 30556104,
      "step": 52670
    },
    {
      "epoch": 7.845546619005064,
      "grad_norm": 13.1875,
      "learning_rate": 3.8079582131129826e-05,
      "loss": 1.2588,
      "num_input_tokens_seen": 30559368,
      "step": 52675
    },
    {
      "epoch": 7.846291331546023,
      "grad_norm": 7.21875,
      "learning_rate": 3.8076812798088796e-05,
      "loss": 1.1599,
      "num_input_tokens_seen": 30562152,
      "step": 52680
    },
    {
      "epoch": 7.847036044086982,
      "grad_norm": 7.375,
      "learning_rate": 3.80740432441283e-05,
      "loss": 1.1783,
      "num_input_tokens_seen": 30565288,
      "step": 52685
    },
    {
      "epoch": 7.847780756627942,
      "grad_norm": 13.3125,
      "learning_rate": 3.807127346929514e-05,
      "loss": 1.0165,
      "num_input_tokens_seen": 30567976,
      "step": 52690
    },
    {
      "epoch": 7.848525469168901,
      "grad_norm": 13.625,
      "learning_rate": 3.806850347363609e-05,
      "loss": 1.0988,
      "num_input_tokens_seen": 30570728,
      "step": 52695
    },
    {
      "epoch": 7.84927018170986,
      "grad_norm": 11.0625,
      "learning_rate": 3.8065733257197964e-05,
      "loss": 1.2465,
      "num_input_tokens_seen": 30573544,
      "step": 52700
    },
    {
      "epoch": 7.850014894250819,
      "grad_norm": 7.46875,
      "learning_rate": 3.806296282002756e-05,
      "loss": 1.267,
      "num_input_tokens_seen": 30576680,
      "step": 52705
    },
    {
      "epoch": 7.850759606791779,
      "grad_norm": 15.3125,
      "learning_rate": 3.8060192162171664e-05,
      "loss": 1.6965,
      "num_input_tokens_seen": 30579688,
      "step": 52710
    },
    {
      "epoch": 7.851504319332737,
      "grad_norm": 8.6875,
      "learning_rate": 3.80574212836771e-05,
      "loss": 0.7075,
      "num_input_tokens_seen": 30582216,
      "step": 52715
    },
    {
      "epoch": 7.852249031873697,
      "grad_norm": 16.0,
      "learning_rate": 3.805465018459067e-05,
      "loss": 1.1943,
      "num_input_tokens_seen": 30585192,
      "step": 52720
    },
    {
      "epoch": 7.852993744414656,
      "grad_norm": 7.9375,
      "learning_rate": 3.8051878864959194e-05,
      "loss": 1.0404,
      "num_input_tokens_seen": 30587944,
      "step": 52725
    },
    {
      "epoch": 7.8537384569556155,
      "grad_norm": 8.4375,
      "learning_rate": 3.804910732482949e-05,
      "loss": 1.1181,
      "num_input_tokens_seen": 30590792,
      "step": 52730
    },
    {
      "epoch": 7.854483169496574,
      "grad_norm": 13.5,
      "learning_rate": 3.804633556424839e-05,
      "loss": 1.3895,
      "num_input_tokens_seen": 30593640,
      "step": 52735
    },
    {
      "epoch": 7.855227882037534,
      "grad_norm": 10.0,
      "learning_rate": 3.804356358326271e-05,
      "loss": 1.0737,
      "num_input_tokens_seen": 30596744,
      "step": 52740
    },
    {
      "epoch": 7.855972594578493,
      "grad_norm": 11.125,
      "learning_rate": 3.804079138191927e-05,
      "loss": 1.1189,
      "num_input_tokens_seen": 30599720,
      "step": 52745
    },
    {
      "epoch": 7.856717307119452,
      "grad_norm": 11.5,
      "learning_rate": 3.803801896026491e-05,
      "loss": 0.9215,
      "num_input_tokens_seen": 30602760,
      "step": 52750
    },
    {
      "epoch": 7.857462019660411,
      "grad_norm": 9.8125,
      "learning_rate": 3.803524631834648e-05,
      "loss": 0.8153,
      "num_input_tokens_seen": 30605544,
      "step": 52755
    },
    {
      "epoch": 7.858206732201371,
      "grad_norm": 15.9375,
      "learning_rate": 3.8032473456210805e-05,
      "loss": 1.0218,
      "num_input_tokens_seen": 30608328,
      "step": 52760
    },
    {
      "epoch": 7.858951444742329,
      "grad_norm": 15.625,
      "learning_rate": 3.8029700373904744e-05,
      "loss": 1.0279,
      "num_input_tokens_seen": 30611112,
      "step": 52765
    },
    {
      "epoch": 7.859696157283288,
      "grad_norm": 5.40625,
      "learning_rate": 3.802692707147514e-05,
      "loss": 1.2236,
      "num_input_tokens_seen": 30613896,
      "step": 52770
    },
    {
      "epoch": 7.860440869824248,
      "grad_norm": 20.375,
      "learning_rate": 3.8024153548968835e-05,
      "loss": 1.4539,
      "num_input_tokens_seen": 30616648,
      "step": 52775
    },
    {
      "epoch": 7.8611855823652075,
      "grad_norm": 10.75,
      "learning_rate": 3.80213798064327e-05,
      "loss": 1.0302,
      "num_input_tokens_seen": 30619336,
      "step": 52780
    },
    {
      "epoch": 7.861930294906166,
      "grad_norm": 6.75,
      "learning_rate": 3.801860584391358e-05,
      "loss": 1.0086,
      "num_input_tokens_seen": 30622056,
      "step": 52785
    },
    {
      "epoch": 7.862675007447125,
      "grad_norm": 7.1875,
      "learning_rate": 3.801583166145835e-05,
      "loss": 1.0892,
      "num_input_tokens_seen": 30624904,
      "step": 52790
    },
    {
      "epoch": 7.863419719988085,
      "grad_norm": 6.0,
      "learning_rate": 3.801305725911387e-05,
      "loss": 0.9988,
      "num_input_tokens_seen": 30628040,
      "step": 52795
    },
    {
      "epoch": 7.864164432529043,
      "grad_norm": 12.0625,
      "learning_rate": 3.8010282636927016e-05,
      "loss": 1.1518,
      "num_input_tokens_seen": 30631016,
      "step": 52800
    },
    {
      "epoch": 7.864909145070003,
      "grad_norm": 11.25,
      "learning_rate": 3.800750779494466e-05,
      "loss": 1.24,
      "num_input_tokens_seen": 30633832,
      "step": 52805
    },
    {
      "epoch": 7.865653857610962,
      "grad_norm": 9.1875,
      "learning_rate": 3.8004732733213674e-05,
      "loss": 1.3243,
      "num_input_tokens_seen": 30636456,
      "step": 52810
    },
    {
      "epoch": 7.8663985701519215,
      "grad_norm": 11.9375,
      "learning_rate": 3.8001957451780956e-05,
      "loss": 0.8635,
      "num_input_tokens_seen": 30639336,
      "step": 52815
    },
    {
      "epoch": 7.86714328269288,
      "grad_norm": 9.625,
      "learning_rate": 3.799918195069338e-05,
      "loss": 1.3704,
      "num_input_tokens_seen": 30642408,
      "step": 52820
    },
    {
      "epoch": 7.86788799523384,
      "grad_norm": 15.375,
      "learning_rate": 3.799640622999784e-05,
      "loss": 1.1148,
      "num_input_tokens_seen": 30644872,
      "step": 52825
    },
    {
      "epoch": 7.868632707774799,
      "grad_norm": 10.875,
      "learning_rate": 3.799363028974121e-05,
      "loss": 1.2496,
      "num_input_tokens_seen": 30647752,
      "step": 52830
    },
    {
      "epoch": 7.869377420315758,
      "grad_norm": 14.1875,
      "learning_rate": 3.799085412997041e-05,
      "loss": 1.3283,
      "num_input_tokens_seen": 30651144,
      "step": 52835
    },
    {
      "epoch": 7.870122132856717,
      "grad_norm": 13.0625,
      "learning_rate": 3.798807775073234e-05,
      "loss": 0.9553,
      "num_input_tokens_seen": 30653800,
      "step": 52840
    },
    {
      "epoch": 7.870866845397677,
      "grad_norm": 11.5625,
      "learning_rate": 3.7985301152073896e-05,
      "loss": 1.1327,
      "num_input_tokens_seen": 30656680,
      "step": 52845
    },
    {
      "epoch": 7.871611557938635,
      "grad_norm": 7.9375,
      "learning_rate": 3.798252433404198e-05,
      "loss": 1.0581,
      "num_input_tokens_seen": 30659528,
      "step": 52850
    },
    {
      "epoch": 7.872356270479595,
      "grad_norm": 3.953125,
      "learning_rate": 3.797974729668351e-05,
      "loss": 0.9685,
      "num_input_tokens_seen": 30662216,
      "step": 52855
    },
    {
      "epoch": 7.873100983020554,
      "grad_norm": 12.5,
      "learning_rate": 3.7976970040045404e-05,
      "loss": 1.206,
      "num_input_tokens_seen": 30665000,
      "step": 52860
    },
    {
      "epoch": 7.8738456955615135,
      "grad_norm": 14.0625,
      "learning_rate": 3.797419256417458e-05,
      "loss": 1.1068,
      "num_input_tokens_seen": 30667912,
      "step": 52865
    },
    {
      "epoch": 7.874590408102472,
      "grad_norm": 9.5,
      "learning_rate": 3.797141486911796e-05,
      "loss": 0.8234,
      "num_input_tokens_seen": 30670824,
      "step": 52870
    },
    {
      "epoch": 7.875335120643432,
      "grad_norm": 11.1875,
      "learning_rate": 3.796863695492247e-05,
      "loss": 0.9498,
      "num_input_tokens_seen": 30673512,
      "step": 52875
    },
    {
      "epoch": 7.876079833184391,
      "grad_norm": 13.625,
      "learning_rate": 3.796585882163503e-05,
      "loss": 0.8324,
      "num_input_tokens_seen": 30676456,
      "step": 52880
    },
    {
      "epoch": 7.87682454572535,
      "grad_norm": 13.5,
      "learning_rate": 3.796308046930258e-05,
      "loss": 1.1091,
      "num_input_tokens_seen": 30679176,
      "step": 52885
    },
    {
      "epoch": 7.877569258266309,
      "grad_norm": 17.125,
      "learning_rate": 3.796030189797207e-05,
      "loss": 0.8884,
      "num_input_tokens_seen": 30682120,
      "step": 52890
    },
    {
      "epoch": 7.878313970807269,
      "grad_norm": 7.78125,
      "learning_rate": 3.795752310769044e-05,
      "loss": 1.231,
      "num_input_tokens_seen": 30684904,
      "step": 52895
    },
    {
      "epoch": 7.8790586833482275,
      "grad_norm": 6.46875,
      "learning_rate": 3.795474409850462e-05,
      "loss": 1.2379,
      "num_input_tokens_seen": 30687624,
      "step": 52900
    },
    {
      "epoch": 7.879803395889187,
      "grad_norm": 6.375,
      "learning_rate": 3.795196487046157e-05,
      "loss": 1.0809,
      "num_input_tokens_seen": 30690504,
      "step": 52905
    },
    {
      "epoch": 7.880548108430146,
      "grad_norm": 12.9375,
      "learning_rate": 3.794918542360822e-05,
      "loss": 1.0466,
      "num_input_tokens_seen": 30693320,
      "step": 52910
    },
    {
      "epoch": 7.8812928209711055,
      "grad_norm": 10.5625,
      "learning_rate": 3.7946405757991556e-05,
      "loss": 1.0203,
      "num_input_tokens_seen": 30696264,
      "step": 52915
    },
    {
      "epoch": 7.882037533512064,
      "grad_norm": 11.0,
      "learning_rate": 3.7943625873658515e-05,
      "loss": 0.9372,
      "num_input_tokens_seen": 30699240,
      "step": 52920
    },
    {
      "epoch": 7.882782246053024,
      "grad_norm": 6.625,
      "learning_rate": 3.7940845770656085e-05,
      "loss": 1.0993,
      "num_input_tokens_seen": 30702056,
      "step": 52925
    },
    {
      "epoch": 7.883526958593983,
      "grad_norm": 9.9375,
      "learning_rate": 3.7938065449031206e-05,
      "loss": 0.9775,
      "num_input_tokens_seen": 30706120,
      "step": 52930
    },
    {
      "epoch": 7.884271671134941,
      "grad_norm": 3.609375,
      "learning_rate": 3.793528490883087e-05,
      "loss": 1.1051,
      "num_input_tokens_seen": 30709288,
      "step": 52935
    },
    {
      "epoch": 7.885016383675901,
      "grad_norm": 5.9375,
      "learning_rate": 3.7932504150102045e-05,
      "loss": 1.2886,
      "num_input_tokens_seen": 30711976,
      "step": 52940
    },
    {
      "epoch": 7.885761096216861,
      "grad_norm": 11.9375,
      "learning_rate": 3.7929723172891696e-05,
      "loss": 1.0007,
      "num_input_tokens_seen": 30714952,
      "step": 52945
    },
    {
      "epoch": 7.8865058087578195,
      "grad_norm": 9.25,
      "learning_rate": 3.792694197724682e-05,
      "loss": 1.3054,
      "num_input_tokens_seen": 30718024,
      "step": 52950
    },
    {
      "epoch": 7.887250521298778,
      "grad_norm": 15.125,
      "learning_rate": 3.7924160563214395e-05,
      "loss": 1.2748,
      "num_input_tokens_seen": 30720840,
      "step": 52955
    },
    {
      "epoch": 7.887995233839738,
      "grad_norm": 10.375,
      "learning_rate": 3.792137893084141e-05,
      "loss": 1.4467,
      "num_input_tokens_seen": 30723400,
      "step": 52960
    },
    {
      "epoch": 7.8887399463806975,
      "grad_norm": 10.9375,
      "learning_rate": 3.791859708017486e-05,
      "loss": 0.9728,
      "num_input_tokens_seen": 30726152,
      "step": 52965
    },
    {
      "epoch": 7.889484658921656,
      "grad_norm": 5.53125,
      "learning_rate": 3.791581501126175e-05,
      "loss": 1.1403,
      "num_input_tokens_seen": 30729224,
      "step": 52970
    },
    {
      "epoch": 7.890229371462615,
      "grad_norm": 5.90625,
      "learning_rate": 3.791303272414907e-05,
      "loss": 1.1688,
      "num_input_tokens_seen": 30732072,
      "step": 52975
    },
    {
      "epoch": 7.890974084003575,
      "grad_norm": 5.65625,
      "learning_rate": 3.791025021888382e-05,
      "loss": 0.8962,
      "num_input_tokens_seen": 30734728,
      "step": 52980
    },
    {
      "epoch": 7.8917187965445335,
      "grad_norm": 9.75,
      "learning_rate": 3.7907467495513026e-05,
      "loss": 1.0797,
      "num_input_tokens_seen": 30737640,
      "step": 52985
    },
    {
      "epoch": 7.892463509085493,
      "grad_norm": 4.78125,
      "learning_rate": 3.790468455408368e-05,
      "loss": 1.0312,
      "num_input_tokens_seen": 30740648,
      "step": 52990
    },
    {
      "epoch": 7.893208221626452,
      "grad_norm": 7.59375,
      "learning_rate": 3.79019013946428e-05,
      "loss": 1.0802,
      "num_input_tokens_seen": 30743336,
      "step": 52995
    },
    {
      "epoch": 7.8939529341674115,
      "grad_norm": 8.875,
      "learning_rate": 3.789911801723742e-05,
      "loss": 1.6683,
      "num_input_tokens_seen": 30746344,
      "step": 53000
    },
    {
      "epoch": 7.89469764670837,
      "grad_norm": 14.5,
      "learning_rate": 3.789633442191455e-05,
      "loss": 1.323,
      "num_input_tokens_seen": 30749384,
      "step": 53005
    },
    {
      "epoch": 7.89544235924933,
      "grad_norm": 14.0,
      "learning_rate": 3.7893550608721206e-05,
      "loss": 0.9064,
      "num_input_tokens_seen": 30752456,
      "step": 53010
    },
    {
      "epoch": 7.896187071790289,
      "grad_norm": 9.125,
      "learning_rate": 3.789076657770444e-05,
      "loss": 1.3352,
      "num_input_tokens_seen": 30755112,
      "step": 53015
    },
    {
      "epoch": 7.896931784331248,
      "grad_norm": 8.25,
      "learning_rate": 3.788798232891127e-05,
      "loss": 0.9759,
      "num_input_tokens_seen": 30757960,
      "step": 53020
    },
    {
      "epoch": 7.897676496872207,
      "grad_norm": 10.25,
      "learning_rate": 3.788519786238873e-05,
      "loss": 1.0724,
      "num_input_tokens_seen": 30760712,
      "step": 53025
    },
    {
      "epoch": 7.898421209413167,
      "grad_norm": 7.53125,
      "learning_rate": 3.788241317818388e-05,
      "loss": 0.9218,
      "num_input_tokens_seen": 30763528,
      "step": 53030
    },
    {
      "epoch": 7.8991659219541255,
      "grad_norm": 10.0625,
      "learning_rate": 3.7879628276343746e-05,
      "loss": 1.1999,
      "num_input_tokens_seen": 30766152,
      "step": 53035
    },
    {
      "epoch": 7.899910634495085,
      "grad_norm": 13.0,
      "learning_rate": 3.787684315691539e-05,
      "loss": 0.9206,
      "num_input_tokens_seen": 30769352,
      "step": 53040
    },
    {
      "epoch": 7.900655347036044,
      "grad_norm": 9.5,
      "learning_rate": 3.787405781994584e-05,
      "loss": 1.0571,
      "num_input_tokens_seen": 30771944,
      "step": 53045
    },
    {
      "epoch": 7.9014000595770035,
      "grad_norm": 7.21875,
      "learning_rate": 3.7871272265482184e-05,
      "loss": 1.3541,
      "num_input_tokens_seen": 30775048,
      "step": 53050
    },
    {
      "epoch": 7.902144772117962,
      "grad_norm": 9.0,
      "learning_rate": 3.786848649357145e-05,
      "loss": 1.1302,
      "num_input_tokens_seen": 30777928,
      "step": 53055
    },
    {
      "epoch": 7.902889484658922,
      "grad_norm": 10.375,
      "learning_rate": 3.786570050426073e-05,
      "loss": 1.1287,
      "num_input_tokens_seen": 30780776,
      "step": 53060
    },
    {
      "epoch": 7.903634197199881,
      "grad_norm": 14.875,
      "learning_rate": 3.7862914297597075e-05,
      "loss": 1.1194,
      "num_input_tokens_seen": 30783624,
      "step": 53065
    },
    {
      "epoch": 7.90437890974084,
      "grad_norm": 5.0625,
      "learning_rate": 3.7860127873627546e-05,
      "loss": 1.2828,
      "num_input_tokens_seen": 30786312,
      "step": 53070
    },
    {
      "epoch": 7.905123622281799,
      "grad_norm": 9.125,
      "learning_rate": 3.785734123239924e-05,
      "loss": 0.7895,
      "num_input_tokens_seen": 30789320,
      "step": 53075
    },
    {
      "epoch": 7.905868334822759,
      "grad_norm": 11.0625,
      "learning_rate": 3.785455437395921e-05,
      "loss": 1.0994,
      "num_input_tokens_seen": 30792584,
      "step": 53080
    },
    {
      "epoch": 7.9066130473637175,
      "grad_norm": 12.4375,
      "learning_rate": 3.7851767298354554e-05,
      "loss": 1.1603,
      "num_input_tokens_seen": 30795176,
      "step": 53085
    },
    {
      "epoch": 7.907357759904677,
      "grad_norm": 9.8125,
      "learning_rate": 3.7848980005632344e-05,
      "loss": 1.3711,
      "num_input_tokens_seen": 30798184,
      "step": 53090
    },
    {
      "epoch": 7.908102472445636,
      "grad_norm": 9.5,
      "learning_rate": 3.7846192495839686e-05,
      "loss": 1.2735,
      "num_input_tokens_seen": 30801192,
      "step": 53095
    },
    {
      "epoch": 7.908847184986596,
      "grad_norm": 8.0625,
      "learning_rate": 3.784340476902366e-05,
      "loss": 0.8964,
      "num_input_tokens_seen": 30804168,
      "step": 53100
    },
    {
      "epoch": 7.909591897527554,
      "grad_norm": 4.875,
      "learning_rate": 3.7840616825231365e-05,
      "loss": 0.7692,
      "num_input_tokens_seen": 30807112,
      "step": 53105
    },
    {
      "epoch": 7.910336610068514,
      "grad_norm": 11.0625,
      "learning_rate": 3.783782866450989e-05,
      "loss": 1.1629,
      "num_input_tokens_seen": 30809768,
      "step": 53110
    },
    {
      "epoch": 7.911081322609473,
      "grad_norm": 11.1875,
      "learning_rate": 3.783504028690635e-05,
      "loss": 0.98,
      "num_input_tokens_seen": 30812552,
      "step": 53115
    },
    {
      "epoch": 7.9118260351504315,
      "grad_norm": 14.9375,
      "learning_rate": 3.783225169246786e-05,
      "loss": 0.774,
      "num_input_tokens_seen": 30815528,
      "step": 53120
    },
    {
      "epoch": 7.912570747691391,
      "grad_norm": 5.90625,
      "learning_rate": 3.782946288124151e-05,
      "loss": 1.1151,
      "num_input_tokens_seen": 30818504,
      "step": 53125
    },
    {
      "epoch": 7.913315460232351,
      "grad_norm": 4.21875,
      "learning_rate": 3.782667385327442e-05,
      "loss": 0.9205,
      "num_input_tokens_seen": 30821416,
      "step": 53130
    },
    {
      "epoch": 7.9140601727733095,
      "grad_norm": 11.5,
      "learning_rate": 3.782388460861372e-05,
      "loss": 1.0577,
      "num_input_tokens_seen": 30824360,
      "step": 53135
    },
    {
      "epoch": 7.914804885314268,
      "grad_norm": 5.5625,
      "learning_rate": 3.7821095147306527e-05,
      "loss": 1.1781,
      "num_input_tokens_seen": 30827432,
      "step": 53140
    },
    {
      "epoch": 7.915549597855228,
      "grad_norm": 15.4375,
      "learning_rate": 3.781830546939996e-05,
      "loss": 1.4831,
      "num_input_tokens_seen": 30830280,
      "step": 53145
    },
    {
      "epoch": 7.916294310396187,
      "grad_norm": 8.0,
      "learning_rate": 3.781551557494115e-05,
      "loss": 1.1763,
      "num_input_tokens_seen": 30832968,
      "step": 53150
    },
    {
      "epoch": 7.917039022937146,
      "grad_norm": 8.5625,
      "learning_rate": 3.7812725463977225e-05,
      "loss": 1.2919,
      "num_input_tokens_seen": 30835752,
      "step": 53155
    },
    {
      "epoch": 7.917783735478105,
      "grad_norm": 14.75,
      "learning_rate": 3.7809935136555326e-05,
      "loss": 0.9454,
      "num_input_tokens_seen": 30838824,
      "step": 53160
    },
    {
      "epoch": 7.918528448019065,
      "grad_norm": 13.625,
      "learning_rate": 3.780714459272259e-05,
      "loss": 1.3485,
      "num_input_tokens_seen": 30841864,
      "step": 53165
    },
    {
      "epoch": 7.9192731605600235,
      "grad_norm": 5.78125,
      "learning_rate": 3.780435383252617e-05,
      "loss": 1.326,
      "num_input_tokens_seen": 30844872,
      "step": 53170
    },
    {
      "epoch": 7.920017873100983,
      "grad_norm": 13.0625,
      "learning_rate": 3.78015628560132e-05,
      "loss": 1.0769,
      "num_input_tokens_seen": 30847784,
      "step": 53175
    },
    {
      "epoch": 7.920762585641942,
      "grad_norm": 9.625,
      "learning_rate": 3.779877166323084e-05,
      "loss": 1.1893,
      "num_input_tokens_seen": 30850696,
      "step": 53180
    },
    {
      "epoch": 7.921507298182902,
      "grad_norm": 9.9375,
      "learning_rate": 3.779598025422624e-05,
      "loss": 1.2277,
      "num_input_tokens_seen": 30853608,
      "step": 53185
    },
    {
      "epoch": 7.92225201072386,
      "grad_norm": 8.375,
      "learning_rate": 3.779318862904656e-05,
      "loss": 0.9854,
      "num_input_tokens_seen": 30856648,
      "step": 53190
    },
    {
      "epoch": 7.92299672326482,
      "grad_norm": 7.34375,
      "learning_rate": 3.779039678773896e-05,
      "loss": 0.858,
      "num_input_tokens_seen": 30859464,
      "step": 53195
    },
    {
      "epoch": 7.923741435805779,
      "grad_norm": 10.125,
      "learning_rate": 3.77876047303506e-05,
      "loss": 1.0976,
      "num_input_tokens_seen": 30862152,
      "step": 53200
    },
    {
      "epoch": 7.924486148346738,
      "grad_norm": 3.953125,
      "learning_rate": 3.778481245692866e-05,
      "loss": 0.938,
      "num_input_tokens_seen": 30865128,
      "step": 53205
    },
    {
      "epoch": 7.925230860887697,
      "grad_norm": 19.375,
      "learning_rate": 3.7782019967520305e-05,
      "loss": 1.3146,
      "num_input_tokens_seen": 30867976,
      "step": 53210
    },
    {
      "epoch": 7.925975573428657,
      "grad_norm": 19.0,
      "learning_rate": 3.777922726217271e-05,
      "loss": 1.0842,
      "num_input_tokens_seen": 30870856,
      "step": 53215
    },
    {
      "epoch": 7.9267202859696155,
      "grad_norm": 9.375,
      "learning_rate": 3.7776434340933065e-05,
      "loss": 1.06,
      "num_input_tokens_seen": 30873736,
      "step": 53220
    },
    {
      "epoch": 7.927464998510575,
      "grad_norm": 13.375,
      "learning_rate": 3.7773641203848554e-05,
      "loss": 1.0556,
      "num_input_tokens_seen": 30876712,
      "step": 53225
    },
    {
      "epoch": 7.928209711051534,
      "grad_norm": 10.625,
      "learning_rate": 3.7770847850966354e-05,
      "loss": 1.0947,
      "num_input_tokens_seen": 30879624,
      "step": 53230
    },
    {
      "epoch": 7.928954423592494,
      "grad_norm": 6.625,
      "learning_rate": 3.7768054282333655e-05,
      "loss": 1.3186,
      "num_input_tokens_seen": 30882088,
      "step": 53235
    },
    {
      "epoch": 7.929699136133452,
      "grad_norm": 13.625,
      "learning_rate": 3.776526049799765e-05,
      "loss": 0.8345,
      "num_input_tokens_seen": 30884680,
      "step": 53240
    },
    {
      "epoch": 7.930443848674412,
      "grad_norm": 13.1875,
      "learning_rate": 3.7762466498005544e-05,
      "loss": 1.1637,
      "num_input_tokens_seen": 30887784,
      "step": 53245
    },
    {
      "epoch": 7.931188561215371,
      "grad_norm": 8.4375,
      "learning_rate": 3.7759672282404546e-05,
      "loss": 0.7195,
      "num_input_tokens_seen": 30890440,
      "step": 53250
    },
    {
      "epoch": 7.93193327375633,
      "grad_norm": 12.8125,
      "learning_rate": 3.775687785124185e-05,
      "loss": 0.8347,
      "num_input_tokens_seen": 30893192,
      "step": 53255
    },
    {
      "epoch": 7.932677986297289,
      "grad_norm": 6.90625,
      "learning_rate": 3.775408320456466e-05,
      "loss": 1.067,
      "num_input_tokens_seen": 30896328,
      "step": 53260
    },
    {
      "epoch": 7.933422698838249,
      "grad_norm": 8.625,
      "learning_rate": 3.775128834242021e-05,
      "loss": 1.2612,
      "num_input_tokens_seen": 30899336,
      "step": 53265
    },
    {
      "epoch": 7.934167411379208,
      "grad_norm": 9.6875,
      "learning_rate": 3.77484932648557e-05,
      "loss": 0.9899,
      "num_input_tokens_seen": 30902248,
      "step": 53270
    },
    {
      "epoch": 7.934912123920167,
      "grad_norm": 10.6875,
      "learning_rate": 3.774569797191835e-05,
      "loss": 1.0245,
      "num_input_tokens_seen": 30905448,
      "step": 53275
    },
    {
      "epoch": 7.935656836461126,
      "grad_norm": 4.875,
      "learning_rate": 3.774290246365539e-05,
      "loss": 1.0293,
      "num_input_tokens_seen": 30908392,
      "step": 53280
    },
    {
      "epoch": 7.936401549002085,
      "grad_norm": 11.625,
      "learning_rate": 3.774010674011404e-05,
      "loss": 1.1767,
      "num_input_tokens_seen": 30911208,
      "step": 53285
    },
    {
      "epoch": 7.937146261543044,
      "grad_norm": 13.625,
      "learning_rate": 3.773731080134154e-05,
      "loss": 1.1685,
      "num_input_tokens_seen": 30914120,
      "step": 53290
    },
    {
      "epoch": 7.937890974084004,
      "grad_norm": 8.5625,
      "learning_rate": 3.7734514647385114e-05,
      "loss": 0.9702,
      "num_input_tokens_seen": 30917096,
      "step": 53295
    },
    {
      "epoch": 7.938635686624963,
      "grad_norm": 22.125,
      "learning_rate": 3.773171827829201e-05,
      "loss": 1.0407,
      "num_input_tokens_seen": 30920200,
      "step": 53300
    },
    {
      "epoch": 7.9393803991659215,
      "grad_norm": 14.625,
      "learning_rate": 3.772892169410947e-05,
      "loss": 0.9166,
      "num_input_tokens_seen": 30923368,
      "step": 53305
    },
    {
      "epoch": 7.940125111706881,
      "grad_norm": 8.875,
      "learning_rate": 3.772612489488473e-05,
      "loss": 0.9168,
      "num_input_tokens_seen": 30926056,
      "step": 53310
    },
    {
      "epoch": 7.940869824247841,
      "grad_norm": 16.25,
      "learning_rate": 3.772332788066504e-05,
      "loss": 1.2334,
      "num_input_tokens_seen": 30929032,
      "step": 53315
    },
    {
      "epoch": 7.9416145367888,
      "grad_norm": 6.78125,
      "learning_rate": 3.772053065149766e-05,
      "loss": 1.2068,
      "num_input_tokens_seen": 30931784,
      "step": 53320
    },
    {
      "epoch": 7.942359249329758,
      "grad_norm": 10.3125,
      "learning_rate": 3.771773320742984e-05,
      "loss": 1.1741,
      "num_input_tokens_seen": 30934600,
      "step": 53325
    },
    {
      "epoch": 7.943103961870718,
      "grad_norm": 9.6875,
      "learning_rate": 3.7714935548508846e-05,
      "loss": 1.1041,
      "num_input_tokens_seen": 30937352,
      "step": 53330
    },
    {
      "epoch": 7.943848674411677,
      "grad_norm": 7.5,
      "learning_rate": 3.771213767478194e-05,
      "loss": 1.3524,
      "num_input_tokens_seen": 30940264,
      "step": 53335
    },
    {
      "epoch": 7.944593386952636,
      "grad_norm": 13.5,
      "learning_rate": 3.770933958629639e-05,
      "loss": 1.2949,
      "num_input_tokens_seen": 30942888,
      "step": 53340
    },
    {
      "epoch": 7.945338099493595,
      "grad_norm": 6.84375,
      "learning_rate": 3.7706541283099466e-05,
      "loss": 1.1143,
      "num_input_tokens_seen": 30945672,
      "step": 53345
    },
    {
      "epoch": 7.946082812034555,
      "grad_norm": 12.4375,
      "learning_rate": 3.7703742765238436e-05,
      "loss": 1.1296,
      "num_input_tokens_seen": 30948520,
      "step": 53350
    },
    {
      "epoch": 7.946827524575514,
      "grad_norm": 6.46875,
      "learning_rate": 3.770094403276059e-05,
      "loss": 0.8872,
      "num_input_tokens_seen": 30951528,
      "step": 53355
    },
    {
      "epoch": 7.947572237116473,
      "grad_norm": 7.375,
      "learning_rate": 3.7698145085713196e-05,
      "loss": 0.864,
      "num_input_tokens_seen": 30954408,
      "step": 53360
    },
    {
      "epoch": 7.948316949657432,
      "grad_norm": 14.375,
      "learning_rate": 3.7695345924143555e-05,
      "loss": 1.1104,
      "num_input_tokens_seen": 30957352,
      "step": 53365
    },
    {
      "epoch": 7.949061662198392,
      "grad_norm": 12.5,
      "learning_rate": 3.769254654809894e-05,
      "loss": 1.0994,
      "num_input_tokens_seen": 30960168,
      "step": 53370
    },
    {
      "epoch": 7.94980637473935,
      "grad_norm": 9.1875,
      "learning_rate": 3.768974695762665e-05,
      "loss": 1.2747,
      "num_input_tokens_seen": 30963208,
      "step": 53375
    },
    {
      "epoch": 7.95055108728031,
      "grad_norm": 17.375,
      "learning_rate": 3.768694715277398e-05,
      "loss": 1.0436,
      "num_input_tokens_seen": 30965928,
      "step": 53380
    },
    {
      "epoch": 7.951295799821269,
      "grad_norm": 15.375,
      "learning_rate": 3.7684147133588245e-05,
      "loss": 1.1985,
      "num_input_tokens_seen": 30968936,
      "step": 53385
    },
    {
      "epoch": 7.952040512362228,
      "grad_norm": 14.25,
      "learning_rate": 3.7681346900116726e-05,
      "loss": 0.8514,
      "num_input_tokens_seen": 30971688,
      "step": 53390
    },
    {
      "epoch": 7.952785224903187,
      "grad_norm": 3.984375,
      "learning_rate": 3.7678546452406736e-05,
      "loss": 0.9853,
      "num_input_tokens_seen": 30974504,
      "step": 53395
    },
    {
      "epoch": 7.953529937444147,
      "grad_norm": 10.6875,
      "learning_rate": 3.76757457905056e-05,
      "loss": 1.0178,
      "num_input_tokens_seen": 30977320,
      "step": 53400
    },
    {
      "epoch": 7.954274649985106,
      "grad_norm": 8.25,
      "learning_rate": 3.767294491446062e-05,
      "loss": 1.2379,
      "num_input_tokens_seen": 30980136,
      "step": 53405
    },
    {
      "epoch": 7.955019362526065,
      "grad_norm": 10.625,
      "learning_rate": 3.7670143824319116e-05,
      "loss": 1.0584,
      "num_input_tokens_seen": 30982824,
      "step": 53410
    },
    {
      "epoch": 7.955764075067024,
      "grad_norm": 12.9375,
      "learning_rate": 3.76673425201284e-05,
      "loss": 1.0943,
      "num_input_tokens_seen": 30985672,
      "step": 53415
    },
    {
      "epoch": 7.956508787607984,
      "grad_norm": 10.75,
      "learning_rate": 3.766454100193581e-05,
      "loss": 0.9539,
      "num_input_tokens_seen": 30988552,
      "step": 53420
    },
    {
      "epoch": 7.957253500148942,
      "grad_norm": 12.1875,
      "learning_rate": 3.7661739269788687e-05,
      "loss": 1.1404,
      "num_input_tokens_seen": 30991432,
      "step": 53425
    },
    {
      "epoch": 7.957998212689902,
      "grad_norm": 15.75,
      "learning_rate": 3.765893732373433e-05,
      "loss": 0.921,
      "num_input_tokens_seen": 30994376,
      "step": 53430
    },
    {
      "epoch": 7.958742925230861,
      "grad_norm": 8.5,
      "learning_rate": 3.7656135163820105e-05,
      "loss": 1.063,
      "num_input_tokens_seen": 30997608,
      "step": 53435
    },
    {
      "epoch": 7.9594876377718204,
      "grad_norm": 11.75,
      "learning_rate": 3.7653332790093334e-05,
      "loss": 1.3858,
      "num_input_tokens_seen": 31000648,
      "step": 53440
    },
    {
      "epoch": 7.960232350312779,
      "grad_norm": 13.0625,
      "learning_rate": 3.765053020260137e-05,
      "loss": 1.1925,
      "num_input_tokens_seen": 31003400,
      "step": 53445
    },
    {
      "epoch": 7.960977062853738,
      "grad_norm": 12.625,
      "learning_rate": 3.764772740139154e-05,
      "loss": 1.2152,
      "num_input_tokens_seen": 31006248,
      "step": 53450
    },
    {
      "epoch": 7.961721775394698,
      "grad_norm": 12.6875,
      "learning_rate": 3.7644924386511225e-05,
      "loss": 1.5128,
      "num_input_tokens_seen": 31009128,
      "step": 53455
    },
    {
      "epoch": 7.962466487935657,
      "grad_norm": 12.75,
      "learning_rate": 3.7642121158007756e-05,
      "loss": 1.418,
      "num_input_tokens_seen": 31011944,
      "step": 53460
    },
    {
      "epoch": 7.963211200476616,
      "grad_norm": 6.84375,
      "learning_rate": 3.7639317715928514e-05,
      "loss": 1.0621,
      "num_input_tokens_seen": 31014696,
      "step": 53465
    },
    {
      "epoch": 7.963955913017575,
      "grad_norm": 7.09375,
      "learning_rate": 3.763651406032083e-05,
      "loss": 1.0537,
      "num_input_tokens_seen": 31017704,
      "step": 53470
    },
    {
      "epoch": 7.964700625558534,
      "grad_norm": 13.5,
      "learning_rate": 3.763371019123209e-05,
      "loss": 1.6796,
      "num_input_tokens_seen": 31020648,
      "step": 53475
    },
    {
      "epoch": 7.965445338099494,
      "grad_norm": 10.1875,
      "learning_rate": 3.7630906108709654e-05,
      "loss": 1.1511,
      "num_input_tokens_seen": 31023528,
      "step": 53480
    },
    {
      "epoch": 7.966190050640453,
      "grad_norm": 12.75,
      "learning_rate": 3.76281018128009e-05,
      "loss": 1.3226,
      "num_input_tokens_seen": 31026056,
      "step": 53485
    },
    {
      "epoch": 7.966934763181412,
      "grad_norm": 11.5,
      "learning_rate": 3.7625297303553195e-05,
      "loss": 0.9168,
      "num_input_tokens_seen": 31028808,
      "step": 53490
    },
    {
      "epoch": 7.967679475722371,
      "grad_norm": 12.25,
      "learning_rate": 3.762249258101392e-05,
      "loss": 0.9805,
      "num_input_tokens_seen": 31031688,
      "step": 53495
    },
    {
      "epoch": 7.96842418826333,
      "grad_norm": 21.5,
      "learning_rate": 3.761968764523048e-05,
      "loss": 1.2888,
      "num_input_tokens_seen": 31034248,
      "step": 53500
    },
    {
      "epoch": 7.96916890080429,
      "grad_norm": 8.6875,
      "learning_rate": 3.761688249625024e-05,
      "loss": 0.8345,
      "num_input_tokens_seen": 31037320,
      "step": 53505
    },
    {
      "epoch": 7.969913613345248,
      "grad_norm": 17.875,
      "learning_rate": 3.761407713412058e-05,
      "loss": 1.2199,
      "num_input_tokens_seen": 31040200,
      "step": 53510
    },
    {
      "epoch": 7.970658325886208,
      "grad_norm": 15.25,
      "learning_rate": 3.761127155888891e-05,
      "loss": 1.1762,
      "num_input_tokens_seen": 31042952,
      "step": 53515
    },
    {
      "epoch": 7.971403038427167,
      "grad_norm": 13.8125,
      "learning_rate": 3.760846577060263e-05,
      "loss": 1.0408,
      "num_input_tokens_seen": 31045832,
      "step": 53520
    },
    {
      "epoch": 7.9721477509681264,
      "grad_norm": 9.5,
      "learning_rate": 3.760565976930913e-05,
      "loss": 1.0792,
      "num_input_tokens_seen": 31049032,
      "step": 53525
    },
    {
      "epoch": 7.972892463509085,
      "grad_norm": 6.65625,
      "learning_rate": 3.760285355505583e-05,
      "loss": 1.0869,
      "num_input_tokens_seen": 31051944,
      "step": 53530
    },
    {
      "epoch": 7.973637176050045,
      "grad_norm": 11.25,
      "learning_rate": 3.760004712789012e-05,
      "loss": 1.1217,
      "num_input_tokens_seen": 31054536,
      "step": 53535
    },
    {
      "epoch": 7.974381888591004,
      "grad_norm": 15.5625,
      "learning_rate": 3.759724048785942e-05,
      "loss": 1.0547,
      "num_input_tokens_seen": 31057288,
      "step": 53540
    },
    {
      "epoch": 7.975126601131963,
      "grad_norm": 21.875,
      "learning_rate": 3.759443363501115e-05,
      "loss": 1.2198,
      "num_input_tokens_seen": 31060264,
      "step": 53545
    },
    {
      "epoch": 7.975871313672922,
      "grad_norm": 9.625,
      "learning_rate": 3.759162656939271e-05,
      "loss": 0.9725,
      "num_input_tokens_seen": 31062920,
      "step": 53550
    },
    {
      "epoch": 7.976616026213882,
      "grad_norm": 9.875,
      "learning_rate": 3.758881929105155e-05,
      "loss": 1.0783,
      "num_input_tokens_seen": 31065704,
      "step": 53555
    },
    {
      "epoch": 7.97736073875484,
      "grad_norm": 17.375,
      "learning_rate": 3.758601180003508e-05,
      "loss": 1.0941,
      "num_input_tokens_seen": 31069064,
      "step": 53560
    },
    {
      "epoch": 7.9781054512958,
      "grad_norm": 12.8125,
      "learning_rate": 3.758320409639074e-05,
      "loss": 1.2654,
      "num_input_tokens_seen": 31071912,
      "step": 53565
    },
    {
      "epoch": 7.978850163836759,
      "grad_norm": 7.1875,
      "learning_rate": 3.758039618016595e-05,
      "loss": 1.3303,
      "num_input_tokens_seen": 31074568,
      "step": 53570
    },
    {
      "epoch": 7.9795948763777185,
      "grad_norm": 6.6875,
      "learning_rate": 3.757758805140814e-05,
      "loss": 1.0986,
      "num_input_tokens_seen": 31077480,
      "step": 53575
    },
    {
      "epoch": 7.980339588918677,
      "grad_norm": 5.28125,
      "learning_rate": 3.757477971016478e-05,
      "loss": 0.9319,
      "num_input_tokens_seen": 31080296,
      "step": 53580
    },
    {
      "epoch": 7.981084301459637,
      "grad_norm": 13.8125,
      "learning_rate": 3.7571971156483285e-05,
      "loss": 1.1265,
      "num_input_tokens_seen": 31083048,
      "step": 53585
    },
    {
      "epoch": 7.981829014000596,
      "grad_norm": 15.3125,
      "learning_rate": 3.756916239041113e-05,
      "loss": 1.0759,
      "num_input_tokens_seen": 31085768,
      "step": 53590
    },
    {
      "epoch": 7.982573726541555,
      "grad_norm": 6.375,
      "learning_rate": 3.756635341199574e-05,
      "loss": 1.1443,
      "num_input_tokens_seen": 31088392,
      "step": 53595
    },
    {
      "epoch": 7.983318439082514,
      "grad_norm": 8.875,
      "learning_rate": 3.756354422128459e-05,
      "loss": 1.0948,
      "num_input_tokens_seen": 31091368,
      "step": 53600
    },
    {
      "epoch": 7.984063151623474,
      "grad_norm": 11.625,
      "learning_rate": 3.756073481832512e-05,
      "loss": 1.2313,
      "num_input_tokens_seen": 31094472,
      "step": 53605
    },
    {
      "epoch": 7.9848078641644324,
      "grad_norm": 7.28125,
      "learning_rate": 3.75579252031648e-05,
      "loss": 1.1995,
      "num_input_tokens_seen": 31097352,
      "step": 53610
    },
    {
      "epoch": 7.985552576705392,
      "grad_norm": 10.0,
      "learning_rate": 3.75551153758511e-05,
      "loss": 1.131,
      "num_input_tokens_seen": 31100168,
      "step": 53615
    },
    {
      "epoch": 7.986297289246351,
      "grad_norm": 11.5625,
      "learning_rate": 3.755230533643148e-05,
      "loss": 1.4357,
      "num_input_tokens_seen": 31103272,
      "step": 53620
    },
    {
      "epoch": 7.9870420017873105,
      "grad_norm": 11.4375,
      "learning_rate": 3.754949508495344e-05,
      "loss": 1.141,
      "num_input_tokens_seen": 31106216,
      "step": 53625
    },
    {
      "epoch": 7.987786714328269,
      "grad_norm": 9.5,
      "learning_rate": 3.7546684621464415e-05,
      "loss": 1.0397,
      "num_input_tokens_seen": 31109064,
      "step": 53630
    },
    {
      "epoch": 7.988531426869228,
      "grad_norm": 13.875,
      "learning_rate": 3.7543873946011916e-05,
      "loss": 1.4239,
      "num_input_tokens_seen": 31111880,
      "step": 53635
    },
    {
      "epoch": 7.989276139410188,
      "grad_norm": 9.8125,
      "learning_rate": 3.754106305864341e-05,
      "loss": 0.9349,
      "num_input_tokens_seen": 31114760,
      "step": 53640
    },
    {
      "epoch": 7.990020851951147,
      "grad_norm": 8.3125,
      "learning_rate": 3.753825195940639e-05,
      "loss": 1.0431,
      "num_input_tokens_seen": 31117800,
      "step": 53645
    },
    {
      "epoch": 7.990765564492106,
      "grad_norm": 10.6875,
      "learning_rate": 3.753544064834835e-05,
      "loss": 1.168,
      "num_input_tokens_seen": 31120616,
      "step": 53650
    },
    {
      "epoch": 7.991510277033065,
      "grad_norm": 7.71875,
      "learning_rate": 3.753262912551677e-05,
      "loss": 1.1903,
      "num_input_tokens_seen": 31123432,
      "step": 53655
    },
    {
      "epoch": 7.9922549895740245,
      "grad_norm": 13.125,
      "learning_rate": 3.7529817390959164e-05,
      "loss": 1.3672,
      "num_input_tokens_seen": 31126216,
      "step": 53660
    },
    {
      "epoch": 7.992999702114983,
      "grad_norm": 10.4375,
      "learning_rate": 3.752700544472304e-05,
      "loss": 0.9498,
      "num_input_tokens_seen": 31128968,
      "step": 53665
    },
    {
      "epoch": 7.993744414655943,
      "grad_norm": 12.75,
      "learning_rate": 3.752419328685588e-05,
      "loss": 1.0177,
      "num_input_tokens_seen": 31131880,
      "step": 53670
    },
    {
      "epoch": 7.994489127196902,
      "grad_norm": 12.25,
      "learning_rate": 3.752138091740521e-05,
      "loss": 1.1278,
      "num_input_tokens_seen": 31134856,
      "step": 53675
    },
    {
      "epoch": 7.995233839737861,
      "grad_norm": 13.0,
      "learning_rate": 3.7518568336418525e-05,
      "loss": 1.0955,
      "num_input_tokens_seen": 31137544,
      "step": 53680
    },
    {
      "epoch": 7.99597855227882,
      "grad_norm": 9.6875,
      "learning_rate": 3.751575554394336e-05,
      "loss": 1.0589,
      "num_input_tokens_seen": 31140296,
      "step": 53685
    },
    {
      "epoch": 7.99672326481978,
      "grad_norm": 17.0,
      "learning_rate": 3.751294254002722e-05,
      "loss": 0.9219,
      "num_input_tokens_seen": 31143080,
      "step": 53690
    },
    {
      "epoch": 7.9974679773607384,
      "grad_norm": 9.25,
      "learning_rate": 3.751012932471764e-05,
      "loss": 1.0864,
      "num_input_tokens_seen": 31145800,
      "step": 53695
    },
    {
      "epoch": 7.998212689901698,
      "grad_norm": 20.625,
      "learning_rate": 3.7507315898062136e-05,
      "loss": 1.1815,
      "num_input_tokens_seen": 31148456,
      "step": 53700
    },
    {
      "epoch": 7.998957402442657,
      "grad_norm": 10.8125,
      "learning_rate": 3.7504502260108245e-05,
      "loss": 1.234,
      "num_input_tokens_seen": 31151336,
      "step": 53705
    },
    {
      "epoch": 7.9997021149836165,
      "grad_norm": 12.5,
      "learning_rate": 3.750168841090349e-05,
      "loss": 1.0843,
      "num_input_tokens_seen": 31154216,
      "step": 53710
    },
    {
      "epoch": 8.0,
      "eval_loss": 1.0823091268539429,
      "eval_runtime": 43.7486,
      "eval_samples_per_second": 68.208,
      "eval_steps_per_second": 17.052,
      "num_input_tokens_seen": 31154960,
      "step": 53712
    },
    {
      "epoch": 8.000446827524575,
      "grad_norm": 8.9375,
      "learning_rate": 3.749887435049541e-05,
      "loss": 0.8657,
      "num_input_tokens_seen": 31156592,
      "step": 53715
    },
    {
      "epoch": 8.001191540065534,
      "grad_norm": 14.5,
      "learning_rate": 3.749606007893157e-05,
      "loss": 1.3807,
      "num_input_tokens_seen": 31159344,
      "step": 53720
    },
    {
      "epoch": 8.001936252606495,
      "grad_norm": 11.4375,
      "learning_rate": 3.7493245596259484e-05,
      "loss": 1.2726,
      "num_input_tokens_seen": 31162032,
      "step": 53725
    },
    {
      "epoch": 8.002680965147453,
      "grad_norm": 11.4375,
      "learning_rate": 3.7490430902526715e-05,
      "loss": 0.9915,
      "num_input_tokens_seen": 31164784,
      "step": 53730
    },
    {
      "epoch": 8.003425677688412,
      "grad_norm": 6.75,
      "learning_rate": 3.7487615997780815e-05,
      "loss": 0.82,
      "num_input_tokens_seen": 31167824,
      "step": 53735
    },
    {
      "epoch": 8.00417039022937,
      "grad_norm": 6.125,
      "learning_rate": 3.7484800882069324e-05,
      "loss": 1.246,
      "num_input_tokens_seen": 31170704,
      "step": 53740
    },
    {
      "epoch": 8.004915102770331,
      "grad_norm": 11.125,
      "learning_rate": 3.748198555543981e-05,
      "loss": 1.1124,
      "num_input_tokens_seen": 31173456,
      "step": 53745
    },
    {
      "epoch": 8.00565981531129,
      "grad_norm": 11.125,
      "learning_rate": 3.747917001793985e-05,
      "loss": 0.8179,
      "num_input_tokens_seen": 31176400,
      "step": 53750
    },
    {
      "epoch": 8.006404527852249,
      "grad_norm": 11.0625,
      "learning_rate": 3.7476354269616984e-05,
      "loss": 1.0257,
      "num_input_tokens_seen": 31179536,
      "step": 53755
    },
    {
      "epoch": 8.007149240393208,
      "grad_norm": 13.6875,
      "learning_rate": 3.747353831051879e-05,
      "loss": 1.0673,
      "num_input_tokens_seen": 31182192,
      "step": 53760
    },
    {
      "epoch": 8.007893952934168,
      "grad_norm": 8.0625,
      "learning_rate": 3.747072214069286e-05,
      "loss": 0.8284,
      "num_input_tokens_seen": 31185232,
      "step": 53765
    },
    {
      "epoch": 8.008638665475127,
      "grad_norm": 9.75,
      "learning_rate": 3.746790576018674e-05,
      "loss": 0.9992,
      "num_input_tokens_seen": 31188400,
      "step": 53770
    },
    {
      "epoch": 8.009383378016086,
      "grad_norm": 11.25,
      "learning_rate": 3.746508916904803e-05,
      "loss": 1.4332,
      "num_input_tokens_seen": 31191728,
      "step": 53775
    },
    {
      "epoch": 8.010128090557044,
      "grad_norm": 8.625,
      "learning_rate": 3.74622723673243e-05,
      "loss": 1.2244,
      "num_input_tokens_seen": 31194576,
      "step": 53780
    },
    {
      "epoch": 8.010872803098005,
      "grad_norm": 12.3125,
      "learning_rate": 3.745945535506315e-05,
      "loss": 1.0316,
      "num_input_tokens_seen": 31197584,
      "step": 53785
    },
    {
      "epoch": 8.011617515638964,
      "grad_norm": 12.25,
      "learning_rate": 3.7456638132312164e-05,
      "loss": 1.3824,
      "num_input_tokens_seen": 31200304,
      "step": 53790
    },
    {
      "epoch": 8.012362228179922,
      "grad_norm": 13.9375,
      "learning_rate": 3.745382069911894e-05,
      "loss": 1.3434,
      "num_input_tokens_seen": 31203248,
      "step": 53795
    },
    {
      "epoch": 8.013106940720881,
      "grad_norm": 14.0625,
      "learning_rate": 3.745100305553107e-05,
      "loss": 1.0839,
      "num_input_tokens_seen": 31206320,
      "step": 53800
    },
    {
      "epoch": 8.013851653261842,
      "grad_norm": 5.5,
      "learning_rate": 3.744818520159616e-05,
      "loss": 1.497,
      "num_input_tokens_seen": 31209072,
      "step": 53805
    },
    {
      "epoch": 8.0145963658028,
      "grad_norm": 6.03125,
      "learning_rate": 3.744536713736182e-05,
      "loss": 1.1415,
      "num_input_tokens_seen": 31212016,
      "step": 53810
    },
    {
      "epoch": 8.01534107834376,
      "grad_norm": 5.9375,
      "learning_rate": 3.744254886287564e-05,
      "loss": 0.814,
      "num_input_tokens_seen": 31214960,
      "step": 53815
    },
    {
      "epoch": 8.016085790884718,
      "grad_norm": 8.4375,
      "learning_rate": 3.743973037818524e-05,
      "loss": 1.1344,
      "num_input_tokens_seen": 31218160,
      "step": 53820
    },
    {
      "epoch": 8.016830503425677,
      "grad_norm": 13.9375,
      "learning_rate": 3.7436911683338244e-05,
      "loss": 0.9934,
      "num_input_tokens_seen": 31221264,
      "step": 53825
    },
    {
      "epoch": 8.017575215966637,
      "grad_norm": 5.25,
      "learning_rate": 3.743409277838227e-05,
      "loss": 1.1786,
      "num_input_tokens_seen": 31224272,
      "step": 53830
    },
    {
      "epoch": 8.018319928507596,
      "grad_norm": 7.5,
      "learning_rate": 3.7431273663364926e-05,
      "loss": 0.9548,
      "num_input_tokens_seen": 31227440,
      "step": 53835
    },
    {
      "epoch": 8.019064641048555,
      "grad_norm": 6.75,
      "learning_rate": 3.742845433833386e-05,
      "loss": 1.111,
      "num_input_tokens_seen": 31230160,
      "step": 53840
    },
    {
      "epoch": 8.019809353589514,
      "grad_norm": 10.5,
      "learning_rate": 3.742563480333668e-05,
      "loss": 1.0955,
      "num_input_tokens_seen": 31233136,
      "step": 53845
    },
    {
      "epoch": 8.020554066130474,
      "grad_norm": 19.0,
      "learning_rate": 3.742281505842103e-05,
      "loss": 1.1465,
      "num_input_tokens_seen": 31235760,
      "step": 53850
    },
    {
      "epoch": 8.021298778671433,
      "grad_norm": 13.9375,
      "learning_rate": 3.7419995103634546e-05,
      "loss": 1.0696,
      "num_input_tokens_seen": 31238416,
      "step": 53855
    },
    {
      "epoch": 8.022043491212392,
      "grad_norm": 8.5625,
      "learning_rate": 3.741717493902488e-05,
      "loss": 0.8229,
      "num_input_tokens_seen": 31241552,
      "step": 53860
    },
    {
      "epoch": 8.02278820375335,
      "grad_norm": 11.9375,
      "learning_rate": 3.741435456463965e-05,
      "loss": 1.1422,
      "num_input_tokens_seen": 31244208,
      "step": 53865
    },
    {
      "epoch": 8.023532916294311,
      "grad_norm": 9.625,
      "learning_rate": 3.741153398052653e-05,
      "loss": 1.2075,
      "num_input_tokens_seen": 31246896,
      "step": 53870
    },
    {
      "epoch": 8.02427762883527,
      "grad_norm": 16.0,
      "learning_rate": 3.740871318673314e-05,
      "loss": 1.0484,
      "num_input_tokens_seen": 31249776,
      "step": 53875
    },
    {
      "epoch": 8.025022341376228,
      "grad_norm": 6.875,
      "learning_rate": 3.740589218330716e-05,
      "loss": 1.068,
      "num_input_tokens_seen": 31253072,
      "step": 53880
    },
    {
      "epoch": 8.025767053917187,
      "grad_norm": 4.46875,
      "learning_rate": 3.740307097029624e-05,
      "loss": 1.0043,
      "num_input_tokens_seen": 31255920,
      "step": 53885
    },
    {
      "epoch": 8.026511766458148,
      "grad_norm": 5.65625,
      "learning_rate": 3.740024954774804e-05,
      "loss": 1.0124,
      "num_input_tokens_seen": 31258992,
      "step": 53890
    },
    {
      "epoch": 8.027256478999107,
      "grad_norm": 14.125,
      "learning_rate": 3.739742791571023e-05,
      "loss": 1.0604,
      "num_input_tokens_seen": 31261680,
      "step": 53895
    },
    {
      "epoch": 8.028001191540065,
      "grad_norm": 9.875,
      "learning_rate": 3.739460607423048e-05,
      "loss": 1.0427,
      "num_input_tokens_seen": 31264848,
      "step": 53900
    },
    {
      "epoch": 8.028745904081024,
      "grad_norm": 12.25,
      "learning_rate": 3.7391784023356445e-05,
      "loss": 0.9664,
      "num_input_tokens_seen": 31268112,
      "step": 53905
    },
    {
      "epoch": 8.029490616621985,
      "grad_norm": 9.75,
      "learning_rate": 3.7388961763135835e-05,
      "loss": 1.1911,
      "num_input_tokens_seen": 31271280,
      "step": 53910
    },
    {
      "epoch": 8.030235329162943,
      "grad_norm": 9.75,
      "learning_rate": 3.7386139293616285e-05,
      "loss": 1.0775,
      "num_input_tokens_seen": 31274096,
      "step": 53915
    },
    {
      "epoch": 8.030980041703902,
      "grad_norm": 9.5625,
      "learning_rate": 3.738331661484551e-05,
      "loss": 0.8698,
      "num_input_tokens_seen": 31276880,
      "step": 53920
    },
    {
      "epoch": 8.03172475424486,
      "grad_norm": 10.8125,
      "learning_rate": 3.7380493726871186e-05,
      "loss": 1.1448,
      "num_input_tokens_seen": 31279792,
      "step": 53925
    },
    {
      "epoch": 8.032469466785821,
      "grad_norm": 7.59375,
      "learning_rate": 3.737767062974101e-05,
      "loss": 1.5594,
      "num_input_tokens_seen": 31283120,
      "step": 53930
    },
    {
      "epoch": 8.03321417932678,
      "grad_norm": 10.375,
      "learning_rate": 3.737484732350266e-05,
      "loss": 1.0146,
      "num_input_tokens_seen": 31286288,
      "step": 53935
    },
    {
      "epoch": 8.033958891867739,
      "grad_norm": 13.0625,
      "learning_rate": 3.7372023808203836e-05,
      "loss": 0.8012,
      "num_input_tokens_seen": 31289040,
      "step": 53940
    },
    {
      "epoch": 8.034703604408698,
      "grad_norm": 12.8125,
      "learning_rate": 3.736920008389225e-05,
      "loss": 0.9557,
      "num_input_tokens_seen": 31292144,
      "step": 53945
    },
    {
      "epoch": 8.035448316949658,
      "grad_norm": 18.25,
      "learning_rate": 3.7366376150615614e-05,
      "loss": 1.4414,
      "num_input_tokens_seen": 31294928,
      "step": 53950
    },
    {
      "epoch": 8.036193029490617,
      "grad_norm": 8.5625,
      "learning_rate": 3.7363552008421606e-05,
      "loss": 0.8714,
      "num_input_tokens_seen": 31297712,
      "step": 53955
    },
    {
      "epoch": 8.036937742031576,
      "grad_norm": 13.0625,
      "learning_rate": 3.7360727657357954e-05,
      "loss": 1.1052,
      "num_input_tokens_seen": 31300528,
      "step": 53960
    },
    {
      "epoch": 8.037682454572534,
      "grad_norm": 22.75,
      "learning_rate": 3.7357903097472376e-05,
      "loss": 1.162,
      "num_input_tokens_seen": 31303472,
      "step": 53965
    },
    {
      "epoch": 8.038427167113495,
      "grad_norm": 9.125,
      "learning_rate": 3.7355078328812583e-05,
      "loss": 1.2509,
      "num_input_tokens_seen": 31306544,
      "step": 53970
    },
    {
      "epoch": 8.039171879654454,
      "grad_norm": 11.375,
      "learning_rate": 3.7352253351426295e-05,
      "loss": 1.3636,
      "num_input_tokens_seen": 31309392,
      "step": 53975
    },
    {
      "epoch": 8.039916592195413,
      "grad_norm": 5.5,
      "learning_rate": 3.734942816536124e-05,
      "loss": 1.0706,
      "num_input_tokens_seen": 31311920,
      "step": 53980
    },
    {
      "epoch": 8.040661304736371,
      "grad_norm": 8.625,
      "learning_rate": 3.734660277066515e-05,
      "loss": 1.1813,
      "num_input_tokens_seen": 31314640,
      "step": 53985
    },
    {
      "epoch": 8.041406017277332,
      "grad_norm": 13.0,
      "learning_rate": 3.734377716738576e-05,
      "loss": 1.2988,
      "num_input_tokens_seen": 31317424,
      "step": 53990
    },
    {
      "epoch": 8.04215072981829,
      "grad_norm": 11.9375,
      "learning_rate": 3.73409513555708e-05,
      "loss": 1.1319,
      "num_input_tokens_seen": 31320752,
      "step": 53995
    },
    {
      "epoch": 8.04289544235925,
      "grad_norm": 10.6875,
      "learning_rate": 3.733812533526801e-05,
      "loss": 1.007,
      "num_input_tokens_seen": 31323728,
      "step": 54000
    },
    {
      "epoch": 8.043640154900208,
      "grad_norm": 10.1875,
      "learning_rate": 3.733529910652513e-05,
      "loss": 1.1916,
      "num_input_tokens_seen": 31326992,
      "step": 54005
    },
    {
      "epoch": 8.044384867441167,
      "grad_norm": 9.4375,
      "learning_rate": 3.73324726693899e-05,
      "loss": 0.8556,
      "num_input_tokens_seen": 31329872,
      "step": 54010
    },
    {
      "epoch": 8.045129579982127,
      "grad_norm": 6.53125,
      "learning_rate": 3.732964602391009e-05,
      "loss": 0.6773,
      "num_input_tokens_seen": 31332688,
      "step": 54015
    },
    {
      "epoch": 8.045874292523086,
      "grad_norm": 15.625,
      "learning_rate": 3.7326819170133434e-05,
      "loss": 1.2358,
      "num_input_tokens_seen": 31335824,
      "step": 54020
    },
    {
      "epoch": 8.046619005064045,
      "grad_norm": 7.46875,
      "learning_rate": 3.7323992108107705e-05,
      "loss": 1.1002,
      "num_input_tokens_seen": 31338704,
      "step": 54025
    },
    {
      "epoch": 8.047363717605004,
      "grad_norm": 9.0,
      "learning_rate": 3.7321164837880654e-05,
      "loss": 1.0453,
      "num_input_tokens_seen": 31341296,
      "step": 54030
    },
    {
      "epoch": 8.048108430145964,
      "grad_norm": 10.1875,
      "learning_rate": 3.731833735950004e-05,
      "loss": 1.231,
      "num_input_tokens_seen": 31343888,
      "step": 54035
    },
    {
      "epoch": 8.048853142686923,
      "grad_norm": 9.8125,
      "learning_rate": 3.731550967301364e-05,
      "loss": 1.0302,
      "num_input_tokens_seen": 31346928,
      "step": 54040
    },
    {
      "epoch": 8.049597855227882,
      "grad_norm": 11.1875,
      "learning_rate": 3.7312681778469216e-05,
      "loss": 1.3655,
      "num_input_tokens_seen": 31350064,
      "step": 54045
    },
    {
      "epoch": 8.05034256776884,
      "grad_norm": 6.09375,
      "learning_rate": 3.730985367591455e-05,
      "loss": 1.2119,
      "num_input_tokens_seen": 31353008,
      "step": 54050
    },
    {
      "epoch": 8.051087280309801,
      "grad_norm": 6.78125,
      "learning_rate": 3.730702536539741e-05,
      "loss": 0.892,
      "num_input_tokens_seen": 31355696,
      "step": 54055
    },
    {
      "epoch": 8.05183199285076,
      "grad_norm": 7.5,
      "learning_rate": 3.73041968469656e-05,
      "loss": 0.8382,
      "num_input_tokens_seen": 31358576,
      "step": 54060
    },
    {
      "epoch": 8.052576705391719,
      "grad_norm": 16.625,
      "learning_rate": 3.730136812066688e-05,
      "loss": 1.1064,
      "num_input_tokens_seen": 31361776,
      "step": 54065
    },
    {
      "epoch": 8.053321417932677,
      "grad_norm": 9.875,
      "learning_rate": 3.7298539186549054e-05,
      "loss": 1.1282,
      "num_input_tokens_seen": 31364688,
      "step": 54070
    },
    {
      "epoch": 8.054066130473638,
      "grad_norm": 16.5,
      "learning_rate": 3.7295710044659904e-05,
      "loss": 1.0101,
      "num_input_tokens_seen": 31367792,
      "step": 54075
    },
    {
      "epoch": 8.054810843014597,
      "grad_norm": 9.875,
      "learning_rate": 3.7292880695047225e-05,
      "loss": 1.1327,
      "num_input_tokens_seen": 31370352,
      "step": 54080
    },
    {
      "epoch": 8.055555555555555,
      "grad_norm": 10.0625,
      "learning_rate": 3.729005113775883e-05,
      "loss": 1.5142,
      "num_input_tokens_seen": 31373104,
      "step": 54085
    },
    {
      "epoch": 8.056300268096514,
      "grad_norm": 11.125,
      "learning_rate": 3.7287221372842506e-05,
      "loss": 0.933,
      "num_input_tokens_seen": 31375760,
      "step": 54090
    },
    {
      "epoch": 8.057044980637475,
      "grad_norm": 11.125,
      "learning_rate": 3.728439140034607e-05,
      "loss": 1.2935,
      "num_input_tokens_seen": 31379024,
      "step": 54095
    },
    {
      "epoch": 8.057789693178433,
      "grad_norm": 8.6875,
      "learning_rate": 3.728156122031732e-05,
      "loss": 1.2067,
      "num_input_tokens_seen": 31381808,
      "step": 54100
    },
    {
      "epoch": 8.058534405719392,
      "grad_norm": 6.96875,
      "learning_rate": 3.7278730832804076e-05,
      "loss": 0.957,
      "num_input_tokens_seen": 31384656,
      "step": 54105
    },
    {
      "epoch": 8.059279118260351,
      "grad_norm": 15.5,
      "learning_rate": 3.727590023785416e-05,
      "loss": 1.1186,
      "num_input_tokens_seen": 31387568,
      "step": 54110
    },
    {
      "epoch": 8.060023830801311,
      "grad_norm": 18.0,
      "learning_rate": 3.727306943551538e-05,
      "loss": 1.1804,
      "num_input_tokens_seen": 31390416,
      "step": 54115
    },
    {
      "epoch": 8.06076854334227,
      "grad_norm": 6.625,
      "learning_rate": 3.727023842583557e-05,
      "loss": 0.8509,
      "num_input_tokens_seen": 31393360,
      "step": 54120
    },
    {
      "epoch": 8.061513255883229,
      "grad_norm": 10.5625,
      "learning_rate": 3.726740720886255e-05,
      "loss": 1.6694,
      "num_input_tokens_seen": 31395984,
      "step": 54125
    },
    {
      "epoch": 8.062257968424188,
      "grad_norm": 13.375,
      "learning_rate": 3.726457578464416e-05,
      "loss": 1.1784,
      "num_input_tokens_seen": 31398704,
      "step": 54130
    },
    {
      "epoch": 8.063002680965148,
      "grad_norm": 17.375,
      "learning_rate": 3.726174415322822e-05,
      "loss": 0.9978,
      "num_input_tokens_seen": 31401552,
      "step": 54135
    },
    {
      "epoch": 8.063747393506107,
      "grad_norm": 13.625,
      "learning_rate": 3.725891231466258e-05,
      "loss": 1.1535,
      "num_input_tokens_seen": 31404624,
      "step": 54140
    },
    {
      "epoch": 8.064492106047066,
      "grad_norm": 16.25,
      "learning_rate": 3.725608026899507e-05,
      "loss": 1.2211,
      "num_input_tokens_seen": 31407760,
      "step": 54145
    },
    {
      "epoch": 8.065236818588025,
      "grad_norm": 10.4375,
      "learning_rate": 3.7253248016273545e-05,
      "loss": 1.0561,
      "num_input_tokens_seen": 31410736,
      "step": 54150
    },
    {
      "epoch": 8.065981531128985,
      "grad_norm": 11.8125,
      "learning_rate": 3.725041555654585e-05,
      "loss": 1.0793,
      "num_input_tokens_seen": 31413808,
      "step": 54155
    },
    {
      "epoch": 8.066726243669944,
      "grad_norm": 7.96875,
      "learning_rate": 3.7247582889859824e-05,
      "loss": 1.3183,
      "num_input_tokens_seen": 31416912,
      "step": 54160
    },
    {
      "epoch": 8.067470956210903,
      "grad_norm": 7.875,
      "learning_rate": 3.724475001626335e-05,
      "loss": 1.3521,
      "num_input_tokens_seen": 31419760,
      "step": 54165
    },
    {
      "epoch": 8.068215668751861,
      "grad_norm": 11.375,
      "learning_rate": 3.7241916935804254e-05,
      "loss": 1.0144,
      "num_input_tokens_seen": 31422544,
      "step": 54170
    },
    {
      "epoch": 8.06896038129282,
      "grad_norm": 8.625,
      "learning_rate": 3.723908364853042e-05,
      "loss": 0.9635,
      "num_input_tokens_seen": 31425360,
      "step": 54175
    },
    {
      "epoch": 8.06970509383378,
      "grad_norm": 12.0,
      "learning_rate": 3.723625015448971e-05,
      "loss": 1.0158,
      "num_input_tokens_seen": 31428112,
      "step": 54180
    },
    {
      "epoch": 8.07044980637474,
      "grad_norm": 6.5625,
      "learning_rate": 3.723341645372998e-05,
      "loss": 1.1673,
      "num_input_tokens_seen": 31431088,
      "step": 54185
    },
    {
      "epoch": 8.071194518915698,
      "grad_norm": 8.875,
      "learning_rate": 3.723058254629912e-05,
      "loss": 1.0023,
      "num_input_tokens_seen": 31434160,
      "step": 54190
    },
    {
      "epoch": 8.071939231456657,
      "grad_norm": 11.5,
      "learning_rate": 3.7227748432245e-05,
      "loss": 1.0325,
      "num_input_tokens_seen": 31436784,
      "step": 54195
    },
    {
      "epoch": 8.072683943997617,
      "grad_norm": 6.84375,
      "learning_rate": 3.722491411161549e-05,
      "loss": 1.0333,
      "num_input_tokens_seen": 31439856,
      "step": 54200
    },
    {
      "epoch": 8.073428656538576,
      "grad_norm": 10.25,
      "learning_rate": 3.722207958445849e-05,
      "loss": 0.813,
      "num_input_tokens_seen": 31442448,
      "step": 54205
    },
    {
      "epoch": 8.074173369079535,
      "grad_norm": 6.34375,
      "learning_rate": 3.721924485082187e-05,
      "loss": 1.0047,
      "num_input_tokens_seen": 31445456,
      "step": 54210
    },
    {
      "epoch": 8.074918081620494,
      "grad_norm": 14.75,
      "learning_rate": 3.721640991075354e-05,
      "loss": 1.1122,
      "num_input_tokens_seen": 31448432,
      "step": 54215
    },
    {
      "epoch": 8.075662794161454,
      "grad_norm": 15.875,
      "learning_rate": 3.7213574764301363e-05,
      "loss": 1.0585,
      "num_input_tokens_seen": 31451248,
      "step": 54220
    },
    {
      "epoch": 8.076407506702413,
      "grad_norm": 7.5,
      "learning_rate": 3.721073941151327e-05,
      "loss": 1.0866,
      "num_input_tokens_seen": 31454160,
      "step": 54225
    },
    {
      "epoch": 8.077152219243372,
      "grad_norm": 7.78125,
      "learning_rate": 3.7207903852437134e-05,
      "loss": 1.1998,
      "num_input_tokens_seen": 31456784,
      "step": 54230
    },
    {
      "epoch": 8.07789693178433,
      "grad_norm": 8.1875,
      "learning_rate": 3.7205068087120876e-05,
      "loss": 0.7824,
      "num_input_tokens_seen": 31459888,
      "step": 54235
    },
    {
      "epoch": 8.078641644325291,
      "grad_norm": 6.5,
      "learning_rate": 3.7202232115612396e-05,
      "loss": 0.7064,
      "num_input_tokens_seen": 31462704,
      "step": 54240
    },
    {
      "epoch": 8.07938635686625,
      "grad_norm": 10.625,
      "learning_rate": 3.7199395937959604e-05,
      "loss": 1.2514,
      "num_input_tokens_seen": 31465520,
      "step": 54245
    },
    {
      "epoch": 8.080131069407209,
      "grad_norm": 4.9375,
      "learning_rate": 3.7196559554210415e-05,
      "loss": 0.7586,
      "num_input_tokens_seen": 31468496,
      "step": 54250
    },
    {
      "epoch": 8.080875781948167,
      "grad_norm": 9.4375,
      "learning_rate": 3.719372296441275e-05,
      "loss": 0.9776,
      "num_input_tokens_seen": 31471248,
      "step": 54255
    },
    {
      "epoch": 8.081620494489128,
      "grad_norm": 6.875,
      "learning_rate": 3.719088616861453e-05,
      "loss": 1.0531,
      "num_input_tokens_seen": 31474128,
      "step": 54260
    },
    {
      "epoch": 8.082365207030087,
      "grad_norm": 8.9375,
      "learning_rate": 3.718804916686368e-05,
      "loss": 1.1598,
      "num_input_tokens_seen": 31477040,
      "step": 54265
    },
    {
      "epoch": 8.083109919571045,
      "grad_norm": 5.84375,
      "learning_rate": 3.7185211959208124e-05,
      "loss": 1.0946,
      "num_input_tokens_seen": 31479920,
      "step": 54270
    },
    {
      "epoch": 8.083854632112004,
      "grad_norm": 10.125,
      "learning_rate": 3.71823745456958e-05,
      "loss": 1.0394,
      "num_input_tokens_seen": 31482896,
      "step": 54275
    },
    {
      "epoch": 8.084599344652965,
      "grad_norm": 6.4375,
      "learning_rate": 3.7179536926374636e-05,
      "loss": 1.0219,
      "num_input_tokens_seen": 31485680,
      "step": 54280
    },
    {
      "epoch": 8.085344057193923,
      "grad_norm": 6.59375,
      "learning_rate": 3.7176699101292574e-05,
      "loss": 0.7123,
      "num_input_tokens_seen": 31488336,
      "step": 54285
    },
    {
      "epoch": 8.086088769734882,
      "grad_norm": 13.875,
      "learning_rate": 3.7173861070497556e-05,
      "loss": 0.996,
      "num_input_tokens_seen": 31491184,
      "step": 54290
    },
    {
      "epoch": 8.086833482275841,
      "grad_norm": 16.5,
      "learning_rate": 3.717102283403753e-05,
      "loss": 1.0848,
      "num_input_tokens_seen": 31494128,
      "step": 54295
    },
    {
      "epoch": 8.087578194816802,
      "grad_norm": 15.8125,
      "learning_rate": 3.716818439196045e-05,
      "loss": 1.2998,
      "num_input_tokens_seen": 31497296,
      "step": 54300
    },
    {
      "epoch": 8.08832290735776,
      "grad_norm": 11.875,
      "learning_rate": 3.716534574431425e-05,
      "loss": 0.9887,
      "num_input_tokens_seen": 31500432,
      "step": 54305
    },
    {
      "epoch": 8.089067619898719,
      "grad_norm": 12.0625,
      "learning_rate": 3.7162506891146896e-05,
      "loss": 1.1857,
      "num_input_tokens_seen": 31503312,
      "step": 54310
    },
    {
      "epoch": 8.089812332439678,
      "grad_norm": 7.1875,
      "learning_rate": 3.7159667832506365e-05,
      "loss": 1.1681,
      "num_input_tokens_seen": 31506320,
      "step": 54315
    },
    {
      "epoch": 8.090557044980638,
      "grad_norm": 20.375,
      "learning_rate": 3.715682856844059e-05,
      "loss": 1.1937,
      "num_input_tokens_seen": 31509328,
      "step": 54320
    },
    {
      "epoch": 8.091301757521597,
      "grad_norm": 19.375,
      "learning_rate": 3.715398909899756e-05,
      "loss": 1.3702,
      "num_input_tokens_seen": 31512336,
      "step": 54325
    },
    {
      "epoch": 8.092046470062556,
      "grad_norm": 19.5,
      "learning_rate": 3.715114942422524e-05,
      "loss": 1.5499,
      "num_input_tokens_seen": 31515312,
      "step": 54330
    },
    {
      "epoch": 8.092791182603515,
      "grad_norm": 6.65625,
      "learning_rate": 3.71483095441716e-05,
      "loss": 1.1586,
      "num_input_tokens_seen": 31517968,
      "step": 54335
    },
    {
      "epoch": 8.093535895144473,
      "grad_norm": 9.9375,
      "learning_rate": 3.7145469458884606e-05,
      "loss": 1.1363,
      "num_input_tokens_seen": 31520848,
      "step": 54340
    },
    {
      "epoch": 8.094280607685434,
      "grad_norm": 9.6875,
      "learning_rate": 3.714262916841226e-05,
      "loss": 0.923,
      "num_input_tokens_seen": 31523888,
      "step": 54345
    },
    {
      "epoch": 8.095025320226393,
      "grad_norm": 8.1875,
      "learning_rate": 3.7139788672802526e-05,
      "loss": 1.0073,
      "num_input_tokens_seen": 31526832,
      "step": 54350
    },
    {
      "epoch": 8.095770032767351,
      "grad_norm": 19.5,
      "learning_rate": 3.713694797210341e-05,
      "loss": 1.1236,
      "num_input_tokens_seen": 31529808,
      "step": 54355
    },
    {
      "epoch": 8.09651474530831,
      "grad_norm": 11.125,
      "learning_rate": 3.713410706636289e-05,
      "loss": 1.177,
      "num_input_tokens_seen": 31532720,
      "step": 54360
    },
    {
      "epoch": 8.09725945784927,
      "grad_norm": 19.625,
      "learning_rate": 3.713126595562896e-05,
      "loss": 1.0694,
      "num_input_tokens_seen": 31535920,
      "step": 54365
    },
    {
      "epoch": 8.09800417039023,
      "grad_norm": 6.65625,
      "learning_rate": 3.712842463994963e-05,
      "loss": 1.3704,
      "num_input_tokens_seen": 31538800,
      "step": 54370
    },
    {
      "epoch": 8.098748882931188,
      "grad_norm": 8.5625,
      "learning_rate": 3.7125583119372884e-05,
      "loss": 1.3358,
      "num_input_tokens_seen": 31541456,
      "step": 54375
    },
    {
      "epoch": 8.099493595472147,
      "grad_norm": 13.875,
      "learning_rate": 3.712274139394674e-05,
      "loss": 1.0567,
      "num_input_tokens_seen": 31544464,
      "step": 54380
    },
    {
      "epoch": 8.100238308013108,
      "grad_norm": 14.1875,
      "learning_rate": 3.71198994637192e-05,
      "loss": 0.9763,
      "num_input_tokens_seen": 31547440,
      "step": 54385
    },
    {
      "epoch": 8.100983020554066,
      "grad_norm": 12.4375,
      "learning_rate": 3.711705732873828e-05,
      "loss": 0.8768,
      "num_input_tokens_seen": 31550512,
      "step": 54390
    },
    {
      "epoch": 8.101727733095025,
      "grad_norm": 5.90625,
      "learning_rate": 3.711421498905198e-05,
      "loss": 0.9177,
      "num_input_tokens_seen": 31553488,
      "step": 54395
    },
    {
      "epoch": 8.102472445635984,
      "grad_norm": 9.5625,
      "learning_rate": 3.7111372444708345e-05,
      "loss": 1.1062,
      "num_input_tokens_seen": 31556368,
      "step": 54400
    },
    {
      "epoch": 8.103217158176944,
      "grad_norm": 8.4375,
      "learning_rate": 3.7108529695755375e-05,
      "loss": 1.04,
      "num_input_tokens_seen": 31558896,
      "step": 54405
    },
    {
      "epoch": 8.103961870717903,
      "grad_norm": 8.5,
      "learning_rate": 3.7105686742241095e-05,
      "loss": 0.7341,
      "num_input_tokens_seen": 31561712,
      "step": 54410
    },
    {
      "epoch": 8.104706583258862,
      "grad_norm": 7.59375,
      "learning_rate": 3.7102843584213556e-05,
      "loss": 1.1528,
      "num_input_tokens_seen": 31564816,
      "step": 54415
    },
    {
      "epoch": 8.10545129579982,
      "grad_norm": 16.875,
      "learning_rate": 3.710000022172076e-05,
      "loss": 1.2677,
      "num_input_tokens_seen": 31567504,
      "step": 54420
    },
    {
      "epoch": 8.106196008340781,
      "grad_norm": 16.25,
      "learning_rate": 3.7097156654810774e-05,
      "loss": 1.0811,
      "num_input_tokens_seen": 31570544,
      "step": 54425
    },
    {
      "epoch": 8.10694072088174,
      "grad_norm": 14.0,
      "learning_rate": 3.709431288353161e-05,
      "loss": 1.3789,
      "num_input_tokens_seen": 31573040,
      "step": 54430
    },
    {
      "epoch": 8.107685433422699,
      "grad_norm": 14.75,
      "learning_rate": 3.7091468907931324e-05,
      "loss": 0.9667,
      "num_input_tokens_seen": 31576112,
      "step": 54435
    },
    {
      "epoch": 8.108430145963657,
      "grad_norm": 17.0,
      "learning_rate": 3.708862472805796e-05,
      "loss": 1.1667,
      "num_input_tokens_seen": 31579216,
      "step": 54440
    },
    {
      "epoch": 8.109174858504618,
      "grad_norm": 6.65625,
      "learning_rate": 3.708578034395957e-05,
      "loss": 1.1194,
      "num_input_tokens_seen": 31582224,
      "step": 54445
    },
    {
      "epoch": 8.109919571045577,
      "grad_norm": 15.3125,
      "learning_rate": 3.70829357556842e-05,
      "loss": 0.9871,
      "num_input_tokens_seen": 31585040,
      "step": 54450
    },
    {
      "epoch": 8.110664283586535,
      "grad_norm": 8.0,
      "learning_rate": 3.7080090963279915e-05,
      "loss": 1.2044,
      "num_input_tokens_seen": 31587920,
      "step": 54455
    },
    {
      "epoch": 8.111408996127494,
      "grad_norm": 13.125,
      "learning_rate": 3.7077245966794774e-05,
      "loss": 0.8733,
      "num_input_tokens_seen": 31590608,
      "step": 54460
    },
    {
      "epoch": 8.112153708668455,
      "grad_norm": 21.25,
      "learning_rate": 3.707440076627683e-05,
      "loss": 1.155,
      "num_input_tokens_seen": 31593520,
      "step": 54465
    },
    {
      "epoch": 8.112898421209414,
      "grad_norm": 9.375,
      "learning_rate": 3.7071555361774165e-05,
      "loss": 0.9432,
      "num_input_tokens_seen": 31596688,
      "step": 54470
    },
    {
      "epoch": 8.113643133750372,
      "grad_norm": 12.125,
      "learning_rate": 3.706870975333484e-05,
      "loss": 0.8808,
      "num_input_tokens_seen": 31599248,
      "step": 54475
    },
    {
      "epoch": 8.114387846291331,
      "grad_norm": 10.5,
      "learning_rate": 3.706586394100692e-05,
      "loss": 1.3771,
      "num_input_tokens_seen": 31602064,
      "step": 54480
    },
    {
      "epoch": 8.115132558832292,
      "grad_norm": 8.5625,
      "learning_rate": 3.70630179248385e-05,
      "loss": 1.346,
      "num_input_tokens_seen": 31605296,
      "step": 54485
    },
    {
      "epoch": 8.11587727137325,
      "grad_norm": 8.5,
      "learning_rate": 3.706017170487765e-05,
      "loss": 1.113,
      "num_input_tokens_seen": 31608432,
      "step": 54490
    },
    {
      "epoch": 8.116621983914209,
      "grad_norm": 15.875,
      "learning_rate": 3.705732528117246e-05,
      "loss": 1.3866,
      "num_input_tokens_seen": 31611280,
      "step": 54495
    },
    {
      "epoch": 8.117366696455168,
      "grad_norm": 6.28125,
      "learning_rate": 3.7054478653771005e-05,
      "loss": 1.015,
      "num_input_tokens_seen": 31614192,
      "step": 54500
    },
    {
      "epoch": 8.118111408996128,
      "grad_norm": 9.625,
      "learning_rate": 3.7051631822721395e-05,
      "loss": 1.0194,
      "num_input_tokens_seen": 31617200,
      "step": 54505
    },
    {
      "epoch": 8.118856121537087,
      "grad_norm": 15.4375,
      "learning_rate": 3.704878478807171e-05,
      "loss": 1.0729,
      "num_input_tokens_seen": 31619984,
      "step": 54510
    },
    {
      "epoch": 8.119600834078046,
      "grad_norm": 8.4375,
      "learning_rate": 3.704593754987005e-05,
      "loss": 0.9389,
      "num_input_tokens_seen": 31622768,
      "step": 54515
    },
    {
      "epoch": 8.120345546619005,
      "grad_norm": 10.8125,
      "learning_rate": 3.704309010816452e-05,
      "loss": 0.9614,
      "num_input_tokens_seen": 31625488,
      "step": 54520
    },
    {
      "epoch": 8.121090259159963,
      "grad_norm": 14.375,
      "learning_rate": 3.7040242463003225e-05,
      "loss": 1.2788,
      "num_input_tokens_seen": 31628688,
      "step": 54525
    },
    {
      "epoch": 8.121834971700924,
      "grad_norm": 13.25,
      "learning_rate": 3.703739461443427e-05,
      "loss": 1.0465,
      "num_input_tokens_seen": 31631664,
      "step": 54530
    },
    {
      "epoch": 8.122579684241883,
      "grad_norm": 14.375,
      "learning_rate": 3.703454656250576e-05,
      "loss": 1.3729,
      "num_input_tokens_seen": 31634448,
      "step": 54535
    },
    {
      "epoch": 8.123324396782841,
      "grad_norm": 15.9375,
      "learning_rate": 3.7031698307265824e-05,
      "loss": 1.2697,
      "num_input_tokens_seen": 31637200,
      "step": 54540
    },
    {
      "epoch": 8.1240691093238,
      "grad_norm": 10.0625,
      "learning_rate": 3.702884984876257e-05,
      "loss": 0.7622,
      "num_input_tokens_seen": 31640016,
      "step": 54545
    },
    {
      "epoch": 8.12481382186476,
      "grad_norm": 12.0,
      "learning_rate": 3.702600118704412e-05,
      "loss": 0.8855,
      "num_input_tokens_seen": 31642992,
      "step": 54550
    },
    {
      "epoch": 8.12555853440572,
      "grad_norm": 13.625,
      "learning_rate": 3.702315232215862e-05,
      "loss": 0.9209,
      "num_input_tokens_seen": 31645840,
      "step": 54555
    },
    {
      "epoch": 8.126303246946678,
      "grad_norm": 8.4375,
      "learning_rate": 3.7020303254154164e-05,
      "loss": 1.0689,
      "num_input_tokens_seen": 31648752,
      "step": 54560
    },
    {
      "epoch": 8.127047959487637,
      "grad_norm": 7.875,
      "learning_rate": 3.701745398307891e-05,
      "loss": 0.9226,
      "num_input_tokens_seen": 31651600,
      "step": 54565
    },
    {
      "epoch": 8.127792672028598,
      "grad_norm": 13.625,
      "learning_rate": 3.701460450898098e-05,
      "loss": 1.2675,
      "num_input_tokens_seen": 31654448,
      "step": 54570
    },
    {
      "epoch": 8.128537384569556,
      "grad_norm": 6.8125,
      "learning_rate": 3.701175483190852e-05,
      "loss": 0.8868,
      "num_input_tokens_seen": 31657296,
      "step": 54575
    },
    {
      "epoch": 8.129282097110515,
      "grad_norm": 5.625,
      "learning_rate": 3.700890495190967e-05,
      "loss": 0.8713,
      "num_input_tokens_seen": 31660144,
      "step": 54580
    },
    {
      "epoch": 8.130026809651474,
      "grad_norm": 17.875,
      "learning_rate": 3.7006054869032574e-05,
      "loss": 1.2741,
      "num_input_tokens_seen": 31663216,
      "step": 54585
    },
    {
      "epoch": 8.130771522192434,
      "grad_norm": 10.4375,
      "learning_rate": 3.700320458332539e-05,
      "loss": 0.998,
      "num_input_tokens_seen": 31666256,
      "step": 54590
    },
    {
      "epoch": 8.131516234733393,
      "grad_norm": 6.09375,
      "learning_rate": 3.700035409483626e-05,
      "loss": 0.7602,
      "num_input_tokens_seen": 31669008,
      "step": 54595
    },
    {
      "epoch": 8.132260947274352,
      "grad_norm": 6.96875,
      "learning_rate": 3.699750340361334e-05,
      "loss": 1.2051,
      "num_input_tokens_seen": 31672144,
      "step": 54600
    },
    {
      "epoch": 8.13300565981531,
      "grad_norm": 7.625,
      "learning_rate": 3.69946525097048e-05,
      "loss": 1.4026,
      "num_input_tokens_seen": 31674864,
      "step": 54605
    },
    {
      "epoch": 8.133750372356271,
      "grad_norm": 6.25,
      "learning_rate": 3.6991801413158795e-05,
      "loss": 0.8597,
      "num_input_tokens_seen": 31677744,
      "step": 54610
    },
    {
      "epoch": 8.13449508489723,
      "grad_norm": 14.5,
      "learning_rate": 3.6988950114023494e-05,
      "loss": 1.0201,
      "num_input_tokens_seen": 31680560,
      "step": 54615
    },
    {
      "epoch": 8.135239797438189,
      "grad_norm": 16.625,
      "learning_rate": 3.6986098612347056e-05,
      "loss": 1.1831,
      "num_input_tokens_seen": 31683728,
      "step": 54620
    },
    {
      "epoch": 8.135984509979147,
      "grad_norm": 5.8125,
      "learning_rate": 3.6983246908177675e-05,
      "loss": 1.1479,
      "num_input_tokens_seen": 31686512,
      "step": 54625
    },
    {
      "epoch": 8.136729222520108,
      "grad_norm": 7.25,
      "learning_rate": 3.698039500156352e-05,
      "loss": 1.5,
      "num_input_tokens_seen": 31689264,
      "step": 54630
    },
    {
      "epoch": 8.137473935061067,
      "grad_norm": 6.0625,
      "learning_rate": 3.697754289255277e-05,
      "loss": 1.2708,
      "num_input_tokens_seen": 31692208,
      "step": 54635
    },
    {
      "epoch": 8.138218647602026,
      "grad_norm": 10.3125,
      "learning_rate": 3.697469058119359e-05,
      "loss": 1.2885,
      "num_input_tokens_seen": 31695024,
      "step": 54640
    },
    {
      "epoch": 8.138963360142984,
      "grad_norm": 12.6875,
      "learning_rate": 3.697183806753419e-05,
      "loss": 1.1598,
      "num_input_tokens_seen": 31698192,
      "step": 54645
    },
    {
      "epoch": 8.139708072683945,
      "grad_norm": 8.1875,
      "learning_rate": 3.696898535162275e-05,
      "loss": 1.0348,
      "num_input_tokens_seen": 31700944,
      "step": 54650
    },
    {
      "epoch": 8.140452785224904,
      "grad_norm": 10.0625,
      "learning_rate": 3.696613243350747e-05,
      "loss": 1.1168,
      "num_input_tokens_seen": 31704016,
      "step": 54655
    },
    {
      "epoch": 8.141197497765862,
      "grad_norm": 12.8125,
      "learning_rate": 3.696327931323655e-05,
      "loss": 1.0814,
      "num_input_tokens_seen": 31707024,
      "step": 54660
    },
    {
      "epoch": 8.141942210306821,
      "grad_norm": 10.6875,
      "learning_rate": 3.696042599085818e-05,
      "loss": 0.8676,
      "num_input_tokens_seen": 31709840,
      "step": 54665
    },
    {
      "epoch": 8.142686922847782,
      "grad_norm": 6.40625,
      "learning_rate": 3.695757246642057e-05,
      "loss": 0.9123,
      "num_input_tokens_seen": 31712816,
      "step": 54670
    },
    {
      "epoch": 8.14343163538874,
      "grad_norm": 9.5625,
      "learning_rate": 3.695471873997193e-05,
      "loss": 1.0311,
      "num_input_tokens_seen": 31715440,
      "step": 54675
    },
    {
      "epoch": 8.1441763479297,
      "grad_norm": 14.5,
      "learning_rate": 3.6951864811560464e-05,
      "loss": 0.7136,
      "num_input_tokens_seen": 31718352,
      "step": 54680
    },
    {
      "epoch": 8.144921060470658,
      "grad_norm": 6.40625,
      "learning_rate": 3.694901068123439e-05,
      "loss": 1.0846,
      "num_input_tokens_seen": 31721424,
      "step": 54685
    },
    {
      "epoch": 8.145665773011617,
      "grad_norm": 13.0,
      "learning_rate": 3.694615634904192e-05,
      "loss": 1.2213,
      "num_input_tokens_seen": 31724496,
      "step": 54690
    },
    {
      "epoch": 8.146410485552577,
      "grad_norm": 3.8125,
      "learning_rate": 3.694330181503128e-05,
      "loss": 0.9795,
      "num_input_tokens_seen": 31727344,
      "step": 54695
    },
    {
      "epoch": 8.147155198093536,
      "grad_norm": 7.28125,
      "learning_rate": 3.69404470792507e-05,
      "loss": 1.1665,
      "num_input_tokens_seen": 31730320,
      "step": 54700
    },
    {
      "epoch": 8.147899910634495,
      "grad_norm": 10.125,
      "learning_rate": 3.69375921417484e-05,
      "loss": 1.0612,
      "num_input_tokens_seen": 31733360,
      "step": 54705
    },
    {
      "epoch": 8.148644623175453,
      "grad_norm": 7.75,
      "learning_rate": 3.6934737002572614e-05,
      "loss": 0.9762,
      "num_input_tokens_seen": 31736112,
      "step": 54710
    },
    {
      "epoch": 8.149389335716414,
      "grad_norm": 13.625,
      "learning_rate": 3.693188166177158e-05,
      "loss": 1.433,
      "num_input_tokens_seen": 31738992,
      "step": 54715
    },
    {
      "epoch": 8.150134048257373,
      "grad_norm": 6.5625,
      "learning_rate": 3.6929026119393525e-05,
      "loss": 0.9506,
      "num_input_tokens_seen": 31742000,
      "step": 54720
    },
    {
      "epoch": 8.150878760798332,
      "grad_norm": 9.125,
      "learning_rate": 3.69261703754867e-05,
      "loss": 1.2068,
      "num_input_tokens_seen": 31744912,
      "step": 54725
    },
    {
      "epoch": 8.15162347333929,
      "grad_norm": 11.0625,
      "learning_rate": 3.6923314430099354e-05,
      "loss": 1.1707,
      "num_input_tokens_seen": 31747856,
      "step": 54730
    },
    {
      "epoch": 8.15236818588025,
      "grad_norm": 7.4375,
      "learning_rate": 3.6920458283279725e-05,
      "loss": 1.3797,
      "num_input_tokens_seen": 31750928,
      "step": 54735
    },
    {
      "epoch": 8.15311289842121,
      "grad_norm": 7.15625,
      "learning_rate": 3.691760193507607e-05,
      "loss": 0.9349,
      "num_input_tokens_seen": 31754032,
      "step": 54740
    },
    {
      "epoch": 8.153857610962168,
      "grad_norm": 9.5625,
      "learning_rate": 3.691474538553664e-05,
      "loss": 1.1526,
      "num_input_tokens_seen": 31756816,
      "step": 54745
    },
    {
      "epoch": 8.154602323503127,
      "grad_norm": 17.25,
      "learning_rate": 3.691188863470969e-05,
      "loss": 1.0897,
      "num_input_tokens_seen": 31759792,
      "step": 54750
    },
    {
      "epoch": 8.155347036044088,
      "grad_norm": 8.75,
      "learning_rate": 3.69090316826435e-05,
      "loss": 1.0544,
      "num_input_tokens_seen": 31762480,
      "step": 54755
    },
    {
      "epoch": 8.156091748585046,
      "grad_norm": 8.0625,
      "learning_rate": 3.690617452938632e-05,
      "loss": 1.0999,
      "num_input_tokens_seen": 31765264,
      "step": 54760
    },
    {
      "epoch": 8.156836461126005,
      "grad_norm": 7.53125,
      "learning_rate": 3.6903317174986425e-05,
      "loss": 1.0006,
      "num_input_tokens_seen": 31768176,
      "step": 54765
    },
    {
      "epoch": 8.157581173666964,
      "grad_norm": 15.875,
      "learning_rate": 3.690045961949208e-05,
      "loss": 1.2543,
      "num_input_tokens_seen": 31770960,
      "step": 54770
    },
    {
      "epoch": 8.158325886207924,
      "grad_norm": 12.5,
      "learning_rate": 3.689760186295156e-05,
      "loss": 1.1207,
      "num_input_tokens_seen": 31773584,
      "step": 54775
    },
    {
      "epoch": 8.159070598748883,
      "grad_norm": 7.4375,
      "learning_rate": 3.689474390541316e-05,
      "loss": 1.0576,
      "num_input_tokens_seen": 31776432,
      "step": 54780
    },
    {
      "epoch": 8.159815311289842,
      "grad_norm": 6.71875,
      "learning_rate": 3.6891885746925136e-05,
      "loss": 0.9143,
      "num_input_tokens_seen": 31779408,
      "step": 54785
    },
    {
      "epoch": 8.1605600238308,
      "grad_norm": 11.375,
      "learning_rate": 3.68890273875358e-05,
      "loss": 1.1077,
      "num_input_tokens_seen": 31782640,
      "step": 54790
    },
    {
      "epoch": 8.161304736371761,
      "grad_norm": 9.9375,
      "learning_rate": 3.688616882729343e-05,
      "loss": 1.1621,
      "num_input_tokens_seen": 31785616,
      "step": 54795
    },
    {
      "epoch": 8.16204944891272,
      "grad_norm": 11.4375,
      "learning_rate": 3.688331006624632e-05,
      "loss": 1.026,
      "num_input_tokens_seen": 31788496,
      "step": 54800
    },
    {
      "epoch": 8.162794161453679,
      "grad_norm": 25.0,
      "learning_rate": 3.688045110444276e-05,
      "loss": 1.2121,
      "num_input_tokens_seen": 31791760,
      "step": 54805
    },
    {
      "epoch": 8.163538873994638,
      "grad_norm": 15.3125,
      "learning_rate": 3.687759194193105e-05,
      "loss": 1.3019,
      "num_input_tokens_seen": 31794576,
      "step": 54810
    },
    {
      "epoch": 8.164283586535598,
      "grad_norm": 3.328125,
      "learning_rate": 3.6874732578759495e-05,
      "loss": 0.6827,
      "num_input_tokens_seen": 31797552,
      "step": 54815
    },
    {
      "epoch": 8.165028299076557,
      "grad_norm": 8.4375,
      "learning_rate": 3.687187301497641e-05,
      "loss": 1.0901,
      "num_input_tokens_seen": 31800912,
      "step": 54820
    },
    {
      "epoch": 8.165773011617516,
      "grad_norm": 18.25,
      "learning_rate": 3.6869013250630094e-05,
      "loss": 1.2163,
      "num_input_tokens_seen": 31803824,
      "step": 54825
    },
    {
      "epoch": 8.166517724158474,
      "grad_norm": 7.46875,
      "learning_rate": 3.686615328576886e-05,
      "loss": 0.9215,
      "num_input_tokens_seen": 31807504,
      "step": 54830
    },
    {
      "epoch": 8.167262436699435,
      "grad_norm": 7.9375,
      "learning_rate": 3.686329312044102e-05,
      "loss": 1.1843,
      "num_input_tokens_seen": 31810576,
      "step": 54835
    },
    {
      "epoch": 8.168007149240394,
      "grad_norm": 15.6875,
      "learning_rate": 3.6860432754694915e-05,
      "loss": 1.2307,
      "num_input_tokens_seen": 31813648,
      "step": 54840
    },
    {
      "epoch": 8.168751861781352,
      "grad_norm": 4.25,
      "learning_rate": 3.685757218857885e-05,
      "loss": 0.9154,
      "num_input_tokens_seen": 31816240,
      "step": 54845
    },
    {
      "epoch": 8.169496574322311,
      "grad_norm": 6.3125,
      "learning_rate": 3.6854711422141144e-05,
      "loss": 0.997,
      "num_input_tokens_seen": 31818800,
      "step": 54850
    },
    {
      "epoch": 8.17024128686327,
      "grad_norm": 17.0,
      "learning_rate": 3.685185045543014e-05,
      "loss": 1.1732,
      "num_input_tokens_seen": 31821520,
      "step": 54855
    },
    {
      "epoch": 8.17098599940423,
      "grad_norm": 15.75,
      "learning_rate": 3.684898928849417e-05,
      "loss": 1.1976,
      "num_input_tokens_seen": 31824560,
      "step": 54860
    },
    {
      "epoch": 8.17173071194519,
      "grad_norm": 5.0,
      "learning_rate": 3.6846127921381576e-05,
      "loss": 0.5871,
      "num_input_tokens_seen": 31827408,
      "step": 54865
    },
    {
      "epoch": 8.172475424486148,
      "grad_norm": 2.71875,
      "learning_rate": 3.684326635414068e-05,
      "loss": 0.6566,
      "num_input_tokens_seen": 31830160,
      "step": 54870
    },
    {
      "epoch": 8.173220137027107,
      "grad_norm": 8.5,
      "learning_rate": 3.684040458681984e-05,
      "loss": 1.3495,
      "num_input_tokens_seen": 31832848,
      "step": 54875
    },
    {
      "epoch": 8.173964849568067,
      "grad_norm": 10.5,
      "learning_rate": 3.6837542619467404e-05,
      "loss": 1.0537,
      "num_input_tokens_seen": 31836080,
      "step": 54880
    },
    {
      "epoch": 8.174709562109026,
      "grad_norm": 17.75,
      "learning_rate": 3.6834680452131707e-05,
      "loss": 0.8944,
      "num_input_tokens_seen": 31838928,
      "step": 54885
    },
    {
      "epoch": 8.175454274649985,
      "grad_norm": 9.8125,
      "learning_rate": 3.683181808486112e-05,
      "loss": 1.099,
      "num_input_tokens_seen": 31841584,
      "step": 54890
    },
    {
      "epoch": 8.176198987190944,
      "grad_norm": 6.4375,
      "learning_rate": 3.682895551770399e-05,
      "loss": 0.9995,
      "num_input_tokens_seen": 31844496,
      "step": 54895
    },
    {
      "epoch": 8.176943699731904,
      "grad_norm": 8.8125,
      "learning_rate": 3.682609275070867e-05,
      "loss": 1.0381,
      "num_input_tokens_seen": 31847184,
      "step": 54900
    },
    {
      "epoch": 8.177688412272863,
      "grad_norm": 7.375,
      "learning_rate": 3.682322978392354e-05,
      "loss": 1.1155,
      "num_input_tokens_seen": 31850128,
      "step": 54905
    },
    {
      "epoch": 8.178433124813822,
      "grad_norm": 8.0,
      "learning_rate": 3.682036661739696e-05,
      "loss": 1.3656,
      "num_input_tokens_seen": 31852912,
      "step": 54910
    },
    {
      "epoch": 8.17917783735478,
      "grad_norm": 10.9375,
      "learning_rate": 3.68175032511773e-05,
      "loss": 0.6763,
      "num_input_tokens_seen": 31855920,
      "step": 54915
    },
    {
      "epoch": 8.17992254989574,
      "grad_norm": 7.0625,
      "learning_rate": 3.6814639685312936e-05,
      "loss": 1.2507,
      "num_input_tokens_seen": 31859088,
      "step": 54920
    },
    {
      "epoch": 8.1806672624367,
      "grad_norm": 6.21875,
      "learning_rate": 3.6811775919852245e-05,
      "loss": 1.1347,
      "num_input_tokens_seen": 31861936,
      "step": 54925
    },
    {
      "epoch": 8.181411974977658,
      "grad_norm": 8.875,
      "learning_rate": 3.6808911954843595e-05,
      "loss": 0.8741,
      "num_input_tokens_seen": 31864816,
      "step": 54930
    },
    {
      "epoch": 8.182156687518617,
      "grad_norm": 9.0625,
      "learning_rate": 3.680604779033538e-05,
      "loss": 0.9823,
      "num_input_tokens_seen": 31867728,
      "step": 54935
    },
    {
      "epoch": 8.182901400059578,
      "grad_norm": 7.59375,
      "learning_rate": 3.680318342637599e-05,
      "loss": 0.9901,
      "num_input_tokens_seen": 31870704,
      "step": 54940
    },
    {
      "epoch": 8.183646112600536,
      "grad_norm": 8.75,
      "learning_rate": 3.6800318863013806e-05,
      "loss": 0.9024,
      "num_input_tokens_seen": 31873616,
      "step": 54945
    },
    {
      "epoch": 8.184390825141495,
      "grad_norm": 17.0,
      "learning_rate": 3.6797454100297234e-05,
      "loss": 1.0376,
      "num_input_tokens_seen": 31876528,
      "step": 54950
    },
    {
      "epoch": 8.185135537682454,
      "grad_norm": 18.0,
      "learning_rate": 3.679458913827467e-05,
      "loss": 1.1484,
      "num_input_tokens_seen": 31879088,
      "step": 54955
    },
    {
      "epoch": 8.185880250223414,
      "grad_norm": 7.09375,
      "learning_rate": 3.6791723976994505e-05,
      "loss": 0.843,
      "num_input_tokens_seen": 31881840,
      "step": 54960
    },
    {
      "epoch": 8.186624962764373,
      "grad_norm": 6.75,
      "learning_rate": 3.678885861650515e-05,
      "loss": 1.0462,
      "num_input_tokens_seen": 31885008,
      "step": 54965
    },
    {
      "epoch": 8.187369675305332,
      "grad_norm": 10.375,
      "learning_rate": 3.6785993056855004e-05,
      "loss": 0.9091,
      "num_input_tokens_seen": 31887952,
      "step": 54970
    },
    {
      "epoch": 8.18811438784629,
      "grad_norm": 8.3125,
      "learning_rate": 3.678312729809249e-05,
      "loss": 0.9756,
      "num_input_tokens_seen": 31891088,
      "step": 54975
    },
    {
      "epoch": 8.188859100387251,
      "grad_norm": 30.0,
      "learning_rate": 3.6780261340266014e-05,
      "loss": 1.4402,
      "num_input_tokens_seen": 31893872,
      "step": 54980
    },
    {
      "epoch": 8.18960381292821,
      "grad_norm": 8.6875,
      "learning_rate": 3.677739518342399e-05,
      "loss": 0.9445,
      "num_input_tokens_seen": 31896688,
      "step": 54985
    },
    {
      "epoch": 8.190348525469169,
      "grad_norm": 17.875,
      "learning_rate": 3.677452882761486e-05,
      "loss": 1.2555,
      "num_input_tokens_seen": 31899536,
      "step": 54990
    },
    {
      "epoch": 8.191093238010128,
      "grad_norm": 15.875,
      "learning_rate": 3.677166227288702e-05,
      "loss": 0.835,
      "num_input_tokens_seen": 31902672,
      "step": 54995
    },
    {
      "epoch": 8.191837950551088,
      "grad_norm": 13.0625,
      "learning_rate": 3.6768795519288916e-05,
      "loss": 1.0928,
      "num_input_tokens_seen": 31905456,
      "step": 55000
    },
    {
      "epoch": 8.192582663092047,
      "grad_norm": 11.3125,
      "learning_rate": 3.6765928566868976e-05,
      "loss": 0.891,
      "num_input_tokens_seen": 31908272,
      "step": 55005
    },
    {
      "epoch": 8.193327375633006,
      "grad_norm": 5.65625,
      "learning_rate": 3.676306141567562e-05,
      "loss": 0.9275,
      "num_input_tokens_seen": 31911216,
      "step": 55010
    },
    {
      "epoch": 8.194072088173964,
      "grad_norm": 5.0625,
      "learning_rate": 3.67601940657573e-05,
      "loss": 1.0132,
      "num_input_tokens_seen": 31914096,
      "step": 55015
    },
    {
      "epoch": 8.194816800714925,
      "grad_norm": 13.1875,
      "learning_rate": 3.675732651716246e-05,
      "loss": 1.2078,
      "num_input_tokens_seen": 31916848,
      "step": 55020
    },
    {
      "epoch": 8.195561513255884,
      "grad_norm": 15.4375,
      "learning_rate": 3.675445876993953e-05,
      "loss": 1.4208,
      "num_input_tokens_seen": 31919792,
      "step": 55025
    },
    {
      "epoch": 8.196306225796842,
      "grad_norm": 6.34375,
      "learning_rate": 3.675159082413697e-05,
      "loss": 1.0379,
      "num_input_tokens_seen": 31922704,
      "step": 55030
    },
    {
      "epoch": 8.197050938337801,
      "grad_norm": 13.4375,
      "learning_rate": 3.674872267980323e-05,
      "loss": 0.9887,
      "num_input_tokens_seen": 31925744,
      "step": 55035
    },
    {
      "epoch": 8.19779565087876,
      "grad_norm": 16.625,
      "learning_rate": 3.674585433698676e-05,
      "loss": 0.9325,
      "num_input_tokens_seen": 31928560,
      "step": 55040
    },
    {
      "epoch": 8.19854036341972,
      "grad_norm": 9.375,
      "learning_rate": 3.674298579573602e-05,
      "loss": 1.2611,
      "num_input_tokens_seen": 31931472,
      "step": 55045
    },
    {
      "epoch": 8.19928507596068,
      "grad_norm": 5.90625,
      "learning_rate": 3.674011705609946e-05,
      "loss": 1.0132,
      "num_input_tokens_seen": 31935152,
      "step": 55050
    },
    {
      "epoch": 8.200029788501638,
      "grad_norm": 9.875,
      "learning_rate": 3.6737248118125564e-05,
      "loss": 1.0835,
      "num_input_tokens_seen": 31938032,
      "step": 55055
    },
    {
      "epoch": 8.200774501042597,
      "grad_norm": 8.875,
      "learning_rate": 3.673437898186279e-05,
      "loss": 1.3545,
      "num_input_tokens_seen": 31940656,
      "step": 55060
    },
    {
      "epoch": 8.201519213583557,
      "grad_norm": 15.125,
      "learning_rate": 3.6731509647359604e-05,
      "loss": 0.9275,
      "num_input_tokens_seen": 31943568,
      "step": 55065
    },
    {
      "epoch": 8.202263926124516,
      "grad_norm": 23.25,
      "learning_rate": 3.6728640114664485e-05,
      "loss": 0.775,
      "num_input_tokens_seen": 31946352,
      "step": 55070
    },
    {
      "epoch": 8.203008638665475,
      "grad_norm": 17.75,
      "learning_rate": 3.672577038382592e-05,
      "loss": 1.5692,
      "num_input_tokens_seen": 31948976,
      "step": 55075
    },
    {
      "epoch": 8.203753351206434,
      "grad_norm": 8.5,
      "learning_rate": 3.672290045489238e-05,
      "loss": 1.0414,
      "num_input_tokens_seen": 31951824,
      "step": 55080
    },
    {
      "epoch": 8.204498063747394,
      "grad_norm": 5.84375,
      "learning_rate": 3.672003032791235e-05,
      "loss": 0.9696,
      "num_input_tokens_seen": 31954800,
      "step": 55085
    },
    {
      "epoch": 8.205242776288353,
      "grad_norm": 10.6875,
      "learning_rate": 3.671716000293432e-05,
      "loss": 0.9481,
      "num_input_tokens_seen": 31957520,
      "step": 55090
    },
    {
      "epoch": 8.205987488829312,
      "grad_norm": 10.9375,
      "learning_rate": 3.671428948000677e-05,
      "loss": 1.1172,
      "num_input_tokens_seen": 31960208,
      "step": 55095
    },
    {
      "epoch": 8.20673220137027,
      "grad_norm": 14.375,
      "learning_rate": 3.671141875917822e-05,
      "loss": 1.2389,
      "num_input_tokens_seen": 31963056,
      "step": 55100
    },
    {
      "epoch": 8.207476913911231,
      "grad_norm": 9.125,
      "learning_rate": 3.6708547840497144e-05,
      "loss": 1.2366,
      "num_input_tokens_seen": 31965616,
      "step": 55105
    },
    {
      "epoch": 8.20822162645219,
      "grad_norm": 5.09375,
      "learning_rate": 3.6705676724012055e-05,
      "loss": 1.0217,
      "num_input_tokens_seen": 31968336,
      "step": 55110
    },
    {
      "epoch": 8.208966338993148,
      "grad_norm": 8.25,
      "learning_rate": 3.670280540977145e-05,
      "loss": 1.3372,
      "num_input_tokens_seen": 31971152,
      "step": 55115
    },
    {
      "epoch": 8.209711051534107,
      "grad_norm": 13.8125,
      "learning_rate": 3.669993389782385e-05,
      "loss": 0.8472,
      "num_input_tokens_seen": 31973904,
      "step": 55120
    },
    {
      "epoch": 8.210455764075068,
      "grad_norm": 9.0,
      "learning_rate": 3.669706218821776e-05,
      "loss": 0.8775,
      "num_input_tokens_seen": 31976784,
      "step": 55125
    },
    {
      "epoch": 8.211200476616026,
      "grad_norm": 11.375,
      "learning_rate": 3.669419028100169e-05,
      "loss": 1.01,
      "num_input_tokens_seen": 31980016,
      "step": 55130
    },
    {
      "epoch": 8.211945189156985,
      "grad_norm": 11.375,
      "learning_rate": 3.6691318176224156e-05,
      "loss": 0.8291,
      "num_input_tokens_seen": 31982896,
      "step": 55135
    },
    {
      "epoch": 8.212689901697944,
      "grad_norm": 7.4375,
      "learning_rate": 3.6688445873933686e-05,
      "loss": 1.1156,
      "num_input_tokens_seen": 31985872,
      "step": 55140
    },
    {
      "epoch": 8.213434614238905,
      "grad_norm": 11.875,
      "learning_rate": 3.66855733741788e-05,
      "loss": 0.7816,
      "num_input_tokens_seen": 31988624,
      "step": 55145
    },
    {
      "epoch": 8.214179326779863,
      "grad_norm": 12.1875,
      "learning_rate": 3.668270067700803e-05,
      "loss": 1.1635,
      "num_input_tokens_seen": 31991664,
      "step": 55150
    },
    {
      "epoch": 8.214924039320822,
      "grad_norm": 15.3125,
      "learning_rate": 3.667982778246991e-05,
      "loss": 1.105,
      "num_input_tokens_seen": 31994768,
      "step": 55155
    },
    {
      "epoch": 8.21566875186178,
      "grad_norm": 7.9375,
      "learning_rate": 3.6676954690612974e-05,
      "loss": 1.0666,
      "num_input_tokens_seen": 31997744,
      "step": 55160
    },
    {
      "epoch": 8.216413464402741,
      "grad_norm": 12.0625,
      "learning_rate": 3.6674081401485746e-05,
      "loss": 1.2057,
      "num_input_tokens_seen": 32000720,
      "step": 55165
    },
    {
      "epoch": 8.2171581769437,
      "grad_norm": 10.0,
      "learning_rate": 3.667120791513678e-05,
      "loss": 1.1064,
      "num_input_tokens_seen": 32003536,
      "step": 55170
    },
    {
      "epoch": 8.217902889484659,
      "grad_norm": 7.75,
      "learning_rate": 3.666833423161462e-05,
      "loss": 1.0469,
      "num_input_tokens_seen": 32006640,
      "step": 55175
    },
    {
      "epoch": 8.218647602025618,
      "grad_norm": 13.875,
      "learning_rate": 3.666546035096781e-05,
      "loss": 1.2333,
      "num_input_tokens_seen": 32009232,
      "step": 55180
    },
    {
      "epoch": 8.219392314566578,
      "grad_norm": 10.0,
      "learning_rate": 3.6662586273244906e-05,
      "loss": 1.2898,
      "num_input_tokens_seen": 32012112,
      "step": 55185
    },
    {
      "epoch": 8.220137027107537,
      "grad_norm": 12.3125,
      "learning_rate": 3.665971199849447e-05,
      "loss": 0.7844,
      "num_input_tokens_seen": 32014768,
      "step": 55190
    },
    {
      "epoch": 8.220881739648496,
      "grad_norm": 11.3125,
      "learning_rate": 3.665683752676503e-05,
      "loss": 1.0918,
      "num_input_tokens_seen": 32017776,
      "step": 55195
    },
    {
      "epoch": 8.221626452189454,
      "grad_norm": 7.96875,
      "learning_rate": 3.665396285810519e-05,
      "loss": 1.2636,
      "num_input_tokens_seen": 32020304,
      "step": 55200
    },
    {
      "epoch": 8.222371164730415,
      "grad_norm": 6.3125,
      "learning_rate": 3.665108799256348e-05,
      "loss": 0.9812,
      "num_input_tokens_seen": 32023152,
      "step": 55205
    },
    {
      "epoch": 8.223115877271374,
      "grad_norm": 12.875,
      "learning_rate": 3.6648212930188474e-05,
      "loss": 1.476,
      "num_input_tokens_seen": 32026032,
      "step": 55210
    },
    {
      "epoch": 8.223860589812332,
      "grad_norm": 9.625,
      "learning_rate": 3.664533767102876e-05,
      "loss": 1.0101,
      "num_input_tokens_seen": 32029168,
      "step": 55215
    },
    {
      "epoch": 8.224605302353291,
      "grad_norm": 10.25,
      "learning_rate": 3.664246221513289e-05,
      "loss": 1.0925,
      "num_input_tokens_seen": 32031952,
      "step": 55220
    },
    {
      "epoch": 8.22535001489425,
      "grad_norm": 14.0625,
      "learning_rate": 3.663958656254947e-05,
      "loss": 1.0518,
      "num_input_tokens_seen": 32034640,
      "step": 55225
    },
    {
      "epoch": 8.22609472743521,
      "grad_norm": 7.875,
      "learning_rate": 3.663671071332705e-05,
      "loss": 0.8837,
      "num_input_tokens_seen": 32037360,
      "step": 55230
    },
    {
      "epoch": 8.22683943997617,
      "grad_norm": 10.5625,
      "learning_rate": 3.6633834667514236e-05,
      "loss": 1.0661,
      "num_input_tokens_seen": 32040112,
      "step": 55235
    },
    {
      "epoch": 8.227584152517128,
      "grad_norm": 5.15625,
      "learning_rate": 3.663095842515961e-05,
      "loss": 0.9231,
      "num_input_tokens_seen": 32043152,
      "step": 55240
    },
    {
      "epoch": 8.228328865058087,
      "grad_norm": 12.0625,
      "learning_rate": 3.662808198631176e-05,
      "loss": 1.1352,
      "num_input_tokens_seen": 32045808,
      "step": 55245
    },
    {
      "epoch": 8.229073577599047,
      "grad_norm": 14.3125,
      "learning_rate": 3.662520535101928e-05,
      "loss": 0.7915,
      "num_input_tokens_seen": 32048656,
      "step": 55250
    },
    {
      "epoch": 8.229818290140006,
      "grad_norm": 9.75,
      "learning_rate": 3.662232851933079e-05,
      "loss": 1.1605,
      "num_input_tokens_seen": 32051696,
      "step": 55255
    },
    {
      "epoch": 8.230563002680965,
      "grad_norm": 10.1875,
      "learning_rate": 3.661945149129485e-05,
      "loss": 1.13,
      "num_input_tokens_seen": 32054768,
      "step": 55260
    },
    {
      "epoch": 8.231307715221924,
      "grad_norm": 19.625,
      "learning_rate": 3.661657426696009e-05,
      "loss": 0.9755,
      "num_input_tokens_seen": 32057392,
      "step": 55265
    },
    {
      "epoch": 8.232052427762884,
      "grad_norm": 4.25,
      "learning_rate": 3.6613696846375115e-05,
      "loss": 0.9758,
      "num_input_tokens_seen": 32060112,
      "step": 55270
    },
    {
      "epoch": 8.232797140303843,
      "grad_norm": 8.8125,
      "learning_rate": 3.661081922958854e-05,
      "loss": 1.1494,
      "num_input_tokens_seen": 32062832,
      "step": 55275
    },
    {
      "epoch": 8.233541852844802,
      "grad_norm": 4.21875,
      "learning_rate": 3.660794141664898e-05,
      "loss": 0.9933,
      "num_input_tokens_seen": 32065776,
      "step": 55280
    },
    {
      "epoch": 8.23428656538576,
      "grad_norm": 15.125,
      "learning_rate": 3.660506340760504e-05,
      "loss": 1.3013,
      "num_input_tokens_seen": 32068848,
      "step": 55285
    },
    {
      "epoch": 8.235031277926721,
      "grad_norm": 16.375,
      "learning_rate": 3.660218520250535e-05,
      "loss": 1.3268,
      "num_input_tokens_seen": 32071440,
      "step": 55290
    },
    {
      "epoch": 8.23577599046768,
      "grad_norm": 14.125,
      "learning_rate": 3.659930680139853e-05,
      "loss": 1.0384,
      "num_input_tokens_seen": 32074032,
      "step": 55295
    },
    {
      "epoch": 8.236520703008638,
      "grad_norm": 7.0,
      "learning_rate": 3.659642820433322e-05,
      "loss": 0.9916,
      "num_input_tokens_seen": 32076880,
      "step": 55300
    },
    {
      "epoch": 8.237265415549597,
      "grad_norm": 9.3125,
      "learning_rate": 3.659354941135803e-05,
      "loss": 1.1498,
      "num_input_tokens_seen": 32079984,
      "step": 55305
    },
    {
      "epoch": 8.238010128090558,
      "grad_norm": 8.25,
      "learning_rate": 3.65906704225216e-05,
      "loss": 0.9878,
      "num_input_tokens_seen": 32083152,
      "step": 55310
    },
    {
      "epoch": 8.238754840631517,
      "grad_norm": 9.5,
      "learning_rate": 3.658779123787259e-05,
      "loss": 1.1569,
      "num_input_tokens_seen": 32086064,
      "step": 55315
    },
    {
      "epoch": 8.239499553172475,
      "grad_norm": 10.125,
      "learning_rate": 3.6584911857459624e-05,
      "loss": 1.3899,
      "num_input_tokens_seen": 32088880,
      "step": 55320
    },
    {
      "epoch": 8.240244265713434,
      "grad_norm": 6.6875,
      "learning_rate": 3.6582032281331345e-05,
      "loss": 1.1995,
      "num_input_tokens_seen": 32091792,
      "step": 55325
    },
    {
      "epoch": 8.240988978254395,
      "grad_norm": 10.75,
      "learning_rate": 3.6579152509536395e-05,
      "loss": 0.8864,
      "num_input_tokens_seen": 32094544,
      "step": 55330
    },
    {
      "epoch": 8.241733690795353,
      "grad_norm": 7.5,
      "learning_rate": 3.6576272542123435e-05,
      "loss": 0.9398,
      "num_input_tokens_seen": 32097392,
      "step": 55335
    },
    {
      "epoch": 8.242478403336312,
      "grad_norm": 9.9375,
      "learning_rate": 3.657339237914111e-05,
      "loss": 0.8139,
      "num_input_tokens_seen": 32100400,
      "step": 55340
    },
    {
      "epoch": 8.24322311587727,
      "grad_norm": 9.0,
      "learning_rate": 3.657051202063809e-05,
      "loss": 1.0674,
      "num_input_tokens_seen": 32103344,
      "step": 55345
    },
    {
      "epoch": 8.243967828418231,
      "grad_norm": 14.4375,
      "learning_rate": 3.656763146666303e-05,
      "loss": 1.0195,
      "num_input_tokens_seen": 32106576,
      "step": 55350
    },
    {
      "epoch": 8.24471254095919,
      "grad_norm": 6.78125,
      "learning_rate": 3.6564750717264595e-05,
      "loss": 1.1275,
      "num_input_tokens_seen": 32109552,
      "step": 55355
    },
    {
      "epoch": 8.245457253500149,
      "grad_norm": 13.25,
      "learning_rate": 3.656186977249145e-05,
      "loss": 1.1138,
      "num_input_tokens_seen": 32112400,
      "step": 55360
    },
    {
      "epoch": 8.246201966041108,
      "grad_norm": 9.875,
      "learning_rate": 3.655898863239226e-05,
      "loss": 1.1674,
      "num_input_tokens_seen": 32115312,
      "step": 55365
    },
    {
      "epoch": 8.246946678582066,
      "grad_norm": 7.03125,
      "learning_rate": 3.655610729701571e-05,
      "loss": 1.1952,
      "num_input_tokens_seen": 32118224,
      "step": 55370
    },
    {
      "epoch": 8.247691391123027,
      "grad_norm": 8.1875,
      "learning_rate": 3.655322576641047e-05,
      "loss": 1.3614,
      "num_input_tokens_seen": 32121072,
      "step": 55375
    },
    {
      "epoch": 8.248436103663986,
      "grad_norm": 7.09375,
      "learning_rate": 3.655034404062522e-05,
      "loss": 0.8798,
      "num_input_tokens_seen": 32124112,
      "step": 55380
    },
    {
      "epoch": 8.249180816204944,
      "grad_norm": 12.0,
      "learning_rate": 3.654746211970865e-05,
      "loss": 1.1078,
      "num_input_tokens_seen": 32126960,
      "step": 55385
    },
    {
      "epoch": 8.249925528745903,
      "grad_norm": 9.0,
      "learning_rate": 3.654458000370945e-05,
      "loss": 1.1798,
      "num_input_tokens_seen": 32129936,
      "step": 55390
    },
    {
      "epoch": 8.250670241286864,
      "grad_norm": 11.5,
      "learning_rate": 3.65416976926763e-05,
      "loss": 0.794,
      "num_input_tokens_seen": 32132560,
      "step": 55395
    },
    {
      "epoch": 8.251414953827823,
      "grad_norm": 6.9375,
      "learning_rate": 3.653881518665789e-05,
      "loss": 0.9798,
      "num_input_tokens_seen": 32135600,
      "step": 55400
    },
    {
      "epoch": 8.252159666368781,
      "grad_norm": 11.1875,
      "learning_rate": 3.6535932485702933e-05,
      "loss": 1.0907,
      "num_input_tokens_seen": 32138512,
      "step": 55405
    },
    {
      "epoch": 8.25290437890974,
      "grad_norm": 7.40625,
      "learning_rate": 3.6533049589860115e-05,
      "loss": 0.7856,
      "num_input_tokens_seen": 32141488,
      "step": 55410
    },
    {
      "epoch": 8.2536490914507,
      "grad_norm": 12.0,
      "learning_rate": 3.653016649917815e-05,
      "loss": 1.0423,
      "num_input_tokens_seen": 32144464,
      "step": 55415
    },
    {
      "epoch": 8.25439380399166,
      "grad_norm": 8.5,
      "learning_rate": 3.652728321370575e-05,
      "loss": 1.074,
      "num_input_tokens_seen": 32147472,
      "step": 55420
    },
    {
      "epoch": 8.255138516532618,
      "grad_norm": 5.59375,
      "learning_rate": 3.6524399733491605e-05,
      "loss": 1.052,
      "num_input_tokens_seen": 32150256,
      "step": 55425
    },
    {
      "epoch": 8.255883229073577,
      "grad_norm": 15.3125,
      "learning_rate": 3.652151605858444e-05,
      "loss": 0.9376,
      "num_input_tokens_seen": 32153296,
      "step": 55430
    },
    {
      "epoch": 8.256627941614537,
      "grad_norm": 17.25,
      "learning_rate": 3.651863218903297e-05,
      "loss": 1.2299,
      "num_input_tokens_seen": 32156304,
      "step": 55435
    },
    {
      "epoch": 8.257372654155496,
      "grad_norm": 8.75,
      "learning_rate": 3.651574812488592e-05,
      "loss": 0.92,
      "num_input_tokens_seen": 32158896,
      "step": 55440
    },
    {
      "epoch": 8.258117366696455,
      "grad_norm": 9.75,
      "learning_rate": 3.651286386619201e-05,
      "loss": 0.7084,
      "num_input_tokens_seen": 32161680,
      "step": 55445
    },
    {
      "epoch": 8.258862079237414,
      "grad_norm": 12.5,
      "learning_rate": 3.650997941299996e-05,
      "loss": 1.3006,
      "num_input_tokens_seen": 32164752,
      "step": 55450
    },
    {
      "epoch": 8.259606791778374,
      "grad_norm": 6.75,
      "learning_rate": 3.650709476535852e-05,
      "loss": 0.7529,
      "num_input_tokens_seen": 32167600,
      "step": 55455
    },
    {
      "epoch": 8.260351504319333,
      "grad_norm": 8.6875,
      "learning_rate": 3.6504209923316396e-05,
      "loss": 1.0338,
      "num_input_tokens_seen": 32170384,
      "step": 55460
    },
    {
      "epoch": 8.261096216860292,
      "grad_norm": 5.875,
      "learning_rate": 3.650132488692234e-05,
      "loss": 1.2739,
      "num_input_tokens_seen": 32173392,
      "step": 55465
    },
    {
      "epoch": 8.26184092940125,
      "grad_norm": 8.3125,
      "learning_rate": 3.649843965622509e-05,
      "loss": 0.8802,
      "num_input_tokens_seen": 32176464,
      "step": 55470
    },
    {
      "epoch": 8.262585641942211,
      "grad_norm": 15.75,
      "learning_rate": 3.6495554231273386e-05,
      "loss": 1.2547,
      "num_input_tokens_seen": 32179632,
      "step": 55475
    },
    {
      "epoch": 8.26333035448317,
      "grad_norm": 8.3125,
      "learning_rate": 3.6492668612115986e-05,
      "loss": 0.9521,
      "num_input_tokens_seen": 32182512,
      "step": 55480
    },
    {
      "epoch": 8.264075067024129,
      "grad_norm": 7.65625,
      "learning_rate": 3.648978279880162e-05,
      "loss": 1.2679,
      "num_input_tokens_seen": 32185264,
      "step": 55485
    },
    {
      "epoch": 8.264819779565087,
      "grad_norm": 10.75,
      "learning_rate": 3.648689679137906e-05,
      "loss": 1.4574,
      "num_input_tokens_seen": 32188272,
      "step": 55490
    },
    {
      "epoch": 8.265564492106048,
      "grad_norm": 10.5,
      "learning_rate": 3.648401058989705e-05,
      "loss": 0.9279,
      "num_input_tokens_seen": 32191248,
      "step": 55495
    },
    {
      "epoch": 8.266309204647007,
      "grad_norm": 12.75,
      "learning_rate": 3.6481124194404345e-05,
      "loss": 1.0989,
      "num_input_tokens_seen": 32193840,
      "step": 55500
    },
    {
      "epoch": 8.267053917187965,
      "grad_norm": 14.125,
      "learning_rate": 3.6478237604949725e-05,
      "loss": 0.953,
      "num_input_tokens_seen": 32196848,
      "step": 55505
    },
    {
      "epoch": 8.267798629728924,
      "grad_norm": 15.4375,
      "learning_rate": 3.647535082158194e-05,
      "loss": 1.0582,
      "num_input_tokens_seen": 32199568,
      "step": 55510
    },
    {
      "epoch": 8.268543342269885,
      "grad_norm": 7.84375,
      "learning_rate": 3.6472463844349776e-05,
      "loss": 0.9934,
      "num_input_tokens_seen": 32202288,
      "step": 55515
    },
    {
      "epoch": 8.269288054810843,
      "grad_norm": 12.3125,
      "learning_rate": 3.646957667330199e-05,
      "loss": 0.8331,
      "num_input_tokens_seen": 32205136,
      "step": 55520
    },
    {
      "epoch": 8.270032767351802,
      "grad_norm": 6.625,
      "learning_rate": 3.6466689308487364e-05,
      "loss": 1.0567,
      "num_input_tokens_seen": 32207952,
      "step": 55525
    },
    {
      "epoch": 8.270777479892761,
      "grad_norm": 8.625,
      "learning_rate": 3.646380174995468e-05,
      "loss": 1.389,
      "num_input_tokens_seen": 32210704,
      "step": 55530
    },
    {
      "epoch": 8.271522192433721,
      "grad_norm": 13.8125,
      "learning_rate": 3.646091399775271e-05,
      "loss": 1.3171,
      "num_input_tokens_seen": 32213392,
      "step": 55535
    },
    {
      "epoch": 8.27226690497468,
      "grad_norm": 9.875,
      "learning_rate": 3.645802605193025e-05,
      "loss": 1.1276,
      "num_input_tokens_seen": 32216336,
      "step": 55540
    },
    {
      "epoch": 8.273011617515639,
      "grad_norm": 14.375,
      "learning_rate": 3.645513791253608e-05,
      "loss": 1.1927,
      "num_input_tokens_seen": 32219536,
      "step": 55545
    },
    {
      "epoch": 8.273756330056598,
      "grad_norm": 5.46875,
      "learning_rate": 3.6452249579619005e-05,
      "loss": 0.9103,
      "num_input_tokens_seen": 32222224,
      "step": 55550
    },
    {
      "epoch": 8.274501042597556,
      "grad_norm": 5.84375,
      "learning_rate": 3.644936105322781e-05,
      "loss": 1.0652,
      "num_input_tokens_seen": 32225072,
      "step": 55555
    },
    {
      "epoch": 8.275245755138517,
      "grad_norm": 7.3125,
      "learning_rate": 3.644647233341129e-05,
      "loss": 0.9534,
      "num_input_tokens_seen": 32227760,
      "step": 55560
    },
    {
      "epoch": 8.275990467679476,
      "grad_norm": 7.78125,
      "learning_rate": 3.644358342021826e-05,
      "loss": 1.0647,
      "num_input_tokens_seen": 32230704,
      "step": 55565
    },
    {
      "epoch": 8.276735180220435,
      "grad_norm": 8.375,
      "learning_rate": 3.644069431369752e-05,
      "loss": 0.8335,
      "num_input_tokens_seen": 32233744,
      "step": 55570
    },
    {
      "epoch": 8.277479892761393,
      "grad_norm": 9.875,
      "learning_rate": 3.643780501389787e-05,
      "loss": 1.1842,
      "num_input_tokens_seen": 32236656,
      "step": 55575
    },
    {
      "epoch": 8.278224605302354,
      "grad_norm": 14.125,
      "learning_rate": 3.643491552086814e-05,
      "loss": 1.2873,
      "num_input_tokens_seen": 32239344,
      "step": 55580
    },
    {
      "epoch": 8.278969317843313,
      "grad_norm": 8.5625,
      "learning_rate": 3.643202583465713e-05,
      "loss": 1.2789,
      "num_input_tokens_seen": 32242352,
      "step": 55585
    },
    {
      "epoch": 8.279714030384271,
      "grad_norm": 11.5,
      "learning_rate": 3.6429135955313664e-05,
      "loss": 0.8936,
      "num_input_tokens_seen": 32245392,
      "step": 55590
    },
    {
      "epoch": 8.28045874292523,
      "grad_norm": 6.8125,
      "learning_rate": 3.6426245882886554e-05,
      "loss": 0.965,
      "num_input_tokens_seen": 32248272,
      "step": 55595
    },
    {
      "epoch": 8.28120345546619,
      "grad_norm": 7.5,
      "learning_rate": 3.6423355617424634e-05,
      "loss": 1.5655,
      "num_input_tokens_seen": 32250928,
      "step": 55600
    },
    {
      "epoch": 8.28194816800715,
      "grad_norm": 14.3125,
      "learning_rate": 3.6420465158976746e-05,
      "loss": 1.25,
      "num_input_tokens_seen": 32254000,
      "step": 55605
    },
    {
      "epoch": 8.282692880548108,
      "grad_norm": 11.0625,
      "learning_rate": 3.6417574507591694e-05,
      "loss": 1.1128,
      "num_input_tokens_seen": 32257008,
      "step": 55610
    },
    {
      "epoch": 8.283437593089067,
      "grad_norm": 8.8125,
      "learning_rate": 3.641468366331833e-05,
      "loss": 0.9049,
      "num_input_tokens_seen": 32259792,
      "step": 55615
    },
    {
      "epoch": 8.284182305630027,
      "grad_norm": 9.8125,
      "learning_rate": 3.641179262620548e-05,
      "loss": 1.0256,
      "num_input_tokens_seen": 32262928,
      "step": 55620
    },
    {
      "epoch": 8.284927018170986,
      "grad_norm": 11.375,
      "learning_rate": 3.6408901396301995e-05,
      "loss": 0.9899,
      "num_input_tokens_seen": 32265840,
      "step": 55625
    },
    {
      "epoch": 8.285671730711945,
      "grad_norm": 9.5625,
      "learning_rate": 3.640600997365672e-05,
      "loss": 0.8133,
      "num_input_tokens_seen": 32268752,
      "step": 55630
    },
    {
      "epoch": 8.286416443252904,
      "grad_norm": 18.5,
      "learning_rate": 3.64031183583185e-05,
      "loss": 1.2298,
      "num_input_tokens_seen": 32271312,
      "step": 55635
    },
    {
      "epoch": 8.287161155793864,
      "grad_norm": 7.40625,
      "learning_rate": 3.6400226550336166e-05,
      "loss": 0.8464,
      "num_input_tokens_seen": 32274352,
      "step": 55640
    },
    {
      "epoch": 8.287905868334823,
      "grad_norm": 16.375,
      "learning_rate": 3.6397334549758614e-05,
      "loss": 1.0983,
      "num_input_tokens_seen": 32277104,
      "step": 55645
    },
    {
      "epoch": 8.288650580875782,
      "grad_norm": 11.375,
      "learning_rate": 3.639444235663467e-05,
      "loss": 0.9283,
      "num_input_tokens_seen": 32279824,
      "step": 55650
    },
    {
      "epoch": 8.28939529341674,
      "grad_norm": 11.0,
      "learning_rate": 3.6391549971013213e-05,
      "loss": 1.0993,
      "num_input_tokens_seen": 32282576,
      "step": 55655
    },
    {
      "epoch": 8.290140005957701,
      "grad_norm": 22.5,
      "learning_rate": 3.638865739294308e-05,
      "loss": 1.1753,
      "num_input_tokens_seen": 32285520,
      "step": 55660
    },
    {
      "epoch": 8.29088471849866,
      "grad_norm": 8.1875,
      "learning_rate": 3.6385764622473164e-05,
      "loss": 1.0715,
      "num_input_tokens_seen": 32288464,
      "step": 55665
    },
    {
      "epoch": 8.291629431039619,
      "grad_norm": 7.25,
      "learning_rate": 3.6382871659652326e-05,
      "loss": 1.0403,
      "num_input_tokens_seen": 32291248,
      "step": 55670
    },
    {
      "epoch": 8.292374143580577,
      "grad_norm": 18.625,
      "learning_rate": 3.6379978504529436e-05,
      "loss": 1.2326,
      "num_input_tokens_seen": 32294352,
      "step": 55675
    },
    {
      "epoch": 8.293118856121538,
      "grad_norm": 8.9375,
      "learning_rate": 3.637708515715338e-05,
      "loss": 0.7678,
      "num_input_tokens_seen": 32297200,
      "step": 55680
    },
    {
      "epoch": 8.293863568662497,
      "grad_norm": 13.5625,
      "learning_rate": 3.637419161757304e-05,
      "loss": 1.2898,
      "num_input_tokens_seen": 32299888,
      "step": 55685
    },
    {
      "epoch": 8.294608281203455,
      "grad_norm": 7.0,
      "learning_rate": 3.637129788583729e-05,
      "loss": 1.0842,
      "num_input_tokens_seen": 32302544,
      "step": 55690
    },
    {
      "epoch": 8.295352993744414,
      "grad_norm": 7.5625,
      "learning_rate": 3.6368403961995014e-05,
      "loss": 0.9074,
      "num_input_tokens_seen": 32305008,
      "step": 55695
    },
    {
      "epoch": 8.296097706285375,
      "grad_norm": 6.8125,
      "learning_rate": 3.63655098460951e-05,
      "loss": 1.0284,
      "num_input_tokens_seen": 32308304,
      "step": 55700
    },
    {
      "epoch": 8.296842418826333,
      "grad_norm": 11.625,
      "learning_rate": 3.636261553818646e-05,
      "loss": 0.9251,
      "num_input_tokens_seen": 32311120,
      "step": 55705
    },
    {
      "epoch": 8.297587131367292,
      "grad_norm": 10.1875,
      "learning_rate": 3.6359721038317976e-05,
      "loss": 1.3504,
      "num_input_tokens_seen": 32313968,
      "step": 55710
    },
    {
      "epoch": 8.298331843908251,
      "grad_norm": 16.375,
      "learning_rate": 3.6356826346538555e-05,
      "loss": 1.3208,
      "num_input_tokens_seen": 32316912,
      "step": 55715
    },
    {
      "epoch": 8.299076556449211,
      "grad_norm": 8.0625,
      "learning_rate": 3.635393146289709e-05,
      "loss": 1.0091,
      "num_input_tokens_seen": 32319888,
      "step": 55720
    },
    {
      "epoch": 8.29982126899017,
      "grad_norm": 18.375,
      "learning_rate": 3.6351036387442496e-05,
      "loss": 1.3269,
      "num_input_tokens_seen": 32322768,
      "step": 55725
    },
    {
      "epoch": 8.300565981531129,
      "grad_norm": 9.9375,
      "learning_rate": 3.634814112022368e-05,
      "loss": 1.1117,
      "num_input_tokens_seen": 32325552,
      "step": 55730
    },
    {
      "epoch": 8.301310694072088,
      "grad_norm": 6.09375,
      "learning_rate": 3.634524566128955e-05,
      "loss": 1.0225,
      "num_input_tokens_seen": 32328528,
      "step": 55735
    },
    {
      "epoch": 8.302055406613047,
      "grad_norm": 12.8125,
      "learning_rate": 3.6342350010689017e-05,
      "loss": 1.2469,
      "num_input_tokens_seen": 32331312,
      "step": 55740
    },
    {
      "epoch": 8.302800119154007,
      "grad_norm": 18.0,
      "learning_rate": 3.633945416847102e-05,
      "loss": 1.1269,
      "num_input_tokens_seen": 32334160,
      "step": 55745
    },
    {
      "epoch": 8.303544831694966,
      "grad_norm": 11.1875,
      "learning_rate": 3.6336558134684465e-05,
      "loss": 1.2781,
      "num_input_tokens_seen": 32336944,
      "step": 55750
    },
    {
      "epoch": 8.304289544235925,
      "grad_norm": 10.4375,
      "learning_rate": 3.6333661909378286e-05,
      "loss": 0.8576,
      "num_input_tokens_seen": 32339952,
      "step": 55755
    },
    {
      "epoch": 8.305034256776883,
      "grad_norm": 7.59375,
      "learning_rate": 3.63307654926014e-05,
      "loss": 1.1373,
      "num_input_tokens_seen": 32342704,
      "step": 55760
    },
    {
      "epoch": 8.305778969317844,
      "grad_norm": 8.625,
      "learning_rate": 3.632786888440276e-05,
      "loss": 1.1454,
      "num_input_tokens_seen": 32345616,
      "step": 55765
    },
    {
      "epoch": 8.306523681858803,
      "grad_norm": 10.125,
      "learning_rate": 3.6324972084831284e-05,
      "loss": 1.3016,
      "num_input_tokens_seen": 32348432,
      "step": 55770
    },
    {
      "epoch": 8.307268394399761,
      "grad_norm": 10.375,
      "learning_rate": 3.632207509393591e-05,
      "loss": 1.1113,
      "num_input_tokens_seen": 32351440,
      "step": 55775
    },
    {
      "epoch": 8.30801310694072,
      "grad_norm": 6.90625,
      "learning_rate": 3.6319177911765583e-05,
      "loss": 0.5805,
      "num_input_tokens_seen": 32354352,
      "step": 55780
    },
    {
      "epoch": 8.30875781948168,
      "grad_norm": 11.25,
      "learning_rate": 3.631628053836926e-05,
      "loss": 1.0852,
      "num_input_tokens_seen": 32357264,
      "step": 55785
    },
    {
      "epoch": 8.30950253202264,
      "grad_norm": 5.375,
      "learning_rate": 3.631338297379587e-05,
      "loss": 0.9356,
      "num_input_tokens_seen": 32360016,
      "step": 55790
    },
    {
      "epoch": 8.310247244563598,
      "grad_norm": 9.625,
      "learning_rate": 3.631048521809437e-05,
      "loss": 1.0879,
      "num_input_tokens_seen": 32363216,
      "step": 55795
    },
    {
      "epoch": 8.310991957104557,
      "grad_norm": 9.875,
      "learning_rate": 3.630758727131373e-05,
      "loss": 0.7848,
      "num_input_tokens_seen": 32366000,
      "step": 55800
    },
    {
      "epoch": 8.311736669645517,
      "grad_norm": 14.6875,
      "learning_rate": 3.6304689133502884e-05,
      "loss": 1.2472,
      "num_input_tokens_seen": 32369040,
      "step": 55805
    },
    {
      "epoch": 8.312481382186476,
      "grad_norm": 8.375,
      "learning_rate": 3.630179080471081e-05,
      "loss": 1.0401,
      "num_input_tokens_seen": 32372208,
      "step": 55810
    },
    {
      "epoch": 8.313226094727435,
      "grad_norm": 9.8125,
      "learning_rate": 3.629889228498646e-05,
      "loss": 1.0312,
      "num_input_tokens_seen": 32375344,
      "step": 55815
    },
    {
      "epoch": 8.313970807268394,
      "grad_norm": 8.5,
      "learning_rate": 3.629599357437882e-05,
      "loss": 1.0978,
      "num_input_tokens_seen": 32378480,
      "step": 55820
    },
    {
      "epoch": 8.314715519809354,
      "grad_norm": 12.3125,
      "learning_rate": 3.629309467293685e-05,
      "loss": 0.8772,
      "num_input_tokens_seen": 32381328,
      "step": 55825
    },
    {
      "epoch": 8.315460232350313,
      "grad_norm": 8.6875,
      "learning_rate": 3.6290195580709505e-05,
      "loss": 1.0478,
      "num_input_tokens_seen": 32384240,
      "step": 55830
    },
    {
      "epoch": 8.316204944891272,
      "grad_norm": 8.8125,
      "learning_rate": 3.62872962977458e-05,
      "loss": 1.3061,
      "num_input_tokens_seen": 32387280,
      "step": 55835
    },
    {
      "epoch": 8.31694965743223,
      "grad_norm": 8.75,
      "learning_rate": 3.628439682409468e-05,
      "loss": 1.1911,
      "num_input_tokens_seen": 32389936,
      "step": 55840
    },
    {
      "epoch": 8.317694369973191,
      "grad_norm": 12.75,
      "learning_rate": 3.628149715980516e-05,
      "loss": 1.0522,
      "num_input_tokens_seen": 32393008,
      "step": 55845
    },
    {
      "epoch": 8.31843908251415,
      "grad_norm": 12.8125,
      "learning_rate": 3.62785973049262e-05,
      "loss": 1.1229,
      "num_input_tokens_seen": 32395792,
      "step": 55850
    },
    {
      "epoch": 8.319183795055109,
      "grad_norm": 10.25,
      "learning_rate": 3.627569725950681e-05,
      "loss": 0.8263,
      "num_input_tokens_seen": 32398800,
      "step": 55855
    },
    {
      "epoch": 8.319928507596067,
      "grad_norm": 15.75,
      "learning_rate": 3.6272797023595974e-05,
      "loss": 0.8378,
      "num_input_tokens_seen": 32401328,
      "step": 55860
    },
    {
      "epoch": 8.320673220137028,
      "grad_norm": 11.0625,
      "learning_rate": 3.626989659724268e-05,
      "loss": 1.2081,
      "num_input_tokens_seen": 32404560,
      "step": 55865
    },
    {
      "epoch": 8.321417932677987,
      "grad_norm": 15.5,
      "learning_rate": 3.626699598049594e-05,
      "loss": 0.965,
      "num_input_tokens_seen": 32407184,
      "step": 55870
    },
    {
      "epoch": 8.322162645218945,
      "grad_norm": 5.4375,
      "learning_rate": 3.626409517340476e-05,
      "loss": 1.2439,
      "num_input_tokens_seen": 32410416,
      "step": 55875
    },
    {
      "epoch": 8.322907357759904,
      "grad_norm": 7.59375,
      "learning_rate": 3.626119417601814e-05,
      "loss": 1.0942,
      "num_input_tokens_seen": 32413648,
      "step": 55880
    },
    {
      "epoch": 8.323652070300863,
      "grad_norm": 13.0,
      "learning_rate": 3.625829298838509e-05,
      "loss": 1.0558,
      "num_input_tokens_seen": 32416368,
      "step": 55885
    },
    {
      "epoch": 8.324396782841823,
      "grad_norm": 12.0625,
      "learning_rate": 3.6255391610554624e-05,
      "loss": 1.2877,
      "num_input_tokens_seen": 32419248,
      "step": 55890
    },
    {
      "epoch": 8.325141495382782,
      "grad_norm": 6.5,
      "learning_rate": 3.625249004257575e-05,
      "loss": 1.005,
      "num_input_tokens_seen": 32422384,
      "step": 55895
    },
    {
      "epoch": 8.325886207923741,
      "grad_norm": 11.5625,
      "learning_rate": 3.6249588284497496e-05,
      "loss": 0.7014,
      "num_input_tokens_seen": 32425232,
      "step": 55900
    },
    {
      "epoch": 8.3266309204647,
      "grad_norm": 16.875,
      "learning_rate": 3.624668633636888e-05,
      "loss": 1.3042,
      "num_input_tokens_seen": 32428240,
      "step": 55905
    },
    {
      "epoch": 8.32737563300566,
      "grad_norm": 11.125,
      "learning_rate": 3.624378419823893e-05,
      "loss": 1.2166,
      "num_input_tokens_seen": 32431088,
      "step": 55910
    },
    {
      "epoch": 8.328120345546619,
      "grad_norm": 9.75,
      "learning_rate": 3.624088187015668e-05,
      "loss": 0.6273,
      "num_input_tokens_seen": 32433744,
      "step": 55915
    },
    {
      "epoch": 8.328865058087578,
      "grad_norm": 8.25,
      "learning_rate": 3.623797935217115e-05,
      "loss": 1.1471,
      "num_input_tokens_seen": 32436976,
      "step": 55920
    },
    {
      "epoch": 8.329609770628537,
      "grad_norm": 17.375,
      "learning_rate": 3.623507664433138e-05,
      "loss": 1.1567,
      "num_input_tokens_seen": 32439696,
      "step": 55925
    },
    {
      "epoch": 8.330354483169497,
      "grad_norm": 13.4375,
      "learning_rate": 3.6232173746686405e-05,
      "loss": 1.2278,
      "num_input_tokens_seen": 32442288,
      "step": 55930
    },
    {
      "epoch": 8.331099195710456,
      "grad_norm": 9.125,
      "learning_rate": 3.6229270659285276e-05,
      "loss": 1.1084,
      "num_input_tokens_seen": 32445136,
      "step": 55935
    },
    {
      "epoch": 8.331843908251415,
      "grad_norm": 12.3125,
      "learning_rate": 3.622636738217703e-05,
      "loss": 0.9356,
      "num_input_tokens_seen": 32448080,
      "step": 55940
    },
    {
      "epoch": 8.332588620792373,
      "grad_norm": 6.21875,
      "learning_rate": 3.6223463915410714e-05,
      "loss": 1.2655,
      "num_input_tokens_seen": 32450928,
      "step": 55945
    },
    {
      "epoch": 8.333333333333334,
      "grad_norm": 36.0,
      "learning_rate": 3.622056025903539e-05,
      "loss": 1.3386,
      "num_input_tokens_seen": 32453808,
      "step": 55950
    },
    {
      "epoch": 8.334078045874293,
      "grad_norm": 7.03125,
      "learning_rate": 3.62176564131001e-05,
      "loss": 1.167,
      "num_input_tokens_seen": 32456816,
      "step": 55955
    },
    {
      "epoch": 8.334822758415251,
      "grad_norm": 9.5,
      "learning_rate": 3.62147523776539e-05,
      "loss": 1.1462,
      "num_input_tokens_seen": 32459760,
      "step": 55960
    },
    {
      "epoch": 8.33556747095621,
      "grad_norm": 7.65625,
      "learning_rate": 3.621184815274587e-05,
      "loss": 1.0867,
      "num_input_tokens_seen": 32462672,
      "step": 55965
    },
    {
      "epoch": 8.33631218349717,
      "grad_norm": 12.9375,
      "learning_rate": 3.620894373842505e-05,
      "loss": 1.1529,
      "num_input_tokens_seen": 32465648,
      "step": 55970
    },
    {
      "epoch": 8.33705689603813,
      "grad_norm": 7.21875,
      "learning_rate": 3.6206039134740525e-05,
      "loss": 1.1448,
      "num_input_tokens_seen": 32468496,
      "step": 55975
    },
    {
      "epoch": 8.337801608579088,
      "grad_norm": 9.0,
      "learning_rate": 3.620313434174135e-05,
      "loss": 1.2027,
      "num_input_tokens_seen": 32471440,
      "step": 55980
    },
    {
      "epoch": 8.338546321120047,
      "grad_norm": 12.3125,
      "learning_rate": 3.6200229359476614e-05,
      "loss": 0.9448,
      "num_input_tokens_seen": 32474384,
      "step": 55985
    },
    {
      "epoch": 8.339291033661008,
      "grad_norm": 10.375,
      "learning_rate": 3.6197324187995384e-05,
      "loss": 1.1985,
      "num_input_tokens_seen": 32477360,
      "step": 55990
    },
    {
      "epoch": 8.340035746201966,
      "grad_norm": 13.25,
      "learning_rate": 3.6194418827346746e-05,
      "loss": 0.9704,
      "num_input_tokens_seen": 32480528,
      "step": 55995
    },
    {
      "epoch": 8.340780458742925,
      "grad_norm": 4.28125,
      "learning_rate": 3.619151327757977e-05,
      "loss": 0.9519,
      "num_input_tokens_seen": 32483152,
      "step": 56000
    },
    {
      "epoch": 8.341525171283884,
      "grad_norm": 6.46875,
      "learning_rate": 3.6188607538743556e-05,
      "loss": 0.7731,
      "num_input_tokens_seen": 32485744,
      "step": 56005
    },
    {
      "epoch": 8.342269883824844,
      "grad_norm": 16.0,
      "learning_rate": 3.618570161088719e-05,
      "loss": 1.3937,
      "num_input_tokens_seen": 32488784,
      "step": 56010
    },
    {
      "epoch": 8.343014596365803,
      "grad_norm": 7.71875,
      "learning_rate": 3.6182795494059764e-05,
      "loss": 1.1674,
      "num_input_tokens_seen": 32491856,
      "step": 56015
    },
    {
      "epoch": 8.343759308906762,
      "grad_norm": 13.0625,
      "learning_rate": 3.617988918831038e-05,
      "loss": 1.035,
      "num_input_tokens_seen": 32494672,
      "step": 56020
    },
    {
      "epoch": 8.34450402144772,
      "grad_norm": 10.0625,
      "learning_rate": 3.617698269368812e-05,
      "loss": 1.005,
      "num_input_tokens_seen": 32497392,
      "step": 56025
    },
    {
      "epoch": 8.345248733988681,
      "grad_norm": 10.875,
      "learning_rate": 3.61740760102421e-05,
      "loss": 0.94,
      "num_input_tokens_seen": 32500528,
      "step": 56030
    },
    {
      "epoch": 8.34599344652964,
      "grad_norm": 14.1875,
      "learning_rate": 3.617116913802143e-05,
      "loss": 1.3275,
      "num_input_tokens_seen": 32503536,
      "step": 56035
    },
    {
      "epoch": 8.346738159070599,
      "grad_norm": 15.25,
      "learning_rate": 3.61682620770752e-05,
      "loss": 1.2905,
      "num_input_tokens_seen": 32506928,
      "step": 56040
    },
    {
      "epoch": 8.347482871611557,
      "grad_norm": 7.21875,
      "learning_rate": 3.616535482745254e-05,
      "loss": 1.29,
      "num_input_tokens_seen": 32509648,
      "step": 56045
    },
    {
      "epoch": 8.348227584152518,
      "grad_norm": 13.25,
      "learning_rate": 3.616244738920256e-05,
      "loss": 0.7738,
      "num_input_tokens_seen": 32512720,
      "step": 56050
    },
    {
      "epoch": 8.348972296693477,
      "grad_norm": 10.625,
      "learning_rate": 3.615953976237438e-05,
      "loss": 1.0803,
      "num_input_tokens_seen": 32515504,
      "step": 56055
    },
    {
      "epoch": 8.349717009234435,
      "grad_norm": 12.375,
      "learning_rate": 3.6156631947017106e-05,
      "loss": 1.1467,
      "num_input_tokens_seen": 32518256,
      "step": 56060
    },
    {
      "epoch": 8.350461721775394,
      "grad_norm": 14.5,
      "learning_rate": 3.6153723943179876e-05,
      "loss": 1.1392,
      "num_input_tokens_seen": 32520816,
      "step": 56065
    },
    {
      "epoch": 8.351206434316353,
      "grad_norm": 13.4375,
      "learning_rate": 3.6150815750911825e-05,
      "loss": 0.9053,
      "num_input_tokens_seen": 32523920,
      "step": 56070
    },
    {
      "epoch": 8.351951146857314,
      "grad_norm": 10.9375,
      "learning_rate": 3.614790737026207e-05,
      "loss": 0.9929,
      "num_input_tokens_seen": 32526960,
      "step": 56075
    },
    {
      "epoch": 8.352695859398272,
      "grad_norm": 16.0,
      "learning_rate": 3.614499880127975e-05,
      "loss": 1.2656,
      "num_input_tokens_seen": 32529936,
      "step": 56080
    },
    {
      "epoch": 8.353440571939231,
      "grad_norm": 13.5625,
      "learning_rate": 3.6142090044014e-05,
      "loss": 1.1981,
      "num_input_tokens_seen": 32533072,
      "step": 56085
    },
    {
      "epoch": 8.35418528448019,
      "grad_norm": 8.5625,
      "learning_rate": 3.613918109851397e-05,
      "loss": 1.0677,
      "num_input_tokens_seen": 32536080,
      "step": 56090
    },
    {
      "epoch": 8.35492999702115,
      "grad_norm": 8.4375,
      "learning_rate": 3.613627196482879e-05,
      "loss": 1.1032,
      "num_input_tokens_seen": 32539056,
      "step": 56095
    },
    {
      "epoch": 8.35567470956211,
      "grad_norm": 12.25,
      "learning_rate": 3.613336264300762e-05,
      "loss": 0.7823,
      "num_input_tokens_seen": 32541872,
      "step": 56100
    },
    {
      "epoch": 8.356419422103068,
      "grad_norm": 8.125,
      "learning_rate": 3.613045313309959e-05,
      "loss": 0.8169,
      "num_input_tokens_seen": 32545872,
      "step": 56105
    },
    {
      "epoch": 8.357164134644027,
      "grad_norm": 31.25,
      "learning_rate": 3.612754343515388e-05,
      "loss": 0.9422,
      "num_input_tokens_seen": 32548592,
      "step": 56110
    },
    {
      "epoch": 8.357908847184987,
      "grad_norm": 5.5625,
      "learning_rate": 3.612463354921963e-05,
      "loss": 0.8988,
      "num_input_tokens_seen": 32551472,
      "step": 56115
    },
    {
      "epoch": 8.358653559725946,
      "grad_norm": 12.9375,
      "learning_rate": 3.6121723475346006e-05,
      "loss": 1.2567,
      "num_input_tokens_seen": 32554320,
      "step": 56120
    },
    {
      "epoch": 8.359398272266905,
      "grad_norm": 21.75,
      "learning_rate": 3.6118813213582156e-05,
      "loss": 1.3764,
      "num_input_tokens_seen": 32557200,
      "step": 56125
    },
    {
      "epoch": 8.360142984807863,
      "grad_norm": 15.9375,
      "learning_rate": 3.611590276397727e-05,
      "loss": 1.3303,
      "num_input_tokens_seen": 32560048,
      "step": 56130
    },
    {
      "epoch": 8.360887697348824,
      "grad_norm": 4.875,
      "learning_rate": 3.6112992126580505e-05,
      "loss": 1.3472,
      "num_input_tokens_seen": 32563056,
      "step": 56135
    },
    {
      "epoch": 8.361632409889783,
      "grad_norm": 15.25,
      "learning_rate": 3.611008130144102e-05,
      "loss": 1.3435,
      "num_input_tokens_seen": 32565808,
      "step": 56140
    },
    {
      "epoch": 8.362377122430741,
      "grad_norm": 11.375,
      "learning_rate": 3.610717028860801e-05,
      "loss": 1.1083,
      "num_input_tokens_seen": 32568560,
      "step": 56145
    },
    {
      "epoch": 8.3631218349717,
      "grad_norm": 12.875,
      "learning_rate": 3.6104259088130655e-05,
      "loss": 0.9535,
      "num_input_tokens_seen": 32571216,
      "step": 56150
    },
    {
      "epoch": 8.36386654751266,
      "grad_norm": 7.875,
      "learning_rate": 3.6101347700058116e-05,
      "loss": 1.1519,
      "num_input_tokens_seen": 32574160,
      "step": 56155
    },
    {
      "epoch": 8.36461126005362,
      "grad_norm": 8.0,
      "learning_rate": 3.6098436124439594e-05,
      "loss": 1.0432,
      "num_input_tokens_seen": 32577168,
      "step": 56160
    },
    {
      "epoch": 8.365355972594578,
      "grad_norm": 7.90625,
      "learning_rate": 3.609552436132427e-05,
      "loss": 0.8721,
      "num_input_tokens_seen": 32579888,
      "step": 56165
    },
    {
      "epoch": 8.366100685135537,
      "grad_norm": 7.78125,
      "learning_rate": 3.609261241076136e-05,
      "loss": 1.3238,
      "num_input_tokens_seen": 32582544,
      "step": 56170
    },
    {
      "epoch": 8.366845397676498,
      "grad_norm": 13.125,
      "learning_rate": 3.608970027280001e-05,
      "loss": 1.2333,
      "num_input_tokens_seen": 32585200,
      "step": 56175
    },
    {
      "epoch": 8.367590110217456,
      "grad_norm": 8.25,
      "learning_rate": 3.608678794748946e-05,
      "loss": 1.0383,
      "num_input_tokens_seen": 32587984,
      "step": 56180
    },
    {
      "epoch": 8.368334822758415,
      "grad_norm": 15.25,
      "learning_rate": 3.608387543487889e-05,
      "loss": 1.0097,
      "num_input_tokens_seen": 32590736,
      "step": 56185
    },
    {
      "epoch": 8.369079535299374,
      "grad_norm": 10.8125,
      "learning_rate": 3.6080962735017514e-05,
      "loss": 1.0379,
      "num_input_tokens_seen": 32593552,
      "step": 56190
    },
    {
      "epoch": 8.369824247840334,
      "grad_norm": 18.0,
      "learning_rate": 3.607804984795453e-05,
      "loss": 1.1634,
      "num_input_tokens_seen": 32596400,
      "step": 56195
    },
    {
      "epoch": 8.370568960381293,
      "grad_norm": 29.875,
      "learning_rate": 3.607513677373916e-05,
      "loss": 1.0898,
      "num_input_tokens_seen": 32599216,
      "step": 56200
    },
    {
      "epoch": 8.371313672922252,
      "grad_norm": 6.09375,
      "learning_rate": 3.60722235124206e-05,
      "loss": 1.1114,
      "num_input_tokens_seen": 32602096,
      "step": 56205
    },
    {
      "epoch": 8.37205838546321,
      "grad_norm": 8.3125,
      "learning_rate": 3.606931006404809e-05,
      "loss": 1.0612,
      "num_input_tokens_seen": 32604912,
      "step": 56210
    },
    {
      "epoch": 8.372803098004171,
      "grad_norm": 16.25,
      "learning_rate": 3.606639642867083e-05,
      "loss": 0.9345,
      "num_input_tokens_seen": 32607920,
      "step": 56215
    },
    {
      "epoch": 8.37354781054513,
      "grad_norm": 11.1875,
      "learning_rate": 3.606348260633805e-05,
      "loss": 0.9499,
      "num_input_tokens_seen": 32610832,
      "step": 56220
    },
    {
      "epoch": 8.374292523086089,
      "grad_norm": 8.75,
      "learning_rate": 3.6060568597098974e-05,
      "loss": 0.889,
      "num_input_tokens_seen": 32613552,
      "step": 56225
    },
    {
      "epoch": 8.375037235627047,
      "grad_norm": 16.75,
      "learning_rate": 3.605765440100283e-05,
      "loss": 1.1324,
      "num_input_tokens_seen": 32616400,
      "step": 56230
    },
    {
      "epoch": 8.375781948168008,
      "grad_norm": 6.9375,
      "learning_rate": 3.605474001809886e-05,
      "loss": 1.2509,
      "num_input_tokens_seen": 32619504,
      "step": 56235
    },
    {
      "epoch": 8.376526660708967,
      "grad_norm": 14.0625,
      "learning_rate": 3.6051825448436286e-05,
      "loss": 1.1278,
      "num_input_tokens_seen": 32622416,
      "step": 56240
    },
    {
      "epoch": 8.377271373249926,
      "grad_norm": 9.375,
      "learning_rate": 3.604891069206437e-05,
      "loss": 0.7003,
      "num_input_tokens_seen": 32625392,
      "step": 56245
    },
    {
      "epoch": 8.378016085790884,
      "grad_norm": 7.96875,
      "learning_rate": 3.6045995749032326e-05,
      "loss": 1.2261,
      "num_input_tokens_seen": 32628048,
      "step": 56250
    },
    {
      "epoch": 8.378760798331843,
      "grad_norm": 9.375,
      "learning_rate": 3.6043080619389406e-05,
      "loss": 1.1561,
      "num_input_tokens_seen": 32630800,
      "step": 56255
    },
    {
      "epoch": 8.379505510872804,
      "grad_norm": 9.5,
      "learning_rate": 3.604016530318487e-05,
      "loss": 1.1866,
      "num_input_tokens_seen": 32634064,
      "step": 56260
    },
    {
      "epoch": 8.380250223413762,
      "grad_norm": 6.96875,
      "learning_rate": 3.6037249800467957e-05,
      "loss": 1.2574,
      "num_input_tokens_seen": 32636816,
      "step": 56265
    },
    {
      "epoch": 8.380994935954721,
      "grad_norm": 17.375,
      "learning_rate": 3.6034334111287926e-05,
      "loss": 0.7908,
      "num_input_tokens_seen": 32639792,
      "step": 56270
    },
    {
      "epoch": 8.38173964849568,
      "grad_norm": 10.6875,
      "learning_rate": 3.603141823569404e-05,
      "loss": 1.0707,
      "num_input_tokens_seen": 32642608,
      "step": 56275
    },
    {
      "epoch": 8.38248436103664,
      "grad_norm": 9.625,
      "learning_rate": 3.602850217373555e-05,
      "loss": 1.2804,
      "num_input_tokens_seen": 32645520,
      "step": 56280
    },
    {
      "epoch": 8.3832290735776,
      "grad_norm": 6.15625,
      "learning_rate": 3.602558592546172e-05,
      "loss": 0.7951,
      "num_input_tokens_seen": 32648528,
      "step": 56285
    },
    {
      "epoch": 8.383973786118558,
      "grad_norm": 14.875,
      "learning_rate": 3.602266949092184e-05,
      "loss": 1.25,
      "num_input_tokens_seen": 32651728,
      "step": 56290
    },
    {
      "epoch": 8.384718498659517,
      "grad_norm": 7.5625,
      "learning_rate": 3.6019752870165145e-05,
      "loss": 1.0504,
      "num_input_tokens_seen": 32654448,
      "step": 56295
    },
    {
      "epoch": 8.385463211200477,
      "grad_norm": 8.1875,
      "learning_rate": 3.601683606324093e-05,
      "loss": 0.9299,
      "num_input_tokens_seen": 32657360,
      "step": 56300
    },
    {
      "epoch": 8.386207923741436,
      "grad_norm": 17.875,
      "learning_rate": 3.601391907019847e-05,
      "loss": 1.0626,
      "num_input_tokens_seen": 32660432,
      "step": 56305
    },
    {
      "epoch": 8.386952636282395,
      "grad_norm": 7.46875,
      "learning_rate": 3.601100189108704e-05,
      "loss": 0.8469,
      "num_input_tokens_seen": 32663536,
      "step": 56310
    },
    {
      "epoch": 8.387697348823353,
      "grad_norm": 9.75,
      "learning_rate": 3.600808452595592e-05,
      "loss": 0.8064,
      "num_input_tokens_seen": 32666576,
      "step": 56315
    },
    {
      "epoch": 8.388442061364314,
      "grad_norm": 8.0625,
      "learning_rate": 3.6005166974854406e-05,
      "loss": 1.1343,
      "num_input_tokens_seen": 32669392,
      "step": 56320
    },
    {
      "epoch": 8.389186773905273,
      "grad_norm": 7.8125,
      "learning_rate": 3.6002249237831774e-05,
      "loss": 1.2424,
      "num_input_tokens_seen": 32672080,
      "step": 56325
    },
    {
      "epoch": 8.389931486446232,
      "grad_norm": 12.0,
      "learning_rate": 3.599933131493733e-05,
      "loss": 0.9421,
      "num_input_tokens_seen": 32674864,
      "step": 56330
    },
    {
      "epoch": 8.39067619898719,
      "grad_norm": 15.4375,
      "learning_rate": 3.599641320622036e-05,
      "loss": 1.2067,
      "num_input_tokens_seen": 32677968,
      "step": 56335
    },
    {
      "epoch": 8.39142091152815,
      "grad_norm": 17.125,
      "learning_rate": 3.599349491173016e-05,
      "loss": 1.0375,
      "num_input_tokens_seen": 32680848,
      "step": 56340
    },
    {
      "epoch": 8.39216562406911,
      "grad_norm": 10.0625,
      "learning_rate": 3.5990576431516044e-05,
      "loss": 1.411,
      "num_input_tokens_seen": 32683536,
      "step": 56345
    },
    {
      "epoch": 8.392910336610068,
      "grad_norm": 4.5,
      "learning_rate": 3.598765776562731e-05,
      "loss": 1.2198,
      "num_input_tokens_seen": 32686320,
      "step": 56350
    },
    {
      "epoch": 8.393655049151027,
      "grad_norm": 5.90625,
      "learning_rate": 3.598473891411326e-05,
      "loss": 1.1032,
      "num_input_tokens_seen": 32689136,
      "step": 56355
    },
    {
      "epoch": 8.394399761691988,
      "grad_norm": 11.75,
      "learning_rate": 3.598181987702321e-05,
      "loss": 0.897,
      "num_input_tokens_seen": 32691728,
      "step": 56360
    },
    {
      "epoch": 8.395144474232946,
      "grad_norm": 11.9375,
      "learning_rate": 3.5978900654406476e-05,
      "loss": 1.0478,
      "num_input_tokens_seen": 32694608,
      "step": 56365
    },
    {
      "epoch": 8.395889186773905,
      "grad_norm": 10.0,
      "learning_rate": 3.597598124631239e-05,
      "loss": 1.0234,
      "num_input_tokens_seen": 32697424,
      "step": 56370
    },
    {
      "epoch": 8.396633899314864,
      "grad_norm": 9.0625,
      "learning_rate": 3.5973061652790237e-05,
      "loss": 0.8547,
      "num_input_tokens_seen": 32700208,
      "step": 56375
    },
    {
      "epoch": 8.397378611855824,
      "grad_norm": 8.5625,
      "learning_rate": 3.597014187388936e-05,
      "loss": 1.4302,
      "num_input_tokens_seen": 32703216,
      "step": 56380
    },
    {
      "epoch": 8.398123324396783,
      "grad_norm": 7.03125,
      "learning_rate": 3.5967221909659095e-05,
      "loss": 0.9543,
      "num_input_tokens_seen": 32705968,
      "step": 56385
    },
    {
      "epoch": 8.398868036937742,
      "grad_norm": 16.5,
      "learning_rate": 3.596430176014875e-05,
      "loss": 1.0696,
      "num_input_tokens_seen": 32709040,
      "step": 56390
    },
    {
      "epoch": 8.3996127494787,
      "grad_norm": 6.4375,
      "learning_rate": 3.596138142540768e-05,
      "loss": 1.0446,
      "num_input_tokens_seen": 32711984,
      "step": 56395
    },
    {
      "epoch": 8.400357462019661,
      "grad_norm": 18.125,
      "learning_rate": 3.5958460905485216e-05,
      "loss": 1.3909,
      "num_input_tokens_seen": 32714704,
      "step": 56400
    },
    {
      "epoch": 8.40110217456062,
      "grad_norm": 8.1875,
      "learning_rate": 3.595554020043068e-05,
      "loss": 1.0959,
      "num_input_tokens_seen": 32717328,
      "step": 56405
    },
    {
      "epoch": 8.401846887101579,
      "grad_norm": 7.46875,
      "learning_rate": 3.5952619310293435e-05,
      "loss": 1.0579,
      "num_input_tokens_seen": 32720528,
      "step": 56410
    },
    {
      "epoch": 8.402591599642538,
      "grad_norm": 11.125,
      "learning_rate": 3.594969823512282e-05,
      "loss": 1.1252,
      "num_input_tokens_seen": 32723248,
      "step": 56415
    },
    {
      "epoch": 8.403336312183498,
      "grad_norm": 8.125,
      "learning_rate": 3.5946776974968174e-05,
      "loss": 0.798,
      "num_input_tokens_seen": 32726416,
      "step": 56420
    },
    {
      "epoch": 8.404081024724457,
      "grad_norm": 6.875,
      "learning_rate": 3.5943855529878865e-05,
      "loss": 0.8005,
      "num_input_tokens_seen": 32729520,
      "step": 56425
    },
    {
      "epoch": 8.404825737265416,
      "grad_norm": 8.625,
      "learning_rate": 3.594093389990424e-05,
      "loss": 1.0368,
      "num_input_tokens_seen": 32732496,
      "step": 56430
    },
    {
      "epoch": 8.405570449806374,
      "grad_norm": 9.375,
      "learning_rate": 3.593801208509365e-05,
      "loss": 0.8502,
      "num_input_tokens_seen": 32735184,
      "step": 56435
    },
    {
      "epoch": 8.406315162347333,
      "grad_norm": 10.3125,
      "learning_rate": 3.593509008549646e-05,
      "loss": 0.7856,
      "num_input_tokens_seen": 32737776,
      "step": 56440
    },
    {
      "epoch": 8.407059874888294,
      "grad_norm": 14.4375,
      "learning_rate": 3.593216790116205e-05,
      "loss": 1.115,
      "num_input_tokens_seen": 32740528,
      "step": 56445
    },
    {
      "epoch": 8.407804587429252,
      "grad_norm": 19.625,
      "learning_rate": 3.5929245532139773e-05,
      "loss": 1.1458,
      "num_input_tokens_seen": 32743504,
      "step": 56450
    },
    {
      "epoch": 8.408549299970211,
      "grad_norm": 12.25,
      "learning_rate": 3.5926322978478985e-05,
      "loss": 1.6876,
      "num_input_tokens_seen": 32746480,
      "step": 56455
    },
    {
      "epoch": 8.40929401251117,
      "grad_norm": 6.28125,
      "learning_rate": 3.592340024022909e-05,
      "loss": 1.2823,
      "num_input_tokens_seen": 32749296,
      "step": 56460
    },
    {
      "epoch": 8.41003872505213,
      "grad_norm": 14.625,
      "learning_rate": 3.592047731743944e-05,
      "loss": 1.2674,
      "num_input_tokens_seen": 32752368,
      "step": 56465
    },
    {
      "epoch": 8.41078343759309,
      "grad_norm": 9.5625,
      "learning_rate": 3.591755421015943e-05,
      "loss": 0.8536,
      "num_input_tokens_seen": 32755216,
      "step": 56470
    },
    {
      "epoch": 8.411528150134048,
      "grad_norm": 22.75,
      "learning_rate": 3.591463091843844e-05,
      "loss": 0.9682,
      "num_input_tokens_seen": 32758064,
      "step": 56475
    },
    {
      "epoch": 8.412272862675007,
      "grad_norm": 7.9375,
      "learning_rate": 3.591170744232585e-05,
      "loss": 0.9353,
      "num_input_tokens_seen": 32761168,
      "step": 56480
    },
    {
      "epoch": 8.413017575215967,
      "grad_norm": 8.0625,
      "learning_rate": 3.590878378187106e-05,
      "loss": 1.2426,
      "num_input_tokens_seen": 32764112,
      "step": 56485
    },
    {
      "epoch": 8.413762287756926,
      "grad_norm": 3.359375,
      "learning_rate": 3.5905859937123445e-05,
      "loss": 1.1197,
      "num_input_tokens_seen": 32767248,
      "step": 56490
    },
    {
      "epoch": 8.414507000297885,
      "grad_norm": 10.625,
      "learning_rate": 3.5902935908132416e-05,
      "loss": 1.1628,
      "num_input_tokens_seen": 32770320,
      "step": 56495
    },
    {
      "epoch": 8.415251712838844,
      "grad_norm": 11.375,
      "learning_rate": 3.5900011694947364e-05,
      "loss": 1.0532,
      "num_input_tokens_seen": 32773264,
      "step": 56500
    },
    {
      "epoch": 8.415996425379804,
      "grad_norm": 11.1875,
      "learning_rate": 3.5897087297617694e-05,
      "loss": 0.9415,
      "num_input_tokens_seen": 32776464,
      "step": 56505
    },
    {
      "epoch": 8.416741137920763,
      "grad_norm": 16.625,
      "learning_rate": 3.589416271619281e-05,
      "loss": 1.1974,
      "num_input_tokens_seen": 32779440,
      "step": 56510
    },
    {
      "epoch": 8.417485850461722,
      "grad_norm": 16.0,
      "learning_rate": 3.589123795072212e-05,
      "loss": 1.4036,
      "num_input_tokens_seen": 32782576,
      "step": 56515
    },
    {
      "epoch": 8.41823056300268,
      "grad_norm": 13.8125,
      "learning_rate": 3.5888313001255034e-05,
      "loss": 1.2315,
      "num_input_tokens_seen": 32785392,
      "step": 56520
    },
    {
      "epoch": 8.418975275543641,
      "grad_norm": 14.9375,
      "learning_rate": 3.588538786784096e-05,
      "loss": 0.8143,
      "num_input_tokens_seen": 32788496,
      "step": 56525
    },
    {
      "epoch": 8.4197199880846,
      "grad_norm": 13.8125,
      "learning_rate": 3.5882462550529325e-05,
      "loss": 0.9008,
      "num_input_tokens_seen": 32791280,
      "step": 56530
    },
    {
      "epoch": 8.420464700625558,
      "grad_norm": 15.5625,
      "learning_rate": 3.587953704936955e-05,
      "loss": 1.0063,
      "num_input_tokens_seen": 32793904,
      "step": 56535
    },
    {
      "epoch": 8.421209413166517,
      "grad_norm": 12.125,
      "learning_rate": 3.587661136441105e-05,
      "loss": 1.1358,
      "num_input_tokens_seen": 32796848,
      "step": 56540
    },
    {
      "epoch": 8.421954125707478,
      "grad_norm": 7.8125,
      "learning_rate": 3.587368549570326e-05,
      "loss": 1.0401,
      "num_input_tokens_seen": 32799856,
      "step": 56545
    },
    {
      "epoch": 8.422698838248436,
      "grad_norm": 9.5,
      "learning_rate": 3.58707594432956e-05,
      "loss": 1.1053,
      "num_input_tokens_seen": 32802864,
      "step": 56550
    },
    {
      "epoch": 8.423443550789395,
      "grad_norm": 13.9375,
      "learning_rate": 3.586783320723751e-05,
      "loss": 1.3669,
      "num_input_tokens_seen": 32805840,
      "step": 56555
    },
    {
      "epoch": 8.424188263330354,
      "grad_norm": 17.125,
      "learning_rate": 3.586490678757842e-05,
      "loss": 1.2293,
      "num_input_tokens_seen": 32808976,
      "step": 56560
    },
    {
      "epoch": 8.424932975871315,
      "grad_norm": 10.125,
      "learning_rate": 3.5861980184367775e-05,
      "loss": 1.3241,
      "num_input_tokens_seen": 32812176,
      "step": 56565
    },
    {
      "epoch": 8.425677688412273,
      "grad_norm": 11.9375,
      "learning_rate": 3.5859053397655014e-05,
      "loss": 1.4672,
      "num_input_tokens_seen": 32815344,
      "step": 56570
    },
    {
      "epoch": 8.426422400953232,
      "grad_norm": 7.8125,
      "learning_rate": 3.585612642748958e-05,
      "loss": 1.4525,
      "num_input_tokens_seen": 32818384,
      "step": 56575
    },
    {
      "epoch": 8.42716711349419,
      "grad_norm": 15.0,
      "learning_rate": 3.585319927392093e-05,
      "loss": 1.1998,
      "num_input_tokens_seen": 32821488,
      "step": 56580
    },
    {
      "epoch": 8.42791182603515,
      "grad_norm": 16.875,
      "learning_rate": 3.585027193699851e-05,
      "loss": 1.2114,
      "num_input_tokens_seen": 32824208,
      "step": 56585
    },
    {
      "epoch": 8.42865653857611,
      "grad_norm": 9.625,
      "learning_rate": 3.5847344416771766e-05,
      "loss": 1.1617,
      "num_input_tokens_seen": 32826832,
      "step": 56590
    },
    {
      "epoch": 8.429401251117069,
      "grad_norm": 9.9375,
      "learning_rate": 3.584441671329016e-05,
      "loss": 1.0293,
      "num_input_tokens_seen": 32829744,
      "step": 56595
    },
    {
      "epoch": 8.430145963658028,
      "grad_norm": 10.625,
      "learning_rate": 3.584148882660316e-05,
      "loss": 1.3946,
      "num_input_tokens_seen": 32832528,
      "step": 56600
    },
    {
      "epoch": 8.430890676198986,
      "grad_norm": 11.9375,
      "learning_rate": 3.583856075676023e-05,
      "loss": 1.5044,
      "num_input_tokens_seen": 32835184,
      "step": 56605
    },
    {
      "epoch": 8.431635388739947,
      "grad_norm": 17.25,
      "learning_rate": 3.5835632503810834e-05,
      "loss": 1.0451,
      "num_input_tokens_seen": 32838000,
      "step": 56610
    },
    {
      "epoch": 8.432380101280906,
      "grad_norm": 11.5,
      "learning_rate": 3.5832704067804436e-05,
      "loss": 1.0155,
      "num_input_tokens_seen": 32840912,
      "step": 56615
    },
    {
      "epoch": 8.433124813821864,
      "grad_norm": 5.5625,
      "learning_rate": 3.582977544879051e-05,
      "loss": 0.8382,
      "num_input_tokens_seen": 32843760,
      "step": 56620
    },
    {
      "epoch": 8.433869526362823,
      "grad_norm": 4.875,
      "learning_rate": 3.5826846646818536e-05,
      "loss": 1.1499,
      "num_input_tokens_seen": 32846800,
      "step": 56625
    },
    {
      "epoch": 8.434614238903784,
      "grad_norm": 8.6875,
      "learning_rate": 3.582391766193799e-05,
      "loss": 1.1866,
      "num_input_tokens_seen": 32849616,
      "step": 56630
    },
    {
      "epoch": 8.435358951444742,
      "grad_norm": 4.8125,
      "learning_rate": 3.582098849419835e-05,
      "loss": 1.399,
      "num_input_tokens_seen": 32852752,
      "step": 56635
    },
    {
      "epoch": 8.436103663985701,
      "grad_norm": 12.1875,
      "learning_rate": 3.581805914364912e-05,
      "loss": 1.0591,
      "num_input_tokens_seen": 32855600,
      "step": 56640
    },
    {
      "epoch": 8.43684837652666,
      "grad_norm": 7.8125,
      "learning_rate": 3.581512961033977e-05,
      "loss": 1.1137,
      "num_input_tokens_seen": 32858512,
      "step": 56645
    },
    {
      "epoch": 8.43759308906762,
      "grad_norm": 12.875,
      "learning_rate": 3.5812199894319795e-05,
      "loss": 1.2929,
      "num_input_tokens_seen": 32861680,
      "step": 56650
    },
    {
      "epoch": 8.43833780160858,
      "grad_norm": 7.5625,
      "learning_rate": 3.58092699956387e-05,
      "loss": 1.1593,
      "num_input_tokens_seen": 32864656,
      "step": 56655
    },
    {
      "epoch": 8.439082514149538,
      "grad_norm": 13.8125,
      "learning_rate": 3.580633991434597e-05,
      "loss": 1.2272,
      "num_input_tokens_seen": 32867952,
      "step": 56660
    },
    {
      "epoch": 8.439827226690497,
      "grad_norm": 11.125,
      "learning_rate": 3.58034096504911e-05,
      "loss": 1.011,
      "num_input_tokens_seen": 32870992,
      "step": 56665
    },
    {
      "epoch": 8.440571939231457,
      "grad_norm": 10.0,
      "learning_rate": 3.580047920412362e-05,
      "loss": 1.0542,
      "num_input_tokens_seen": 32873744,
      "step": 56670
    },
    {
      "epoch": 8.441316651772416,
      "grad_norm": 14.4375,
      "learning_rate": 3.579754857529301e-05,
      "loss": 1.2101,
      "num_input_tokens_seen": 32876688,
      "step": 56675
    },
    {
      "epoch": 8.442061364313375,
      "grad_norm": 10.6875,
      "learning_rate": 3.57946177640488e-05,
      "loss": 0.7938,
      "num_input_tokens_seen": 32879728,
      "step": 56680
    },
    {
      "epoch": 8.442806076854334,
      "grad_norm": 8.6875,
      "learning_rate": 3.579168677044049e-05,
      "loss": 1.2429,
      "num_input_tokens_seen": 32882448,
      "step": 56685
    },
    {
      "epoch": 8.443550789395294,
      "grad_norm": 6.125,
      "learning_rate": 3.5788755594517595e-05,
      "loss": 1.4068,
      "num_input_tokens_seen": 32885680,
      "step": 56690
    },
    {
      "epoch": 8.444295501936253,
      "grad_norm": 21.375,
      "learning_rate": 3.5785824236329644e-05,
      "loss": 1.1313,
      "num_input_tokens_seen": 32888592,
      "step": 56695
    },
    {
      "epoch": 8.445040214477212,
      "grad_norm": 12.5,
      "learning_rate": 3.578289269592615e-05,
      "loss": 0.905,
      "num_input_tokens_seen": 32891536,
      "step": 56700
    },
    {
      "epoch": 8.44578492701817,
      "grad_norm": 10.6875,
      "learning_rate": 3.577996097335665e-05,
      "loss": 1.1145,
      "num_input_tokens_seen": 32894544,
      "step": 56705
    },
    {
      "epoch": 8.446529639559131,
      "grad_norm": 26.25,
      "learning_rate": 3.577702906867066e-05,
      "loss": 1.0239,
      "num_input_tokens_seen": 32897520,
      "step": 56710
    },
    {
      "epoch": 8.44727435210009,
      "grad_norm": 11.5625,
      "learning_rate": 3.577409698191773e-05,
      "loss": 1.2304,
      "num_input_tokens_seen": 32900400,
      "step": 56715
    },
    {
      "epoch": 8.448019064641048,
      "grad_norm": 14.875,
      "learning_rate": 3.5771164713147364e-05,
      "loss": 0.9924,
      "num_input_tokens_seen": 32903024,
      "step": 56720
    },
    {
      "epoch": 8.448763777182007,
      "grad_norm": 19.5,
      "learning_rate": 3.576823226240913e-05,
      "loss": 1.3603,
      "num_input_tokens_seen": 32906288,
      "step": 56725
    },
    {
      "epoch": 8.449508489722968,
      "grad_norm": 7.90625,
      "learning_rate": 3.576529962975255e-05,
      "loss": 0.8652,
      "num_input_tokens_seen": 32909072,
      "step": 56730
    },
    {
      "epoch": 8.450253202263927,
      "grad_norm": 13.5625,
      "learning_rate": 3.576236681522718e-05,
      "loss": 0.9912,
      "num_input_tokens_seen": 32911792,
      "step": 56735
    },
    {
      "epoch": 8.450997914804885,
      "grad_norm": 11.5,
      "learning_rate": 3.575943381888255e-05,
      "loss": 1.0178,
      "num_input_tokens_seen": 32914672,
      "step": 56740
    },
    {
      "epoch": 8.451742627345844,
      "grad_norm": 4.3125,
      "learning_rate": 3.575650064076823e-05,
      "loss": 0.9873,
      "num_input_tokens_seen": 32917424,
      "step": 56745
    },
    {
      "epoch": 8.452487339886805,
      "grad_norm": 7.28125,
      "learning_rate": 3.575356728093376e-05,
      "loss": 0.7562,
      "num_input_tokens_seen": 32920240,
      "step": 56750
    },
    {
      "epoch": 8.453232052427763,
      "grad_norm": 8.25,
      "learning_rate": 3.575063373942871e-05,
      "loss": 1.0383,
      "num_input_tokens_seen": 32923056,
      "step": 56755
    },
    {
      "epoch": 8.453976764968722,
      "grad_norm": 10.0625,
      "learning_rate": 3.5747700016302616e-05,
      "loss": 1.0845,
      "num_input_tokens_seen": 32926000,
      "step": 56760
    },
    {
      "epoch": 8.45472147750968,
      "grad_norm": 13.5625,
      "learning_rate": 3.574476611160506e-05,
      "loss": 1.1262,
      "num_input_tokens_seen": 32928592,
      "step": 56765
    },
    {
      "epoch": 8.45546619005064,
      "grad_norm": 9.6875,
      "learning_rate": 3.5741832025385596e-05,
      "loss": 1.0917,
      "num_input_tokens_seen": 32931728,
      "step": 56770
    },
    {
      "epoch": 8.4562109025916,
      "grad_norm": 8.75,
      "learning_rate": 3.57388977576938e-05,
      "loss": 1.1479,
      "num_input_tokens_seen": 32934640,
      "step": 56775
    },
    {
      "epoch": 8.456955615132559,
      "grad_norm": 8.8125,
      "learning_rate": 3.5735963308579256e-05,
      "loss": 0.9212,
      "num_input_tokens_seen": 32937520,
      "step": 56780
    },
    {
      "epoch": 8.457700327673518,
      "grad_norm": 7.5625,
      "learning_rate": 3.573302867809151e-05,
      "loss": 1.1393,
      "num_input_tokens_seen": 32940624,
      "step": 56785
    },
    {
      "epoch": 8.458445040214476,
      "grad_norm": 14.1875,
      "learning_rate": 3.573009386628015e-05,
      "loss": 1.2347,
      "num_input_tokens_seen": 32943600,
      "step": 56790
    },
    {
      "epoch": 8.459189752755437,
      "grad_norm": 10.3125,
      "learning_rate": 3.5727158873194763e-05,
      "loss": 1.1881,
      "num_input_tokens_seen": 32946352,
      "step": 56795
    },
    {
      "epoch": 8.459934465296396,
      "grad_norm": 11.1875,
      "learning_rate": 3.572422369888493e-05,
      "loss": 0.9467,
      "num_input_tokens_seen": 32949392,
      "step": 56800
    },
    {
      "epoch": 8.460679177837354,
      "grad_norm": 11.9375,
      "learning_rate": 3.5721288343400235e-05,
      "loss": 1.045,
      "num_input_tokens_seen": 32952240,
      "step": 56805
    },
    {
      "epoch": 8.461423890378313,
      "grad_norm": 9.9375,
      "learning_rate": 3.571835280679027e-05,
      "loss": 0.9697,
      "num_input_tokens_seen": 32955216,
      "step": 56810
    },
    {
      "epoch": 8.462168602919274,
      "grad_norm": 10.25,
      "learning_rate": 3.5715417089104634e-05,
      "loss": 1.1802,
      "num_input_tokens_seen": 32958128,
      "step": 56815
    },
    {
      "epoch": 8.462913315460233,
      "grad_norm": 10.5625,
      "learning_rate": 3.571248119039291e-05,
      "loss": 1.0086,
      "num_input_tokens_seen": 32961136,
      "step": 56820
    },
    {
      "epoch": 8.463658028001191,
      "grad_norm": 6.4375,
      "learning_rate": 3.570954511070471e-05,
      "loss": 0.8611,
      "num_input_tokens_seen": 32963920,
      "step": 56825
    },
    {
      "epoch": 8.46440274054215,
      "grad_norm": 15.3125,
      "learning_rate": 3.570660885008962e-05,
      "loss": 1.3649,
      "num_input_tokens_seen": 32966864,
      "step": 56830
    },
    {
      "epoch": 8.46514745308311,
      "grad_norm": 8.0625,
      "learning_rate": 3.570367240859727e-05,
      "loss": 1.1785,
      "num_input_tokens_seen": 32969520,
      "step": 56835
    },
    {
      "epoch": 8.46589216562407,
      "grad_norm": 5.03125,
      "learning_rate": 3.570073578627724e-05,
      "loss": 0.9059,
      "num_input_tokens_seen": 32972432,
      "step": 56840
    },
    {
      "epoch": 8.466636878165028,
      "grad_norm": 20.375,
      "learning_rate": 3.5697798983179165e-05,
      "loss": 1.1861,
      "num_input_tokens_seen": 32975312,
      "step": 56845
    },
    {
      "epoch": 8.467381590705987,
      "grad_norm": 10.4375,
      "learning_rate": 3.569486199935264e-05,
      "loss": 0.8382,
      "num_input_tokens_seen": 32978224,
      "step": 56850
    },
    {
      "epoch": 8.468126303246947,
      "grad_norm": 10.4375,
      "learning_rate": 3.56919248348473e-05,
      "loss": 0.9649,
      "num_input_tokens_seen": 32981008,
      "step": 56855
    },
    {
      "epoch": 8.468871015787906,
      "grad_norm": 8.6875,
      "learning_rate": 3.568898748971275e-05,
      "loss": 1.1616,
      "num_input_tokens_seen": 32984304,
      "step": 56860
    },
    {
      "epoch": 8.469615728328865,
      "grad_norm": 9.9375,
      "learning_rate": 3.568604996399862e-05,
      "loss": 0.993,
      "num_input_tokens_seen": 32987248,
      "step": 56865
    },
    {
      "epoch": 8.470360440869824,
      "grad_norm": 6.03125,
      "learning_rate": 3.5683112257754535e-05,
      "loss": 0.8264,
      "num_input_tokens_seen": 32990256,
      "step": 56870
    },
    {
      "epoch": 8.471105153410784,
      "grad_norm": 6.03125,
      "learning_rate": 3.568017437103013e-05,
      "loss": 1.5418,
      "num_input_tokens_seen": 32993104,
      "step": 56875
    },
    {
      "epoch": 8.471849865951743,
      "grad_norm": 7.28125,
      "learning_rate": 3.567723630387504e-05,
      "loss": 1.1259,
      "num_input_tokens_seen": 32995696,
      "step": 56880
    },
    {
      "epoch": 8.472594578492702,
      "grad_norm": 7.125,
      "learning_rate": 3.5674298056338885e-05,
      "loss": 0.798,
      "num_input_tokens_seen": 32998416,
      "step": 56885
    },
    {
      "epoch": 8.47333929103366,
      "grad_norm": 11.0,
      "learning_rate": 3.5671359628471315e-05,
      "loss": 1.8876,
      "num_input_tokens_seen": 33001872,
      "step": 56890
    },
    {
      "epoch": 8.474084003574621,
      "grad_norm": 6.34375,
      "learning_rate": 3.566842102032198e-05,
      "loss": 0.9859,
      "num_input_tokens_seen": 33004528,
      "step": 56895
    },
    {
      "epoch": 8.47482871611558,
      "grad_norm": 6.5,
      "learning_rate": 3.56654822319405e-05,
      "loss": 1.2417,
      "num_input_tokens_seen": 33007632,
      "step": 56900
    },
    {
      "epoch": 8.475573428656539,
      "grad_norm": 12.3125,
      "learning_rate": 3.5662543263376544e-05,
      "loss": 1.2736,
      "num_input_tokens_seen": 33010640,
      "step": 56905
    },
    {
      "epoch": 8.476318141197497,
      "grad_norm": 4.6875,
      "learning_rate": 3.5659604114679754e-05,
      "loss": 0.8272,
      "num_input_tokens_seen": 33013936,
      "step": 56910
    },
    {
      "epoch": 8.477062853738458,
      "grad_norm": 8.9375,
      "learning_rate": 3.565666478589979e-05,
      "loss": 0.9046,
      "num_input_tokens_seen": 33017104,
      "step": 56915
    },
    {
      "epoch": 8.477807566279417,
      "grad_norm": 17.375,
      "learning_rate": 3.5653725277086306e-05,
      "loss": 1.3806,
      "num_input_tokens_seen": 33019888,
      "step": 56920
    },
    {
      "epoch": 8.478552278820375,
      "grad_norm": 11.375,
      "learning_rate": 3.565078558828896e-05,
      "loss": 0.9845,
      "num_input_tokens_seen": 33022640,
      "step": 56925
    },
    {
      "epoch": 8.479296991361334,
      "grad_norm": 8.9375,
      "learning_rate": 3.564784571955741e-05,
      "loss": 1.2723,
      "num_input_tokens_seen": 33025648,
      "step": 56930
    },
    {
      "epoch": 8.480041703902295,
      "grad_norm": 7.6875,
      "learning_rate": 3.5644905670941345e-05,
      "loss": 0.8811,
      "num_input_tokens_seen": 33028336,
      "step": 56935
    },
    {
      "epoch": 8.480786416443253,
      "grad_norm": 9.4375,
      "learning_rate": 3.56419654424904e-05,
      "loss": 1.1052,
      "num_input_tokens_seen": 33031248,
      "step": 56940
    },
    {
      "epoch": 8.481531128984212,
      "grad_norm": 26.25,
      "learning_rate": 3.5639025034254274e-05,
      "loss": 1.2152,
      "num_input_tokens_seen": 33034416,
      "step": 56945
    },
    {
      "epoch": 8.482275841525171,
      "grad_norm": 5.5625,
      "learning_rate": 3.563608444628264e-05,
      "loss": 0.9382,
      "num_input_tokens_seen": 33037328,
      "step": 56950
    },
    {
      "epoch": 8.48302055406613,
      "grad_norm": 9.9375,
      "learning_rate": 3.563314367862515e-05,
      "loss": 0.8555,
      "num_input_tokens_seen": 33040016,
      "step": 56955
    },
    {
      "epoch": 8.48376526660709,
      "grad_norm": 15.125,
      "learning_rate": 3.5630202731331515e-05,
      "loss": 0.859,
      "num_input_tokens_seen": 33042704,
      "step": 56960
    },
    {
      "epoch": 8.484509979148049,
      "grad_norm": 23.0,
      "learning_rate": 3.562726160445141e-05,
      "loss": 1.1855,
      "num_input_tokens_seen": 33045520,
      "step": 56965
    },
    {
      "epoch": 8.485254691689008,
      "grad_norm": 12.5,
      "learning_rate": 3.562432029803452e-05,
      "loss": 1.4086,
      "num_input_tokens_seen": 33048432,
      "step": 56970
    },
    {
      "epoch": 8.485999404229966,
      "grad_norm": 11.1875,
      "learning_rate": 3.562137881213053e-05,
      "loss": 1.0189,
      "num_input_tokens_seen": 33051408,
      "step": 56975
    },
    {
      "epoch": 8.486744116770927,
      "grad_norm": 7.6875,
      "learning_rate": 3.5618437146789155e-05,
      "loss": 1.3571,
      "num_input_tokens_seen": 33054288,
      "step": 56980
    },
    {
      "epoch": 8.487488829311886,
      "grad_norm": 8.375,
      "learning_rate": 3.561549530206007e-05,
      "loss": 1.3377,
      "num_input_tokens_seen": 33057456,
      "step": 56985
    },
    {
      "epoch": 8.488233541852845,
      "grad_norm": 6.9375,
      "learning_rate": 3.561255327799298e-05,
      "loss": 0.9792,
      "num_input_tokens_seen": 33060400,
      "step": 56990
    },
    {
      "epoch": 8.488978254393803,
      "grad_norm": 12.125,
      "learning_rate": 3.5609611074637584e-05,
      "loss": 1.4011,
      "num_input_tokens_seen": 33063504,
      "step": 56995
    },
    {
      "epoch": 8.489722966934764,
      "grad_norm": 17.375,
      "learning_rate": 3.5606668692043595e-05,
      "loss": 1.1512,
      "num_input_tokens_seen": 33066192,
      "step": 57000
    },
    {
      "epoch": 8.490467679475723,
      "grad_norm": 11.3125,
      "learning_rate": 3.5603726130260715e-05,
      "loss": 0.8693,
      "num_input_tokens_seen": 33069104,
      "step": 57005
    },
    {
      "epoch": 8.491212392016681,
      "grad_norm": 7.9375,
      "learning_rate": 3.5600783389338674e-05,
      "loss": 1.2733,
      "num_input_tokens_seen": 33071920,
      "step": 57010
    },
    {
      "epoch": 8.49195710455764,
      "grad_norm": 8.1875,
      "learning_rate": 3.559784046932716e-05,
      "loss": 1.0149,
      "num_input_tokens_seen": 33074736,
      "step": 57015
    },
    {
      "epoch": 8.4927018170986,
      "grad_norm": 17.375,
      "learning_rate": 3.5594897370275905e-05,
      "loss": 1.0297,
      "num_input_tokens_seen": 33077712,
      "step": 57020
    },
    {
      "epoch": 8.49344652963956,
      "grad_norm": 8.9375,
      "learning_rate": 3.5591954092234625e-05,
      "loss": 1.0736,
      "num_input_tokens_seen": 33080592,
      "step": 57025
    },
    {
      "epoch": 8.494191242180518,
      "grad_norm": 5.71875,
      "learning_rate": 3.558901063525305e-05,
      "loss": 1.0938,
      "num_input_tokens_seen": 33083856,
      "step": 57030
    },
    {
      "epoch": 8.494935954721477,
      "grad_norm": 11.875,
      "learning_rate": 3.55860669993809e-05,
      "loss": 0.8371,
      "num_input_tokens_seen": 33086608,
      "step": 57035
    },
    {
      "epoch": 8.495680667262437,
      "grad_norm": 17.875,
      "learning_rate": 3.55831231846679e-05,
      "loss": 1.1454,
      "num_input_tokens_seen": 33089392,
      "step": 57040
    },
    {
      "epoch": 8.496425379803396,
      "grad_norm": 5.28125,
      "learning_rate": 3.55801791911638e-05,
      "loss": 1.0571,
      "num_input_tokens_seen": 33092432,
      "step": 57045
    },
    {
      "epoch": 8.497170092344355,
      "grad_norm": 4.71875,
      "learning_rate": 3.557723501891832e-05,
      "loss": 0.7333,
      "num_input_tokens_seen": 33095536,
      "step": 57050
    },
    {
      "epoch": 8.497914804885314,
      "grad_norm": 9.0625,
      "learning_rate": 3.557429066798121e-05,
      "loss": 1.2788,
      "num_input_tokens_seen": 33098480,
      "step": 57055
    },
    {
      "epoch": 8.498659517426274,
      "grad_norm": 14.1875,
      "learning_rate": 3.5571346138402204e-05,
      "loss": 1.2215,
      "num_input_tokens_seen": 33101456,
      "step": 57060
    },
    {
      "epoch": 8.499404229967233,
      "grad_norm": 10.625,
      "learning_rate": 3.5568401430231045e-05,
      "loss": 0.9748,
      "num_input_tokens_seen": 33104272,
      "step": 57065
    },
    {
      "epoch": 8.500148942508192,
      "grad_norm": 9.4375,
      "learning_rate": 3.556545654351749e-05,
      "loss": 1.2019,
      "num_input_tokens_seen": 33107088,
      "step": 57070
    },
    {
      "epoch": 8.50089365504915,
      "grad_norm": 9.0625,
      "learning_rate": 3.556251147831128e-05,
      "loss": 1.5392,
      "num_input_tokens_seen": 33109808,
      "step": 57075
    },
    {
      "epoch": 8.501638367590111,
      "grad_norm": 10.25,
      "learning_rate": 3.5559566234662175e-05,
      "loss": 1.2783,
      "num_input_tokens_seen": 33112688,
      "step": 57080
    },
    {
      "epoch": 8.50238308013107,
      "grad_norm": 10.5,
      "learning_rate": 3.555662081261994e-05,
      "loss": 0.9726,
      "num_input_tokens_seen": 33115984,
      "step": 57085
    },
    {
      "epoch": 8.503127792672029,
      "grad_norm": 14.1875,
      "learning_rate": 3.555367521223431e-05,
      "loss": 1.0796,
      "num_input_tokens_seen": 33119024,
      "step": 57090
    },
    {
      "epoch": 8.503872505212987,
      "grad_norm": 7.53125,
      "learning_rate": 3.555072943355508e-05,
      "loss": 1.0166,
      "num_input_tokens_seen": 33122064,
      "step": 57095
    },
    {
      "epoch": 8.504617217753946,
      "grad_norm": 9.4375,
      "learning_rate": 3.554778347663199e-05,
      "loss": 0.8807,
      "num_input_tokens_seen": 33125296,
      "step": 57100
    },
    {
      "epoch": 8.505361930294907,
      "grad_norm": 11.0,
      "learning_rate": 3.554483734151482e-05,
      "loss": 0.86,
      "num_input_tokens_seen": 33128080,
      "step": 57105
    },
    {
      "epoch": 8.506106642835865,
      "grad_norm": 6.53125,
      "learning_rate": 3.554189102825334e-05,
      "loss": 0.8381,
      "num_input_tokens_seen": 33131056,
      "step": 57110
    },
    {
      "epoch": 8.506851355376824,
      "grad_norm": 7.65625,
      "learning_rate": 3.553894453689733e-05,
      "loss": 0.9582,
      "num_input_tokens_seen": 33133840,
      "step": 57115
    },
    {
      "epoch": 8.507596067917785,
      "grad_norm": 14.375,
      "learning_rate": 3.553599786749656e-05,
      "loss": 1.4311,
      "num_input_tokens_seen": 33136752,
      "step": 57120
    },
    {
      "epoch": 8.508340780458743,
      "grad_norm": 11.5625,
      "learning_rate": 3.553305102010081e-05,
      "loss": 1.0422,
      "num_input_tokens_seen": 33139664,
      "step": 57125
    },
    {
      "epoch": 8.509085492999702,
      "grad_norm": 11.375,
      "learning_rate": 3.553010399475987e-05,
      "loss": 0.9865,
      "num_input_tokens_seen": 33142416,
      "step": 57130
    },
    {
      "epoch": 8.509830205540661,
      "grad_norm": 14.5,
      "learning_rate": 3.552715679152353e-05,
      "loss": 0.9848,
      "num_input_tokens_seen": 33145584,
      "step": 57135
    },
    {
      "epoch": 8.51057491808162,
      "grad_norm": 15.4375,
      "learning_rate": 3.552420941044157e-05,
      "loss": 1.1546,
      "num_input_tokens_seen": 33148464,
      "step": 57140
    },
    {
      "epoch": 8.51131963062258,
      "grad_norm": 14.0625,
      "learning_rate": 3.5521261851563796e-05,
      "loss": 1.2575,
      "num_input_tokens_seen": 33151440,
      "step": 57145
    },
    {
      "epoch": 8.512064343163539,
      "grad_norm": 8.25,
      "learning_rate": 3.551831411493999e-05,
      "loss": 1.2522,
      "num_input_tokens_seen": 33154352,
      "step": 57150
    },
    {
      "epoch": 8.512809055704498,
      "grad_norm": 5.15625,
      "learning_rate": 3.551536620061996e-05,
      "loss": 1.129,
      "num_input_tokens_seen": 33157424,
      "step": 57155
    },
    {
      "epoch": 8.513553768245457,
      "grad_norm": 12.4375,
      "learning_rate": 3.55124181086535e-05,
      "loss": 1.2403,
      "num_input_tokens_seen": 33160208,
      "step": 57160
    },
    {
      "epoch": 8.514298480786417,
      "grad_norm": 18.625,
      "learning_rate": 3.5509469839090426e-05,
      "loss": 1.0643,
      "num_input_tokens_seen": 33162672,
      "step": 57165
    },
    {
      "epoch": 8.515043193327376,
      "grad_norm": 9.5625,
      "learning_rate": 3.550652139198054e-05,
      "loss": 1.1212,
      "num_input_tokens_seen": 33165584,
      "step": 57170
    },
    {
      "epoch": 8.515787905868335,
      "grad_norm": 14.4375,
      "learning_rate": 3.550357276737365e-05,
      "loss": 1.2221,
      "num_input_tokens_seen": 33168400,
      "step": 57175
    },
    {
      "epoch": 8.516532618409293,
      "grad_norm": 5.625,
      "learning_rate": 3.550062396531959e-05,
      "loss": 0.8938,
      "num_input_tokens_seen": 33171312,
      "step": 57180
    },
    {
      "epoch": 8.517277330950254,
      "grad_norm": 18.0,
      "learning_rate": 3.549767498586814e-05,
      "loss": 1.4367,
      "num_input_tokens_seen": 33174224,
      "step": 57185
    },
    {
      "epoch": 8.518022043491213,
      "grad_norm": 9.875,
      "learning_rate": 3.549472582906914e-05,
      "loss": 1.0121,
      "num_input_tokens_seen": 33176880,
      "step": 57190
    },
    {
      "epoch": 8.518766756032171,
      "grad_norm": 17.875,
      "learning_rate": 3.549177649497242e-05,
      "loss": 1.4022,
      "num_input_tokens_seen": 33179824,
      "step": 57195
    },
    {
      "epoch": 8.51951146857313,
      "grad_norm": 8.6875,
      "learning_rate": 3.54888269836278e-05,
      "loss": 1.3007,
      "num_input_tokens_seen": 33182864,
      "step": 57200
    },
    {
      "epoch": 8.52025618111409,
      "grad_norm": 19.5,
      "learning_rate": 3.5485877295085105e-05,
      "loss": 1.2698,
      "num_input_tokens_seen": 33185488,
      "step": 57205
    },
    {
      "epoch": 8.52100089365505,
      "grad_norm": 8.1875,
      "learning_rate": 3.5482927429394184e-05,
      "loss": 1.2397,
      "num_input_tokens_seen": 33188656,
      "step": 57210
    },
    {
      "epoch": 8.521745606196008,
      "grad_norm": 10.9375,
      "learning_rate": 3.547997738660485e-05,
      "loss": 1.333,
      "num_input_tokens_seen": 33191664,
      "step": 57215
    },
    {
      "epoch": 8.522490318736967,
      "grad_norm": 15.75,
      "learning_rate": 3.547702716676694e-05,
      "loss": 0.9488,
      "num_input_tokens_seen": 33194320,
      "step": 57220
    },
    {
      "epoch": 8.523235031277927,
      "grad_norm": 15.4375,
      "learning_rate": 3.547407676993032e-05,
      "loss": 1.1993,
      "num_input_tokens_seen": 33197072,
      "step": 57225
    },
    {
      "epoch": 8.523979743818886,
      "grad_norm": 24.375,
      "learning_rate": 3.54711261961448e-05,
      "loss": 1.4016,
      "num_input_tokens_seen": 33199856,
      "step": 57230
    },
    {
      "epoch": 8.524724456359845,
      "grad_norm": 14.6875,
      "learning_rate": 3.5468175445460263e-05,
      "loss": 0.9333,
      "num_input_tokens_seen": 33202672,
      "step": 57235
    },
    {
      "epoch": 8.525469168900804,
      "grad_norm": 12.8125,
      "learning_rate": 3.546522451792653e-05,
      "loss": 0.9029,
      "num_input_tokens_seen": 33205520,
      "step": 57240
    },
    {
      "epoch": 8.526213881441764,
      "grad_norm": 13.375,
      "learning_rate": 3.546227341359347e-05,
      "loss": 0.9243,
      "num_input_tokens_seen": 33208048,
      "step": 57245
    },
    {
      "epoch": 8.526958593982723,
      "grad_norm": 15.0625,
      "learning_rate": 3.545932213251093e-05,
      "loss": 0.9514,
      "num_input_tokens_seen": 33211088,
      "step": 57250
    },
    {
      "epoch": 8.527703306523682,
      "grad_norm": 13.875,
      "learning_rate": 3.545637067472878e-05,
      "loss": 1.0429,
      "num_input_tokens_seen": 33213808,
      "step": 57255
    },
    {
      "epoch": 8.52844801906464,
      "grad_norm": 11.875,
      "learning_rate": 3.545341904029687e-05,
      "loss": 1.3027,
      "num_input_tokens_seen": 33216656,
      "step": 57260
    },
    {
      "epoch": 8.529192731605601,
      "grad_norm": 8.75,
      "learning_rate": 3.545046722926507e-05,
      "loss": 1.1909,
      "num_input_tokens_seen": 33219632,
      "step": 57265
    },
    {
      "epoch": 8.52993744414656,
      "grad_norm": 16.625,
      "learning_rate": 3.544751524168325e-05,
      "loss": 1.3757,
      "num_input_tokens_seen": 33222352,
      "step": 57270
    },
    {
      "epoch": 8.530682156687519,
      "grad_norm": 7.0,
      "learning_rate": 3.544456307760128e-05,
      "loss": 0.7749,
      "num_input_tokens_seen": 33225136,
      "step": 57275
    },
    {
      "epoch": 8.531426869228477,
      "grad_norm": 13.6875,
      "learning_rate": 3.5441610737069026e-05,
      "loss": 1.1173,
      "num_input_tokens_seen": 33227984,
      "step": 57280
    },
    {
      "epoch": 8.532171581769436,
      "grad_norm": 7.96875,
      "learning_rate": 3.543865822013637e-05,
      "loss": 1.0347,
      "num_input_tokens_seen": 33230800,
      "step": 57285
    },
    {
      "epoch": 8.532916294310397,
      "grad_norm": 14.75,
      "learning_rate": 3.5435705526853196e-05,
      "loss": 1.1286,
      "num_input_tokens_seen": 33233584,
      "step": 57290
    },
    {
      "epoch": 8.533661006851355,
      "grad_norm": 14.4375,
      "learning_rate": 3.5432752657269384e-05,
      "loss": 1.3212,
      "num_input_tokens_seen": 33236336,
      "step": 57295
    },
    {
      "epoch": 8.534405719392314,
      "grad_norm": 7.0625,
      "learning_rate": 3.542979961143482e-05,
      "loss": 1.1993,
      "num_input_tokens_seen": 33239664,
      "step": 57300
    },
    {
      "epoch": 8.535150431933273,
      "grad_norm": 13.5,
      "learning_rate": 3.542684638939939e-05,
      "loss": 1.3276,
      "num_input_tokens_seen": 33243248,
      "step": 57305
    },
    {
      "epoch": 8.535895144474233,
      "grad_norm": 13.875,
      "learning_rate": 3.5423892991212994e-05,
      "loss": 1.3257,
      "num_input_tokens_seen": 33246192,
      "step": 57310
    },
    {
      "epoch": 8.536639857015192,
      "grad_norm": 8.5625,
      "learning_rate": 3.542093941692551e-05,
      "loss": 1.1625,
      "num_input_tokens_seen": 33249264,
      "step": 57315
    },
    {
      "epoch": 8.537384569556151,
      "grad_norm": 12.0625,
      "learning_rate": 3.541798566658685e-05,
      "loss": 0.9525,
      "num_input_tokens_seen": 33252272,
      "step": 57320
    },
    {
      "epoch": 8.53812928209711,
      "grad_norm": 12.875,
      "learning_rate": 3.541503174024691e-05,
      "loss": 0.8801,
      "num_input_tokens_seen": 33255056,
      "step": 57325
    },
    {
      "epoch": 8.53887399463807,
      "grad_norm": 10.8125,
      "learning_rate": 3.54120776379556e-05,
      "loss": 1.0664,
      "num_input_tokens_seen": 33257840,
      "step": 57330
    },
    {
      "epoch": 8.539618707179029,
      "grad_norm": 15.625,
      "learning_rate": 3.540912335976281e-05,
      "loss": 1.3726,
      "num_input_tokens_seen": 33260624,
      "step": 57335
    },
    {
      "epoch": 8.540363419719988,
      "grad_norm": 14.4375,
      "learning_rate": 3.540616890571847e-05,
      "loss": 1.0129,
      "num_input_tokens_seen": 33263696,
      "step": 57340
    },
    {
      "epoch": 8.541108132260947,
      "grad_norm": 4.96875,
      "learning_rate": 3.540321427587249e-05,
      "loss": 1.1796,
      "num_input_tokens_seen": 33267024,
      "step": 57345
    },
    {
      "epoch": 8.541852844801907,
      "grad_norm": 10.1875,
      "learning_rate": 3.540025947027476e-05,
      "loss": 0.8566,
      "num_input_tokens_seen": 33269488,
      "step": 57350
    },
    {
      "epoch": 8.542597557342866,
      "grad_norm": 16.125,
      "learning_rate": 3.5397304488975226e-05,
      "loss": 1.3674,
      "num_input_tokens_seen": 33272528,
      "step": 57355
    },
    {
      "epoch": 8.543342269883825,
      "grad_norm": 14.625,
      "learning_rate": 3.53943493320238e-05,
      "loss": 0.9017,
      "num_input_tokens_seen": 33275888,
      "step": 57360
    },
    {
      "epoch": 8.544086982424783,
      "grad_norm": 10.9375,
      "learning_rate": 3.539139399947039e-05,
      "loss": 0.9435,
      "num_input_tokens_seen": 33278672,
      "step": 57365
    },
    {
      "epoch": 8.544831694965744,
      "grad_norm": 18.875,
      "learning_rate": 3.5388438491364963e-05,
      "loss": 1.2578,
      "num_input_tokens_seen": 33281488,
      "step": 57370
    },
    {
      "epoch": 8.545576407506703,
      "grad_norm": 11.8125,
      "learning_rate": 3.538548280775742e-05,
      "loss": 1.5106,
      "num_input_tokens_seen": 33284432,
      "step": 57375
    },
    {
      "epoch": 8.546321120047661,
      "grad_norm": 10.75,
      "learning_rate": 3.53825269486977e-05,
      "loss": 1.1214,
      "num_input_tokens_seen": 33287408,
      "step": 57380
    },
    {
      "epoch": 8.54706583258862,
      "grad_norm": 5.40625,
      "learning_rate": 3.5379570914235735e-05,
      "loss": 1.1619,
      "num_input_tokens_seen": 33290320,
      "step": 57385
    },
    {
      "epoch": 8.54781054512958,
      "grad_norm": 14.4375,
      "learning_rate": 3.537661470442147e-05,
      "loss": 1.3963,
      "num_input_tokens_seen": 33293232,
      "step": 57390
    },
    {
      "epoch": 8.54855525767054,
      "grad_norm": 8.375,
      "learning_rate": 3.537365831930484e-05,
      "loss": 0.897,
      "num_input_tokens_seen": 33296144,
      "step": 57395
    },
    {
      "epoch": 8.549299970211498,
      "grad_norm": 7.40625,
      "learning_rate": 3.53707017589358e-05,
      "loss": 1.0258,
      "num_input_tokens_seen": 33298704,
      "step": 57400
    },
    {
      "epoch": 8.550044682752457,
      "grad_norm": 11.875,
      "learning_rate": 3.53677450233643e-05,
      "loss": 0.9308,
      "num_input_tokens_seen": 33301648,
      "step": 57405
    },
    {
      "epoch": 8.550789395293418,
      "grad_norm": 15.1875,
      "learning_rate": 3.536478811264028e-05,
      "loss": 1.2317,
      "num_input_tokens_seen": 33305040,
      "step": 57410
    },
    {
      "epoch": 8.551534107834376,
      "grad_norm": 5.90625,
      "learning_rate": 3.5361831026813704e-05,
      "loss": 0.7131,
      "num_input_tokens_seen": 33308016,
      "step": 57415
    },
    {
      "epoch": 8.552278820375335,
      "grad_norm": 16.375,
      "learning_rate": 3.535887376593453e-05,
      "loss": 1.107,
      "num_input_tokens_seen": 33311088,
      "step": 57420
    },
    {
      "epoch": 8.553023532916294,
      "grad_norm": 6.78125,
      "learning_rate": 3.53559163300527e-05,
      "loss": 0.973,
      "num_input_tokens_seen": 33313904,
      "step": 57425
    },
    {
      "epoch": 8.553768245457253,
      "grad_norm": 15.0625,
      "learning_rate": 3.5352958719218186e-05,
      "loss": 1.3101,
      "num_input_tokens_seen": 33316624,
      "step": 57430
    },
    {
      "epoch": 8.554512957998213,
      "grad_norm": 9.0625,
      "learning_rate": 3.5350000933480966e-05,
      "loss": 1.0933,
      "num_input_tokens_seen": 33319280,
      "step": 57435
    },
    {
      "epoch": 8.555257670539172,
      "grad_norm": 12.9375,
      "learning_rate": 3.5347042972891e-05,
      "loss": 0.9911,
      "num_input_tokens_seen": 33322256,
      "step": 57440
    },
    {
      "epoch": 8.55600238308013,
      "grad_norm": 12.375,
      "learning_rate": 3.5344084837498245e-05,
      "loss": 0.87,
      "num_input_tokens_seen": 33324944,
      "step": 57445
    },
    {
      "epoch": 8.556747095621091,
      "grad_norm": 10.3125,
      "learning_rate": 3.53411265273527e-05,
      "loss": 0.9964,
      "num_input_tokens_seen": 33327504,
      "step": 57450
    },
    {
      "epoch": 8.55749180816205,
      "grad_norm": 7.21875,
      "learning_rate": 3.5338168042504336e-05,
      "loss": 0.8641,
      "num_input_tokens_seen": 33330352,
      "step": 57455
    },
    {
      "epoch": 8.558236520703009,
      "grad_norm": 8.8125,
      "learning_rate": 3.533520938300313e-05,
      "loss": 0.9225,
      "num_input_tokens_seen": 33333200,
      "step": 57460
    },
    {
      "epoch": 8.558981233243967,
      "grad_norm": 9.1875,
      "learning_rate": 3.533225054889906e-05,
      "loss": 0.8427,
      "num_input_tokens_seen": 33336368,
      "step": 57465
    },
    {
      "epoch": 8.559725945784926,
      "grad_norm": 13.5625,
      "learning_rate": 3.532929154024212e-05,
      "loss": 0.9723,
      "num_input_tokens_seen": 33339344,
      "step": 57470
    },
    {
      "epoch": 8.560470658325887,
      "grad_norm": 11.5625,
      "learning_rate": 3.5326332357082306e-05,
      "loss": 1.2138,
      "num_input_tokens_seen": 33342352,
      "step": 57475
    },
    {
      "epoch": 8.561215370866845,
      "grad_norm": 21.25,
      "learning_rate": 3.532337299946959e-05,
      "loss": 1.158,
      "num_input_tokens_seen": 33345136,
      "step": 57480
    },
    {
      "epoch": 8.561960083407804,
      "grad_norm": 11.75,
      "learning_rate": 3.532041346745398e-05,
      "loss": 1.1096,
      "num_input_tokens_seen": 33348016,
      "step": 57485
    },
    {
      "epoch": 8.562704795948763,
      "grad_norm": 8.9375,
      "learning_rate": 3.5317453761085476e-05,
      "loss": 1.138,
      "num_input_tokens_seen": 33351024,
      "step": 57490
    },
    {
      "epoch": 8.563449508489724,
      "grad_norm": 8.6875,
      "learning_rate": 3.531449388041408e-05,
      "loss": 0.9055,
      "num_input_tokens_seen": 33353776,
      "step": 57495
    },
    {
      "epoch": 8.564194221030682,
      "grad_norm": 11.125,
      "learning_rate": 3.5311533825489795e-05,
      "loss": 1.0138,
      "num_input_tokens_seen": 33356592,
      "step": 57500
    },
    {
      "epoch": 8.564938933571641,
      "grad_norm": 10.1875,
      "learning_rate": 3.530857359636262e-05,
      "loss": 1.1664,
      "num_input_tokens_seen": 33359856,
      "step": 57505
    },
    {
      "epoch": 8.5656836461126,
      "grad_norm": 10.1875,
      "learning_rate": 3.5305613193082575e-05,
      "loss": 1.1465,
      "num_input_tokens_seen": 33362800,
      "step": 57510
    },
    {
      "epoch": 8.56642835865356,
      "grad_norm": 10.0625,
      "learning_rate": 3.530265261569967e-05,
      "loss": 1.1482,
      "num_input_tokens_seen": 33365616,
      "step": 57515
    },
    {
      "epoch": 8.567173071194519,
      "grad_norm": 15.875,
      "learning_rate": 3.529969186426392e-05,
      "loss": 1.2854,
      "num_input_tokens_seen": 33368592,
      "step": 57520
    },
    {
      "epoch": 8.567917783735478,
      "grad_norm": 8.625,
      "learning_rate": 3.529673093882534e-05,
      "loss": 0.7524,
      "num_input_tokens_seen": 33371408,
      "step": 57525
    },
    {
      "epoch": 8.568662496276437,
      "grad_norm": 11.9375,
      "learning_rate": 3.5293769839433956e-05,
      "loss": 1.123,
      "num_input_tokens_seen": 33374000,
      "step": 57530
    },
    {
      "epoch": 8.569407208817397,
      "grad_norm": 11.9375,
      "learning_rate": 3.529080856613979e-05,
      "loss": 1.2881,
      "num_input_tokens_seen": 33377328,
      "step": 57535
    },
    {
      "epoch": 8.570151921358356,
      "grad_norm": 11.8125,
      "learning_rate": 3.528784711899288e-05,
      "loss": 1.1997,
      "num_input_tokens_seen": 33380400,
      "step": 57540
    },
    {
      "epoch": 8.570896633899315,
      "grad_norm": 4.75,
      "learning_rate": 3.5284885498043254e-05,
      "loss": 0.9115,
      "num_input_tokens_seen": 33383152,
      "step": 57545
    },
    {
      "epoch": 8.571641346440273,
      "grad_norm": 7.8125,
      "learning_rate": 3.528192370334094e-05,
      "loss": 0.8788,
      "num_input_tokens_seen": 33385968,
      "step": 57550
    },
    {
      "epoch": 8.572386058981234,
      "grad_norm": 16.5,
      "learning_rate": 3.527896173493596e-05,
      "loss": 1.0796,
      "num_input_tokens_seen": 33388720,
      "step": 57555
    },
    {
      "epoch": 8.573130771522193,
      "grad_norm": 6.625,
      "learning_rate": 3.527599959287838e-05,
      "loss": 1.0696,
      "num_input_tokens_seen": 33391632,
      "step": 57560
    },
    {
      "epoch": 8.573875484063151,
      "grad_norm": 13.625,
      "learning_rate": 3.5273037277218224e-05,
      "loss": 0.9571,
      "num_input_tokens_seen": 33394384,
      "step": 57565
    },
    {
      "epoch": 8.57462019660411,
      "grad_norm": 6.34375,
      "learning_rate": 3.527007478800555e-05,
      "loss": 0.9603,
      "num_input_tokens_seen": 33397104,
      "step": 57570
    },
    {
      "epoch": 8.57536490914507,
      "grad_norm": 11.125,
      "learning_rate": 3.5267112125290396e-05,
      "loss": 1.3201,
      "num_input_tokens_seen": 33400272,
      "step": 57575
    },
    {
      "epoch": 8.57610962168603,
      "grad_norm": 6.15625,
      "learning_rate": 3.5264149289122825e-05,
      "loss": 0.7162,
      "num_input_tokens_seen": 33403344,
      "step": 57580
    },
    {
      "epoch": 8.576854334226988,
      "grad_norm": 4.90625,
      "learning_rate": 3.526118627955288e-05,
      "loss": 1.2251,
      "num_input_tokens_seen": 33406160,
      "step": 57585
    },
    {
      "epoch": 8.577599046767947,
      "grad_norm": 13.8125,
      "learning_rate": 3.525822309663061e-05,
      "loss": 1.3992,
      "num_input_tokens_seen": 33409072,
      "step": 57590
    },
    {
      "epoch": 8.578343759308908,
      "grad_norm": 14.125,
      "learning_rate": 3.5255259740406104e-05,
      "loss": 0.9135,
      "num_input_tokens_seen": 33411888,
      "step": 57595
    },
    {
      "epoch": 8.579088471849866,
      "grad_norm": 8.3125,
      "learning_rate": 3.52522962109294e-05,
      "loss": 0.9433,
      "num_input_tokens_seen": 33414736,
      "step": 57600
    },
    {
      "epoch": 8.579833184390825,
      "grad_norm": 8.375,
      "learning_rate": 3.5249332508250576e-05,
      "loss": 0.7857,
      "num_input_tokens_seen": 33417552,
      "step": 57605
    },
    {
      "epoch": 8.580577896931784,
      "grad_norm": 16.375,
      "learning_rate": 3.524636863241969e-05,
      "loss": 1.1253,
      "num_input_tokens_seen": 33420208,
      "step": 57610
    },
    {
      "epoch": 8.581322609472743,
      "grad_norm": 4.96875,
      "learning_rate": 3.5243404583486824e-05,
      "loss": 0.7326,
      "num_input_tokens_seen": 33422896,
      "step": 57615
    },
    {
      "epoch": 8.582067322013703,
      "grad_norm": 9.375,
      "learning_rate": 3.5240440361502046e-05,
      "loss": 1.3435,
      "num_input_tokens_seen": 33425968,
      "step": 57620
    },
    {
      "epoch": 8.582812034554662,
      "grad_norm": 12.6875,
      "learning_rate": 3.523747596651544e-05,
      "loss": 1.4619,
      "num_input_tokens_seen": 33429040,
      "step": 57625
    },
    {
      "epoch": 8.58355674709562,
      "grad_norm": 4.84375,
      "learning_rate": 3.523451139857708e-05,
      "loss": 0.8295,
      "num_input_tokens_seen": 33431952,
      "step": 57630
    },
    {
      "epoch": 8.584301459636581,
      "grad_norm": 14.375,
      "learning_rate": 3.5231546657737044e-05,
      "loss": 0.992,
      "num_input_tokens_seen": 33434960,
      "step": 57635
    },
    {
      "epoch": 8.58504617217754,
      "grad_norm": 13.0,
      "learning_rate": 3.522858174404544e-05,
      "loss": 1.1989,
      "num_input_tokens_seen": 33438064,
      "step": 57640
    },
    {
      "epoch": 8.585790884718499,
      "grad_norm": 9.6875,
      "learning_rate": 3.522561665755234e-05,
      "loss": 1.2795,
      "num_input_tokens_seen": 33440848,
      "step": 57645
    },
    {
      "epoch": 8.586535597259457,
      "grad_norm": 12.3125,
      "learning_rate": 3.5222651398307835e-05,
      "loss": 1.3566,
      "num_input_tokens_seen": 33443696,
      "step": 57650
    },
    {
      "epoch": 8.587280309800416,
      "grad_norm": 16.375,
      "learning_rate": 3.5219685966362024e-05,
      "loss": 1.1227,
      "num_input_tokens_seen": 33446288,
      "step": 57655
    },
    {
      "epoch": 8.588025022341377,
      "grad_norm": 8.5,
      "learning_rate": 3.521672036176501e-05,
      "loss": 1.3682,
      "num_input_tokens_seen": 33449072,
      "step": 57660
    },
    {
      "epoch": 8.588769734882336,
      "grad_norm": 20.125,
      "learning_rate": 3.5213754584566886e-05,
      "loss": 1.0482,
      "num_input_tokens_seen": 33451760,
      "step": 57665
    },
    {
      "epoch": 8.589514447423294,
      "grad_norm": 14.5625,
      "learning_rate": 3.521078863481776e-05,
      "loss": 0.8406,
      "num_input_tokens_seen": 33454576,
      "step": 57670
    },
    {
      "epoch": 8.590259159964253,
      "grad_norm": 12.1875,
      "learning_rate": 3.5207822512567736e-05,
      "loss": 1.28,
      "num_input_tokens_seen": 33457744,
      "step": 57675
    },
    {
      "epoch": 8.591003872505214,
      "grad_norm": 11.6875,
      "learning_rate": 3.520485621786693e-05,
      "loss": 1.1551,
      "num_input_tokens_seen": 33460816,
      "step": 57680
    },
    {
      "epoch": 8.591748585046172,
      "grad_norm": 8.875,
      "learning_rate": 3.5201889750765446e-05,
      "loss": 1.185,
      "num_input_tokens_seen": 33463568,
      "step": 57685
    },
    {
      "epoch": 8.592493297587131,
      "grad_norm": 12.1875,
      "learning_rate": 3.51989231113134e-05,
      "loss": 0.6976,
      "num_input_tokens_seen": 33466672,
      "step": 57690
    },
    {
      "epoch": 8.59323801012809,
      "grad_norm": 10.1875,
      "learning_rate": 3.519595629956092e-05,
      "loss": 1.3552,
      "num_input_tokens_seen": 33469712,
      "step": 57695
    },
    {
      "epoch": 8.59398272266905,
      "grad_norm": 7.09375,
      "learning_rate": 3.519298931555812e-05,
      "loss": 1.0441,
      "num_input_tokens_seen": 33472464,
      "step": 57700
    },
    {
      "epoch": 8.59472743521001,
      "grad_norm": 10.0625,
      "learning_rate": 3.519002215935512e-05,
      "loss": 0.9625,
      "num_input_tokens_seen": 33475184,
      "step": 57705
    },
    {
      "epoch": 8.595472147750968,
      "grad_norm": 11.75,
      "learning_rate": 3.5187054831002064e-05,
      "loss": 1.0942,
      "num_input_tokens_seen": 33478256,
      "step": 57710
    },
    {
      "epoch": 8.596216860291927,
      "grad_norm": 9.0,
      "learning_rate": 3.5184087330549056e-05,
      "loss": 1.1891,
      "num_input_tokens_seen": 33481104,
      "step": 57715
    },
    {
      "epoch": 8.596961572832887,
      "grad_norm": 9.25,
      "learning_rate": 3.518111965804625e-05,
      "loss": 1.2075,
      "num_input_tokens_seen": 33483856,
      "step": 57720
    },
    {
      "epoch": 8.597706285373846,
      "grad_norm": 16.75,
      "learning_rate": 3.517815181354378e-05,
      "loss": 1.3152,
      "num_input_tokens_seen": 33486672,
      "step": 57725
    },
    {
      "epoch": 8.598450997914805,
      "grad_norm": 12.25,
      "learning_rate": 3.517518379709177e-05,
      "loss": 1.0439,
      "num_input_tokens_seen": 33489328,
      "step": 57730
    },
    {
      "epoch": 8.599195710455763,
      "grad_norm": 10.375,
      "learning_rate": 3.5172215608740376e-05,
      "loss": 0.7857,
      "num_input_tokens_seen": 33492048,
      "step": 57735
    },
    {
      "epoch": 8.599940422996724,
      "grad_norm": 9.6875,
      "learning_rate": 3.516924724853974e-05,
      "loss": 1.1325,
      "num_input_tokens_seen": 33494640,
      "step": 57740
    },
    {
      "epoch": 8.600685135537683,
      "grad_norm": 11.3125,
      "learning_rate": 3.5166278716540016e-05,
      "loss": 0.9575,
      "num_input_tokens_seen": 33497296,
      "step": 57745
    },
    {
      "epoch": 8.601429848078642,
      "grad_norm": 10.9375,
      "learning_rate": 3.5163310012791326e-05,
      "loss": 1.4808,
      "num_input_tokens_seen": 33499984,
      "step": 57750
    },
    {
      "epoch": 8.6021745606196,
      "grad_norm": 5.59375,
      "learning_rate": 3.516034113734385e-05,
      "loss": 1.1843,
      "num_input_tokens_seen": 33502736,
      "step": 57755
    },
    {
      "epoch": 8.60291927316056,
      "grad_norm": 17.0,
      "learning_rate": 3.515737209024774e-05,
      "loss": 1.388,
      "num_input_tokens_seen": 33505712,
      "step": 57760
    },
    {
      "epoch": 8.60366398570152,
      "grad_norm": 11.1875,
      "learning_rate": 3.515440287155315e-05,
      "loss": 1.2336,
      "num_input_tokens_seen": 33508592,
      "step": 57765
    },
    {
      "epoch": 8.604408698242478,
      "grad_norm": 15.9375,
      "learning_rate": 3.515143348131025e-05,
      "loss": 1.0157,
      "num_input_tokens_seen": 33511504,
      "step": 57770
    },
    {
      "epoch": 8.605153410783437,
      "grad_norm": 3.21875,
      "learning_rate": 3.514846391956919e-05,
      "loss": 0.8752,
      "num_input_tokens_seen": 33514288,
      "step": 57775
    },
    {
      "epoch": 8.605898123324398,
      "grad_norm": 17.5,
      "learning_rate": 3.514549418638015e-05,
      "loss": 1.2826,
      "num_input_tokens_seen": 33517104,
      "step": 57780
    },
    {
      "epoch": 8.606642835865356,
      "grad_norm": 8.9375,
      "learning_rate": 3.5142524281793296e-05,
      "loss": 0.982,
      "num_input_tokens_seen": 33520112,
      "step": 57785
    },
    {
      "epoch": 8.607387548406315,
      "grad_norm": 4.8125,
      "learning_rate": 3.513955420585881e-05,
      "loss": 1.1175,
      "num_input_tokens_seen": 33523056,
      "step": 57790
    },
    {
      "epoch": 8.608132260947274,
      "grad_norm": 5.40625,
      "learning_rate": 3.513658395862685e-05,
      "loss": 0.916,
      "num_input_tokens_seen": 33525904,
      "step": 57795
    },
    {
      "epoch": 8.608876973488233,
      "grad_norm": 15.75,
      "learning_rate": 3.5133613540147605e-05,
      "loss": 1.0195,
      "num_input_tokens_seen": 33528688,
      "step": 57800
    },
    {
      "epoch": 8.609621686029193,
      "grad_norm": 9.5625,
      "learning_rate": 3.513064295047127e-05,
      "loss": 1.0949,
      "num_input_tokens_seen": 33531472,
      "step": 57805
    },
    {
      "epoch": 8.610366398570152,
      "grad_norm": 7.0625,
      "learning_rate": 3.5127672189648016e-05,
      "loss": 0.8152,
      "num_input_tokens_seen": 33534256,
      "step": 57810
    },
    {
      "epoch": 8.61111111111111,
      "grad_norm": 15.5,
      "learning_rate": 3.512470125772803e-05,
      "loss": 1.0714,
      "num_input_tokens_seen": 33536912,
      "step": 57815
    },
    {
      "epoch": 8.61185582365207,
      "grad_norm": 8.25,
      "learning_rate": 3.5121730154761507e-05,
      "loss": 1.1596,
      "num_input_tokens_seen": 33539920,
      "step": 57820
    },
    {
      "epoch": 8.61260053619303,
      "grad_norm": 10.0625,
      "learning_rate": 3.511875888079864e-05,
      "loss": 1.0019,
      "num_input_tokens_seen": 33542768,
      "step": 57825
    },
    {
      "epoch": 8.613345248733989,
      "grad_norm": 9.125,
      "learning_rate": 3.511578743588963e-05,
      "loss": 1.179,
      "num_input_tokens_seen": 33545872,
      "step": 57830
    },
    {
      "epoch": 8.614089961274948,
      "grad_norm": 12.0,
      "learning_rate": 3.511281582008466e-05,
      "loss": 1.192,
      "num_input_tokens_seen": 33548816,
      "step": 57835
    },
    {
      "epoch": 8.614834673815906,
      "grad_norm": 10.5625,
      "learning_rate": 3.5109844033433963e-05,
      "loss": 1.2375,
      "num_input_tokens_seen": 33551632,
      "step": 57840
    },
    {
      "epoch": 8.615579386356867,
      "grad_norm": 13.625,
      "learning_rate": 3.5106872075987716e-05,
      "loss": 0.9994,
      "num_input_tokens_seen": 33554576,
      "step": 57845
    },
    {
      "epoch": 8.616324098897826,
      "grad_norm": 10.1875,
      "learning_rate": 3.5103899947796134e-05,
      "loss": 1.233,
      "num_input_tokens_seen": 33557296,
      "step": 57850
    },
    {
      "epoch": 8.617068811438784,
      "grad_norm": 9.6875,
      "learning_rate": 3.510092764890944e-05,
      "loss": 1.2675,
      "num_input_tokens_seen": 33560272,
      "step": 57855
    },
    {
      "epoch": 8.617813523979743,
      "grad_norm": 15.5625,
      "learning_rate": 3.509795517937784e-05,
      "loss": 1.5515,
      "num_input_tokens_seen": 33563216,
      "step": 57860
    },
    {
      "epoch": 8.618558236520704,
      "grad_norm": 8.4375,
      "learning_rate": 3.5094982539251545e-05,
      "loss": 0.825,
      "num_input_tokens_seen": 33566032,
      "step": 57865
    },
    {
      "epoch": 8.619302949061662,
      "grad_norm": 7.3125,
      "learning_rate": 3.5092009728580784e-05,
      "loss": 1.4222,
      "num_input_tokens_seen": 33568976,
      "step": 57870
    },
    {
      "epoch": 8.620047661602621,
      "grad_norm": 6.34375,
      "learning_rate": 3.5089036747415775e-05,
      "loss": 1.0719,
      "num_input_tokens_seen": 33571696,
      "step": 57875
    },
    {
      "epoch": 8.62079237414358,
      "grad_norm": 9.125,
      "learning_rate": 3.508606359580674e-05,
      "loss": 1.152,
      "num_input_tokens_seen": 33574768,
      "step": 57880
    },
    {
      "epoch": 8.62153708668454,
      "grad_norm": 7.09375,
      "learning_rate": 3.508309027380392e-05,
      "loss": 1.2033,
      "num_input_tokens_seen": 33578096,
      "step": 57885
    },
    {
      "epoch": 8.6222817992255,
      "grad_norm": 6.90625,
      "learning_rate": 3.508011678145752e-05,
      "loss": 0.8482,
      "num_input_tokens_seen": 33580688,
      "step": 57890
    },
    {
      "epoch": 8.623026511766458,
      "grad_norm": 5.75,
      "learning_rate": 3.5077143118817805e-05,
      "loss": 1.0097,
      "num_input_tokens_seen": 33583984,
      "step": 57895
    },
    {
      "epoch": 8.623771224307417,
      "grad_norm": 14.625,
      "learning_rate": 3.5074169285935e-05,
      "loss": 1.2478,
      "num_input_tokens_seen": 33586864,
      "step": 57900
    },
    {
      "epoch": 8.624515936848377,
      "grad_norm": 3.484375,
      "learning_rate": 3.5071195282859345e-05,
      "loss": 1.2439,
      "num_input_tokens_seen": 33590032,
      "step": 57905
    },
    {
      "epoch": 8.625260649389336,
      "grad_norm": 13.625,
      "learning_rate": 3.506822110964108e-05,
      "loss": 1.2612,
      "num_input_tokens_seen": 33593232,
      "step": 57910
    },
    {
      "epoch": 8.626005361930295,
      "grad_norm": 13.625,
      "learning_rate": 3.506524676633045e-05,
      "loss": 1.3958,
      "num_input_tokens_seen": 33595792,
      "step": 57915
    },
    {
      "epoch": 8.626750074471254,
      "grad_norm": 8.8125,
      "learning_rate": 3.50622722529777e-05,
      "loss": 1.1294,
      "num_input_tokens_seen": 33598512,
      "step": 57920
    },
    {
      "epoch": 8.627494787012214,
      "grad_norm": 6.34375,
      "learning_rate": 3.5059297569633096e-05,
      "loss": 0.9682,
      "num_input_tokens_seen": 33601232,
      "step": 57925
    },
    {
      "epoch": 8.628239499553173,
      "grad_norm": 10.75,
      "learning_rate": 3.505632271634688e-05,
      "loss": 1.3212,
      "num_input_tokens_seen": 33604048,
      "step": 57930
    },
    {
      "epoch": 8.628984212094132,
      "grad_norm": 6.9375,
      "learning_rate": 3.505334769316931e-05,
      "loss": 1.128,
      "num_input_tokens_seen": 33607184,
      "step": 57935
    },
    {
      "epoch": 8.62972892463509,
      "grad_norm": 7.1875,
      "learning_rate": 3.505037250015066e-05,
      "loss": 0.9838,
      "num_input_tokens_seen": 33609680,
      "step": 57940
    },
    {
      "epoch": 8.63047363717605,
      "grad_norm": 8.5625,
      "learning_rate": 3.504739713734118e-05,
      "loss": 0.948,
      "num_input_tokens_seen": 33612560,
      "step": 57945
    },
    {
      "epoch": 8.63121834971701,
      "grad_norm": 9.25,
      "learning_rate": 3.504442160479112e-05,
      "loss": 1.3481,
      "num_input_tokens_seen": 33615728,
      "step": 57950
    },
    {
      "epoch": 8.631963062257968,
      "grad_norm": 7.1875,
      "learning_rate": 3.5041445902550776e-05,
      "loss": 0.8213,
      "num_input_tokens_seen": 33618416,
      "step": 57955
    },
    {
      "epoch": 8.632707774798927,
      "grad_norm": 6.28125,
      "learning_rate": 3.503847003067041e-05,
      "loss": 1.1365,
      "num_input_tokens_seen": 33621232,
      "step": 57960
    },
    {
      "epoch": 8.633452487339888,
      "grad_norm": 15.25,
      "learning_rate": 3.503549398920029e-05,
      "loss": 1.0553,
      "num_input_tokens_seen": 33624592,
      "step": 57965
    },
    {
      "epoch": 8.634197199880846,
      "grad_norm": 11.75,
      "learning_rate": 3.503251777819071e-05,
      "loss": 1.161,
      "num_input_tokens_seen": 33627696,
      "step": 57970
    },
    {
      "epoch": 8.634941912421805,
      "grad_norm": 5.46875,
      "learning_rate": 3.502954139769193e-05,
      "loss": 1.2013,
      "num_input_tokens_seen": 33630544,
      "step": 57975
    },
    {
      "epoch": 8.635686624962764,
      "grad_norm": 4.5,
      "learning_rate": 3.502656484775424e-05,
      "loss": 1.0155,
      "num_input_tokens_seen": 33633424,
      "step": 57980
    },
    {
      "epoch": 8.636431337503723,
      "grad_norm": 11.375,
      "learning_rate": 3.502358812842794e-05,
      "loss": 1.2476,
      "num_input_tokens_seen": 33636176,
      "step": 57985
    },
    {
      "epoch": 8.637176050044683,
      "grad_norm": 11.0625,
      "learning_rate": 3.502061123976329e-05,
      "loss": 0.9264,
      "num_input_tokens_seen": 33639120,
      "step": 57990
    },
    {
      "epoch": 8.637920762585642,
      "grad_norm": 12.0625,
      "learning_rate": 3.50176341818106e-05,
      "loss": 0.9788,
      "num_input_tokens_seen": 33641872,
      "step": 57995
    },
    {
      "epoch": 8.6386654751266,
      "grad_norm": 11.125,
      "learning_rate": 3.5014656954620174e-05,
      "loss": 1.2721,
      "num_input_tokens_seen": 33644816,
      "step": 58000
    },
    {
      "epoch": 8.63941018766756,
      "grad_norm": 8.0625,
      "learning_rate": 3.5011679558242286e-05,
      "loss": 0.8833,
      "num_input_tokens_seen": 33647664,
      "step": 58005
    },
    {
      "epoch": 8.64015490020852,
      "grad_norm": 14.25,
      "learning_rate": 3.5008701992727254e-05,
      "loss": 1.1274,
      "num_input_tokens_seen": 33650288,
      "step": 58010
    },
    {
      "epoch": 8.640899612749479,
      "grad_norm": 11.3125,
      "learning_rate": 3.500572425812537e-05,
      "loss": 1.0454,
      "num_input_tokens_seen": 33653552,
      "step": 58015
    },
    {
      "epoch": 8.641644325290438,
      "grad_norm": 13.0,
      "learning_rate": 3.500274635448694e-05,
      "loss": 1.4814,
      "num_input_tokens_seen": 33656336,
      "step": 58020
    },
    {
      "epoch": 8.642389037831396,
      "grad_norm": 10.5,
      "learning_rate": 3.499976828186229e-05,
      "loss": 1.0285,
      "num_input_tokens_seen": 33659376,
      "step": 58025
    },
    {
      "epoch": 8.643133750372357,
      "grad_norm": 7.3125,
      "learning_rate": 3.499679004030171e-05,
      "loss": 1.117,
      "num_input_tokens_seen": 33662320,
      "step": 58030
    },
    {
      "epoch": 8.643878462913316,
      "grad_norm": 12.8125,
      "learning_rate": 3.499381162985552e-05,
      "loss": 0.9935,
      "num_input_tokens_seen": 33665200,
      "step": 58035
    },
    {
      "epoch": 8.644623175454274,
      "grad_norm": 4.625,
      "learning_rate": 3.499083305057405e-05,
      "loss": 0.8694,
      "num_input_tokens_seen": 33668272,
      "step": 58040
    },
    {
      "epoch": 8.645367887995233,
      "grad_norm": 8.625,
      "learning_rate": 3.49878543025076e-05,
      "loss": 1.1166,
      "num_input_tokens_seen": 33671088,
      "step": 58045
    },
    {
      "epoch": 8.646112600536194,
      "grad_norm": 11.0,
      "learning_rate": 3.49848753857065e-05,
      "loss": 0.9115,
      "num_input_tokens_seen": 33673936,
      "step": 58050
    },
    {
      "epoch": 8.646857313077152,
      "grad_norm": 8.9375,
      "learning_rate": 3.4981896300221084e-05,
      "loss": 1.068,
      "num_input_tokens_seen": 33676784,
      "step": 58055
    },
    {
      "epoch": 8.647602025618111,
      "grad_norm": 14.0625,
      "learning_rate": 3.497891704610167e-05,
      "loss": 1.4426,
      "num_input_tokens_seen": 33679632,
      "step": 58060
    },
    {
      "epoch": 8.64834673815907,
      "grad_norm": 10.625,
      "learning_rate": 3.49759376233986e-05,
      "loss": 1.2102,
      "num_input_tokens_seen": 33682448,
      "step": 58065
    },
    {
      "epoch": 8.64909145070003,
      "grad_norm": 13.6875,
      "learning_rate": 3.4972958032162204e-05,
      "loss": 1.2797,
      "num_input_tokens_seen": 33685328,
      "step": 58070
    },
    {
      "epoch": 8.64983616324099,
      "grad_norm": 15.0,
      "learning_rate": 3.496997827244282e-05,
      "loss": 1.2619,
      "num_input_tokens_seen": 33688048,
      "step": 58075
    },
    {
      "epoch": 8.650580875781948,
      "grad_norm": 10.25,
      "learning_rate": 3.496699834429078e-05,
      "loss": 1.1497,
      "num_input_tokens_seen": 33690800,
      "step": 58080
    },
    {
      "epoch": 8.651325588322907,
      "grad_norm": 13.875,
      "learning_rate": 3.4964018247756434e-05,
      "loss": 1.1354,
      "num_input_tokens_seen": 33693648,
      "step": 58085
    },
    {
      "epoch": 8.652070300863867,
      "grad_norm": 14.1875,
      "learning_rate": 3.4961037982890135e-05,
      "loss": 1.0672,
      "num_input_tokens_seen": 33696368,
      "step": 58090
    },
    {
      "epoch": 8.652815013404826,
      "grad_norm": 11.1875,
      "learning_rate": 3.495805754974221e-05,
      "loss": 1.0261,
      "num_input_tokens_seen": 33699056,
      "step": 58095
    },
    {
      "epoch": 8.653559725945785,
      "grad_norm": 3.890625,
      "learning_rate": 3.495507694836304e-05,
      "loss": 1.0347,
      "num_input_tokens_seen": 33702096,
      "step": 58100
    },
    {
      "epoch": 8.654304438486744,
      "grad_norm": 11.0,
      "learning_rate": 3.4952096178802946e-05,
      "loss": 0.9098,
      "num_input_tokens_seen": 33704784,
      "step": 58105
    },
    {
      "epoch": 8.655049151027704,
      "grad_norm": 10.625,
      "learning_rate": 3.4949115241112314e-05,
      "loss": 1.2017,
      "num_input_tokens_seen": 33707472,
      "step": 58110
    },
    {
      "epoch": 8.655793863568663,
      "grad_norm": 9.25,
      "learning_rate": 3.4946134135341486e-05,
      "loss": 0.9282,
      "num_input_tokens_seen": 33710320,
      "step": 58115
    },
    {
      "epoch": 8.656538576109622,
      "grad_norm": 10.875,
      "learning_rate": 3.494315286154083e-05,
      "loss": 1.3675,
      "num_input_tokens_seen": 33713072,
      "step": 58120
    },
    {
      "epoch": 8.65728328865058,
      "grad_norm": 11.875,
      "learning_rate": 3.494017141976071e-05,
      "loss": 1.1857,
      "num_input_tokens_seen": 33716016,
      "step": 58125
    },
    {
      "epoch": 8.65802800119154,
      "grad_norm": 14.6875,
      "learning_rate": 3.4937189810051494e-05,
      "loss": 1.4305,
      "num_input_tokens_seen": 33718832,
      "step": 58130
    },
    {
      "epoch": 8.6587727137325,
      "grad_norm": 8.25,
      "learning_rate": 3.4934208032463565e-05,
      "loss": 1.3851,
      "num_input_tokens_seen": 33722096,
      "step": 58135
    },
    {
      "epoch": 8.659517426273458,
      "grad_norm": 19.125,
      "learning_rate": 3.4931226087047285e-05,
      "loss": 1.5756,
      "num_input_tokens_seen": 33724816,
      "step": 58140
    },
    {
      "epoch": 8.660262138814417,
      "grad_norm": 11.3125,
      "learning_rate": 3.4928243973853044e-05,
      "loss": 1.1128,
      "num_input_tokens_seen": 33727888,
      "step": 58145
    },
    {
      "epoch": 8.661006851355378,
      "grad_norm": 6.75,
      "learning_rate": 3.49252616929312e-05,
      "loss": 0.8423,
      "num_input_tokens_seen": 33730576,
      "step": 58150
    },
    {
      "epoch": 8.661751563896336,
      "grad_norm": 10.625,
      "learning_rate": 3.492227924433215e-05,
      "loss": 0.9839,
      "num_input_tokens_seen": 33733200,
      "step": 58155
    },
    {
      "epoch": 8.662496276437295,
      "grad_norm": 9.875,
      "learning_rate": 3.491929662810627e-05,
      "loss": 0.8746,
      "num_input_tokens_seen": 33736016,
      "step": 58160
    },
    {
      "epoch": 8.663240988978254,
      "grad_norm": 11.3125,
      "learning_rate": 3.491631384430396e-05,
      "loss": 0.7668,
      "num_input_tokens_seen": 33738704,
      "step": 58165
    },
    {
      "epoch": 8.663985701519213,
      "grad_norm": 7.125,
      "learning_rate": 3.4913330892975606e-05,
      "loss": 1.0431,
      "num_input_tokens_seen": 33741488,
      "step": 58170
    },
    {
      "epoch": 8.664730414060173,
      "grad_norm": 8.5,
      "learning_rate": 3.4910347774171606e-05,
      "loss": 0.9995,
      "num_input_tokens_seen": 33744272,
      "step": 58175
    },
    {
      "epoch": 8.665475126601132,
      "grad_norm": 9.1875,
      "learning_rate": 3.490736448794235e-05,
      "loss": 1.3226,
      "num_input_tokens_seen": 33746864,
      "step": 58180
    },
    {
      "epoch": 8.66621983914209,
      "grad_norm": 6.59375,
      "learning_rate": 3.490438103433824e-05,
      "loss": 1.1882,
      "num_input_tokens_seen": 33750000,
      "step": 58185
    },
    {
      "epoch": 8.66696455168305,
      "grad_norm": 12.375,
      "learning_rate": 3.490139741340967e-05,
      "loss": 1.208,
      "num_input_tokens_seen": 33753392,
      "step": 58190
    },
    {
      "epoch": 8.66770926422401,
      "grad_norm": 20.875,
      "learning_rate": 3.4898413625207067e-05,
      "loss": 1.0919,
      "num_input_tokens_seen": 33756144,
      "step": 58195
    },
    {
      "epoch": 8.668453976764969,
      "grad_norm": 13.75,
      "learning_rate": 3.4895429669780824e-05,
      "loss": 1.3116,
      "num_input_tokens_seen": 33758864,
      "step": 58200
    },
    {
      "epoch": 8.669198689305928,
      "grad_norm": 11.625,
      "learning_rate": 3.4892445547181354e-05,
      "loss": 1.1947,
      "num_input_tokens_seen": 33761648,
      "step": 58205
    },
    {
      "epoch": 8.669943401846886,
      "grad_norm": 5.40625,
      "learning_rate": 3.4889461257459065e-05,
      "loss": 0.9459,
      "num_input_tokens_seen": 33764304,
      "step": 58210
    },
    {
      "epoch": 8.670688114387847,
      "grad_norm": 15.25,
      "learning_rate": 3.488647680066438e-05,
      "loss": 1.0068,
      "num_input_tokens_seen": 33767280,
      "step": 58215
    },
    {
      "epoch": 8.671432826928806,
      "grad_norm": 17.875,
      "learning_rate": 3.4883492176847724e-05,
      "loss": 1.1484,
      "num_input_tokens_seen": 33770192,
      "step": 58220
    },
    {
      "epoch": 8.672177539469764,
      "grad_norm": 16.0,
      "learning_rate": 3.488050738605951e-05,
      "loss": 1.3714,
      "num_input_tokens_seen": 33772912,
      "step": 58225
    },
    {
      "epoch": 8.672922252010723,
      "grad_norm": 12.5,
      "learning_rate": 3.4877522428350165e-05,
      "loss": 0.9441,
      "num_input_tokens_seen": 33775856,
      "step": 58230
    },
    {
      "epoch": 8.673666964551684,
      "grad_norm": 14.125,
      "learning_rate": 3.487453730377011e-05,
      "loss": 1.0534,
      "num_input_tokens_seen": 33778928,
      "step": 58235
    },
    {
      "epoch": 8.674411677092642,
      "grad_norm": 12.0625,
      "learning_rate": 3.4871552012369793e-05,
      "loss": 1.0112,
      "num_input_tokens_seen": 33781680,
      "step": 58240
    },
    {
      "epoch": 8.675156389633601,
      "grad_norm": 13.625,
      "learning_rate": 3.486856655419964e-05,
      "loss": 1.2521,
      "num_input_tokens_seen": 33784528,
      "step": 58245
    },
    {
      "epoch": 8.67590110217456,
      "grad_norm": 6.0625,
      "learning_rate": 3.4865580929310074e-05,
      "loss": 1.0038,
      "num_input_tokens_seen": 33787408,
      "step": 58250
    },
    {
      "epoch": 8.67664581471552,
      "grad_norm": 12.0,
      "learning_rate": 3.486259513775155e-05,
      "loss": 1.1569,
      "num_input_tokens_seen": 33790352,
      "step": 58255
    },
    {
      "epoch": 8.67739052725648,
      "grad_norm": 11.3125,
      "learning_rate": 3.485960917957451e-05,
      "loss": 1.0507,
      "num_input_tokens_seen": 33793168,
      "step": 58260
    },
    {
      "epoch": 8.678135239797438,
      "grad_norm": 8.375,
      "learning_rate": 3.4856623054829395e-05,
      "loss": 0.9203,
      "num_input_tokens_seen": 33795952,
      "step": 58265
    },
    {
      "epoch": 8.678879952338397,
      "grad_norm": 16.125,
      "learning_rate": 3.4853636763566646e-05,
      "loss": 1.3276,
      "num_input_tokens_seen": 33798704,
      "step": 58270
    },
    {
      "epoch": 8.679624664879357,
      "grad_norm": 12.9375,
      "learning_rate": 3.485065030583672e-05,
      "loss": 0.9806,
      "num_input_tokens_seen": 33801616,
      "step": 58275
    },
    {
      "epoch": 8.680369377420316,
      "grad_norm": 7.5625,
      "learning_rate": 3.484766368169007e-05,
      "loss": 1.0022,
      "num_input_tokens_seen": 33804752,
      "step": 58280
    },
    {
      "epoch": 8.681114089961275,
      "grad_norm": 9.1875,
      "learning_rate": 3.484467689117715e-05,
      "loss": 1.2614,
      "num_input_tokens_seen": 33807856,
      "step": 58285
    },
    {
      "epoch": 8.681858802502234,
      "grad_norm": 8.0625,
      "learning_rate": 3.4841689934348416e-05,
      "loss": 0.8545,
      "num_input_tokens_seen": 33810480,
      "step": 58290
    },
    {
      "epoch": 8.682603515043194,
      "grad_norm": 15.1875,
      "learning_rate": 3.483870281125433e-05,
      "loss": 1.1183,
      "num_input_tokens_seen": 33813360,
      "step": 58295
    },
    {
      "epoch": 8.683348227584153,
      "grad_norm": 9.5625,
      "learning_rate": 3.483571552194537e-05,
      "loss": 0.9886,
      "num_input_tokens_seen": 33816304,
      "step": 58300
    },
    {
      "epoch": 8.684092940125112,
      "grad_norm": 9.5625,
      "learning_rate": 3.4832728066471994e-05,
      "loss": 0.9343,
      "num_input_tokens_seen": 33819312,
      "step": 58305
    },
    {
      "epoch": 8.68483765266607,
      "grad_norm": 8.125,
      "learning_rate": 3.482974044488466e-05,
      "loss": 1.1491,
      "num_input_tokens_seen": 33822352,
      "step": 58310
    },
    {
      "epoch": 8.68558236520703,
      "grad_norm": 10.5625,
      "learning_rate": 3.4826752657233855e-05,
      "loss": 0.9487,
      "num_input_tokens_seen": 33825104,
      "step": 58315
    },
    {
      "epoch": 8.68632707774799,
      "grad_norm": 9.8125,
      "learning_rate": 3.4823764703570054e-05,
      "loss": 1.6039,
      "num_input_tokens_seen": 33827856,
      "step": 58320
    },
    {
      "epoch": 8.687071790288948,
      "grad_norm": 9.9375,
      "learning_rate": 3.482077658394373e-05,
      "loss": 1.1145,
      "num_input_tokens_seen": 33830640,
      "step": 58325
    },
    {
      "epoch": 8.687816502829907,
      "grad_norm": 7.09375,
      "learning_rate": 3.481778829840537e-05,
      "loss": 0.9205,
      "num_input_tokens_seen": 33833520,
      "step": 58330
    },
    {
      "epoch": 8.688561215370868,
      "grad_norm": 10.625,
      "learning_rate": 3.481479984700546e-05,
      "loss": 1.0333,
      "num_input_tokens_seen": 33836496,
      "step": 58335
    },
    {
      "epoch": 8.689305927911827,
      "grad_norm": 11.5625,
      "learning_rate": 3.481181122979447e-05,
      "loss": 0.8954,
      "num_input_tokens_seen": 33839504,
      "step": 58340
    },
    {
      "epoch": 8.690050640452785,
      "grad_norm": 4.53125,
      "learning_rate": 3.480882244682291e-05,
      "loss": 0.9955,
      "num_input_tokens_seen": 33842384,
      "step": 58345
    },
    {
      "epoch": 8.690795352993744,
      "grad_norm": 8.75,
      "learning_rate": 3.480583349814126e-05,
      "loss": 1.3663,
      "num_input_tokens_seen": 33845168,
      "step": 58350
    },
    {
      "epoch": 8.691540065534703,
      "grad_norm": 7.21875,
      "learning_rate": 3.480284438380002e-05,
      "loss": 1.1223,
      "num_input_tokens_seen": 33848304,
      "step": 58355
    },
    {
      "epoch": 8.692284778075663,
      "grad_norm": 11.0625,
      "learning_rate": 3.479985510384969e-05,
      "loss": 1.0365,
      "num_input_tokens_seen": 33851088,
      "step": 58360
    },
    {
      "epoch": 8.693029490616622,
      "grad_norm": 6.28125,
      "learning_rate": 3.479686565834077e-05,
      "loss": 1.2824,
      "num_input_tokens_seen": 33854192,
      "step": 58365
    },
    {
      "epoch": 8.69377420315758,
      "grad_norm": 9.1875,
      "learning_rate": 3.479387604732376e-05,
      "loss": 0.9499,
      "num_input_tokens_seen": 33857392,
      "step": 58370
    },
    {
      "epoch": 8.69451891569854,
      "grad_norm": 9.8125,
      "learning_rate": 3.479088627084916e-05,
      "loss": 1.0477,
      "num_input_tokens_seen": 33859984,
      "step": 58375
    },
    {
      "epoch": 8.6952636282395,
      "grad_norm": 9.25,
      "learning_rate": 3.4787896328967493e-05,
      "loss": 1.028,
      "num_input_tokens_seen": 33863056,
      "step": 58380
    },
    {
      "epoch": 8.696008340780459,
      "grad_norm": 9.75,
      "learning_rate": 3.478490622172926e-05,
      "loss": 1.0572,
      "num_input_tokens_seen": 33865872,
      "step": 58385
    },
    {
      "epoch": 8.696753053321418,
      "grad_norm": 7.8125,
      "learning_rate": 3.478191594918499e-05,
      "loss": 1.1143,
      "num_input_tokens_seen": 33868592,
      "step": 58390
    },
    {
      "epoch": 8.697497765862376,
      "grad_norm": 9.1875,
      "learning_rate": 3.477892551138519e-05,
      "loss": 1.2052,
      "num_input_tokens_seen": 33871696,
      "step": 58395
    },
    {
      "epoch": 8.698242478403337,
      "grad_norm": 10.5625,
      "learning_rate": 3.4775934908380386e-05,
      "loss": 1.0946,
      "num_input_tokens_seen": 33874672,
      "step": 58400
    },
    {
      "epoch": 8.698987190944296,
      "grad_norm": 16.0,
      "learning_rate": 3.4772944140221094e-05,
      "loss": 1.271,
      "num_input_tokens_seen": 33877552,
      "step": 58405
    },
    {
      "epoch": 8.699731903485254,
      "grad_norm": 7.21875,
      "learning_rate": 3.476995320695784e-05,
      "loss": 1.1167,
      "num_input_tokens_seen": 33880528,
      "step": 58410
    },
    {
      "epoch": 8.700476616026213,
      "grad_norm": 7.34375,
      "learning_rate": 3.476696210864116e-05,
      "loss": 0.9543,
      "num_input_tokens_seen": 33883600,
      "step": 58415
    },
    {
      "epoch": 8.701221328567174,
      "grad_norm": 9.4375,
      "learning_rate": 3.476397084532158e-05,
      "loss": 0.8129,
      "num_input_tokens_seen": 33886512,
      "step": 58420
    },
    {
      "epoch": 8.701966041108133,
      "grad_norm": 8.75,
      "learning_rate": 3.476097941704964e-05,
      "loss": 1.1013,
      "num_input_tokens_seen": 33889456,
      "step": 58425
    },
    {
      "epoch": 8.702710753649091,
      "grad_norm": 7.9375,
      "learning_rate": 3.475798782387587e-05,
      "loss": 0.887,
      "num_input_tokens_seen": 33892304,
      "step": 58430
    },
    {
      "epoch": 8.70345546619005,
      "grad_norm": 8.8125,
      "learning_rate": 3.475499606585081e-05,
      "loss": 1.1848,
      "num_input_tokens_seen": 33894960,
      "step": 58435
    },
    {
      "epoch": 8.70420017873101,
      "grad_norm": 4.53125,
      "learning_rate": 3.4752004143025016e-05,
      "loss": 0.9068,
      "num_input_tokens_seen": 33897808,
      "step": 58440
    },
    {
      "epoch": 8.70494489127197,
      "grad_norm": 9.0,
      "learning_rate": 3.4749012055449015e-05,
      "loss": 0.8988,
      "num_input_tokens_seen": 33900624,
      "step": 58445
    },
    {
      "epoch": 8.705689603812928,
      "grad_norm": 6.03125,
      "learning_rate": 3.4746019803173365e-05,
      "loss": 0.9187,
      "num_input_tokens_seen": 33903472,
      "step": 58450
    },
    {
      "epoch": 8.706434316353887,
      "grad_norm": 5.34375,
      "learning_rate": 3.474302738624862e-05,
      "loss": 0.9261,
      "num_input_tokens_seen": 33906192,
      "step": 58455
    },
    {
      "epoch": 8.707179028894847,
      "grad_norm": 9.1875,
      "learning_rate": 3.474003480472532e-05,
      "loss": 1.1148,
      "num_input_tokens_seen": 33908880,
      "step": 58460
    },
    {
      "epoch": 8.707923741435806,
      "grad_norm": 4.40625,
      "learning_rate": 3.473704205865405e-05,
      "loss": 0.9321,
      "num_input_tokens_seen": 33912432,
      "step": 58465
    },
    {
      "epoch": 8.708668453976765,
      "grad_norm": 9.9375,
      "learning_rate": 3.473404914808534e-05,
      "loss": 1.3329,
      "num_input_tokens_seen": 33915440,
      "step": 58470
    },
    {
      "epoch": 8.709413166517724,
      "grad_norm": 17.25,
      "learning_rate": 3.4731056073069754e-05,
      "loss": 1.0881,
      "num_input_tokens_seen": 33918704,
      "step": 58475
    },
    {
      "epoch": 8.710157879058684,
      "grad_norm": 14.125,
      "learning_rate": 3.472806283365788e-05,
      "loss": 0.9998,
      "num_input_tokens_seen": 33921584,
      "step": 58480
    },
    {
      "epoch": 8.710902591599643,
      "grad_norm": 12.5,
      "learning_rate": 3.472506942990026e-05,
      "loss": 0.9449,
      "num_input_tokens_seen": 33924880,
      "step": 58485
    },
    {
      "epoch": 8.711647304140602,
      "grad_norm": 15.3125,
      "learning_rate": 3.472207586184748e-05,
      "loss": 1.132,
      "num_input_tokens_seen": 33927760,
      "step": 58490
    },
    {
      "epoch": 8.71239201668156,
      "grad_norm": 8.875,
      "learning_rate": 3.4719082129550106e-05,
      "loss": 0.9374,
      "num_input_tokens_seen": 33930736,
      "step": 58495
    },
    {
      "epoch": 8.71313672922252,
      "grad_norm": 11.875,
      "learning_rate": 3.471608823305873e-05,
      "loss": 1.137,
      "num_input_tokens_seen": 33933680,
      "step": 58500
    },
    {
      "epoch": 8.71388144176348,
      "grad_norm": 9.375,
      "learning_rate": 3.471309417242391e-05,
      "loss": 1.0795,
      "num_input_tokens_seen": 33936688,
      "step": 58505
    },
    {
      "epoch": 8.714626154304439,
      "grad_norm": 8.375,
      "learning_rate": 3.471009994769624e-05,
      "loss": 1.292,
      "num_input_tokens_seen": 33939696,
      "step": 58510
    },
    {
      "epoch": 8.715370866845397,
      "grad_norm": 16.625,
      "learning_rate": 3.470710555892629e-05,
      "loss": 1.1293,
      "num_input_tokens_seen": 33942896,
      "step": 58515
    },
    {
      "epoch": 8.716115579386356,
      "grad_norm": 8.75,
      "learning_rate": 3.470411100616466e-05,
      "loss": 1.0626,
      "num_input_tokens_seen": 33945936,
      "step": 58520
    },
    {
      "epoch": 8.716860291927317,
      "grad_norm": 5.8125,
      "learning_rate": 3.4701116289461945e-05,
      "loss": 0.974,
      "num_input_tokens_seen": 33948912,
      "step": 58525
    },
    {
      "epoch": 8.717605004468275,
      "grad_norm": 10.125,
      "learning_rate": 3.469812140886872e-05,
      "loss": 1.032,
      "num_input_tokens_seen": 33951792,
      "step": 58530
    },
    {
      "epoch": 8.718349717009234,
      "grad_norm": 13.9375,
      "learning_rate": 3.4695126364435604e-05,
      "loss": 1.3468,
      "num_input_tokens_seen": 33955088,
      "step": 58535
    },
    {
      "epoch": 8.719094429550193,
      "grad_norm": 10.0,
      "learning_rate": 3.4692131156213175e-05,
      "loss": 0.9246,
      "num_input_tokens_seen": 33958160,
      "step": 58540
    },
    {
      "epoch": 8.719839142091153,
      "grad_norm": 5.78125,
      "learning_rate": 3.468913578425203e-05,
      "loss": 0.802,
      "num_input_tokens_seen": 33961104,
      "step": 58545
    },
    {
      "epoch": 8.720583854632112,
      "grad_norm": 6.6875,
      "learning_rate": 3.4686140248602804e-05,
      "loss": 0.897,
      "num_input_tokens_seen": 33963888,
      "step": 58550
    },
    {
      "epoch": 8.721328567173071,
      "grad_norm": 13.1875,
      "learning_rate": 3.468314454931607e-05,
      "loss": 0.9626,
      "num_input_tokens_seen": 33966544,
      "step": 58555
    },
    {
      "epoch": 8.72207327971403,
      "grad_norm": 9.6875,
      "learning_rate": 3.468014868644245e-05,
      "loss": 0.9819,
      "num_input_tokens_seen": 33969392,
      "step": 58560
    },
    {
      "epoch": 8.72281799225499,
      "grad_norm": 12.25,
      "learning_rate": 3.4677152660032565e-05,
      "loss": 0.9731,
      "num_input_tokens_seen": 33972112,
      "step": 58565
    },
    {
      "epoch": 8.723562704795949,
      "grad_norm": 5.4375,
      "learning_rate": 3.467415647013702e-05,
      "loss": 0.9356,
      "num_input_tokens_seen": 33974896,
      "step": 58570
    },
    {
      "epoch": 8.724307417336908,
      "grad_norm": 8.0,
      "learning_rate": 3.467116011680643e-05,
      "loss": 0.8598,
      "num_input_tokens_seen": 33977872,
      "step": 58575
    },
    {
      "epoch": 8.725052129877866,
      "grad_norm": 9.9375,
      "learning_rate": 3.4668163600091415e-05,
      "loss": 0.8823,
      "num_input_tokens_seen": 33980624,
      "step": 58580
    },
    {
      "epoch": 8.725796842418827,
      "grad_norm": 19.75,
      "learning_rate": 3.46651669200426e-05,
      "loss": 1.0583,
      "num_input_tokens_seen": 33983344,
      "step": 58585
    },
    {
      "epoch": 8.726541554959786,
      "grad_norm": 7.375,
      "learning_rate": 3.4662170076710624e-05,
      "loss": 1.4258,
      "num_input_tokens_seen": 33986096,
      "step": 58590
    },
    {
      "epoch": 8.727286267500745,
      "grad_norm": 12.6875,
      "learning_rate": 3.46591730701461e-05,
      "loss": 1.0105,
      "num_input_tokens_seen": 33989200,
      "step": 58595
    },
    {
      "epoch": 8.728030980041703,
      "grad_norm": 7.21875,
      "learning_rate": 3.465617590039967e-05,
      "loss": 0.974,
      "num_input_tokens_seen": 33992240,
      "step": 58600
    },
    {
      "epoch": 8.728775692582664,
      "grad_norm": 5.40625,
      "learning_rate": 3.4653178567521956e-05,
      "loss": 0.8964,
      "num_input_tokens_seen": 33994896,
      "step": 58605
    },
    {
      "epoch": 8.729520405123623,
      "grad_norm": 7.5,
      "learning_rate": 3.4650181071563595e-05,
      "loss": 1.0667,
      "num_input_tokens_seen": 33998000,
      "step": 58610
    },
    {
      "epoch": 8.730265117664581,
      "grad_norm": 19.0,
      "learning_rate": 3.4647183412575243e-05,
      "loss": 1.0874,
      "num_input_tokens_seen": 34000944,
      "step": 58615
    },
    {
      "epoch": 8.73100983020554,
      "grad_norm": 8.5,
      "learning_rate": 3.464418559060753e-05,
      "loss": 0.7958,
      "num_input_tokens_seen": 34003696,
      "step": 58620
    },
    {
      "epoch": 8.7317545427465,
      "grad_norm": 13.625,
      "learning_rate": 3.464118760571109e-05,
      "loss": 0.9276,
      "num_input_tokens_seen": 34006576,
      "step": 58625
    },
    {
      "epoch": 8.73249925528746,
      "grad_norm": 7.84375,
      "learning_rate": 3.463818945793661e-05,
      "loss": 1.1807,
      "num_input_tokens_seen": 34009328,
      "step": 58630
    },
    {
      "epoch": 8.733243967828418,
      "grad_norm": 14.9375,
      "learning_rate": 3.46351911473347e-05,
      "loss": 1.0935,
      "num_input_tokens_seen": 34012304,
      "step": 58635
    },
    {
      "epoch": 8.733988680369377,
      "grad_norm": 19.375,
      "learning_rate": 3.463219267395603e-05,
      "loss": 0.8709,
      "num_input_tokens_seen": 34015024,
      "step": 58640
    },
    {
      "epoch": 8.734733392910336,
      "grad_norm": 12.25,
      "learning_rate": 3.4629194037851254e-05,
      "loss": 1.3095,
      "num_input_tokens_seen": 34017904,
      "step": 58645
    },
    {
      "epoch": 8.735478105451296,
      "grad_norm": 6.75,
      "learning_rate": 3.462619523907103e-05,
      "loss": 0.8776,
      "num_input_tokens_seen": 34021072,
      "step": 58650
    },
    {
      "epoch": 8.736222817992255,
      "grad_norm": 5.03125,
      "learning_rate": 3.462319627766602e-05,
      "loss": 1.1831,
      "num_input_tokens_seen": 34023888,
      "step": 58655
    },
    {
      "epoch": 8.736967530533214,
      "grad_norm": 13.3125,
      "learning_rate": 3.462019715368689e-05,
      "loss": 1.0303,
      "num_input_tokens_seen": 34026928,
      "step": 58660
    },
    {
      "epoch": 8.737712243074174,
      "grad_norm": 9.4375,
      "learning_rate": 3.461719786718431e-05,
      "loss": 1.2215,
      "num_input_tokens_seen": 34029744,
      "step": 58665
    },
    {
      "epoch": 8.738456955615133,
      "grad_norm": 6.71875,
      "learning_rate": 3.461419841820895e-05,
      "loss": 0.8154,
      "num_input_tokens_seen": 34032656,
      "step": 58670
    },
    {
      "epoch": 8.739201668156092,
      "grad_norm": 9.0,
      "learning_rate": 3.461119880681147e-05,
      "loss": 1.0856,
      "num_input_tokens_seen": 34035600,
      "step": 58675
    },
    {
      "epoch": 8.73994638069705,
      "grad_norm": 6.84375,
      "learning_rate": 3.460819903304256e-05,
      "loss": 1.0128,
      "num_input_tokens_seen": 34038288,
      "step": 58680
    },
    {
      "epoch": 8.74069109323801,
      "grad_norm": 8.625,
      "learning_rate": 3.460519909695289e-05,
      "loss": 1.4596,
      "num_input_tokens_seen": 34040880,
      "step": 58685
    },
    {
      "epoch": 8.74143580577897,
      "grad_norm": 14.0625,
      "learning_rate": 3.460219899859314e-05,
      "loss": 1.2033,
      "num_input_tokens_seen": 34044080,
      "step": 58690
    },
    {
      "epoch": 8.742180518319929,
      "grad_norm": 7.75,
      "learning_rate": 3.459919873801401e-05,
      "loss": 1.0335,
      "num_input_tokens_seen": 34047024,
      "step": 58695
    },
    {
      "epoch": 8.742925230860887,
      "grad_norm": 16.875,
      "learning_rate": 3.4596198315266165e-05,
      "loss": 1.0799,
      "num_input_tokens_seen": 34049968,
      "step": 58700
    },
    {
      "epoch": 8.743669943401846,
      "grad_norm": 10.5,
      "learning_rate": 3.45931977304003e-05,
      "loss": 1.1512,
      "num_input_tokens_seen": 34053104,
      "step": 58705
    },
    {
      "epoch": 8.744414655942807,
      "grad_norm": 11.8125,
      "learning_rate": 3.4590196983467114e-05,
      "loss": 1.2075,
      "num_input_tokens_seen": 34056112,
      "step": 58710
    },
    {
      "epoch": 8.745159368483765,
      "grad_norm": 8.25,
      "learning_rate": 3.45871960745173e-05,
      "loss": 0.7914,
      "num_input_tokens_seen": 34058960,
      "step": 58715
    },
    {
      "epoch": 8.745904081024724,
      "grad_norm": 10.4375,
      "learning_rate": 3.458419500360154e-05,
      "loss": 1.1266,
      "num_input_tokens_seen": 34061872,
      "step": 58720
    },
    {
      "epoch": 8.746648793565683,
      "grad_norm": 16.625,
      "learning_rate": 3.458119377077056e-05,
      "loss": 0.9804,
      "num_input_tokens_seen": 34064720,
      "step": 58725
    },
    {
      "epoch": 8.747393506106643,
      "grad_norm": 10.625,
      "learning_rate": 3.4578192376075044e-05,
      "loss": 1.0451,
      "num_input_tokens_seen": 34067408,
      "step": 58730
    },
    {
      "epoch": 8.748138218647602,
      "grad_norm": 9.0,
      "learning_rate": 3.45751908195657e-05,
      "loss": 1.0625,
      "num_input_tokens_seen": 34070544,
      "step": 58735
    },
    {
      "epoch": 8.748882931188561,
      "grad_norm": 12.125,
      "learning_rate": 3.457218910129324e-05,
      "loss": 0.8895,
      "num_input_tokens_seen": 34073424,
      "step": 58740
    },
    {
      "epoch": 8.74962764372952,
      "grad_norm": 9.0625,
      "learning_rate": 3.4569187221308376e-05,
      "loss": 1.1817,
      "num_input_tokens_seen": 34076080,
      "step": 58745
    },
    {
      "epoch": 8.75037235627048,
      "grad_norm": 13.5625,
      "learning_rate": 3.456618517966183e-05,
      "loss": 1.0133,
      "num_input_tokens_seen": 34078864,
      "step": 58750
    },
    {
      "epoch": 8.751117068811439,
      "grad_norm": 9.0,
      "learning_rate": 3.4563182976404286e-05,
      "loss": 1.2124,
      "num_input_tokens_seen": 34082000,
      "step": 58755
    },
    {
      "epoch": 8.751861781352398,
      "grad_norm": 23.625,
      "learning_rate": 3.456018061158649e-05,
      "loss": 1.187,
      "num_input_tokens_seen": 34085200,
      "step": 58760
    },
    {
      "epoch": 8.752606493893357,
      "grad_norm": 21.75,
      "learning_rate": 3.455717808525917e-05,
      "loss": 1.1851,
      "num_input_tokens_seen": 34088016,
      "step": 58765
    },
    {
      "epoch": 8.753351206434317,
      "grad_norm": 19.5,
      "learning_rate": 3.4554175397473036e-05,
      "loss": 1.1354,
      "num_input_tokens_seen": 34090864,
      "step": 58770
    },
    {
      "epoch": 8.754095918975276,
      "grad_norm": 4.53125,
      "learning_rate": 3.455117254827882e-05,
      "loss": 1.0431,
      "num_input_tokens_seen": 34094128,
      "step": 58775
    },
    {
      "epoch": 8.754840631516235,
      "grad_norm": 10.75,
      "learning_rate": 3.454816953772724e-05,
      "loss": 1.0056,
      "num_input_tokens_seen": 34097232,
      "step": 58780
    },
    {
      "epoch": 8.755585344057193,
      "grad_norm": 12.1875,
      "learning_rate": 3.4545166365869054e-05,
      "loss": 1.4096,
      "num_input_tokens_seen": 34100016,
      "step": 58785
    },
    {
      "epoch": 8.756330056598154,
      "grad_norm": 10.25,
      "learning_rate": 3.454216303275498e-05,
      "loss": 1.3155,
      "num_input_tokens_seen": 34102864,
      "step": 58790
    },
    {
      "epoch": 8.757074769139113,
      "grad_norm": 10.75,
      "learning_rate": 3.4539159538435755e-05,
      "loss": 1.1277,
      "num_input_tokens_seen": 34105744,
      "step": 58795
    },
    {
      "epoch": 8.757819481680071,
      "grad_norm": 18.625,
      "learning_rate": 3.453615588296213e-05,
      "loss": 1.4187,
      "num_input_tokens_seen": 34108368,
      "step": 58800
    },
    {
      "epoch": 8.75856419422103,
      "grad_norm": 9.375,
      "learning_rate": 3.4533152066384844e-05,
      "loss": 1.0644,
      "num_input_tokens_seen": 34111280,
      "step": 58805
    },
    {
      "epoch": 8.75930890676199,
      "grad_norm": 18.625,
      "learning_rate": 3.453014808875464e-05,
      "loss": 0.9936,
      "num_input_tokens_seen": 34114096,
      "step": 58810
    },
    {
      "epoch": 8.76005361930295,
      "grad_norm": 7.5625,
      "learning_rate": 3.4527143950122266e-05,
      "loss": 1.0936,
      "num_input_tokens_seen": 34116976,
      "step": 58815
    },
    {
      "epoch": 8.760798331843908,
      "grad_norm": 11.125,
      "learning_rate": 3.4524139650538485e-05,
      "loss": 1.459,
      "num_input_tokens_seen": 34120016,
      "step": 58820
    },
    {
      "epoch": 8.761543044384867,
      "grad_norm": 14.5625,
      "learning_rate": 3.452113519005404e-05,
      "loss": 1.1744,
      "num_input_tokens_seen": 34122832,
      "step": 58825
    },
    {
      "epoch": 8.762287756925826,
      "grad_norm": 6.4375,
      "learning_rate": 3.45181305687197e-05,
      "loss": 1.168,
      "num_input_tokens_seen": 34125712,
      "step": 58830
    },
    {
      "epoch": 8.763032469466786,
      "grad_norm": 9.25,
      "learning_rate": 3.451512578658621e-05,
      "loss": 1.2655,
      "num_input_tokens_seen": 34128656,
      "step": 58835
    },
    {
      "epoch": 8.763777182007745,
      "grad_norm": 8.5,
      "learning_rate": 3.4512120843704344e-05,
      "loss": 1.4905,
      "num_input_tokens_seen": 34131632,
      "step": 58840
    },
    {
      "epoch": 8.764521894548704,
      "grad_norm": 10.3125,
      "learning_rate": 3.4509115740124866e-05,
      "loss": 0.9163,
      "num_input_tokens_seen": 34134704,
      "step": 58845
    },
    {
      "epoch": 8.765266607089664,
      "grad_norm": 6.59375,
      "learning_rate": 3.4506110475898535e-05,
      "loss": 1.0874,
      "num_input_tokens_seen": 34137808,
      "step": 58850
    },
    {
      "epoch": 8.766011319630623,
      "grad_norm": 10.4375,
      "learning_rate": 3.4503105051076126e-05,
      "loss": 1.2562,
      "num_input_tokens_seen": 34140560,
      "step": 58855
    },
    {
      "epoch": 8.766756032171582,
      "grad_norm": 10.75,
      "learning_rate": 3.450009946570843e-05,
      "loss": 1.4534,
      "num_input_tokens_seen": 34143376,
      "step": 58860
    },
    {
      "epoch": 8.76750074471254,
      "grad_norm": 9.3125,
      "learning_rate": 3.44970937198462e-05,
      "loss": 0.8971,
      "num_input_tokens_seen": 34145936,
      "step": 58865
    },
    {
      "epoch": 8.7682454572535,
      "grad_norm": 10.375,
      "learning_rate": 3.449408781354023e-05,
      "loss": 1.0235,
      "num_input_tokens_seen": 34149232,
      "step": 58870
    },
    {
      "epoch": 8.76899016979446,
      "grad_norm": 7.1875,
      "learning_rate": 3.449108174684129e-05,
      "loss": 0.8655,
      "num_input_tokens_seen": 34152176,
      "step": 58875
    },
    {
      "epoch": 8.769734882335419,
      "grad_norm": 13.125,
      "learning_rate": 3.448807551980017e-05,
      "loss": 1.1292,
      "num_input_tokens_seen": 34155120,
      "step": 58880
    },
    {
      "epoch": 8.770479594876377,
      "grad_norm": 8.25,
      "learning_rate": 3.448506913246766e-05,
      "loss": 1.0913,
      "num_input_tokens_seen": 34158128,
      "step": 58885
    },
    {
      "epoch": 8.771224307417336,
      "grad_norm": 12.3125,
      "learning_rate": 3.448206258489455e-05,
      "loss": 0.9564,
      "num_input_tokens_seen": 34161168,
      "step": 58890
    },
    {
      "epoch": 8.771969019958297,
      "grad_norm": 11.0625,
      "learning_rate": 3.4479055877131616e-05,
      "loss": 1.4746,
      "num_input_tokens_seen": 34164592,
      "step": 58895
    },
    {
      "epoch": 8.772713732499255,
      "grad_norm": 22.625,
      "learning_rate": 3.4476049009229685e-05,
      "loss": 1.1426,
      "num_input_tokens_seen": 34167504,
      "step": 58900
    },
    {
      "epoch": 8.773458445040214,
      "grad_norm": 14.0625,
      "learning_rate": 3.447304198123953e-05,
      "loss": 1.3341,
      "num_input_tokens_seen": 34170320,
      "step": 58905
    },
    {
      "epoch": 8.774203157581173,
      "grad_norm": 8.0625,
      "learning_rate": 3.447003479321196e-05,
      "loss": 1.0819,
      "num_input_tokens_seen": 34173488,
      "step": 58910
    },
    {
      "epoch": 8.774947870122134,
      "grad_norm": 13.625,
      "learning_rate": 3.4467027445197774e-05,
      "loss": 1.3183,
      "num_input_tokens_seen": 34176304,
      "step": 58915
    },
    {
      "epoch": 8.775692582663092,
      "grad_norm": 10.5625,
      "learning_rate": 3.446401993724778e-05,
      "loss": 1.1183,
      "num_input_tokens_seen": 34178992,
      "step": 58920
    },
    {
      "epoch": 8.776437295204051,
      "grad_norm": 17.375,
      "learning_rate": 3.446101226941279e-05,
      "loss": 1.3742,
      "num_input_tokens_seen": 34181776,
      "step": 58925
    },
    {
      "epoch": 8.77718200774501,
      "grad_norm": 10.4375,
      "learning_rate": 3.4458004441743605e-05,
      "loss": 1.104,
      "num_input_tokens_seen": 34184624,
      "step": 58930
    },
    {
      "epoch": 8.77792672028597,
      "grad_norm": 12.25,
      "learning_rate": 3.445499645429107e-05,
      "loss": 0.8926,
      "num_input_tokens_seen": 34187472,
      "step": 58935
    },
    {
      "epoch": 8.778671432826929,
      "grad_norm": 15.9375,
      "learning_rate": 3.445198830710596e-05,
      "loss": 0.8149,
      "num_input_tokens_seen": 34190160,
      "step": 58940
    },
    {
      "epoch": 8.779416145367888,
      "grad_norm": 10.625,
      "learning_rate": 3.4448980000239114e-05,
      "loss": 0.8242,
      "num_input_tokens_seen": 34192944,
      "step": 58945
    },
    {
      "epoch": 8.780160857908847,
      "grad_norm": 18.125,
      "learning_rate": 3.444597153374136e-05,
      "loss": 1.1975,
      "num_input_tokens_seen": 34195952,
      "step": 58950
    },
    {
      "epoch": 8.780905570449807,
      "grad_norm": 8.0,
      "learning_rate": 3.444296290766352e-05,
      "loss": 1.1483,
      "num_input_tokens_seen": 34198896,
      "step": 58955
    },
    {
      "epoch": 8.781650282990766,
      "grad_norm": 37.75,
      "learning_rate": 3.443995412205642e-05,
      "loss": 1.2055,
      "num_input_tokens_seen": 34201744,
      "step": 58960
    },
    {
      "epoch": 8.782394995531725,
      "grad_norm": 10.0625,
      "learning_rate": 3.443694517697089e-05,
      "loss": 1.3068,
      "num_input_tokens_seen": 34204720,
      "step": 58965
    },
    {
      "epoch": 8.783139708072683,
      "grad_norm": 8.4375,
      "learning_rate": 3.4433936072457754e-05,
      "loss": 1.1436,
      "num_input_tokens_seen": 34207760,
      "step": 58970
    },
    {
      "epoch": 8.783884420613644,
      "grad_norm": 11.3125,
      "learning_rate": 3.443092680856787e-05,
      "loss": 1.0284,
      "num_input_tokens_seen": 34210800,
      "step": 58975
    },
    {
      "epoch": 8.784629133154603,
      "grad_norm": 14.9375,
      "learning_rate": 3.442791738535205e-05,
      "loss": 1.0594,
      "num_input_tokens_seen": 34213712,
      "step": 58980
    },
    {
      "epoch": 8.785373845695561,
      "grad_norm": 15.3125,
      "learning_rate": 3.4424907802861143e-05,
      "loss": 1.0738,
      "num_input_tokens_seen": 34216464,
      "step": 58985
    },
    {
      "epoch": 8.78611855823652,
      "grad_norm": 10.3125,
      "learning_rate": 3.4421898061146005e-05,
      "loss": 0.9881,
      "num_input_tokens_seen": 34219536,
      "step": 58990
    },
    {
      "epoch": 8.78686327077748,
      "grad_norm": 6.90625,
      "learning_rate": 3.4418888160257486e-05,
      "loss": 1.0897,
      "num_input_tokens_seen": 34222416,
      "step": 58995
    },
    {
      "epoch": 8.78760798331844,
      "grad_norm": 3.5625,
      "learning_rate": 3.441587810024642e-05,
      "loss": 0.6201,
      "num_input_tokens_seen": 34225296,
      "step": 59000
    },
    {
      "epoch": 8.788352695859398,
      "grad_norm": 7.15625,
      "learning_rate": 3.441286788116365e-05,
      "loss": 1.0727,
      "num_input_tokens_seen": 34228176,
      "step": 59005
    },
    {
      "epoch": 8.789097408400357,
      "grad_norm": 5.875,
      "learning_rate": 3.440985750306006e-05,
      "loss": 0.9611,
      "num_input_tokens_seen": 34231056,
      "step": 59010
    },
    {
      "epoch": 8.789842120941316,
      "grad_norm": 9.4375,
      "learning_rate": 3.4406846965986476e-05,
      "loss": 1.3021,
      "num_input_tokens_seen": 34234064,
      "step": 59015
    },
    {
      "epoch": 8.790586833482276,
      "grad_norm": 7.5625,
      "learning_rate": 3.440383626999378e-05,
      "loss": 0.9612,
      "num_input_tokens_seen": 34236752,
      "step": 59020
    },
    {
      "epoch": 8.791331546023235,
      "grad_norm": 6.28125,
      "learning_rate": 3.440082541513283e-05,
      "loss": 0.8587,
      "num_input_tokens_seen": 34239504,
      "step": 59025
    },
    {
      "epoch": 8.792076258564194,
      "grad_norm": 6.1875,
      "learning_rate": 3.439781440145449e-05,
      "loss": 1.1786,
      "num_input_tokens_seen": 34242480,
      "step": 59030
    },
    {
      "epoch": 8.792820971105153,
      "grad_norm": 11.1875,
      "learning_rate": 3.4394803229009634e-05,
      "loss": 0.9799,
      "num_input_tokens_seen": 34245712,
      "step": 59035
    },
    {
      "epoch": 8.793565683646113,
      "grad_norm": 8.25,
      "learning_rate": 3.439179189784911e-05,
      "loss": 0.869,
      "num_input_tokens_seen": 34248912,
      "step": 59040
    },
    {
      "epoch": 8.794310396187072,
      "grad_norm": 20.0,
      "learning_rate": 3.438878040802381e-05,
      "loss": 1.1258,
      "num_input_tokens_seen": 34251760,
      "step": 59045
    },
    {
      "epoch": 8.79505510872803,
      "grad_norm": 10.3125,
      "learning_rate": 3.438576875958461e-05,
      "loss": 0.8225,
      "num_input_tokens_seen": 34254544,
      "step": 59050
    },
    {
      "epoch": 8.79579982126899,
      "grad_norm": 15.125,
      "learning_rate": 3.438275695258239e-05,
      "loss": 1.3354,
      "num_input_tokens_seen": 34257456,
      "step": 59055
    },
    {
      "epoch": 8.79654453380995,
      "grad_norm": 6.15625,
      "learning_rate": 3.4379744987068025e-05,
      "loss": 0.9954,
      "num_input_tokens_seen": 34260304,
      "step": 59060
    },
    {
      "epoch": 8.797289246350909,
      "grad_norm": 6.84375,
      "learning_rate": 3.43767328630924e-05,
      "loss": 1.3372,
      "num_input_tokens_seen": 34262992,
      "step": 59065
    },
    {
      "epoch": 8.798033958891867,
      "grad_norm": 10.875,
      "learning_rate": 3.437372058070641e-05,
      "loss": 0.9511,
      "num_input_tokens_seen": 34265776,
      "step": 59070
    },
    {
      "epoch": 8.798778671432826,
      "grad_norm": 12.4375,
      "learning_rate": 3.4370708139960934e-05,
      "loss": 1.2014,
      "num_input_tokens_seen": 34268784,
      "step": 59075
    },
    {
      "epoch": 8.799523383973787,
      "grad_norm": 12.8125,
      "learning_rate": 3.4367695540906864e-05,
      "loss": 0.9936,
      "num_input_tokens_seen": 34271888,
      "step": 59080
    },
    {
      "epoch": 8.800268096514746,
      "grad_norm": 22.625,
      "learning_rate": 3.43646827835951e-05,
      "loss": 1.2845,
      "num_input_tokens_seen": 34275120,
      "step": 59085
    },
    {
      "epoch": 8.801012809055704,
      "grad_norm": 7.5625,
      "learning_rate": 3.436166986807654e-05,
      "loss": 1.03,
      "num_input_tokens_seen": 34278000,
      "step": 59090
    },
    {
      "epoch": 8.801757521596663,
      "grad_norm": 6.25,
      "learning_rate": 3.435865679440208e-05,
      "loss": 1.0883,
      "num_input_tokens_seen": 34280624,
      "step": 59095
    },
    {
      "epoch": 8.802502234137624,
      "grad_norm": 13.0,
      "learning_rate": 3.435564356262263e-05,
      "loss": 0.9943,
      "num_input_tokens_seen": 34283696,
      "step": 59100
    },
    {
      "epoch": 8.803246946678582,
      "grad_norm": 8.0,
      "learning_rate": 3.435263017278909e-05,
      "loss": 0.8489,
      "num_input_tokens_seen": 34286832,
      "step": 59105
    },
    {
      "epoch": 8.803991659219541,
      "grad_norm": 11.25,
      "learning_rate": 3.4349616624952365e-05,
      "loss": 0.9228,
      "num_input_tokens_seen": 34289776,
      "step": 59110
    },
    {
      "epoch": 8.8047363717605,
      "grad_norm": 9.0625,
      "learning_rate": 3.434660291916337e-05,
      "loss": 1.1366,
      "num_input_tokens_seen": 34292432,
      "step": 59115
    },
    {
      "epoch": 8.80548108430146,
      "grad_norm": 5.875,
      "learning_rate": 3.4343589055473025e-05,
      "loss": 1.2158,
      "num_input_tokens_seen": 34295632,
      "step": 59120
    },
    {
      "epoch": 8.80622579684242,
      "grad_norm": 10.5625,
      "learning_rate": 3.4340575033932234e-05,
      "loss": 0.8659,
      "num_input_tokens_seen": 34298512,
      "step": 59125
    },
    {
      "epoch": 8.806970509383378,
      "grad_norm": 8.875,
      "learning_rate": 3.433756085459192e-05,
      "loss": 0.9488,
      "num_input_tokens_seen": 34301360,
      "step": 59130
    },
    {
      "epoch": 8.807715221924337,
      "grad_norm": 9.8125,
      "learning_rate": 3.4334546517503006e-05,
      "loss": 1.1192,
      "num_input_tokens_seen": 34303984,
      "step": 59135
    },
    {
      "epoch": 8.808459934465297,
      "grad_norm": 14.25,
      "learning_rate": 3.4331532022716416e-05,
      "loss": 1.1084,
      "num_input_tokens_seen": 34306992,
      "step": 59140
    },
    {
      "epoch": 8.809204647006256,
      "grad_norm": 7.90625,
      "learning_rate": 3.432851737028308e-05,
      "loss": 1.1969,
      "num_input_tokens_seen": 34309712,
      "step": 59145
    },
    {
      "epoch": 8.809949359547215,
      "grad_norm": 12.875,
      "learning_rate": 3.432550256025391e-05,
      "loss": 0.681,
      "num_input_tokens_seen": 34312528,
      "step": 59150
    },
    {
      "epoch": 8.810694072088173,
      "grad_norm": 10.3125,
      "learning_rate": 3.4322487592679876e-05,
      "loss": 1.0804,
      "num_input_tokens_seen": 34315056,
      "step": 59155
    },
    {
      "epoch": 8.811438784629132,
      "grad_norm": 9.0,
      "learning_rate": 3.4319472467611876e-05,
      "loss": 1.2911,
      "num_input_tokens_seen": 34317968,
      "step": 59160
    },
    {
      "epoch": 8.812183497170093,
      "grad_norm": 12.3125,
      "learning_rate": 3.431645718510086e-05,
      "loss": 0.786,
      "num_input_tokens_seen": 34320944,
      "step": 59165
    },
    {
      "epoch": 8.812928209711052,
      "grad_norm": 7.1875,
      "learning_rate": 3.431344174519777e-05,
      "loss": 1.0531,
      "num_input_tokens_seen": 34323824,
      "step": 59170
    },
    {
      "epoch": 8.81367292225201,
      "grad_norm": 8.5625,
      "learning_rate": 3.431042614795354e-05,
      "loss": 1.0319,
      "num_input_tokens_seen": 34326768,
      "step": 59175
    },
    {
      "epoch": 8.81441763479297,
      "grad_norm": 11.0,
      "learning_rate": 3.430741039341914e-05,
      "loss": 1.4142,
      "num_input_tokens_seen": 34329360,
      "step": 59180
    },
    {
      "epoch": 8.81516234733393,
      "grad_norm": 5.15625,
      "learning_rate": 3.4304394481645485e-05,
      "loss": 1.2839,
      "num_input_tokens_seen": 34332592,
      "step": 59185
    },
    {
      "epoch": 8.815907059874888,
      "grad_norm": 12.0625,
      "learning_rate": 3.430137841268355e-05,
      "loss": 0.6477,
      "num_input_tokens_seen": 34335408,
      "step": 59190
    },
    {
      "epoch": 8.816651772415847,
      "grad_norm": 9.6875,
      "learning_rate": 3.4298362186584275e-05,
      "loss": 1.1844,
      "num_input_tokens_seen": 34338320,
      "step": 59195
    },
    {
      "epoch": 8.817396484956806,
      "grad_norm": 9.9375,
      "learning_rate": 3.4295345803398634e-05,
      "loss": 0.928,
      "num_input_tokens_seen": 34341392,
      "step": 59200
    },
    {
      "epoch": 8.818141197497766,
      "grad_norm": 9.9375,
      "learning_rate": 3.429232926317756e-05,
      "loss": 0.9732,
      "num_input_tokens_seen": 34344080,
      "step": 59205
    },
    {
      "epoch": 8.818885910038725,
      "grad_norm": 10.0,
      "learning_rate": 3.428931256597203e-05,
      "loss": 0.7922,
      "num_input_tokens_seen": 34346864,
      "step": 59210
    },
    {
      "epoch": 8.819630622579684,
      "grad_norm": 9.0,
      "learning_rate": 3.428629571183301e-05,
      "loss": 1.2637,
      "num_input_tokens_seen": 34349712,
      "step": 59215
    },
    {
      "epoch": 8.820375335120643,
      "grad_norm": 7.46875,
      "learning_rate": 3.428327870081145e-05,
      "loss": 0.9359,
      "num_input_tokens_seen": 34352784,
      "step": 59220
    },
    {
      "epoch": 8.821120047661603,
      "grad_norm": 9.625,
      "learning_rate": 3.428026153295834e-05,
      "loss": 1.1151,
      "num_input_tokens_seen": 34355600,
      "step": 59225
    },
    {
      "epoch": 8.821864760202562,
      "grad_norm": 11.125,
      "learning_rate": 3.427724420832464e-05,
      "loss": 1.0177,
      "num_input_tokens_seen": 34358288,
      "step": 59230
    },
    {
      "epoch": 8.82260947274352,
      "grad_norm": 7.3125,
      "learning_rate": 3.427422672696135e-05,
      "loss": 0.7534,
      "num_input_tokens_seen": 34361360,
      "step": 59235
    },
    {
      "epoch": 8.82335418528448,
      "grad_norm": 9.8125,
      "learning_rate": 3.42712090889194e-05,
      "loss": 0.9887,
      "num_input_tokens_seen": 34364144,
      "step": 59240
    },
    {
      "epoch": 8.82409889782544,
      "grad_norm": 8.5,
      "learning_rate": 3.426819129424979e-05,
      "loss": 1.0981,
      "num_input_tokens_seen": 34366864,
      "step": 59245
    },
    {
      "epoch": 8.824843610366399,
      "grad_norm": 7.21875,
      "learning_rate": 3.426517334300352e-05,
      "loss": 1.1369,
      "num_input_tokens_seen": 34369808,
      "step": 59250
    },
    {
      "epoch": 8.825588322907358,
      "grad_norm": 4.65625,
      "learning_rate": 3.426215523523157e-05,
      "loss": 0.8036,
      "num_input_tokens_seen": 34372912,
      "step": 59255
    },
    {
      "epoch": 8.826333035448316,
      "grad_norm": 10.75,
      "learning_rate": 3.425913697098491e-05,
      "loss": 0.9873,
      "num_input_tokens_seen": 34376016,
      "step": 59260
    },
    {
      "epoch": 8.827077747989277,
      "grad_norm": 7.625,
      "learning_rate": 3.4256118550314556e-05,
      "loss": 1.1269,
      "num_input_tokens_seen": 34379024,
      "step": 59265
    },
    {
      "epoch": 8.827822460530236,
      "grad_norm": 8.9375,
      "learning_rate": 3.425309997327147e-05,
      "loss": 0.935,
      "num_input_tokens_seen": 34381744,
      "step": 59270
    },
    {
      "epoch": 8.828567173071194,
      "grad_norm": 13.8125,
      "learning_rate": 3.4250081239906674e-05,
      "loss": 0.9558,
      "num_input_tokens_seen": 34384496,
      "step": 59275
    },
    {
      "epoch": 8.829311885612153,
      "grad_norm": 4.375,
      "learning_rate": 3.424706235027115e-05,
      "loss": 1.0642,
      "num_input_tokens_seen": 34387344,
      "step": 59280
    },
    {
      "epoch": 8.830056598153114,
      "grad_norm": 7.96875,
      "learning_rate": 3.4244043304415907e-05,
      "loss": 1.2287,
      "num_input_tokens_seen": 34390160,
      "step": 59285
    },
    {
      "epoch": 8.830801310694072,
      "grad_norm": 4.09375,
      "learning_rate": 3.424102410239195e-05,
      "loss": 1.1269,
      "num_input_tokens_seen": 34393040,
      "step": 59290
    },
    {
      "epoch": 8.831546023235031,
      "grad_norm": 18.375,
      "learning_rate": 3.423800474425029e-05,
      "loss": 1.262,
      "num_input_tokens_seen": 34396016,
      "step": 59295
    },
    {
      "epoch": 8.83229073577599,
      "grad_norm": 6.75,
      "learning_rate": 3.4234985230041916e-05,
      "loss": 1.1373,
      "num_input_tokens_seen": 34398608,
      "step": 59300
    },
    {
      "epoch": 8.83303544831695,
      "grad_norm": 6.84375,
      "learning_rate": 3.4231965559817856e-05,
      "loss": 1.3257,
      "num_input_tokens_seen": 34401520,
      "step": 59305
    },
    {
      "epoch": 8.83378016085791,
      "grad_norm": 9.1875,
      "learning_rate": 3.4228945733629124e-05,
      "loss": 1.4747,
      "num_input_tokens_seen": 34404432,
      "step": 59310
    },
    {
      "epoch": 8.834524873398868,
      "grad_norm": 7.09375,
      "learning_rate": 3.422592575152673e-05,
      "loss": 0.9063,
      "num_input_tokens_seen": 34407312,
      "step": 59315
    },
    {
      "epoch": 8.835269585939827,
      "grad_norm": 12.1875,
      "learning_rate": 3.4222905613561706e-05,
      "loss": 1.1603,
      "num_input_tokens_seen": 34410256,
      "step": 59320
    },
    {
      "epoch": 8.836014298480787,
      "grad_norm": 6.375,
      "learning_rate": 3.421988531978506e-05,
      "loss": 1.214,
      "num_input_tokens_seen": 34412944,
      "step": 59325
    },
    {
      "epoch": 8.836759011021746,
      "grad_norm": 10.1875,
      "learning_rate": 3.421686487024782e-05,
      "loss": 1.019,
      "num_input_tokens_seen": 34415568,
      "step": 59330
    },
    {
      "epoch": 8.837503723562705,
      "grad_norm": 7.71875,
      "learning_rate": 3.4213844265001015e-05,
      "loss": 1.2068,
      "num_input_tokens_seen": 34418320,
      "step": 59335
    },
    {
      "epoch": 8.838248436103664,
      "grad_norm": 9.375,
      "learning_rate": 3.421082350409568e-05,
      "loss": 1.1315,
      "num_input_tokens_seen": 34421200,
      "step": 59340
    },
    {
      "epoch": 8.838993148644622,
      "grad_norm": 8.5625,
      "learning_rate": 3.420780258758284e-05,
      "loss": 1.0973,
      "num_input_tokens_seen": 34424240,
      "step": 59345
    },
    {
      "epoch": 8.839737861185583,
      "grad_norm": 14.3125,
      "learning_rate": 3.420478151551353e-05,
      "loss": 1.1169,
      "num_input_tokens_seen": 34427376,
      "step": 59350
    },
    {
      "epoch": 8.840482573726542,
      "grad_norm": 5.53125,
      "learning_rate": 3.42017602879388e-05,
      "loss": 0.8935,
      "num_input_tokens_seen": 34430384,
      "step": 59355
    },
    {
      "epoch": 8.8412272862675,
      "grad_norm": 17.125,
      "learning_rate": 3.419873890490968e-05,
      "loss": 1.1451,
      "num_input_tokens_seen": 34433200,
      "step": 59360
    },
    {
      "epoch": 8.84197199880846,
      "grad_norm": 10.0625,
      "learning_rate": 3.4195717366477216e-05,
      "loss": 0.9253,
      "num_input_tokens_seen": 34435760,
      "step": 59365
    },
    {
      "epoch": 8.84271671134942,
      "grad_norm": 14.5625,
      "learning_rate": 3.419269567269245e-05,
      "loss": 1.1378,
      "num_input_tokens_seen": 34438480,
      "step": 59370
    },
    {
      "epoch": 8.843461423890378,
      "grad_norm": 5.4375,
      "learning_rate": 3.418967382360643e-05,
      "loss": 0.7861,
      "num_input_tokens_seen": 34441520,
      "step": 59375
    },
    {
      "epoch": 8.844206136431337,
      "grad_norm": 10.3125,
      "learning_rate": 3.4186651819270224e-05,
      "loss": 0.9737,
      "num_input_tokens_seen": 34444432,
      "step": 59380
    },
    {
      "epoch": 8.844950848972296,
      "grad_norm": 14.375,
      "learning_rate": 3.4183629659734855e-05,
      "loss": 1.1945,
      "num_input_tokens_seen": 34447376,
      "step": 59385
    },
    {
      "epoch": 8.845695561513256,
      "grad_norm": 13.0,
      "learning_rate": 3.418060734505141e-05,
      "loss": 1.1033,
      "num_input_tokens_seen": 34450160,
      "step": 59390
    },
    {
      "epoch": 8.846440274054215,
      "grad_norm": 5.75,
      "learning_rate": 3.417758487527093e-05,
      "loss": 0.977,
      "num_input_tokens_seen": 34453168,
      "step": 59395
    },
    {
      "epoch": 8.847184986595174,
      "grad_norm": 10.3125,
      "learning_rate": 3.417456225044449e-05,
      "loss": 1.069,
      "num_input_tokens_seen": 34456272,
      "step": 59400
    },
    {
      "epoch": 8.847929699136133,
      "grad_norm": 7.75,
      "learning_rate": 3.417153947062313e-05,
      "loss": 0.928,
      "num_input_tokens_seen": 34459184,
      "step": 59405
    },
    {
      "epoch": 8.848674411677093,
      "grad_norm": 5.78125,
      "learning_rate": 3.416851653585794e-05,
      "loss": 1.1888,
      "num_input_tokens_seen": 34462000,
      "step": 59410
    },
    {
      "epoch": 8.849419124218052,
      "grad_norm": 11.1875,
      "learning_rate": 3.416549344619998e-05,
      "loss": 1.2257,
      "num_input_tokens_seen": 34465040,
      "step": 59415
    },
    {
      "epoch": 8.85016383675901,
      "grad_norm": 9.5625,
      "learning_rate": 3.416247020170032e-05,
      "loss": 0.9973,
      "num_input_tokens_seen": 34467984,
      "step": 59420
    },
    {
      "epoch": 8.85090854929997,
      "grad_norm": 8.8125,
      "learning_rate": 3.415944680241004e-05,
      "loss": 1.1516,
      "num_input_tokens_seen": 34470576,
      "step": 59425
    },
    {
      "epoch": 8.85165326184093,
      "grad_norm": 10.125,
      "learning_rate": 3.415642324838023e-05,
      "loss": 0.7644,
      "num_input_tokens_seen": 34473392,
      "step": 59430
    },
    {
      "epoch": 8.852397974381889,
      "grad_norm": 11.3125,
      "learning_rate": 3.415339953966194e-05,
      "loss": 1.0997,
      "num_input_tokens_seen": 34476336,
      "step": 59435
    },
    {
      "epoch": 8.853142686922848,
      "grad_norm": 7.0,
      "learning_rate": 3.4150375676306276e-05,
      "loss": 1.1778,
      "num_input_tokens_seen": 34479280,
      "step": 59440
    },
    {
      "epoch": 8.853887399463806,
      "grad_norm": 9.375,
      "learning_rate": 3.4147351658364304e-05,
      "loss": 0.9938,
      "num_input_tokens_seen": 34482096,
      "step": 59445
    },
    {
      "epoch": 8.854632112004767,
      "grad_norm": 10.0,
      "learning_rate": 3.4144327485887126e-05,
      "loss": 1.1525,
      "num_input_tokens_seen": 34484784,
      "step": 59450
    },
    {
      "epoch": 8.855376824545726,
      "grad_norm": 8.625,
      "learning_rate": 3.414130315892583e-05,
      "loss": 0.9936,
      "num_input_tokens_seen": 34487888,
      "step": 59455
    },
    {
      "epoch": 8.856121537086684,
      "grad_norm": 10.4375,
      "learning_rate": 3.4138278677531515e-05,
      "loss": 0.8101,
      "num_input_tokens_seen": 34490800,
      "step": 59460
    },
    {
      "epoch": 8.856866249627643,
      "grad_norm": 11.0625,
      "learning_rate": 3.413525404175527e-05,
      "loss": 1.1861,
      "num_input_tokens_seen": 34493680,
      "step": 59465
    },
    {
      "epoch": 8.857610962168604,
      "grad_norm": 14.375,
      "learning_rate": 3.413222925164818e-05,
      "loss": 1.2249,
      "num_input_tokens_seen": 34496720,
      "step": 59470
    },
    {
      "epoch": 8.858355674709562,
      "grad_norm": 9.6875,
      "learning_rate": 3.412920430726137e-05,
      "loss": 1.2406,
      "num_input_tokens_seen": 34499664,
      "step": 59475
    },
    {
      "epoch": 8.859100387250521,
      "grad_norm": 11.5,
      "learning_rate": 3.412617920864593e-05,
      "loss": 0.8043,
      "num_input_tokens_seen": 34502416,
      "step": 59480
    },
    {
      "epoch": 8.85984509979148,
      "grad_norm": 22.625,
      "learning_rate": 3.412315395585296e-05,
      "loss": 1.0252,
      "num_input_tokens_seen": 34505328,
      "step": 59485
    },
    {
      "epoch": 8.86058981233244,
      "grad_norm": 10.75,
      "learning_rate": 3.4120128548933575e-05,
      "loss": 1.0268,
      "num_input_tokens_seen": 34508176,
      "step": 59490
    },
    {
      "epoch": 8.8613345248734,
      "grad_norm": 10.3125,
      "learning_rate": 3.4117102987938895e-05,
      "loss": 1.3389,
      "num_input_tokens_seen": 34511152,
      "step": 59495
    },
    {
      "epoch": 8.862079237414358,
      "grad_norm": 11.9375,
      "learning_rate": 3.411407727292003e-05,
      "loss": 1.0744,
      "num_input_tokens_seen": 34513840,
      "step": 59500
    },
    {
      "epoch": 8.862823949955317,
      "grad_norm": 5.96875,
      "learning_rate": 3.411105140392808e-05,
      "loss": 0.8286,
      "num_input_tokens_seen": 34516848,
      "step": 59505
    },
    {
      "epoch": 8.863568662496277,
      "grad_norm": 8.375,
      "learning_rate": 3.4108025381014184e-05,
      "loss": 1.3784,
      "num_input_tokens_seen": 34519696,
      "step": 59510
    },
    {
      "epoch": 8.864313375037236,
      "grad_norm": 6.65625,
      "learning_rate": 3.4104999204229466e-05,
      "loss": 1.1203,
      "num_input_tokens_seen": 34522800,
      "step": 59515
    },
    {
      "epoch": 8.865058087578195,
      "grad_norm": 7.28125,
      "learning_rate": 3.410197287362503e-05,
      "loss": 1.0939,
      "num_input_tokens_seen": 34525680,
      "step": 59520
    },
    {
      "epoch": 8.865802800119154,
      "grad_norm": 8.5,
      "learning_rate": 3.409894638925201e-05,
      "loss": 1.0872,
      "num_input_tokens_seen": 34528592,
      "step": 59525
    },
    {
      "epoch": 8.866547512660112,
      "grad_norm": 5.21875,
      "learning_rate": 3.409591975116155e-05,
      "loss": 0.7894,
      "num_input_tokens_seen": 34531216,
      "step": 59530
    },
    {
      "epoch": 8.867292225201073,
      "grad_norm": 8.8125,
      "learning_rate": 3.409289295940476e-05,
      "loss": 0.9917,
      "num_input_tokens_seen": 34534064,
      "step": 59535
    },
    {
      "epoch": 8.868036937742032,
      "grad_norm": 13.625,
      "learning_rate": 3.408986601403278e-05,
      "loss": 1.0791,
      "num_input_tokens_seen": 34537040,
      "step": 59540
    },
    {
      "epoch": 8.86878165028299,
      "grad_norm": 5.75,
      "learning_rate": 3.4086838915096765e-05,
      "loss": 0.9689,
      "num_input_tokens_seen": 34539728,
      "step": 59545
    },
    {
      "epoch": 8.86952636282395,
      "grad_norm": 4.96875,
      "learning_rate": 3.408381166264784e-05,
      "loss": 1.0618,
      "num_input_tokens_seen": 34542384,
      "step": 59550
    },
    {
      "epoch": 8.87027107536491,
      "grad_norm": 7.78125,
      "learning_rate": 3.408078425673714e-05,
      "loss": 0.951,
      "num_input_tokens_seen": 34545296,
      "step": 59555
    },
    {
      "epoch": 8.871015787905868,
      "grad_norm": 8.125,
      "learning_rate": 3.407775669741583e-05,
      "loss": 0.9104,
      "num_input_tokens_seen": 34548208,
      "step": 59560
    },
    {
      "epoch": 8.871760500446827,
      "grad_norm": 10.9375,
      "learning_rate": 3.4074728984735043e-05,
      "loss": 1.337,
      "num_input_tokens_seen": 34551120,
      "step": 59565
    },
    {
      "epoch": 8.872505212987786,
      "grad_norm": 8.1875,
      "learning_rate": 3.407170111874593e-05,
      "loss": 1.2257,
      "num_input_tokens_seen": 34553936,
      "step": 59570
    },
    {
      "epoch": 8.873249925528746,
      "grad_norm": 9.0625,
      "learning_rate": 3.4068673099499646e-05,
      "loss": 1.1881,
      "num_input_tokens_seen": 34556592,
      "step": 59575
    },
    {
      "epoch": 8.873994638069705,
      "grad_norm": 20.125,
      "learning_rate": 3.4065644927047354e-05,
      "loss": 1.0083,
      "num_input_tokens_seen": 34559440,
      "step": 59580
    },
    {
      "epoch": 8.874739350610664,
      "grad_norm": 8.6875,
      "learning_rate": 3.40626166014402e-05,
      "loss": 0.9947,
      "num_input_tokens_seen": 34562032,
      "step": 59585
    },
    {
      "epoch": 8.875484063151623,
      "grad_norm": 9.1875,
      "learning_rate": 3.4059588122729344e-05,
      "loss": 1.023,
      "num_input_tokens_seen": 34564848,
      "step": 59590
    },
    {
      "epoch": 8.876228775692583,
      "grad_norm": 9.4375,
      "learning_rate": 3.405655949096597e-05,
      "loss": 0.9924,
      "num_input_tokens_seen": 34567696,
      "step": 59595
    },
    {
      "epoch": 8.876973488233542,
      "grad_norm": 15.0625,
      "learning_rate": 3.405353070620122e-05,
      "loss": 1.1626,
      "num_input_tokens_seen": 34570832,
      "step": 59600
    },
    {
      "epoch": 8.8777182007745,
      "grad_norm": 14.5,
      "learning_rate": 3.4050501768486266e-05,
      "loss": 0.8824,
      "num_input_tokens_seen": 34573744,
      "step": 59605
    },
    {
      "epoch": 8.87846291331546,
      "grad_norm": 11.5625,
      "learning_rate": 3.404747267787228e-05,
      "loss": 1.0222,
      "num_input_tokens_seen": 34576464,
      "step": 59610
    },
    {
      "epoch": 8.87920762585642,
      "grad_norm": 10.625,
      "learning_rate": 3.404444343441045e-05,
      "loss": 1.2846,
      "num_input_tokens_seen": 34579408,
      "step": 59615
    },
    {
      "epoch": 8.879952338397379,
      "grad_norm": 5.375,
      "learning_rate": 3.404141403815193e-05,
      "loss": 0.9845,
      "num_input_tokens_seen": 34582416,
      "step": 59620
    },
    {
      "epoch": 8.880697050938338,
      "grad_norm": 12.0,
      "learning_rate": 3.4038384489147926e-05,
      "loss": 1.3332,
      "num_input_tokens_seen": 34585200,
      "step": 59625
    },
    {
      "epoch": 8.881441763479296,
      "grad_norm": 9.8125,
      "learning_rate": 3.4035354787449584e-05,
      "loss": 1.5329,
      "num_input_tokens_seen": 34587920,
      "step": 59630
    },
    {
      "epoch": 8.882186476020257,
      "grad_norm": 8.1875,
      "learning_rate": 3.403232493310811e-05,
      "loss": 1.2191,
      "num_input_tokens_seen": 34590736,
      "step": 59635
    },
    {
      "epoch": 8.882931188561216,
      "grad_norm": 8.6875,
      "learning_rate": 3.402929492617469e-05,
      "loss": 1.1178,
      "num_input_tokens_seen": 34594096,
      "step": 59640
    },
    {
      "epoch": 8.883675901102174,
      "grad_norm": 15.9375,
      "learning_rate": 3.402626476670051e-05,
      "loss": 0.9946,
      "num_input_tokens_seen": 34596944,
      "step": 59645
    },
    {
      "epoch": 8.884420613643133,
      "grad_norm": 5.34375,
      "learning_rate": 3.4023234454736756e-05,
      "loss": 1.0821,
      "num_input_tokens_seen": 34600368,
      "step": 59650
    },
    {
      "epoch": 8.885165326184094,
      "grad_norm": 10.0,
      "learning_rate": 3.402020399033463e-05,
      "loss": 1.1153,
      "num_input_tokens_seen": 34603248,
      "step": 59655
    },
    {
      "epoch": 8.885910038725052,
      "grad_norm": 8.4375,
      "learning_rate": 3.401717337354533e-05,
      "loss": 0.9289,
      "num_input_tokens_seen": 34606000,
      "step": 59660
    },
    {
      "epoch": 8.886654751266011,
      "grad_norm": 5.71875,
      "learning_rate": 3.401414260442004e-05,
      "loss": 1.1436,
      "num_input_tokens_seen": 34608976,
      "step": 59665
    },
    {
      "epoch": 8.88739946380697,
      "grad_norm": 10.375,
      "learning_rate": 3.401111168300998e-05,
      "loss": 1.0001,
      "num_input_tokens_seen": 34611824,
      "step": 59670
    },
    {
      "epoch": 8.88814417634793,
      "grad_norm": 7.8125,
      "learning_rate": 3.400808060936635e-05,
      "loss": 0.8968,
      "num_input_tokens_seen": 34614704,
      "step": 59675
    },
    {
      "epoch": 8.88888888888889,
      "grad_norm": 20.0,
      "learning_rate": 3.4005049383540345e-05,
      "loss": 1.2015,
      "num_input_tokens_seen": 34617520,
      "step": 59680
    },
    {
      "epoch": 8.889633601429848,
      "grad_norm": 10.375,
      "learning_rate": 3.400201800558318e-05,
      "loss": 1.0325,
      "num_input_tokens_seen": 34620336,
      "step": 59685
    },
    {
      "epoch": 8.890378313970807,
      "grad_norm": 16.0,
      "learning_rate": 3.399898647554608e-05,
      "loss": 1.2785,
      "num_input_tokens_seen": 34623408,
      "step": 59690
    },
    {
      "epoch": 8.891123026511767,
      "grad_norm": 2.8125,
      "learning_rate": 3.399595479348024e-05,
      "loss": 1.23,
      "num_input_tokens_seen": 34626288,
      "step": 59695
    },
    {
      "epoch": 8.891867739052726,
      "grad_norm": 5.25,
      "learning_rate": 3.3992922959436894e-05,
      "loss": 0.6644,
      "num_input_tokens_seen": 34629296,
      "step": 59700
    },
    {
      "epoch": 8.892612451593685,
      "grad_norm": 15.625,
      "learning_rate": 3.3989890973467255e-05,
      "loss": 1.1222,
      "num_input_tokens_seen": 34632272,
      "step": 59705
    },
    {
      "epoch": 8.893357164134644,
      "grad_norm": 10.6875,
      "learning_rate": 3.3986858835622536e-05,
      "loss": 1.0719,
      "num_input_tokens_seen": 34634800,
      "step": 59710
    },
    {
      "epoch": 8.894101876675602,
      "grad_norm": 7.03125,
      "learning_rate": 3.398382654595398e-05,
      "loss": 0.915,
      "num_input_tokens_seen": 34637680,
      "step": 59715
    },
    {
      "epoch": 8.894846589216563,
      "grad_norm": 9.3125,
      "learning_rate": 3.39807941045128e-05,
      "loss": 1.084,
      "num_input_tokens_seen": 34640752,
      "step": 59720
    },
    {
      "epoch": 8.895591301757522,
      "grad_norm": 13.0,
      "learning_rate": 3.397776151135024e-05,
      "loss": 0.7109,
      "num_input_tokens_seen": 34643472,
      "step": 59725
    },
    {
      "epoch": 8.89633601429848,
      "grad_norm": 8.25,
      "learning_rate": 3.397472876651752e-05,
      "loss": 1.0825,
      "num_input_tokens_seen": 34646544,
      "step": 59730
    },
    {
      "epoch": 8.89708072683944,
      "grad_norm": 13.75,
      "learning_rate": 3.397169587006588e-05,
      "loss": 1.0488,
      "num_input_tokens_seen": 34649648,
      "step": 59735
    },
    {
      "epoch": 8.8978254393804,
      "grad_norm": 10.0,
      "learning_rate": 3.396866282204655e-05,
      "loss": 1.5244,
      "num_input_tokens_seen": 34652496,
      "step": 59740
    },
    {
      "epoch": 8.898570151921358,
      "grad_norm": 8.9375,
      "learning_rate": 3.3965629622510776e-05,
      "loss": 0.9863,
      "num_input_tokens_seen": 34655632,
      "step": 59745
    },
    {
      "epoch": 8.899314864462317,
      "grad_norm": 9.1875,
      "learning_rate": 3.3962596271509806e-05,
      "loss": 1.3208,
      "num_input_tokens_seen": 34658512,
      "step": 59750
    },
    {
      "epoch": 8.900059577003276,
      "grad_norm": 8.0625,
      "learning_rate": 3.395956276909488e-05,
      "loss": 1.2952,
      "num_input_tokens_seen": 34661616,
      "step": 59755
    },
    {
      "epoch": 8.900804289544237,
      "grad_norm": 8.375,
      "learning_rate": 3.395652911531725e-05,
      "loss": 0.7988,
      "num_input_tokens_seen": 34664272,
      "step": 59760
    },
    {
      "epoch": 8.901549002085195,
      "grad_norm": 12.875,
      "learning_rate": 3.395349531022817e-05,
      "loss": 1.0323,
      "num_input_tokens_seen": 34667120,
      "step": 59765
    },
    {
      "epoch": 8.902293714626154,
      "grad_norm": 11.75,
      "learning_rate": 3.395046135387888e-05,
      "loss": 1.2484,
      "num_input_tokens_seen": 34669840,
      "step": 59770
    },
    {
      "epoch": 8.903038427167113,
      "grad_norm": 16.625,
      "learning_rate": 3.394742724632064e-05,
      "loss": 1.0073,
      "num_input_tokens_seen": 34672528,
      "step": 59775
    },
    {
      "epoch": 8.903783139708073,
      "grad_norm": 9.625,
      "learning_rate": 3.3944392987604703e-05,
      "loss": 0.9936,
      "num_input_tokens_seen": 34675312,
      "step": 59780
    },
    {
      "epoch": 8.904527852249032,
      "grad_norm": 6.84375,
      "learning_rate": 3.394135857778235e-05,
      "loss": 1.1848,
      "num_input_tokens_seen": 34678288,
      "step": 59785
    },
    {
      "epoch": 8.90527256478999,
      "grad_norm": 7.65625,
      "learning_rate": 3.3938324016904825e-05,
      "loss": 0.8455,
      "num_input_tokens_seen": 34681456,
      "step": 59790
    },
    {
      "epoch": 8.90601727733095,
      "grad_norm": 9.5625,
      "learning_rate": 3.3935289305023405e-05,
      "loss": 1.1404,
      "num_input_tokens_seen": 34684304,
      "step": 59795
    },
    {
      "epoch": 8.90676198987191,
      "grad_norm": 11.125,
      "learning_rate": 3.393225444218936e-05,
      "loss": 1.1471,
      "num_input_tokens_seen": 34687056,
      "step": 59800
    },
    {
      "epoch": 8.907506702412869,
      "grad_norm": 9.5625,
      "learning_rate": 3.392921942845394e-05,
      "loss": 1.1336,
      "num_input_tokens_seen": 34689936,
      "step": 59805
    },
    {
      "epoch": 8.908251414953828,
      "grad_norm": 15.8125,
      "learning_rate": 3.392618426386843e-05,
      "loss": 1.1339,
      "num_input_tokens_seen": 34692752,
      "step": 59810
    },
    {
      "epoch": 8.908996127494786,
      "grad_norm": 7.375,
      "learning_rate": 3.3923148948484115e-05,
      "loss": 0.9004,
      "num_input_tokens_seen": 34695568,
      "step": 59815
    },
    {
      "epoch": 8.909740840035747,
      "grad_norm": 10.4375,
      "learning_rate": 3.392011348235228e-05,
      "loss": 1.1659,
      "num_input_tokens_seen": 34698960,
      "step": 59820
    },
    {
      "epoch": 8.910485552576706,
      "grad_norm": 15.75,
      "learning_rate": 3.391707786552418e-05,
      "loss": 1.1605,
      "num_input_tokens_seen": 34701808,
      "step": 59825
    },
    {
      "epoch": 8.911230265117664,
      "grad_norm": 6.65625,
      "learning_rate": 3.391404209805112e-05,
      "loss": 0.9695,
      "num_input_tokens_seen": 34704752,
      "step": 59830
    },
    {
      "epoch": 8.911974977658623,
      "grad_norm": 15.5625,
      "learning_rate": 3.3911006179984375e-05,
      "loss": 1.1835,
      "num_input_tokens_seen": 34707664,
      "step": 59835
    },
    {
      "epoch": 8.912719690199584,
      "grad_norm": 17.375,
      "learning_rate": 3.3907970111375237e-05,
      "loss": 1.1876,
      "num_input_tokens_seen": 34710704,
      "step": 59840
    },
    {
      "epoch": 8.913464402740543,
      "grad_norm": 12.375,
      "learning_rate": 3.3904933892275e-05,
      "loss": 0.9355,
      "num_input_tokens_seen": 34713616,
      "step": 59845
    },
    {
      "epoch": 8.914209115281501,
      "grad_norm": 12.6875,
      "learning_rate": 3.390189752273495e-05,
      "loss": 1.4559,
      "num_input_tokens_seen": 34716336,
      "step": 59850
    },
    {
      "epoch": 8.91495382782246,
      "grad_norm": 6.625,
      "learning_rate": 3.389886100280639e-05,
      "loss": 1.1166,
      "num_input_tokens_seen": 34719344,
      "step": 59855
    },
    {
      "epoch": 8.915698540363419,
      "grad_norm": 10.4375,
      "learning_rate": 3.389582433254062e-05,
      "loss": 0.9743,
      "num_input_tokens_seen": 34722288,
      "step": 59860
    },
    {
      "epoch": 8.91644325290438,
      "grad_norm": 9.5625,
      "learning_rate": 3.3892787511988936e-05,
      "loss": 0.7954,
      "num_input_tokens_seen": 34725392,
      "step": 59865
    },
    {
      "epoch": 8.917187965445338,
      "grad_norm": 9.125,
      "learning_rate": 3.3889750541202654e-05,
      "loss": 0.9008,
      "num_input_tokens_seen": 34728112,
      "step": 59870
    },
    {
      "epoch": 8.917932677986297,
      "grad_norm": 7.875,
      "learning_rate": 3.388671342023306e-05,
      "loss": 0.9959,
      "num_input_tokens_seen": 34731344,
      "step": 59875
    },
    {
      "epoch": 8.918677390527257,
      "grad_norm": 10.1875,
      "learning_rate": 3.388367614913149e-05,
      "loss": 0.6986,
      "num_input_tokens_seen": 34734320,
      "step": 59880
    },
    {
      "epoch": 8.919422103068216,
      "grad_norm": 11.125,
      "learning_rate": 3.388063872794923e-05,
      "loss": 1.0568,
      "num_input_tokens_seen": 34737168,
      "step": 59885
    },
    {
      "epoch": 8.920166815609175,
      "grad_norm": 12.3125,
      "learning_rate": 3.3877601156737604e-05,
      "loss": 0.7784,
      "num_input_tokens_seen": 34740208,
      "step": 59890
    },
    {
      "epoch": 8.920911528150134,
      "grad_norm": 9.4375,
      "learning_rate": 3.3874563435547934e-05,
      "loss": 1.4007,
      "num_input_tokens_seen": 34743152,
      "step": 59895
    },
    {
      "epoch": 8.921656240691092,
      "grad_norm": 20.625,
      "learning_rate": 3.387152556443153e-05,
      "loss": 1.0518,
      "num_input_tokens_seen": 34746128,
      "step": 59900
    },
    {
      "epoch": 8.922400953232053,
      "grad_norm": 6.3125,
      "learning_rate": 3.386848754343972e-05,
      "loss": 1.1403,
      "num_input_tokens_seen": 34749296,
      "step": 59905
    },
    {
      "epoch": 8.923145665773012,
      "grad_norm": 11.0625,
      "learning_rate": 3.386544937262382e-05,
      "loss": 1.0448,
      "num_input_tokens_seen": 34752176,
      "step": 59910
    },
    {
      "epoch": 8.92389037831397,
      "grad_norm": 9.0,
      "learning_rate": 3.386241105203517e-05,
      "loss": 1.264,
      "num_input_tokens_seen": 34755280,
      "step": 59915
    },
    {
      "epoch": 8.92463509085493,
      "grad_norm": 17.375,
      "learning_rate": 3.38593725817251e-05,
      "loss": 1.5905,
      "num_input_tokens_seen": 34758256,
      "step": 59920
    },
    {
      "epoch": 8.92537980339589,
      "grad_norm": 3.609375,
      "learning_rate": 3.385633396174492e-05,
      "loss": 1.0053,
      "num_input_tokens_seen": 34761104,
      "step": 59925
    },
    {
      "epoch": 8.926124515936849,
      "grad_norm": 13.9375,
      "learning_rate": 3.385329519214599e-05,
      "loss": 1.0115,
      "num_input_tokens_seen": 34764048,
      "step": 59930
    },
    {
      "epoch": 8.926869228477807,
      "grad_norm": 9.625,
      "learning_rate": 3.385025627297963e-05,
      "loss": 0.8078,
      "num_input_tokens_seen": 34766896,
      "step": 59935
    },
    {
      "epoch": 8.927613941018766,
      "grad_norm": 14.875,
      "learning_rate": 3.384721720429718e-05,
      "loss": 1.2833,
      "num_input_tokens_seen": 34769616,
      "step": 59940
    },
    {
      "epoch": 8.928358653559727,
      "grad_norm": 10.3125,
      "learning_rate": 3.384417798614999e-05,
      "loss": 1.0272,
      "num_input_tokens_seen": 34772848,
      "step": 59945
    },
    {
      "epoch": 8.929103366100685,
      "grad_norm": 10.25,
      "learning_rate": 3.3841138618589416e-05,
      "loss": 1.0638,
      "num_input_tokens_seen": 34775760,
      "step": 59950
    },
    {
      "epoch": 8.929848078641644,
      "grad_norm": 16.125,
      "learning_rate": 3.383809910166678e-05,
      "loss": 1.2638,
      "num_input_tokens_seen": 34778768,
      "step": 59955
    },
    {
      "epoch": 8.930592791182603,
      "grad_norm": 10.5625,
      "learning_rate": 3.383505943543344e-05,
      "loss": 1.1436,
      "num_input_tokens_seen": 34781872,
      "step": 59960
    },
    {
      "epoch": 8.931337503723563,
      "grad_norm": 9.0,
      "learning_rate": 3.383201961994076e-05,
      "loss": 1.1559,
      "num_input_tokens_seen": 34784624,
      "step": 59965
    },
    {
      "epoch": 8.932082216264522,
      "grad_norm": 12.25,
      "learning_rate": 3.382897965524007e-05,
      "loss": 1.03,
      "num_input_tokens_seen": 34787760,
      "step": 59970
    },
    {
      "epoch": 8.932826928805481,
      "grad_norm": 5.90625,
      "learning_rate": 3.382593954138276e-05,
      "loss": 0.9218,
      "num_input_tokens_seen": 34790512,
      "step": 59975
    },
    {
      "epoch": 8.93357164134644,
      "grad_norm": 10.25,
      "learning_rate": 3.382289927842015e-05,
      "loss": 1.2249,
      "num_input_tokens_seen": 34793232,
      "step": 59980
    },
    {
      "epoch": 8.9343163538874,
      "grad_norm": 14.0625,
      "learning_rate": 3.381985886640364e-05,
      "loss": 1.5368,
      "num_input_tokens_seen": 34796336,
      "step": 59985
    },
    {
      "epoch": 8.935061066428359,
      "grad_norm": 18.5,
      "learning_rate": 3.381681830538458e-05,
      "loss": 1.0423,
      "num_input_tokens_seen": 34799184,
      "step": 59990
    },
    {
      "epoch": 8.935805778969318,
      "grad_norm": 7.875,
      "learning_rate": 3.3813777595414324e-05,
      "loss": 0.7967,
      "num_input_tokens_seen": 34802384,
      "step": 59995
    },
    {
      "epoch": 8.936550491510276,
      "grad_norm": 14.75,
      "learning_rate": 3.3810736736544265e-05,
      "loss": 1.4022,
      "num_input_tokens_seen": 34805296,
      "step": 60000
    },
    {
      "epoch": 8.937295204051237,
      "grad_norm": 3.71875,
      "learning_rate": 3.380769572882576e-05,
      "loss": 1.0304,
      "num_input_tokens_seen": 34808048,
      "step": 60005
    },
    {
      "epoch": 8.938039916592196,
      "grad_norm": 7.65625,
      "learning_rate": 3.380465457231018e-05,
      "loss": 0.9112,
      "num_input_tokens_seen": 34810864,
      "step": 60010
    },
    {
      "epoch": 8.938784629133155,
      "grad_norm": 11.375,
      "learning_rate": 3.3801613267048916e-05,
      "loss": 1.1076,
      "num_input_tokens_seen": 34813712,
      "step": 60015
    },
    {
      "epoch": 8.939529341674113,
      "grad_norm": 9.0,
      "learning_rate": 3.379857181309334e-05,
      "loss": 0.8791,
      "num_input_tokens_seen": 34816592,
      "step": 60020
    },
    {
      "epoch": 8.940274054215074,
      "grad_norm": 12.5,
      "learning_rate": 3.379553021049484e-05,
      "loss": 0.8075,
      "num_input_tokens_seen": 34819344,
      "step": 60025
    },
    {
      "epoch": 8.941018766756033,
      "grad_norm": 15.125,
      "learning_rate": 3.379248845930479e-05,
      "loss": 0.9882,
      "num_input_tokens_seen": 34822256,
      "step": 60030
    },
    {
      "epoch": 8.941763479296991,
      "grad_norm": 20.75,
      "learning_rate": 3.378944655957458e-05,
      "loss": 1.3192,
      "num_input_tokens_seen": 34825040,
      "step": 60035
    },
    {
      "epoch": 8.94250819183795,
      "grad_norm": 19.625,
      "learning_rate": 3.3786404511355616e-05,
      "loss": 1.4968,
      "num_input_tokens_seen": 34828304,
      "step": 60040
    },
    {
      "epoch": 8.943252904378909,
      "grad_norm": 13.0,
      "learning_rate": 3.378336231469927e-05,
      "loss": 1.2829,
      "num_input_tokens_seen": 34831184,
      "step": 60045
    },
    {
      "epoch": 8.94399761691987,
      "grad_norm": 7.40625,
      "learning_rate": 3.378031996965695e-05,
      "loss": 0.8732,
      "num_input_tokens_seen": 34833840,
      "step": 60050
    },
    {
      "epoch": 8.944742329460828,
      "grad_norm": 13.4375,
      "learning_rate": 3.3777277476280036e-05,
      "loss": 1.3518,
      "num_input_tokens_seen": 34836592,
      "step": 60055
    },
    {
      "epoch": 8.945487042001787,
      "grad_norm": 16.625,
      "learning_rate": 3.3774234834619956e-05,
      "loss": 1.1064,
      "num_input_tokens_seen": 34839440,
      "step": 60060
    },
    {
      "epoch": 8.946231754542747,
      "grad_norm": 12.6875,
      "learning_rate": 3.377119204472809e-05,
      "loss": 1.1009,
      "num_input_tokens_seen": 34842064,
      "step": 60065
    },
    {
      "epoch": 8.946976467083706,
      "grad_norm": 7.625,
      "learning_rate": 3.376814910665584e-05,
      "loss": 1.038,
      "num_input_tokens_seen": 34844688,
      "step": 60070
    },
    {
      "epoch": 8.947721179624665,
      "grad_norm": 7.4375,
      "learning_rate": 3.3765106020454636e-05,
      "loss": 0.805,
      "num_input_tokens_seen": 34847696,
      "step": 60075
    },
    {
      "epoch": 8.948465892165624,
      "grad_norm": 11.9375,
      "learning_rate": 3.376206278617587e-05,
      "loss": 1.1434,
      "num_input_tokens_seen": 34851600,
      "step": 60080
    },
    {
      "epoch": 8.949210604706582,
      "grad_norm": 9.6875,
      "learning_rate": 3.375901940387096e-05,
      "loss": 1.0643,
      "num_input_tokens_seen": 34854800,
      "step": 60085
    },
    {
      "epoch": 8.949955317247543,
      "grad_norm": 6.90625,
      "learning_rate": 3.375597587359131e-05,
      "loss": 1.1534,
      "num_input_tokens_seen": 34857680,
      "step": 60090
    },
    {
      "epoch": 8.950700029788502,
      "grad_norm": 9.0625,
      "learning_rate": 3.375293219538836e-05,
      "loss": 1.1793,
      "num_input_tokens_seen": 34860624,
      "step": 60095
    },
    {
      "epoch": 8.95144474232946,
      "grad_norm": 7.8125,
      "learning_rate": 3.374988836931351e-05,
      "loss": 0.9608,
      "num_input_tokens_seen": 34863408,
      "step": 60100
    },
    {
      "epoch": 8.95218945487042,
      "grad_norm": 11.0625,
      "learning_rate": 3.374684439541819e-05,
      "loss": 1.164,
      "num_input_tokens_seen": 34866320,
      "step": 60105
    },
    {
      "epoch": 8.95293416741138,
      "grad_norm": 17.375,
      "learning_rate": 3.374380027375382e-05,
      "loss": 1.3196,
      "num_input_tokens_seen": 34869072,
      "step": 60110
    },
    {
      "epoch": 8.953678879952339,
      "grad_norm": 4.625,
      "learning_rate": 3.374075600437183e-05,
      "loss": 0.9648,
      "num_input_tokens_seen": 34872080,
      "step": 60115
    },
    {
      "epoch": 8.954423592493297,
      "grad_norm": 14.0,
      "learning_rate": 3.373771158732366e-05,
      "loss": 1.02,
      "num_input_tokens_seen": 34875120,
      "step": 60120
    },
    {
      "epoch": 8.955168305034256,
      "grad_norm": 6.59375,
      "learning_rate": 3.3734667022660724e-05,
      "loss": 1.0503,
      "num_input_tokens_seen": 34878128,
      "step": 60125
    },
    {
      "epoch": 8.955913017575217,
      "grad_norm": 12.9375,
      "learning_rate": 3.373162231043447e-05,
      "loss": 1.1548,
      "num_input_tokens_seen": 34880752,
      "step": 60130
    },
    {
      "epoch": 8.956657730116175,
      "grad_norm": 10.6875,
      "learning_rate": 3.3728577450696336e-05,
      "loss": 1.0304,
      "num_input_tokens_seen": 34883856,
      "step": 60135
    },
    {
      "epoch": 8.957402442657134,
      "grad_norm": 8.0625,
      "learning_rate": 3.372553244349775e-05,
      "loss": 1.1623,
      "num_input_tokens_seen": 34887024,
      "step": 60140
    },
    {
      "epoch": 8.958147155198093,
      "grad_norm": 6.5625,
      "learning_rate": 3.372248728889017e-05,
      "loss": 1.0976,
      "num_input_tokens_seen": 34890032,
      "step": 60145
    },
    {
      "epoch": 8.958891867739053,
      "grad_norm": 7.625,
      "learning_rate": 3.371944198692502e-05,
      "loss": 1.073,
      "num_input_tokens_seen": 34893040,
      "step": 60150
    },
    {
      "epoch": 8.959636580280012,
      "grad_norm": 13.25,
      "learning_rate": 3.3716396537653774e-05,
      "loss": 0.9472,
      "num_input_tokens_seen": 34895856,
      "step": 60155
    },
    {
      "epoch": 8.960381292820971,
      "grad_norm": 17.625,
      "learning_rate": 3.371335094112786e-05,
      "loss": 1.2515,
      "num_input_tokens_seen": 34898928,
      "step": 60160
    },
    {
      "epoch": 8.96112600536193,
      "grad_norm": 14.9375,
      "learning_rate": 3.371030519739874e-05,
      "loss": 1.1483,
      "num_input_tokens_seen": 34901808,
      "step": 60165
    },
    {
      "epoch": 8.96187071790289,
      "grad_norm": 7.4375,
      "learning_rate": 3.370725930651786e-05,
      "loss": 1.0884,
      "num_input_tokens_seen": 34904592,
      "step": 60170
    },
    {
      "epoch": 8.962615430443849,
      "grad_norm": 5.25,
      "learning_rate": 3.370421326853669e-05,
      "loss": 0.9285,
      "num_input_tokens_seen": 34907536,
      "step": 60175
    },
    {
      "epoch": 8.963360142984808,
      "grad_norm": 9.6875,
      "learning_rate": 3.370116708350668e-05,
      "loss": 1.0384,
      "num_input_tokens_seen": 34910640,
      "step": 60180
    },
    {
      "epoch": 8.964104855525767,
      "grad_norm": 4.5625,
      "learning_rate": 3.369812075147929e-05,
      "loss": 1.0089,
      "num_input_tokens_seen": 34913552,
      "step": 60185
    },
    {
      "epoch": 8.964849568066727,
      "grad_norm": 11.4375,
      "learning_rate": 3.369507427250601e-05,
      "loss": 0.7977,
      "num_input_tokens_seen": 34916464,
      "step": 60190
    },
    {
      "epoch": 8.965594280607686,
      "grad_norm": 13.8125,
      "learning_rate": 3.369202764663827e-05,
      "loss": 0.7934,
      "num_input_tokens_seen": 34919504,
      "step": 60195
    },
    {
      "epoch": 8.966338993148645,
      "grad_norm": 11.8125,
      "learning_rate": 3.368898087392756e-05,
      "loss": 1.5374,
      "num_input_tokens_seen": 34922512,
      "step": 60200
    },
    {
      "epoch": 8.967083705689603,
      "grad_norm": 8.8125,
      "learning_rate": 3.3685933954425353e-05,
      "loss": 0.9584,
      "num_input_tokens_seen": 34925296,
      "step": 60205
    },
    {
      "epoch": 8.967828418230564,
      "grad_norm": 11.6875,
      "learning_rate": 3.368288688818312e-05,
      "loss": 1.0892,
      "num_input_tokens_seen": 34927952,
      "step": 60210
    },
    {
      "epoch": 8.968573130771523,
      "grad_norm": 14.5,
      "learning_rate": 3.367983967525234e-05,
      "loss": 1.0323,
      "num_input_tokens_seen": 34930704,
      "step": 60215
    },
    {
      "epoch": 8.969317843312481,
      "grad_norm": 13.0625,
      "learning_rate": 3.367679231568448e-05,
      "loss": 1.3351,
      "num_input_tokens_seen": 34933648,
      "step": 60220
    },
    {
      "epoch": 8.97006255585344,
      "grad_norm": 17.25,
      "learning_rate": 3.367374480953104e-05,
      "loss": 0.9564,
      "num_input_tokens_seen": 34936944,
      "step": 60225
    },
    {
      "epoch": 8.970807268394399,
      "grad_norm": 11.0,
      "learning_rate": 3.367069715684349e-05,
      "loss": 1.1757,
      "num_input_tokens_seen": 34940016,
      "step": 60230
    },
    {
      "epoch": 8.97155198093536,
      "grad_norm": 5.28125,
      "learning_rate": 3.366764935767333e-05,
      "loss": 0.7037,
      "num_input_tokens_seen": 34942992,
      "step": 60235
    },
    {
      "epoch": 8.972296693476318,
      "grad_norm": 6.96875,
      "learning_rate": 3.366460141207205e-05,
      "loss": 0.8256,
      "num_input_tokens_seen": 34946032,
      "step": 60240
    },
    {
      "epoch": 8.973041406017277,
      "grad_norm": 10.3125,
      "learning_rate": 3.366155332009113e-05,
      "loss": 1.2345,
      "num_input_tokens_seen": 34949008,
      "step": 60245
    },
    {
      "epoch": 8.973786118558236,
      "grad_norm": 8.625,
      "learning_rate": 3.3658505081782064e-05,
      "loss": 1.2461,
      "num_input_tokens_seen": 34951696,
      "step": 60250
    },
    {
      "epoch": 8.974530831099196,
      "grad_norm": 10.4375,
      "learning_rate": 3.3655456697196366e-05,
      "loss": 1.0634,
      "num_input_tokens_seen": 34954672,
      "step": 60255
    },
    {
      "epoch": 8.975275543640155,
      "grad_norm": 18.5,
      "learning_rate": 3.365240816638552e-05,
      "loss": 1.1731,
      "num_input_tokens_seen": 34957520,
      "step": 60260
    },
    {
      "epoch": 8.976020256181114,
      "grad_norm": 17.0,
      "learning_rate": 3.3649359489401025e-05,
      "loss": 0.9923,
      "num_input_tokens_seen": 34960496,
      "step": 60265
    },
    {
      "epoch": 8.976764968722073,
      "grad_norm": 7.65625,
      "learning_rate": 3.3646310666294396e-05,
      "loss": 1.0113,
      "num_input_tokens_seen": 34963120,
      "step": 60270
    },
    {
      "epoch": 8.977509681263033,
      "grad_norm": 13.5,
      "learning_rate": 3.364326169711713e-05,
      "loss": 1.2785,
      "num_input_tokens_seen": 34966128,
      "step": 60275
    },
    {
      "epoch": 8.978254393803992,
      "grad_norm": 11.8125,
      "learning_rate": 3.364021258192075e-05,
      "loss": 1.2381,
      "num_input_tokens_seen": 34969008,
      "step": 60280
    },
    {
      "epoch": 8.97899910634495,
      "grad_norm": 9.25,
      "learning_rate": 3.363716332075676e-05,
      "loss": 0.9372,
      "num_input_tokens_seen": 34972304,
      "step": 60285
    },
    {
      "epoch": 8.97974381888591,
      "grad_norm": 12.625,
      "learning_rate": 3.363411391367668e-05,
      "loss": 1.1305,
      "num_input_tokens_seen": 34975536,
      "step": 60290
    },
    {
      "epoch": 8.98048853142687,
      "grad_norm": 16.375,
      "learning_rate": 3.363106436073202e-05,
      "loss": 0.9698,
      "num_input_tokens_seen": 34978416,
      "step": 60295
    },
    {
      "epoch": 8.981233243967829,
      "grad_norm": 11.5625,
      "learning_rate": 3.362801466197429e-05,
      "loss": 0.9658,
      "num_input_tokens_seen": 34981424,
      "step": 60300
    },
    {
      "epoch": 8.981977956508787,
      "grad_norm": 4.4375,
      "learning_rate": 3.362496481745502e-05,
      "loss": 0.997,
      "num_input_tokens_seen": 34984272,
      "step": 60305
    },
    {
      "epoch": 8.982722669049746,
      "grad_norm": 10.625,
      "learning_rate": 3.362191482722574e-05,
      "loss": 0.8013,
      "num_input_tokens_seen": 34987056,
      "step": 60310
    },
    {
      "epoch": 8.983467381590707,
      "grad_norm": 7.53125,
      "learning_rate": 3.361886469133798e-05,
      "loss": 1.089,
      "num_input_tokens_seen": 34989904,
      "step": 60315
    },
    {
      "epoch": 8.984212094131665,
      "grad_norm": 9.125,
      "learning_rate": 3.361581440984325e-05,
      "loss": 1.154,
      "num_input_tokens_seen": 34992720,
      "step": 60320
    },
    {
      "epoch": 8.984956806672624,
      "grad_norm": 9.5,
      "learning_rate": 3.3612763982793094e-05,
      "loss": 1.8897,
      "num_input_tokens_seen": 34995600,
      "step": 60325
    },
    {
      "epoch": 8.985701519213583,
      "grad_norm": 6.15625,
      "learning_rate": 3.360971341023905e-05,
      "loss": 0.9587,
      "num_input_tokens_seen": 34998672,
      "step": 60330
    },
    {
      "epoch": 8.986446231754543,
      "grad_norm": 7.46875,
      "learning_rate": 3.360666269223264e-05,
      "loss": 0.7435,
      "num_input_tokens_seen": 35001776,
      "step": 60335
    },
    {
      "epoch": 8.987190944295502,
      "grad_norm": 7.28125,
      "learning_rate": 3.360361182882542e-05,
      "loss": 0.9233,
      "num_input_tokens_seen": 35004720,
      "step": 60340
    },
    {
      "epoch": 8.987935656836461,
      "grad_norm": 8.5,
      "learning_rate": 3.3600560820068916e-05,
      "loss": 1.1783,
      "num_input_tokens_seen": 35007760,
      "step": 60345
    },
    {
      "epoch": 8.98868036937742,
      "grad_norm": 10.125,
      "learning_rate": 3.3597509666014684e-05,
      "loss": 0.9938,
      "num_input_tokens_seen": 35010352,
      "step": 60350
    },
    {
      "epoch": 8.98942508191838,
      "grad_norm": 6.59375,
      "learning_rate": 3.359445836671426e-05,
      "loss": 0.9239,
      "num_input_tokens_seen": 35012912,
      "step": 60355
    },
    {
      "epoch": 8.990169794459339,
      "grad_norm": 17.125,
      "learning_rate": 3.359140692221919e-05,
      "loss": 1.4711,
      "num_input_tokens_seen": 35015600,
      "step": 60360
    },
    {
      "epoch": 8.990914507000298,
      "grad_norm": 14.125,
      "learning_rate": 3.3588355332581045e-05,
      "loss": 1.1901,
      "num_input_tokens_seen": 35018768,
      "step": 60365
    },
    {
      "epoch": 8.991659219541257,
      "grad_norm": 12.1875,
      "learning_rate": 3.358530359785136e-05,
      "loss": 1.2542,
      "num_input_tokens_seen": 35021936,
      "step": 60370
    },
    {
      "epoch": 8.992403932082215,
      "grad_norm": 7.4375,
      "learning_rate": 3.35822517180817e-05,
      "loss": 0.9817,
      "num_input_tokens_seen": 35024816,
      "step": 60375
    },
    {
      "epoch": 8.993148644623176,
      "grad_norm": 18.25,
      "learning_rate": 3.357919969332361e-05,
      "loss": 0.7946,
      "num_input_tokens_seen": 35027728,
      "step": 60380
    },
    {
      "epoch": 8.993893357164135,
      "grad_norm": 10.8125,
      "learning_rate": 3.357614752362867e-05,
      "loss": 1.408,
      "num_input_tokens_seen": 35030448,
      "step": 60385
    },
    {
      "epoch": 8.994638069705093,
      "grad_norm": 7.15625,
      "learning_rate": 3.3573095209048435e-05,
      "loss": 1.4622,
      "num_input_tokens_seen": 35033552,
      "step": 60390
    },
    {
      "epoch": 8.995382782246054,
      "grad_norm": 4.84375,
      "learning_rate": 3.357004274963446e-05,
      "loss": 0.862,
      "num_input_tokens_seen": 35036240,
      "step": 60395
    },
    {
      "epoch": 8.996127494787013,
      "grad_norm": 7.5,
      "learning_rate": 3.356699014543833e-05,
      "loss": 1.2048,
      "num_input_tokens_seen": 35039504,
      "step": 60400
    },
    {
      "epoch": 8.996872207327971,
      "grad_norm": 12.0,
      "learning_rate": 3.3563937396511607e-05,
      "loss": 0.9381,
      "num_input_tokens_seen": 35042288,
      "step": 60405
    },
    {
      "epoch": 8.99761691986893,
      "grad_norm": 11.4375,
      "learning_rate": 3.3560884502905865e-05,
      "loss": 0.9381,
      "num_input_tokens_seen": 35044944,
      "step": 60410
    },
    {
      "epoch": 8.998361632409889,
      "grad_norm": 11.9375,
      "learning_rate": 3.355783146467268e-05,
      "loss": 1.3289,
      "num_input_tokens_seen": 35047824,
      "step": 60415
    },
    {
      "epoch": 8.99910634495085,
      "grad_norm": 6.59375,
      "learning_rate": 3.355477828186363e-05,
      "loss": 0.9425,
      "num_input_tokens_seen": 35050704,
      "step": 60420
    },
    {
      "epoch": 8.999851057491808,
      "grad_norm": 8.125,
      "learning_rate": 3.3551724954530303e-05,
      "loss": 1.2063,
      "num_input_tokens_seen": 35053648,
      "step": 60425
    },
    {
      "epoch": 9.0,
      "eval_loss": 1.08272123336792,
      "eval_runtime": 43.8973,
      "eval_samples_per_second": 67.977,
      "eval_steps_per_second": 16.994,
      "num_input_tokens_seen": 35053760,
      "step": 60426
    },
    {
      "epoch": 9.000595770032767,
      "grad_norm": 6.125,
      "learning_rate": 3.3548671482724267e-05,
      "loss": 0.9789,
      "num_input_tokens_seen": 35056000,
      "step": 60430
    },
    {
      "epoch": 9.001340482573726,
      "grad_norm": 11.125,
      "learning_rate": 3.354561786649711e-05,
      "loss": 1.4242,
      "num_input_tokens_seen": 35058880,
      "step": 60435
    },
    {
      "epoch": 9.002085195114686,
      "grad_norm": 8.0625,
      "learning_rate": 3.354256410590043e-05,
      "loss": 1.4042,
      "num_input_tokens_seen": 35062048,
      "step": 60440
    },
    {
      "epoch": 9.002829907655645,
      "grad_norm": 11.4375,
      "learning_rate": 3.353951020098582e-05,
      "loss": 1.2178,
      "num_input_tokens_seen": 35065120,
      "step": 60445
    },
    {
      "epoch": 9.003574620196604,
      "grad_norm": 8.75,
      "learning_rate": 3.353645615180485e-05,
      "loss": 1.1613,
      "num_input_tokens_seen": 35067840,
      "step": 60450
    },
    {
      "epoch": 9.004319332737563,
      "grad_norm": 11.375,
      "learning_rate": 3.3533401958409136e-05,
      "loss": 1.1863,
      "num_input_tokens_seen": 35070752,
      "step": 60455
    },
    {
      "epoch": 9.005064045278523,
      "grad_norm": 13.3125,
      "learning_rate": 3.3530347620850276e-05,
      "loss": 1.2125,
      "num_input_tokens_seen": 35073728,
      "step": 60460
    },
    {
      "epoch": 9.005808757819482,
      "grad_norm": 11.125,
      "learning_rate": 3.3527293139179854e-05,
      "loss": 0.9801,
      "num_input_tokens_seen": 35076800,
      "step": 60465
    },
    {
      "epoch": 9.00655347036044,
      "grad_norm": 14.5,
      "learning_rate": 3.352423851344948e-05,
      "loss": 1.0268,
      "num_input_tokens_seen": 35079712,
      "step": 60470
    },
    {
      "epoch": 9.0072981829014,
      "grad_norm": 6.53125,
      "learning_rate": 3.352118374371076e-05,
      "loss": 0.8669,
      "num_input_tokens_seen": 35082528,
      "step": 60475
    },
    {
      "epoch": 9.00804289544236,
      "grad_norm": 11.375,
      "learning_rate": 3.351812883001531e-05,
      "loss": 1.1132,
      "num_input_tokens_seen": 35085376,
      "step": 60480
    },
    {
      "epoch": 9.008787607983319,
      "grad_norm": 12.6875,
      "learning_rate": 3.3515073772414725e-05,
      "loss": 1.0179,
      "num_input_tokens_seen": 35088064,
      "step": 60485
    },
    {
      "epoch": 9.009532320524277,
      "grad_norm": 11.0,
      "learning_rate": 3.351201857096062e-05,
      "loss": 1.0226,
      "num_input_tokens_seen": 35090688,
      "step": 60490
    },
    {
      "epoch": 9.010277033065236,
      "grad_norm": 9.0625,
      "learning_rate": 3.350896322570462e-05,
      "loss": 1.0055,
      "num_input_tokens_seen": 35093760,
      "step": 60495
    },
    {
      "epoch": 9.011021745606197,
      "grad_norm": 9.375,
      "learning_rate": 3.350590773669833e-05,
      "loss": 1.1448,
      "num_input_tokens_seen": 35096352,
      "step": 60500
    },
    {
      "epoch": 9.011766458147155,
      "grad_norm": 15.375,
      "learning_rate": 3.350285210399337e-05,
      "loss": 1.2266,
      "num_input_tokens_seen": 35099392,
      "step": 60505
    },
    {
      "epoch": 9.012511170688114,
      "grad_norm": 14.75,
      "learning_rate": 3.3499796327641366e-05,
      "loss": 1.0554,
      "num_input_tokens_seen": 35102144,
      "step": 60510
    },
    {
      "epoch": 9.013255883229073,
      "grad_norm": 12.4375,
      "learning_rate": 3.349674040769394e-05,
      "loss": 1.3008,
      "num_input_tokens_seen": 35104992,
      "step": 60515
    },
    {
      "epoch": 9.014000595770034,
      "grad_norm": 10.6875,
      "learning_rate": 3.349368434420274e-05,
      "loss": 0.9178,
      "num_input_tokens_seen": 35108064,
      "step": 60520
    },
    {
      "epoch": 9.014745308310992,
      "grad_norm": 14.9375,
      "learning_rate": 3.349062813721936e-05,
      "loss": 1.0181,
      "num_input_tokens_seen": 35110944,
      "step": 60525
    },
    {
      "epoch": 9.015490020851951,
      "grad_norm": 12.25,
      "learning_rate": 3.348757178679545e-05,
      "loss": 1.0823,
      "num_input_tokens_seen": 35113824,
      "step": 60530
    },
    {
      "epoch": 9.01623473339291,
      "grad_norm": 14.8125,
      "learning_rate": 3.3484515292982634e-05,
      "loss": 1.1579,
      "num_input_tokens_seen": 35116640,
      "step": 60535
    },
    {
      "epoch": 9.01697944593387,
      "grad_norm": 8.875,
      "learning_rate": 3.348145865583256e-05,
      "loss": 1.1331,
      "num_input_tokens_seen": 35119616,
      "step": 60540
    },
    {
      "epoch": 9.017724158474829,
      "grad_norm": 14.125,
      "learning_rate": 3.347840187539686e-05,
      "loss": 1.2003,
      "num_input_tokens_seen": 35122272,
      "step": 60545
    },
    {
      "epoch": 9.018468871015788,
      "grad_norm": 12.1875,
      "learning_rate": 3.347534495172718e-05,
      "loss": 1.3415,
      "num_input_tokens_seen": 35125056,
      "step": 60550
    },
    {
      "epoch": 9.019213583556747,
      "grad_norm": 7.6875,
      "learning_rate": 3.3472287884875167e-05,
      "loss": 1.2188,
      "num_input_tokens_seen": 35128064,
      "step": 60555
    },
    {
      "epoch": 9.019958296097707,
      "grad_norm": 9.625,
      "learning_rate": 3.346923067489245e-05,
      "loss": 1.0185,
      "num_input_tokens_seen": 35131136,
      "step": 60560
    },
    {
      "epoch": 9.020703008638666,
      "grad_norm": 8.625,
      "learning_rate": 3.3466173321830705e-05,
      "loss": 0.8746,
      "num_input_tokens_seen": 35133952,
      "step": 60565
    },
    {
      "epoch": 9.021447721179625,
      "grad_norm": 8.625,
      "learning_rate": 3.346311582574155e-05,
      "loss": 1.1253,
      "num_input_tokens_seen": 35136992,
      "step": 60570
    },
    {
      "epoch": 9.022192433720583,
      "grad_norm": 7.1875,
      "learning_rate": 3.3460058186676656e-05,
      "loss": 0.8167,
      "num_input_tokens_seen": 35139712,
      "step": 60575
    },
    {
      "epoch": 9.022937146261542,
      "grad_norm": 7.9375,
      "learning_rate": 3.345700040468768e-05,
      "loss": 0.9935,
      "num_input_tokens_seen": 35142656,
      "step": 60580
    },
    {
      "epoch": 9.023681858802503,
      "grad_norm": 8.8125,
      "learning_rate": 3.345394247982628e-05,
      "loss": 1.026,
      "num_input_tokens_seen": 35145536,
      "step": 60585
    },
    {
      "epoch": 9.024426571343461,
      "grad_norm": 7.53125,
      "learning_rate": 3.345088441214411e-05,
      "loss": 1.1358,
      "num_input_tokens_seen": 35148416,
      "step": 60590
    },
    {
      "epoch": 9.02517128388442,
      "grad_norm": 10.5625,
      "learning_rate": 3.344782620169284e-05,
      "loss": 1.1931,
      "num_input_tokens_seen": 35151168,
      "step": 60595
    },
    {
      "epoch": 9.025915996425379,
      "grad_norm": 21.125,
      "learning_rate": 3.344476784852413e-05,
      "loss": 1.2347,
      "num_input_tokens_seen": 35154016,
      "step": 60600
    },
    {
      "epoch": 9.02666070896634,
      "grad_norm": 10.375,
      "learning_rate": 3.344170935268966e-05,
      "loss": 1.0224,
      "num_input_tokens_seen": 35156992,
      "step": 60605
    },
    {
      "epoch": 9.027405421507298,
      "grad_norm": 20.0,
      "learning_rate": 3.3438650714241084e-05,
      "loss": 1.171,
      "num_input_tokens_seen": 35160160,
      "step": 60610
    },
    {
      "epoch": 9.028150134048257,
      "grad_norm": 5.71875,
      "learning_rate": 3.3435591933230074e-05,
      "loss": 0.6665,
      "num_input_tokens_seen": 35162720,
      "step": 60615
    },
    {
      "epoch": 9.028894846589216,
      "grad_norm": 14.5625,
      "learning_rate": 3.343253300970832e-05,
      "loss": 1.1201,
      "num_input_tokens_seen": 35165568,
      "step": 60620
    },
    {
      "epoch": 9.029639559130176,
      "grad_norm": 15.9375,
      "learning_rate": 3.3429473943727486e-05,
      "loss": 1.2384,
      "num_input_tokens_seen": 35168320,
      "step": 60625
    },
    {
      "epoch": 9.030384271671135,
      "grad_norm": 12.25,
      "learning_rate": 3.342641473533926e-05,
      "loss": 1.405,
      "num_input_tokens_seen": 35171200,
      "step": 60630
    },
    {
      "epoch": 9.031128984212094,
      "grad_norm": 13.8125,
      "learning_rate": 3.3423355384595316e-05,
      "loss": 1.0587,
      "num_input_tokens_seen": 35174048,
      "step": 60635
    },
    {
      "epoch": 9.031873696753053,
      "grad_norm": 7.3125,
      "learning_rate": 3.342029589154735e-05,
      "loss": 0.9932,
      "num_input_tokens_seen": 35176960,
      "step": 60640
    },
    {
      "epoch": 9.032618409294013,
      "grad_norm": 7.84375,
      "learning_rate": 3.3417236256247044e-05,
      "loss": 1.0767,
      "num_input_tokens_seen": 35179648,
      "step": 60645
    },
    {
      "epoch": 9.033363121834972,
      "grad_norm": 9.0625,
      "learning_rate": 3.341417647874608e-05,
      "loss": 1.3154,
      "num_input_tokens_seen": 35182784,
      "step": 60650
    },
    {
      "epoch": 9.03410783437593,
      "grad_norm": 9.8125,
      "learning_rate": 3.341111655909616e-05,
      "loss": 0.9162,
      "num_input_tokens_seen": 35185856,
      "step": 60655
    },
    {
      "epoch": 9.03485254691689,
      "grad_norm": 11.5,
      "learning_rate": 3.340805649734898e-05,
      "loss": 0.767,
      "num_input_tokens_seen": 35188896,
      "step": 60660
    },
    {
      "epoch": 9.03559725945785,
      "grad_norm": 10.5625,
      "learning_rate": 3.340499629355622e-05,
      "loss": 0.817,
      "num_input_tokens_seen": 35191648,
      "step": 60665
    },
    {
      "epoch": 9.036341971998809,
      "grad_norm": 3.96875,
      "learning_rate": 3.34019359477696e-05,
      "loss": 0.7258,
      "num_input_tokens_seen": 35194816,
      "step": 60670
    },
    {
      "epoch": 9.037086684539767,
      "grad_norm": 11.75,
      "learning_rate": 3.33988754600408e-05,
      "loss": 0.8279,
      "num_input_tokens_seen": 35197568,
      "step": 60675
    },
    {
      "epoch": 9.037831397080726,
      "grad_norm": 13.0,
      "learning_rate": 3.339581483042155e-05,
      "loss": 0.984,
      "num_input_tokens_seen": 35200480,
      "step": 60680
    },
    {
      "epoch": 9.038576109621687,
      "grad_norm": 10.375,
      "learning_rate": 3.339275405896353e-05,
      "loss": 1.0577,
      "num_input_tokens_seen": 35203552,
      "step": 60685
    },
    {
      "epoch": 9.039320822162646,
      "grad_norm": 16.0,
      "learning_rate": 3.338969314571847e-05,
      "loss": 1.202,
      "num_input_tokens_seen": 35206336,
      "step": 60690
    },
    {
      "epoch": 9.040065534703604,
      "grad_norm": 15.375,
      "learning_rate": 3.338663209073806e-05,
      "loss": 1.2889,
      "num_input_tokens_seen": 35209280,
      "step": 60695
    },
    {
      "epoch": 9.040810247244563,
      "grad_norm": 12.375,
      "learning_rate": 3.338357089407403e-05,
      "loss": 1.0084,
      "num_input_tokens_seen": 35212320,
      "step": 60700
    },
    {
      "epoch": 9.041554959785524,
      "grad_norm": 8.4375,
      "learning_rate": 3.338050955577809e-05,
      "loss": 0.858,
      "num_input_tokens_seen": 35215424,
      "step": 60705
    },
    {
      "epoch": 9.042299672326482,
      "grad_norm": 9.1875,
      "learning_rate": 3.337744807590196e-05,
      "loss": 0.8522,
      "num_input_tokens_seen": 35218336,
      "step": 60710
    },
    {
      "epoch": 9.043044384867441,
      "grad_norm": 12.9375,
      "learning_rate": 3.337438645449735e-05,
      "loss": 1.1856,
      "num_input_tokens_seen": 35221472,
      "step": 60715
    },
    {
      "epoch": 9.0437890974084,
      "grad_norm": 7.84375,
      "learning_rate": 3.3371324691616004e-05,
      "loss": 1.0407,
      "num_input_tokens_seen": 35224320,
      "step": 60720
    },
    {
      "epoch": 9.04453380994936,
      "grad_norm": 7.71875,
      "learning_rate": 3.3368262787309636e-05,
      "loss": 1.197,
      "num_input_tokens_seen": 35227168,
      "step": 60725
    },
    {
      "epoch": 9.04527852249032,
      "grad_norm": 6.6875,
      "learning_rate": 3.3365200741629973e-05,
      "loss": 1.0602,
      "num_input_tokens_seen": 35229856,
      "step": 60730
    },
    {
      "epoch": 9.046023235031278,
      "grad_norm": 8.875,
      "learning_rate": 3.336213855462874e-05,
      "loss": 1.4616,
      "num_input_tokens_seen": 35232896,
      "step": 60735
    },
    {
      "epoch": 9.046767947572237,
      "grad_norm": 17.875,
      "learning_rate": 3.3359076226357675e-05,
      "loss": 1.5529,
      "num_input_tokens_seen": 35235616,
      "step": 60740
    },
    {
      "epoch": 9.047512660113195,
      "grad_norm": 12.875,
      "learning_rate": 3.335601375686851e-05,
      "loss": 0.906,
      "num_input_tokens_seen": 35238240,
      "step": 60745
    },
    {
      "epoch": 9.048257372654156,
      "grad_norm": 17.5,
      "learning_rate": 3.335295114621299e-05,
      "loss": 0.8912,
      "num_input_tokens_seen": 35241056,
      "step": 60750
    },
    {
      "epoch": 9.049002085195115,
      "grad_norm": 14.0,
      "learning_rate": 3.334988839444285e-05,
      "loss": 1.0691,
      "num_input_tokens_seen": 35243808,
      "step": 60755
    },
    {
      "epoch": 9.049746797736073,
      "grad_norm": 10.875,
      "learning_rate": 3.3346825501609834e-05,
      "loss": 0.9364,
      "num_input_tokens_seen": 35246944,
      "step": 60760
    },
    {
      "epoch": 9.050491510277032,
      "grad_norm": 9.6875,
      "learning_rate": 3.3343762467765685e-05,
      "loss": 1.1221,
      "num_input_tokens_seen": 35250048,
      "step": 60765
    },
    {
      "epoch": 9.051236222817993,
      "grad_norm": 14.8125,
      "learning_rate": 3.334069929296215e-05,
      "loss": 1.2788,
      "num_input_tokens_seen": 35253376,
      "step": 60770
    },
    {
      "epoch": 9.051980935358952,
      "grad_norm": 8.6875,
      "learning_rate": 3.333763597725097e-05,
      "loss": 1.0089,
      "num_input_tokens_seen": 35256256,
      "step": 60775
    },
    {
      "epoch": 9.05272564789991,
      "grad_norm": 11.125,
      "learning_rate": 3.333457252068391e-05,
      "loss": 0.9954,
      "num_input_tokens_seen": 35259264,
      "step": 60780
    },
    {
      "epoch": 9.053470360440869,
      "grad_norm": 8.9375,
      "learning_rate": 3.333150892331271e-05,
      "loss": 0.9344,
      "num_input_tokens_seen": 35262144,
      "step": 60785
    },
    {
      "epoch": 9.05421507298183,
      "grad_norm": 15.375,
      "learning_rate": 3.3328445185189145e-05,
      "loss": 0.873,
      "num_input_tokens_seen": 35265280,
      "step": 60790
    },
    {
      "epoch": 9.054959785522788,
      "grad_norm": 9.9375,
      "learning_rate": 3.332538130636496e-05,
      "loss": 1.2747,
      "num_input_tokens_seen": 35268000,
      "step": 60795
    },
    {
      "epoch": 9.055704498063747,
      "grad_norm": 15.625,
      "learning_rate": 3.3322317286891913e-05,
      "loss": 0.984,
      "num_input_tokens_seen": 35270944,
      "step": 60800
    },
    {
      "epoch": 9.056449210604706,
      "grad_norm": 11.875,
      "learning_rate": 3.331925312682178e-05,
      "loss": 0.8766,
      "num_input_tokens_seen": 35273760,
      "step": 60805
    },
    {
      "epoch": 9.057193923145666,
      "grad_norm": 16.875,
      "learning_rate": 3.331618882620632e-05,
      "loss": 1.3657,
      "num_input_tokens_seen": 35276640,
      "step": 60810
    },
    {
      "epoch": 9.057938635686625,
      "grad_norm": 10.0,
      "learning_rate": 3.3313124385097306e-05,
      "loss": 1.0542,
      "num_input_tokens_seen": 35279488,
      "step": 60815
    },
    {
      "epoch": 9.058683348227584,
      "grad_norm": 9.25,
      "learning_rate": 3.33100598035465e-05,
      "loss": 1.2183,
      "num_input_tokens_seen": 35282016,
      "step": 60820
    },
    {
      "epoch": 9.059428060768543,
      "grad_norm": 5.25,
      "learning_rate": 3.3306995081605686e-05,
      "loss": 1.0099,
      "num_input_tokens_seen": 35285088,
      "step": 60825
    },
    {
      "epoch": 9.060172773309503,
      "grad_norm": 11.375,
      "learning_rate": 3.3303930219326625e-05,
      "loss": 0.674,
      "num_input_tokens_seen": 35288128,
      "step": 60830
    },
    {
      "epoch": 9.060917485850462,
      "grad_norm": 7.40625,
      "learning_rate": 3.33008652167611e-05,
      "loss": 1.3893,
      "num_input_tokens_seen": 35290976,
      "step": 60835
    },
    {
      "epoch": 9.06166219839142,
      "grad_norm": 17.625,
      "learning_rate": 3.32978000739609e-05,
      "loss": 1.3521,
      "num_input_tokens_seen": 35293728,
      "step": 60840
    },
    {
      "epoch": 9.06240691093238,
      "grad_norm": 17.75,
      "learning_rate": 3.32947347909778e-05,
      "loss": 0.9924,
      "num_input_tokens_seen": 35296608,
      "step": 60845
    },
    {
      "epoch": 9.06315162347334,
      "grad_norm": 9.0625,
      "learning_rate": 3.329166936786359e-05,
      "loss": 1.1201,
      "num_input_tokens_seen": 35299488,
      "step": 60850
    },
    {
      "epoch": 9.063896336014299,
      "grad_norm": 3.734375,
      "learning_rate": 3.328860380467005e-05,
      "loss": 0.7685,
      "num_input_tokens_seen": 35302624,
      "step": 60855
    },
    {
      "epoch": 9.064641048555258,
      "grad_norm": 9.25,
      "learning_rate": 3.328553810144897e-05,
      "loss": 1.04,
      "num_input_tokens_seen": 35305536,
      "step": 60860
    },
    {
      "epoch": 9.065385761096216,
      "grad_norm": 15.1875,
      "learning_rate": 3.328247225825215e-05,
      "loss": 0.9867,
      "num_input_tokens_seen": 35308224,
      "step": 60865
    },
    {
      "epoch": 9.066130473637177,
      "grad_norm": 9.0,
      "learning_rate": 3.327940627513137e-05,
      "loss": 1.2551,
      "num_input_tokens_seen": 35310976,
      "step": 60870
    },
    {
      "epoch": 9.066875186178136,
      "grad_norm": 18.0,
      "learning_rate": 3.327634015213844e-05,
      "loss": 0.8942,
      "num_input_tokens_seen": 35314016,
      "step": 60875
    },
    {
      "epoch": 9.067619898719094,
      "grad_norm": 6.21875,
      "learning_rate": 3.327327388932516e-05,
      "loss": 0.9878,
      "num_input_tokens_seen": 35317184,
      "step": 60880
    },
    {
      "epoch": 9.068364611260053,
      "grad_norm": 8.5625,
      "learning_rate": 3.327020748674333e-05,
      "loss": 1.0443,
      "num_input_tokens_seen": 35319744,
      "step": 60885
    },
    {
      "epoch": 9.069109323801014,
      "grad_norm": 12.875,
      "learning_rate": 3.326714094444474e-05,
      "loss": 0.9794,
      "num_input_tokens_seen": 35322400,
      "step": 60890
    },
    {
      "epoch": 9.069854036341972,
      "grad_norm": 6.90625,
      "learning_rate": 3.326407426248121e-05,
      "loss": 1.0227,
      "num_input_tokens_seen": 35325472,
      "step": 60895
    },
    {
      "epoch": 9.070598748882931,
      "grad_norm": 6.0625,
      "learning_rate": 3.326100744090455e-05,
      "loss": 1.1262,
      "num_input_tokens_seen": 35328352,
      "step": 60900
    },
    {
      "epoch": 9.07134346142389,
      "grad_norm": 5.0,
      "learning_rate": 3.3257940479766544e-05,
      "loss": 1.0026,
      "num_input_tokens_seen": 35331360,
      "step": 60905
    },
    {
      "epoch": 9.07208817396485,
      "grad_norm": 10.6875,
      "learning_rate": 3.3254873379119044e-05,
      "loss": 0.9243,
      "num_input_tokens_seen": 35334240,
      "step": 60910
    },
    {
      "epoch": 9.07283288650581,
      "grad_norm": 7.40625,
      "learning_rate": 3.325180613901385e-05,
      "loss": 1.0656,
      "num_input_tokens_seen": 35337216,
      "step": 60915
    },
    {
      "epoch": 9.073577599046768,
      "grad_norm": 13.5625,
      "learning_rate": 3.3248738759502775e-05,
      "loss": 1.3073,
      "num_input_tokens_seen": 35340000,
      "step": 60920
    },
    {
      "epoch": 9.074322311587727,
      "grad_norm": 19.25,
      "learning_rate": 3.3245671240637635e-05,
      "loss": 1.4767,
      "num_input_tokens_seen": 35343040,
      "step": 60925
    },
    {
      "epoch": 9.075067024128685,
      "grad_norm": 14.125,
      "learning_rate": 3.324260358247028e-05,
      "loss": 0.9409,
      "num_input_tokens_seen": 35345824,
      "step": 60930
    },
    {
      "epoch": 9.075811736669646,
      "grad_norm": 13.8125,
      "learning_rate": 3.323953578505249e-05,
      "loss": 1.2837,
      "num_input_tokens_seen": 35348416,
      "step": 60935
    },
    {
      "epoch": 9.076556449210605,
      "grad_norm": 11.1875,
      "learning_rate": 3.323646784843613e-05,
      "loss": 1.0275,
      "num_input_tokens_seen": 35351488,
      "step": 60940
    },
    {
      "epoch": 9.077301161751564,
      "grad_norm": 9.5,
      "learning_rate": 3.323339977267301e-05,
      "loss": 0.8904,
      "num_input_tokens_seen": 35354112,
      "step": 60945
    },
    {
      "epoch": 9.078045874292522,
      "grad_norm": 7.34375,
      "learning_rate": 3.3230331557814975e-05,
      "loss": 1.4163,
      "num_input_tokens_seen": 35357120,
      "step": 60950
    },
    {
      "epoch": 9.078790586833483,
      "grad_norm": 5.65625,
      "learning_rate": 3.322726320391386e-05,
      "loss": 1.0036,
      "num_input_tokens_seen": 35360000,
      "step": 60955
    },
    {
      "epoch": 9.079535299374442,
      "grad_norm": 5.0,
      "learning_rate": 3.322419471102148e-05,
      "loss": 0.9793,
      "num_input_tokens_seen": 35363008,
      "step": 60960
    },
    {
      "epoch": 9.0802800119154,
      "grad_norm": 4.03125,
      "learning_rate": 3.3221126079189704e-05,
      "loss": 0.7925,
      "num_input_tokens_seen": 35365984,
      "step": 60965
    },
    {
      "epoch": 9.081024724456359,
      "grad_norm": 10.25,
      "learning_rate": 3.321805730847035e-05,
      "loss": 0.8715,
      "num_input_tokens_seen": 35369056,
      "step": 60970
    },
    {
      "epoch": 9.08176943699732,
      "grad_norm": 14.4375,
      "learning_rate": 3.321498839891527e-05,
      "loss": 1.4011,
      "num_input_tokens_seen": 35371968,
      "step": 60975
    },
    {
      "epoch": 9.082514149538278,
      "grad_norm": 15.8125,
      "learning_rate": 3.321191935057631e-05,
      "loss": 1.0552,
      "num_input_tokens_seen": 35374816,
      "step": 60980
    },
    {
      "epoch": 9.083258862079237,
      "grad_norm": 7.28125,
      "learning_rate": 3.3208850163505314e-05,
      "loss": 1.2368,
      "num_input_tokens_seen": 35378048,
      "step": 60985
    },
    {
      "epoch": 9.084003574620196,
      "grad_norm": 20.5,
      "learning_rate": 3.3205780837754154e-05,
      "loss": 0.7682,
      "num_input_tokens_seen": 35380576,
      "step": 60990
    },
    {
      "epoch": 9.084748287161156,
      "grad_norm": 10.625,
      "learning_rate": 3.3202711373374654e-05,
      "loss": 1.2166,
      "num_input_tokens_seen": 35383488,
      "step": 60995
    },
    {
      "epoch": 9.085492999702115,
      "grad_norm": 10.25,
      "learning_rate": 3.319964177041868e-05,
      "loss": 1.1748,
      "num_input_tokens_seen": 35386528,
      "step": 61000
    },
    {
      "epoch": 9.086237712243074,
      "grad_norm": 6.53125,
      "learning_rate": 3.31965720289381e-05,
      "loss": 1.1418,
      "num_input_tokens_seen": 35389216,
      "step": 61005
    },
    {
      "epoch": 9.086982424784033,
      "grad_norm": 8.8125,
      "learning_rate": 3.319350214898476e-05,
      "loss": 1.0119,
      "num_input_tokens_seen": 35391808,
      "step": 61010
    },
    {
      "epoch": 9.087727137324993,
      "grad_norm": 6.5,
      "learning_rate": 3.319043213061053e-05,
      "loss": 1.2329,
      "num_input_tokens_seen": 35394720,
      "step": 61015
    },
    {
      "epoch": 9.088471849865952,
      "grad_norm": 7.65625,
      "learning_rate": 3.318736197386728e-05,
      "loss": 1.1023,
      "num_input_tokens_seen": 35397664,
      "step": 61020
    },
    {
      "epoch": 9.08921656240691,
      "grad_norm": 10.625,
      "learning_rate": 3.3184291678806866e-05,
      "loss": 0.9934,
      "num_input_tokens_seen": 35400448,
      "step": 61025
    },
    {
      "epoch": 9.08996127494787,
      "grad_norm": 10.9375,
      "learning_rate": 3.3181221245481164e-05,
      "loss": 0.8543,
      "num_input_tokens_seen": 35403232,
      "step": 61030
    },
    {
      "epoch": 9.09070598748883,
      "grad_norm": 7.46875,
      "learning_rate": 3.317815067394204e-05,
      "loss": 0.9555,
      "num_input_tokens_seen": 35406400,
      "step": 61035
    },
    {
      "epoch": 9.091450700029789,
      "grad_norm": 21.5,
      "learning_rate": 3.317507996424137e-05,
      "loss": 1.1454,
      "num_input_tokens_seen": 35409504,
      "step": 61040
    },
    {
      "epoch": 9.092195412570748,
      "grad_norm": 12.9375,
      "learning_rate": 3.317200911643103e-05,
      "loss": 1.0716,
      "num_input_tokens_seen": 35412608,
      "step": 61045
    },
    {
      "epoch": 9.092940125111706,
      "grad_norm": 8.125,
      "learning_rate": 3.316893813056292e-05,
      "loss": 1.42,
      "num_input_tokens_seen": 35415776,
      "step": 61050
    },
    {
      "epoch": 9.093684837652667,
      "grad_norm": 13.1875,
      "learning_rate": 3.3165867006688894e-05,
      "loss": 1.0734,
      "num_input_tokens_seen": 35418368,
      "step": 61055
    },
    {
      "epoch": 9.094429550193626,
      "grad_norm": 8.0625,
      "learning_rate": 3.3162795744860845e-05,
      "loss": 1.1457,
      "num_input_tokens_seen": 35421152,
      "step": 61060
    },
    {
      "epoch": 9.095174262734584,
      "grad_norm": 6.3125,
      "learning_rate": 3.315972434513065e-05,
      "loss": 1.2232,
      "num_input_tokens_seen": 35424224,
      "step": 61065
    },
    {
      "epoch": 9.095918975275543,
      "grad_norm": 15.625,
      "learning_rate": 3.315665280755021e-05,
      "loss": 1.0559,
      "num_input_tokens_seen": 35427040,
      "step": 61070
    },
    {
      "epoch": 9.096663687816504,
      "grad_norm": 6.5625,
      "learning_rate": 3.315358113217141e-05,
      "loss": 0.9704,
      "num_input_tokens_seen": 35429856,
      "step": 61075
    },
    {
      "epoch": 9.097408400357462,
      "grad_norm": 10.8125,
      "learning_rate": 3.315050931904614e-05,
      "loss": 1.0841,
      "num_input_tokens_seen": 35432512,
      "step": 61080
    },
    {
      "epoch": 9.098153112898421,
      "grad_norm": 14.75,
      "learning_rate": 3.314743736822631e-05,
      "loss": 1.1705,
      "num_input_tokens_seen": 35435552,
      "step": 61085
    },
    {
      "epoch": 9.09889782543938,
      "grad_norm": 21.875,
      "learning_rate": 3.314436527976381e-05,
      "loss": 1.1891,
      "num_input_tokens_seen": 35438496,
      "step": 61090
    },
    {
      "epoch": 9.099642537980339,
      "grad_norm": 8.25,
      "learning_rate": 3.314129305371052e-05,
      "loss": 0.7468,
      "num_input_tokens_seen": 35441312,
      "step": 61095
    },
    {
      "epoch": 9.1003872505213,
      "grad_norm": 9.0,
      "learning_rate": 3.313822069011837e-05,
      "loss": 1.2436,
      "num_input_tokens_seen": 35444160,
      "step": 61100
    },
    {
      "epoch": 9.101131963062258,
      "grad_norm": 13.125,
      "learning_rate": 3.313514818903924e-05,
      "loss": 1.2858,
      "num_input_tokens_seen": 35446976,
      "step": 61105
    },
    {
      "epoch": 9.101876675603217,
      "grad_norm": 7.3125,
      "learning_rate": 3.313207555052505e-05,
      "loss": 0.9874,
      "num_input_tokens_seen": 35449952,
      "step": 61110
    },
    {
      "epoch": 9.102621388144176,
      "grad_norm": 9.9375,
      "learning_rate": 3.3129002774627723e-05,
      "loss": 1.1021,
      "num_input_tokens_seen": 35452576,
      "step": 61115
    },
    {
      "epoch": 9.103366100685136,
      "grad_norm": 7.9375,
      "learning_rate": 3.3125929861399155e-05,
      "loss": 1.2425,
      "num_input_tokens_seen": 35455584,
      "step": 61120
    },
    {
      "epoch": 9.104110813226095,
      "grad_norm": 14.5,
      "learning_rate": 3.3122856810891245e-05,
      "loss": 1.0922,
      "num_input_tokens_seen": 35458464,
      "step": 61125
    },
    {
      "epoch": 9.104855525767054,
      "grad_norm": 5.34375,
      "learning_rate": 3.311978362315594e-05,
      "loss": 0.6913,
      "num_input_tokens_seen": 35461152,
      "step": 61130
    },
    {
      "epoch": 9.105600238308012,
      "grad_norm": 11.1875,
      "learning_rate": 3.3116710298245134e-05,
      "loss": 1.0536,
      "num_input_tokens_seen": 35464288,
      "step": 61135
    },
    {
      "epoch": 9.106344950848973,
      "grad_norm": 9.375,
      "learning_rate": 3.311363683621076e-05,
      "loss": 1.4179,
      "num_input_tokens_seen": 35467040,
      "step": 61140
    },
    {
      "epoch": 9.107089663389932,
      "grad_norm": 15.0,
      "learning_rate": 3.311056323710474e-05,
      "loss": 1.0511,
      "num_input_tokens_seen": 35469792,
      "step": 61145
    },
    {
      "epoch": 9.10783437593089,
      "grad_norm": 13.25,
      "learning_rate": 3.3107489500978996e-05,
      "loss": 1.2563,
      "num_input_tokens_seen": 35472768,
      "step": 61150
    },
    {
      "epoch": 9.10857908847185,
      "grad_norm": 11.1875,
      "learning_rate": 3.310441562788546e-05,
      "loss": 0.8721,
      "num_input_tokens_seen": 35475296,
      "step": 61155
    },
    {
      "epoch": 9.10932380101281,
      "grad_norm": 8.75,
      "learning_rate": 3.310134161787605e-05,
      "loss": 0.994,
      "num_input_tokens_seen": 35478048,
      "step": 61160
    },
    {
      "epoch": 9.110068513553768,
      "grad_norm": 9.0,
      "learning_rate": 3.309826747100272e-05,
      "loss": 0.9176,
      "num_input_tokens_seen": 35481184,
      "step": 61165
    },
    {
      "epoch": 9.110813226094727,
      "grad_norm": 5.53125,
      "learning_rate": 3.309519318731739e-05,
      "loss": 1.0366,
      "num_input_tokens_seen": 35484320,
      "step": 61170
    },
    {
      "epoch": 9.111557938635686,
      "grad_norm": 4.75,
      "learning_rate": 3.309211876687199e-05,
      "loss": 0.9463,
      "num_input_tokens_seen": 35487136,
      "step": 61175
    },
    {
      "epoch": 9.112302651176647,
      "grad_norm": 7.25,
      "learning_rate": 3.308904420971847e-05,
      "loss": 0.884,
      "num_input_tokens_seen": 35489664,
      "step": 61180
    },
    {
      "epoch": 9.113047363717605,
      "grad_norm": 5.84375,
      "learning_rate": 3.308596951590877e-05,
      "loss": 1.0348,
      "num_input_tokens_seen": 35492672,
      "step": 61185
    },
    {
      "epoch": 9.113792076258564,
      "grad_norm": 10.75,
      "learning_rate": 3.308289468549484e-05,
      "loss": 1.1266,
      "num_input_tokens_seen": 35495488,
      "step": 61190
    },
    {
      "epoch": 9.114536788799523,
      "grad_norm": 17.125,
      "learning_rate": 3.30798197185286e-05,
      "loss": 1.3327,
      "num_input_tokens_seen": 35498336,
      "step": 61195
    },
    {
      "epoch": 9.115281501340483,
      "grad_norm": 4.8125,
      "learning_rate": 3.307674461506204e-05,
      "loss": 0.7587,
      "num_input_tokens_seen": 35501088,
      "step": 61200
    },
    {
      "epoch": 9.116026213881442,
      "grad_norm": 10.375,
      "learning_rate": 3.3073669375147074e-05,
      "loss": 1.1246,
      "num_input_tokens_seen": 35504096,
      "step": 61205
    },
    {
      "epoch": 9.1167709264224,
      "grad_norm": 18.125,
      "learning_rate": 3.307059399883568e-05,
      "loss": 1.5169,
      "num_input_tokens_seen": 35506848,
      "step": 61210
    },
    {
      "epoch": 9.11751563896336,
      "grad_norm": 14.3125,
      "learning_rate": 3.30675184861798e-05,
      "loss": 1.3439,
      "num_input_tokens_seen": 35509696,
      "step": 61215
    },
    {
      "epoch": 9.11826035150432,
      "grad_norm": 4.125,
      "learning_rate": 3.30644428372314e-05,
      "loss": 1.1663,
      "num_input_tokens_seen": 35512704,
      "step": 61220
    },
    {
      "epoch": 9.119005064045279,
      "grad_norm": 8.1875,
      "learning_rate": 3.306136705204242e-05,
      "loss": 1.1942,
      "num_input_tokens_seen": 35515488,
      "step": 61225
    },
    {
      "epoch": 9.119749776586238,
      "grad_norm": 6.25,
      "learning_rate": 3.3058291130664844e-05,
      "loss": 0.6164,
      "num_input_tokens_seen": 35518240,
      "step": 61230
    },
    {
      "epoch": 9.120494489127196,
      "grad_norm": 8.9375,
      "learning_rate": 3.305521507315063e-05,
      "loss": 1.293,
      "num_input_tokens_seen": 35521568,
      "step": 61235
    },
    {
      "epoch": 9.121239201668157,
      "grad_norm": 6.3125,
      "learning_rate": 3.305213887955174e-05,
      "loss": 0.8968,
      "num_input_tokens_seen": 35524416,
      "step": 61240
    },
    {
      "epoch": 9.121983914209116,
      "grad_norm": 9.75,
      "learning_rate": 3.3049062549920154e-05,
      "loss": 0.8098,
      "num_input_tokens_seen": 35527168,
      "step": 61245
    },
    {
      "epoch": 9.122728626750074,
      "grad_norm": 6.28125,
      "learning_rate": 3.3045986084307835e-05,
      "loss": 0.951,
      "num_input_tokens_seen": 35530496,
      "step": 61250
    },
    {
      "epoch": 9.123473339291033,
      "grad_norm": 10.0,
      "learning_rate": 3.304290948276677e-05,
      "loss": 0.8573,
      "num_input_tokens_seen": 35533280,
      "step": 61255
    },
    {
      "epoch": 9.124218051831992,
      "grad_norm": 17.125,
      "learning_rate": 3.30398327453489e-05,
      "loss": 1.0242,
      "num_input_tokens_seen": 35536416,
      "step": 61260
    },
    {
      "epoch": 9.124962764372953,
      "grad_norm": 15.1875,
      "learning_rate": 3.303675587210624e-05,
      "loss": 1.0805,
      "num_input_tokens_seen": 35539552,
      "step": 61265
    },
    {
      "epoch": 9.125707476913911,
      "grad_norm": 7.53125,
      "learning_rate": 3.3033678863090756e-05,
      "loss": 0.8882,
      "num_input_tokens_seen": 35542432,
      "step": 61270
    },
    {
      "epoch": 9.12645218945487,
      "grad_norm": 9.1875,
      "learning_rate": 3.303060171835444e-05,
      "loss": 1.2214,
      "num_input_tokens_seen": 35545280,
      "step": 61275
    },
    {
      "epoch": 9.127196901995829,
      "grad_norm": 14.0,
      "learning_rate": 3.302752443794925e-05,
      "loss": 0.8801,
      "num_input_tokens_seen": 35548000,
      "step": 61280
    },
    {
      "epoch": 9.12794161453679,
      "grad_norm": 8.8125,
      "learning_rate": 3.302444702192722e-05,
      "loss": 1.3271,
      "num_input_tokens_seen": 35550976,
      "step": 61285
    },
    {
      "epoch": 9.128686327077748,
      "grad_norm": 10.125,
      "learning_rate": 3.30213694703403e-05,
      "loss": 1.0253,
      "num_input_tokens_seen": 35553632,
      "step": 61290
    },
    {
      "epoch": 9.129431039618707,
      "grad_norm": 18.625,
      "learning_rate": 3.3018291783240495e-05,
      "loss": 0.9078,
      "num_input_tokens_seen": 35556608,
      "step": 61295
    },
    {
      "epoch": 9.130175752159666,
      "grad_norm": 11.1875,
      "learning_rate": 3.3015213960679796e-05,
      "loss": 0.8571,
      "num_input_tokens_seen": 35559392,
      "step": 61300
    },
    {
      "epoch": 9.130920464700626,
      "grad_norm": 15.0625,
      "learning_rate": 3.301213600271021e-05,
      "loss": 1.0395,
      "num_input_tokens_seen": 35562432,
      "step": 61305
    },
    {
      "epoch": 9.131665177241585,
      "grad_norm": 6.28125,
      "learning_rate": 3.3009057909383725e-05,
      "loss": 0.9147,
      "num_input_tokens_seen": 35564960,
      "step": 61310
    },
    {
      "epoch": 9.132409889782544,
      "grad_norm": 4.3125,
      "learning_rate": 3.300597968075235e-05,
      "loss": 1.1021,
      "num_input_tokens_seen": 35568512,
      "step": 61315
    },
    {
      "epoch": 9.133154602323502,
      "grad_norm": 5.09375,
      "learning_rate": 3.3002901316868085e-05,
      "loss": 0.9798,
      "num_input_tokens_seen": 35571296,
      "step": 61320
    },
    {
      "epoch": 9.133899314864463,
      "grad_norm": 7.0625,
      "learning_rate": 3.299982281778293e-05,
      "loss": 1.0642,
      "num_input_tokens_seen": 35574208,
      "step": 61325
    },
    {
      "epoch": 9.134644027405422,
      "grad_norm": 5.84375,
      "learning_rate": 3.2996744183548905e-05,
      "loss": 1.0009,
      "num_input_tokens_seen": 35577184,
      "step": 61330
    },
    {
      "epoch": 9.13538873994638,
      "grad_norm": 20.125,
      "learning_rate": 3.2993665414218024e-05,
      "loss": 1.2029,
      "num_input_tokens_seen": 35580096,
      "step": 61335
    },
    {
      "epoch": 9.13613345248734,
      "grad_norm": 16.375,
      "learning_rate": 3.2990586509842274e-05,
      "loss": 1.1319,
      "num_input_tokens_seen": 35583168,
      "step": 61340
    },
    {
      "epoch": 9.1368781650283,
      "grad_norm": 7.5625,
      "learning_rate": 3.298750747047369e-05,
      "loss": 0.9244,
      "num_input_tokens_seen": 35586112,
      "step": 61345
    },
    {
      "epoch": 9.137622877569259,
      "grad_norm": 8.25,
      "learning_rate": 3.2984428296164296e-05,
      "loss": 0.9185,
      "num_input_tokens_seen": 35589056,
      "step": 61350
    },
    {
      "epoch": 9.138367590110217,
      "grad_norm": 9.1875,
      "learning_rate": 3.298134898696609e-05,
      "loss": 1.1582,
      "num_input_tokens_seen": 35591904,
      "step": 61355
    },
    {
      "epoch": 9.139112302651176,
      "grad_norm": 10.1875,
      "learning_rate": 3.297826954293111e-05,
      "loss": 1.011,
      "num_input_tokens_seen": 35594912,
      "step": 61360
    },
    {
      "epoch": 9.139857015192137,
      "grad_norm": 17.625,
      "learning_rate": 3.2975189964111365e-05,
      "loss": 0.8621,
      "num_input_tokens_seen": 35597696,
      "step": 61365
    },
    {
      "epoch": 9.140601727733095,
      "grad_norm": 9.625,
      "learning_rate": 3.2972110250558895e-05,
      "loss": 1.1879,
      "num_input_tokens_seen": 35600480,
      "step": 61370
    },
    {
      "epoch": 9.141346440274054,
      "grad_norm": 14.9375,
      "learning_rate": 3.296903040232573e-05,
      "loss": 0.9624,
      "num_input_tokens_seen": 35603360,
      "step": 61375
    },
    {
      "epoch": 9.142091152815013,
      "grad_norm": 13.625,
      "learning_rate": 3.29659504194639e-05,
      "loss": 1.162,
      "num_input_tokens_seen": 35606432,
      "step": 61380
    },
    {
      "epoch": 9.142835865355973,
      "grad_norm": 12.6875,
      "learning_rate": 3.296287030202543e-05,
      "loss": 0.9118,
      "num_input_tokens_seen": 35609664,
      "step": 61385
    },
    {
      "epoch": 9.143580577896932,
      "grad_norm": 8.8125,
      "learning_rate": 3.295979005006235e-05,
      "loss": 1.1864,
      "num_input_tokens_seen": 35612256,
      "step": 61390
    },
    {
      "epoch": 9.14432529043789,
      "grad_norm": 10.25,
      "learning_rate": 3.295670966362672e-05,
      "loss": 0.9836,
      "num_input_tokens_seen": 35614912,
      "step": 61395
    },
    {
      "epoch": 9.14507000297885,
      "grad_norm": 11.4375,
      "learning_rate": 3.2953629142770556e-05,
      "loss": 0.915,
      "num_input_tokens_seen": 35617728,
      "step": 61400
    },
    {
      "epoch": 9.14581471551981,
      "grad_norm": 6.75,
      "learning_rate": 3.295054848754591e-05,
      "loss": 0.9902,
      "num_input_tokens_seen": 35620480,
      "step": 61405
    },
    {
      "epoch": 9.146559428060769,
      "grad_norm": 7.46875,
      "learning_rate": 3.294746769800484e-05,
      "loss": 1.4027,
      "num_input_tokens_seen": 35623296,
      "step": 61410
    },
    {
      "epoch": 9.147304140601728,
      "grad_norm": 5.8125,
      "learning_rate": 3.2944386774199373e-05,
      "loss": 0.9108,
      "num_input_tokens_seen": 35626432,
      "step": 61415
    },
    {
      "epoch": 9.148048853142686,
      "grad_norm": 5.6875,
      "learning_rate": 3.294130571618157e-05,
      "loss": 0.9163,
      "num_input_tokens_seen": 35629120,
      "step": 61420
    },
    {
      "epoch": 9.148793565683647,
      "grad_norm": 10.0625,
      "learning_rate": 3.2938224524003483e-05,
      "loss": 0.9348,
      "num_input_tokens_seen": 35632032,
      "step": 61425
    },
    {
      "epoch": 9.149538278224606,
      "grad_norm": 5.78125,
      "learning_rate": 3.293514319771715e-05,
      "loss": 1.0986,
      "num_input_tokens_seen": 35634976,
      "step": 61430
    },
    {
      "epoch": 9.150282990765565,
      "grad_norm": 11.125,
      "learning_rate": 3.2932061737374635e-05,
      "loss": 0.9543,
      "num_input_tokens_seen": 35637632,
      "step": 61435
    },
    {
      "epoch": 9.151027703306523,
      "grad_norm": 15.875,
      "learning_rate": 3.292898014302801e-05,
      "loss": 0.9868,
      "num_input_tokens_seen": 35640480,
      "step": 61440
    },
    {
      "epoch": 9.151772415847482,
      "grad_norm": 6.65625,
      "learning_rate": 3.292589841472932e-05,
      "loss": 1.0373,
      "num_input_tokens_seen": 35643392,
      "step": 61445
    },
    {
      "epoch": 9.152517128388443,
      "grad_norm": 13.4375,
      "learning_rate": 3.292281655253063e-05,
      "loss": 0.9917,
      "num_input_tokens_seen": 35646304,
      "step": 61450
    },
    {
      "epoch": 9.153261840929401,
      "grad_norm": 7.53125,
      "learning_rate": 3.291973455648401e-05,
      "loss": 0.8433,
      "num_input_tokens_seen": 35649440,
      "step": 61455
    },
    {
      "epoch": 9.15400655347036,
      "grad_norm": 11.75,
      "learning_rate": 3.291665242664152e-05,
      "loss": 0.8972,
      "num_input_tokens_seen": 35652064,
      "step": 61460
    },
    {
      "epoch": 9.154751266011319,
      "grad_norm": 16.875,
      "learning_rate": 3.291357016305523e-05,
      "loss": 1.2517,
      "num_input_tokens_seen": 35654848,
      "step": 61465
    },
    {
      "epoch": 9.15549597855228,
      "grad_norm": 10.0625,
      "learning_rate": 3.291048776577722e-05,
      "loss": 0.9307,
      "num_input_tokens_seen": 35657664,
      "step": 61470
    },
    {
      "epoch": 9.156240691093238,
      "grad_norm": 8.625,
      "learning_rate": 3.290740523485956e-05,
      "loss": 0.8934,
      "num_input_tokens_seen": 35660864,
      "step": 61475
    },
    {
      "epoch": 9.156985403634197,
      "grad_norm": 7.3125,
      "learning_rate": 3.290432257035432e-05,
      "loss": 1.1344,
      "num_input_tokens_seen": 35663552,
      "step": 61480
    },
    {
      "epoch": 9.157730116175156,
      "grad_norm": 14.0,
      "learning_rate": 3.29012397723136e-05,
      "loss": 1.1448,
      "num_input_tokens_seen": 35666688,
      "step": 61485
    },
    {
      "epoch": 9.158474828716116,
      "grad_norm": 20.125,
      "learning_rate": 3.289815684078944e-05,
      "loss": 1.2348,
      "num_input_tokens_seen": 35669568,
      "step": 61490
    },
    {
      "epoch": 9.159219541257075,
      "grad_norm": 15.3125,
      "learning_rate": 3.2895073775833976e-05,
      "loss": 1.4971,
      "num_input_tokens_seen": 35672768,
      "step": 61495
    },
    {
      "epoch": 9.159964253798034,
      "grad_norm": 8.3125,
      "learning_rate": 3.2891990577499246e-05,
      "loss": 0.8508,
      "num_input_tokens_seen": 35675968,
      "step": 61500
    },
    {
      "epoch": 9.160708966338992,
      "grad_norm": 10.625,
      "learning_rate": 3.2888907245837356e-05,
      "loss": 1.1935,
      "num_input_tokens_seen": 35678688,
      "step": 61505
    },
    {
      "epoch": 9.161453678879953,
      "grad_norm": 7.53125,
      "learning_rate": 3.2885823780900395e-05,
      "loss": 0.9542,
      "num_input_tokens_seen": 35681536,
      "step": 61510
    },
    {
      "epoch": 9.162198391420912,
      "grad_norm": 10.6875,
      "learning_rate": 3.2882740182740466e-05,
      "loss": 1.0755,
      "num_input_tokens_seen": 35684352,
      "step": 61515
    },
    {
      "epoch": 9.16294310396187,
      "grad_norm": 9.8125,
      "learning_rate": 3.2879656451409644e-05,
      "loss": 0.9302,
      "num_input_tokens_seen": 35687488,
      "step": 61520
    },
    {
      "epoch": 9.16368781650283,
      "grad_norm": 9.25,
      "learning_rate": 3.287657258696004e-05,
      "loss": 1.1249,
      "num_input_tokens_seen": 35690304,
      "step": 61525
    },
    {
      "epoch": 9.16443252904379,
      "grad_norm": 10.875,
      "learning_rate": 3.2873488589443747e-05,
      "loss": 0.9899,
      "num_input_tokens_seen": 35693472,
      "step": 61530
    },
    {
      "epoch": 9.165177241584749,
      "grad_norm": 14.125,
      "learning_rate": 3.287040445891286e-05,
      "loss": 1.3298,
      "num_input_tokens_seen": 35696448,
      "step": 61535
    },
    {
      "epoch": 9.165921954125707,
      "grad_norm": 14.3125,
      "learning_rate": 3.28673201954195e-05,
      "loss": 1.3506,
      "num_input_tokens_seen": 35699264,
      "step": 61540
    },
    {
      "epoch": 9.166666666666666,
      "grad_norm": 10.75,
      "learning_rate": 3.286423579901575e-05,
      "loss": 1.013,
      "num_input_tokens_seen": 35702048,
      "step": 61545
    },
    {
      "epoch": 9.167411379207627,
      "grad_norm": 5.75,
      "learning_rate": 3.2861151269753745e-05,
      "loss": 0.821,
      "num_input_tokens_seen": 35704928,
      "step": 61550
    },
    {
      "epoch": 9.168156091748585,
      "grad_norm": 7.65625,
      "learning_rate": 3.285806660768556e-05,
      "loss": 0.8151,
      "num_input_tokens_seen": 35707616,
      "step": 61555
    },
    {
      "epoch": 9.168900804289544,
      "grad_norm": 6.75,
      "learning_rate": 3.285498181286334e-05,
      "loss": 1.1493,
      "num_input_tokens_seen": 35710528,
      "step": 61560
    },
    {
      "epoch": 9.169645516830503,
      "grad_norm": 14.5625,
      "learning_rate": 3.285189688533917e-05,
      "loss": 1.1029,
      "num_input_tokens_seen": 35713216,
      "step": 61565
    },
    {
      "epoch": 9.170390229371463,
      "grad_norm": 9.8125,
      "learning_rate": 3.284881182516519e-05,
      "loss": 1.0176,
      "num_input_tokens_seen": 35716160,
      "step": 61570
    },
    {
      "epoch": 9.171134941912422,
      "grad_norm": 7.34375,
      "learning_rate": 3.2845726632393525e-05,
      "loss": 1.1897,
      "num_input_tokens_seen": 35718944,
      "step": 61575
    },
    {
      "epoch": 9.171879654453381,
      "grad_norm": 10.9375,
      "learning_rate": 3.284264130707627e-05,
      "loss": 1.1577,
      "num_input_tokens_seen": 35721792,
      "step": 61580
    },
    {
      "epoch": 9.17262436699434,
      "grad_norm": 8.75,
      "learning_rate": 3.283955584926557e-05,
      "loss": 0.9196,
      "num_input_tokens_seen": 35724608,
      "step": 61585
    },
    {
      "epoch": 9.1733690795353,
      "grad_norm": 11.5625,
      "learning_rate": 3.283647025901353e-05,
      "loss": 1.0079,
      "num_input_tokens_seen": 35727712,
      "step": 61590
    },
    {
      "epoch": 9.174113792076259,
      "grad_norm": 17.125,
      "learning_rate": 3.283338453637229e-05,
      "loss": 1.3445,
      "num_input_tokens_seen": 35731072,
      "step": 61595
    },
    {
      "epoch": 9.174858504617218,
      "grad_norm": 24.5,
      "learning_rate": 3.2830298681393985e-05,
      "loss": 1.4769,
      "num_input_tokens_seen": 35733824,
      "step": 61600
    },
    {
      "epoch": 9.175603217158177,
      "grad_norm": 11.625,
      "learning_rate": 3.2827212694130736e-05,
      "loss": 0.9098,
      "num_input_tokens_seen": 35736704,
      "step": 61605
    },
    {
      "epoch": 9.176347929699135,
      "grad_norm": 10.8125,
      "learning_rate": 3.282412657463469e-05,
      "loss": 1.029,
      "num_input_tokens_seen": 35739648,
      "step": 61610
    },
    {
      "epoch": 9.177092642240096,
      "grad_norm": 8.0625,
      "learning_rate": 3.282104032295798e-05,
      "loss": 0.7909,
      "num_input_tokens_seen": 35742560,
      "step": 61615
    },
    {
      "epoch": 9.177837354781055,
      "grad_norm": 9.5,
      "learning_rate": 3.281795393915275e-05,
      "loss": 1.2299,
      "num_input_tokens_seen": 35745440,
      "step": 61620
    },
    {
      "epoch": 9.178582067322013,
      "grad_norm": 8.625,
      "learning_rate": 3.281486742327112e-05,
      "loss": 1.1614,
      "num_input_tokens_seen": 35748096,
      "step": 61625
    },
    {
      "epoch": 9.179326779862972,
      "grad_norm": 12.0,
      "learning_rate": 3.281178077536525e-05,
      "loss": 0.9995,
      "num_input_tokens_seen": 35751296,
      "step": 61630
    },
    {
      "epoch": 9.180071492403933,
      "grad_norm": 8.4375,
      "learning_rate": 3.280869399548728e-05,
      "loss": 1.1595,
      "num_input_tokens_seen": 35754336,
      "step": 61635
    },
    {
      "epoch": 9.180816204944891,
      "grad_norm": 8.1875,
      "learning_rate": 3.280560708368936e-05,
      "loss": 1.1757,
      "num_input_tokens_seen": 35757536,
      "step": 61640
    },
    {
      "epoch": 9.18156091748585,
      "grad_norm": 8.0,
      "learning_rate": 3.2802520040023646e-05,
      "loss": 0.9408,
      "num_input_tokens_seen": 35760416,
      "step": 61645
    },
    {
      "epoch": 9.182305630026809,
      "grad_norm": 4.75,
      "learning_rate": 3.279943286454229e-05,
      "loss": 0.9728,
      "num_input_tokens_seen": 35763424,
      "step": 61650
    },
    {
      "epoch": 9.18305034256777,
      "grad_norm": 7.0,
      "learning_rate": 3.2796345557297446e-05,
      "loss": 0.7522,
      "num_input_tokens_seen": 35766176,
      "step": 61655
    },
    {
      "epoch": 9.183795055108728,
      "grad_norm": 7.15625,
      "learning_rate": 3.2793258118341265e-05,
      "loss": 0.8908,
      "num_input_tokens_seen": 35768896,
      "step": 61660
    },
    {
      "epoch": 9.184539767649687,
      "grad_norm": 13.25,
      "learning_rate": 3.2790170547725894e-05,
      "loss": 1.038,
      "num_input_tokens_seen": 35772064,
      "step": 61665
    },
    {
      "epoch": 9.185284480190646,
      "grad_norm": 10.5,
      "learning_rate": 3.2787082845503525e-05,
      "loss": 1.1099,
      "num_input_tokens_seen": 35774880,
      "step": 61670
    },
    {
      "epoch": 9.186029192731606,
      "grad_norm": 5.21875,
      "learning_rate": 3.27839950117263e-05,
      "loss": 1.0452,
      "num_input_tokens_seen": 35777664,
      "step": 61675
    },
    {
      "epoch": 9.186773905272565,
      "grad_norm": 10.5625,
      "learning_rate": 3.27809070464464e-05,
      "loss": 1.2123,
      "num_input_tokens_seen": 35780608,
      "step": 61680
    },
    {
      "epoch": 9.187518617813524,
      "grad_norm": 18.375,
      "learning_rate": 3.2777818949715965e-05,
      "loss": 1.0107,
      "num_input_tokens_seen": 35783584,
      "step": 61685
    },
    {
      "epoch": 9.188263330354483,
      "grad_norm": 13.25,
      "learning_rate": 3.27747307215872e-05,
      "loss": 1.1434,
      "num_input_tokens_seen": 35786592,
      "step": 61690
    },
    {
      "epoch": 9.189008042895443,
      "grad_norm": 13.875,
      "learning_rate": 3.2771642362112255e-05,
      "loss": 1.0587,
      "num_input_tokens_seen": 35789344,
      "step": 61695
    },
    {
      "epoch": 9.189752755436402,
      "grad_norm": 7.625,
      "learning_rate": 3.276855387134331e-05,
      "loss": 1.1141,
      "num_input_tokens_seen": 35792224,
      "step": 61700
    },
    {
      "epoch": 9.19049746797736,
      "grad_norm": 7.65625,
      "learning_rate": 3.2765465249332545e-05,
      "loss": 1.2889,
      "num_input_tokens_seen": 35795360,
      "step": 61705
    },
    {
      "epoch": 9.19124218051832,
      "grad_norm": 9.5,
      "learning_rate": 3.276237649613214e-05,
      "loss": 0.9007,
      "num_input_tokens_seen": 35798848,
      "step": 61710
    },
    {
      "epoch": 9.19198689305928,
      "grad_norm": 14.6875,
      "learning_rate": 3.275928761179427e-05,
      "loss": 1.4456,
      "num_input_tokens_seen": 35801824,
      "step": 61715
    },
    {
      "epoch": 9.192731605600239,
      "grad_norm": 8.4375,
      "learning_rate": 3.2756198596371115e-05,
      "loss": 1.1549,
      "num_input_tokens_seen": 35804672,
      "step": 61720
    },
    {
      "epoch": 9.193476318141197,
      "grad_norm": 4.3125,
      "learning_rate": 3.275310944991487e-05,
      "loss": 1.1367,
      "num_input_tokens_seen": 35807680,
      "step": 61725
    },
    {
      "epoch": 9.194221030682156,
      "grad_norm": 17.5,
      "learning_rate": 3.275002017247773e-05,
      "loss": 1.475,
      "num_input_tokens_seen": 35810272,
      "step": 61730
    },
    {
      "epoch": 9.194965743223117,
      "grad_norm": 15.625,
      "learning_rate": 3.2746930764111876e-05,
      "loss": 1.2511,
      "num_input_tokens_seen": 35813056,
      "step": 61735
    },
    {
      "epoch": 9.195710455764075,
      "grad_norm": 13.875,
      "learning_rate": 3.2743841224869496e-05,
      "loss": 1.05,
      "num_input_tokens_seen": 35816160,
      "step": 61740
    },
    {
      "epoch": 9.196455168305034,
      "grad_norm": 9.5625,
      "learning_rate": 3.274075155480278e-05,
      "loss": 1.0847,
      "num_input_tokens_seen": 35819104,
      "step": 61745
    },
    {
      "epoch": 9.197199880845993,
      "grad_norm": 12.625,
      "learning_rate": 3.273766175396395e-05,
      "loss": 1.2401,
      "num_input_tokens_seen": 35822080,
      "step": 61750
    },
    {
      "epoch": 9.197944593386953,
      "grad_norm": 14.75,
      "learning_rate": 3.273457182240518e-05,
      "loss": 1.2322,
      "num_input_tokens_seen": 35825088,
      "step": 61755
    },
    {
      "epoch": 9.198689305927912,
      "grad_norm": 9.875,
      "learning_rate": 3.273148176017868e-05,
      "loss": 1.3789,
      "num_input_tokens_seen": 35828128,
      "step": 61760
    },
    {
      "epoch": 9.199434018468871,
      "grad_norm": 15.625,
      "learning_rate": 3.2728391567336656e-05,
      "loss": 1.1991,
      "num_input_tokens_seen": 35831072,
      "step": 61765
    },
    {
      "epoch": 9.20017873100983,
      "grad_norm": 16.25,
      "learning_rate": 3.272530124393131e-05,
      "loss": 1.2425,
      "num_input_tokens_seen": 35834080,
      "step": 61770
    },
    {
      "epoch": 9.200923443550789,
      "grad_norm": 8.125,
      "learning_rate": 3.2722210790014854e-05,
      "loss": 1.221,
      "num_input_tokens_seen": 35836768,
      "step": 61775
    },
    {
      "epoch": 9.201668156091749,
      "grad_norm": 11.875,
      "learning_rate": 3.271912020563949e-05,
      "loss": 1.2222,
      "num_input_tokens_seen": 35839520,
      "step": 61780
    },
    {
      "epoch": 9.202412868632708,
      "grad_norm": 8.875,
      "learning_rate": 3.2716029490857445e-05,
      "loss": 1.1967,
      "num_input_tokens_seen": 35842528,
      "step": 61785
    },
    {
      "epoch": 9.203157581173667,
      "grad_norm": 11.4375,
      "learning_rate": 3.271293864572092e-05,
      "loss": 1.076,
      "num_input_tokens_seen": 35845664,
      "step": 61790
    },
    {
      "epoch": 9.203902293714625,
      "grad_norm": 6.6875,
      "learning_rate": 3.2709847670282126e-05,
      "loss": 1.2833,
      "num_input_tokens_seen": 35848704,
      "step": 61795
    },
    {
      "epoch": 9.204647006255586,
      "grad_norm": 10.875,
      "learning_rate": 3.27067565645933e-05,
      "loss": 1.3361,
      "num_input_tokens_seen": 35851872,
      "step": 61800
    },
    {
      "epoch": 9.205391718796545,
      "grad_norm": 9.8125,
      "learning_rate": 3.2703665328706654e-05,
      "loss": 0.8541,
      "num_input_tokens_seen": 35854592,
      "step": 61805
    },
    {
      "epoch": 9.206136431337503,
      "grad_norm": 7.9375,
      "learning_rate": 3.270057396267441e-05,
      "loss": 1.2084,
      "num_input_tokens_seen": 35857440,
      "step": 61810
    },
    {
      "epoch": 9.206881143878462,
      "grad_norm": 9.8125,
      "learning_rate": 3.26974824665488e-05,
      "loss": 1.0125,
      "num_input_tokens_seen": 35860672,
      "step": 61815
    },
    {
      "epoch": 9.207625856419423,
      "grad_norm": 18.5,
      "learning_rate": 3.269439084038205e-05,
      "loss": 1.0584,
      "num_input_tokens_seen": 35863584,
      "step": 61820
    },
    {
      "epoch": 9.208370568960381,
      "grad_norm": 10.125,
      "learning_rate": 3.2691299084226375e-05,
      "loss": 0.8806,
      "num_input_tokens_seen": 35866144,
      "step": 61825
    },
    {
      "epoch": 9.20911528150134,
      "grad_norm": 12.1875,
      "learning_rate": 3.2688207198134026e-05,
      "loss": 1.0758,
      "num_input_tokens_seen": 35869248,
      "step": 61830
    },
    {
      "epoch": 9.209859994042299,
      "grad_norm": 9.1875,
      "learning_rate": 3.2685115182157225e-05,
      "loss": 0.9856,
      "num_input_tokens_seen": 35872256,
      "step": 61835
    },
    {
      "epoch": 9.21060470658326,
      "grad_norm": 6.40625,
      "learning_rate": 3.2682023036348216e-05,
      "loss": 1.2075,
      "num_input_tokens_seen": 35875296,
      "step": 61840
    },
    {
      "epoch": 9.211349419124218,
      "grad_norm": 7.0625,
      "learning_rate": 3.267893076075924e-05,
      "loss": 1.2179,
      "num_input_tokens_seen": 35878048,
      "step": 61845
    },
    {
      "epoch": 9.212094131665177,
      "grad_norm": 7.0,
      "learning_rate": 3.267583835544253e-05,
      "loss": 1.1875,
      "num_input_tokens_seen": 35881056,
      "step": 61850
    },
    {
      "epoch": 9.212838844206136,
      "grad_norm": 10.25,
      "learning_rate": 3.2672745820450336e-05,
      "loss": 1.2228,
      "num_input_tokens_seen": 35883648,
      "step": 61855
    },
    {
      "epoch": 9.213583556747096,
      "grad_norm": 10.25,
      "learning_rate": 3.2669653155834894e-05,
      "loss": 1.2547,
      "num_input_tokens_seen": 35886304,
      "step": 61860
    },
    {
      "epoch": 9.214328269288055,
      "grad_norm": 9.0625,
      "learning_rate": 3.2666560361648456e-05,
      "loss": 0.7252,
      "num_input_tokens_seen": 35889248,
      "step": 61865
    },
    {
      "epoch": 9.215072981829014,
      "grad_norm": 12.1875,
      "learning_rate": 3.266346743794328e-05,
      "loss": 1.0986,
      "num_input_tokens_seen": 35892000,
      "step": 61870
    },
    {
      "epoch": 9.215817694369973,
      "grad_norm": 15.875,
      "learning_rate": 3.26603743847716e-05,
      "loss": 1.2464,
      "num_input_tokens_seen": 35894880,
      "step": 61875
    },
    {
      "epoch": 9.216562406910933,
      "grad_norm": 6.15625,
      "learning_rate": 3.26572812021857e-05,
      "loss": 0.9304,
      "num_input_tokens_seen": 35897792,
      "step": 61880
    },
    {
      "epoch": 9.217307119451892,
      "grad_norm": 8.25,
      "learning_rate": 3.2654187890237795e-05,
      "loss": 0.9851,
      "num_input_tokens_seen": 35900480,
      "step": 61885
    },
    {
      "epoch": 9.21805183199285,
      "grad_norm": 15.125,
      "learning_rate": 3.2651094448980175e-05,
      "loss": 1.4314,
      "num_input_tokens_seen": 35903232,
      "step": 61890
    },
    {
      "epoch": 9.21879654453381,
      "grad_norm": 10.6875,
      "learning_rate": 3.264800087846509e-05,
      "loss": 0.9285,
      "num_input_tokens_seen": 35906144,
      "step": 61895
    },
    {
      "epoch": 9.21954125707477,
      "grad_norm": 9.875,
      "learning_rate": 3.2644907178744805e-05,
      "loss": 0.7612,
      "num_input_tokens_seen": 35909184,
      "step": 61900
    },
    {
      "epoch": 9.220285969615729,
      "grad_norm": 9.0,
      "learning_rate": 3.264181334987157e-05,
      "loss": 1.167,
      "num_input_tokens_seen": 35911840,
      "step": 61905
    },
    {
      "epoch": 9.221030682156687,
      "grad_norm": 10.8125,
      "learning_rate": 3.2638719391897684e-05,
      "loss": 1.1082,
      "num_input_tokens_seen": 35914592,
      "step": 61910
    },
    {
      "epoch": 9.221775394697646,
      "grad_norm": 8.875,
      "learning_rate": 3.2635625304875386e-05,
      "loss": 1.4443,
      "num_input_tokens_seen": 35917376,
      "step": 61915
    },
    {
      "epoch": 9.222520107238607,
      "grad_norm": 8.9375,
      "learning_rate": 3.263253108885696e-05,
      "loss": 1.1931,
      "num_input_tokens_seen": 35920320,
      "step": 61920
    },
    {
      "epoch": 9.223264819779565,
      "grad_norm": 17.75,
      "learning_rate": 3.262943674389469e-05,
      "loss": 0.8712,
      "num_input_tokens_seen": 35923040,
      "step": 61925
    },
    {
      "epoch": 9.224009532320524,
      "grad_norm": 12.9375,
      "learning_rate": 3.2626342270040823e-05,
      "loss": 1.2817,
      "num_input_tokens_seen": 35926304,
      "step": 61930
    },
    {
      "epoch": 9.224754244861483,
      "grad_norm": 13.4375,
      "learning_rate": 3.262324766734766e-05,
      "loss": 1.3088,
      "num_input_tokens_seen": 35928960,
      "step": 61935
    },
    {
      "epoch": 9.225498957402444,
      "grad_norm": 17.125,
      "learning_rate": 3.2620152935867484e-05,
      "loss": 1.0836,
      "num_input_tokens_seen": 35931904,
      "step": 61940
    },
    {
      "epoch": 9.226243669943402,
      "grad_norm": 15.125,
      "learning_rate": 3.261705807565256e-05,
      "loss": 1.149,
      "num_input_tokens_seen": 35934848,
      "step": 61945
    },
    {
      "epoch": 9.226988382484361,
      "grad_norm": 18.375,
      "learning_rate": 3.26139630867552e-05,
      "loss": 1.5867,
      "num_input_tokens_seen": 35937376,
      "step": 61950
    },
    {
      "epoch": 9.22773309502532,
      "grad_norm": 10.5,
      "learning_rate": 3.261086796922765e-05,
      "loss": 1.204,
      "num_input_tokens_seen": 35940160,
      "step": 61955
    },
    {
      "epoch": 9.228477807566279,
      "grad_norm": 12.125,
      "learning_rate": 3.260777272312222e-05,
      "loss": 1.069,
      "num_input_tokens_seen": 35942976,
      "step": 61960
    },
    {
      "epoch": 9.229222520107239,
      "grad_norm": 11.125,
      "learning_rate": 3.2604677348491215e-05,
      "loss": 1.1425,
      "num_input_tokens_seen": 35945888,
      "step": 61965
    },
    {
      "epoch": 9.229967232648198,
      "grad_norm": 8.0625,
      "learning_rate": 3.260158184538691e-05,
      "loss": 1.2068,
      "num_input_tokens_seen": 35948864,
      "step": 61970
    },
    {
      "epoch": 9.230711945189157,
      "grad_norm": 18.75,
      "learning_rate": 3.25984862138616e-05,
      "loss": 1.007,
      "num_input_tokens_seen": 35951680,
      "step": 61975
    },
    {
      "epoch": 9.231456657730115,
      "grad_norm": 11.4375,
      "learning_rate": 3.25953904539676e-05,
      "loss": 1.2567,
      "num_input_tokens_seen": 35954336,
      "step": 61980
    },
    {
      "epoch": 9.232201370271076,
      "grad_norm": 9.5,
      "learning_rate": 3.259229456575719e-05,
      "loss": 1.1724,
      "num_input_tokens_seen": 35956864,
      "step": 61985
    },
    {
      "epoch": 9.232946082812035,
      "grad_norm": 4.90625,
      "learning_rate": 3.258919854928268e-05,
      "loss": 0.9762,
      "num_input_tokens_seen": 35959616,
      "step": 61990
    },
    {
      "epoch": 9.233690795352993,
      "grad_norm": 9.5625,
      "learning_rate": 3.2586102404596375e-05,
      "loss": 0.9041,
      "num_input_tokens_seen": 35962368,
      "step": 61995
    },
    {
      "epoch": 9.234435507893952,
      "grad_norm": 11.125,
      "learning_rate": 3.258300613175058e-05,
      "loss": 1.5666,
      "num_input_tokens_seen": 35965344,
      "step": 62000
    },
    {
      "epoch": 9.235180220434913,
      "grad_norm": 12.1875,
      "learning_rate": 3.2579909730797605e-05,
      "loss": 1.1293,
      "num_input_tokens_seen": 35968224,
      "step": 62005
    },
    {
      "epoch": 9.235924932975871,
      "grad_norm": 8.6875,
      "learning_rate": 3.2576813201789755e-05,
      "loss": 0.8516,
      "num_input_tokens_seen": 35971200,
      "step": 62010
    },
    {
      "epoch": 9.23666964551683,
      "grad_norm": 6.5625,
      "learning_rate": 3.257371654477935e-05,
      "loss": 1.2269,
      "num_input_tokens_seen": 35974144,
      "step": 62015
    },
    {
      "epoch": 9.237414358057789,
      "grad_norm": 7.15625,
      "learning_rate": 3.257061975981871e-05,
      "loss": 0.7886,
      "num_input_tokens_seen": 35977056,
      "step": 62020
    },
    {
      "epoch": 9.23815907059875,
      "grad_norm": 14.4375,
      "learning_rate": 3.256752284696013e-05,
      "loss": 1.3456,
      "num_input_tokens_seen": 35980256,
      "step": 62025
    },
    {
      "epoch": 9.238903783139708,
      "grad_norm": 9.75,
      "learning_rate": 3.256442580625595e-05,
      "loss": 1.2199,
      "num_input_tokens_seen": 35983200,
      "step": 62030
    },
    {
      "epoch": 9.239648495680667,
      "grad_norm": 9.4375,
      "learning_rate": 3.2561328637758475e-05,
      "loss": 1.3952,
      "num_input_tokens_seen": 35986144,
      "step": 62035
    },
    {
      "epoch": 9.240393208221626,
      "grad_norm": 17.125,
      "learning_rate": 3.2558231341520046e-05,
      "loss": 0.936,
      "num_input_tokens_seen": 35988896,
      "step": 62040
    },
    {
      "epoch": 9.241137920762586,
      "grad_norm": 5.1875,
      "learning_rate": 3.255513391759299e-05,
      "loss": 1.105,
      "num_input_tokens_seen": 35992320,
      "step": 62045
    },
    {
      "epoch": 9.241882633303545,
      "grad_norm": 12.5625,
      "learning_rate": 3.25520363660296e-05,
      "loss": 1.3015,
      "num_input_tokens_seen": 35995040,
      "step": 62050
    },
    {
      "epoch": 9.242627345844504,
      "grad_norm": 15.75,
      "learning_rate": 3.2548938686882246e-05,
      "loss": 1.2915,
      "num_input_tokens_seen": 35997856,
      "step": 62055
    },
    {
      "epoch": 9.243372058385463,
      "grad_norm": 7.59375,
      "learning_rate": 3.254584088020325e-05,
      "loss": 1.23,
      "num_input_tokens_seen": 36000800,
      "step": 62060
    },
    {
      "epoch": 9.244116770926423,
      "grad_norm": 8.25,
      "learning_rate": 3.254274294604494e-05,
      "loss": 1.3874,
      "num_input_tokens_seen": 36003424,
      "step": 62065
    },
    {
      "epoch": 9.244861483467382,
      "grad_norm": 8.3125,
      "learning_rate": 3.253964488445964e-05,
      "loss": 1.2078,
      "num_input_tokens_seen": 36006400,
      "step": 62070
    },
    {
      "epoch": 9.24560619600834,
      "grad_norm": 16.25,
      "learning_rate": 3.253654669549972e-05,
      "loss": 1.2863,
      "num_input_tokens_seen": 36009440,
      "step": 62075
    },
    {
      "epoch": 9.2463509085493,
      "grad_norm": 12.75,
      "learning_rate": 3.253344837921749e-05,
      "loss": 1.3179,
      "num_input_tokens_seen": 36012160,
      "step": 62080
    },
    {
      "epoch": 9.24709562109026,
      "grad_norm": 10.5625,
      "learning_rate": 3.253034993566532e-05,
      "loss": 1.0282,
      "num_input_tokens_seen": 36015072,
      "step": 62085
    },
    {
      "epoch": 9.247840333631219,
      "grad_norm": 13.5,
      "learning_rate": 3.252725136489553e-05,
      "loss": 1.1817,
      "num_input_tokens_seen": 36018144,
      "step": 62090
    },
    {
      "epoch": 9.248585046172177,
      "grad_norm": 9.3125,
      "learning_rate": 3.2524152666960476e-05,
      "loss": 1.0131,
      "num_input_tokens_seen": 36021184,
      "step": 62095
    },
    {
      "epoch": 9.249329758713136,
      "grad_norm": 10.5625,
      "learning_rate": 3.252105384191252e-05,
      "loss": 0.9439,
      "num_input_tokens_seen": 36023840,
      "step": 62100
    },
    {
      "epoch": 9.250074471254097,
      "grad_norm": 13.9375,
      "learning_rate": 3.2517954889803995e-05,
      "loss": 1.2437,
      "num_input_tokens_seen": 36027104,
      "step": 62105
    },
    {
      "epoch": 9.250819183795056,
      "grad_norm": 19.75,
      "learning_rate": 3.2514855810687265e-05,
      "loss": 1.0888,
      "num_input_tokens_seen": 36029984,
      "step": 62110
    },
    {
      "epoch": 9.251563896336014,
      "grad_norm": 7.6875,
      "learning_rate": 3.2511756604614695e-05,
      "loss": 1.2553,
      "num_input_tokens_seen": 36032960,
      "step": 62115
    },
    {
      "epoch": 9.252308608876973,
      "grad_norm": 11.6875,
      "learning_rate": 3.250865727163862e-05,
      "loss": 1.2519,
      "num_input_tokens_seen": 36035808,
      "step": 62120
    },
    {
      "epoch": 9.253053321417934,
      "grad_norm": 7.6875,
      "learning_rate": 3.250555781181142e-05,
      "loss": 0.9009,
      "num_input_tokens_seen": 36038592,
      "step": 62125
    },
    {
      "epoch": 9.253798033958892,
      "grad_norm": 10.25,
      "learning_rate": 3.250245822518544e-05,
      "loss": 0.9238,
      "num_input_tokens_seen": 36041440,
      "step": 62130
    },
    {
      "epoch": 9.254542746499851,
      "grad_norm": 15.75,
      "learning_rate": 3.249935851181305e-05,
      "loss": 1.3164,
      "num_input_tokens_seen": 36044288,
      "step": 62135
    },
    {
      "epoch": 9.25528745904081,
      "grad_norm": 10.5625,
      "learning_rate": 3.2496258671746636e-05,
      "loss": 1.1311,
      "num_input_tokens_seen": 36046848,
      "step": 62140
    },
    {
      "epoch": 9.256032171581769,
      "grad_norm": 12.1875,
      "learning_rate": 3.249315870503854e-05,
      "loss": 0.8644,
      "num_input_tokens_seen": 36049792,
      "step": 62145
    },
    {
      "epoch": 9.25677688412273,
      "grad_norm": 7.0625,
      "learning_rate": 3.249005861174115e-05,
      "loss": 1.2008,
      "num_input_tokens_seen": 36052576,
      "step": 62150
    },
    {
      "epoch": 9.257521596663688,
      "grad_norm": 9.8125,
      "learning_rate": 3.2486958391906825e-05,
      "loss": 0.8164,
      "num_input_tokens_seen": 36055328,
      "step": 62155
    },
    {
      "epoch": 9.258266309204647,
      "grad_norm": 4.5,
      "learning_rate": 3.2483858045587944e-05,
      "loss": 0.8737,
      "num_input_tokens_seen": 36058144,
      "step": 62160
    },
    {
      "epoch": 9.259011021745605,
      "grad_norm": 17.125,
      "learning_rate": 3.2480757572836895e-05,
      "loss": 1.2715,
      "num_input_tokens_seen": 36061056,
      "step": 62165
    },
    {
      "epoch": 9.259755734286566,
      "grad_norm": 16.125,
      "learning_rate": 3.247765697370604e-05,
      "loss": 0.8784,
      "num_input_tokens_seen": 36064160,
      "step": 62170
    },
    {
      "epoch": 9.260500446827525,
      "grad_norm": 13.25,
      "learning_rate": 3.247455624824779e-05,
      "loss": 1.2189,
      "num_input_tokens_seen": 36067008,
      "step": 62175
    },
    {
      "epoch": 9.261245159368483,
      "grad_norm": 16.0,
      "learning_rate": 3.247145539651449e-05,
      "loss": 1.4403,
      "num_input_tokens_seen": 36069984,
      "step": 62180
    },
    {
      "epoch": 9.261989871909442,
      "grad_norm": 8.25,
      "learning_rate": 3.246835441855856e-05,
      "loss": 0.8599,
      "num_input_tokens_seen": 36073088,
      "step": 62185
    },
    {
      "epoch": 9.262734584450403,
      "grad_norm": 6.15625,
      "learning_rate": 3.2465253314432366e-05,
      "loss": 0.9043,
      "num_input_tokens_seen": 36075904,
      "step": 62190
    },
    {
      "epoch": 9.263479296991362,
      "grad_norm": 17.0,
      "learning_rate": 3.24621520841883e-05,
      "loss": 1.3662,
      "num_input_tokens_seen": 36078656,
      "step": 62195
    },
    {
      "epoch": 9.26422400953232,
      "grad_norm": 21.75,
      "learning_rate": 3.245905072787876e-05,
      "loss": 1.2905,
      "num_input_tokens_seen": 36081568,
      "step": 62200
    },
    {
      "epoch": 9.264968722073279,
      "grad_norm": 14.1875,
      "learning_rate": 3.245594924555614e-05,
      "loss": 1.0457,
      "num_input_tokens_seen": 36084384,
      "step": 62205
    },
    {
      "epoch": 9.26571343461424,
      "grad_norm": 22.375,
      "learning_rate": 3.2452847637272845e-05,
      "loss": 1.55,
      "num_input_tokens_seen": 36087232,
      "step": 62210
    },
    {
      "epoch": 9.266458147155198,
      "grad_norm": 6.28125,
      "learning_rate": 3.244974590308125e-05,
      "loss": 1.0088,
      "num_input_tokens_seen": 36090016,
      "step": 62215
    },
    {
      "epoch": 9.267202859696157,
      "grad_norm": 10.4375,
      "learning_rate": 3.244664404303378e-05,
      "loss": 0.9683,
      "num_input_tokens_seen": 36092576,
      "step": 62220
    },
    {
      "epoch": 9.267947572237116,
      "grad_norm": 14.75,
      "learning_rate": 3.2443542057182825e-05,
      "loss": 0.9304,
      "num_input_tokens_seen": 36095264,
      "step": 62225
    },
    {
      "epoch": 9.268692284778076,
      "grad_norm": 11.1875,
      "learning_rate": 3.244043994558079e-05,
      "loss": 1.2593,
      "num_input_tokens_seen": 36098208,
      "step": 62230
    },
    {
      "epoch": 9.269436997319035,
      "grad_norm": 7.40625,
      "learning_rate": 3.243733770828008e-05,
      "loss": 0.9913,
      "num_input_tokens_seen": 36101152,
      "step": 62235
    },
    {
      "epoch": 9.270181709859994,
      "grad_norm": 3.703125,
      "learning_rate": 3.243423534533311e-05,
      "loss": 1.344,
      "num_input_tokens_seen": 36103680,
      "step": 62240
    },
    {
      "epoch": 9.270926422400953,
      "grad_norm": 12.125,
      "learning_rate": 3.2431132856792294e-05,
      "loss": 1.0429,
      "num_input_tokens_seen": 36106432,
      "step": 62245
    },
    {
      "epoch": 9.271671134941913,
      "grad_norm": 9.125,
      "learning_rate": 3.242803024271004e-05,
      "loss": 1.1338,
      "num_input_tokens_seen": 36109088,
      "step": 62250
    },
    {
      "epoch": 9.272415847482872,
      "grad_norm": 11.0625,
      "learning_rate": 3.2424927503138766e-05,
      "loss": 1.0694,
      "num_input_tokens_seen": 36112288,
      "step": 62255
    },
    {
      "epoch": 9.27316056002383,
      "grad_norm": 10.875,
      "learning_rate": 3.242182463813088e-05,
      "loss": 0.8919,
      "num_input_tokens_seen": 36115360,
      "step": 62260
    },
    {
      "epoch": 9.27390527256479,
      "grad_norm": 14.4375,
      "learning_rate": 3.241872164773882e-05,
      "loss": 1.1067,
      "num_input_tokens_seen": 36118208,
      "step": 62265
    },
    {
      "epoch": 9.27464998510575,
      "grad_norm": 5.5,
      "learning_rate": 3.241561853201499e-05,
      "loss": 1.1084,
      "num_input_tokens_seen": 36121152,
      "step": 62270
    },
    {
      "epoch": 9.275394697646709,
      "grad_norm": 12.5,
      "learning_rate": 3.2412515291011826e-05,
      "loss": 0.9996,
      "num_input_tokens_seen": 36124192,
      "step": 62275
    },
    {
      "epoch": 9.276139410187668,
      "grad_norm": 12.0625,
      "learning_rate": 3.2409411924781754e-05,
      "loss": 0.9991,
      "num_input_tokens_seen": 36127488,
      "step": 62280
    },
    {
      "epoch": 9.276884122728626,
      "grad_norm": 7.46875,
      "learning_rate": 3.2406308433377194e-05,
      "loss": 1.1766,
      "num_input_tokens_seen": 36130304,
      "step": 62285
    },
    {
      "epoch": 9.277628835269585,
      "grad_norm": 9.75,
      "learning_rate": 3.2403204816850574e-05,
      "loss": 1.2286,
      "num_input_tokens_seen": 36133568,
      "step": 62290
    },
    {
      "epoch": 9.278373547810546,
      "grad_norm": 15.8125,
      "learning_rate": 3.240010107525434e-05,
      "loss": 0.887,
      "num_input_tokens_seen": 36136576,
      "step": 62295
    },
    {
      "epoch": 9.279118260351504,
      "grad_norm": 13.25,
      "learning_rate": 3.2396997208640925e-05,
      "loss": 1.2866,
      "num_input_tokens_seen": 36139264,
      "step": 62300
    },
    {
      "epoch": 9.279862972892463,
      "grad_norm": 8.25,
      "learning_rate": 3.2393893217062746e-05,
      "loss": 0.8994,
      "num_input_tokens_seen": 36142112,
      "step": 62305
    },
    {
      "epoch": 9.280607685433422,
      "grad_norm": 5.96875,
      "learning_rate": 3.239078910057226e-05,
      "loss": 1.1795,
      "num_input_tokens_seen": 36145120,
      "step": 62310
    },
    {
      "epoch": 9.281352397974382,
      "grad_norm": 14.6875,
      "learning_rate": 3.238768485922191e-05,
      "loss": 1.301,
      "num_input_tokens_seen": 36147872,
      "step": 62315
    },
    {
      "epoch": 9.282097110515341,
      "grad_norm": 7.28125,
      "learning_rate": 3.238458049306413e-05,
      "loss": 1.0285,
      "num_input_tokens_seen": 36150720,
      "step": 62320
    },
    {
      "epoch": 9.2828418230563,
      "grad_norm": 5.9375,
      "learning_rate": 3.2381476002151365e-05,
      "loss": 0.9073,
      "num_input_tokens_seen": 36153824,
      "step": 62325
    },
    {
      "epoch": 9.283586535597259,
      "grad_norm": 8.375,
      "learning_rate": 3.2378371386536074e-05,
      "loss": 1.4027,
      "num_input_tokens_seen": 36156576,
      "step": 62330
    },
    {
      "epoch": 9.28433124813822,
      "grad_norm": 7.5,
      "learning_rate": 3.2375266646270684e-05,
      "loss": 1.0093,
      "num_input_tokens_seen": 36159200,
      "step": 62335
    },
    {
      "epoch": 9.285075960679178,
      "grad_norm": 4.46875,
      "learning_rate": 3.2372161781407675e-05,
      "loss": 0.9929,
      "num_input_tokens_seen": 36162144,
      "step": 62340
    },
    {
      "epoch": 9.285820673220137,
      "grad_norm": 11.625,
      "learning_rate": 3.2369056791999476e-05,
      "loss": 1.3345,
      "num_input_tokens_seen": 36164928,
      "step": 62345
    },
    {
      "epoch": 9.286565385761095,
      "grad_norm": 14.375,
      "learning_rate": 3.236595167809856e-05,
      "loss": 1.4498,
      "num_input_tokens_seen": 36167936,
      "step": 62350
    },
    {
      "epoch": 9.287310098302056,
      "grad_norm": 8.125,
      "learning_rate": 3.236284643975737e-05,
      "loss": 1.0762,
      "num_input_tokens_seen": 36171008,
      "step": 62355
    },
    {
      "epoch": 9.288054810843015,
      "grad_norm": 11.6875,
      "learning_rate": 3.235974107702837e-05,
      "loss": 1.7131,
      "num_input_tokens_seen": 36173888,
      "step": 62360
    },
    {
      "epoch": 9.288799523383974,
      "grad_norm": 13.8125,
      "learning_rate": 3.235663558996402e-05,
      "loss": 1.2389,
      "num_input_tokens_seen": 36176896,
      "step": 62365
    },
    {
      "epoch": 9.289544235924932,
      "grad_norm": 10.625,
      "learning_rate": 3.2353529978616806e-05,
      "loss": 1.0001,
      "num_input_tokens_seen": 36179584,
      "step": 62370
    },
    {
      "epoch": 9.290288948465893,
      "grad_norm": 12.125,
      "learning_rate": 3.235042424303917e-05,
      "loss": 1.1125,
      "num_input_tokens_seen": 36182656,
      "step": 62375
    },
    {
      "epoch": 9.291033661006852,
      "grad_norm": 14.625,
      "learning_rate": 3.2347318383283585e-05,
      "loss": 1.3228,
      "num_input_tokens_seen": 36185600,
      "step": 62380
    },
    {
      "epoch": 9.29177837354781,
      "grad_norm": 11.5625,
      "learning_rate": 3.234421239940252e-05,
      "loss": 1.0537,
      "num_input_tokens_seen": 36188704,
      "step": 62385
    },
    {
      "epoch": 9.292523086088769,
      "grad_norm": 20.625,
      "learning_rate": 3.2341106291448456e-05,
      "loss": 0.9929,
      "num_input_tokens_seen": 36191392,
      "step": 62390
    },
    {
      "epoch": 9.29326779862973,
      "grad_norm": 6.5625,
      "learning_rate": 3.233800005947386e-05,
      "loss": 0.9594,
      "num_input_tokens_seen": 36194208,
      "step": 62395
    },
    {
      "epoch": 9.294012511170688,
      "grad_norm": 13.4375,
      "learning_rate": 3.23348937035312e-05,
      "loss": 0.8145,
      "num_input_tokens_seen": 36196992,
      "step": 62400
    },
    {
      "epoch": 9.294757223711647,
      "grad_norm": 12.3125,
      "learning_rate": 3.233178722367298e-05,
      "loss": 1.067,
      "num_input_tokens_seen": 36199648,
      "step": 62405
    },
    {
      "epoch": 9.295501936252606,
      "grad_norm": 7.21875,
      "learning_rate": 3.232868061995167e-05,
      "loss": 0.8314,
      "num_input_tokens_seen": 36202848,
      "step": 62410
    },
    {
      "epoch": 9.296246648793566,
      "grad_norm": 13.125,
      "learning_rate": 3.2325573892419745e-05,
      "loss": 1.1103,
      "num_input_tokens_seen": 36205728,
      "step": 62415
    },
    {
      "epoch": 9.296991361334525,
      "grad_norm": 14.1875,
      "learning_rate": 3.232246704112969e-05,
      "loss": 1.2483,
      "num_input_tokens_seen": 36208352,
      "step": 62420
    },
    {
      "epoch": 9.297736073875484,
      "grad_norm": 8.1875,
      "learning_rate": 3.2319360066134e-05,
      "loss": 0.928,
      "num_input_tokens_seen": 36211648,
      "step": 62425
    },
    {
      "epoch": 9.298480786416443,
      "grad_norm": 9.1875,
      "learning_rate": 3.2316252967485155e-05,
      "loss": 0.8159,
      "num_input_tokens_seen": 36214272,
      "step": 62430
    },
    {
      "epoch": 9.299225498957403,
      "grad_norm": 10.375,
      "learning_rate": 3.231314574523566e-05,
      "loss": 1.3024,
      "num_input_tokens_seen": 36217376,
      "step": 62435
    },
    {
      "epoch": 9.299970211498362,
      "grad_norm": 6.25,
      "learning_rate": 3.2310038399437995e-05,
      "loss": 1.1162,
      "num_input_tokens_seen": 36220288,
      "step": 62440
    },
    {
      "epoch": 9.30071492403932,
      "grad_norm": 10.625,
      "learning_rate": 3.230693093014466e-05,
      "loss": 1.3393,
      "num_input_tokens_seen": 36223200,
      "step": 62445
    },
    {
      "epoch": 9.30145963658028,
      "grad_norm": 22.875,
      "learning_rate": 3.230382333740816e-05,
      "loss": 1.5955,
      "num_input_tokens_seen": 36226080,
      "step": 62450
    },
    {
      "epoch": 9.30220434912124,
      "grad_norm": 10.25,
      "learning_rate": 3.230071562128098e-05,
      "loss": 0.7638,
      "num_input_tokens_seen": 36228896,
      "step": 62455
    },
    {
      "epoch": 9.302949061662199,
      "grad_norm": 7.46875,
      "learning_rate": 3.2297607781815645e-05,
      "loss": 1.1597,
      "num_input_tokens_seen": 36231872,
      "step": 62460
    },
    {
      "epoch": 9.303693774203158,
      "grad_norm": 9.8125,
      "learning_rate": 3.229449981906463e-05,
      "loss": 1.1502,
      "num_input_tokens_seen": 36234848,
      "step": 62465
    },
    {
      "epoch": 9.304438486744116,
      "grad_norm": 12.0,
      "learning_rate": 3.229139173308045e-05,
      "loss": 0.9137,
      "num_input_tokens_seen": 36237696,
      "step": 62470
    },
    {
      "epoch": 9.305183199285075,
      "grad_norm": 8.0625,
      "learning_rate": 3.228828352391562e-05,
      "loss": 1.1074,
      "num_input_tokens_seen": 36240928,
      "step": 62475
    },
    {
      "epoch": 9.305927911826036,
      "grad_norm": 15.6875,
      "learning_rate": 3.2285175191622656e-05,
      "loss": 1.3051,
      "num_input_tokens_seen": 36243584,
      "step": 62480
    },
    {
      "epoch": 9.306672624366994,
      "grad_norm": 12.5625,
      "learning_rate": 3.2282066736254056e-05,
      "loss": 1.0441,
      "num_input_tokens_seen": 36246720,
      "step": 62485
    },
    {
      "epoch": 9.307417336907953,
      "grad_norm": 7.5625,
      "learning_rate": 3.2278958157862336e-05,
      "loss": 0.6706,
      "num_input_tokens_seen": 36249472,
      "step": 62490
    },
    {
      "epoch": 9.308162049448912,
      "grad_norm": 16.375,
      "learning_rate": 3.2275849456500026e-05,
      "loss": 0.8545,
      "num_input_tokens_seen": 36252224,
      "step": 62495
    },
    {
      "epoch": 9.308906761989872,
      "grad_norm": 9.3125,
      "learning_rate": 3.2272740632219635e-05,
      "loss": 1.1093,
      "num_input_tokens_seen": 36254912,
      "step": 62500
    },
    {
      "epoch": 9.309651474530831,
      "grad_norm": 17.5,
      "learning_rate": 3.226963168507367e-05,
      "loss": 1.4886,
      "num_input_tokens_seen": 36257824,
      "step": 62505
    },
    {
      "epoch": 9.31039618707179,
      "grad_norm": 13.1875,
      "learning_rate": 3.226652261511467e-05,
      "loss": 1.2138,
      "num_input_tokens_seen": 36261056,
      "step": 62510
    },
    {
      "epoch": 9.311140899612749,
      "grad_norm": 8.5,
      "learning_rate": 3.226341342239516e-05,
      "loss": 1.3218,
      "num_input_tokens_seen": 36263808,
      "step": 62515
    },
    {
      "epoch": 9.31188561215371,
      "grad_norm": 9.4375,
      "learning_rate": 3.226030410696766e-05,
      "loss": 0.8534,
      "num_input_tokens_seen": 36267072,
      "step": 62520
    },
    {
      "epoch": 9.312630324694668,
      "grad_norm": 12.0625,
      "learning_rate": 3.2257194668884704e-05,
      "loss": 1.3069,
      "num_input_tokens_seen": 36269760,
      "step": 62525
    },
    {
      "epoch": 9.313375037235627,
      "grad_norm": 8.25,
      "learning_rate": 3.2254085108198815e-05,
      "loss": 1.4047,
      "num_input_tokens_seen": 36272800,
      "step": 62530
    },
    {
      "epoch": 9.314119749776586,
      "grad_norm": 16.125,
      "learning_rate": 3.225097542496254e-05,
      "loss": 1.1672,
      "num_input_tokens_seen": 36275744,
      "step": 62535
    },
    {
      "epoch": 9.314864462317546,
      "grad_norm": 9.125,
      "learning_rate": 3.2247865619228394e-05,
      "loss": 0.9591,
      "num_input_tokens_seen": 36278592,
      "step": 62540
    },
    {
      "epoch": 9.315609174858505,
      "grad_norm": 9.8125,
      "learning_rate": 3.2244755691048933e-05,
      "loss": 1.0945,
      "num_input_tokens_seen": 36281344,
      "step": 62545
    },
    {
      "epoch": 9.316353887399464,
      "grad_norm": 9.8125,
      "learning_rate": 3.224164564047669e-05,
      "loss": 1.1136,
      "num_input_tokens_seen": 36284192,
      "step": 62550
    },
    {
      "epoch": 9.317098599940422,
      "grad_norm": 10.6875,
      "learning_rate": 3.223853546756419e-05,
      "loss": 1.3211,
      "num_input_tokens_seen": 36287296,
      "step": 62555
    },
    {
      "epoch": 9.317843312481383,
      "grad_norm": 6.375,
      "learning_rate": 3.2235425172363996e-05,
      "loss": 0.8786,
      "num_input_tokens_seen": 36290240,
      "step": 62560
    },
    {
      "epoch": 9.318588025022342,
      "grad_norm": 9.5625,
      "learning_rate": 3.223231475492865e-05,
      "loss": 1.1831,
      "num_input_tokens_seen": 36293216,
      "step": 62565
    },
    {
      "epoch": 9.3193327375633,
      "grad_norm": 11.9375,
      "learning_rate": 3.222920421531069e-05,
      "loss": 0.9026,
      "num_input_tokens_seen": 36295808,
      "step": 62570
    },
    {
      "epoch": 9.32007745010426,
      "grad_norm": 12.6875,
      "learning_rate": 3.222609355356269e-05,
      "loss": 1.267,
      "num_input_tokens_seen": 36298784,
      "step": 62575
    },
    {
      "epoch": 9.32082216264522,
      "grad_norm": 10.5,
      "learning_rate": 3.222298276973717e-05,
      "loss": 1.1439,
      "num_input_tokens_seen": 36302048,
      "step": 62580
    },
    {
      "epoch": 9.321566875186178,
      "grad_norm": 9.625,
      "learning_rate": 3.22198718638867e-05,
      "loss": 1.1802,
      "num_input_tokens_seen": 36304672,
      "step": 62585
    },
    {
      "epoch": 9.322311587727137,
      "grad_norm": 6.46875,
      "learning_rate": 3.2216760836063834e-05,
      "loss": 0.7202,
      "num_input_tokens_seen": 36307552,
      "step": 62590
    },
    {
      "epoch": 9.323056300268096,
      "grad_norm": 14.5625,
      "learning_rate": 3.2213649686321124e-05,
      "loss": 1.1924,
      "num_input_tokens_seen": 36310496,
      "step": 62595
    },
    {
      "epoch": 9.323801012809056,
      "grad_norm": 12.3125,
      "learning_rate": 3.2210538414711136e-05,
      "loss": 0.9907,
      "num_input_tokens_seen": 36313344,
      "step": 62600
    },
    {
      "epoch": 9.324545725350015,
      "grad_norm": 12.8125,
      "learning_rate": 3.220742702128643e-05,
      "loss": 1.1788,
      "num_input_tokens_seen": 36316256,
      "step": 62605
    },
    {
      "epoch": 9.325290437890974,
      "grad_norm": 10.875,
      "learning_rate": 3.220431550609958e-05,
      "loss": 0.9369,
      "num_input_tokens_seen": 36319168,
      "step": 62610
    },
    {
      "epoch": 9.326035150431933,
      "grad_norm": 9.75,
      "learning_rate": 3.220120386920313e-05,
      "loss": 0.909,
      "num_input_tokens_seen": 36322016,
      "step": 62615
    },
    {
      "epoch": 9.326779862972893,
      "grad_norm": 16.75,
      "learning_rate": 3.219809211064966e-05,
      "loss": 1.072,
      "num_input_tokens_seen": 36324768,
      "step": 62620
    },
    {
      "epoch": 9.327524575513852,
      "grad_norm": 7.625,
      "learning_rate": 3.2194980230491744e-05,
      "loss": 1.1517,
      "num_input_tokens_seen": 36327456,
      "step": 62625
    },
    {
      "epoch": 9.32826928805481,
      "grad_norm": 7.65625,
      "learning_rate": 3.2191868228781944e-05,
      "loss": 1.12,
      "num_input_tokens_seen": 36330432,
      "step": 62630
    },
    {
      "epoch": 9.32901400059577,
      "grad_norm": 15.9375,
      "learning_rate": 3.2188756105572844e-05,
      "loss": 1.0255,
      "num_input_tokens_seen": 36333600,
      "step": 62635
    },
    {
      "epoch": 9.32975871313673,
      "grad_norm": 14.0,
      "learning_rate": 3.218564386091701e-05,
      "loss": 1.078,
      "num_input_tokens_seen": 36336384,
      "step": 62640
    },
    {
      "epoch": 9.330503425677689,
      "grad_norm": 20.125,
      "learning_rate": 3.218253149486704e-05,
      "loss": 1.2335,
      "num_input_tokens_seen": 36339488,
      "step": 62645
    },
    {
      "epoch": 9.331248138218648,
      "grad_norm": 4.0,
      "learning_rate": 3.2179419007475483e-05,
      "loss": 0.8192,
      "num_input_tokens_seen": 36342368,
      "step": 62650
    },
    {
      "epoch": 9.331992850759606,
      "grad_norm": 10.1875,
      "learning_rate": 3.217630639879495e-05,
      "loss": 0.9881,
      "num_input_tokens_seen": 36345472,
      "step": 62655
    },
    {
      "epoch": 9.332737563300565,
      "grad_norm": 5.59375,
      "learning_rate": 3.217319366887801e-05,
      "loss": 1.4312,
      "num_input_tokens_seen": 36348480,
      "step": 62660
    },
    {
      "epoch": 9.333482275841526,
      "grad_norm": 15.5625,
      "learning_rate": 3.217008081777726e-05,
      "loss": 1.5516,
      "num_input_tokens_seen": 36351808,
      "step": 62665
    },
    {
      "epoch": 9.334226988382484,
      "grad_norm": 6.625,
      "learning_rate": 3.2166967845545275e-05,
      "loss": 1.0168,
      "num_input_tokens_seen": 36354560,
      "step": 62670
    },
    {
      "epoch": 9.334971700923443,
      "grad_norm": 4.65625,
      "learning_rate": 3.216385475223465e-05,
      "loss": 1.1035,
      "num_input_tokens_seen": 36357600,
      "step": 62675
    },
    {
      "epoch": 9.335716413464402,
      "grad_norm": 8.3125,
      "learning_rate": 3.216074153789799e-05,
      "loss": 0.9904,
      "num_input_tokens_seen": 36360320,
      "step": 62680
    },
    {
      "epoch": 9.336461126005362,
      "grad_norm": 21.625,
      "learning_rate": 3.2157628202587874e-05,
      "loss": 1.3298,
      "num_input_tokens_seen": 36363168,
      "step": 62685
    },
    {
      "epoch": 9.337205838546321,
      "grad_norm": 12.1875,
      "learning_rate": 3.21545147463569e-05,
      "loss": 0.94,
      "num_input_tokens_seen": 36366112,
      "step": 62690
    },
    {
      "epoch": 9.33795055108728,
      "grad_norm": 6.34375,
      "learning_rate": 3.2151401169257676e-05,
      "loss": 0.8705,
      "num_input_tokens_seen": 36368960,
      "step": 62695
    },
    {
      "epoch": 9.338695263628239,
      "grad_norm": 4.21875,
      "learning_rate": 3.2148287471342796e-05,
      "loss": 0.9926,
      "num_input_tokens_seen": 36371872,
      "step": 62700
    },
    {
      "epoch": 9.3394399761692,
      "grad_norm": 12.4375,
      "learning_rate": 3.2145173652664864e-05,
      "loss": 1.0037,
      "num_input_tokens_seen": 36374528,
      "step": 62705
    },
    {
      "epoch": 9.340184688710158,
      "grad_norm": 5.9375,
      "learning_rate": 3.21420597132765e-05,
      "loss": 0.9149,
      "num_input_tokens_seen": 36377792,
      "step": 62710
    },
    {
      "epoch": 9.340929401251117,
      "grad_norm": 10.875,
      "learning_rate": 3.213894565323027e-05,
      "loss": 1.1412,
      "num_input_tokens_seen": 36380800,
      "step": 62715
    },
    {
      "epoch": 9.341674113792076,
      "grad_norm": 6.0,
      "learning_rate": 3.213583147257883e-05,
      "loss": 0.7396,
      "num_input_tokens_seen": 36383360,
      "step": 62720
    },
    {
      "epoch": 9.342418826333036,
      "grad_norm": 9.3125,
      "learning_rate": 3.213271717137475e-05,
      "loss": 1.0147,
      "num_input_tokens_seen": 36385984,
      "step": 62725
    },
    {
      "epoch": 9.343163538873995,
      "grad_norm": 10.625,
      "learning_rate": 3.2129602749670674e-05,
      "loss": 1.1806,
      "num_input_tokens_seen": 36388832,
      "step": 62730
    },
    {
      "epoch": 9.343908251414954,
      "grad_norm": 11.4375,
      "learning_rate": 3.212648820751921e-05,
      "loss": 0.95,
      "num_input_tokens_seen": 36391968,
      "step": 62735
    },
    {
      "epoch": 9.344652963955912,
      "grad_norm": 8.375,
      "learning_rate": 3.212337354497296e-05,
      "loss": 1.0259,
      "num_input_tokens_seen": 36394688,
      "step": 62740
    },
    {
      "epoch": 9.345397676496873,
      "grad_norm": 14.6875,
      "learning_rate": 3.2120258762084565e-05,
      "loss": 1.0071,
      "num_input_tokens_seen": 36397408,
      "step": 62745
    },
    {
      "epoch": 9.346142389037832,
      "grad_norm": 9.875,
      "learning_rate": 3.211714385890663e-05,
      "loss": 1.5071,
      "num_input_tokens_seen": 36400512,
      "step": 62750
    },
    {
      "epoch": 9.34688710157879,
      "grad_norm": 5.5,
      "learning_rate": 3.2114028835491786e-05,
      "loss": 1.2526,
      "num_input_tokens_seen": 36403296,
      "step": 62755
    },
    {
      "epoch": 9.34763181411975,
      "grad_norm": 10.875,
      "learning_rate": 3.211091369189265e-05,
      "loss": 1.3353,
      "num_input_tokens_seen": 36405984,
      "step": 62760
    },
    {
      "epoch": 9.34837652666071,
      "grad_norm": 19.375,
      "learning_rate": 3.210779842816185e-05,
      "loss": 1.2281,
      "num_input_tokens_seen": 36408928,
      "step": 62765
    },
    {
      "epoch": 9.349121239201668,
      "grad_norm": 16.75,
      "learning_rate": 3.2104683044352025e-05,
      "loss": 1.336,
      "num_input_tokens_seen": 36411744,
      "step": 62770
    },
    {
      "epoch": 9.349865951742627,
      "grad_norm": 8.0,
      "learning_rate": 3.210156754051581e-05,
      "loss": 1.0164,
      "num_input_tokens_seen": 36414592,
      "step": 62775
    },
    {
      "epoch": 9.350610664283586,
      "grad_norm": 8.75,
      "learning_rate": 3.2098451916705815e-05,
      "loss": 0.9448,
      "num_input_tokens_seen": 36417696,
      "step": 62780
    },
    {
      "epoch": 9.351355376824547,
      "grad_norm": 12.25,
      "learning_rate": 3.20953361729747e-05,
      "loss": 1.5736,
      "num_input_tokens_seen": 36420448,
      "step": 62785
    },
    {
      "epoch": 9.352100089365505,
      "grad_norm": 14.6875,
      "learning_rate": 3.209222030937509e-05,
      "loss": 1.1962,
      "num_input_tokens_seen": 36423424,
      "step": 62790
    },
    {
      "epoch": 9.352844801906464,
      "grad_norm": 7.03125,
      "learning_rate": 3.208910432595962e-05,
      "loss": 0.8805,
      "num_input_tokens_seen": 36426336,
      "step": 62795
    },
    {
      "epoch": 9.353589514447423,
      "grad_norm": 13.0625,
      "learning_rate": 3.208598822278094e-05,
      "loss": 1.1817,
      "num_input_tokens_seen": 36429024,
      "step": 62800
    },
    {
      "epoch": 9.354334226988382,
      "grad_norm": 4.9375,
      "learning_rate": 3.208287199989169e-05,
      "loss": 0.9892,
      "num_input_tokens_seen": 36432160,
      "step": 62805
    },
    {
      "epoch": 9.355078939529342,
      "grad_norm": 15.375,
      "learning_rate": 3.207975565734452e-05,
      "loss": 1.0238,
      "num_input_tokens_seen": 36435104,
      "step": 62810
    },
    {
      "epoch": 9.3558236520703,
      "grad_norm": 12.0625,
      "learning_rate": 3.207663919519207e-05,
      "loss": 0.9709,
      "num_input_tokens_seen": 36438016,
      "step": 62815
    },
    {
      "epoch": 9.35656836461126,
      "grad_norm": 11.5,
      "learning_rate": 3.2073522613486994e-05,
      "loss": 1.1324,
      "num_input_tokens_seen": 36440864,
      "step": 62820
    },
    {
      "epoch": 9.357313077152218,
      "grad_norm": 14.6875,
      "learning_rate": 3.207040591228194e-05,
      "loss": 1.1553,
      "num_input_tokens_seen": 36443680,
      "step": 62825
    },
    {
      "epoch": 9.358057789693179,
      "grad_norm": 19.125,
      "learning_rate": 3.206728909162957e-05,
      "loss": 1.0405,
      "num_input_tokens_seen": 36446496,
      "step": 62830
    },
    {
      "epoch": 9.358802502234138,
      "grad_norm": 15.8125,
      "learning_rate": 3.206417215158253e-05,
      "loss": 1.3344,
      "num_input_tokens_seen": 36449248,
      "step": 62835
    },
    {
      "epoch": 9.359547214775096,
      "grad_norm": 5.8125,
      "learning_rate": 3.206105509219348e-05,
      "loss": 1.0304,
      "num_input_tokens_seen": 36452096,
      "step": 62840
    },
    {
      "epoch": 9.360291927316055,
      "grad_norm": 12.625,
      "learning_rate": 3.205793791351509e-05,
      "loss": 0.7394,
      "num_input_tokens_seen": 36455392,
      "step": 62845
    },
    {
      "epoch": 9.361036639857016,
      "grad_norm": 10.1875,
      "learning_rate": 3.2054820615600003e-05,
      "loss": 0.9755,
      "num_input_tokens_seen": 36458112,
      "step": 62850
    },
    {
      "epoch": 9.361781352397974,
      "grad_norm": 12.875,
      "learning_rate": 3.2051703198500896e-05,
      "loss": 1.0365,
      "num_input_tokens_seen": 36461088,
      "step": 62855
    },
    {
      "epoch": 9.362526064938933,
      "grad_norm": 20.0,
      "learning_rate": 3.2048585662270425e-05,
      "loss": 1.4083,
      "num_input_tokens_seen": 36464608,
      "step": 62860
    },
    {
      "epoch": 9.363270777479892,
      "grad_norm": 11.625,
      "learning_rate": 3.204546800696127e-05,
      "loss": 0.9995,
      "num_input_tokens_seen": 36467488,
      "step": 62865
    },
    {
      "epoch": 9.364015490020853,
      "grad_norm": 17.5,
      "learning_rate": 3.2042350232626086e-05,
      "loss": 1.3295,
      "num_input_tokens_seen": 36470592,
      "step": 62870
    },
    {
      "epoch": 9.364760202561811,
      "grad_norm": 15.125,
      "learning_rate": 3.203923233931757e-05,
      "loss": 0.8408,
      "num_input_tokens_seen": 36473856,
      "step": 62875
    },
    {
      "epoch": 9.36550491510277,
      "grad_norm": 11.4375,
      "learning_rate": 3.2036114327088354e-05,
      "loss": 1.2387,
      "num_input_tokens_seen": 36476768,
      "step": 62880
    },
    {
      "epoch": 9.366249627643729,
      "grad_norm": 12.25,
      "learning_rate": 3.203299619599115e-05,
      "loss": 1.5429,
      "num_input_tokens_seen": 36479840,
      "step": 62885
    },
    {
      "epoch": 9.36699434018469,
      "grad_norm": 12.1875,
      "learning_rate": 3.2029877946078624e-05,
      "loss": 1.1645,
      "num_input_tokens_seen": 36483488,
      "step": 62890
    },
    {
      "epoch": 9.367739052725648,
      "grad_norm": 8.5,
      "learning_rate": 3.2026759577403445e-05,
      "loss": 0.9571,
      "num_input_tokens_seen": 36486080,
      "step": 62895
    },
    {
      "epoch": 9.368483765266607,
      "grad_norm": 7.125,
      "learning_rate": 3.202364109001831e-05,
      "loss": 1.144,
      "num_input_tokens_seen": 36489312,
      "step": 62900
    },
    {
      "epoch": 9.369228477807566,
      "grad_norm": 13.1875,
      "learning_rate": 3.2020522483975906e-05,
      "loss": 1.0207,
      "num_input_tokens_seen": 36492256,
      "step": 62905
    },
    {
      "epoch": 9.369973190348526,
      "grad_norm": 7.65625,
      "learning_rate": 3.201740375932891e-05,
      "loss": 1.1757,
      "num_input_tokens_seen": 36495040,
      "step": 62910
    },
    {
      "epoch": 9.370717902889485,
      "grad_norm": 11.875,
      "learning_rate": 3.201428491613e-05,
      "loss": 1.0925,
      "num_input_tokens_seen": 36497952,
      "step": 62915
    },
    {
      "epoch": 9.371462615430444,
      "grad_norm": 7.90625,
      "learning_rate": 3.2011165954431873e-05,
      "loss": 1.0626,
      "num_input_tokens_seen": 36500992,
      "step": 62920
    },
    {
      "epoch": 9.372207327971402,
      "grad_norm": 14.0625,
      "learning_rate": 3.200804687428724e-05,
      "loss": 1.117,
      "num_input_tokens_seen": 36503616,
      "step": 62925
    },
    {
      "epoch": 9.372952040512363,
      "grad_norm": 14.6875,
      "learning_rate": 3.200492767574876e-05,
      "loss": 0.8974,
      "num_input_tokens_seen": 36506368,
      "step": 62930
    },
    {
      "epoch": 9.373696753053322,
      "grad_norm": 10.1875,
      "learning_rate": 3.200180835886915e-05,
      "loss": 0.8366,
      "num_input_tokens_seen": 36509216,
      "step": 62935
    },
    {
      "epoch": 9.37444146559428,
      "grad_norm": 11.375,
      "learning_rate": 3.199868892370111e-05,
      "loss": 0.9745,
      "num_input_tokens_seen": 36511936,
      "step": 62940
    },
    {
      "epoch": 9.37518617813524,
      "grad_norm": 7.3125,
      "learning_rate": 3.199556937029734e-05,
      "loss": 1.0402,
      "num_input_tokens_seen": 36514848,
      "step": 62945
    },
    {
      "epoch": 9.3759308906762,
      "grad_norm": 13.625,
      "learning_rate": 3.199244969871052e-05,
      "loss": 0.9008,
      "num_input_tokens_seen": 36518016,
      "step": 62950
    },
    {
      "epoch": 9.376675603217159,
      "grad_norm": 15.6875,
      "learning_rate": 3.198932990899337e-05,
      "loss": 0.871,
      "num_input_tokens_seen": 36520960,
      "step": 62955
    },
    {
      "epoch": 9.377420315758117,
      "grad_norm": 15.5625,
      "learning_rate": 3.19862100011986e-05,
      "loss": 1.3135,
      "num_input_tokens_seen": 36523840,
      "step": 62960
    },
    {
      "epoch": 9.378165028299076,
      "grad_norm": 13.125,
      "learning_rate": 3.198308997537891e-05,
      "loss": 1.1826,
      "num_input_tokens_seen": 36526752,
      "step": 62965
    },
    {
      "epoch": 9.378909740840037,
      "grad_norm": 16.5,
      "learning_rate": 3.1979969831587014e-05,
      "loss": 1.2662,
      "num_input_tokens_seen": 36529984,
      "step": 62970
    },
    {
      "epoch": 9.379654453380995,
      "grad_norm": 21.5,
      "learning_rate": 3.1976849569875624e-05,
      "loss": 1.1232,
      "num_input_tokens_seen": 36532864,
      "step": 62975
    },
    {
      "epoch": 9.380399165921954,
      "grad_norm": 13.0,
      "learning_rate": 3.197372919029745e-05,
      "loss": 1.0753,
      "num_input_tokens_seen": 36535520,
      "step": 62980
    },
    {
      "epoch": 9.381143878462913,
      "grad_norm": 15.75,
      "learning_rate": 3.1970608692905216e-05,
      "loss": 1.0408,
      "num_input_tokens_seen": 36538368,
      "step": 62985
    },
    {
      "epoch": 9.381888591003872,
      "grad_norm": 7.5625,
      "learning_rate": 3.196748807775162e-05,
      "loss": 0.9986,
      "num_input_tokens_seen": 36541408,
      "step": 62990
    },
    {
      "epoch": 9.382633303544832,
      "grad_norm": 14.25,
      "learning_rate": 3.19643673448894e-05,
      "loss": 1.1878,
      "num_input_tokens_seen": 36544256,
      "step": 62995
    },
    {
      "epoch": 9.383378016085791,
      "grad_norm": 19.75,
      "learning_rate": 3.1961246494371275e-05,
      "loss": 0.9858,
      "num_input_tokens_seen": 36547008,
      "step": 63000
    },
    {
      "epoch": 9.38412272862675,
      "grad_norm": 13.25,
      "learning_rate": 3.195812552624996e-05,
      "loss": 1.0816,
      "num_input_tokens_seen": 36550240,
      "step": 63005
    },
    {
      "epoch": 9.384867441167708,
      "grad_norm": 11.9375,
      "learning_rate": 3.1955004440578196e-05,
      "loss": 1.1031,
      "num_input_tokens_seen": 36552992,
      "step": 63010
    },
    {
      "epoch": 9.385612153708669,
      "grad_norm": 4.21875,
      "learning_rate": 3.195188323740869e-05,
      "loss": 1.0687,
      "num_input_tokens_seen": 36555712,
      "step": 63015
    },
    {
      "epoch": 9.386356866249628,
      "grad_norm": 15.6875,
      "learning_rate": 3.194876191679418e-05,
      "loss": 1.3346,
      "num_input_tokens_seen": 36558624,
      "step": 63020
    },
    {
      "epoch": 9.387101578790586,
      "grad_norm": 5.53125,
      "learning_rate": 3.194564047878742e-05,
      "loss": 0.8122,
      "num_input_tokens_seen": 36561344,
      "step": 63025
    },
    {
      "epoch": 9.387846291331545,
      "grad_norm": 14.4375,
      "learning_rate": 3.19425189234411e-05,
      "loss": 0.835,
      "num_input_tokens_seen": 36564320,
      "step": 63030
    },
    {
      "epoch": 9.388591003872506,
      "grad_norm": 10.4375,
      "learning_rate": 3.193939725080799e-05,
      "loss": 0.9271,
      "num_input_tokens_seen": 36567200,
      "step": 63035
    },
    {
      "epoch": 9.389335716413465,
      "grad_norm": 7.4375,
      "learning_rate": 3.1936275460940815e-05,
      "loss": 1.1117,
      "num_input_tokens_seen": 36570080,
      "step": 63040
    },
    {
      "epoch": 9.390080428954423,
      "grad_norm": 9.8125,
      "learning_rate": 3.193315355389231e-05,
      "loss": 0.8785,
      "num_input_tokens_seen": 36573056,
      "step": 63045
    },
    {
      "epoch": 9.390825141495382,
      "grad_norm": 8.6875,
      "learning_rate": 3.1930031529715234e-05,
      "loss": 1.3876,
      "num_input_tokens_seen": 36576096,
      "step": 63050
    },
    {
      "epoch": 9.391569854036343,
      "grad_norm": 5.625,
      "learning_rate": 3.192690938846231e-05,
      "loss": 1.3372,
      "num_input_tokens_seen": 36578912,
      "step": 63055
    },
    {
      "epoch": 9.392314566577301,
      "grad_norm": 11.75,
      "learning_rate": 3.1923787130186286e-05,
      "loss": 1.2345,
      "num_input_tokens_seen": 36581696,
      "step": 63060
    },
    {
      "epoch": 9.39305927911826,
      "grad_norm": 7.4375,
      "learning_rate": 3.1920664754939936e-05,
      "loss": 1.3822,
      "num_input_tokens_seen": 36584736,
      "step": 63065
    },
    {
      "epoch": 9.393803991659219,
      "grad_norm": 9.5625,
      "learning_rate": 3.1917542262775975e-05,
      "loss": 0.7872,
      "num_input_tokens_seen": 36587712,
      "step": 63070
    },
    {
      "epoch": 9.39454870420018,
      "grad_norm": 11.0,
      "learning_rate": 3.191441965374717e-05,
      "loss": 1.028,
      "num_input_tokens_seen": 36590304,
      "step": 63075
    },
    {
      "epoch": 9.395293416741138,
      "grad_norm": 5.09375,
      "learning_rate": 3.191129692790627e-05,
      "loss": 1.4507,
      "num_input_tokens_seen": 36593056,
      "step": 63080
    },
    {
      "epoch": 9.396038129282097,
      "grad_norm": 5.40625,
      "learning_rate": 3.190817408530604e-05,
      "loss": 1.0168,
      "num_input_tokens_seen": 36595808,
      "step": 63085
    },
    {
      "epoch": 9.396782841823056,
      "grad_norm": 14.125,
      "learning_rate": 3.190505112599922e-05,
      "loss": 1.259,
      "num_input_tokens_seen": 36598784,
      "step": 63090
    },
    {
      "epoch": 9.397527554364016,
      "grad_norm": 6.59375,
      "learning_rate": 3.190192805003858e-05,
      "loss": 1.2032,
      "num_input_tokens_seen": 36601504,
      "step": 63095
    },
    {
      "epoch": 9.398272266904975,
      "grad_norm": 14.0,
      "learning_rate": 3.189880485747688e-05,
      "loss": 1.0023,
      "num_input_tokens_seen": 36604288,
      "step": 63100
    },
    {
      "epoch": 9.399016979445934,
      "grad_norm": 10.375,
      "learning_rate": 3.1895681548366896e-05,
      "loss": 1.0884,
      "num_input_tokens_seen": 36607104,
      "step": 63105
    },
    {
      "epoch": 9.399761691986892,
      "grad_norm": 8.8125,
      "learning_rate": 3.189255812276137e-05,
      "loss": 1.1172,
      "num_input_tokens_seen": 36609952,
      "step": 63110
    },
    {
      "epoch": 9.400506404527853,
      "grad_norm": 12.0,
      "learning_rate": 3.188943458071308e-05,
      "loss": 1.0523,
      "num_input_tokens_seen": 36612704,
      "step": 63115
    },
    {
      "epoch": 9.401251117068812,
      "grad_norm": 12.0625,
      "learning_rate": 3.18863109222748e-05,
      "loss": 1.4003,
      "num_input_tokens_seen": 36615360,
      "step": 63120
    },
    {
      "epoch": 9.40199582960977,
      "grad_norm": 17.125,
      "learning_rate": 3.188318714749929e-05,
      "loss": 1.0866,
      "num_input_tokens_seen": 36618528,
      "step": 63125
    },
    {
      "epoch": 9.40274054215073,
      "grad_norm": 12.3125,
      "learning_rate": 3.188006325643934e-05,
      "loss": 0.8091,
      "num_input_tokens_seen": 36621472,
      "step": 63130
    },
    {
      "epoch": 9.40348525469169,
      "grad_norm": 14.25,
      "learning_rate": 3.1876939249147694e-05,
      "loss": 1.1524,
      "num_input_tokens_seen": 36624256,
      "step": 63135
    },
    {
      "epoch": 9.404229967232649,
      "grad_norm": 11.3125,
      "learning_rate": 3.187381512567717e-05,
      "loss": 1.1213,
      "num_input_tokens_seen": 36627168,
      "step": 63140
    },
    {
      "epoch": 9.404974679773607,
      "grad_norm": 10.375,
      "learning_rate": 3.1870690886080515e-05,
      "loss": 1.1113,
      "num_input_tokens_seen": 36630144,
      "step": 63145
    },
    {
      "epoch": 9.405719392314566,
      "grad_norm": 8.4375,
      "learning_rate": 3.186756653041053e-05,
      "loss": 1.1147,
      "num_input_tokens_seen": 36633024,
      "step": 63150
    },
    {
      "epoch": 9.406464104855527,
      "grad_norm": 18.0,
      "learning_rate": 3.186444205871997e-05,
      "loss": 1.2478,
      "num_input_tokens_seen": 36635904,
      "step": 63155
    },
    {
      "epoch": 9.407208817396485,
      "grad_norm": 19.625,
      "learning_rate": 3.186131747106165e-05,
      "loss": 1.1214,
      "num_input_tokens_seen": 36638816,
      "step": 63160
    },
    {
      "epoch": 9.407953529937444,
      "grad_norm": 15.875,
      "learning_rate": 3.1858192767488345e-05,
      "loss": 1.0443,
      "num_input_tokens_seen": 36641600,
      "step": 63165
    },
    {
      "epoch": 9.408698242478403,
      "grad_norm": 13.375,
      "learning_rate": 3.185506794805284e-05,
      "loss": 1.1783,
      "num_input_tokens_seen": 36644384,
      "step": 63170
    },
    {
      "epoch": 9.409442955019362,
      "grad_norm": 9.6875,
      "learning_rate": 3.1851943012807934e-05,
      "loss": 0.9008,
      "num_input_tokens_seen": 36647104,
      "step": 63175
    },
    {
      "epoch": 9.410187667560322,
      "grad_norm": 12.5,
      "learning_rate": 3.184881796180641e-05,
      "loss": 0.9157,
      "num_input_tokens_seen": 36649824,
      "step": 63180
    },
    {
      "epoch": 9.410932380101281,
      "grad_norm": 13.0,
      "learning_rate": 3.184569279510107e-05,
      "loss": 1.4791,
      "num_input_tokens_seen": 36652800,
      "step": 63185
    },
    {
      "epoch": 9.41167709264224,
      "grad_norm": 11.9375,
      "learning_rate": 3.184256751274471e-05,
      "loss": 1.041,
      "num_input_tokens_seen": 36655584,
      "step": 63190
    },
    {
      "epoch": 9.412421805183198,
      "grad_norm": 8.8125,
      "learning_rate": 3.183944211479012e-05,
      "loss": 1.1358,
      "num_input_tokens_seen": 36658368,
      "step": 63195
    },
    {
      "epoch": 9.413166517724159,
      "grad_norm": 12.1875,
      "learning_rate": 3.183631660129011e-05,
      "loss": 0.9705,
      "num_input_tokens_seen": 36661184,
      "step": 63200
    },
    {
      "epoch": 9.413911230265118,
      "grad_norm": 14.0625,
      "learning_rate": 3.183319097229748e-05,
      "loss": 1.0978,
      "num_input_tokens_seen": 36664160,
      "step": 63205
    },
    {
      "epoch": 9.414655942806077,
      "grad_norm": 11.4375,
      "learning_rate": 3.183006522786504e-05,
      "loss": 1.4839,
      "num_input_tokens_seen": 36666976,
      "step": 63210
    },
    {
      "epoch": 9.415400655347035,
      "grad_norm": 17.125,
      "learning_rate": 3.182693936804558e-05,
      "loss": 1.187,
      "num_input_tokens_seen": 36669984,
      "step": 63215
    },
    {
      "epoch": 9.416145367887996,
      "grad_norm": 9.125,
      "learning_rate": 3.182381339289192e-05,
      "loss": 1.044,
      "num_input_tokens_seen": 36673088,
      "step": 63220
    },
    {
      "epoch": 9.416890080428955,
      "grad_norm": 14.3125,
      "learning_rate": 3.182068730245686e-05,
      "loss": 1.4204,
      "num_input_tokens_seen": 36675648,
      "step": 63225
    },
    {
      "epoch": 9.417634792969913,
      "grad_norm": 11.0625,
      "learning_rate": 3.181756109679324e-05,
      "loss": 0.951,
      "num_input_tokens_seen": 36678368,
      "step": 63230
    },
    {
      "epoch": 9.418379505510872,
      "grad_norm": 12.0,
      "learning_rate": 3.1814434775953837e-05,
      "loss": 1.1168,
      "num_input_tokens_seen": 36681120,
      "step": 63235
    },
    {
      "epoch": 9.419124218051833,
      "grad_norm": 15.3125,
      "learning_rate": 3.18113083399915e-05,
      "loss": 1.0642,
      "num_input_tokens_seen": 36683840,
      "step": 63240
    },
    {
      "epoch": 9.419868930592791,
      "grad_norm": 11.9375,
      "learning_rate": 3.180818178895901e-05,
      "loss": 1.1659,
      "num_input_tokens_seen": 36686880,
      "step": 63245
    },
    {
      "epoch": 9.42061364313375,
      "grad_norm": 7.65625,
      "learning_rate": 3.180505512290922e-05,
      "loss": 1.1654,
      "num_input_tokens_seen": 36689568,
      "step": 63250
    },
    {
      "epoch": 9.421358355674709,
      "grad_norm": 10.5,
      "learning_rate": 3.1801928341894943e-05,
      "loss": 0.8582,
      "num_input_tokens_seen": 36692288,
      "step": 63255
    },
    {
      "epoch": 9.42210306821567,
      "grad_norm": 7.21875,
      "learning_rate": 3.1798801445968993e-05,
      "loss": 1.177,
      "num_input_tokens_seen": 36695392,
      "step": 63260
    },
    {
      "epoch": 9.422847780756628,
      "grad_norm": 13.75,
      "learning_rate": 3.179567443518421e-05,
      "loss": 1.0536,
      "num_input_tokens_seen": 36698176,
      "step": 63265
    },
    {
      "epoch": 9.423592493297587,
      "grad_norm": 11.4375,
      "learning_rate": 3.1792547309593415e-05,
      "loss": 1.0215,
      "num_input_tokens_seen": 36700992,
      "step": 63270
    },
    {
      "epoch": 9.424337205838546,
      "grad_norm": 10.5,
      "learning_rate": 3.178942006924943e-05,
      "loss": 1.0282,
      "num_input_tokens_seen": 36703680,
      "step": 63275
    },
    {
      "epoch": 9.425081918379506,
      "grad_norm": 15.9375,
      "learning_rate": 3.17862927142051e-05,
      "loss": 1.294,
      "num_input_tokens_seen": 36706496,
      "step": 63280
    },
    {
      "epoch": 9.425826630920465,
      "grad_norm": 7.78125,
      "learning_rate": 3.178316524451325e-05,
      "loss": 1.3142,
      "num_input_tokens_seen": 36709408,
      "step": 63285
    },
    {
      "epoch": 9.426571343461424,
      "grad_norm": 7.25,
      "learning_rate": 3.178003766022671e-05,
      "loss": 1.1656,
      "num_input_tokens_seen": 36712256,
      "step": 63290
    },
    {
      "epoch": 9.427316056002383,
      "grad_norm": 9.5625,
      "learning_rate": 3.177690996139833e-05,
      "loss": 1.5289,
      "num_input_tokens_seen": 36715168,
      "step": 63295
    },
    {
      "epoch": 9.428060768543343,
      "grad_norm": 11.0625,
      "learning_rate": 3.177378214808094e-05,
      "loss": 1.0424,
      "num_input_tokens_seen": 36718208,
      "step": 63300
    },
    {
      "epoch": 9.428805481084302,
      "grad_norm": 13.5625,
      "learning_rate": 3.177065422032739e-05,
      "loss": 1.7055,
      "num_input_tokens_seen": 36721344,
      "step": 63305
    },
    {
      "epoch": 9.42955019362526,
      "grad_norm": 9.6875,
      "learning_rate": 3.176752617819052e-05,
      "loss": 0.7664,
      "num_input_tokens_seen": 36724064,
      "step": 63310
    },
    {
      "epoch": 9.43029490616622,
      "grad_norm": 14.3125,
      "learning_rate": 3.1764398021723175e-05,
      "loss": 1.0996,
      "num_input_tokens_seen": 36727168,
      "step": 63315
    },
    {
      "epoch": 9.43103961870718,
      "grad_norm": 14.5,
      "learning_rate": 3.17612697509782e-05,
      "loss": 1.3225,
      "num_input_tokens_seen": 36729984,
      "step": 63320
    },
    {
      "epoch": 9.431784331248139,
      "grad_norm": 9.25,
      "learning_rate": 3.1758141366008434e-05,
      "loss": 0.7802,
      "num_input_tokens_seen": 36732704,
      "step": 63325
    },
    {
      "epoch": 9.432529043789097,
      "grad_norm": 8.625,
      "learning_rate": 3.175501286686674e-05,
      "loss": 1.0421,
      "num_input_tokens_seen": 36735488,
      "step": 63330
    },
    {
      "epoch": 9.433273756330056,
      "grad_norm": 16.5,
      "learning_rate": 3.1751884253605974e-05,
      "loss": 1.4715,
      "num_input_tokens_seen": 36738368,
      "step": 63335
    },
    {
      "epoch": 9.434018468871017,
      "grad_norm": 11.125,
      "learning_rate": 3.174875552627899e-05,
      "loss": 1.0052,
      "num_input_tokens_seen": 36741152,
      "step": 63340
    },
    {
      "epoch": 9.434763181411975,
      "grad_norm": 8.8125,
      "learning_rate": 3.174562668493863e-05,
      "loss": 0.9538,
      "num_input_tokens_seen": 36743872,
      "step": 63345
    },
    {
      "epoch": 9.435507893952934,
      "grad_norm": 10.1875,
      "learning_rate": 3.1742497729637774e-05,
      "loss": 0.959,
      "num_input_tokens_seen": 36746784,
      "step": 63350
    },
    {
      "epoch": 9.436252606493893,
      "grad_norm": 9.1875,
      "learning_rate": 3.173936866042927e-05,
      "loss": 0.8889,
      "num_input_tokens_seen": 36749504,
      "step": 63355
    },
    {
      "epoch": 9.436997319034852,
      "grad_norm": 15.5,
      "learning_rate": 3.173623947736597e-05,
      "loss": 1.4958,
      "num_input_tokens_seen": 36752384,
      "step": 63360
    },
    {
      "epoch": 9.437742031575812,
      "grad_norm": 8.5625,
      "learning_rate": 3.1733110180500766e-05,
      "loss": 1.3637,
      "num_input_tokens_seen": 36755104,
      "step": 63365
    },
    {
      "epoch": 9.438486744116771,
      "grad_norm": 9.25,
      "learning_rate": 3.17299807698865e-05,
      "loss": 0.8846,
      "num_input_tokens_seen": 36757664,
      "step": 63370
    },
    {
      "epoch": 9.43923145665773,
      "grad_norm": 15.75,
      "learning_rate": 3.1726851245576056e-05,
      "loss": 0.9415,
      "num_input_tokens_seen": 36760704,
      "step": 63375
    },
    {
      "epoch": 9.439976169198689,
      "grad_norm": 11.375,
      "learning_rate": 3.172372160762229e-05,
      "loss": 1.027,
      "num_input_tokens_seen": 36763552,
      "step": 63380
    },
    {
      "epoch": 9.440720881739649,
      "grad_norm": 7.40625,
      "learning_rate": 3.172059185607808e-05,
      "loss": 0.9781,
      "num_input_tokens_seen": 36766176,
      "step": 63385
    },
    {
      "epoch": 9.441465594280608,
      "grad_norm": 15.375,
      "learning_rate": 3.171746199099631e-05,
      "loss": 1.2299,
      "num_input_tokens_seen": 36768960,
      "step": 63390
    },
    {
      "epoch": 9.442210306821567,
      "grad_norm": 8.9375,
      "learning_rate": 3.171433201242984e-05,
      "loss": 0.8473,
      "num_input_tokens_seen": 36771840,
      "step": 63395
    },
    {
      "epoch": 9.442955019362525,
      "grad_norm": 6.09375,
      "learning_rate": 3.1711201920431556e-05,
      "loss": 1.1156,
      "num_input_tokens_seen": 36774464,
      "step": 63400
    },
    {
      "epoch": 9.443699731903486,
      "grad_norm": 8.5,
      "learning_rate": 3.170807171505434e-05,
      "loss": 1.1153,
      "num_input_tokens_seen": 36777408,
      "step": 63405
    },
    {
      "epoch": 9.444444444444445,
      "grad_norm": 11.4375,
      "learning_rate": 3.1704941396351064e-05,
      "loss": 1.1951,
      "num_input_tokens_seen": 36780256,
      "step": 63410
    },
    {
      "epoch": 9.445189156985403,
      "grad_norm": 6.65625,
      "learning_rate": 3.1701810964374626e-05,
      "loss": 0.9825,
      "num_input_tokens_seen": 36783296,
      "step": 63415
    },
    {
      "epoch": 9.445933869526362,
      "grad_norm": 17.25,
      "learning_rate": 3.169868041917789e-05,
      "loss": 0.8792,
      "num_input_tokens_seen": 36786176,
      "step": 63420
    },
    {
      "epoch": 9.446678582067323,
      "grad_norm": 13.0625,
      "learning_rate": 3.1695549760813764e-05,
      "loss": 0.83,
      "num_input_tokens_seen": 36789152,
      "step": 63425
    },
    {
      "epoch": 9.447423294608281,
      "grad_norm": 6.0,
      "learning_rate": 3.169241898933514e-05,
      "loss": 0.8762,
      "num_input_tokens_seen": 36791840,
      "step": 63430
    },
    {
      "epoch": 9.44816800714924,
      "grad_norm": 9.1875,
      "learning_rate": 3.168928810479488e-05,
      "loss": 1.0082,
      "num_input_tokens_seen": 36794880,
      "step": 63435
    },
    {
      "epoch": 9.448912719690199,
      "grad_norm": 9.0625,
      "learning_rate": 3.1686157107245915e-05,
      "loss": 1.0675,
      "num_input_tokens_seen": 36797856,
      "step": 63440
    },
    {
      "epoch": 9.44965743223116,
      "grad_norm": 21.375,
      "learning_rate": 3.1683025996741104e-05,
      "loss": 1.4786,
      "num_input_tokens_seen": 36800640,
      "step": 63445
    },
    {
      "epoch": 9.450402144772118,
      "grad_norm": 6.0625,
      "learning_rate": 3.167989477333337e-05,
      "loss": 1.4836,
      "num_input_tokens_seen": 36803552,
      "step": 63450
    },
    {
      "epoch": 9.451146857313077,
      "grad_norm": 9.25,
      "learning_rate": 3.16767634370756e-05,
      "loss": 1.307,
      "num_input_tokens_seen": 36806432,
      "step": 63455
    },
    {
      "epoch": 9.451891569854036,
      "grad_norm": 8.0,
      "learning_rate": 3.16736319880207e-05,
      "loss": 1.0687,
      "num_input_tokens_seen": 36809536,
      "step": 63460
    },
    {
      "epoch": 9.452636282394996,
      "grad_norm": 10.125,
      "learning_rate": 3.1670500426221566e-05,
      "loss": 1.1988,
      "num_input_tokens_seen": 36812672,
      "step": 63465
    },
    {
      "epoch": 9.453380994935955,
      "grad_norm": 13.375,
      "learning_rate": 3.1667368751731116e-05,
      "loss": 1.1215,
      "num_input_tokens_seen": 36815424,
      "step": 63470
    },
    {
      "epoch": 9.454125707476914,
      "grad_norm": 9.0625,
      "learning_rate": 3.1664236964602244e-05,
      "loss": 0.8125,
      "num_input_tokens_seen": 36818304,
      "step": 63475
    },
    {
      "epoch": 9.454870420017873,
      "grad_norm": 4.15625,
      "learning_rate": 3.166110506488786e-05,
      "loss": 0.9438,
      "num_input_tokens_seen": 36821088,
      "step": 63480
    },
    {
      "epoch": 9.455615132558833,
      "grad_norm": 13.25,
      "learning_rate": 3.165797305264087e-05,
      "loss": 0.7094,
      "num_input_tokens_seen": 36823904,
      "step": 63485
    },
    {
      "epoch": 9.456359845099792,
      "grad_norm": 11.5,
      "learning_rate": 3.1654840927914196e-05,
      "loss": 1.0686,
      "num_input_tokens_seen": 36826464,
      "step": 63490
    },
    {
      "epoch": 9.45710455764075,
      "grad_norm": 9.3125,
      "learning_rate": 3.165170869076075e-05,
      "loss": 0.8652,
      "num_input_tokens_seen": 36829088,
      "step": 63495
    },
    {
      "epoch": 9.45784927018171,
      "grad_norm": 11.75,
      "learning_rate": 3.164857634123345e-05,
      "loss": 1.0111,
      "num_input_tokens_seen": 36831968,
      "step": 63500
    },
    {
      "epoch": 9.458593982722668,
      "grad_norm": 12.1875,
      "learning_rate": 3.1645443879385206e-05,
      "loss": 1.0704,
      "num_input_tokens_seen": 36835040,
      "step": 63505
    },
    {
      "epoch": 9.459338695263629,
      "grad_norm": 9.3125,
      "learning_rate": 3.164231130526894e-05,
      "loss": 0.9397,
      "num_input_tokens_seen": 36838336,
      "step": 63510
    },
    {
      "epoch": 9.460083407804587,
      "grad_norm": 12.5,
      "learning_rate": 3.163917861893758e-05,
      "loss": 1.1974,
      "num_input_tokens_seen": 36841056,
      "step": 63515
    },
    {
      "epoch": 9.460828120345546,
      "grad_norm": 12.25,
      "learning_rate": 3.1636045820444044e-05,
      "loss": 1.3878,
      "num_input_tokens_seen": 36843904,
      "step": 63520
    },
    {
      "epoch": 9.461572832886505,
      "grad_norm": 10.3125,
      "learning_rate": 3.163291290984125e-05,
      "loss": 1.0624,
      "num_input_tokens_seen": 36846752,
      "step": 63525
    },
    {
      "epoch": 9.462317545427466,
      "grad_norm": 5.3125,
      "learning_rate": 3.162977988718214e-05,
      "loss": 0.9923,
      "num_input_tokens_seen": 36849824,
      "step": 63530
    },
    {
      "epoch": 9.463062257968424,
      "grad_norm": 15.3125,
      "learning_rate": 3.162664675251965e-05,
      "loss": 1.2885,
      "num_input_tokens_seen": 36852704,
      "step": 63535
    },
    {
      "epoch": 9.463806970509383,
      "grad_norm": 22.125,
      "learning_rate": 3.162351350590668e-05,
      "loss": 1.0483,
      "num_input_tokens_seen": 36855584,
      "step": 63540
    },
    {
      "epoch": 9.464551683050342,
      "grad_norm": 12.875,
      "learning_rate": 3.1620380147396186e-05,
      "loss": 1.1105,
      "num_input_tokens_seen": 36858272,
      "step": 63545
    },
    {
      "epoch": 9.465296395591302,
      "grad_norm": 8.375,
      "learning_rate": 3.1617246677041104e-05,
      "loss": 1.0764,
      "num_input_tokens_seen": 36861728,
      "step": 63550
    },
    {
      "epoch": 9.466041108132261,
      "grad_norm": 11.25,
      "learning_rate": 3.161411309489436e-05,
      "loss": 0.8798,
      "num_input_tokens_seen": 36864896,
      "step": 63555
    },
    {
      "epoch": 9.46678582067322,
      "grad_norm": 11.125,
      "learning_rate": 3.161097940100889e-05,
      "loss": 1.076,
      "num_input_tokens_seen": 36867872,
      "step": 63560
    },
    {
      "epoch": 9.467530533214179,
      "grad_norm": 3.875,
      "learning_rate": 3.160784559543765e-05,
      "loss": 0.9245,
      "num_input_tokens_seen": 36870944,
      "step": 63565
    },
    {
      "epoch": 9.46827524575514,
      "grad_norm": 9.75,
      "learning_rate": 3.160471167823358e-05,
      "loss": 1.0751,
      "num_input_tokens_seen": 36873856,
      "step": 63570
    },
    {
      "epoch": 9.469019958296098,
      "grad_norm": 16.75,
      "learning_rate": 3.1601577649449606e-05,
      "loss": 1.2221,
      "num_input_tokens_seen": 36876384,
      "step": 63575
    },
    {
      "epoch": 9.469764670837057,
      "grad_norm": 14.5,
      "learning_rate": 3.15984435091387e-05,
      "loss": 0.7714,
      "num_input_tokens_seen": 36878976,
      "step": 63580
    },
    {
      "epoch": 9.470509383378015,
      "grad_norm": 8.5,
      "learning_rate": 3.159530925735379e-05,
      "loss": 0.767,
      "num_input_tokens_seen": 36881920,
      "step": 63585
    },
    {
      "epoch": 9.471254095918976,
      "grad_norm": 9.125,
      "learning_rate": 3.1592174894147835e-05,
      "loss": 0.8628,
      "num_input_tokens_seen": 36884736,
      "step": 63590
    },
    {
      "epoch": 9.471998808459935,
      "grad_norm": 8.1875,
      "learning_rate": 3.158904041957379e-05,
      "loss": 1.2181,
      "num_input_tokens_seen": 36887488,
      "step": 63595
    },
    {
      "epoch": 9.472743521000893,
      "grad_norm": 16.125,
      "learning_rate": 3.1585905833684595e-05,
      "loss": 1.1137,
      "num_input_tokens_seen": 36890752,
      "step": 63600
    },
    {
      "epoch": 9.473488233541852,
      "grad_norm": 14.5,
      "learning_rate": 3.158277113653322e-05,
      "loss": 1.1108,
      "num_input_tokens_seen": 36893504,
      "step": 63605
    },
    {
      "epoch": 9.474232946082813,
      "grad_norm": 4.59375,
      "learning_rate": 3.157963632817261e-05,
      "loss": 1.2513,
      "num_input_tokens_seen": 36896192,
      "step": 63610
    },
    {
      "epoch": 9.474977658623772,
      "grad_norm": 8.9375,
      "learning_rate": 3.157650140865574e-05,
      "loss": 1.0996,
      "num_input_tokens_seen": 36899072,
      "step": 63615
    },
    {
      "epoch": 9.47572237116473,
      "grad_norm": 13.125,
      "learning_rate": 3.157336637803556e-05,
      "loss": 1.3014,
      "num_input_tokens_seen": 36901888,
      "step": 63620
    },
    {
      "epoch": 9.476467083705689,
      "grad_norm": 5.8125,
      "learning_rate": 3.1570231236365035e-05,
      "loss": 0.9183,
      "num_input_tokens_seen": 36904544,
      "step": 63625
    },
    {
      "epoch": 9.47721179624665,
      "grad_norm": 8.1875,
      "learning_rate": 3.156709598369713e-05,
      "loss": 1.1487,
      "num_input_tokens_seen": 36907584,
      "step": 63630
    },
    {
      "epoch": 9.477956508787608,
      "grad_norm": 13.5625,
      "learning_rate": 3.1563960620084816e-05,
      "loss": 1.3776,
      "num_input_tokens_seen": 36910336,
      "step": 63635
    },
    {
      "epoch": 9.478701221328567,
      "grad_norm": 12.3125,
      "learning_rate": 3.1560825145581056e-05,
      "loss": 1.0598,
      "num_input_tokens_seen": 36912928,
      "step": 63640
    },
    {
      "epoch": 9.479445933869526,
      "grad_norm": 11.875,
      "learning_rate": 3.155768956023882e-05,
      "loss": 1.1076,
      "num_input_tokens_seen": 36915744,
      "step": 63645
    },
    {
      "epoch": 9.480190646410486,
      "grad_norm": 7.28125,
      "learning_rate": 3.155455386411109e-05,
      "loss": 1.1965,
      "num_input_tokens_seen": 36918496,
      "step": 63650
    },
    {
      "epoch": 9.480935358951445,
      "grad_norm": 6.84375,
      "learning_rate": 3.1551418057250835e-05,
      "loss": 0.8189,
      "num_input_tokens_seen": 36921664,
      "step": 63655
    },
    {
      "epoch": 9.481680071492404,
      "grad_norm": 11.375,
      "learning_rate": 3.1548282139711025e-05,
      "loss": 0.9697,
      "num_input_tokens_seen": 36924608,
      "step": 63660
    },
    {
      "epoch": 9.482424784033363,
      "grad_norm": 9.9375,
      "learning_rate": 3.154514611154464e-05,
      "loss": 1.1313,
      "num_input_tokens_seen": 36927680,
      "step": 63665
    },
    {
      "epoch": 9.483169496574323,
      "grad_norm": 11.0,
      "learning_rate": 3.154200997280468e-05,
      "loss": 1.2777,
      "num_input_tokens_seen": 36930528,
      "step": 63670
    },
    {
      "epoch": 9.483914209115282,
      "grad_norm": 9.6875,
      "learning_rate": 3.1538873723544105e-05,
      "loss": 0.843,
      "num_input_tokens_seen": 36933440,
      "step": 63675
    },
    {
      "epoch": 9.48465892165624,
      "grad_norm": 14.1875,
      "learning_rate": 3.1535737363815896e-05,
      "loss": 1.1348,
      "num_input_tokens_seen": 36936000,
      "step": 63680
    },
    {
      "epoch": 9.4854036341972,
      "grad_norm": 7.34375,
      "learning_rate": 3.1532600893673045e-05,
      "loss": 0.9664,
      "num_input_tokens_seen": 36939424,
      "step": 63685
    },
    {
      "epoch": 9.486148346738158,
      "grad_norm": 7.5,
      "learning_rate": 3.152946431316855e-05,
      "loss": 0.9564,
      "num_input_tokens_seen": 36942240,
      "step": 63690
    },
    {
      "epoch": 9.486893059279119,
      "grad_norm": 5.59375,
      "learning_rate": 3.152632762235539e-05,
      "loss": 1.0419,
      "num_input_tokens_seen": 36944960,
      "step": 63695
    },
    {
      "epoch": 9.487637771820078,
      "grad_norm": 6.46875,
      "learning_rate": 3.152319082128656e-05,
      "loss": 0.9966,
      "num_input_tokens_seen": 36947904,
      "step": 63700
    },
    {
      "epoch": 9.488382484361036,
      "grad_norm": 6.28125,
      "learning_rate": 3.1520053910015046e-05,
      "loss": 0.8367,
      "num_input_tokens_seen": 36950912,
      "step": 63705
    },
    {
      "epoch": 9.489127196901995,
      "grad_norm": 7.6875,
      "learning_rate": 3.151691688859385e-05,
      "loss": 0.6955,
      "num_input_tokens_seen": 36953568,
      "step": 63710
    },
    {
      "epoch": 9.489871909442956,
      "grad_norm": 7.84375,
      "learning_rate": 3.151377975707597e-05,
      "loss": 1.0032,
      "num_input_tokens_seen": 36956672,
      "step": 63715
    },
    {
      "epoch": 9.490616621983914,
      "grad_norm": 9.375,
      "learning_rate": 3.1510642515514393e-05,
      "loss": 0.986,
      "num_input_tokens_seen": 36959776,
      "step": 63720
    },
    {
      "epoch": 9.491361334524873,
      "grad_norm": 3.796875,
      "learning_rate": 3.150750516396213e-05,
      "loss": 1.2493,
      "num_input_tokens_seen": 36962592,
      "step": 63725
    },
    {
      "epoch": 9.492106047065832,
      "grad_norm": 5.71875,
      "learning_rate": 3.1504367702472185e-05,
      "loss": 1.0514,
      "num_input_tokens_seen": 36965216,
      "step": 63730
    },
    {
      "epoch": 9.492850759606792,
      "grad_norm": 15.125,
      "learning_rate": 3.150123013109756e-05,
      "loss": 1.2728,
      "num_input_tokens_seen": 36968064,
      "step": 63735
    },
    {
      "epoch": 9.493595472147751,
      "grad_norm": 14.625,
      "learning_rate": 3.149809244989125e-05,
      "loss": 1.3351,
      "num_input_tokens_seen": 36970784,
      "step": 63740
    },
    {
      "epoch": 9.49434018468871,
      "grad_norm": 11.5,
      "learning_rate": 3.149495465890628e-05,
      "loss": 1.3588,
      "num_input_tokens_seen": 36973504,
      "step": 63745
    },
    {
      "epoch": 9.495084897229669,
      "grad_norm": 12.25,
      "learning_rate": 3.149181675819565e-05,
      "loss": 1.0249,
      "num_input_tokens_seen": 36976256,
      "step": 63750
    },
    {
      "epoch": 9.49582960977063,
      "grad_norm": 15.25,
      "learning_rate": 3.148867874781238e-05,
      "loss": 1.0414,
      "num_input_tokens_seen": 36979424,
      "step": 63755
    },
    {
      "epoch": 9.496574322311588,
      "grad_norm": 16.125,
      "learning_rate": 3.148554062780947e-05,
      "loss": 1.22,
      "num_input_tokens_seen": 36982464,
      "step": 63760
    },
    {
      "epoch": 9.497319034852547,
      "grad_norm": 6.5625,
      "learning_rate": 3.148240239823994e-05,
      "loss": 1.2838,
      "num_input_tokens_seen": 36984992,
      "step": 63765
    },
    {
      "epoch": 9.498063747393505,
      "grad_norm": 9.875,
      "learning_rate": 3.147926405915682e-05,
      "loss": 1.2969,
      "num_input_tokens_seen": 36988160,
      "step": 63770
    },
    {
      "epoch": 9.498808459934466,
      "grad_norm": 7.6875,
      "learning_rate": 3.147612561061312e-05,
      "loss": 1.1487,
      "num_input_tokens_seen": 36990816,
      "step": 63775
    },
    {
      "epoch": 9.499553172475425,
      "grad_norm": 20.375,
      "learning_rate": 3.147298705266185e-05,
      "loss": 0.9029,
      "num_input_tokens_seen": 36993408,
      "step": 63780
    },
    {
      "epoch": 9.500297885016384,
      "grad_norm": 4.59375,
      "learning_rate": 3.146984838535604e-05,
      "loss": 1.0404,
      "num_input_tokens_seen": 36995968,
      "step": 63785
    },
    {
      "epoch": 9.501042597557342,
      "grad_norm": 10.125,
      "learning_rate": 3.146670960874872e-05,
      "loss": 1.0348,
      "num_input_tokens_seen": 36998784,
      "step": 63790
    },
    {
      "epoch": 9.501787310098303,
      "grad_norm": 12.25,
      "learning_rate": 3.146357072289292e-05,
      "loss": 0.9331,
      "num_input_tokens_seen": 37001888,
      "step": 63795
    },
    {
      "epoch": 9.502532022639262,
      "grad_norm": 11.9375,
      "learning_rate": 3.146043172784166e-05,
      "loss": 1.6347,
      "num_input_tokens_seen": 37004576,
      "step": 63800
    },
    {
      "epoch": 9.50327673518022,
      "grad_norm": 4.75,
      "learning_rate": 3.1457292623647976e-05,
      "loss": 0.7973,
      "num_input_tokens_seen": 37007392,
      "step": 63805
    },
    {
      "epoch": 9.504021447721179,
      "grad_norm": 12.625,
      "learning_rate": 3.145415341036489e-05,
      "loss": 1.1462,
      "num_input_tokens_seen": 37010240,
      "step": 63810
    },
    {
      "epoch": 9.50476616026214,
      "grad_norm": 3.9375,
      "learning_rate": 3.1451014088045435e-05,
      "loss": 1.0318,
      "num_input_tokens_seen": 37012928,
      "step": 63815
    },
    {
      "epoch": 9.505510872803098,
      "grad_norm": 10.8125,
      "learning_rate": 3.144787465674266e-05,
      "loss": 0.9472,
      "num_input_tokens_seen": 37015584,
      "step": 63820
    },
    {
      "epoch": 9.506255585344057,
      "grad_norm": 11.5,
      "learning_rate": 3.14447351165096e-05,
      "loss": 1.254,
      "num_input_tokens_seen": 37018368,
      "step": 63825
    },
    {
      "epoch": 9.507000297885016,
      "grad_norm": 12.75,
      "learning_rate": 3.1441595467399286e-05,
      "loss": 1.068,
      "num_input_tokens_seen": 37021152,
      "step": 63830
    },
    {
      "epoch": 9.507745010425975,
      "grad_norm": 8.3125,
      "learning_rate": 3.143845570946477e-05,
      "loss": 0.9348,
      "num_input_tokens_seen": 37024032,
      "step": 63835
    },
    {
      "epoch": 9.508489722966935,
      "grad_norm": 10.0,
      "learning_rate": 3.143531584275909e-05,
      "loss": 0.9408,
      "num_input_tokens_seen": 37026944,
      "step": 63840
    },
    {
      "epoch": 9.509234435507894,
      "grad_norm": 8.25,
      "learning_rate": 3.1432175867335275e-05,
      "loss": 0.8528,
      "num_input_tokens_seen": 37029920,
      "step": 63845
    },
    {
      "epoch": 9.509979148048853,
      "grad_norm": 6.28125,
      "learning_rate": 3.1429035783246395e-05,
      "loss": 0.9659,
      "num_input_tokens_seen": 37032608,
      "step": 63850
    },
    {
      "epoch": 9.510723860589813,
      "grad_norm": 15.3125,
      "learning_rate": 3.142589559054549e-05,
      "loss": 1.4146,
      "num_input_tokens_seen": 37035360,
      "step": 63855
    },
    {
      "epoch": 9.511468573130772,
      "grad_norm": 6.375,
      "learning_rate": 3.142275528928561e-05,
      "loss": 1.1056,
      "num_input_tokens_seen": 37038048,
      "step": 63860
    },
    {
      "epoch": 9.51221328567173,
      "grad_norm": 14.75,
      "learning_rate": 3.141961487951981e-05,
      "loss": 1.4274,
      "num_input_tokens_seen": 37040544,
      "step": 63865
    },
    {
      "epoch": 9.51295799821269,
      "grad_norm": 4.65625,
      "learning_rate": 3.141647436130113e-05,
      "loss": 1.1213,
      "num_input_tokens_seen": 37043264,
      "step": 63870
    },
    {
      "epoch": 9.513702710753648,
      "grad_norm": 16.75,
      "learning_rate": 3.1413333734682656e-05,
      "loss": 0.8638,
      "num_input_tokens_seen": 37046240,
      "step": 63875
    },
    {
      "epoch": 9.514447423294609,
      "grad_norm": 12.25,
      "learning_rate": 3.141019299971741e-05,
      "loss": 0.8049,
      "num_input_tokens_seen": 37049408,
      "step": 63880
    },
    {
      "epoch": 9.515192135835568,
      "grad_norm": 16.5,
      "learning_rate": 3.140705215645847e-05,
      "loss": 1.0718,
      "num_input_tokens_seen": 37052448,
      "step": 63885
    },
    {
      "epoch": 9.515936848376526,
      "grad_norm": 23.0,
      "learning_rate": 3.14039112049589e-05,
      "loss": 1.1713,
      "num_input_tokens_seen": 37055104,
      "step": 63890
    },
    {
      "epoch": 9.516681560917485,
      "grad_norm": 11.5,
      "learning_rate": 3.140077014527176e-05,
      "loss": 1.1123,
      "num_input_tokens_seen": 37057792,
      "step": 63895
    },
    {
      "epoch": 9.517426273458446,
      "grad_norm": 17.5,
      "learning_rate": 3.139762897745011e-05,
      "loss": 1.1455,
      "num_input_tokens_seen": 37060896,
      "step": 63900
    },
    {
      "epoch": 9.518170985999404,
      "grad_norm": 11.8125,
      "learning_rate": 3.139448770154702e-05,
      "loss": 1.123,
      "num_input_tokens_seen": 37063744,
      "step": 63905
    },
    {
      "epoch": 9.518915698540363,
      "grad_norm": 14.6875,
      "learning_rate": 3.139134631761557e-05,
      "loss": 1.0313,
      "num_input_tokens_seen": 37066624,
      "step": 63910
    },
    {
      "epoch": 9.519660411081322,
      "grad_norm": 13.4375,
      "learning_rate": 3.1388204825708815e-05,
      "loss": 1.1115,
      "num_input_tokens_seen": 37069216,
      "step": 63915
    },
    {
      "epoch": 9.520405123622282,
      "grad_norm": 13.5625,
      "learning_rate": 3.138506322587982e-05,
      "loss": 1.0625,
      "num_input_tokens_seen": 37071936,
      "step": 63920
    },
    {
      "epoch": 9.521149836163241,
      "grad_norm": 7.21875,
      "learning_rate": 3.138192151818168e-05,
      "loss": 1.0739,
      "num_input_tokens_seen": 37075072,
      "step": 63925
    },
    {
      "epoch": 9.5218945487042,
      "grad_norm": 8.3125,
      "learning_rate": 3.137877970266746e-05,
      "loss": 1.0024,
      "num_input_tokens_seen": 37077952,
      "step": 63930
    },
    {
      "epoch": 9.522639261245159,
      "grad_norm": 14.6875,
      "learning_rate": 3.1375637779390244e-05,
      "loss": 1.369,
      "num_input_tokens_seen": 37081024,
      "step": 63935
    },
    {
      "epoch": 9.52338397378612,
      "grad_norm": 8.875,
      "learning_rate": 3.137249574840311e-05,
      "loss": 0.9715,
      "num_input_tokens_seen": 37083936,
      "step": 63940
    },
    {
      "epoch": 9.524128686327078,
      "grad_norm": 6.375,
      "learning_rate": 3.136935360975913e-05,
      "loss": 0.8975,
      "num_input_tokens_seen": 37086592,
      "step": 63945
    },
    {
      "epoch": 9.524873398868037,
      "grad_norm": 14.375,
      "learning_rate": 3.1366211363511394e-05,
      "loss": 1.042,
      "num_input_tokens_seen": 37089120,
      "step": 63950
    },
    {
      "epoch": 9.525618111408996,
      "grad_norm": 13.75,
      "learning_rate": 3.1363069009712994e-05,
      "loss": 1.3237,
      "num_input_tokens_seen": 37092352,
      "step": 63955
    },
    {
      "epoch": 9.526362823949956,
      "grad_norm": 6.5,
      "learning_rate": 3.1359926548417007e-05,
      "loss": 1.0104,
      "num_input_tokens_seen": 37095424,
      "step": 63960
    },
    {
      "epoch": 9.527107536490915,
      "grad_norm": 12.4375,
      "learning_rate": 3.135678397967652e-05,
      "loss": 1.2386,
      "num_input_tokens_seen": 37098304,
      "step": 63965
    },
    {
      "epoch": 9.527852249031874,
      "grad_norm": 15.625,
      "learning_rate": 3.135364130354464e-05,
      "loss": 0.9192,
      "num_input_tokens_seen": 37101088,
      "step": 63970
    },
    {
      "epoch": 9.528596961572832,
      "grad_norm": 7.53125,
      "learning_rate": 3.135049852007444e-05,
      "loss": 1.142,
      "num_input_tokens_seen": 37103744,
      "step": 63975
    },
    {
      "epoch": 9.529341674113793,
      "grad_norm": 10.625,
      "learning_rate": 3.134735562931902e-05,
      "loss": 0.8508,
      "num_input_tokens_seen": 37106528,
      "step": 63980
    },
    {
      "epoch": 9.530086386654752,
      "grad_norm": 10.375,
      "learning_rate": 3.1344212631331484e-05,
      "loss": 1.1264,
      "num_input_tokens_seen": 37109632,
      "step": 63985
    },
    {
      "epoch": 9.53083109919571,
      "grad_norm": 11.5625,
      "learning_rate": 3.134106952616491e-05,
      "loss": 0.9526,
      "num_input_tokens_seen": 37112384,
      "step": 63990
    },
    {
      "epoch": 9.53157581173667,
      "grad_norm": 8.875,
      "learning_rate": 3.133792631387243e-05,
      "loss": 1.0248,
      "num_input_tokens_seen": 37115136,
      "step": 63995
    },
    {
      "epoch": 9.53232052427763,
      "grad_norm": 7.75,
      "learning_rate": 3.133478299450712e-05,
      "loss": 1.1254,
      "num_input_tokens_seen": 37117920,
      "step": 64000
    },
    {
      "epoch": 9.533065236818588,
      "grad_norm": 7.65625,
      "learning_rate": 3.1331639568122084e-05,
      "loss": 1.2266,
      "num_input_tokens_seen": 37121024,
      "step": 64005
    },
    {
      "epoch": 9.533809949359547,
      "grad_norm": 14.75,
      "learning_rate": 3.132849603477044e-05,
      "loss": 1.1897,
      "num_input_tokens_seen": 37123936,
      "step": 64010
    },
    {
      "epoch": 9.534554661900506,
      "grad_norm": 11.75,
      "learning_rate": 3.132535239450528e-05,
      "loss": 1.0727,
      "num_input_tokens_seen": 37126944,
      "step": 64015
    },
    {
      "epoch": 9.535299374441465,
      "grad_norm": 7.5625,
      "learning_rate": 3.1322208647379724e-05,
      "loss": 1.4459,
      "num_input_tokens_seen": 37129888,
      "step": 64020
    },
    {
      "epoch": 9.536044086982425,
      "grad_norm": 7.03125,
      "learning_rate": 3.1319064793446876e-05,
      "loss": 0.9578,
      "num_input_tokens_seen": 37132768,
      "step": 64025
    },
    {
      "epoch": 9.536788799523384,
      "grad_norm": 8.75,
      "learning_rate": 3.131592083275986e-05,
      "loss": 1.0275,
      "num_input_tokens_seen": 37135584,
      "step": 64030
    },
    {
      "epoch": 9.537533512064343,
      "grad_norm": 10.375,
      "learning_rate": 3.1312776765371765e-05,
      "loss": 1.1263,
      "num_input_tokens_seen": 37138560,
      "step": 64035
    },
    {
      "epoch": 9.538278224605303,
      "grad_norm": 14.4375,
      "learning_rate": 3.1309632591335734e-05,
      "loss": 1.3972,
      "num_input_tokens_seen": 37141440,
      "step": 64040
    },
    {
      "epoch": 9.539022937146262,
      "grad_norm": 9.4375,
      "learning_rate": 3.1306488310704875e-05,
      "loss": 0.7631,
      "num_input_tokens_seen": 37144384,
      "step": 64045
    },
    {
      "epoch": 9.53976764968722,
      "grad_norm": 8.1875,
      "learning_rate": 3.1303343923532294e-05,
      "loss": 0.8457,
      "num_input_tokens_seen": 37147744,
      "step": 64050
    },
    {
      "epoch": 9.54051236222818,
      "grad_norm": 7.6875,
      "learning_rate": 3.130019942987114e-05,
      "loss": 1.0861,
      "num_input_tokens_seen": 37150624,
      "step": 64055
    },
    {
      "epoch": 9.541257074769138,
      "grad_norm": 13.9375,
      "learning_rate": 3.1297054829774505e-05,
      "loss": 1.0807,
      "num_input_tokens_seen": 37153600,
      "step": 64060
    },
    {
      "epoch": 9.542001787310099,
      "grad_norm": 11.25,
      "learning_rate": 3.1293910123295535e-05,
      "loss": 1.2365,
      "num_input_tokens_seen": 37156768,
      "step": 64065
    },
    {
      "epoch": 9.542746499851058,
      "grad_norm": 17.75,
      "learning_rate": 3.1290765310487346e-05,
      "loss": 1.2563,
      "num_input_tokens_seen": 37159552,
      "step": 64070
    },
    {
      "epoch": 9.543491212392016,
      "grad_norm": 15.6875,
      "learning_rate": 3.1287620391403086e-05,
      "loss": 1.045,
      "num_input_tokens_seen": 37162592,
      "step": 64075
    },
    {
      "epoch": 9.544235924932975,
      "grad_norm": 10.5,
      "learning_rate": 3.128447536609585e-05,
      "loss": 0.8684,
      "num_input_tokens_seen": 37165344,
      "step": 64080
    },
    {
      "epoch": 9.544980637473936,
      "grad_norm": 19.25,
      "learning_rate": 3.12813302346188e-05,
      "loss": 1.3393,
      "num_input_tokens_seen": 37168192,
      "step": 64085
    },
    {
      "epoch": 9.545725350014894,
      "grad_norm": 19.25,
      "learning_rate": 3.127818499702506e-05,
      "loss": 0.988,
      "num_input_tokens_seen": 37171424,
      "step": 64090
    },
    {
      "epoch": 9.546470062555853,
      "grad_norm": 9.4375,
      "learning_rate": 3.127503965336776e-05,
      "loss": 0.9355,
      "num_input_tokens_seen": 37174240,
      "step": 64095
    },
    {
      "epoch": 9.547214775096812,
      "grad_norm": 6.375,
      "learning_rate": 3.1271894203700045e-05,
      "loss": 1.1039,
      "num_input_tokens_seen": 37177024,
      "step": 64100
    },
    {
      "epoch": 9.547959487637772,
      "grad_norm": 8.375,
      "learning_rate": 3.126874864807505e-05,
      "loss": 1.006,
      "num_input_tokens_seen": 37179776,
      "step": 64105
    },
    {
      "epoch": 9.548704200178731,
      "grad_norm": 13.375,
      "learning_rate": 3.126560298654593e-05,
      "loss": 0.9721,
      "num_input_tokens_seen": 37182592,
      "step": 64110
    },
    {
      "epoch": 9.54944891271969,
      "grad_norm": 6.875,
      "learning_rate": 3.126245721916581e-05,
      "loss": 0.8896,
      "num_input_tokens_seen": 37185696,
      "step": 64115
    },
    {
      "epoch": 9.550193625260649,
      "grad_norm": 13.5,
      "learning_rate": 3.125931134598783e-05,
      "loss": 1.291,
      "num_input_tokens_seen": 37188160,
      "step": 64120
    },
    {
      "epoch": 9.55093833780161,
      "grad_norm": 12.4375,
      "learning_rate": 3.1256165367065155e-05,
      "loss": 1.0079,
      "num_input_tokens_seen": 37191040,
      "step": 64125
    },
    {
      "epoch": 9.551683050342568,
      "grad_norm": 13.9375,
      "learning_rate": 3.125301928245092e-05,
      "loss": 1.0412,
      "num_input_tokens_seen": 37193984,
      "step": 64130
    },
    {
      "epoch": 9.552427762883527,
      "grad_norm": 9.0,
      "learning_rate": 3.124987309219828e-05,
      "loss": 1.4197,
      "num_input_tokens_seen": 37196864,
      "step": 64135
    },
    {
      "epoch": 9.553172475424486,
      "grad_norm": 9.4375,
      "learning_rate": 3.124672679636039e-05,
      "loss": 0.9448,
      "num_input_tokens_seen": 37199776,
      "step": 64140
    },
    {
      "epoch": 9.553917187965446,
      "grad_norm": 11.0,
      "learning_rate": 3.12435803949904e-05,
      "loss": 0.9704,
      "num_input_tokens_seen": 37202656,
      "step": 64145
    },
    {
      "epoch": 9.554661900506405,
      "grad_norm": 9.5625,
      "learning_rate": 3.1240433888141466e-05,
      "loss": 1.3222,
      "num_input_tokens_seen": 37205408,
      "step": 64150
    },
    {
      "epoch": 9.555406613047364,
      "grad_norm": 14.5,
      "learning_rate": 3.1237287275866736e-05,
      "loss": 1.1505,
      "num_input_tokens_seen": 37208576,
      "step": 64155
    },
    {
      "epoch": 9.556151325588322,
      "grad_norm": 11.6875,
      "learning_rate": 3.123414055821938e-05,
      "loss": 1.3025,
      "num_input_tokens_seen": 37211360,
      "step": 64160
    },
    {
      "epoch": 9.556896038129283,
      "grad_norm": 6.65625,
      "learning_rate": 3.1230993735252564e-05,
      "loss": 1.1832,
      "num_input_tokens_seen": 37214400,
      "step": 64165
    },
    {
      "epoch": 9.557640750670242,
      "grad_norm": 9.0,
      "learning_rate": 3.1227846807019435e-05,
      "loss": 1.0616,
      "num_input_tokens_seen": 37217376,
      "step": 64170
    },
    {
      "epoch": 9.5583854632112,
      "grad_norm": 10.6875,
      "learning_rate": 3.1224699773573164e-05,
      "loss": 1.2608,
      "num_input_tokens_seen": 37220224,
      "step": 64175
    },
    {
      "epoch": 9.55913017575216,
      "grad_norm": 3.203125,
      "learning_rate": 3.1221552634966914e-05,
      "loss": 1.0753,
      "num_input_tokens_seen": 37223200,
      "step": 64180
    },
    {
      "epoch": 9.55987488829312,
      "grad_norm": 11.875,
      "learning_rate": 3.1218405391253856e-05,
      "loss": 0.8388,
      "num_input_tokens_seen": 37225888,
      "step": 64185
    },
    {
      "epoch": 9.560619600834078,
      "grad_norm": 10.6875,
      "learning_rate": 3.121525804248716e-05,
      "loss": 1.2643,
      "num_input_tokens_seen": 37228832,
      "step": 64190
    },
    {
      "epoch": 9.561364313375037,
      "grad_norm": 11.125,
      "learning_rate": 3.1212110588720004e-05,
      "loss": 1.0389,
      "num_input_tokens_seen": 37231680,
      "step": 64195
    },
    {
      "epoch": 9.562109025915996,
      "grad_norm": 9.875,
      "learning_rate": 3.1208963030005543e-05,
      "loss": 1.4077,
      "num_input_tokens_seen": 37234432,
      "step": 64200
    },
    {
      "epoch": 9.562853738456955,
      "grad_norm": 5.84375,
      "learning_rate": 3.120581536639697e-05,
      "loss": 0.921,
      "num_input_tokens_seen": 37237216,
      "step": 64205
    },
    {
      "epoch": 9.563598450997915,
      "grad_norm": 10.75,
      "learning_rate": 3.120266759794745e-05,
      "loss": 0.8515,
      "num_input_tokens_seen": 37239840,
      "step": 64210
    },
    {
      "epoch": 9.564343163538874,
      "grad_norm": 14.5625,
      "learning_rate": 3.119951972471016e-05,
      "loss": 1.2828,
      "num_input_tokens_seen": 37242752,
      "step": 64215
    },
    {
      "epoch": 9.565087876079833,
      "grad_norm": 14.1875,
      "learning_rate": 3.119637174673829e-05,
      "loss": 1.502,
      "num_input_tokens_seen": 37246048,
      "step": 64220
    },
    {
      "epoch": 9.565832588620792,
      "grad_norm": 10.0,
      "learning_rate": 3.119322366408501e-05,
      "loss": 1.0396,
      "num_input_tokens_seen": 37249024,
      "step": 64225
    },
    {
      "epoch": 9.566577301161752,
      "grad_norm": 17.75,
      "learning_rate": 3.119007547680353e-05,
      "loss": 1.3381,
      "num_input_tokens_seen": 37251808,
      "step": 64230
    },
    {
      "epoch": 9.56732201370271,
      "grad_norm": 11.625,
      "learning_rate": 3.1186927184947e-05,
      "loss": 1.0984,
      "num_input_tokens_seen": 37254784,
      "step": 64235
    },
    {
      "epoch": 9.56806672624367,
      "grad_norm": 10.3125,
      "learning_rate": 3.118377878856863e-05,
      "loss": 0.9685,
      "num_input_tokens_seen": 37257760,
      "step": 64240
    },
    {
      "epoch": 9.568811438784628,
      "grad_norm": 11.1875,
      "learning_rate": 3.1180630287721595e-05,
      "loss": 1.4664,
      "num_input_tokens_seen": 37260640,
      "step": 64245
    },
    {
      "epoch": 9.569556151325589,
      "grad_norm": 12.8125,
      "learning_rate": 3.11774816824591e-05,
      "loss": 0.8223,
      "num_input_tokens_seen": 37263648,
      "step": 64250
    },
    {
      "epoch": 9.570300863866548,
      "grad_norm": 16.25,
      "learning_rate": 3.1174332972834326e-05,
      "loss": 0.9781,
      "num_input_tokens_seen": 37266624,
      "step": 64255
    },
    {
      "epoch": 9.571045576407506,
      "grad_norm": 13.0,
      "learning_rate": 3.117118415890047e-05,
      "loss": 1.0469,
      "num_input_tokens_seen": 37269568,
      "step": 64260
    },
    {
      "epoch": 9.571790288948465,
      "grad_norm": 8.3125,
      "learning_rate": 3.116803524071074e-05,
      "loss": 1.1177,
      "num_input_tokens_seen": 37272480,
      "step": 64265
    },
    {
      "epoch": 9.572535001489426,
      "grad_norm": 10.875,
      "learning_rate": 3.116488621831831e-05,
      "loss": 1.2345,
      "num_input_tokens_seen": 37275200,
      "step": 64270
    },
    {
      "epoch": 9.573279714030384,
      "grad_norm": 8.5625,
      "learning_rate": 3.1161737091776404e-05,
      "loss": 1.103,
      "num_input_tokens_seen": 37278208,
      "step": 64275
    },
    {
      "epoch": 9.574024426571343,
      "grad_norm": 7.3125,
      "learning_rate": 3.115858786113821e-05,
      "loss": 1.1225,
      "num_input_tokens_seen": 37281280,
      "step": 64280
    },
    {
      "epoch": 9.574769139112302,
      "grad_norm": 9.25,
      "learning_rate": 3.115543852645693e-05,
      "loss": 1.4167,
      "num_input_tokens_seen": 37284224,
      "step": 64285
    },
    {
      "epoch": 9.575513851653263,
      "grad_norm": 16.625,
      "learning_rate": 3.1152289087785776e-05,
      "loss": 1.1903,
      "num_input_tokens_seen": 37287808,
      "step": 64290
    },
    {
      "epoch": 9.576258564194221,
      "grad_norm": 11.5,
      "learning_rate": 3.114913954517794e-05,
      "loss": 0.8191,
      "num_input_tokens_seen": 37290336,
      "step": 64295
    },
    {
      "epoch": 9.57700327673518,
      "grad_norm": 13.125,
      "learning_rate": 3.1145989898686656e-05,
      "loss": 1.347,
      "num_input_tokens_seen": 37293344,
      "step": 64300
    },
    {
      "epoch": 9.577747989276139,
      "grad_norm": 7.90625,
      "learning_rate": 3.114284014836512e-05,
      "loss": 1.0779,
      "num_input_tokens_seen": 37296192,
      "step": 64305
    },
    {
      "epoch": 9.5784927018171,
      "grad_norm": 12.6875,
      "learning_rate": 3.1139690294266526e-05,
      "loss": 1.0997,
      "num_input_tokens_seen": 37298784,
      "step": 64310
    },
    {
      "epoch": 9.579237414358058,
      "grad_norm": 7.34375,
      "learning_rate": 3.1136540336444114e-05,
      "loss": 0.9394,
      "num_input_tokens_seen": 37301824,
      "step": 64315
    },
    {
      "epoch": 9.579982126899017,
      "grad_norm": 6.4375,
      "learning_rate": 3.11333902749511e-05,
      "loss": 0.966,
      "num_input_tokens_seen": 37304640,
      "step": 64320
    },
    {
      "epoch": 9.580726839439976,
      "grad_norm": 8.3125,
      "learning_rate": 3.1130240109840676e-05,
      "loss": 1.0585,
      "num_input_tokens_seen": 37307328,
      "step": 64325
    },
    {
      "epoch": 9.581471551980936,
      "grad_norm": 13.5,
      "learning_rate": 3.112708984116608e-05,
      "loss": 1.0765,
      "num_input_tokens_seen": 37310464,
      "step": 64330
    },
    {
      "epoch": 9.582216264521895,
      "grad_norm": 16.5,
      "learning_rate": 3.1123939468980535e-05,
      "loss": 1.1533,
      "num_input_tokens_seen": 37313408,
      "step": 64335
    },
    {
      "epoch": 9.582960977062854,
      "grad_norm": 6.9375,
      "learning_rate": 3.1120788993337244e-05,
      "loss": 1.0029,
      "num_input_tokens_seen": 37316288,
      "step": 64340
    },
    {
      "epoch": 9.583705689603812,
      "grad_norm": 11.75,
      "learning_rate": 3.111763841428945e-05,
      "loss": 1.3655,
      "num_input_tokens_seen": 37318816,
      "step": 64345
    },
    {
      "epoch": 9.584450402144771,
      "grad_norm": 10.5625,
      "learning_rate": 3.111448773189037e-05,
      "loss": 0.8781,
      "num_input_tokens_seen": 37322080,
      "step": 64350
    },
    {
      "epoch": 9.585195114685732,
      "grad_norm": 12.875,
      "learning_rate": 3.1111336946193244e-05,
      "loss": 1.0918,
      "num_input_tokens_seen": 37324800,
      "step": 64355
    },
    {
      "epoch": 9.58593982722669,
      "grad_norm": 7.3125,
      "learning_rate": 3.1108186057251285e-05,
      "loss": 1.1289,
      "num_input_tokens_seen": 37327552,
      "step": 64360
    },
    {
      "epoch": 9.58668453976765,
      "grad_norm": 12.375,
      "learning_rate": 3.1105035065117735e-05,
      "loss": 0.9583,
      "num_input_tokens_seen": 37330304,
      "step": 64365
    },
    {
      "epoch": 9.58742925230861,
      "grad_norm": 10.3125,
      "learning_rate": 3.110188396984582e-05,
      "loss": 1.4806,
      "num_input_tokens_seen": 37333056,
      "step": 64370
    },
    {
      "epoch": 9.588173964849569,
      "grad_norm": 11.875,
      "learning_rate": 3.109873277148878e-05,
      "loss": 1.1151,
      "num_input_tokens_seen": 37335840,
      "step": 64375
    },
    {
      "epoch": 9.588918677390527,
      "grad_norm": 19.25,
      "learning_rate": 3.109558147009984e-05,
      "loss": 1.0325,
      "num_input_tokens_seen": 37338784,
      "step": 64380
    },
    {
      "epoch": 9.589663389931486,
      "grad_norm": 8.1875,
      "learning_rate": 3.1092430065732246e-05,
      "loss": 0.8121,
      "num_input_tokens_seen": 37341664,
      "step": 64385
    },
    {
      "epoch": 9.590408102472445,
      "grad_norm": 5.5625,
      "learning_rate": 3.108927855843924e-05,
      "loss": 0.9546,
      "num_input_tokens_seen": 37344608,
      "step": 64390
    },
    {
      "epoch": 9.591152815013405,
      "grad_norm": 10.4375,
      "learning_rate": 3.108612694827407e-05,
      "loss": 0.9664,
      "num_input_tokens_seen": 37347328,
      "step": 64395
    },
    {
      "epoch": 9.591897527554364,
      "grad_norm": 4.90625,
      "learning_rate": 3.108297523528997e-05,
      "loss": 1.2137,
      "num_input_tokens_seen": 37349952,
      "step": 64400
    },
    {
      "epoch": 9.592642240095323,
      "grad_norm": 9.6875,
      "learning_rate": 3.107982341954018e-05,
      "loss": 0.9985,
      "num_input_tokens_seen": 37353152,
      "step": 64405
    },
    {
      "epoch": 9.593386952636282,
      "grad_norm": 9.0625,
      "learning_rate": 3.1076671501077946e-05,
      "loss": 0.9358,
      "num_input_tokens_seen": 37355968,
      "step": 64410
    },
    {
      "epoch": 9.594131665177242,
      "grad_norm": 11.4375,
      "learning_rate": 3.1073519479956534e-05,
      "loss": 1.2532,
      "num_input_tokens_seen": 37358688,
      "step": 64415
    },
    {
      "epoch": 9.594876377718201,
      "grad_norm": 5.0,
      "learning_rate": 3.107036735622918e-05,
      "loss": 0.8347,
      "num_input_tokens_seen": 37361632,
      "step": 64420
    },
    {
      "epoch": 9.59562109025916,
      "grad_norm": 10.625,
      "learning_rate": 3.106721512994913e-05,
      "loss": 1.0835,
      "num_input_tokens_seen": 37364352,
      "step": 64425
    },
    {
      "epoch": 9.596365802800118,
      "grad_norm": 9.0625,
      "learning_rate": 3.1064062801169666e-05,
      "loss": 1.023,
      "num_input_tokens_seen": 37367424,
      "step": 64430
    },
    {
      "epoch": 9.597110515341079,
      "grad_norm": 8.3125,
      "learning_rate": 3.106091036994401e-05,
      "loss": 1.0696,
      "num_input_tokens_seen": 37370624,
      "step": 64435
    },
    {
      "epoch": 9.597855227882038,
      "grad_norm": 14.0,
      "learning_rate": 3.105775783632544e-05,
      "loss": 0.7968,
      "num_input_tokens_seen": 37373376,
      "step": 64440
    },
    {
      "epoch": 9.598599940422996,
      "grad_norm": 11.25,
      "learning_rate": 3.10546052003672e-05,
      "loss": 1.0179,
      "num_input_tokens_seen": 37376352,
      "step": 64445
    },
    {
      "epoch": 9.599344652963955,
      "grad_norm": 9.3125,
      "learning_rate": 3.105145246212257e-05,
      "loss": 1.1211,
      "num_input_tokens_seen": 37379232,
      "step": 64450
    },
    {
      "epoch": 9.600089365504916,
      "grad_norm": 5.71875,
      "learning_rate": 3.1048299621644794e-05,
      "loss": 0.9666,
      "num_input_tokens_seen": 37382240,
      "step": 64455
    },
    {
      "epoch": 9.600834078045875,
      "grad_norm": 9.0,
      "learning_rate": 3.1045146678987144e-05,
      "loss": 1.2522,
      "num_input_tokens_seen": 37385024,
      "step": 64460
    },
    {
      "epoch": 9.601578790586833,
      "grad_norm": 11.4375,
      "learning_rate": 3.104199363420289e-05,
      "loss": 1.2351,
      "num_input_tokens_seen": 37388032,
      "step": 64465
    },
    {
      "epoch": 9.602323503127792,
      "grad_norm": 10.625,
      "learning_rate": 3.1038840487345286e-05,
      "loss": 0.9009,
      "num_input_tokens_seen": 37390976,
      "step": 64470
    },
    {
      "epoch": 9.603068215668753,
      "grad_norm": 8.375,
      "learning_rate": 3.103568723846761e-05,
      "loss": 0.8668,
      "num_input_tokens_seen": 37393920,
      "step": 64475
    },
    {
      "epoch": 9.603812928209711,
      "grad_norm": 9.1875,
      "learning_rate": 3.103253388762314e-05,
      "loss": 0.9896,
      "num_input_tokens_seen": 37396768,
      "step": 64480
    },
    {
      "epoch": 9.60455764075067,
      "grad_norm": 12.1875,
      "learning_rate": 3.102938043486513e-05,
      "loss": 1.2536,
      "num_input_tokens_seen": 37399712,
      "step": 64485
    },
    {
      "epoch": 9.605302353291629,
      "grad_norm": 15.125,
      "learning_rate": 3.102622688024688e-05,
      "loss": 1.0515,
      "num_input_tokens_seen": 37402720,
      "step": 64490
    },
    {
      "epoch": 9.60604706583259,
      "grad_norm": 6.9375,
      "learning_rate": 3.1023073223821643e-05,
      "loss": 1.0696,
      "num_input_tokens_seen": 37405440,
      "step": 64495
    },
    {
      "epoch": 9.606791778373548,
      "grad_norm": 11.0,
      "learning_rate": 3.101991946564271e-05,
      "loss": 0.9906,
      "num_input_tokens_seen": 37408512,
      "step": 64500
    },
    {
      "epoch": 9.607536490914507,
      "grad_norm": 8.9375,
      "learning_rate": 3.1016765605763346e-05,
      "loss": 1.1513,
      "num_input_tokens_seen": 37412064,
      "step": 64505
    },
    {
      "epoch": 9.608281203455466,
      "grad_norm": 7.78125,
      "learning_rate": 3.101361164423685e-05,
      "loss": 0.8094,
      "num_input_tokens_seen": 37415040,
      "step": 64510
    },
    {
      "epoch": 9.609025915996426,
      "grad_norm": 5.90625,
      "learning_rate": 3.1010457581116494e-05,
      "loss": 0.8378,
      "num_input_tokens_seen": 37417792,
      "step": 64515
    },
    {
      "epoch": 9.609770628537385,
      "grad_norm": 8.25,
      "learning_rate": 3.100730341645557e-05,
      "loss": 0.9955,
      "num_input_tokens_seen": 37420768,
      "step": 64520
    },
    {
      "epoch": 9.610515341078344,
      "grad_norm": 10.3125,
      "learning_rate": 3.100414915030736e-05,
      "loss": 1.2009,
      "num_input_tokens_seen": 37423424,
      "step": 64525
    },
    {
      "epoch": 9.611260053619302,
      "grad_norm": 14.125,
      "learning_rate": 3.100099478272515e-05,
      "loss": 1.1562,
      "num_input_tokens_seen": 37426560,
      "step": 64530
    },
    {
      "epoch": 9.612004766160261,
      "grad_norm": 10.5625,
      "learning_rate": 3.099784031376224e-05,
      "loss": 1.097,
      "num_input_tokens_seen": 37429152,
      "step": 64535
    },
    {
      "epoch": 9.612749478701222,
      "grad_norm": 11.75,
      "learning_rate": 3.09946857434719e-05,
      "loss": 0.9727,
      "num_input_tokens_seen": 37432224,
      "step": 64540
    },
    {
      "epoch": 9.61349419124218,
      "grad_norm": 16.875,
      "learning_rate": 3.099153107190744e-05,
      "loss": 1.0849,
      "num_input_tokens_seen": 37434976,
      "step": 64545
    },
    {
      "epoch": 9.61423890378314,
      "grad_norm": 7.8125,
      "learning_rate": 3.0988376299122154e-05,
      "loss": 1.0931,
      "num_input_tokens_seen": 37437952,
      "step": 64550
    },
    {
      "epoch": 9.6149836163241,
      "grad_norm": 11.5,
      "learning_rate": 3.098522142516934e-05,
      "loss": 1.1034,
      "num_input_tokens_seen": 37440736,
      "step": 64555
    },
    {
      "epoch": 9.615728328865059,
      "grad_norm": 6.6875,
      "learning_rate": 3.09820664501023e-05,
      "loss": 0.8745,
      "num_input_tokens_seen": 37443744,
      "step": 64560
    },
    {
      "epoch": 9.616473041406017,
      "grad_norm": 9.4375,
      "learning_rate": 3.097891137397432e-05,
      "loss": 0.9642,
      "num_input_tokens_seen": 37446432,
      "step": 64565
    },
    {
      "epoch": 9.617217753946976,
      "grad_norm": 9.0625,
      "learning_rate": 3.097575619683871e-05,
      "loss": 0.9221,
      "num_input_tokens_seen": 37449664,
      "step": 64570
    },
    {
      "epoch": 9.617962466487935,
      "grad_norm": 11.9375,
      "learning_rate": 3.097260091874877e-05,
      "loss": 1.0909,
      "num_input_tokens_seen": 37452608,
      "step": 64575
    },
    {
      "epoch": 9.618707179028895,
      "grad_norm": 6.90625,
      "learning_rate": 3.0969445539757805e-05,
      "loss": 0.9193,
      "num_input_tokens_seen": 37455360,
      "step": 64580
    },
    {
      "epoch": 9.619451891569854,
      "grad_norm": 14.0625,
      "learning_rate": 3.0966290059919126e-05,
      "loss": 1.0588,
      "num_input_tokens_seen": 37458528,
      "step": 64585
    },
    {
      "epoch": 9.620196604110813,
      "grad_norm": 12.5625,
      "learning_rate": 3.096313447928604e-05,
      "loss": 1.3632,
      "num_input_tokens_seen": 37461408,
      "step": 64590
    },
    {
      "epoch": 9.620941316651772,
      "grad_norm": 8.9375,
      "learning_rate": 3.095997879791187e-05,
      "loss": 0.9205,
      "num_input_tokens_seen": 37464384,
      "step": 64595
    },
    {
      "epoch": 9.621686029192732,
      "grad_norm": 6.75,
      "learning_rate": 3.095682301584991e-05,
      "loss": 1.022,
      "num_input_tokens_seen": 37467456,
      "step": 64600
    },
    {
      "epoch": 9.622430741733691,
      "grad_norm": 14.0625,
      "learning_rate": 3.095366713315347e-05,
      "loss": 0.8866,
      "num_input_tokens_seen": 37470368,
      "step": 64605
    },
    {
      "epoch": 9.62317545427465,
      "grad_norm": 10.875,
      "learning_rate": 3.095051114987588e-05,
      "loss": 0.8587,
      "num_input_tokens_seen": 37473152,
      "step": 64610
    },
    {
      "epoch": 9.623920166815608,
      "grad_norm": 12.5625,
      "learning_rate": 3.094735506607045e-05,
      "loss": 1.1239,
      "num_input_tokens_seen": 37476096,
      "step": 64615
    },
    {
      "epoch": 9.624664879356569,
      "grad_norm": 3.125,
      "learning_rate": 3.09441988817905e-05,
      "loss": 0.8238,
      "num_input_tokens_seen": 37478784,
      "step": 64620
    },
    {
      "epoch": 9.625409591897528,
      "grad_norm": 8.5625,
      "learning_rate": 3.0941042597089356e-05,
      "loss": 1.059,
      "num_input_tokens_seen": 37481632,
      "step": 64625
    },
    {
      "epoch": 9.626154304438487,
      "grad_norm": 6.59375,
      "learning_rate": 3.093788621202033e-05,
      "loss": 1.084,
      "num_input_tokens_seen": 37484608,
      "step": 64630
    },
    {
      "epoch": 9.626899016979445,
      "grad_norm": 8.0625,
      "learning_rate": 3.0934729726636755e-05,
      "loss": 1.0453,
      "num_input_tokens_seen": 37487904,
      "step": 64635
    },
    {
      "epoch": 9.627643729520406,
      "grad_norm": 6.8125,
      "learning_rate": 3.093157314099196e-05,
      "loss": 1.0316,
      "num_input_tokens_seen": 37490752,
      "step": 64640
    },
    {
      "epoch": 9.628388442061365,
      "grad_norm": 7.5625,
      "learning_rate": 3.092841645513925e-05,
      "loss": 1.0338,
      "num_input_tokens_seen": 37493600,
      "step": 64645
    },
    {
      "epoch": 9.629133154602323,
      "grad_norm": 12.625,
      "learning_rate": 3.092525966913198e-05,
      "loss": 1.2828,
      "num_input_tokens_seen": 37496480,
      "step": 64650
    },
    {
      "epoch": 9.629877867143282,
      "grad_norm": 10.875,
      "learning_rate": 3.0922102783023466e-05,
      "loss": 0.9542,
      "num_input_tokens_seen": 37499520,
      "step": 64655
    },
    {
      "epoch": 9.630622579684243,
      "grad_norm": 5.34375,
      "learning_rate": 3.0918945796867044e-05,
      "loss": 1.0638,
      "num_input_tokens_seen": 37502432,
      "step": 64660
    },
    {
      "epoch": 9.631367292225201,
      "grad_norm": 13.625,
      "learning_rate": 3.091578871071605e-05,
      "loss": 1.1584,
      "num_input_tokens_seen": 37505248,
      "step": 64665
    },
    {
      "epoch": 9.63211200476616,
      "grad_norm": 10.0,
      "learning_rate": 3.0912631524623826e-05,
      "loss": 1.2884,
      "num_input_tokens_seen": 37508544,
      "step": 64670
    },
    {
      "epoch": 9.632856717307119,
      "grad_norm": 6.8125,
      "learning_rate": 3.0909474238643694e-05,
      "loss": 0.9545,
      "num_input_tokens_seen": 37511680,
      "step": 64675
    },
    {
      "epoch": 9.63360142984808,
      "grad_norm": 9.5625,
      "learning_rate": 3.090631685282901e-05,
      "loss": 0.9155,
      "num_input_tokens_seen": 37514464,
      "step": 64680
    },
    {
      "epoch": 9.634346142389038,
      "grad_norm": 19.625,
      "learning_rate": 3.0903159367233086e-05,
      "loss": 1.2979,
      "num_input_tokens_seen": 37517408,
      "step": 64685
    },
    {
      "epoch": 9.635090854929997,
      "grad_norm": 10.0625,
      "learning_rate": 3.09000017819093e-05,
      "loss": 1.2498,
      "num_input_tokens_seen": 37520480,
      "step": 64690
    },
    {
      "epoch": 9.635835567470956,
      "grad_norm": 9.8125,
      "learning_rate": 3.089684409691097e-05,
      "loss": 0.994,
      "num_input_tokens_seen": 37523456,
      "step": 64695
    },
    {
      "epoch": 9.636580280011916,
      "grad_norm": 10.625,
      "learning_rate": 3.0893686312291466e-05,
      "loss": 1.061,
      "num_input_tokens_seen": 37526592,
      "step": 64700
    },
    {
      "epoch": 9.637324992552875,
      "grad_norm": 9.8125,
      "learning_rate": 3.089052842810411e-05,
      "loss": 0.9369,
      "num_input_tokens_seen": 37529728,
      "step": 64705
    },
    {
      "epoch": 9.638069705093834,
      "grad_norm": 16.25,
      "learning_rate": 3.088737044440226e-05,
      "loss": 1.2564,
      "num_input_tokens_seen": 37532480,
      "step": 64710
    },
    {
      "epoch": 9.638814417634793,
      "grad_norm": 12.25,
      "learning_rate": 3.088421236123928e-05,
      "loss": 1.1513,
      "num_input_tokens_seen": 37535680,
      "step": 64715
    },
    {
      "epoch": 9.639559130175751,
      "grad_norm": 6.28125,
      "learning_rate": 3.0881054178668514e-05,
      "loss": 1.0474,
      "num_input_tokens_seen": 37538240,
      "step": 64720
    },
    {
      "epoch": 9.640303842716712,
      "grad_norm": 12.5625,
      "learning_rate": 3.087789589674331e-05,
      "loss": 1.2019,
      "num_input_tokens_seen": 37542336,
      "step": 64725
    },
    {
      "epoch": 9.64104855525767,
      "grad_norm": 6.09375,
      "learning_rate": 3.087473751551703e-05,
      "loss": 0.8285,
      "num_input_tokens_seen": 37545600,
      "step": 64730
    },
    {
      "epoch": 9.64179326779863,
      "grad_norm": 12.6875,
      "learning_rate": 3.087157903504303e-05,
      "loss": 1.2984,
      "num_input_tokens_seen": 37548352,
      "step": 64735
    },
    {
      "epoch": 9.642537980339588,
      "grad_norm": 7.78125,
      "learning_rate": 3.086842045537467e-05,
      "loss": 1.1556,
      "num_input_tokens_seen": 37551072,
      "step": 64740
    },
    {
      "epoch": 9.643282692880549,
      "grad_norm": 12.375,
      "learning_rate": 3.0865261776565306e-05,
      "loss": 1.1627,
      "num_input_tokens_seen": 37554080,
      "step": 64745
    },
    {
      "epoch": 9.644027405421507,
      "grad_norm": 9.5,
      "learning_rate": 3.0862102998668314e-05,
      "loss": 1.1235,
      "num_input_tokens_seen": 37557024,
      "step": 64750
    },
    {
      "epoch": 9.644772117962466,
      "grad_norm": 17.375,
      "learning_rate": 3.085894412173704e-05,
      "loss": 1.0543,
      "num_input_tokens_seen": 37559776,
      "step": 64755
    },
    {
      "epoch": 9.645516830503425,
      "grad_norm": 8.125,
      "learning_rate": 3.085578514582487e-05,
      "loss": 1.1701,
      "num_input_tokens_seen": 37562464,
      "step": 64760
    },
    {
      "epoch": 9.646261543044385,
      "grad_norm": 9.3125,
      "learning_rate": 3.0852626070985164e-05,
      "loss": 0.8457,
      "num_input_tokens_seen": 37565408,
      "step": 64765
    },
    {
      "epoch": 9.647006255585344,
      "grad_norm": 6.96875,
      "learning_rate": 3.084946689727128e-05,
      "loss": 0.9949,
      "num_input_tokens_seen": 37568256,
      "step": 64770
    },
    {
      "epoch": 9.647750968126303,
      "grad_norm": 8.0625,
      "learning_rate": 3.08463076247366e-05,
      "loss": 0.9081,
      "num_input_tokens_seen": 37571488,
      "step": 64775
    },
    {
      "epoch": 9.648495680667262,
      "grad_norm": 16.125,
      "learning_rate": 3.084314825343449e-05,
      "loss": 1.2793,
      "num_input_tokens_seen": 37574368,
      "step": 64780
    },
    {
      "epoch": 9.649240393208222,
      "grad_norm": 8.3125,
      "learning_rate": 3.083998878341833e-05,
      "loss": 1.0756,
      "num_input_tokens_seen": 37577120,
      "step": 64785
    },
    {
      "epoch": 9.649985105749181,
      "grad_norm": 13.125,
      "learning_rate": 3.0836829214741496e-05,
      "loss": 0.9927,
      "num_input_tokens_seen": 37580352,
      "step": 64790
    },
    {
      "epoch": 9.65072981829014,
      "grad_norm": 12.0,
      "learning_rate": 3.0833669547457375e-05,
      "loss": 1.293,
      "num_input_tokens_seen": 37583136,
      "step": 64795
    },
    {
      "epoch": 9.651474530831099,
      "grad_norm": 7.84375,
      "learning_rate": 3.083050978161933e-05,
      "loss": 0.9887,
      "num_input_tokens_seen": 37586208,
      "step": 64800
    },
    {
      "epoch": 9.652219243372059,
      "grad_norm": 12.1875,
      "learning_rate": 3.082734991728075e-05,
      "loss": 1.0741,
      "num_input_tokens_seen": 37589152,
      "step": 64805
    },
    {
      "epoch": 9.652963955913018,
      "grad_norm": 15.875,
      "learning_rate": 3.0824189954495006e-05,
      "loss": 1.1114,
      "num_input_tokens_seen": 37592128,
      "step": 64810
    },
    {
      "epoch": 9.653708668453977,
      "grad_norm": 8.125,
      "learning_rate": 3.08210298933155e-05,
      "loss": 1.0471,
      "num_input_tokens_seen": 37594944,
      "step": 64815
    },
    {
      "epoch": 9.654453380994935,
      "grad_norm": 10.5,
      "learning_rate": 3.081786973379561e-05,
      "loss": 1.0054,
      "num_input_tokens_seen": 37598016,
      "step": 64820
    },
    {
      "epoch": 9.655198093535896,
      "grad_norm": 10.5,
      "learning_rate": 3.081470947598872e-05,
      "loss": 1.1881,
      "num_input_tokens_seen": 37601344,
      "step": 64825
    },
    {
      "epoch": 9.655942806076855,
      "grad_norm": 8.5625,
      "learning_rate": 3.081154911994822e-05,
      "loss": 1.1257,
      "num_input_tokens_seen": 37604448,
      "step": 64830
    },
    {
      "epoch": 9.656687518617813,
      "grad_norm": 12.5,
      "learning_rate": 3.080838866572752e-05,
      "loss": 1.0031,
      "num_input_tokens_seen": 37607232,
      "step": 64835
    },
    {
      "epoch": 9.657432231158772,
      "grad_norm": 10.375,
      "learning_rate": 3.0805228113379986e-05,
      "loss": 1.094,
      "num_input_tokens_seen": 37610144,
      "step": 64840
    },
    {
      "epoch": 9.658176943699733,
      "grad_norm": 8.0,
      "learning_rate": 3.080206746295902e-05,
      "loss": 1.0429,
      "num_input_tokens_seen": 37613184,
      "step": 64845
    },
    {
      "epoch": 9.658921656240691,
      "grad_norm": 10.25,
      "learning_rate": 3.079890671451802e-05,
      "loss": 1.1153,
      "num_input_tokens_seen": 37615840,
      "step": 64850
    },
    {
      "epoch": 9.65966636878165,
      "grad_norm": 7.40625,
      "learning_rate": 3.079574586811039e-05,
      "loss": 1.226,
      "num_input_tokens_seen": 37618656,
      "step": 64855
    },
    {
      "epoch": 9.660411081322609,
      "grad_norm": 20.875,
      "learning_rate": 3.0792584923789525e-05,
      "loss": 1.1018,
      "num_input_tokens_seen": 37621696,
      "step": 64860
    },
    {
      "epoch": 9.66115579386357,
      "grad_norm": 13.125,
      "learning_rate": 3.078942388160883e-05,
      "loss": 1.2014,
      "num_input_tokens_seen": 37624672,
      "step": 64865
    },
    {
      "epoch": 9.661900506404528,
      "grad_norm": 8.5,
      "learning_rate": 3.07862627416217e-05,
      "loss": 0.8688,
      "num_input_tokens_seen": 37627840,
      "step": 64870
    },
    {
      "epoch": 9.662645218945487,
      "grad_norm": 16.625,
      "learning_rate": 3.0783101503881526e-05,
      "loss": 1.2517,
      "num_input_tokens_seen": 37631264,
      "step": 64875
    },
    {
      "epoch": 9.663389931486446,
      "grad_norm": 10.4375,
      "learning_rate": 3.0779940168441754e-05,
      "loss": 0.9854,
      "num_input_tokens_seen": 37634240,
      "step": 64880
    },
    {
      "epoch": 9.664134644027406,
      "grad_norm": 12.375,
      "learning_rate": 3.077677873535575e-05,
      "loss": 0.8426,
      "num_input_tokens_seen": 37637312,
      "step": 64885
    },
    {
      "epoch": 9.664879356568365,
      "grad_norm": 4.875,
      "learning_rate": 3.0773617204676946e-05,
      "loss": 0.9024,
      "num_input_tokens_seen": 37640416,
      "step": 64890
    },
    {
      "epoch": 9.665624069109324,
      "grad_norm": 10.0,
      "learning_rate": 3.077045557645875e-05,
      "loss": 1.0757,
      "num_input_tokens_seen": 37643552,
      "step": 64895
    },
    {
      "epoch": 9.666368781650283,
      "grad_norm": 8.5,
      "learning_rate": 3.0767293850754566e-05,
      "loss": 0.854,
      "num_input_tokens_seen": 37646208,
      "step": 64900
    },
    {
      "epoch": 9.667113494191241,
      "grad_norm": 8.8125,
      "learning_rate": 3.076413202761782e-05,
      "loss": 0.987,
      "num_input_tokens_seen": 37649088,
      "step": 64905
    },
    {
      "epoch": 9.667858206732202,
      "grad_norm": 10.5625,
      "learning_rate": 3.076097010710192e-05,
      "loss": 1.1574,
      "num_input_tokens_seen": 37651904,
      "step": 64910
    },
    {
      "epoch": 9.66860291927316,
      "grad_norm": 10.5625,
      "learning_rate": 3.075780808926028e-05,
      "loss": 1.0531,
      "num_input_tokens_seen": 37654624,
      "step": 64915
    },
    {
      "epoch": 9.66934763181412,
      "grad_norm": 13.8125,
      "learning_rate": 3.075464597414632e-05,
      "loss": 0.728,
      "num_input_tokens_seen": 37657312,
      "step": 64920
    },
    {
      "epoch": 9.670092344355078,
      "grad_norm": 8.8125,
      "learning_rate": 3.075148376181348e-05,
      "loss": 1.2975,
      "num_input_tokens_seen": 37660288,
      "step": 64925
    },
    {
      "epoch": 9.670837056896039,
      "grad_norm": 7.0,
      "learning_rate": 3.074832145231517e-05,
      "loss": 1.0429,
      "num_input_tokens_seen": 37663392,
      "step": 64930
    },
    {
      "epoch": 9.671581769436997,
      "grad_norm": 6.9375,
      "learning_rate": 3.07451590457048e-05,
      "loss": 1.0078,
      "num_input_tokens_seen": 37666272,
      "step": 64935
    },
    {
      "epoch": 9.672326481977956,
      "grad_norm": 12.9375,
      "learning_rate": 3.0741996542035804e-05,
      "loss": 0.7983,
      "num_input_tokens_seen": 37669312,
      "step": 64940
    },
    {
      "epoch": 9.673071194518915,
      "grad_norm": 10.9375,
      "learning_rate": 3.073883394136162e-05,
      "loss": 1.0642,
      "num_input_tokens_seen": 37672064,
      "step": 64945
    },
    {
      "epoch": 9.673815907059875,
      "grad_norm": 4.28125,
      "learning_rate": 3.073567124373567e-05,
      "loss": 0.7629,
      "num_input_tokens_seen": 37674720,
      "step": 64950
    },
    {
      "epoch": 9.674560619600834,
      "grad_norm": 8.375,
      "learning_rate": 3.0732508449211373e-05,
      "loss": 1.1291,
      "num_input_tokens_seen": 37677536,
      "step": 64955
    },
    {
      "epoch": 9.675305332141793,
      "grad_norm": 7.84375,
      "learning_rate": 3.0729345557842184e-05,
      "loss": 0.9004,
      "num_input_tokens_seen": 37680192,
      "step": 64960
    },
    {
      "epoch": 9.676050044682752,
      "grad_norm": 7.21875,
      "learning_rate": 3.072618256968153e-05,
      "loss": 0.9246,
      "num_input_tokens_seen": 37684064,
      "step": 64965
    },
    {
      "epoch": 9.676794757223712,
      "grad_norm": 7.75,
      "learning_rate": 3.072301948478283e-05,
      "loss": 1.0005,
      "num_input_tokens_seen": 37686848,
      "step": 64970
    },
    {
      "epoch": 9.677539469764671,
      "grad_norm": 23.375,
      "learning_rate": 3.0719856303199526e-05,
      "loss": 1.3656,
      "num_input_tokens_seen": 37689824,
      "step": 64975
    },
    {
      "epoch": 9.67828418230563,
      "grad_norm": 10.0,
      "learning_rate": 3.071669302498508e-05,
      "loss": 0.9015,
      "num_input_tokens_seen": 37693184,
      "step": 64980
    },
    {
      "epoch": 9.679028894846589,
      "grad_norm": 20.0,
      "learning_rate": 3.07135296501929e-05,
      "loss": 1.238,
      "num_input_tokens_seen": 37696256,
      "step": 64985
    },
    {
      "epoch": 9.679773607387549,
      "grad_norm": 12.0,
      "learning_rate": 3.071036617887645e-05,
      "loss": 0.8709,
      "num_input_tokens_seen": 37698912,
      "step": 64990
    },
    {
      "epoch": 9.680518319928508,
      "grad_norm": 19.125,
      "learning_rate": 3.070720261108917e-05,
      "loss": 1.4879,
      "num_input_tokens_seen": 37702016,
      "step": 64995
    },
    {
      "epoch": 9.681263032469467,
      "grad_norm": 11.0,
      "learning_rate": 3.07040389468845e-05,
      "loss": 0.98,
      "num_input_tokens_seen": 37704768,
      "step": 65000
    },
    {
      "epoch": 9.682007745010425,
      "grad_norm": 13.75,
      "learning_rate": 3.070087518631589e-05,
      "loss": 1.4033,
      "num_input_tokens_seen": 37707424,
      "step": 65005
    },
    {
      "epoch": 9.682752457551386,
      "grad_norm": 6.40625,
      "learning_rate": 3.069771132943679e-05,
      "loss": 1.2175,
      "num_input_tokens_seen": 37710400,
      "step": 65010
    },
    {
      "epoch": 9.683497170092345,
      "grad_norm": 7.6875,
      "learning_rate": 3.069454737630064e-05,
      "loss": 1.465,
      "num_input_tokens_seen": 37712928,
      "step": 65015
    },
    {
      "epoch": 9.684241882633303,
      "grad_norm": 11.0,
      "learning_rate": 3.069138332696091e-05,
      "loss": 1.1352,
      "num_input_tokens_seen": 37716128,
      "step": 65020
    },
    {
      "epoch": 9.684986595174262,
      "grad_norm": 11.8125,
      "learning_rate": 3.0688219181471036e-05,
      "loss": 0.7709,
      "num_input_tokens_seen": 37718912,
      "step": 65025
    },
    {
      "epoch": 9.685731307715223,
      "grad_norm": 6.34375,
      "learning_rate": 3.0685054939884485e-05,
      "loss": 0.9234,
      "num_input_tokens_seen": 37722304,
      "step": 65030
    },
    {
      "epoch": 9.686476020256181,
      "grad_norm": 12.25,
      "learning_rate": 3.0681890602254704e-05,
      "loss": 1.0873,
      "num_input_tokens_seen": 37725248,
      "step": 65035
    },
    {
      "epoch": 9.68722073279714,
      "grad_norm": 7.28125,
      "learning_rate": 3.067872616863516e-05,
      "loss": 1.2378,
      "num_input_tokens_seen": 37728224,
      "step": 65040
    },
    {
      "epoch": 9.687965445338099,
      "grad_norm": 8.4375,
      "learning_rate": 3.0675561639079306e-05,
      "loss": 0.9477,
      "num_input_tokens_seen": 37731072,
      "step": 65045
    },
    {
      "epoch": 9.688710157879058,
      "grad_norm": 8.75,
      "learning_rate": 3.0672397013640605e-05,
      "loss": 1.1002,
      "num_input_tokens_seen": 37734240,
      "step": 65050
    },
    {
      "epoch": 9.689454870420018,
      "grad_norm": 9.1875,
      "learning_rate": 3.066923229237253e-05,
      "loss": 0.9946,
      "num_input_tokens_seen": 37736992,
      "step": 65055
    },
    {
      "epoch": 9.690199582960977,
      "grad_norm": 9.0625,
      "learning_rate": 3.0666067475328534e-05,
      "loss": 1.0792,
      "num_input_tokens_seen": 37739712,
      "step": 65060
    },
    {
      "epoch": 9.690944295501936,
      "grad_norm": 9.375,
      "learning_rate": 3.066290256256208e-05,
      "loss": 0.9144,
      "num_input_tokens_seen": 37742720,
      "step": 65065
    },
    {
      "epoch": 9.691689008042896,
      "grad_norm": 12.6875,
      "learning_rate": 3.065973755412665e-05,
      "loss": 1.0486,
      "num_input_tokens_seen": 37745696,
      "step": 65070
    },
    {
      "epoch": 9.692433720583855,
      "grad_norm": 16.25,
      "learning_rate": 3.06565724500757e-05,
      "loss": 0.954,
      "num_input_tokens_seen": 37748416,
      "step": 65075
    },
    {
      "epoch": 9.693178433124814,
      "grad_norm": 10.4375,
      "learning_rate": 3.0653407250462716e-05,
      "loss": 0.9041,
      "num_input_tokens_seen": 37751360,
      "step": 65080
    },
    {
      "epoch": 9.693923145665773,
      "grad_norm": 13.25,
      "learning_rate": 3.065024195534116e-05,
      "loss": 1.1046,
      "num_input_tokens_seen": 37754272,
      "step": 65085
    },
    {
      "epoch": 9.694667858206731,
      "grad_norm": 12.8125,
      "learning_rate": 3.06470765647645e-05,
      "loss": 1.0683,
      "num_input_tokens_seen": 37757088,
      "step": 65090
    },
    {
      "epoch": 9.695412570747692,
      "grad_norm": 4.71875,
      "learning_rate": 3.064391107878623e-05,
      "loss": 0.8489,
      "num_input_tokens_seen": 37760128,
      "step": 65095
    },
    {
      "epoch": 9.69615728328865,
      "grad_norm": 10.5625,
      "learning_rate": 3.064074549745982e-05,
      "loss": 0.941,
      "num_input_tokens_seen": 37762976,
      "step": 65100
    },
    {
      "epoch": 9.69690199582961,
      "grad_norm": 9.875,
      "learning_rate": 3.063757982083874e-05,
      "loss": 1.2344,
      "num_input_tokens_seen": 37765920,
      "step": 65105
    },
    {
      "epoch": 9.697646708370568,
      "grad_norm": 14.375,
      "learning_rate": 3.063441404897648e-05,
      "loss": 1.0707,
      "num_input_tokens_seen": 37768864,
      "step": 65110
    },
    {
      "epoch": 9.698391420911529,
      "grad_norm": 8.25,
      "learning_rate": 3.063124818192652e-05,
      "loss": 1.062,
      "num_input_tokens_seen": 37772000,
      "step": 65115
    },
    {
      "epoch": 9.699136133452487,
      "grad_norm": 11.25,
      "learning_rate": 3.062808221974235e-05,
      "loss": 1.1185,
      "num_input_tokens_seen": 37774624,
      "step": 65120
    },
    {
      "epoch": 9.699880845993446,
      "grad_norm": 7.1875,
      "learning_rate": 3.062491616247745e-05,
      "loss": 1.1369,
      "num_input_tokens_seen": 37777536,
      "step": 65125
    },
    {
      "epoch": 9.700625558534405,
      "grad_norm": 10.125,
      "learning_rate": 3.0621750010185316e-05,
      "loss": 1.2444,
      "num_input_tokens_seen": 37780480,
      "step": 65130
    },
    {
      "epoch": 9.701370271075366,
      "grad_norm": 16.875,
      "learning_rate": 3.0618583762919417e-05,
      "loss": 1.2501,
      "num_input_tokens_seen": 37783552,
      "step": 65135
    },
    {
      "epoch": 9.702114983616324,
      "grad_norm": 14.0625,
      "learning_rate": 3.0615417420733264e-05,
      "loss": 0.9014,
      "num_input_tokens_seen": 37786432,
      "step": 65140
    },
    {
      "epoch": 9.702859696157283,
      "grad_norm": 8.5625,
      "learning_rate": 3.0612250983680336e-05,
      "loss": 1.1268,
      "num_input_tokens_seen": 37789408,
      "step": 65145
    },
    {
      "epoch": 9.703604408698242,
      "grad_norm": 12.75,
      "learning_rate": 3.060908445181413e-05,
      "loss": 0.912,
      "num_input_tokens_seen": 37792448,
      "step": 65150
    },
    {
      "epoch": 9.704349121239202,
      "grad_norm": 9.0,
      "learning_rate": 3.060591782518815e-05,
      "loss": 1.1926,
      "num_input_tokens_seen": 37795424,
      "step": 65155
    },
    {
      "epoch": 9.705093833780161,
      "grad_norm": 7.25,
      "learning_rate": 3.060275110385588e-05,
      "loss": 1.1549,
      "num_input_tokens_seen": 37798208,
      "step": 65160
    },
    {
      "epoch": 9.70583854632112,
      "grad_norm": 8.0,
      "learning_rate": 3.059958428787083e-05,
      "loss": 0.8624,
      "num_input_tokens_seen": 37801024,
      "step": 65165
    },
    {
      "epoch": 9.706583258862079,
      "grad_norm": 8.5,
      "learning_rate": 3.059641737728649e-05,
      "loss": 1.407,
      "num_input_tokens_seen": 37804160,
      "step": 65170
    },
    {
      "epoch": 9.70732797140304,
      "grad_norm": 9.875,
      "learning_rate": 3.059325037215637e-05,
      "loss": 1.2522,
      "num_input_tokens_seen": 37807168,
      "step": 65175
    },
    {
      "epoch": 9.708072683943998,
      "grad_norm": 7.59375,
      "learning_rate": 3.059008327253396e-05,
      "loss": 1.0084,
      "num_input_tokens_seen": 37809760,
      "step": 65180
    },
    {
      "epoch": 9.708817396484957,
      "grad_norm": 6.8125,
      "learning_rate": 3.0586916078472785e-05,
      "loss": 0.9659,
      "num_input_tokens_seen": 37812800,
      "step": 65185
    },
    {
      "epoch": 9.709562109025915,
      "grad_norm": 8.5625,
      "learning_rate": 3.058374879002634e-05,
      "loss": 0.9401,
      "num_input_tokens_seen": 37815584,
      "step": 65190
    },
    {
      "epoch": 9.710306821566876,
      "grad_norm": 6.6875,
      "learning_rate": 3.0580581407248126e-05,
      "loss": 1.0051,
      "num_input_tokens_seen": 37818336,
      "step": 65195
    },
    {
      "epoch": 9.711051534107835,
      "grad_norm": 7.40625,
      "learning_rate": 3.0577413930191666e-05,
      "loss": 1.2416,
      "num_input_tokens_seen": 37821120,
      "step": 65200
    },
    {
      "epoch": 9.711796246648793,
      "grad_norm": 18.125,
      "learning_rate": 3.0574246358910474e-05,
      "loss": 1.5111,
      "num_input_tokens_seen": 37823872,
      "step": 65205
    },
    {
      "epoch": 9.712540959189752,
      "grad_norm": 12.25,
      "learning_rate": 3.057107869345804e-05,
      "loss": 1.0386,
      "num_input_tokens_seen": 37826880,
      "step": 65210
    },
    {
      "epoch": 9.713285671730713,
      "grad_norm": 11.125,
      "learning_rate": 3.0567910933887905e-05,
      "loss": 1.1463,
      "num_input_tokens_seen": 37829664,
      "step": 65215
    },
    {
      "epoch": 9.714030384271672,
      "grad_norm": 9.6875,
      "learning_rate": 3.056474308025357e-05,
      "loss": 1.1654,
      "num_input_tokens_seen": 37832576,
      "step": 65220
    },
    {
      "epoch": 9.71477509681263,
      "grad_norm": 14.4375,
      "learning_rate": 3.056157513260856e-05,
      "loss": 1.0486,
      "num_input_tokens_seen": 37835424,
      "step": 65225
    },
    {
      "epoch": 9.715519809353589,
      "grad_norm": 7.71875,
      "learning_rate": 3.055840709100639e-05,
      "loss": 1.2382,
      "num_input_tokens_seen": 37838272,
      "step": 65230
    },
    {
      "epoch": 9.716264521894548,
      "grad_norm": 7.90625,
      "learning_rate": 3.055523895550058e-05,
      "loss": 1.5321,
      "num_input_tokens_seen": 37841184,
      "step": 65235
    },
    {
      "epoch": 9.717009234435508,
      "grad_norm": 6.40625,
      "learning_rate": 3.055207072614465e-05,
      "loss": 0.9605,
      "num_input_tokens_seen": 37844192,
      "step": 65240
    },
    {
      "epoch": 9.717753946976467,
      "grad_norm": 25.0,
      "learning_rate": 3.0548902402992134e-05,
      "loss": 1.0879,
      "num_input_tokens_seen": 37846848,
      "step": 65245
    },
    {
      "epoch": 9.718498659517426,
      "grad_norm": 8.1875,
      "learning_rate": 3.0545733986096545e-05,
      "loss": 0.8253,
      "num_input_tokens_seen": 37850144,
      "step": 65250
    },
    {
      "epoch": 9.719243372058386,
      "grad_norm": 11.625,
      "learning_rate": 3.054256547551142e-05,
      "loss": 1.1255,
      "num_input_tokens_seen": 37852800,
      "step": 65255
    },
    {
      "epoch": 9.719988084599345,
      "grad_norm": 7.65625,
      "learning_rate": 3.0539396871290294e-05,
      "loss": 1.1585,
      "num_input_tokens_seen": 37855520,
      "step": 65260
    },
    {
      "epoch": 9.720732797140304,
      "grad_norm": 18.375,
      "learning_rate": 3.053622817348668e-05,
      "loss": 0.8381,
      "num_input_tokens_seen": 37858112,
      "step": 65265
    },
    {
      "epoch": 9.721477509681263,
      "grad_norm": 32.25,
      "learning_rate": 3.053305938215411e-05,
      "loss": 1.323,
      "num_input_tokens_seen": 37861152,
      "step": 65270
    },
    {
      "epoch": 9.722222222222221,
      "grad_norm": 9.625,
      "learning_rate": 3.052989049734613e-05,
      "loss": 1.0109,
      "num_input_tokens_seen": 37864064,
      "step": 65275
    },
    {
      "epoch": 9.722966934763182,
      "grad_norm": 6.875,
      "learning_rate": 3.052672151911627e-05,
      "loss": 1.0314,
      "num_input_tokens_seen": 37866880,
      "step": 65280
    },
    {
      "epoch": 9.72371164730414,
      "grad_norm": 8.8125,
      "learning_rate": 3.052355244751807e-05,
      "loss": 0.8096,
      "num_input_tokens_seen": 37870208,
      "step": 65285
    },
    {
      "epoch": 9.7244563598451,
      "grad_norm": 7.4375,
      "learning_rate": 3.052038328260507e-05,
      "loss": 1.0037,
      "num_input_tokens_seen": 37872960,
      "step": 65290
    },
    {
      "epoch": 9.725201072386058,
      "grad_norm": 12.125,
      "learning_rate": 3.05172140244308e-05,
      "loss": 0.844,
      "num_input_tokens_seen": 37875936,
      "step": 65295
    },
    {
      "epoch": 9.725945784927019,
      "grad_norm": 7.4375,
      "learning_rate": 3.051404467304881e-05,
      "loss": 1.0867,
      "num_input_tokens_seen": 37878816,
      "step": 65300
    },
    {
      "epoch": 9.726690497467978,
      "grad_norm": 10.0,
      "learning_rate": 3.051087522851263e-05,
      "loss": 0.9932,
      "num_input_tokens_seen": 37881536,
      "step": 65305
    },
    {
      "epoch": 9.727435210008936,
      "grad_norm": 11.625,
      "learning_rate": 3.050770569087582e-05,
      "loss": 1.121,
      "num_input_tokens_seen": 37884448,
      "step": 65310
    },
    {
      "epoch": 9.728179922549895,
      "grad_norm": 8.625,
      "learning_rate": 3.0504536060191917e-05,
      "loss": 1.1856,
      "num_input_tokens_seen": 37887296,
      "step": 65315
    },
    {
      "epoch": 9.728924635090856,
      "grad_norm": 7.125,
      "learning_rate": 3.0501366336514477e-05,
      "loss": 0.7632,
      "num_input_tokens_seen": 37890016,
      "step": 65320
    },
    {
      "epoch": 9.729669347631814,
      "grad_norm": 7.03125,
      "learning_rate": 3.0498196519897044e-05,
      "loss": 0.9709,
      "num_input_tokens_seen": 37892832,
      "step": 65325
    },
    {
      "epoch": 9.730414060172773,
      "grad_norm": 14.4375,
      "learning_rate": 3.0495026610393168e-05,
      "loss": 1.0608,
      "num_input_tokens_seen": 37895712,
      "step": 65330
    },
    {
      "epoch": 9.731158772713732,
      "grad_norm": 6.96875,
      "learning_rate": 3.0491856608056403e-05,
      "loss": 0.9884,
      "num_input_tokens_seen": 37898496,
      "step": 65335
    },
    {
      "epoch": 9.731903485254692,
      "grad_norm": 6.90625,
      "learning_rate": 3.0488686512940297e-05,
      "loss": 1.07,
      "num_input_tokens_seen": 37901248,
      "step": 65340
    },
    {
      "epoch": 9.732648197795651,
      "grad_norm": 10.8125,
      "learning_rate": 3.0485516325098413e-05,
      "loss": 0.9297,
      "num_input_tokens_seen": 37904352,
      "step": 65345
    },
    {
      "epoch": 9.73339291033661,
      "grad_norm": 13.375,
      "learning_rate": 3.0482346044584305e-05,
      "loss": 0.8683,
      "num_input_tokens_seen": 37907296,
      "step": 65350
    },
    {
      "epoch": 9.734137622877569,
      "grad_norm": 13.3125,
      "learning_rate": 3.047917567145153e-05,
      "loss": 1.02,
      "num_input_tokens_seen": 37909920,
      "step": 65355
    },
    {
      "epoch": 9.73488233541853,
      "grad_norm": 14.0,
      "learning_rate": 3.0476005205753666e-05,
      "loss": 1.2442,
      "num_input_tokens_seen": 37912832,
      "step": 65360
    },
    {
      "epoch": 9.735627047959488,
      "grad_norm": 5.625,
      "learning_rate": 3.047283464754425e-05,
      "loss": 0.9638,
      "num_input_tokens_seen": 37915392,
      "step": 65365
    },
    {
      "epoch": 9.736371760500447,
      "grad_norm": 12.5625,
      "learning_rate": 3.0469663996876853e-05,
      "loss": 0.9676,
      "num_input_tokens_seen": 37918368,
      "step": 65370
    },
    {
      "epoch": 9.737116473041405,
      "grad_norm": 12.6875,
      "learning_rate": 3.046649325380504e-05,
      "loss": 1.061,
      "num_input_tokens_seen": 37920992,
      "step": 65375
    },
    {
      "epoch": 9.737861185582366,
      "grad_norm": 11.125,
      "learning_rate": 3.0463322418382384e-05,
      "loss": 1.1303,
      "num_input_tokens_seen": 37923968,
      "step": 65380
    },
    {
      "epoch": 9.738605898123325,
      "grad_norm": 8.5,
      "learning_rate": 3.0460151490662442e-05,
      "loss": 1.1572,
      "num_input_tokens_seen": 37926752,
      "step": 65385
    },
    {
      "epoch": 9.739350610664284,
      "grad_norm": 7.3125,
      "learning_rate": 3.0456980470698803e-05,
      "loss": 0.7521,
      "num_input_tokens_seen": 37929760,
      "step": 65390
    },
    {
      "epoch": 9.740095323205242,
      "grad_norm": 16.75,
      "learning_rate": 3.0453809358545016e-05,
      "loss": 1.4299,
      "num_input_tokens_seen": 37932416,
      "step": 65395
    },
    {
      "epoch": 9.740840035746203,
      "grad_norm": 13.8125,
      "learning_rate": 3.0450638154254664e-05,
      "loss": 1.0058,
      "num_input_tokens_seen": 37935104,
      "step": 65400
    },
    {
      "epoch": 9.741584748287162,
      "grad_norm": 19.5,
      "learning_rate": 3.0447466857881318e-05,
      "loss": 1.302,
      "num_input_tokens_seen": 37937920,
      "step": 65405
    },
    {
      "epoch": 9.74232946082812,
      "grad_norm": 9.0625,
      "learning_rate": 3.0444295469478557e-05,
      "loss": 0.8755,
      "num_input_tokens_seen": 37940960,
      "step": 65410
    },
    {
      "epoch": 9.743074173369079,
      "grad_norm": 9.1875,
      "learning_rate": 3.0441123989099958e-05,
      "loss": 0.8628,
      "num_input_tokens_seen": 37943872,
      "step": 65415
    },
    {
      "epoch": 9.743818885910038,
      "grad_norm": 6.0625,
      "learning_rate": 3.0437952416799097e-05,
      "loss": 1.0292,
      "num_input_tokens_seen": 37946784,
      "step": 65420
    },
    {
      "epoch": 9.744563598450998,
      "grad_norm": 6.40625,
      "learning_rate": 3.0434780752629567e-05,
      "loss": 0.9754,
      "num_input_tokens_seen": 37949696,
      "step": 65425
    },
    {
      "epoch": 9.745308310991957,
      "grad_norm": 7.65625,
      "learning_rate": 3.043160899664493e-05,
      "loss": 0.974,
      "num_input_tokens_seen": 37952736,
      "step": 65430
    },
    {
      "epoch": 9.746053023532916,
      "grad_norm": 10.9375,
      "learning_rate": 3.042843714889878e-05,
      "loss": 0.9249,
      "num_input_tokens_seen": 37955616,
      "step": 65435
    },
    {
      "epoch": 9.746797736073875,
      "grad_norm": 18.375,
      "learning_rate": 3.0425265209444704e-05,
      "loss": 1.3646,
      "num_input_tokens_seen": 37958496,
      "step": 65440
    },
    {
      "epoch": 9.747542448614835,
      "grad_norm": 12.75,
      "learning_rate": 3.0422093178336287e-05,
      "loss": 1.0335,
      "num_input_tokens_seen": 37961376,
      "step": 65445
    },
    {
      "epoch": 9.748287161155794,
      "grad_norm": 11.4375,
      "learning_rate": 3.0418921055627115e-05,
      "loss": 1.0658,
      "num_input_tokens_seen": 37964288,
      "step": 65450
    },
    {
      "epoch": 9.749031873696753,
      "grad_norm": 5.9375,
      "learning_rate": 3.0415748841370778e-05,
      "loss": 0.823,
      "num_input_tokens_seen": 37967200,
      "step": 65455
    },
    {
      "epoch": 9.749776586237711,
      "grad_norm": 10.25,
      "learning_rate": 3.0412576535620873e-05,
      "loss": 1.1941,
      "num_input_tokens_seen": 37969952,
      "step": 65460
    },
    {
      "epoch": 9.750521298778672,
      "grad_norm": 13.4375,
      "learning_rate": 3.0409404138430986e-05,
      "loss": 0.9926,
      "num_input_tokens_seen": 37973088,
      "step": 65465
    },
    {
      "epoch": 9.75126601131963,
      "grad_norm": 11.125,
      "learning_rate": 3.040623164985471e-05,
      "loss": 1.2418,
      "num_input_tokens_seen": 37976416,
      "step": 65470
    },
    {
      "epoch": 9.75201072386059,
      "grad_norm": 5.53125,
      "learning_rate": 3.040305906994565e-05,
      "loss": 0.717,
      "num_input_tokens_seen": 37979424,
      "step": 65475
    },
    {
      "epoch": 9.752755436401548,
      "grad_norm": 12.375,
      "learning_rate": 3.0399886398757394e-05,
      "loss": 0.9585,
      "num_input_tokens_seen": 37982432,
      "step": 65480
    },
    {
      "epoch": 9.753500148942509,
      "grad_norm": 8.375,
      "learning_rate": 3.0396713636343553e-05,
      "loss": 0.729,
      "num_input_tokens_seen": 37985632,
      "step": 65485
    },
    {
      "epoch": 9.754244861483468,
      "grad_norm": 8.9375,
      "learning_rate": 3.039354078275771e-05,
      "loss": 1.0803,
      "num_input_tokens_seen": 37988320,
      "step": 65490
    },
    {
      "epoch": 9.754989574024426,
      "grad_norm": 9.4375,
      "learning_rate": 3.039036783805349e-05,
      "loss": 0.9898,
      "num_input_tokens_seen": 37991168,
      "step": 65495
    },
    {
      "epoch": 9.755734286565385,
      "grad_norm": 18.0,
      "learning_rate": 3.0387194802284474e-05,
      "loss": 0.965,
      "num_input_tokens_seen": 37994272,
      "step": 65500
    },
    {
      "epoch": 9.756478999106346,
      "grad_norm": 8.625,
      "learning_rate": 3.0384021675504283e-05,
      "loss": 1.2609,
      "num_input_tokens_seen": 37996960,
      "step": 65505
    },
    {
      "epoch": 9.757223711647304,
      "grad_norm": 8.5625,
      "learning_rate": 3.038084845776651e-05,
      "loss": 0.7794,
      "num_input_tokens_seen": 37999808,
      "step": 65510
    },
    {
      "epoch": 9.757968424188263,
      "grad_norm": 10.3125,
      "learning_rate": 3.0377675149124772e-05,
      "loss": 1.1159,
      "num_input_tokens_seen": 38002784,
      "step": 65515
    },
    {
      "epoch": 9.758713136729222,
      "grad_norm": 14.25,
      "learning_rate": 3.0374501749632684e-05,
      "loss": 0.9618,
      "num_input_tokens_seen": 38005632,
      "step": 65520
    },
    {
      "epoch": 9.759457849270182,
      "grad_norm": 6.5625,
      "learning_rate": 3.037132825934385e-05,
      "loss": 1.3342,
      "num_input_tokens_seen": 38008288,
      "step": 65525
    },
    {
      "epoch": 9.760202561811141,
      "grad_norm": 9.0625,
      "learning_rate": 3.0368154678311888e-05,
      "loss": 1.2718,
      "num_input_tokens_seen": 38011712,
      "step": 65530
    },
    {
      "epoch": 9.7609472743521,
      "grad_norm": 6.1875,
      "learning_rate": 3.0364981006590404e-05,
      "loss": 0.9671,
      "num_input_tokens_seen": 38014784,
      "step": 65535
    },
    {
      "epoch": 9.761691986893059,
      "grad_norm": 10.375,
      "learning_rate": 3.0361807244233016e-05,
      "loss": 1.0258,
      "num_input_tokens_seen": 38017472,
      "step": 65540
    },
    {
      "epoch": 9.76243669943402,
      "grad_norm": 15.9375,
      "learning_rate": 3.0358633391293346e-05,
      "loss": 1.2559,
      "num_input_tokens_seen": 38020064,
      "step": 65545
    },
    {
      "epoch": 9.763181411974978,
      "grad_norm": 6.9375,
      "learning_rate": 3.0355459447825014e-05,
      "loss": 1.3168,
      "num_input_tokens_seen": 38023008,
      "step": 65550
    },
    {
      "epoch": 9.763926124515937,
      "grad_norm": 8.4375,
      "learning_rate": 3.0352285413881636e-05,
      "loss": 1.1082,
      "num_input_tokens_seen": 38026016,
      "step": 65555
    },
    {
      "epoch": 9.764670837056896,
      "grad_norm": 6.8125,
      "learning_rate": 3.0349111289516834e-05,
      "loss": 0.9074,
      "num_input_tokens_seen": 38028992,
      "step": 65560
    },
    {
      "epoch": 9.765415549597854,
      "grad_norm": 5.3125,
      "learning_rate": 3.0345937074784235e-05,
      "loss": 0.7933,
      "num_input_tokens_seen": 38032064,
      "step": 65565
    },
    {
      "epoch": 9.766160262138815,
      "grad_norm": 9.75,
      "learning_rate": 3.0342762769737464e-05,
      "loss": 0.9301,
      "num_input_tokens_seen": 38035296,
      "step": 65570
    },
    {
      "epoch": 9.766904974679774,
      "grad_norm": 11.625,
      "learning_rate": 3.0339588374430146e-05,
      "loss": 1.0426,
      "num_input_tokens_seen": 38038208,
      "step": 65575
    },
    {
      "epoch": 9.767649687220732,
      "grad_norm": 9.25,
      "learning_rate": 3.0336413888915903e-05,
      "loss": 1.4313,
      "num_input_tokens_seen": 38040992,
      "step": 65580
    },
    {
      "epoch": 9.768394399761693,
      "grad_norm": 13.9375,
      "learning_rate": 3.0333239313248372e-05,
      "loss": 0.8501,
      "num_input_tokens_seen": 38044352,
      "step": 65585
    },
    {
      "epoch": 9.769139112302652,
      "grad_norm": 12.5625,
      "learning_rate": 3.0330064647481195e-05,
      "loss": 1.1441,
      "num_input_tokens_seen": 38047136,
      "step": 65590
    },
    {
      "epoch": 9.76988382484361,
      "grad_norm": 9.4375,
      "learning_rate": 3.032688989166798e-05,
      "loss": 1.5263,
      "num_input_tokens_seen": 38049888,
      "step": 65595
    },
    {
      "epoch": 9.77062853738457,
      "grad_norm": 11.625,
      "learning_rate": 3.0323715045862382e-05,
      "loss": 1.2453,
      "num_input_tokens_seen": 38052864,
      "step": 65600
    },
    {
      "epoch": 9.771373249925528,
      "grad_norm": 9.125,
      "learning_rate": 3.0320540110118022e-05,
      "loss": 0.8568,
      "num_input_tokens_seen": 38055744,
      "step": 65605
    },
    {
      "epoch": 9.772117962466488,
      "grad_norm": 12.5625,
      "learning_rate": 3.031736508448855e-05,
      "loss": 1.159,
      "num_input_tokens_seen": 38058528,
      "step": 65610
    },
    {
      "epoch": 9.772862675007447,
      "grad_norm": 10.0625,
      "learning_rate": 3.03141899690276e-05,
      "loss": 0.9658,
      "num_input_tokens_seen": 38061440,
      "step": 65615
    },
    {
      "epoch": 9.773607387548406,
      "grad_norm": 11.4375,
      "learning_rate": 3.031101476378881e-05,
      "loss": 1.1772,
      "num_input_tokens_seen": 38064512,
      "step": 65620
    },
    {
      "epoch": 9.774352100089365,
      "grad_norm": 12.8125,
      "learning_rate": 3.0307839468825826e-05,
      "loss": 1.1971,
      "num_input_tokens_seen": 38067360,
      "step": 65625
    },
    {
      "epoch": 9.775096812630325,
      "grad_norm": 10.5,
      "learning_rate": 3.0304664084192286e-05,
      "loss": 0.9578,
      "num_input_tokens_seen": 38070528,
      "step": 65630
    },
    {
      "epoch": 9.775841525171284,
      "grad_norm": 17.125,
      "learning_rate": 3.0301488609941837e-05,
      "loss": 1.2572,
      "num_input_tokens_seen": 38073536,
      "step": 65635
    },
    {
      "epoch": 9.776586237712243,
      "grad_norm": 7.375,
      "learning_rate": 3.0298313046128123e-05,
      "loss": 1.0765,
      "num_input_tokens_seen": 38076288,
      "step": 65640
    },
    {
      "epoch": 9.777330950253202,
      "grad_norm": 8.25,
      "learning_rate": 3.0295137392804796e-05,
      "loss": 0.7439,
      "num_input_tokens_seen": 38079264,
      "step": 65645
    },
    {
      "epoch": 9.778075662794162,
      "grad_norm": 8.875,
      "learning_rate": 3.0291961650025512e-05,
      "loss": 1.0141,
      "num_input_tokens_seen": 38082400,
      "step": 65650
    },
    {
      "epoch": 9.77882037533512,
      "grad_norm": 12.25,
      "learning_rate": 3.0288785817843907e-05,
      "loss": 1.0301,
      "num_input_tokens_seen": 38085280,
      "step": 65655
    },
    {
      "epoch": 9.77956508787608,
      "grad_norm": 11.4375,
      "learning_rate": 3.028560989631365e-05,
      "loss": 1.0527,
      "num_input_tokens_seen": 38088288,
      "step": 65660
    },
    {
      "epoch": 9.780309800417038,
      "grad_norm": 8.4375,
      "learning_rate": 3.0282433885488375e-05,
      "loss": 1.0624,
      "num_input_tokens_seen": 38090976,
      "step": 65665
    },
    {
      "epoch": 9.781054512957999,
      "grad_norm": 15.4375,
      "learning_rate": 3.0279257785421755e-05,
      "loss": 1.5265,
      "num_input_tokens_seen": 38093824,
      "step": 65670
    },
    {
      "epoch": 9.781799225498958,
      "grad_norm": 7.40625,
      "learning_rate": 3.0276081596167434e-05,
      "loss": 1.113,
      "num_input_tokens_seen": 38097024,
      "step": 65675
    },
    {
      "epoch": 9.782543938039916,
      "grad_norm": 12.125,
      "learning_rate": 3.027290531777908e-05,
      "loss": 1.2443,
      "num_input_tokens_seen": 38099936,
      "step": 65680
    },
    {
      "epoch": 9.783288650580875,
      "grad_norm": 7.40625,
      "learning_rate": 3.026972895031035e-05,
      "loss": 0.8378,
      "num_input_tokens_seen": 38102560,
      "step": 65685
    },
    {
      "epoch": 9.784033363121836,
      "grad_norm": 9.0625,
      "learning_rate": 3.026655249381491e-05,
      "loss": 1.0211,
      "num_input_tokens_seen": 38105600,
      "step": 65690
    },
    {
      "epoch": 9.784778075662794,
      "grad_norm": 12.75,
      "learning_rate": 3.0263375948346416e-05,
      "loss": 0.9672,
      "num_input_tokens_seen": 38108672,
      "step": 65695
    },
    {
      "epoch": 9.785522788203753,
      "grad_norm": 8.875,
      "learning_rate": 3.026019931395853e-05,
      "loss": 1.1537,
      "num_input_tokens_seen": 38111520,
      "step": 65700
    },
    {
      "epoch": 9.786267500744712,
      "grad_norm": 6.0,
      "learning_rate": 3.0257022590704926e-05,
      "loss": 1.3617,
      "num_input_tokens_seen": 38114432,
      "step": 65705
    },
    {
      "epoch": 9.787012213285673,
      "grad_norm": 10.4375,
      "learning_rate": 3.0253845778639267e-05,
      "loss": 0.7836,
      "num_input_tokens_seen": 38116992,
      "step": 65710
    },
    {
      "epoch": 9.787756925826631,
      "grad_norm": 9.375,
      "learning_rate": 3.0250668877815226e-05,
      "loss": 1.0578,
      "num_input_tokens_seen": 38120032,
      "step": 65715
    },
    {
      "epoch": 9.78850163836759,
      "grad_norm": 9.5,
      "learning_rate": 3.024749188828647e-05,
      "loss": 0.858,
      "num_input_tokens_seen": 38122976,
      "step": 65720
    },
    {
      "epoch": 9.789246350908549,
      "grad_norm": 11.8125,
      "learning_rate": 3.024431481010667e-05,
      "loss": 0.7293,
      "num_input_tokens_seen": 38126016,
      "step": 65725
    },
    {
      "epoch": 9.78999106344951,
      "grad_norm": 9.0625,
      "learning_rate": 3.0241137643329508e-05,
      "loss": 1.3349,
      "num_input_tokens_seen": 38128992,
      "step": 65730
    },
    {
      "epoch": 9.790735775990468,
      "grad_norm": 7.1875,
      "learning_rate": 3.0237960388008647e-05,
      "loss": 1.2024,
      "num_input_tokens_seen": 38132000,
      "step": 65735
    },
    {
      "epoch": 9.791480488531427,
      "grad_norm": 15.25,
      "learning_rate": 3.0234783044197767e-05,
      "loss": 1.2556,
      "num_input_tokens_seen": 38135104,
      "step": 65740
    },
    {
      "epoch": 9.792225201072386,
      "grad_norm": 11.375,
      "learning_rate": 3.0231605611950548e-05,
      "loss": 1.3936,
      "num_input_tokens_seen": 38137952,
      "step": 65745
    },
    {
      "epoch": 9.792969913613344,
      "grad_norm": 7.25,
      "learning_rate": 3.0228428091320672e-05,
      "loss": 0.8627,
      "num_input_tokens_seen": 38140672,
      "step": 65750
    },
    {
      "epoch": 9.793714626154305,
      "grad_norm": 5.875,
      "learning_rate": 3.0225250482361818e-05,
      "loss": 1.0536,
      "num_input_tokens_seen": 38143520,
      "step": 65755
    },
    {
      "epoch": 9.794459338695264,
      "grad_norm": 10.8125,
      "learning_rate": 3.0222072785127663e-05,
      "loss": 1.0689,
      "num_input_tokens_seen": 38146336,
      "step": 65760
    },
    {
      "epoch": 9.795204051236222,
      "grad_norm": 16.875,
      "learning_rate": 3.02188949996719e-05,
      "loss": 1.3094,
      "num_input_tokens_seen": 38149152,
      "step": 65765
    },
    {
      "epoch": 9.795948763777183,
      "grad_norm": 15.8125,
      "learning_rate": 3.021571712604821e-05,
      "loss": 0.9459,
      "num_input_tokens_seen": 38152064,
      "step": 65770
    },
    {
      "epoch": 9.796693476318142,
      "grad_norm": 9.875,
      "learning_rate": 3.0212539164310276e-05,
      "loss": 1.4197,
      "num_input_tokens_seen": 38154976,
      "step": 65775
    },
    {
      "epoch": 9.7974381888591,
      "grad_norm": 12.6875,
      "learning_rate": 3.0209361114511796e-05,
      "loss": 1.0037,
      "num_input_tokens_seen": 38157888,
      "step": 65780
    },
    {
      "epoch": 9.79818290140006,
      "grad_norm": 15.5625,
      "learning_rate": 3.0206182976706447e-05,
      "loss": 1.2142,
      "num_input_tokens_seen": 38160928,
      "step": 65785
    },
    {
      "epoch": 9.798927613941018,
      "grad_norm": 10.5,
      "learning_rate": 3.0203004750947938e-05,
      "loss": 0.9258,
      "num_input_tokens_seen": 38163648,
      "step": 65790
    },
    {
      "epoch": 9.799672326481979,
      "grad_norm": 13.1875,
      "learning_rate": 3.0199826437289947e-05,
      "loss": 1.2732,
      "num_input_tokens_seen": 38166624,
      "step": 65795
    },
    {
      "epoch": 9.800417039022937,
      "grad_norm": 8.0,
      "learning_rate": 3.0196648035786173e-05,
      "loss": 0.972,
      "num_input_tokens_seen": 38169600,
      "step": 65800
    },
    {
      "epoch": 9.801161751563896,
      "grad_norm": 19.125,
      "learning_rate": 3.019346954649031e-05,
      "loss": 1.1486,
      "num_input_tokens_seen": 38172480,
      "step": 65805
    },
    {
      "epoch": 9.801906464104855,
      "grad_norm": 10.125,
      "learning_rate": 3.0190290969456063e-05,
      "loss": 0.9774,
      "num_input_tokens_seen": 38175136,
      "step": 65810
    },
    {
      "epoch": 9.802651176645815,
      "grad_norm": 10.9375,
      "learning_rate": 3.0187112304737125e-05,
      "loss": 1.2536,
      "num_input_tokens_seen": 38178112,
      "step": 65815
    },
    {
      "epoch": 9.803395889186774,
      "grad_norm": 20.125,
      "learning_rate": 3.0183933552387188e-05,
      "loss": 1.1403,
      "num_input_tokens_seen": 38180864,
      "step": 65820
    },
    {
      "epoch": 9.804140601727733,
      "grad_norm": 7.53125,
      "learning_rate": 3.0180754712459973e-05,
      "loss": 1.0788,
      "num_input_tokens_seen": 38183904,
      "step": 65825
    },
    {
      "epoch": 9.804885314268692,
      "grad_norm": 15.375,
      "learning_rate": 3.0177575785009172e-05,
      "loss": 1.1576,
      "num_input_tokens_seen": 38186752,
      "step": 65830
    },
    {
      "epoch": 9.805630026809652,
      "grad_norm": 14.1875,
      "learning_rate": 3.017439677008848e-05,
      "loss": 1.2991,
      "num_input_tokens_seen": 38189632,
      "step": 65835
    },
    {
      "epoch": 9.80637473935061,
      "grad_norm": 8.3125,
      "learning_rate": 3.0171217667751617e-05,
      "loss": 1.0158,
      "num_input_tokens_seen": 38192608,
      "step": 65840
    },
    {
      "epoch": 9.80711945189157,
      "grad_norm": 12.0625,
      "learning_rate": 3.016803847805229e-05,
      "loss": 1.2295,
      "num_input_tokens_seen": 38195584,
      "step": 65845
    },
    {
      "epoch": 9.807864164432528,
      "grad_norm": 5.5,
      "learning_rate": 3.016485920104421e-05,
      "loss": 1.06,
      "num_input_tokens_seen": 38198624,
      "step": 65850
    },
    {
      "epoch": 9.808608876973489,
      "grad_norm": 10.6875,
      "learning_rate": 3.0161679836781076e-05,
      "loss": 0.9845,
      "num_input_tokens_seen": 38201408,
      "step": 65855
    },
    {
      "epoch": 9.809353589514448,
      "grad_norm": 6.1875,
      "learning_rate": 3.0158500385316612e-05,
      "loss": 0.9534,
      "num_input_tokens_seen": 38203968,
      "step": 65860
    },
    {
      "epoch": 9.810098302055406,
      "grad_norm": 14.5625,
      "learning_rate": 3.0155320846704526e-05,
      "loss": 1.3842,
      "num_input_tokens_seen": 38207008,
      "step": 65865
    },
    {
      "epoch": 9.810843014596365,
      "grad_norm": 16.75,
      "learning_rate": 3.015214122099853e-05,
      "loss": 1.4037,
      "num_input_tokens_seen": 38209600,
      "step": 65870
    },
    {
      "epoch": 9.811587727137326,
      "grad_norm": 13.625,
      "learning_rate": 3.0148961508252347e-05,
      "loss": 1.1647,
      "num_input_tokens_seen": 38212416,
      "step": 65875
    },
    {
      "epoch": 9.812332439678285,
      "grad_norm": 7.53125,
      "learning_rate": 3.0145781708519692e-05,
      "loss": 0.9612,
      "num_input_tokens_seen": 38215456,
      "step": 65880
    },
    {
      "epoch": 9.813077152219243,
      "grad_norm": 18.125,
      "learning_rate": 3.0142601821854288e-05,
      "loss": 1.3316,
      "num_input_tokens_seen": 38218624,
      "step": 65885
    },
    {
      "epoch": 9.813821864760202,
      "grad_norm": 5.6875,
      "learning_rate": 3.0139421848309852e-05,
      "loss": 1.2404,
      "num_input_tokens_seen": 38221440,
      "step": 65890
    },
    {
      "epoch": 9.814566577301163,
      "grad_norm": 5.09375,
      "learning_rate": 3.0136241787940107e-05,
      "loss": 1.1794,
      "num_input_tokens_seen": 38224608,
      "step": 65895
    },
    {
      "epoch": 9.815311289842121,
      "grad_norm": 10.25,
      "learning_rate": 3.0133061640798776e-05,
      "loss": 0.9269,
      "num_input_tokens_seen": 38227520,
      "step": 65900
    },
    {
      "epoch": 9.81605600238308,
      "grad_norm": 5.65625,
      "learning_rate": 3.0129881406939587e-05,
      "loss": 1.0252,
      "num_input_tokens_seen": 38230176,
      "step": 65905
    },
    {
      "epoch": 9.816800714924039,
      "grad_norm": 8.375,
      "learning_rate": 3.012670108641626e-05,
      "loss": 1.3018,
      "num_input_tokens_seen": 38233376,
      "step": 65910
    },
    {
      "epoch": 9.817545427465,
      "grad_norm": 9.125,
      "learning_rate": 3.012352067928253e-05,
      "loss": 1.6531,
      "num_input_tokens_seen": 38236416,
      "step": 65915
    },
    {
      "epoch": 9.818290140005958,
      "grad_norm": 14.8125,
      "learning_rate": 3.0120340185592132e-05,
      "loss": 1.3218,
      "num_input_tokens_seen": 38239232,
      "step": 65920
    },
    {
      "epoch": 9.819034852546917,
      "grad_norm": 9.3125,
      "learning_rate": 3.0117159605398786e-05,
      "loss": 1.1116,
      "num_input_tokens_seen": 38242176,
      "step": 65925
    },
    {
      "epoch": 9.819779565087876,
      "grad_norm": 10.5625,
      "learning_rate": 3.0113978938756237e-05,
      "loss": 1.0954,
      "num_input_tokens_seen": 38245120,
      "step": 65930
    },
    {
      "epoch": 9.820524277628834,
      "grad_norm": 12.375,
      "learning_rate": 3.0110798185718202e-05,
      "loss": 1.1749,
      "num_input_tokens_seen": 38248128,
      "step": 65935
    },
    {
      "epoch": 9.821268990169795,
      "grad_norm": 11.4375,
      "learning_rate": 3.0107617346338422e-05,
      "loss": 0.9522,
      "num_input_tokens_seen": 38250976,
      "step": 65940
    },
    {
      "epoch": 9.822013702710754,
      "grad_norm": 12.75,
      "learning_rate": 3.0104436420670644e-05,
      "loss": 1.2698,
      "num_input_tokens_seen": 38253696,
      "step": 65945
    },
    {
      "epoch": 9.822758415251712,
      "grad_norm": 9.625,
      "learning_rate": 3.0101255408768603e-05,
      "loss": 0.9861,
      "num_input_tokens_seen": 38256384,
      "step": 65950
    },
    {
      "epoch": 9.823503127792671,
      "grad_norm": 13.9375,
      "learning_rate": 3.0098074310686042e-05,
      "loss": 1.0241,
      "num_input_tokens_seen": 38259424,
      "step": 65955
    },
    {
      "epoch": 9.824247840333632,
      "grad_norm": 6.375,
      "learning_rate": 3.0094893126476686e-05,
      "loss": 0.9999,
      "num_input_tokens_seen": 38262016,
      "step": 65960
    },
    {
      "epoch": 9.82499255287459,
      "grad_norm": 13.9375,
      "learning_rate": 3.0091711856194295e-05,
      "loss": 0.9279,
      "num_input_tokens_seen": 38264928,
      "step": 65965
    },
    {
      "epoch": 9.82573726541555,
      "grad_norm": 15.125,
      "learning_rate": 3.0088530499892605e-05,
      "loss": 1.393,
      "num_input_tokens_seen": 38267872,
      "step": 65970
    },
    {
      "epoch": 9.826481977956508,
      "grad_norm": 7.78125,
      "learning_rate": 3.008534905762536e-05,
      "loss": 1.0634,
      "num_input_tokens_seen": 38270688,
      "step": 65975
    },
    {
      "epoch": 9.827226690497469,
      "grad_norm": 5.0,
      "learning_rate": 3.0082167529446314e-05,
      "loss": 0.8163,
      "num_input_tokens_seen": 38273472,
      "step": 65980
    },
    {
      "epoch": 9.827971403038427,
      "grad_norm": 12.8125,
      "learning_rate": 3.0078985915409214e-05,
      "loss": 1.145,
      "num_input_tokens_seen": 38276256,
      "step": 65985
    },
    {
      "epoch": 9.828716115579386,
      "grad_norm": 8.125,
      "learning_rate": 3.0075804215567817e-05,
      "loss": 1.0105,
      "num_input_tokens_seen": 38279520,
      "step": 65990
    },
    {
      "epoch": 9.829460828120345,
      "grad_norm": 13.875,
      "learning_rate": 3.0072622429975856e-05,
      "loss": 0.9635,
      "num_input_tokens_seen": 38282304,
      "step": 65995
    },
    {
      "epoch": 9.830205540661305,
      "grad_norm": 12.0625,
      "learning_rate": 3.006944055868709e-05,
      "loss": 1.052,
      "num_input_tokens_seen": 38285248,
      "step": 66000
    },
    {
      "epoch": 9.830950253202264,
      "grad_norm": 14.125,
      "learning_rate": 3.0066258601755288e-05,
      "loss": 1.4701,
      "num_input_tokens_seen": 38288192,
      "step": 66005
    },
    {
      "epoch": 9.831694965743223,
      "grad_norm": 10.5625,
      "learning_rate": 3.0063076559234192e-05,
      "loss": 1.6016,
      "num_input_tokens_seen": 38291104,
      "step": 66010
    },
    {
      "epoch": 9.832439678284182,
      "grad_norm": 11.875,
      "learning_rate": 3.0059894431177565e-05,
      "loss": 1.2314,
      "num_input_tokens_seen": 38293856,
      "step": 66015
    },
    {
      "epoch": 9.833184390825142,
      "grad_norm": 12.0,
      "learning_rate": 3.0056712217639165e-05,
      "loss": 1.0937,
      "num_input_tokens_seen": 38296736,
      "step": 66020
    },
    {
      "epoch": 9.833929103366101,
      "grad_norm": 12.5625,
      "learning_rate": 3.005352991867275e-05,
      "loss": 1.1642,
      "num_input_tokens_seen": 38300128,
      "step": 66025
    },
    {
      "epoch": 9.83467381590706,
      "grad_norm": 5.53125,
      "learning_rate": 3.0050347534332084e-05,
      "loss": 1.0263,
      "num_input_tokens_seen": 38302816,
      "step": 66030
    },
    {
      "epoch": 9.835418528448018,
      "grad_norm": 9.25,
      "learning_rate": 3.004716506467093e-05,
      "loss": 0.7646,
      "num_input_tokens_seen": 38305600,
      "step": 66035
    },
    {
      "epoch": 9.836163240988979,
      "grad_norm": 6.0625,
      "learning_rate": 3.0043982509743052e-05,
      "loss": 1.0668,
      "num_input_tokens_seen": 38308800,
      "step": 66040
    },
    {
      "epoch": 9.836907953529938,
      "grad_norm": 16.125,
      "learning_rate": 3.004079986960221e-05,
      "loss": 1.1041,
      "num_input_tokens_seen": 38311968,
      "step": 66045
    },
    {
      "epoch": 9.837652666070897,
      "grad_norm": 14.9375,
      "learning_rate": 3.0037617144302188e-05,
      "loss": 1.0435,
      "num_input_tokens_seen": 38314592,
      "step": 66050
    },
    {
      "epoch": 9.838397378611855,
      "grad_norm": 9.625,
      "learning_rate": 3.0034434333896737e-05,
      "loss": 1.3623,
      "num_input_tokens_seen": 38317504,
      "step": 66055
    },
    {
      "epoch": 9.839142091152816,
      "grad_norm": 13.75,
      "learning_rate": 3.003125143843964e-05,
      "loss": 0.9839,
      "num_input_tokens_seen": 38320256,
      "step": 66060
    },
    {
      "epoch": 9.839886803693775,
      "grad_norm": 9.4375,
      "learning_rate": 3.002806845798466e-05,
      "loss": 0.9189,
      "num_input_tokens_seen": 38322848,
      "step": 66065
    },
    {
      "epoch": 9.840631516234733,
      "grad_norm": 17.0,
      "learning_rate": 3.002488539258557e-05,
      "loss": 1.284,
      "num_input_tokens_seen": 38325856,
      "step": 66070
    },
    {
      "epoch": 9.841376228775692,
      "grad_norm": 9.0,
      "learning_rate": 3.0021702242296153e-05,
      "loss": 1.1868,
      "num_input_tokens_seen": 38328800,
      "step": 66075
    },
    {
      "epoch": 9.842120941316653,
      "grad_norm": 9.625,
      "learning_rate": 3.0018519007170177e-05,
      "loss": 1.2675,
      "num_input_tokens_seen": 38331904,
      "step": 66080
    },
    {
      "epoch": 9.842865653857611,
      "grad_norm": 15.8125,
      "learning_rate": 3.0015335687261425e-05,
      "loss": 1.0796,
      "num_input_tokens_seen": 38334912,
      "step": 66085
    },
    {
      "epoch": 9.84361036639857,
      "grad_norm": 12.1875,
      "learning_rate": 3.001215228262368e-05,
      "loss": 1.5387,
      "num_input_tokens_seen": 38337696,
      "step": 66090
    },
    {
      "epoch": 9.844355078939529,
      "grad_norm": 4.90625,
      "learning_rate": 3.000896879331071e-05,
      "loss": 0.9044,
      "num_input_tokens_seen": 38340512,
      "step": 66095
    },
    {
      "epoch": 9.84509979148049,
      "grad_norm": 7.4375,
      "learning_rate": 3.0005785219376304e-05,
      "loss": 1.0206,
      "num_input_tokens_seen": 38343680,
      "step": 66100
    },
    {
      "epoch": 9.845844504021448,
      "grad_norm": 12.875,
      "learning_rate": 3.000260156087424e-05,
      "loss": 0.8282,
      "num_input_tokens_seen": 38346368,
      "step": 66105
    },
    {
      "epoch": 9.846589216562407,
      "grad_norm": 7.125,
      "learning_rate": 2.999941781785831e-05,
      "loss": 1.3276,
      "num_input_tokens_seen": 38349216,
      "step": 66110
    },
    {
      "epoch": 9.847333929103366,
      "grad_norm": 7.75,
      "learning_rate": 2.9996233990382296e-05,
      "loss": 0.7309,
      "num_input_tokens_seen": 38351840,
      "step": 66115
    },
    {
      "epoch": 9.848078641644324,
      "grad_norm": 7.625,
      "learning_rate": 2.9993050078499997e-05,
      "loss": 0.8715,
      "num_input_tokens_seen": 38354816,
      "step": 66120
    },
    {
      "epoch": 9.848823354185285,
      "grad_norm": 9.5625,
      "learning_rate": 2.9989866082265177e-05,
      "loss": 1.2193,
      "num_input_tokens_seen": 38357696,
      "step": 66125
    },
    {
      "epoch": 9.849568066726244,
      "grad_norm": 17.125,
      "learning_rate": 2.9986682001731647e-05,
      "loss": 1.3171,
      "num_input_tokens_seen": 38360704,
      "step": 66130
    },
    {
      "epoch": 9.850312779267203,
      "grad_norm": 12.0,
      "learning_rate": 2.99834978369532e-05,
      "loss": 1.1703,
      "num_input_tokens_seen": 38363776,
      "step": 66135
    },
    {
      "epoch": 9.851057491808161,
      "grad_norm": 9.6875,
      "learning_rate": 2.998031358798361e-05,
      "loss": 1.2191,
      "num_input_tokens_seen": 38366816,
      "step": 66140
    },
    {
      "epoch": 9.851802204349122,
      "grad_norm": 11.6875,
      "learning_rate": 2.997712925487669e-05,
      "loss": 0.8994,
      "num_input_tokens_seen": 38369920,
      "step": 66145
    },
    {
      "epoch": 9.85254691689008,
      "grad_norm": 15.3125,
      "learning_rate": 2.9973944837686228e-05,
      "loss": 1.2822,
      "num_input_tokens_seen": 38373120,
      "step": 66150
    },
    {
      "epoch": 9.85329162943104,
      "grad_norm": 7.625,
      "learning_rate": 2.9970760336466032e-05,
      "loss": 0.9849,
      "num_input_tokens_seen": 38375872,
      "step": 66155
    },
    {
      "epoch": 9.854036341971998,
      "grad_norm": 14.1875,
      "learning_rate": 2.9967575751269878e-05,
      "loss": 1.1425,
      "num_input_tokens_seen": 38378624,
      "step": 66160
    },
    {
      "epoch": 9.854781054512959,
      "grad_norm": 11.1875,
      "learning_rate": 2.9964391082151587e-05,
      "loss": 1.1902,
      "num_input_tokens_seen": 38381216,
      "step": 66165
    },
    {
      "epoch": 9.855525767053917,
      "grad_norm": 7.125,
      "learning_rate": 2.9961206329164952e-05,
      "loss": 1.0088,
      "num_input_tokens_seen": 38384000,
      "step": 66170
    },
    {
      "epoch": 9.856270479594876,
      "grad_norm": 10.4375,
      "learning_rate": 2.9958021492363787e-05,
      "loss": 1.4435,
      "num_input_tokens_seen": 38386560,
      "step": 66175
    },
    {
      "epoch": 9.857015192135835,
      "grad_norm": 8.8125,
      "learning_rate": 2.9954836571801875e-05,
      "loss": 0.9438,
      "num_input_tokens_seen": 38389312,
      "step": 66180
    },
    {
      "epoch": 9.857759904676795,
      "grad_norm": 15.0,
      "learning_rate": 2.9951651567533046e-05,
      "loss": 1.3391,
      "num_input_tokens_seen": 38392064,
      "step": 66185
    },
    {
      "epoch": 9.858504617217754,
      "grad_norm": 7.53125,
      "learning_rate": 2.994846647961109e-05,
      "loss": 1.0509,
      "num_input_tokens_seen": 38394976,
      "step": 66190
    },
    {
      "epoch": 9.859249329758713,
      "grad_norm": 7.09375,
      "learning_rate": 2.9945281308089824e-05,
      "loss": 0.8376,
      "num_input_tokens_seen": 38398208,
      "step": 66195
    },
    {
      "epoch": 9.859994042299672,
      "grad_norm": 11.1875,
      "learning_rate": 2.9942096053023055e-05,
      "loss": 1.5063,
      "num_input_tokens_seen": 38401344,
      "step": 66200
    },
    {
      "epoch": 9.860738754840632,
      "grad_norm": 5.5625,
      "learning_rate": 2.9938910714464596e-05,
      "loss": 0.8565,
      "num_input_tokens_seen": 38404320,
      "step": 66205
    },
    {
      "epoch": 9.861483467381591,
      "grad_norm": 13.0,
      "learning_rate": 2.9935725292468263e-05,
      "loss": 1.1372,
      "num_input_tokens_seen": 38407360,
      "step": 66210
    },
    {
      "epoch": 9.86222817992255,
      "grad_norm": 5.125,
      "learning_rate": 2.9932539787087872e-05,
      "loss": 0.952,
      "num_input_tokens_seen": 38410112,
      "step": 66215
    },
    {
      "epoch": 9.862972892463509,
      "grad_norm": 15.3125,
      "learning_rate": 2.9929354198377223e-05,
      "loss": 1.277,
      "num_input_tokens_seen": 38413344,
      "step": 66220
    },
    {
      "epoch": 9.863717605004469,
      "grad_norm": 18.75,
      "learning_rate": 2.9926168526390157e-05,
      "loss": 0.9496,
      "num_input_tokens_seen": 38416448,
      "step": 66225
    },
    {
      "epoch": 9.864462317545428,
      "grad_norm": 8.3125,
      "learning_rate": 2.9922982771180475e-05,
      "loss": 1.0777,
      "num_input_tokens_seen": 38419456,
      "step": 66230
    },
    {
      "epoch": 9.865207030086387,
      "grad_norm": 10.75,
      "learning_rate": 2.9919796932801996e-05,
      "loss": 0.8259,
      "num_input_tokens_seen": 38422432,
      "step": 66235
    },
    {
      "epoch": 9.865951742627345,
      "grad_norm": 12.25,
      "learning_rate": 2.9916611011308555e-05,
      "loss": 1.0455,
      "num_input_tokens_seen": 38425152,
      "step": 66240
    },
    {
      "epoch": 9.866696455168306,
      "grad_norm": 4.0625,
      "learning_rate": 2.9913425006753965e-05,
      "loss": 0.8569,
      "num_input_tokens_seen": 38428224,
      "step": 66245
    },
    {
      "epoch": 9.867441167709265,
      "grad_norm": 13.5625,
      "learning_rate": 2.9910238919192058e-05,
      "loss": 1.2128,
      "num_input_tokens_seen": 38431008,
      "step": 66250
    },
    {
      "epoch": 9.868185880250223,
      "grad_norm": 5.59375,
      "learning_rate": 2.9907052748676656e-05,
      "loss": 1.1865,
      "num_input_tokens_seen": 38433696,
      "step": 66255
    },
    {
      "epoch": 9.868930592791182,
      "grad_norm": 4.1875,
      "learning_rate": 2.9903866495261578e-05,
      "loss": 0.8324,
      "num_input_tokens_seen": 38436736,
      "step": 66260
    },
    {
      "epoch": 9.86967530533214,
      "grad_norm": 19.625,
      "learning_rate": 2.9900680159000666e-05,
      "loss": 1.0778,
      "num_input_tokens_seen": 38439456,
      "step": 66265
    },
    {
      "epoch": 9.870420017873101,
      "grad_norm": 9.0625,
      "learning_rate": 2.9897493739947736e-05,
      "loss": 1.0977,
      "num_input_tokens_seen": 38442272,
      "step": 66270
    },
    {
      "epoch": 9.87116473041406,
      "grad_norm": 16.25,
      "learning_rate": 2.9894307238156634e-05,
      "loss": 1.2633,
      "num_input_tokens_seen": 38444960,
      "step": 66275
    },
    {
      "epoch": 9.871909442955019,
      "grad_norm": 9.0,
      "learning_rate": 2.989112065368118e-05,
      "loss": 0.9767,
      "num_input_tokens_seen": 38447872,
      "step": 66280
    },
    {
      "epoch": 9.87265415549598,
      "grad_norm": 17.5,
      "learning_rate": 2.9887933986575218e-05,
      "loss": 1.2567,
      "num_input_tokens_seen": 38450816,
      "step": 66285
    },
    {
      "epoch": 9.873398868036938,
      "grad_norm": 7.75,
      "learning_rate": 2.9884747236892578e-05,
      "loss": 0.8451,
      "num_input_tokens_seen": 38453760,
      "step": 66290
    },
    {
      "epoch": 9.874143580577897,
      "grad_norm": 8.8125,
      "learning_rate": 2.9881560404687103e-05,
      "loss": 0.9919,
      "num_input_tokens_seen": 38456736,
      "step": 66295
    },
    {
      "epoch": 9.874888293118856,
      "grad_norm": 4.75,
      "learning_rate": 2.9878373490012617e-05,
      "loss": 0.9991,
      "num_input_tokens_seen": 38459520,
      "step": 66300
    },
    {
      "epoch": 9.875633005659815,
      "grad_norm": 3.734375,
      "learning_rate": 2.9875186492922973e-05,
      "loss": 1.0293,
      "num_input_tokens_seen": 38462560,
      "step": 66305
    },
    {
      "epoch": 9.876377718200775,
      "grad_norm": 16.875,
      "learning_rate": 2.9871999413472006e-05,
      "loss": 1.1463,
      "num_input_tokens_seen": 38465568,
      "step": 66310
    },
    {
      "epoch": 9.877122430741734,
      "grad_norm": 6.875,
      "learning_rate": 2.9868812251713564e-05,
      "loss": 1.1151,
      "num_input_tokens_seen": 38468384,
      "step": 66315
    },
    {
      "epoch": 9.877867143282693,
      "grad_norm": 5.5625,
      "learning_rate": 2.9865625007701487e-05,
      "loss": 0.9367,
      "num_input_tokens_seen": 38471328,
      "step": 66320
    },
    {
      "epoch": 9.878611855823651,
      "grad_norm": 9.75,
      "learning_rate": 2.986243768148962e-05,
      "loss": 1.1454,
      "num_input_tokens_seen": 38474304,
      "step": 66325
    },
    {
      "epoch": 9.879356568364612,
      "grad_norm": 9.875,
      "learning_rate": 2.9859250273131812e-05,
      "loss": 1.1324,
      "num_input_tokens_seen": 38477120,
      "step": 66330
    },
    {
      "epoch": 9.88010128090557,
      "grad_norm": 17.625,
      "learning_rate": 2.985606278268191e-05,
      "loss": 0.8902,
      "num_input_tokens_seen": 38480096,
      "step": 66335
    },
    {
      "epoch": 9.88084599344653,
      "grad_norm": 8.3125,
      "learning_rate": 2.985287521019376e-05,
      "loss": 0.8763,
      "num_input_tokens_seen": 38482784,
      "step": 66340
    },
    {
      "epoch": 9.881590705987488,
      "grad_norm": 11.9375,
      "learning_rate": 2.984968755572121e-05,
      "loss": 0.9672,
      "num_input_tokens_seen": 38485856,
      "step": 66345
    },
    {
      "epoch": 9.882335418528449,
      "grad_norm": 11.4375,
      "learning_rate": 2.9846499819318124e-05,
      "loss": 0.981,
      "num_input_tokens_seen": 38488512,
      "step": 66350
    },
    {
      "epoch": 9.883080131069407,
      "grad_norm": 5.375,
      "learning_rate": 2.9843312001038353e-05,
      "loss": 0.9709,
      "num_input_tokens_seen": 38491296,
      "step": 66355
    },
    {
      "epoch": 9.883824843610366,
      "grad_norm": 8.0,
      "learning_rate": 2.9840124100935744e-05,
      "loss": 1.0805,
      "num_input_tokens_seen": 38494304,
      "step": 66360
    },
    {
      "epoch": 9.884569556151325,
      "grad_norm": 7.78125,
      "learning_rate": 2.9836936119064156e-05,
      "loss": 1.1312,
      "num_input_tokens_seen": 38497312,
      "step": 66365
    },
    {
      "epoch": 9.885314268692285,
      "grad_norm": 12.4375,
      "learning_rate": 2.983374805547745e-05,
      "loss": 1.1074,
      "num_input_tokens_seen": 38500352,
      "step": 66370
    },
    {
      "epoch": 9.886058981233244,
      "grad_norm": 11.5625,
      "learning_rate": 2.983055991022949e-05,
      "loss": 0.8725,
      "num_input_tokens_seen": 38503232,
      "step": 66375
    },
    {
      "epoch": 9.886803693774203,
      "grad_norm": 11.3125,
      "learning_rate": 2.9827371683374116e-05,
      "loss": 1.1476,
      "num_input_tokens_seen": 38505856,
      "step": 66380
    },
    {
      "epoch": 9.887548406315162,
      "grad_norm": 10.625,
      "learning_rate": 2.9824183374965214e-05,
      "loss": 1.0163,
      "num_input_tokens_seen": 38508800,
      "step": 66385
    },
    {
      "epoch": 9.888293118856122,
      "grad_norm": 16.125,
      "learning_rate": 2.982099498505664e-05,
      "loss": 1.1739,
      "num_input_tokens_seen": 38511712,
      "step": 66390
    },
    {
      "epoch": 9.889037831397081,
      "grad_norm": 3.875,
      "learning_rate": 2.9817806513702244e-05,
      "loss": 0.9059,
      "num_input_tokens_seen": 38514400,
      "step": 66395
    },
    {
      "epoch": 9.88978254393804,
      "grad_norm": 14.75,
      "learning_rate": 2.9814617960955908e-05,
      "loss": 1.4196,
      "num_input_tokens_seen": 38517888,
      "step": 66400
    },
    {
      "epoch": 9.890527256478999,
      "grad_norm": 11.4375,
      "learning_rate": 2.9811429326871498e-05,
      "loss": 0.9731,
      "num_input_tokens_seen": 38520864,
      "step": 66405
    },
    {
      "epoch": 9.891271969019959,
      "grad_norm": 13.25,
      "learning_rate": 2.9808240611502873e-05,
      "loss": 1.0847,
      "num_input_tokens_seen": 38524032,
      "step": 66410
    },
    {
      "epoch": 9.892016681560918,
      "grad_norm": 18.5,
      "learning_rate": 2.9805051814903923e-05,
      "loss": 1.0582,
      "num_input_tokens_seen": 38527008,
      "step": 66415
    },
    {
      "epoch": 9.892761394101877,
      "grad_norm": 5.3125,
      "learning_rate": 2.98018629371285e-05,
      "loss": 1.1176,
      "num_input_tokens_seen": 38529824,
      "step": 66420
    },
    {
      "epoch": 9.893506106642835,
      "grad_norm": 11.0,
      "learning_rate": 2.979867397823048e-05,
      "loss": 0.9692,
      "num_input_tokens_seen": 38532928,
      "step": 66425
    },
    {
      "epoch": 9.894250819183796,
      "grad_norm": 8.5625,
      "learning_rate": 2.979548493826374e-05,
      "loss": 0.8735,
      "num_input_tokens_seen": 38535520,
      "step": 66430
    },
    {
      "epoch": 9.894995531724755,
      "grad_norm": 7.15625,
      "learning_rate": 2.9792295817282157e-05,
      "loss": 1.1083,
      "num_input_tokens_seen": 38538848,
      "step": 66435
    },
    {
      "epoch": 9.895740244265713,
      "grad_norm": 7.84375,
      "learning_rate": 2.9789106615339603e-05,
      "loss": 1.1924,
      "num_input_tokens_seen": 38541664,
      "step": 66440
    },
    {
      "epoch": 9.896484956806672,
      "grad_norm": 5.9375,
      "learning_rate": 2.9785917332489965e-05,
      "loss": 1.1226,
      "num_input_tokens_seen": 38544416,
      "step": 66445
    },
    {
      "epoch": 9.897229669347631,
      "grad_norm": 11.8125,
      "learning_rate": 2.9782727968787116e-05,
      "loss": 1.1218,
      "num_input_tokens_seen": 38547360,
      "step": 66450
    },
    {
      "epoch": 9.897974381888591,
      "grad_norm": 6.5,
      "learning_rate": 2.9779538524284943e-05,
      "loss": 0.7852,
      "num_input_tokens_seen": 38550464,
      "step": 66455
    },
    {
      "epoch": 9.89871909442955,
      "grad_norm": 9.625,
      "learning_rate": 2.9776348999037322e-05,
      "loss": 1.1795,
      "num_input_tokens_seen": 38553248,
      "step": 66460
    },
    {
      "epoch": 9.899463806970509,
      "grad_norm": 11.5625,
      "learning_rate": 2.9773159393098137e-05,
      "loss": 1.3275,
      "num_input_tokens_seen": 38555872,
      "step": 66465
    },
    {
      "epoch": 9.900208519511468,
      "grad_norm": 7.75,
      "learning_rate": 2.9769969706521277e-05,
      "loss": 0.9431,
      "num_input_tokens_seen": 38558752,
      "step": 66470
    },
    {
      "epoch": 9.900953232052428,
      "grad_norm": 7.03125,
      "learning_rate": 2.9766779939360623e-05,
      "loss": 1.0537,
      "num_input_tokens_seen": 38561664,
      "step": 66475
    },
    {
      "epoch": 9.901697944593387,
      "grad_norm": 10.25,
      "learning_rate": 2.976359009167007e-05,
      "loss": 0.7202,
      "num_input_tokens_seen": 38564736,
      "step": 66480
    },
    {
      "epoch": 9.902442657134346,
      "grad_norm": 5.40625,
      "learning_rate": 2.976040016350351e-05,
      "loss": 1.0183,
      "num_input_tokens_seen": 38567872,
      "step": 66485
    },
    {
      "epoch": 9.903187369675305,
      "grad_norm": 24.75,
      "learning_rate": 2.9757210154914816e-05,
      "loss": 1.3863,
      "num_input_tokens_seen": 38570880,
      "step": 66490
    },
    {
      "epoch": 9.903932082216265,
      "grad_norm": 9.875,
      "learning_rate": 2.9754020065957905e-05,
      "loss": 0.9372,
      "num_input_tokens_seen": 38574016,
      "step": 66495
    },
    {
      "epoch": 9.904676794757224,
      "grad_norm": 9.875,
      "learning_rate": 2.9750829896686645e-05,
      "loss": 0.9042,
      "num_input_tokens_seen": 38576896,
      "step": 66500
    },
    {
      "epoch": 9.905421507298183,
      "grad_norm": 9.4375,
      "learning_rate": 2.9747639647154947e-05,
      "loss": 1.2012,
      "num_input_tokens_seen": 38579616,
      "step": 66505
    },
    {
      "epoch": 9.906166219839141,
      "grad_norm": 12.25,
      "learning_rate": 2.97444493174167e-05,
      "loss": 1.0646,
      "num_input_tokens_seen": 38582272,
      "step": 66510
    },
    {
      "epoch": 9.906910932380102,
      "grad_norm": 12.4375,
      "learning_rate": 2.9741258907525805e-05,
      "loss": 1.4972,
      "num_input_tokens_seen": 38584960,
      "step": 66515
    },
    {
      "epoch": 9.90765564492106,
      "grad_norm": 23.625,
      "learning_rate": 2.9738068417536165e-05,
      "loss": 1.4376,
      "num_input_tokens_seen": 38587808,
      "step": 66520
    },
    {
      "epoch": 9.90840035746202,
      "grad_norm": 19.75,
      "learning_rate": 2.9734877847501664e-05,
      "loss": 1.0633,
      "num_input_tokens_seen": 38590592,
      "step": 66525
    },
    {
      "epoch": 9.909145070002978,
      "grad_norm": 6.96875,
      "learning_rate": 2.973168719747622e-05,
      "loss": 0.8458,
      "num_input_tokens_seen": 38593568,
      "step": 66530
    },
    {
      "epoch": 9.909889782543939,
      "grad_norm": 15.875,
      "learning_rate": 2.9728496467513734e-05,
      "loss": 1.1155,
      "num_input_tokens_seen": 38596160,
      "step": 66535
    },
    {
      "epoch": 9.910634495084897,
      "grad_norm": 9.5,
      "learning_rate": 2.9725305657668102e-05,
      "loss": 0.8938,
      "num_input_tokens_seen": 38599296,
      "step": 66540
    },
    {
      "epoch": 9.911379207625856,
      "grad_norm": 6.90625,
      "learning_rate": 2.9722114767993226e-05,
      "loss": 1.0095,
      "num_input_tokens_seen": 38602176,
      "step": 66545
    },
    {
      "epoch": 9.912123920166815,
      "grad_norm": 10.0,
      "learning_rate": 2.971892379854303e-05,
      "loss": 1.1949,
      "num_input_tokens_seen": 38604960,
      "step": 66550
    },
    {
      "epoch": 9.912868632707776,
      "grad_norm": 16.125,
      "learning_rate": 2.9715732749371412e-05,
      "loss": 1.1414,
      "num_input_tokens_seen": 38608128,
      "step": 66555
    },
    {
      "epoch": 9.913613345248734,
      "grad_norm": 8.625,
      "learning_rate": 2.971254162053228e-05,
      "loss": 1.0526,
      "num_input_tokens_seen": 38610976,
      "step": 66560
    },
    {
      "epoch": 9.914358057789693,
      "grad_norm": 5.90625,
      "learning_rate": 2.9709350412079544e-05,
      "loss": 0.8953,
      "num_input_tokens_seen": 38613600,
      "step": 66565
    },
    {
      "epoch": 9.915102770330652,
      "grad_norm": 12.3125,
      "learning_rate": 2.9706159124067123e-05,
      "loss": 1.1092,
      "num_input_tokens_seen": 38616544,
      "step": 66570
    },
    {
      "epoch": 9.915847482871612,
      "grad_norm": 5.375,
      "learning_rate": 2.9702967756548927e-05,
      "loss": 1.0514,
      "num_input_tokens_seen": 38619264,
      "step": 66575
    },
    {
      "epoch": 9.916592195412571,
      "grad_norm": 11.3125,
      "learning_rate": 2.969977630957887e-05,
      "loss": 1.0466,
      "num_input_tokens_seen": 38622208,
      "step": 66580
    },
    {
      "epoch": 9.91733690795353,
      "grad_norm": 13.625,
      "learning_rate": 2.9696584783210874e-05,
      "loss": 1.022,
      "num_input_tokens_seen": 38625056,
      "step": 66585
    },
    {
      "epoch": 9.918081620494489,
      "grad_norm": 12.875,
      "learning_rate": 2.969339317749884e-05,
      "loss": 1.0673,
      "num_input_tokens_seen": 38627904,
      "step": 66590
    },
    {
      "epoch": 9.91882633303545,
      "grad_norm": 8.1875,
      "learning_rate": 2.9690201492496704e-05,
      "loss": 0.9899,
      "num_input_tokens_seen": 38630880,
      "step": 66595
    },
    {
      "epoch": 9.919571045576408,
      "grad_norm": 12.0625,
      "learning_rate": 2.968700972825838e-05,
      "loss": 1.2361,
      "num_input_tokens_seen": 38633888,
      "step": 66600
    },
    {
      "epoch": 9.920315758117367,
      "grad_norm": 11.25,
      "learning_rate": 2.9683817884837788e-05,
      "loss": 1.125,
      "num_input_tokens_seen": 38636608,
      "step": 66605
    },
    {
      "epoch": 9.921060470658325,
      "grad_norm": 13.3125,
      "learning_rate": 2.9680625962288856e-05,
      "loss": 1.1287,
      "num_input_tokens_seen": 38639200,
      "step": 66610
    },
    {
      "epoch": 9.921805183199286,
      "grad_norm": 13.6875,
      "learning_rate": 2.9677433960665512e-05,
      "loss": 1.253,
      "num_input_tokens_seen": 38642240,
      "step": 66615
    },
    {
      "epoch": 9.922549895740245,
      "grad_norm": 7.78125,
      "learning_rate": 2.967424188002167e-05,
      "loss": 1.0348,
      "num_input_tokens_seen": 38645088,
      "step": 66620
    },
    {
      "epoch": 9.923294608281203,
      "grad_norm": 8.0,
      "learning_rate": 2.967104972041126e-05,
      "loss": 0.9521,
      "num_input_tokens_seen": 38647808,
      "step": 66625
    },
    {
      "epoch": 9.924039320822162,
      "grad_norm": 7.6875,
      "learning_rate": 2.9667857481888218e-05,
      "loss": 1.0579,
      "num_input_tokens_seen": 38650624,
      "step": 66630
    },
    {
      "epoch": 9.924784033363121,
      "grad_norm": 9.5625,
      "learning_rate": 2.9664665164506455e-05,
      "loss": 1.0875,
      "num_input_tokens_seen": 38653376,
      "step": 66635
    },
    {
      "epoch": 9.925528745904082,
      "grad_norm": 7.375,
      "learning_rate": 2.9661472768319924e-05,
      "loss": 1.2666,
      "num_input_tokens_seen": 38656160,
      "step": 66640
    },
    {
      "epoch": 9.92627345844504,
      "grad_norm": 14.75,
      "learning_rate": 2.9658280293382545e-05,
      "loss": 1.1624,
      "num_input_tokens_seen": 38658944,
      "step": 66645
    },
    {
      "epoch": 9.927018170985999,
      "grad_norm": 4.875,
      "learning_rate": 2.9655087739748267e-05,
      "loss": 1.1449,
      "num_input_tokens_seen": 38662176,
      "step": 66650
    },
    {
      "epoch": 9.927762883526958,
      "grad_norm": 5.9375,
      "learning_rate": 2.9651895107471004e-05,
      "loss": 0.824,
      "num_input_tokens_seen": 38664992,
      "step": 66655
    },
    {
      "epoch": 9.928507596067918,
      "grad_norm": 13.625,
      "learning_rate": 2.96487023966047e-05,
      "loss": 1.3339,
      "num_input_tokens_seen": 38667744,
      "step": 66660
    },
    {
      "epoch": 9.929252308608877,
      "grad_norm": 12.3125,
      "learning_rate": 2.9645509607203294e-05,
      "loss": 1.188,
      "num_input_tokens_seen": 38670432,
      "step": 66665
    },
    {
      "epoch": 9.929997021149836,
      "grad_norm": 16.25,
      "learning_rate": 2.9642316739320724e-05,
      "loss": 1.2339,
      "num_input_tokens_seen": 38673440,
      "step": 66670
    },
    {
      "epoch": 9.930741733690795,
      "grad_norm": 4.75,
      "learning_rate": 2.9639123793010933e-05,
      "loss": 1.2009,
      "num_input_tokens_seen": 38676608,
      "step": 66675
    },
    {
      "epoch": 9.931486446231755,
      "grad_norm": 6.875,
      "learning_rate": 2.9635930768327856e-05,
      "loss": 1.1108,
      "num_input_tokens_seen": 38679680,
      "step": 66680
    },
    {
      "epoch": 9.932231158772714,
      "grad_norm": 11.6875,
      "learning_rate": 2.963273766532545e-05,
      "loss": 1.1381,
      "num_input_tokens_seen": 38682624,
      "step": 66685
    },
    {
      "epoch": 9.932975871313673,
      "grad_norm": 9.875,
      "learning_rate": 2.962954448405764e-05,
      "loss": 1.0794,
      "num_input_tokens_seen": 38685248,
      "step": 66690
    },
    {
      "epoch": 9.933720583854631,
      "grad_norm": 18.25,
      "learning_rate": 2.9626351224578386e-05,
      "loss": 1.3911,
      "num_input_tokens_seen": 38688256,
      "step": 66695
    },
    {
      "epoch": 9.934465296395592,
      "grad_norm": 7.84375,
      "learning_rate": 2.9623157886941633e-05,
      "loss": 1.1098,
      "num_input_tokens_seen": 38691072,
      "step": 66700
    },
    {
      "epoch": 9.93521000893655,
      "grad_norm": 11.4375,
      "learning_rate": 2.961996447120132e-05,
      "loss": 1.1608,
      "num_input_tokens_seen": 38694144,
      "step": 66705
    },
    {
      "epoch": 9.93595472147751,
      "grad_norm": 8.5625,
      "learning_rate": 2.9616770977411408e-05,
      "loss": 1.2817,
      "num_input_tokens_seen": 38696768,
      "step": 66710
    },
    {
      "epoch": 9.936699434018468,
      "grad_norm": 13.125,
      "learning_rate": 2.9613577405625838e-05,
      "loss": 0.8842,
      "num_input_tokens_seen": 38699456,
      "step": 66715
    },
    {
      "epoch": 9.937444146559429,
      "grad_norm": 9.875,
      "learning_rate": 2.961038375589857e-05,
      "loss": 1.0616,
      "num_input_tokens_seen": 38702528,
      "step": 66720
    },
    {
      "epoch": 9.938188859100388,
      "grad_norm": 7.78125,
      "learning_rate": 2.9607190028283548e-05,
      "loss": 0.9103,
      "num_input_tokens_seen": 38705856,
      "step": 66725
    },
    {
      "epoch": 9.938933571641346,
      "grad_norm": 14.1875,
      "learning_rate": 2.960399622283474e-05,
      "loss": 1.0542,
      "num_input_tokens_seen": 38708768,
      "step": 66730
    },
    {
      "epoch": 9.939678284182305,
      "grad_norm": 8.75,
      "learning_rate": 2.960080233960609e-05,
      "loss": 1.0171,
      "num_input_tokens_seen": 38711616,
      "step": 66735
    },
    {
      "epoch": 9.940422996723266,
      "grad_norm": 12.125,
      "learning_rate": 2.959760837865157e-05,
      "loss": 1.0307,
      "num_input_tokens_seen": 38714560,
      "step": 66740
    },
    {
      "epoch": 9.941167709264224,
      "grad_norm": 13.625,
      "learning_rate": 2.9594414340025118e-05,
      "loss": 1.0771,
      "num_input_tokens_seen": 38717504,
      "step": 66745
    },
    {
      "epoch": 9.941912421805183,
      "grad_norm": 23.125,
      "learning_rate": 2.9591220223780714e-05,
      "loss": 1.1376,
      "num_input_tokens_seen": 38720288,
      "step": 66750
    },
    {
      "epoch": 9.942657134346142,
      "grad_norm": 5.625,
      "learning_rate": 2.9588026029972305e-05,
      "loss": 0.8862,
      "num_input_tokens_seen": 38723072,
      "step": 66755
    },
    {
      "epoch": 9.943401846887102,
      "grad_norm": 6.15625,
      "learning_rate": 2.9584831758653865e-05,
      "loss": 0.9912,
      "num_input_tokens_seen": 38726144,
      "step": 66760
    },
    {
      "epoch": 9.944146559428061,
      "grad_norm": 12.0625,
      "learning_rate": 2.9581637409879344e-05,
      "loss": 0.8123,
      "num_input_tokens_seen": 38729184,
      "step": 66765
    },
    {
      "epoch": 9.94489127196902,
      "grad_norm": 9.3125,
      "learning_rate": 2.9578442983702716e-05,
      "loss": 1.4491,
      "num_input_tokens_seen": 38732096,
      "step": 66770
    },
    {
      "epoch": 9.945635984509979,
      "grad_norm": 10.125,
      "learning_rate": 2.9575248480177952e-05,
      "loss": 0.5791,
      "num_input_tokens_seen": 38734752,
      "step": 66775
    },
    {
      "epoch": 9.946380697050937,
      "grad_norm": 7.03125,
      "learning_rate": 2.9572053899359013e-05,
      "loss": 0.8001,
      "num_input_tokens_seen": 38737664,
      "step": 66780
    },
    {
      "epoch": 9.947125409591898,
      "grad_norm": 12.0625,
      "learning_rate": 2.9568859241299878e-05,
      "loss": 1.1977,
      "num_input_tokens_seen": 38740640,
      "step": 66785
    },
    {
      "epoch": 9.947870122132857,
      "grad_norm": 10.0,
      "learning_rate": 2.9565664506054503e-05,
      "loss": 1.2025,
      "num_input_tokens_seen": 38743904,
      "step": 66790
    },
    {
      "epoch": 9.948614834673815,
      "grad_norm": 10.5,
      "learning_rate": 2.9562469693676865e-05,
      "loss": 0.8519,
      "num_input_tokens_seen": 38747200,
      "step": 66795
    },
    {
      "epoch": 9.949359547214776,
      "grad_norm": 10.9375,
      "learning_rate": 2.9559274804220936e-05,
      "loss": 0.9635,
      "num_input_tokens_seen": 38749920,
      "step": 66800
    },
    {
      "epoch": 9.950104259755735,
      "grad_norm": 9.5,
      "learning_rate": 2.9556079837740697e-05,
      "loss": 1.1604,
      "num_input_tokens_seen": 38752736,
      "step": 66805
    },
    {
      "epoch": 9.950848972296694,
      "grad_norm": 15.5,
      "learning_rate": 2.9552884794290116e-05,
      "loss": 0.976,
      "num_input_tokens_seen": 38755584,
      "step": 66810
    },
    {
      "epoch": 9.951593684837652,
      "grad_norm": 14.0,
      "learning_rate": 2.954968967392318e-05,
      "loss": 1.1168,
      "num_input_tokens_seen": 38758368,
      "step": 66815
    },
    {
      "epoch": 9.952338397378611,
      "grad_norm": 13.4375,
      "learning_rate": 2.9546494476693865e-05,
      "loss": 1.1077,
      "num_input_tokens_seen": 38761184,
      "step": 66820
    },
    {
      "epoch": 9.953083109919572,
      "grad_norm": 10.625,
      "learning_rate": 2.954329920265614e-05,
      "loss": 1.2367,
      "num_input_tokens_seen": 38763936,
      "step": 66825
    },
    {
      "epoch": 9.95382782246053,
      "grad_norm": 9.125,
      "learning_rate": 2.9540103851863986e-05,
      "loss": 1.1143,
      "num_input_tokens_seen": 38766912,
      "step": 66830
    },
    {
      "epoch": 9.954572535001489,
      "grad_norm": 12.1875,
      "learning_rate": 2.95369084243714e-05,
      "loss": 1.3345,
      "num_input_tokens_seen": 38769888,
      "step": 66835
    },
    {
      "epoch": 9.955317247542448,
      "grad_norm": 8.125,
      "learning_rate": 2.9533712920232353e-05,
      "loss": 1.1578,
      "num_input_tokens_seen": 38773152,
      "step": 66840
    },
    {
      "epoch": 9.956061960083408,
      "grad_norm": 16.125,
      "learning_rate": 2.9530517339500835e-05,
      "loss": 0.9735,
      "num_input_tokens_seen": 38776096,
      "step": 66845
    },
    {
      "epoch": 9.956806672624367,
      "grad_norm": 15.625,
      "learning_rate": 2.952732168223084e-05,
      "loss": 1.5862,
      "num_input_tokens_seen": 38778912,
      "step": 66850
    },
    {
      "epoch": 9.957551385165326,
      "grad_norm": 12.625,
      "learning_rate": 2.952412594847634e-05,
      "loss": 0.9943,
      "num_input_tokens_seen": 38781792,
      "step": 66855
    },
    {
      "epoch": 9.958296097706285,
      "grad_norm": 5.625,
      "learning_rate": 2.952093013829133e-05,
      "loss": 1.0231,
      "num_input_tokens_seen": 38784576,
      "step": 66860
    },
    {
      "epoch": 9.959040810247245,
      "grad_norm": 9.25,
      "learning_rate": 2.95177342517298e-05,
      "loss": 0.9204,
      "num_input_tokens_seen": 38787648,
      "step": 66865
    },
    {
      "epoch": 9.959785522788204,
      "grad_norm": 6.65625,
      "learning_rate": 2.951453828884574e-05,
      "loss": 1.0863,
      "num_input_tokens_seen": 38790272,
      "step": 66870
    },
    {
      "epoch": 9.960530235329163,
      "grad_norm": 7.8125,
      "learning_rate": 2.951134224969314e-05,
      "loss": 1.1851,
      "num_input_tokens_seen": 38793216,
      "step": 66875
    },
    {
      "epoch": 9.961274947870121,
      "grad_norm": 13.1875,
      "learning_rate": 2.9508146134326004e-05,
      "loss": 1.2584,
      "num_input_tokens_seen": 38796096,
      "step": 66880
    },
    {
      "epoch": 9.962019660411082,
      "grad_norm": 12.0,
      "learning_rate": 2.950494994279832e-05,
      "loss": 1.3601,
      "num_input_tokens_seen": 38798880,
      "step": 66885
    },
    {
      "epoch": 9.96276437295204,
      "grad_norm": 7.46875,
      "learning_rate": 2.950175367516409e-05,
      "loss": 0.7582,
      "num_input_tokens_seen": 38801632,
      "step": 66890
    },
    {
      "epoch": 9.963509085493,
      "grad_norm": 12.0625,
      "learning_rate": 2.94985573314773e-05,
      "loss": 0.8807,
      "num_input_tokens_seen": 38804512,
      "step": 66895
    },
    {
      "epoch": 9.964253798033958,
      "grad_norm": 17.625,
      "learning_rate": 2.949536091179196e-05,
      "loss": 0.9829,
      "num_input_tokens_seen": 38807200,
      "step": 66900
    },
    {
      "epoch": 9.964998510574919,
      "grad_norm": 11.0,
      "learning_rate": 2.9492164416162066e-05,
      "loss": 1.0657,
      "num_input_tokens_seen": 38810304,
      "step": 66905
    },
    {
      "epoch": 9.965743223115878,
      "grad_norm": 7.375,
      "learning_rate": 2.9488967844641612e-05,
      "loss": 1.462,
      "num_input_tokens_seen": 38813248,
      "step": 66910
    },
    {
      "epoch": 9.966487935656836,
      "grad_norm": 18.0,
      "learning_rate": 2.9485771197284625e-05,
      "loss": 1.159,
      "num_input_tokens_seen": 38816288,
      "step": 66915
    },
    {
      "epoch": 9.967232648197795,
      "grad_norm": 5.65625,
      "learning_rate": 2.948257447414508e-05,
      "loss": 1.0429,
      "num_input_tokens_seen": 38819040,
      "step": 66920
    },
    {
      "epoch": 9.967977360738756,
      "grad_norm": 12.5,
      "learning_rate": 2.9479377675276998e-05,
      "loss": 1.4204,
      "num_input_tokens_seen": 38822016,
      "step": 66925
    },
    {
      "epoch": 9.968722073279714,
      "grad_norm": 7.09375,
      "learning_rate": 2.9476180800734376e-05,
      "loss": 1.0324,
      "num_input_tokens_seen": 38825056,
      "step": 66930
    },
    {
      "epoch": 9.969466785820673,
      "grad_norm": 13.3125,
      "learning_rate": 2.9472983850571235e-05,
      "loss": 1.1163,
      "num_input_tokens_seen": 38828032,
      "step": 66935
    },
    {
      "epoch": 9.970211498361632,
      "grad_norm": 20.375,
      "learning_rate": 2.9469786824841584e-05,
      "loss": 1.3307,
      "num_input_tokens_seen": 38831136,
      "step": 66940
    },
    {
      "epoch": 9.970956210902592,
      "grad_norm": 5.53125,
      "learning_rate": 2.946658972359942e-05,
      "loss": 0.8247,
      "num_input_tokens_seen": 38833728,
      "step": 66945
    },
    {
      "epoch": 9.971700923443551,
      "grad_norm": 12.6875,
      "learning_rate": 2.946339254689877e-05,
      "loss": 1.0673,
      "num_input_tokens_seen": 38836960,
      "step": 66950
    },
    {
      "epoch": 9.97244563598451,
      "grad_norm": 7.09375,
      "learning_rate": 2.946019529479363e-05,
      "loss": 0.9955,
      "num_input_tokens_seen": 38840192,
      "step": 66955
    },
    {
      "epoch": 9.973190348525469,
      "grad_norm": 7.15625,
      "learning_rate": 2.945699796733803e-05,
      "loss": 1.0462,
      "num_input_tokens_seen": 38843296,
      "step": 66960
    },
    {
      "epoch": 9.973935061066427,
      "grad_norm": 19.5,
      "learning_rate": 2.945380056458597e-05,
      "loss": 1.4858,
      "num_input_tokens_seen": 38845952,
      "step": 66965
    },
    {
      "epoch": 9.974679773607388,
      "grad_norm": 8.25,
      "learning_rate": 2.9450603086591484e-05,
      "loss": 1.1459,
      "num_input_tokens_seen": 38848896,
      "step": 66970
    },
    {
      "epoch": 9.975424486148347,
      "grad_norm": 8.6875,
      "learning_rate": 2.944740553340858e-05,
      "loss": 0.8556,
      "num_input_tokens_seen": 38851776,
      "step": 66975
    },
    {
      "epoch": 9.976169198689306,
      "grad_norm": 19.75,
      "learning_rate": 2.944420790509128e-05,
      "loss": 1.1826,
      "num_input_tokens_seen": 38854560,
      "step": 66980
    },
    {
      "epoch": 9.976913911230266,
      "grad_norm": 8.75,
      "learning_rate": 2.9441010201693614e-05,
      "loss": 0.943,
      "num_input_tokens_seen": 38857312,
      "step": 66985
    },
    {
      "epoch": 9.977658623771225,
      "grad_norm": 11.3125,
      "learning_rate": 2.9437812423269585e-05,
      "loss": 1.0061,
      "num_input_tokens_seen": 38860064,
      "step": 66990
    },
    {
      "epoch": 9.978403336312184,
      "grad_norm": 10.6875,
      "learning_rate": 2.943461456987322e-05,
      "loss": 0.8824,
      "num_input_tokens_seen": 38862848,
      "step": 66995
    },
    {
      "epoch": 9.979148048853142,
      "grad_norm": 9.3125,
      "learning_rate": 2.9431416641558558e-05,
      "loss": 1.2757,
      "num_input_tokens_seen": 38865824,
      "step": 67000
    },
    {
      "epoch": 9.979892761394101,
      "grad_norm": 11.6875,
      "learning_rate": 2.9428218638379608e-05,
      "loss": 1.1537,
      "num_input_tokens_seen": 38868832,
      "step": 67005
    },
    {
      "epoch": 9.980637473935062,
      "grad_norm": 11.1875,
      "learning_rate": 2.942502056039041e-05,
      "loss": 1.0715,
      "num_input_tokens_seen": 38871488,
      "step": 67010
    },
    {
      "epoch": 9.98138218647602,
      "grad_norm": 14.5625,
      "learning_rate": 2.9421822407644987e-05,
      "loss": 1.0475,
      "num_input_tokens_seen": 38874624,
      "step": 67015
    },
    {
      "epoch": 9.98212689901698,
      "grad_norm": 8.5625,
      "learning_rate": 2.9418624180197375e-05,
      "loss": 1.0102,
      "num_input_tokens_seen": 38877536,
      "step": 67020
    },
    {
      "epoch": 9.982871611557938,
      "grad_norm": 7.9375,
      "learning_rate": 2.941542587810159e-05,
      "loss": 0.7855,
      "num_input_tokens_seen": 38880384,
      "step": 67025
    },
    {
      "epoch": 9.983616324098898,
      "grad_norm": 7.9375,
      "learning_rate": 2.941222750141167e-05,
      "loss": 1.5593,
      "num_input_tokens_seen": 38883168,
      "step": 67030
    },
    {
      "epoch": 9.984361036639857,
      "grad_norm": 5.90625,
      "learning_rate": 2.9409029050181652e-05,
      "loss": 1.1595,
      "num_input_tokens_seen": 38885920,
      "step": 67035
    },
    {
      "epoch": 9.985105749180816,
      "grad_norm": 13.125,
      "learning_rate": 2.9405830524465573e-05,
      "loss": 1.4396,
      "num_input_tokens_seen": 38888640,
      "step": 67040
    },
    {
      "epoch": 9.985850461721775,
      "grad_norm": 14.3125,
      "learning_rate": 2.940263192431746e-05,
      "loss": 1.2667,
      "num_input_tokens_seen": 38891424,
      "step": 67045
    },
    {
      "epoch": 9.986595174262735,
      "grad_norm": 12.0,
      "learning_rate": 2.9399433249791363e-05,
      "loss": 0.9843,
      "num_input_tokens_seen": 38894336,
      "step": 67050
    },
    {
      "epoch": 9.987339886803694,
      "grad_norm": 13.3125,
      "learning_rate": 2.9396234500941307e-05,
      "loss": 1.1474,
      "num_input_tokens_seen": 38897280,
      "step": 67055
    },
    {
      "epoch": 9.988084599344653,
      "grad_norm": 6.875,
      "learning_rate": 2.939303567782134e-05,
      "loss": 1.1407,
      "num_input_tokens_seen": 38900320,
      "step": 67060
    },
    {
      "epoch": 9.988829311885612,
      "grad_norm": 6.3125,
      "learning_rate": 2.9389836780485502e-05,
      "loss": 0.9582,
      "num_input_tokens_seen": 38903296,
      "step": 67065
    },
    {
      "epoch": 9.989574024426572,
      "grad_norm": 8.5,
      "learning_rate": 2.9386637808987828e-05,
      "loss": 0.902,
      "num_input_tokens_seen": 38906368,
      "step": 67070
    },
    {
      "epoch": 9.99031873696753,
      "grad_norm": 8.4375,
      "learning_rate": 2.9383438763382363e-05,
      "loss": 0.7999,
      "num_input_tokens_seen": 38909536,
      "step": 67075
    },
    {
      "epoch": 9.99106344950849,
      "grad_norm": 12.5,
      "learning_rate": 2.9380239643723167e-05,
      "loss": 1.32,
      "num_input_tokens_seen": 38912512,
      "step": 67080
    },
    {
      "epoch": 9.991808162049448,
      "grad_norm": 12.6875,
      "learning_rate": 2.9377040450064268e-05,
      "loss": 1.6433,
      "num_input_tokens_seen": 38915520,
      "step": 67085
    },
    {
      "epoch": 9.992552874590409,
      "grad_norm": 9.75,
      "learning_rate": 2.9373841182459715e-05,
      "loss": 1.1978,
      "num_input_tokens_seen": 38918400,
      "step": 67090
    },
    {
      "epoch": 9.993297587131368,
      "grad_norm": 7.625,
      "learning_rate": 2.9370641840963565e-05,
      "loss": 0.8971,
      "num_input_tokens_seen": 38921472,
      "step": 67095
    },
    {
      "epoch": 9.994042299672326,
      "grad_norm": 5.40625,
      "learning_rate": 2.9367442425629866e-05,
      "loss": 1.0148,
      "num_input_tokens_seen": 38924352,
      "step": 67100
    },
    {
      "epoch": 9.994787012213285,
      "grad_norm": 16.25,
      "learning_rate": 2.9364242936512665e-05,
      "loss": 1.1505,
      "num_input_tokens_seen": 38926976,
      "step": 67105
    },
    {
      "epoch": 9.995531724754246,
      "grad_norm": 12.5,
      "learning_rate": 2.936104337366601e-05,
      "loss": 1.1948,
      "num_input_tokens_seen": 38929952,
      "step": 67110
    },
    {
      "epoch": 9.996276437295204,
      "grad_norm": 10.6875,
      "learning_rate": 2.935784373714397e-05,
      "loss": 1.0314,
      "num_input_tokens_seen": 38932544,
      "step": 67115
    },
    {
      "epoch": 9.997021149836163,
      "grad_norm": 9.5625,
      "learning_rate": 2.9354644027000577e-05,
      "loss": 1.0533,
      "num_input_tokens_seen": 38935840,
      "step": 67120
    },
    {
      "epoch": 9.997765862377122,
      "grad_norm": 14.0625,
      "learning_rate": 2.9351444243289904e-05,
      "loss": 1.1572,
      "num_input_tokens_seen": 38938944,
      "step": 67125
    },
    {
      "epoch": 9.998510574918082,
      "grad_norm": 9.5,
      "learning_rate": 2.9348244386066005e-05,
      "loss": 1.0285,
      "num_input_tokens_seen": 38941696,
      "step": 67130
    },
    {
      "epoch": 9.999255287459041,
      "grad_norm": 5.125,
      "learning_rate": 2.9345044455382932e-05,
      "loss": 1.2336,
      "num_input_tokens_seen": 38944672,
      "step": 67135
    },
    {
      "epoch": 10.0,
      "grad_norm": 22.75,
      "learning_rate": 2.9341844451294754e-05,
      "loss": 1.4493,
      "num_input_tokens_seen": 38947216,
      "step": 67140
    },
    {
      "epoch": 10.0,
      "eval_loss": 1.0846481323242188,
      "eval_runtime": 43.7689,
      "eval_samples_per_second": 68.176,
      "eval_steps_per_second": 17.044,
      "num_input_tokens_seen": 38947216,
      "step": 67140
    },
    {
      "epoch": 10.000744712540959,
      "grad_norm": 7.84375,
      "learning_rate": 2.9338644373855522e-05,
      "loss": 0.9463,
      "num_input_tokens_seen": 38950352,
      "step": 67145
    },
    {
      "epoch": 10.001489425081918,
      "grad_norm": 19.5,
      "learning_rate": 2.9335444223119314e-05,
      "loss": 1.2447,
      "num_input_tokens_seen": 38953328,
      "step": 67150
    },
    {
      "epoch": 10.002234137622878,
      "grad_norm": 8.25,
      "learning_rate": 2.9332243999140167e-05,
      "loss": 0.927,
      "num_input_tokens_seen": 38956208,
      "step": 67155
    },
    {
      "epoch": 10.002978850163837,
      "grad_norm": 20.0,
      "learning_rate": 2.932904370197217e-05,
      "loss": 0.9174,
      "num_input_tokens_seen": 38959120,
      "step": 67160
    },
    {
      "epoch": 10.003723562704796,
      "grad_norm": 10.125,
      "learning_rate": 2.932584333166937e-05,
      "loss": 0.8033,
      "num_input_tokens_seen": 38961872,
      "step": 67165
    },
    {
      "epoch": 10.004468275245754,
      "grad_norm": 16.375,
      "learning_rate": 2.9322642888285855e-05,
      "loss": 1.1411,
      "num_input_tokens_seen": 38964816,
      "step": 67170
    },
    {
      "epoch": 10.005212987786715,
      "grad_norm": 15.5625,
      "learning_rate": 2.931944237187567e-05,
      "loss": 1.0942,
      "num_input_tokens_seen": 38967632,
      "step": 67175
    },
    {
      "epoch": 10.005957700327674,
      "grad_norm": 5.125,
      "learning_rate": 2.931624178249291e-05,
      "loss": 1.0812,
      "num_input_tokens_seen": 38970640,
      "step": 67180
    },
    {
      "epoch": 10.006702412868632,
      "grad_norm": 12.375,
      "learning_rate": 2.931304112019163e-05,
      "loss": 1.3707,
      "num_input_tokens_seen": 38973584,
      "step": 67185
    },
    {
      "epoch": 10.007447125409591,
      "grad_norm": 12.0,
      "learning_rate": 2.93098403850259e-05,
      "loss": 1.2114,
      "num_input_tokens_seen": 38976400,
      "step": 67190
    },
    {
      "epoch": 10.008191837950552,
      "grad_norm": 10.0625,
      "learning_rate": 2.9306639577049793e-05,
      "loss": 1.0762,
      "num_input_tokens_seen": 38979312,
      "step": 67195
    },
    {
      "epoch": 10.00893655049151,
      "grad_norm": 9.875,
      "learning_rate": 2.9303438696317385e-05,
      "loss": 1.1859,
      "num_input_tokens_seen": 38982160,
      "step": 67200
    },
    {
      "epoch": 10.00968126303247,
      "grad_norm": 22.75,
      "learning_rate": 2.9300237742882764e-05,
      "loss": 1.0933,
      "num_input_tokens_seen": 38985040,
      "step": 67205
    },
    {
      "epoch": 10.010425975573428,
      "grad_norm": 7.96875,
      "learning_rate": 2.929703671679999e-05,
      "loss": 1.1093,
      "num_input_tokens_seen": 38987664,
      "step": 67210
    },
    {
      "epoch": 10.011170688114388,
      "grad_norm": 10.125,
      "learning_rate": 2.9293835618123157e-05,
      "loss": 1.219,
      "num_input_tokens_seen": 38990544,
      "step": 67215
    },
    {
      "epoch": 10.011915400655347,
      "grad_norm": 8.4375,
      "learning_rate": 2.929063444690633e-05,
      "loss": 0.9545,
      "num_input_tokens_seen": 38993520,
      "step": 67220
    },
    {
      "epoch": 10.012660113196306,
      "grad_norm": 5.53125,
      "learning_rate": 2.9287433203203598e-05,
      "loss": 0.9523,
      "num_input_tokens_seen": 38996432,
      "step": 67225
    },
    {
      "epoch": 10.013404825737265,
      "grad_norm": 9.75,
      "learning_rate": 2.928423188706903e-05,
      "loss": 0.9275,
      "num_input_tokens_seen": 38999248,
      "step": 67230
    },
    {
      "epoch": 10.014149538278225,
      "grad_norm": 15.0,
      "learning_rate": 2.9281030498556723e-05,
      "loss": 0.9387,
      "num_input_tokens_seen": 39002128,
      "step": 67235
    },
    {
      "epoch": 10.014894250819184,
      "grad_norm": 9.375,
      "learning_rate": 2.9277829037720754e-05,
      "loss": 1.2256,
      "num_input_tokens_seen": 39005072,
      "step": 67240
    },
    {
      "epoch": 10.015638963360143,
      "grad_norm": 12.25,
      "learning_rate": 2.927462750461522e-05,
      "loss": 0.7853,
      "num_input_tokens_seen": 39008112,
      "step": 67245
    },
    {
      "epoch": 10.016383675901102,
      "grad_norm": 21.625,
      "learning_rate": 2.9271425899294193e-05,
      "loss": 1.1905,
      "num_input_tokens_seen": 39010832,
      "step": 67250
    },
    {
      "epoch": 10.017128388442062,
      "grad_norm": 9.3125,
      "learning_rate": 2.9268224221811763e-05,
      "loss": 1.2826,
      "num_input_tokens_seen": 39013424,
      "step": 67255
    },
    {
      "epoch": 10.01787310098302,
      "grad_norm": 17.875,
      "learning_rate": 2.9265022472222032e-05,
      "loss": 1.409,
      "num_input_tokens_seen": 39016304,
      "step": 67260
    },
    {
      "epoch": 10.01861781352398,
      "grad_norm": 9.25,
      "learning_rate": 2.9261820650579074e-05,
      "loss": 1.0668,
      "num_input_tokens_seen": 39019280,
      "step": 67265
    },
    {
      "epoch": 10.019362526064938,
      "grad_norm": 11.4375,
      "learning_rate": 2.925861875693699e-05,
      "loss": 0.7694,
      "num_input_tokens_seen": 39022160,
      "step": 67270
    },
    {
      "epoch": 10.020107238605899,
      "grad_norm": 9.6875,
      "learning_rate": 2.9255416791349867e-05,
      "loss": 1.0104,
      "num_input_tokens_seen": 39024752,
      "step": 67275
    },
    {
      "epoch": 10.020851951146858,
      "grad_norm": 14.625,
      "learning_rate": 2.925221475387181e-05,
      "loss": 0.9512,
      "num_input_tokens_seen": 39027760,
      "step": 67280
    },
    {
      "epoch": 10.021596663687816,
      "grad_norm": 6.34375,
      "learning_rate": 2.92490126445569e-05,
      "loss": 1.2795,
      "num_input_tokens_seen": 39030768,
      "step": 67285
    },
    {
      "epoch": 10.022341376228775,
      "grad_norm": 11.5,
      "learning_rate": 2.9245810463459245e-05,
      "loss": 0.958,
      "num_input_tokens_seen": 39033488,
      "step": 67290
    },
    {
      "epoch": 10.023086088769736,
      "grad_norm": 16.125,
      "learning_rate": 2.9242608210632932e-05,
      "loss": 1.1524,
      "num_input_tokens_seen": 39036656,
      "step": 67295
    },
    {
      "epoch": 10.023830801310694,
      "grad_norm": 10.5625,
      "learning_rate": 2.9239405886132066e-05,
      "loss": 0.8081,
      "num_input_tokens_seen": 39039568,
      "step": 67300
    },
    {
      "epoch": 10.024575513851653,
      "grad_norm": 7.5625,
      "learning_rate": 2.923620349001075e-05,
      "loss": 0.9745,
      "num_input_tokens_seen": 39042640,
      "step": 67305
    },
    {
      "epoch": 10.025320226392612,
      "grad_norm": 12.5625,
      "learning_rate": 2.923300102232308e-05,
      "loss": 1.242,
      "num_input_tokens_seen": 39045744,
      "step": 67310
    },
    {
      "epoch": 10.02606493893357,
      "grad_norm": 10.625,
      "learning_rate": 2.9229798483123162e-05,
      "loss": 1.2077,
      "num_input_tokens_seen": 39048528,
      "step": 67315
    },
    {
      "epoch": 10.026809651474531,
      "grad_norm": 15.5625,
      "learning_rate": 2.9226595872465097e-05,
      "loss": 1.2353,
      "num_input_tokens_seen": 39051280,
      "step": 67320
    },
    {
      "epoch": 10.02755436401549,
      "grad_norm": 8.8125,
      "learning_rate": 2.922339319040298e-05,
      "loss": 0.9609,
      "num_input_tokens_seen": 39054224,
      "step": 67325
    },
    {
      "epoch": 10.028299076556449,
      "grad_norm": 7.65625,
      "learning_rate": 2.922019043699094e-05,
      "loss": 1.1212,
      "num_input_tokens_seen": 39057072,
      "step": 67330
    },
    {
      "epoch": 10.029043789097408,
      "grad_norm": 7.09375,
      "learning_rate": 2.9216987612283064e-05,
      "loss": 0.9133,
      "num_input_tokens_seen": 39060080,
      "step": 67335
    },
    {
      "epoch": 10.029788501638368,
      "grad_norm": 14.5,
      "learning_rate": 2.921378471633347e-05,
      "loss": 1.2621,
      "num_input_tokens_seen": 39063152,
      "step": 67340
    },
    {
      "epoch": 10.030533214179327,
      "grad_norm": 9.9375,
      "learning_rate": 2.9210581749196274e-05,
      "loss": 1.3546,
      "num_input_tokens_seen": 39066064,
      "step": 67345
    },
    {
      "epoch": 10.031277926720286,
      "grad_norm": 11.125,
      "learning_rate": 2.9207378710925575e-05,
      "loss": 1.1246,
      "num_input_tokens_seen": 39069008,
      "step": 67350
    },
    {
      "epoch": 10.032022639261244,
      "grad_norm": 9.0625,
      "learning_rate": 2.920417560157549e-05,
      "loss": 0.9862,
      "num_input_tokens_seen": 39071824,
      "step": 67355
    },
    {
      "epoch": 10.032767351802205,
      "grad_norm": 8.1875,
      "learning_rate": 2.9200972421200124e-05,
      "loss": 1.3179,
      "num_input_tokens_seen": 39074672,
      "step": 67360
    },
    {
      "epoch": 10.033512064343164,
      "grad_norm": 9.4375,
      "learning_rate": 2.91977691698536e-05,
      "loss": 0.9603,
      "num_input_tokens_seen": 39077552,
      "step": 67365
    },
    {
      "epoch": 10.034256776884122,
      "grad_norm": 11.6875,
      "learning_rate": 2.919456584759003e-05,
      "loss": 1.1655,
      "num_input_tokens_seen": 39080464,
      "step": 67370
    },
    {
      "epoch": 10.035001489425081,
      "grad_norm": 10.25,
      "learning_rate": 2.919136245446354e-05,
      "loss": 1.2143,
      "num_input_tokens_seen": 39083472,
      "step": 67375
    },
    {
      "epoch": 10.035746201966042,
      "grad_norm": 12.8125,
      "learning_rate": 2.918815899052824e-05,
      "loss": 1.2103,
      "num_input_tokens_seen": 39086224,
      "step": 67380
    },
    {
      "epoch": 10.036490914507,
      "grad_norm": 6.09375,
      "learning_rate": 2.9184955455838258e-05,
      "loss": 0.7187,
      "num_input_tokens_seen": 39089296,
      "step": 67385
    },
    {
      "epoch": 10.03723562704796,
      "grad_norm": 9.8125,
      "learning_rate": 2.9181751850447698e-05,
      "loss": 1.2262,
      "num_input_tokens_seen": 39092432,
      "step": 67390
    },
    {
      "epoch": 10.037980339588918,
      "grad_norm": 9.25,
      "learning_rate": 2.9178548174410687e-05,
      "loss": 0.835,
      "num_input_tokens_seen": 39095152,
      "step": 67395
    },
    {
      "epoch": 10.038725052129879,
      "grad_norm": 13.25,
      "learning_rate": 2.9175344427781354e-05,
      "loss": 1.0294,
      "num_input_tokens_seen": 39098032,
      "step": 67400
    },
    {
      "epoch": 10.039469764670837,
      "grad_norm": 6.71875,
      "learning_rate": 2.9172140610613825e-05,
      "loss": 1.0096,
      "num_input_tokens_seen": 39101104,
      "step": 67405
    },
    {
      "epoch": 10.040214477211796,
      "grad_norm": 12.6875,
      "learning_rate": 2.916893672296222e-05,
      "loss": 1.123,
      "num_input_tokens_seen": 39104016,
      "step": 67410
    },
    {
      "epoch": 10.040959189752755,
      "grad_norm": 5.59375,
      "learning_rate": 2.916573276488066e-05,
      "loss": 1.2231,
      "num_input_tokens_seen": 39106864,
      "step": 67415
    },
    {
      "epoch": 10.041703902293715,
      "grad_norm": 9.4375,
      "learning_rate": 2.9162528736423283e-05,
      "loss": 0.9293,
      "num_input_tokens_seen": 39109776,
      "step": 67420
    },
    {
      "epoch": 10.042448614834674,
      "grad_norm": 14.3125,
      "learning_rate": 2.915932463764422e-05,
      "loss": 1.15,
      "num_input_tokens_seen": 39112912,
      "step": 67425
    },
    {
      "epoch": 10.043193327375633,
      "grad_norm": 8.25,
      "learning_rate": 2.9156120468597588e-05,
      "loss": 1.2795,
      "num_input_tokens_seen": 39115792,
      "step": 67430
    },
    {
      "epoch": 10.043938039916592,
      "grad_norm": 10.1875,
      "learning_rate": 2.9152916229337525e-05,
      "loss": 1.1801,
      "num_input_tokens_seen": 39118608,
      "step": 67435
    },
    {
      "epoch": 10.044682752457552,
      "grad_norm": 11.125,
      "learning_rate": 2.9149711919918154e-05,
      "loss": 0.8483,
      "num_input_tokens_seen": 39121488,
      "step": 67440
    },
    {
      "epoch": 10.045427464998511,
      "grad_norm": 16.625,
      "learning_rate": 2.9146507540393636e-05,
      "loss": 1.0343,
      "num_input_tokens_seen": 39124560,
      "step": 67445
    },
    {
      "epoch": 10.04617217753947,
      "grad_norm": 15.125,
      "learning_rate": 2.9143303090818074e-05,
      "loss": 0.8588,
      "num_input_tokens_seen": 39127504,
      "step": 67450
    },
    {
      "epoch": 10.046916890080428,
      "grad_norm": 3.84375,
      "learning_rate": 2.9140098571245623e-05,
      "loss": 0.7565,
      "num_input_tokens_seen": 39130320,
      "step": 67455
    },
    {
      "epoch": 10.047661602621389,
      "grad_norm": 8.5,
      "learning_rate": 2.9136893981730406e-05,
      "loss": 1.0748,
      "num_input_tokens_seen": 39133072,
      "step": 67460
    },
    {
      "epoch": 10.048406315162348,
      "grad_norm": 6.75,
      "learning_rate": 2.9133689322326586e-05,
      "loss": 1.1635,
      "num_input_tokens_seen": 39136016,
      "step": 67465
    },
    {
      "epoch": 10.049151027703306,
      "grad_norm": 16.75,
      "learning_rate": 2.9130484593088276e-05,
      "loss": 1.1548,
      "num_input_tokens_seen": 39138992,
      "step": 67470
    },
    {
      "epoch": 10.049895740244265,
      "grad_norm": 9.3125,
      "learning_rate": 2.9127279794069624e-05,
      "loss": 0.8466,
      "num_input_tokens_seen": 39141840,
      "step": 67475
    },
    {
      "epoch": 10.050640452785226,
      "grad_norm": 13.75,
      "learning_rate": 2.9124074925324785e-05,
      "loss": 1.0434,
      "num_input_tokens_seen": 39144624,
      "step": 67480
    },
    {
      "epoch": 10.051385165326185,
      "grad_norm": 6.09375,
      "learning_rate": 2.9120869986907885e-05,
      "loss": 0.8615,
      "num_input_tokens_seen": 39147440,
      "step": 67485
    },
    {
      "epoch": 10.052129877867143,
      "grad_norm": 5.96875,
      "learning_rate": 2.9117664978873072e-05,
      "loss": 1.0026,
      "num_input_tokens_seen": 39150192,
      "step": 67490
    },
    {
      "epoch": 10.052874590408102,
      "grad_norm": 13.5625,
      "learning_rate": 2.9114459901274493e-05,
      "loss": 1.1669,
      "num_input_tokens_seen": 39153136,
      "step": 67495
    },
    {
      "epoch": 10.05361930294906,
      "grad_norm": 8.9375,
      "learning_rate": 2.91112547541663e-05,
      "loss": 1.1082,
      "num_input_tokens_seen": 39156176,
      "step": 67500
    },
    {
      "epoch": 10.054364015490021,
      "grad_norm": 19.75,
      "learning_rate": 2.9108049537602637e-05,
      "loss": 1.0323,
      "num_input_tokens_seen": 39159504,
      "step": 67505
    },
    {
      "epoch": 10.05510872803098,
      "grad_norm": 16.5,
      "learning_rate": 2.9104844251637652e-05,
      "loss": 0.8959,
      "num_input_tokens_seen": 39162480,
      "step": 67510
    },
    {
      "epoch": 10.055853440571939,
      "grad_norm": 8.625,
      "learning_rate": 2.91016388963255e-05,
      "loss": 1.207,
      "num_input_tokens_seen": 39165232,
      "step": 67515
    },
    {
      "epoch": 10.056598153112898,
      "grad_norm": 12.0625,
      "learning_rate": 2.9098433471720322e-05,
      "loss": 1.1397,
      "num_input_tokens_seen": 39168176,
      "step": 67520
    },
    {
      "epoch": 10.057342865653858,
      "grad_norm": 10.75,
      "learning_rate": 2.909522797787627e-05,
      "loss": 1.0772,
      "num_input_tokens_seen": 39171024,
      "step": 67525
    },
    {
      "epoch": 10.058087578194817,
      "grad_norm": 12.0625,
      "learning_rate": 2.9092022414847514e-05,
      "loss": 1.2891,
      "num_input_tokens_seen": 39174192,
      "step": 67530
    },
    {
      "epoch": 10.058832290735776,
      "grad_norm": 21.5,
      "learning_rate": 2.908881678268819e-05,
      "loss": 1.1327,
      "num_input_tokens_seen": 39177328,
      "step": 67535
    },
    {
      "epoch": 10.059577003276734,
      "grad_norm": 22.875,
      "learning_rate": 2.908561108145247e-05,
      "loss": 0.9081,
      "num_input_tokens_seen": 39180016,
      "step": 67540
    },
    {
      "epoch": 10.060321715817695,
      "grad_norm": 8.5,
      "learning_rate": 2.90824053111945e-05,
      "loss": 1.1368,
      "num_input_tokens_seen": 39183056,
      "step": 67545
    },
    {
      "epoch": 10.061066428358654,
      "grad_norm": 4.03125,
      "learning_rate": 2.9079199471968444e-05,
      "loss": 0.8506,
      "num_input_tokens_seen": 39185968,
      "step": 67550
    },
    {
      "epoch": 10.061811140899612,
      "grad_norm": 7.5,
      "learning_rate": 2.9075993563828452e-05,
      "loss": 1.007,
      "num_input_tokens_seen": 39189168,
      "step": 67555
    },
    {
      "epoch": 10.062555853440571,
      "grad_norm": 15.25,
      "learning_rate": 2.9072787586828697e-05,
      "loss": 1.3409,
      "num_input_tokens_seen": 39191984,
      "step": 67560
    },
    {
      "epoch": 10.063300565981532,
      "grad_norm": 7.78125,
      "learning_rate": 2.9069581541023333e-05,
      "loss": 0.6927,
      "num_input_tokens_seen": 39194928,
      "step": 67565
    },
    {
      "epoch": 10.06404527852249,
      "grad_norm": 14.1875,
      "learning_rate": 2.9066375426466518e-05,
      "loss": 0.7578,
      "num_input_tokens_seen": 39197808,
      "step": 67570
    },
    {
      "epoch": 10.06478999106345,
      "grad_norm": 12.9375,
      "learning_rate": 2.906316924321244e-05,
      "loss": 1.0829,
      "num_input_tokens_seen": 39200752,
      "step": 67575
    },
    {
      "epoch": 10.065534703604408,
      "grad_norm": 4.0625,
      "learning_rate": 2.9059962991315237e-05,
      "loss": 0.795,
      "num_input_tokens_seen": 39203504,
      "step": 67580
    },
    {
      "epoch": 10.066279416145369,
      "grad_norm": 14.4375,
      "learning_rate": 2.9056756670829087e-05,
      "loss": 1.1517,
      "num_input_tokens_seen": 39206192,
      "step": 67585
    },
    {
      "epoch": 10.067024128686327,
      "grad_norm": 11.5625,
      "learning_rate": 2.9053550281808155e-05,
      "loss": 1.2731,
      "num_input_tokens_seen": 39209008,
      "step": 67590
    },
    {
      "epoch": 10.067768841227286,
      "grad_norm": 7.84375,
      "learning_rate": 2.905034382430661e-05,
      "loss": 1.174,
      "num_input_tokens_seen": 39212208,
      "step": 67595
    },
    {
      "epoch": 10.068513553768245,
      "grad_norm": 6.90625,
      "learning_rate": 2.9047137298378624e-05,
      "loss": 0.9791,
      "num_input_tokens_seen": 39214992,
      "step": 67600
    },
    {
      "epoch": 10.069258266309205,
      "grad_norm": 9.125,
      "learning_rate": 2.9043930704078364e-05,
      "loss": 1.1303,
      "num_input_tokens_seen": 39217808,
      "step": 67605
    },
    {
      "epoch": 10.070002978850164,
      "grad_norm": 15.6875,
      "learning_rate": 2.904072404146001e-05,
      "loss": 1.1823,
      "num_input_tokens_seen": 39220560,
      "step": 67610
    },
    {
      "epoch": 10.070747691391123,
      "grad_norm": 13.5625,
      "learning_rate": 2.9037517310577726e-05,
      "loss": 1.3033,
      "num_input_tokens_seen": 39223664,
      "step": 67615
    },
    {
      "epoch": 10.071492403932082,
      "grad_norm": 13.8125,
      "learning_rate": 2.9034310511485692e-05,
      "loss": 1.0141,
      "num_input_tokens_seen": 39226736,
      "step": 67620
    },
    {
      "epoch": 10.072237116473042,
      "grad_norm": 8.6875,
      "learning_rate": 2.903110364423809e-05,
      "loss": 1.1118,
      "num_input_tokens_seen": 39229776,
      "step": 67625
    },
    {
      "epoch": 10.072981829014001,
      "grad_norm": 8.5,
      "learning_rate": 2.9027896708889073e-05,
      "loss": 0.9072,
      "num_input_tokens_seen": 39232656,
      "step": 67630
    },
    {
      "epoch": 10.07372654155496,
      "grad_norm": 13.4375,
      "learning_rate": 2.9024689705492847e-05,
      "loss": 0.9251,
      "num_input_tokens_seen": 39235792,
      "step": 67635
    },
    {
      "epoch": 10.074471254095918,
      "grad_norm": 14.5,
      "learning_rate": 2.902148263410357e-05,
      "loss": 1.3417,
      "num_input_tokens_seen": 39238800,
      "step": 67640
    },
    {
      "epoch": 10.075215966636879,
      "grad_norm": 3.4375,
      "learning_rate": 2.9018275494775442e-05,
      "loss": 0.8777,
      "num_input_tokens_seen": 39241840,
      "step": 67645
    },
    {
      "epoch": 10.075960679177838,
      "grad_norm": 10.75,
      "learning_rate": 2.9015068287562626e-05,
      "loss": 1.2865,
      "num_input_tokens_seen": 39244752,
      "step": 67650
    },
    {
      "epoch": 10.076705391718797,
      "grad_norm": 6.6875,
      "learning_rate": 2.9011861012519316e-05,
      "loss": 0.9222,
      "num_input_tokens_seen": 39247568,
      "step": 67655
    },
    {
      "epoch": 10.077450104259755,
      "grad_norm": 10.8125,
      "learning_rate": 2.900865366969968e-05,
      "loss": 0.9609,
      "num_input_tokens_seen": 39250576,
      "step": 67660
    },
    {
      "epoch": 10.078194816800714,
      "grad_norm": 7.8125,
      "learning_rate": 2.900544625915793e-05,
      "loss": 0.7272,
      "num_input_tokens_seen": 39253232,
      "step": 67665
    },
    {
      "epoch": 10.078939529341675,
      "grad_norm": 9.3125,
      "learning_rate": 2.9002238780948232e-05,
      "loss": 1.0842,
      "num_input_tokens_seen": 39256112,
      "step": 67670
    },
    {
      "epoch": 10.079684241882633,
      "grad_norm": 2.359375,
      "learning_rate": 2.8999031235124775e-05,
      "loss": 0.981,
      "num_input_tokens_seen": 39259152,
      "step": 67675
    },
    {
      "epoch": 10.080428954423592,
      "grad_norm": 10.1875,
      "learning_rate": 2.8995823621741754e-05,
      "loss": 1.2459,
      "num_input_tokens_seen": 39261872,
      "step": 67680
    },
    {
      "epoch": 10.08117366696455,
      "grad_norm": 12.9375,
      "learning_rate": 2.8992615940853347e-05,
      "loss": 1.1621,
      "num_input_tokens_seen": 39264752,
      "step": 67685
    },
    {
      "epoch": 10.081918379505511,
      "grad_norm": 9.6875,
      "learning_rate": 2.8989408192513756e-05,
      "loss": 0.9047,
      "num_input_tokens_seen": 39267504,
      "step": 67690
    },
    {
      "epoch": 10.08266309204647,
      "grad_norm": 6.3125,
      "learning_rate": 2.898620037677717e-05,
      "loss": 0.9459,
      "num_input_tokens_seen": 39270608,
      "step": 67695
    },
    {
      "epoch": 10.083407804587429,
      "grad_norm": 12.75,
      "learning_rate": 2.898299249369777e-05,
      "loss": 1.0012,
      "num_input_tokens_seen": 39273712,
      "step": 67700
    },
    {
      "epoch": 10.084152517128388,
      "grad_norm": 14.25,
      "learning_rate": 2.8979784543329775e-05,
      "loss": 1.3336,
      "num_input_tokens_seen": 39276784,
      "step": 67705
    },
    {
      "epoch": 10.084897229669348,
      "grad_norm": 12.0,
      "learning_rate": 2.897657652572735e-05,
      "loss": 1.0649,
      "num_input_tokens_seen": 39279792,
      "step": 67710
    },
    {
      "epoch": 10.085641942210307,
      "grad_norm": 14.0625,
      "learning_rate": 2.897336844094472e-05,
      "loss": 0.8636,
      "num_input_tokens_seen": 39282576,
      "step": 67715
    },
    {
      "epoch": 10.086386654751266,
      "grad_norm": 8.375,
      "learning_rate": 2.8970160289036064e-05,
      "loss": 1.4043,
      "num_input_tokens_seen": 39285712,
      "step": 67720
    },
    {
      "epoch": 10.087131367292224,
      "grad_norm": 8.125,
      "learning_rate": 2.8966952070055582e-05,
      "loss": 1.0152,
      "num_input_tokens_seen": 39288592,
      "step": 67725
    },
    {
      "epoch": 10.087876079833185,
      "grad_norm": 5.4375,
      "learning_rate": 2.8963743784057474e-05,
      "loss": 1.0302,
      "num_input_tokens_seen": 39291440,
      "step": 67730
    },
    {
      "epoch": 10.088620792374144,
      "grad_norm": 17.25,
      "learning_rate": 2.896053543109595e-05,
      "loss": 1.2188,
      "num_input_tokens_seen": 39294352,
      "step": 67735
    },
    {
      "epoch": 10.089365504915103,
      "grad_norm": 11.0625,
      "learning_rate": 2.8957327011225198e-05,
      "loss": 0.9847,
      "num_input_tokens_seen": 39297520,
      "step": 67740
    },
    {
      "epoch": 10.090110217456061,
      "grad_norm": 7.9375,
      "learning_rate": 2.8954118524499434e-05,
      "loss": 1.0022,
      "num_input_tokens_seen": 39300592,
      "step": 67745
    },
    {
      "epoch": 10.090854929997022,
      "grad_norm": 12.5,
      "learning_rate": 2.895090997097286e-05,
      "loss": 1.2686,
      "num_input_tokens_seen": 39303696,
      "step": 67750
    },
    {
      "epoch": 10.09159964253798,
      "grad_norm": 22.875,
      "learning_rate": 2.894770135069967e-05,
      "loss": 1.3174,
      "num_input_tokens_seen": 39306512,
      "step": 67755
    },
    {
      "epoch": 10.09234435507894,
      "grad_norm": 8.1875,
      "learning_rate": 2.894449266373408e-05,
      "loss": 1.1248,
      "num_input_tokens_seen": 39309136,
      "step": 67760
    },
    {
      "epoch": 10.093089067619898,
      "grad_norm": 9.3125,
      "learning_rate": 2.8941283910130295e-05,
      "loss": 1.3283,
      "num_input_tokens_seen": 39311920,
      "step": 67765
    },
    {
      "epoch": 10.093833780160859,
      "grad_norm": 9.25,
      "learning_rate": 2.8938075089942524e-05,
      "loss": 1.159,
      "num_input_tokens_seen": 39314544,
      "step": 67770
    },
    {
      "epoch": 10.094578492701817,
      "grad_norm": 14.5,
      "learning_rate": 2.893486620322498e-05,
      "loss": 1.0831,
      "num_input_tokens_seen": 39317456,
      "step": 67775
    },
    {
      "epoch": 10.095323205242776,
      "grad_norm": 12.125,
      "learning_rate": 2.893165725003187e-05,
      "loss": 0.9333,
      "num_input_tokens_seen": 39320368,
      "step": 67780
    },
    {
      "epoch": 10.096067917783735,
      "grad_norm": 7.96875,
      "learning_rate": 2.8928448230417404e-05,
      "loss": 0.9988,
      "num_input_tokens_seen": 39322992,
      "step": 67785
    },
    {
      "epoch": 10.096812630324695,
      "grad_norm": 9.625,
      "learning_rate": 2.89252391444358e-05,
      "loss": 1.0895,
      "num_input_tokens_seen": 39325712,
      "step": 67790
    },
    {
      "epoch": 10.097557342865654,
      "grad_norm": 9.5625,
      "learning_rate": 2.892202999214127e-05,
      "loss": 1.0971,
      "num_input_tokens_seen": 39328528,
      "step": 67795
    },
    {
      "epoch": 10.098302055406613,
      "grad_norm": 8.625,
      "learning_rate": 2.8918820773588025e-05,
      "loss": 0.9499,
      "num_input_tokens_seen": 39331312,
      "step": 67800
    },
    {
      "epoch": 10.099046767947572,
      "grad_norm": 11.8125,
      "learning_rate": 2.8915611488830284e-05,
      "loss": 0.9031,
      "num_input_tokens_seen": 39334128,
      "step": 67805
    },
    {
      "epoch": 10.099791480488532,
      "grad_norm": 7.9375,
      "learning_rate": 2.891240213792228e-05,
      "loss": 1.0341,
      "num_input_tokens_seen": 39337008,
      "step": 67810
    },
    {
      "epoch": 10.100536193029491,
      "grad_norm": 9.0625,
      "learning_rate": 2.890919272091821e-05,
      "loss": 1.1289,
      "num_input_tokens_seen": 39339920,
      "step": 67815
    },
    {
      "epoch": 10.10128090557045,
      "grad_norm": 11.5,
      "learning_rate": 2.8905983237872304e-05,
      "loss": 1.2178,
      "num_input_tokens_seen": 39342640,
      "step": 67820
    },
    {
      "epoch": 10.102025618111409,
      "grad_norm": 12.3125,
      "learning_rate": 2.890277368883878e-05,
      "loss": 1.406,
      "num_input_tokens_seen": 39345584,
      "step": 67825
    },
    {
      "epoch": 10.102770330652369,
      "grad_norm": 13.75,
      "learning_rate": 2.889956407387186e-05,
      "loss": 1.1962,
      "num_input_tokens_seen": 39348400,
      "step": 67830
    },
    {
      "epoch": 10.103515043193328,
      "grad_norm": 7.90625,
      "learning_rate": 2.8896354393025765e-05,
      "loss": 1.1767,
      "num_input_tokens_seen": 39351440,
      "step": 67835
    },
    {
      "epoch": 10.104259755734287,
      "grad_norm": 9.5,
      "learning_rate": 2.8893144646354725e-05,
      "loss": 1.155,
      "num_input_tokens_seen": 39354480,
      "step": 67840
    },
    {
      "epoch": 10.105004468275245,
      "grad_norm": 10.375,
      "learning_rate": 2.888993483391297e-05,
      "loss": 1.0291,
      "num_input_tokens_seen": 39357744,
      "step": 67845
    },
    {
      "epoch": 10.105749180816204,
      "grad_norm": 6.59375,
      "learning_rate": 2.8886724955754713e-05,
      "loss": 1.2061,
      "num_input_tokens_seen": 39360752,
      "step": 67850
    },
    {
      "epoch": 10.106493893357165,
      "grad_norm": 14.3125,
      "learning_rate": 2.8883515011934186e-05,
      "loss": 1.3762,
      "num_input_tokens_seen": 39363632,
      "step": 67855
    },
    {
      "epoch": 10.107238605898123,
      "grad_norm": 5.625,
      "learning_rate": 2.8880305002505624e-05,
      "loss": 1.1931,
      "num_input_tokens_seen": 39366672,
      "step": 67860
    },
    {
      "epoch": 10.107983318439082,
      "grad_norm": 9.0625,
      "learning_rate": 2.887709492752325e-05,
      "loss": 1.1484,
      "num_input_tokens_seen": 39369328,
      "step": 67865
    },
    {
      "epoch": 10.108728030980041,
      "grad_norm": 10.0625,
      "learning_rate": 2.8873884787041304e-05,
      "loss": 1.1016,
      "num_input_tokens_seen": 39372240,
      "step": 67870
    },
    {
      "epoch": 10.109472743521001,
      "grad_norm": 9.3125,
      "learning_rate": 2.8870674581114004e-05,
      "loss": 1.3853,
      "num_input_tokens_seen": 39375376,
      "step": 67875
    },
    {
      "epoch": 10.11021745606196,
      "grad_norm": 8.1875,
      "learning_rate": 2.88674643097956e-05,
      "loss": 1.3159,
      "num_input_tokens_seen": 39378512,
      "step": 67880
    },
    {
      "epoch": 10.110962168602919,
      "grad_norm": 3.34375,
      "learning_rate": 2.886425397314031e-05,
      "loss": 0.8526,
      "num_input_tokens_seen": 39381328,
      "step": 67885
    },
    {
      "epoch": 10.111706881143878,
      "grad_norm": 13.0,
      "learning_rate": 2.886104357120237e-05,
      "loss": 1.2054,
      "num_input_tokens_seen": 39384272,
      "step": 67890
    },
    {
      "epoch": 10.112451593684838,
      "grad_norm": 8.1875,
      "learning_rate": 2.8857833104036036e-05,
      "loss": 1.0147,
      "num_input_tokens_seen": 39386992,
      "step": 67895
    },
    {
      "epoch": 10.113196306225797,
      "grad_norm": 3.140625,
      "learning_rate": 2.8854622571695526e-05,
      "loss": 0.634,
      "num_input_tokens_seen": 39389904,
      "step": 67900
    },
    {
      "epoch": 10.113941018766756,
      "grad_norm": 12.1875,
      "learning_rate": 2.8851411974235086e-05,
      "loss": 1.046,
      "num_input_tokens_seen": 39392976,
      "step": 67905
    },
    {
      "epoch": 10.114685731307715,
      "grad_norm": 13.875,
      "learning_rate": 2.884820131170896e-05,
      "loss": 1.3512,
      "num_input_tokens_seen": 39395664,
      "step": 67910
    },
    {
      "epoch": 10.115430443848675,
      "grad_norm": 8.6875,
      "learning_rate": 2.884499058417138e-05,
      "loss": 1.5446,
      "num_input_tokens_seen": 39398352,
      "step": 67915
    },
    {
      "epoch": 10.116175156389634,
      "grad_norm": 13.75,
      "learning_rate": 2.8841779791676594e-05,
      "loss": 1.2116,
      "num_input_tokens_seen": 39401200,
      "step": 67920
    },
    {
      "epoch": 10.116919868930593,
      "grad_norm": 13.5625,
      "learning_rate": 2.8838568934278843e-05,
      "loss": 1.0021,
      "num_input_tokens_seen": 39403856,
      "step": 67925
    },
    {
      "epoch": 10.117664581471551,
      "grad_norm": 7.8125,
      "learning_rate": 2.8835358012032364e-05,
      "loss": 0.8421,
      "num_input_tokens_seen": 39406672,
      "step": 67930
    },
    {
      "epoch": 10.118409294012512,
      "grad_norm": 17.25,
      "learning_rate": 2.8832147024991412e-05,
      "loss": 1.0034,
      "num_input_tokens_seen": 39409680,
      "step": 67935
    },
    {
      "epoch": 10.11915400655347,
      "grad_norm": 7.21875,
      "learning_rate": 2.882893597321024e-05,
      "loss": 1.1346,
      "num_input_tokens_seen": 39412336,
      "step": 67940
    },
    {
      "epoch": 10.11989871909443,
      "grad_norm": 18.75,
      "learning_rate": 2.8825724856743075e-05,
      "loss": 1.092,
      "num_input_tokens_seen": 39415856,
      "step": 67945
    },
    {
      "epoch": 10.120643431635388,
      "grad_norm": 9.75,
      "learning_rate": 2.8822513675644192e-05,
      "loss": 1.2705,
      "num_input_tokens_seen": 39418640,
      "step": 67950
    },
    {
      "epoch": 10.121388144176349,
      "grad_norm": 9.8125,
      "learning_rate": 2.8819302429967808e-05,
      "loss": 1.0246,
      "num_input_tokens_seen": 39421520,
      "step": 67955
    },
    {
      "epoch": 10.122132856717307,
      "grad_norm": 14.75,
      "learning_rate": 2.88160911197682e-05,
      "loss": 0.9064,
      "num_input_tokens_seen": 39424528,
      "step": 67960
    },
    {
      "epoch": 10.122877569258266,
      "grad_norm": 9.0,
      "learning_rate": 2.881287974509961e-05,
      "loss": 0.7616,
      "num_input_tokens_seen": 39427184,
      "step": 67965
    },
    {
      "epoch": 10.123622281799225,
      "grad_norm": 5.65625,
      "learning_rate": 2.8809668306016286e-05,
      "loss": 1.1192,
      "num_input_tokens_seen": 39430288,
      "step": 67970
    },
    {
      "epoch": 10.124366994340185,
      "grad_norm": 13.6875,
      "learning_rate": 2.8806456802572502e-05,
      "loss": 0.8137,
      "num_input_tokens_seen": 39433168,
      "step": 67975
    },
    {
      "epoch": 10.125111706881144,
      "grad_norm": 6.46875,
      "learning_rate": 2.8803245234822485e-05,
      "loss": 0.7503,
      "num_input_tokens_seen": 39435920,
      "step": 67980
    },
    {
      "epoch": 10.125856419422103,
      "grad_norm": 9.75,
      "learning_rate": 2.880003360282051e-05,
      "loss": 0.9261,
      "num_input_tokens_seen": 39438768,
      "step": 67985
    },
    {
      "epoch": 10.126601131963062,
      "grad_norm": 7.21875,
      "learning_rate": 2.8796821906620837e-05,
      "loss": 1.2585,
      "num_input_tokens_seen": 39441712,
      "step": 67990
    },
    {
      "epoch": 10.127345844504022,
      "grad_norm": 11.0625,
      "learning_rate": 2.8793610146277707e-05,
      "loss": 1.2319,
      "num_input_tokens_seen": 39444496,
      "step": 67995
    },
    {
      "epoch": 10.128090557044981,
      "grad_norm": 11.0625,
      "learning_rate": 2.879039832184539e-05,
      "loss": 0.887,
      "num_input_tokens_seen": 39447216,
      "step": 68000
    },
    {
      "epoch": 10.12883526958594,
      "grad_norm": 9.1875,
      "learning_rate": 2.8787186433378142e-05,
      "loss": 1.2826,
      "num_input_tokens_seen": 39450256,
      "step": 68005
    },
    {
      "epoch": 10.129579982126899,
      "grad_norm": 5.59375,
      "learning_rate": 2.8783974480930244e-05,
      "loss": 1.0679,
      "num_input_tokens_seen": 39453104,
      "step": 68010
    },
    {
      "epoch": 10.130324694667857,
      "grad_norm": 8.1875,
      "learning_rate": 2.8780762464555928e-05,
      "loss": 1.4513,
      "num_input_tokens_seen": 39455888,
      "step": 68015
    },
    {
      "epoch": 10.131069407208818,
      "grad_norm": 11.25,
      "learning_rate": 2.8777550384309477e-05,
      "loss": 1.16,
      "num_input_tokens_seen": 39458736,
      "step": 68020
    },
    {
      "epoch": 10.131814119749777,
      "grad_norm": 7.46875,
      "learning_rate": 2.877433824024515e-05,
      "loss": 0.9353,
      "num_input_tokens_seen": 39461616,
      "step": 68025
    },
    {
      "epoch": 10.132558832290735,
      "grad_norm": 6.375,
      "learning_rate": 2.8771126032417222e-05,
      "loss": 0.9547,
      "num_input_tokens_seen": 39464656,
      "step": 68030
    },
    {
      "epoch": 10.133303544831694,
      "grad_norm": 7.3125,
      "learning_rate": 2.876791376087995e-05,
      "loss": 1.1297,
      "num_input_tokens_seen": 39467696,
      "step": 68035
    },
    {
      "epoch": 10.134048257372655,
      "grad_norm": 9.6875,
      "learning_rate": 2.8764701425687597e-05,
      "loss": 1.2355,
      "num_input_tokens_seen": 39470672,
      "step": 68040
    },
    {
      "epoch": 10.134792969913613,
      "grad_norm": 18.5,
      "learning_rate": 2.876148902689445e-05,
      "loss": 1.3043,
      "num_input_tokens_seen": 39473456,
      "step": 68045
    },
    {
      "epoch": 10.135537682454572,
      "grad_norm": 8.25,
      "learning_rate": 2.875827656455476e-05,
      "loss": 1.3082,
      "num_input_tokens_seen": 39476112,
      "step": 68050
    },
    {
      "epoch": 10.136282394995531,
      "grad_norm": 13.3125,
      "learning_rate": 2.8755064038722813e-05,
      "loss": 0.9538,
      "num_input_tokens_seen": 39479120,
      "step": 68055
    },
    {
      "epoch": 10.137027107536491,
      "grad_norm": 20.5,
      "learning_rate": 2.875185144945287e-05,
      "loss": 1.0844,
      "num_input_tokens_seen": 39482064,
      "step": 68060
    },
    {
      "epoch": 10.13777182007745,
      "grad_norm": 8.125,
      "learning_rate": 2.874863879679921e-05,
      "loss": 1.0519,
      "num_input_tokens_seen": 39484880,
      "step": 68065
    },
    {
      "epoch": 10.138516532618409,
      "grad_norm": 10.5,
      "learning_rate": 2.8745426080816117e-05,
      "loss": 1.4317,
      "num_input_tokens_seen": 39487952,
      "step": 68070
    },
    {
      "epoch": 10.139261245159368,
      "grad_norm": 7.6875,
      "learning_rate": 2.8742213301557847e-05,
      "loss": 1.0002,
      "num_input_tokens_seen": 39490928,
      "step": 68075
    },
    {
      "epoch": 10.140005957700328,
      "grad_norm": 8.375,
      "learning_rate": 2.8739000459078695e-05,
      "loss": 1.1187,
      "num_input_tokens_seen": 39494000,
      "step": 68080
    },
    {
      "epoch": 10.140750670241287,
      "grad_norm": 14.75,
      "learning_rate": 2.8735787553432925e-05,
      "loss": 1.5295,
      "num_input_tokens_seen": 39497104,
      "step": 68085
    },
    {
      "epoch": 10.141495382782246,
      "grad_norm": 11.6875,
      "learning_rate": 2.873257458467482e-05,
      "loss": 1.1603,
      "num_input_tokens_seen": 39500016,
      "step": 68090
    },
    {
      "epoch": 10.142240095323205,
      "grad_norm": 12.3125,
      "learning_rate": 2.8729361552858662e-05,
      "loss": 0.9407,
      "num_input_tokens_seen": 39502704,
      "step": 68095
    },
    {
      "epoch": 10.142984807864165,
      "grad_norm": 21.125,
      "learning_rate": 2.8726148458038732e-05,
      "loss": 1.2622,
      "num_input_tokens_seen": 39505296,
      "step": 68100
    },
    {
      "epoch": 10.143729520405124,
      "grad_norm": 8.0,
      "learning_rate": 2.8722935300269315e-05,
      "loss": 1.161,
      "num_input_tokens_seen": 39507984,
      "step": 68105
    },
    {
      "epoch": 10.144474232946083,
      "grad_norm": 11.5625,
      "learning_rate": 2.8719722079604684e-05,
      "loss": 0.954,
      "num_input_tokens_seen": 39510640,
      "step": 68110
    },
    {
      "epoch": 10.145218945487041,
      "grad_norm": 5.96875,
      "learning_rate": 2.8716508796099135e-05,
      "loss": 1.246,
      "num_input_tokens_seen": 39513552,
      "step": 68115
    },
    {
      "epoch": 10.145963658028002,
      "grad_norm": 9.0,
      "learning_rate": 2.8713295449806944e-05,
      "loss": 0.9696,
      "num_input_tokens_seen": 39516624,
      "step": 68120
    },
    {
      "epoch": 10.14670837056896,
      "grad_norm": 11.625,
      "learning_rate": 2.8710082040782392e-05,
      "loss": 1.1328,
      "num_input_tokens_seen": 39519504,
      "step": 68125
    },
    {
      "epoch": 10.14745308310992,
      "grad_norm": 12.5625,
      "learning_rate": 2.870686856907978e-05,
      "loss": 1.21,
      "num_input_tokens_seen": 39522608,
      "step": 68130
    },
    {
      "epoch": 10.148197795650878,
      "grad_norm": 11.4375,
      "learning_rate": 2.8703655034753397e-05,
      "loss": 1.464,
      "num_input_tokens_seen": 39525168,
      "step": 68135
    },
    {
      "epoch": 10.148942508191839,
      "grad_norm": 11.6875,
      "learning_rate": 2.8700441437857527e-05,
      "loss": 0.9806,
      "num_input_tokens_seen": 39528272,
      "step": 68140
    },
    {
      "epoch": 10.149687220732797,
      "grad_norm": 8.25,
      "learning_rate": 2.869722777844645e-05,
      "loss": 0.8666,
      "num_input_tokens_seen": 39531056,
      "step": 68145
    },
    {
      "epoch": 10.150431933273756,
      "grad_norm": 9.125,
      "learning_rate": 2.869401405657448e-05,
      "loss": 0.9471,
      "num_input_tokens_seen": 39533744,
      "step": 68150
    },
    {
      "epoch": 10.151176645814715,
      "grad_norm": 6.3125,
      "learning_rate": 2.8690800272295888e-05,
      "loss": 0.9723,
      "num_input_tokens_seen": 39536624,
      "step": 68155
    },
    {
      "epoch": 10.151921358355676,
      "grad_norm": 8.4375,
      "learning_rate": 2.8687586425664974e-05,
      "loss": 1.401,
      "num_input_tokens_seen": 39539728,
      "step": 68160
    },
    {
      "epoch": 10.152666070896634,
      "grad_norm": 9.0625,
      "learning_rate": 2.868437251673604e-05,
      "loss": 1.4198,
      "num_input_tokens_seen": 39542576,
      "step": 68165
    },
    {
      "epoch": 10.153410783437593,
      "grad_norm": 13.5625,
      "learning_rate": 2.8681158545563375e-05,
      "loss": 0.7159,
      "num_input_tokens_seen": 39545360,
      "step": 68170
    },
    {
      "epoch": 10.154155495978552,
      "grad_norm": 8.3125,
      "learning_rate": 2.8677944512201283e-05,
      "loss": 1.0477,
      "num_input_tokens_seen": 39548144,
      "step": 68175
    },
    {
      "epoch": 10.15490020851951,
      "grad_norm": 18.875,
      "learning_rate": 2.8674730416704056e-05,
      "loss": 1.3639,
      "num_input_tokens_seen": 39550800,
      "step": 68180
    },
    {
      "epoch": 10.155644921060471,
      "grad_norm": 16.375,
      "learning_rate": 2.8671516259125985e-05,
      "loss": 1.2359,
      "num_input_tokens_seen": 39553872,
      "step": 68185
    },
    {
      "epoch": 10.15638963360143,
      "grad_norm": 5.78125,
      "learning_rate": 2.866830203952139e-05,
      "loss": 1.2227,
      "num_input_tokens_seen": 39556688,
      "step": 68190
    },
    {
      "epoch": 10.157134346142389,
      "grad_norm": 11.6875,
      "learning_rate": 2.866508775794455e-05,
      "loss": 0.8721,
      "num_input_tokens_seen": 39559760,
      "step": 68195
    },
    {
      "epoch": 10.157879058683347,
      "grad_norm": 8.0625,
      "learning_rate": 2.866187341444978e-05,
      "loss": 1.1141,
      "num_input_tokens_seen": 39562448,
      "step": 68200
    },
    {
      "epoch": 10.158623771224308,
      "grad_norm": 13.8125,
      "learning_rate": 2.8658659009091383e-05,
      "loss": 1.0428,
      "num_input_tokens_seen": 39565136,
      "step": 68205
    },
    {
      "epoch": 10.159368483765267,
      "grad_norm": 9.375,
      "learning_rate": 2.865544454192366e-05,
      "loss": 0.919,
      "num_input_tokens_seen": 39568016,
      "step": 68210
    },
    {
      "epoch": 10.160113196306225,
      "grad_norm": 12.875,
      "learning_rate": 2.8652230013000914e-05,
      "loss": 1.3263,
      "num_input_tokens_seen": 39570768,
      "step": 68215
    },
    {
      "epoch": 10.160857908847184,
      "grad_norm": 4.875,
      "learning_rate": 2.8649015422377456e-05,
      "loss": 1.202,
      "num_input_tokens_seen": 39573552,
      "step": 68220
    },
    {
      "epoch": 10.161602621388145,
      "grad_norm": 16.25,
      "learning_rate": 2.864580077010759e-05,
      "loss": 0.876,
      "num_input_tokens_seen": 39576464,
      "step": 68225
    },
    {
      "epoch": 10.162347333929103,
      "grad_norm": 5.59375,
      "learning_rate": 2.8642586056245628e-05,
      "loss": 0.9983,
      "num_input_tokens_seen": 39579664,
      "step": 68230
    },
    {
      "epoch": 10.163092046470062,
      "grad_norm": 6.1875,
      "learning_rate": 2.8639371280845872e-05,
      "loss": 0.9358,
      "num_input_tokens_seen": 39582800,
      "step": 68235
    },
    {
      "epoch": 10.163836759011021,
      "grad_norm": 8.5,
      "learning_rate": 2.863615644396264e-05,
      "loss": 1.1762,
      "num_input_tokens_seen": 39585776,
      "step": 68240
    },
    {
      "epoch": 10.164581471551982,
      "grad_norm": 9.75,
      "learning_rate": 2.863294154565025e-05,
      "loss": 1.4015,
      "num_input_tokens_seen": 39588688,
      "step": 68245
    },
    {
      "epoch": 10.16532618409294,
      "grad_norm": 12.625,
      "learning_rate": 2.862972658596299e-05,
      "loss": 1.2391,
      "num_input_tokens_seen": 39591472,
      "step": 68250
    },
    {
      "epoch": 10.166070896633899,
      "grad_norm": 7.5625,
      "learning_rate": 2.8626511564955195e-05,
      "loss": 0.9187,
      "num_input_tokens_seen": 39594448,
      "step": 68255
    },
    {
      "epoch": 10.166815609174858,
      "grad_norm": 11.75,
      "learning_rate": 2.8623296482681166e-05,
      "loss": 1.2442,
      "num_input_tokens_seen": 39597360,
      "step": 68260
    },
    {
      "epoch": 10.167560321715818,
      "grad_norm": 10.125,
      "learning_rate": 2.862008133919523e-05,
      "loss": 0.9708,
      "num_input_tokens_seen": 39600176,
      "step": 68265
    },
    {
      "epoch": 10.168305034256777,
      "grad_norm": 6.78125,
      "learning_rate": 2.8616866134551706e-05,
      "loss": 1.0135,
      "num_input_tokens_seen": 39603184,
      "step": 68270
    },
    {
      "epoch": 10.169049746797736,
      "grad_norm": 10.25,
      "learning_rate": 2.86136508688049e-05,
      "loss": 1.1056,
      "num_input_tokens_seen": 39605840,
      "step": 68275
    },
    {
      "epoch": 10.169794459338695,
      "grad_norm": 7.71875,
      "learning_rate": 2.861043554200914e-05,
      "loss": 0.8802,
      "num_input_tokens_seen": 39608912,
      "step": 68280
    },
    {
      "epoch": 10.170539171879655,
      "grad_norm": 10.0,
      "learning_rate": 2.8607220154218734e-05,
      "loss": 0.7206,
      "num_input_tokens_seen": 39611664,
      "step": 68285
    },
    {
      "epoch": 10.171283884420614,
      "grad_norm": 10.625,
      "learning_rate": 2.860400470548801e-05,
      "loss": 1.2441,
      "num_input_tokens_seen": 39614512,
      "step": 68290
    },
    {
      "epoch": 10.172028596961573,
      "grad_norm": 11.3125,
      "learning_rate": 2.8600789195871286e-05,
      "loss": 1.0779,
      "num_input_tokens_seen": 39617264,
      "step": 68295
    },
    {
      "epoch": 10.172773309502531,
      "grad_norm": 8.875,
      "learning_rate": 2.8597573625422892e-05,
      "loss": 1.2235,
      "num_input_tokens_seen": 39619888,
      "step": 68300
    },
    {
      "epoch": 10.173518022043492,
      "grad_norm": 7.96875,
      "learning_rate": 2.859435799419715e-05,
      "loss": 0.798,
      "num_input_tokens_seen": 39622928,
      "step": 68305
    },
    {
      "epoch": 10.17426273458445,
      "grad_norm": 6.375,
      "learning_rate": 2.8591142302248392e-05,
      "loss": 0.9745,
      "num_input_tokens_seen": 39625904,
      "step": 68310
    },
    {
      "epoch": 10.17500744712541,
      "grad_norm": 4.46875,
      "learning_rate": 2.8587926549630923e-05,
      "loss": 0.8728,
      "num_input_tokens_seen": 39628656,
      "step": 68315
    },
    {
      "epoch": 10.175752159666368,
      "grad_norm": 11.0625,
      "learning_rate": 2.858471073639908e-05,
      "loss": 1.2153,
      "num_input_tokens_seen": 39631760,
      "step": 68320
    },
    {
      "epoch": 10.176496872207329,
      "grad_norm": 9.5625,
      "learning_rate": 2.8581494862607194e-05,
      "loss": 1.2841,
      "num_input_tokens_seen": 39634928,
      "step": 68325
    },
    {
      "epoch": 10.177241584748288,
      "grad_norm": 5.53125,
      "learning_rate": 2.8578278928309594e-05,
      "loss": 1.3761,
      "num_input_tokens_seen": 39637776,
      "step": 68330
    },
    {
      "epoch": 10.177986297289246,
      "grad_norm": 17.375,
      "learning_rate": 2.8575062933560605e-05,
      "loss": 1.0544,
      "num_input_tokens_seen": 39641008,
      "step": 68335
    },
    {
      "epoch": 10.178731009830205,
      "grad_norm": 8.3125,
      "learning_rate": 2.8571846878414565e-05,
      "loss": 0.7592,
      "num_input_tokens_seen": 39643952,
      "step": 68340
    },
    {
      "epoch": 10.179475722371166,
      "grad_norm": 8.1875,
      "learning_rate": 2.8568630762925803e-05,
      "loss": 1.3127,
      "num_input_tokens_seen": 39646448,
      "step": 68345
    },
    {
      "epoch": 10.180220434912124,
      "grad_norm": 9.5625,
      "learning_rate": 2.8565414587148654e-05,
      "loss": 0.9199,
      "num_input_tokens_seen": 39649200,
      "step": 68350
    },
    {
      "epoch": 10.180965147453083,
      "grad_norm": 12.4375,
      "learning_rate": 2.856219835113744e-05,
      "loss": 1.0003,
      "num_input_tokens_seen": 39652208,
      "step": 68355
    },
    {
      "epoch": 10.181709859994042,
      "grad_norm": 17.0,
      "learning_rate": 2.8558982054946515e-05,
      "loss": 1.3448,
      "num_input_tokens_seen": 39655344,
      "step": 68360
    },
    {
      "epoch": 10.182454572535,
      "grad_norm": 15.5,
      "learning_rate": 2.85557656986302e-05,
      "loss": 1.025,
      "num_input_tokens_seen": 39658384,
      "step": 68365
    },
    {
      "epoch": 10.183199285075961,
      "grad_norm": 15.0,
      "learning_rate": 2.8552549282242836e-05,
      "loss": 1.0225,
      "num_input_tokens_seen": 39661264,
      "step": 68370
    },
    {
      "epoch": 10.18394399761692,
      "grad_norm": 16.5,
      "learning_rate": 2.854933280583877e-05,
      "loss": 1.0177,
      "num_input_tokens_seen": 39664240,
      "step": 68375
    },
    {
      "epoch": 10.184688710157879,
      "grad_norm": 9.375,
      "learning_rate": 2.8546116269472322e-05,
      "loss": 1.3173,
      "num_input_tokens_seen": 39666768,
      "step": 68380
    },
    {
      "epoch": 10.185433422698837,
      "grad_norm": 9.6875,
      "learning_rate": 2.8542899673197847e-05,
      "loss": 1.2314,
      "num_input_tokens_seen": 39669616,
      "step": 68385
    },
    {
      "epoch": 10.186178135239798,
      "grad_norm": 12.6875,
      "learning_rate": 2.8539683017069697e-05,
      "loss": 0.8497,
      "num_input_tokens_seen": 39672784,
      "step": 68390
    },
    {
      "epoch": 10.186922847780757,
      "grad_norm": 21.5,
      "learning_rate": 2.8536466301142185e-05,
      "loss": 1.0516,
      "num_input_tokens_seen": 39675664,
      "step": 68395
    },
    {
      "epoch": 10.187667560321715,
      "grad_norm": 7.25,
      "learning_rate": 2.853324952546967e-05,
      "loss": 1.3792,
      "num_input_tokens_seen": 39678512,
      "step": 68400
    },
    {
      "epoch": 10.188412272862674,
      "grad_norm": 7.3125,
      "learning_rate": 2.8530032690106494e-05,
      "loss": 1.1965,
      "num_input_tokens_seen": 39681424,
      "step": 68405
    },
    {
      "epoch": 10.189156985403635,
      "grad_norm": 7.03125,
      "learning_rate": 2.8526815795107016e-05,
      "loss": 0.7663,
      "num_input_tokens_seen": 39684112,
      "step": 68410
    },
    {
      "epoch": 10.189901697944594,
      "grad_norm": 10.875,
      "learning_rate": 2.8523598840525563e-05,
      "loss": 1.0784,
      "num_input_tokens_seen": 39686800,
      "step": 68415
    },
    {
      "epoch": 10.190646410485552,
      "grad_norm": 12.0,
      "learning_rate": 2.852038182641648e-05,
      "loss": 1.2739,
      "num_input_tokens_seen": 39689616,
      "step": 68420
    },
    {
      "epoch": 10.191391123026511,
      "grad_norm": 9.125,
      "learning_rate": 2.8517164752834136e-05,
      "loss": 1.0273,
      "num_input_tokens_seen": 39692496,
      "step": 68425
    },
    {
      "epoch": 10.192135835567472,
      "grad_norm": 8.8125,
      "learning_rate": 2.8513947619832866e-05,
      "loss": 1.0055,
      "num_input_tokens_seen": 39695280,
      "step": 68430
    },
    {
      "epoch": 10.19288054810843,
      "grad_norm": 12.125,
      "learning_rate": 2.8510730427467015e-05,
      "loss": 0.9725,
      "num_input_tokens_seen": 39698448,
      "step": 68435
    },
    {
      "epoch": 10.19362526064939,
      "grad_norm": 5.71875,
      "learning_rate": 2.8507513175790944e-05,
      "loss": 0.9136,
      "num_input_tokens_seen": 39701456,
      "step": 68440
    },
    {
      "epoch": 10.194369973190348,
      "grad_norm": 6.78125,
      "learning_rate": 2.850429586485901e-05,
      "loss": 1.127,
      "num_input_tokens_seen": 39704528,
      "step": 68445
    },
    {
      "epoch": 10.195114685731308,
      "grad_norm": 11.3125,
      "learning_rate": 2.850107849472555e-05,
      "loss": 1.276,
      "num_input_tokens_seen": 39707440,
      "step": 68450
    },
    {
      "epoch": 10.195859398272267,
      "grad_norm": 8.0625,
      "learning_rate": 2.8497861065444937e-05,
      "loss": 0.9882,
      "num_input_tokens_seen": 39710320,
      "step": 68455
    },
    {
      "epoch": 10.196604110813226,
      "grad_norm": 6.34375,
      "learning_rate": 2.8494643577071506e-05,
      "loss": 1.052,
      "num_input_tokens_seen": 39713360,
      "step": 68460
    },
    {
      "epoch": 10.197348823354185,
      "grad_norm": 8.4375,
      "learning_rate": 2.849142602965963e-05,
      "loss": 1.2983,
      "num_input_tokens_seen": 39716272,
      "step": 68465
    },
    {
      "epoch": 10.198093535895145,
      "grad_norm": 19.125,
      "learning_rate": 2.8488208423263663e-05,
      "loss": 1.1593,
      "num_input_tokens_seen": 39718992,
      "step": 68470
    },
    {
      "epoch": 10.198838248436104,
      "grad_norm": 14.75,
      "learning_rate": 2.8484990757937958e-05,
      "loss": 1.4633,
      "num_input_tokens_seen": 39721936,
      "step": 68475
    },
    {
      "epoch": 10.199582960977063,
      "grad_norm": 10.9375,
      "learning_rate": 2.848177303373687e-05,
      "loss": 1.2327,
      "num_input_tokens_seen": 39725264,
      "step": 68480
    },
    {
      "epoch": 10.200327673518021,
      "grad_norm": 12.3125,
      "learning_rate": 2.847855525071477e-05,
      "loss": 0.9051,
      "num_input_tokens_seen": 39728048,
      "step": 68485
    },
    {
      "epoch": 10.201072386058982,
      "grad_norm": 14.375,
      "learning_rate": 2.8475337408926005e-05,
      "loss": 1.0797,
      "num_input_tokens_seen": 39730800,
      "step": 68490
    },
    {
      "epoch": 10.20181709859994,
      "grad_norm": 6.75,
      "learning_rate": 2.8472119508424954e-05,
      "loss": 0.8766,
      "num_input_tokens_seen": 39733680,
      "step": 68495
    },
    {
      "epoch": 10.2025618111409,
      "grad_norm": 9.3125,
      "learning_rate": 2.8468901549265976e-05,
      "loss": 0.8874,
      "num_input_tokens_seen": 39736720,
      "step": 68500
    },
    {
      "epoch": 10.203306523681858,
      "grad_norm": 20.5,
      "learning_rate": 2.8465683531503435e-05,
      "loss": 1.3053,
      "num_input_tokens_seen": 39739248,
      "step": 68505
    },
    {
      "epoch": 10.204051236222819,
      "grad_norm": 13.1875,
      "learning_rate": 2.8462465455191682e-05,
      "loss": 1.2825,
      "num_input_tokens_seen": 39742000,
      "step": 68510
    },
    {
      "epoch": 10.204795948763778,
      "grad_norm": 12.75,
      "learning_rate": 2.845924732038511e-05,
      "loss": 1.1988,
      "num_input_tokens_seen": 39744784,
      "step": 68515
    },
    {
      "epoch": 10.205540661304736,
      "grad_norm": 5.71875,
      "learning_rate": 2.8456029127138056e-05,
      "loss": 1.1691,
      "num_input_tokens_seen": 39747600,
      "step": 68520
    },
    {
      "epoch": 10.206285373845695,
      "grad_norm": 11.125,
      "learning_rate": 2.8452810875504903e-05,
      "loss": 1.0389,
      "num_input_tokens_seen": 39750448,
      "step": 68525
    },
    {
      "epoch": 10.207030086386654,
      "grad_norm": 9.5625,
      "learning_rate": 2.8449592565540024e-05,
      "loss": 1.2268,
      "num_input_tokens_seen": 39753168,
      "step": 68530
    },
    {
      "epoch": 10.207774798927614,
      "grad_norm": 8.9375,
      "learning_rate": 2.844637419729778e-05,
      "loss": 0.7802,
      "num_input_tokens_seen": 39755984,
      "step": 68535
    },
    {
      "epoch": 10.208519511468573,
      "grad_norm": 12.8125,
      "learning_rate": 2.844315577083255e-05,
      "loss": 0.8726,
      "num_input_tokens_seen": 39758736,
      "step": 68540
    },
    {
      "epoch": 10.209264224009532,
      "grad_norm": 18.125,
      "learning_rate": 2.8439937286198704e-05,
      "loss": 1.2307,
      "num_input_tokens_seen": 39761808,
      "step": 68545
    },
    {
      "epoch": 10.21000893655049,
      "grad_norm": 11.8125,
      "learning_rate": 2.8436718743450614e-05,
      "loss": 1.2101,
      "num_input_tokens_seen": 39764656,
      "step": 68550
    },
    {
      "epoch": 10.210753649091451,
      "grad_norm": 9.3125,
      "learning_rate": 2.8433500142642654e-05,
      "loss": 1.0583,
      "num_input_tokens_seen": 39767600,
      "step": 68555
    },
    {
      "epoch": 10.21149836163241,
      "grad_norm": 4.03125,
      "learning_rate": 2.8430281483829196e-05,
      "loss": 1.1527,
      "num_input_tokens_seen": 39770416,
      "step": 68560
    },
    {
      "epoch": 10.212243074173369,
      "grad_norm": 6.46875,
      "learning_rate": 2.842706276706462e-05,
      "loss": 0.9247,
      "num_input_tokens_seen": 39773200,
      "step": 68565
    },
    {
      "epoch": 10.212987786714327,
      "grad_norm": 9.625,
      "learning_rate": 2.8423843992403298e-05,
      "loss": 0.938,
      "num_input_tokens_seen": 39776272,
      "step": 68570
    },
    {
      "epoch": 10.213732499255288,
      "grad_norm": 7.40625,
      "learning_rate": 2.8420625159899622e-05,
      "loss": 1.3379,
      "num_input_tokens_seen": 39779280,
      "step": 68575
    },
    {
      "epoch": 10.214477211796247,
      "grad_norm": 8.625,
      "learning_rate": 2.8417406269607954e-05,
      "loss": 1.0549,
      "num_input_tokens_seen": 39782384,
      "step": 68580
    },
    {
      "epoch": 10.215221924337206,
      "grad_norm": 8.625,
      "learning_rate": 2.8414187321582676e-05,
      "loss": 0.8905,
      "num_input_tokens_seen": 39785616,
      "step": 68585
    },
    {
      "epoch": 10.215966636878164,
      "grad_norm": 11.0625,
      "learning_rate": 2.8410968315878178e-05,
      "loss": 1.0528,
      "num_input_tokens_seen": 39788816,
      "step": 68590
    },
    {
      "epoch": 10.216711349419125,
      "grad_norm": 9.9375,
      "learning_rate": 2.8407749252548843e-05,
      "loss": 0.9083,
      "num_input_tokens_seen": 39791504,
      "step": 68595
    },
    {
      "epoch": 10.217456061960084,
      "grad_norm": 8.8125,
      "learning_rate": 2.8404530131649036e-05,
      "loss": 0.7612,
      "num_input_tokens_seen": 39794320,
      "step": 68600
    },
    {
      "epoch": 10.218200774501042,
      "grad_norm": 17.25,
      "learning_rate": 2.8401310953233158e-05,
      "loss": 0.9648,
      "num_input_tokens_seen": 39797200,
      "step": 68605
    },
    {
      "epoch": 10.218945487042001,
      "grad_norm": 17.625,
      "learning_rate": 2.839809171735559e-05,
      "loss": 1.073,
      "num_input_tokens_seen": 39800304,
      "step": 68610
    },
    {
      "epoch": 10.219690199582962,
      "grad_norm": 11.375,
      "learning_rate": 2.8394872424070716e-05,
      "loss": 1.1509,
      "num_input_tokens_seen": 39803088,
      "step": 68615
    },
    {
      "epoch": 10.22043491212392,
      "grad_norm": 14.4375,
      "learning_rate": 2.8391653073432918e-05,
      "loss": 1.0131,
      "num_input_tokens_seen": 39806160,
      "step": 68620
    },
    {
      "epoch": 10.22117962466488,
      "grad_norm": 15.8125,
      "learning_rate": 2.838843366549659e-05,
      "loss": 0.82,
      "num_input_tokens_seen": 39809136,
      "step": 68625
    },
    {
      "epoch": 10.221924337205838,
      "grad_norm": 6.9375,
      "learning_rate": 2.8385214200316118e-05,
      "loss": 1.046,
      "num_input_tokens_seen": 39811856,
      "step": 68630
    },
    {
      "epoch": 10.222669049746798,
      "grad_norm": 5.875,
      "learning_rate": 2.83819946779459e-05,
      "loss": 0.8001,
      "num_input_tokens_seen": 39814768,
      "step": 68635
    },
    {
      "epoch": 10.223413762287757,
      "grad_norm": 9.0,
      "learning_rate": 2.8378775098440318e-05,
      "loss": 1.3129,
      "num_input_tokens_seen": 39818032,
      "step": 68640
    },
    {
      "epoch": 10.224158474828716,
      "grad_norm": 8.125,
      "learning_rate": 2.8375555461853764e-05,
      "loss": 1.037,
      "num_input_tokens_seen": 39820912,
      "step": 68645
    },
    {
      "epoch": 10.224903187369675,
      "grad_norm": 6.0,
      "learning_rate": 2.8372335768240626e-05,
      "loss": 0.8678,
      "num_input_tokens_seen": 39823824,
      "step": 68650
    },
    {
      "epoch": 10.225647899910635,
      "grad_norm": 12.0625,
      "learning_rate": 2.8369116017655307e-05,
      "loss": 1.2498,
      "num_input_tokens_seen": 39826896,
      "step": 68655
    },
    {
      "epoch": 10.226392612451594,
      "grad_norm": 17.125,
      "learning_rate": 2.836589621015219e-05,
      "loss": 0.9375,
      "num_input_tokens_seen": 39830128,
      "step": 68660
    },
    {
      "epoch": 10.227137324992553,
      "grad_norm": 11.125,
      "learning_rate": 2.8362676345785683e-05,
      "loss": 1.4147,
      "num_input_tokens_seen": 39832784,
      "step": 68665
    },
    {
      "epoch": 10.227882037533512,
      "grad_norm": 15.5625,
      "learning_rate": 2.835945642461018e-05,
      "loss": 0.9618,
      "num_input_tokens_seen": 39835664,
      "step": 68670
    },
    {
      "epoch": 10.228626750074472,
      "grad_norm": 6.8125,
      "learning_rate": 2.8356236446680073e-05,
      "loss": 0.8996,
      "num_input_tokens_seen": 39838480,
      "step": 68675
    },
    {
      "epoch": 10.22937146261543,
      "grad_norm": 14.8125,
      "learning_rate": 2.835301641204976e-05,
      "loss": 1.2305,
      "num_input_tokens_seen": 39841360,
      "step": 68680
    },
    {
      "epoch": 10.23011617515639,
      "grad_norm": 8.25,
      "learning_rate": 2.834979632077364e-05,
      "loss": 0.8525,
      "num_input_tokens_seen": 39844304,
      "step": 68685
    },
    {
      "epoch": 10.230860887697348,
      "grad_norm": 7.625,
      "learning_rate": 2.834657617290612e-05,
      "loss": 1.0597,
      "num_input_tokens_seen": 39847056,
      "step": 68690
    },
    {
      "epoch": 10.231605600238307,
      "grad_norm": 16.875,
      "learning_rate": 2.8343355968501596e-05,
      "loss": 1.0436,
      "num_input_tokens_seen": 39849840,
      "step": 68695
    },
    {
      "epoch": 10.232350312779268,
      "grad_norm": 13.75,
      "learning_rate": 2.8340135707614467e-05,
      "loss": 1.1921,
      "num_input_tokens_seen": 39852720,
      "step": 68700
    },
    {
      "epoch": 10.233095025320226,
      "grad_norm": 13.0625,
      "learning_rate": 2.8336915390299152e-05,
      "loss": 1.4014,
      "num_input_tokens_seen": 39855280,
      "step": 68705
    },
    {
      "epoch": 10.233839737861185,
      "grad_norm": 9.9375,
      "learning_rate": 2.8333695016610034e-05,
      "loss": 0.9882,
      "num_input_tokens_seen": 39858192,
      "step": 68710
    },
    {
      "epoch": 10.234584450402144,
      "grad_norm": 7.65625,
      "learning_rate": 2.833047458660153e-05,
      "loss": 1.2893,
      "num_input_tokens_seen": 39860848,
      "step": 68715
    },
    {
      "epoch": 10.235329162943104,
      "grad_norm": 5.6875,
      "learning_rate": 2.8327254100328044e-05,
      "loss": 0.7383,
      "num_input_tokens_seen": 39863952,
      "step": 68720
    },
    {
      "epoch": 10.236073875484063,
      "grad_norm": 8.0625,
      "learning_rate": 2.8324033557843975e-05,
      "loss": 1.1999,
      "num_input_tokens_seen": 39866896,
      "step": 68725
    },
    {
      "epoch": 10.236818588025022,
      "grad_norm": 5.96875,
      "learning_rate": 2.832081295920374e-05,
      "loss": 1.0379,
      "num_input_tokens_seen": 39869936,
      "step": 68730
    },
    {
      "epoch": 10.23756330056598,
      "grad_norm": 4.53125,
      "learning_rate": 2.8317592304461744e-05,
      "loss": 0.8668,
      "num_input_tokens_seen": 39872848,
      "step": 68735
    },
    {
      "epoch": 10.238308013106941,
      "grad_norm": 13.6875,
      "learning_rate": 2.8314371593672408e-05,
      "loss": 1.0509,
      "num_input_tokens_seen": 39875600,
      "step": 68740
    },
    {
      "epoch": 10.2390527256479,
      "grad_norm": 14.0,
      "learning_rate": 2.8311150826890122e-05,
      "loss": 0.5756,
      "num_input_tokens_seen": 39878544,
      "step": 68745
    },
    {
      "epoch": 10.239797438188859,
      "grad_norm": 7.96875,
      "learning_rate": 2.830793000416931e-05,
      "loss": 0.9971,
      "num_input_tokens_seen": 39881456,
      "step": 68750
    },
    {
      "epoch": 10.240542150729818,
      "grad_norm": 11.125,
      "learning_rate": 2.8304709125564382e-05,
      "loss": 1.0204,
      "num_input_tokens_seen": 39884240,
      "step": 68755
    },
    {
      "epoch": 10.241286863270778,
      "grad_norm": 14.0,
      "learning_rate": 2.8301488191129756e-05,
      "loss": 1.0603,
      "num_input_tokens_seen": 39886992,
      "step": 68760
    },
    {
      "epoch": 10.242031575811737,
      "grad_norm": 13.875,
      "learning_rate": 2.8298267200919836e-05,
      "loss": 1.1457,
      "num_input_tokens_seen": 39889968,
      "step": 68765
    },
    {
      "epoch": 10.242776288352696,
      "grad_norm": 4.625,
      "learning_rate": 2.8295046154989047e-05,
      "loss": 1.3586,
      "num_input_tokens_seen": 39892752,
      "step": 68770
    },
    {
      "epoch": 10.243521000893654,
      "grad_norm": 7.71875,
      "learning_rate": 2.8291825053391808e-05,
      "loss": 1.1688,
      "num_input_tokens_seen": 39895824,
      "step": 68775
    },
    {
      "epoch": 10.244265713434615,
      "grad_norm": 8.1875,
      "learning_rate": 2.828860389618252e-05,
      "loss": 0.9577,
      "num_input_tokens_seen": 39898640,
      "step": 68780
    },
    {
      "epoch": 10.245010425975574,
      "grad_norm": 7.3125,
      "learning_rate": 2.8285382683415617e-05,
      "loss": 1.2377,
      "num_input_tokens_seen": 39901424,
      "step": 68785
    },
    {
      "epoch": 10.245755138516532,
      "grad_norm": 6.15625,
      "learning_rate": 2.8282161415145513e-05,
      "loss": 0.8852,
      "num_input_tokens_seen": 39904368,
      "step": 68790
    },
    {
      "epoch": 10.246499851057491,
      "grad_norm": 8.5,
      "learning_rate": 2.827894009142663e-05,
      "loss": 1.1094,
      "num_input_tokens_seen": 39907216,
      "step": 68795
    },
    {
      "epoch": 10.247244563598452,
      "grad_norm": 11.6875,
      "learning_rate": 2.827571871231338e-05,
      "loss": 1.4601,
      "num_input_tokens_seen": 39910160,
      "step": 68800
    },
    {
      "epoch": 10.24798927613941,
      "grad_norm": 15.25,
      "learning_rate": 2.82724972778602e-05,
      "loss": 1.3986,
      "num_input_tokens_seen": 39913008,
      "step": 68805
    },
    {
      "epoch": 10.24873398868037,
      "grad_norm": 12.75,
      "learning_rate": 2.8269275788121503e-05,
      "loss": 1.1656,
      "num_input_tokens_seen": 39915824,
      "step": 68810
    },
    {
      "epoch": 10.249478701221328,
      "grad_norm": 11.5,
      "learning_rate": 2.8266054243151708e-05,
      "loss": 0.9561,
      "num_input_tokens_seen": 39918736,
      "step": 68815
    },
    {
      "epoch": 10.250223413762289,
      "grad_norm": 15.6875,
      "learning_rate": 2.8262832643005242e-05,
      "loss": 0.8107,
      "num_input_tokens_seen": 39922096,
      "step": 68820
    },
    {
      "epoch": 10.250968126303247,
      "grad_norm": 4.375,
      "learning_rate": 2.8259610987736545e-05,
      "loss": 0.8786,
      "num_input_tokens_seen": 39925104,
      "step": 68825
    },
    {
      "epoch": 10.251712838844206,
      "grad_norm": 6.3125,
      "learning_rate": 2.825638927740003e-05,
      "loss": 0.9787,
      "num_input_tokens_seen": 39927792,
      "step": 68830
    },
    {
      "epoch": 10.252457551385165,
      "grad_norm": 6.34375,
      "learning_rate": 2.825316751205013e-05,
      "loss": 1.1628,
      "num_input_tokens_seen": 39930736,
      "step": 68835
    },
    {
      "epoch": 10.253202263926125,
      "grad_norm": 9.75,
      "learning_rate": 2.8249945691741276e-05,
      "loss": 1.0632,
      "num_input_tokens_seen": 39933616,
      "step": 68840
    },
    {
      "epoch": 10.253946976467084,
      "grad_norm": 5.84375,
      "learning_rate": 2.824672381652788e-05,
      "loss": 1.2353,
      "num_input_tokens_seen": 39936688,
      "step": 68845
    },
    {
      "epoch": 10.254691689008043,
      "grad_norm": 13.4375,
      "learning_rate": 2.8243501886464392e-05,
      "loss": 1.2246,
      "num_input_tokens_seen": 39939536,
      "step": 68850
    },
    {
      "epoch": 10.255436401549002,
      "grad_norm": 4.5,
      "learning_rate": 2.8240279901605238e-05,
      "loss": 0.8598,
      "num_input_tokens_seen": 39942672,
      "step": 68855
    },
    {
      "epoch": 10.256181114089962,
      "grad_norm": 10.25,
      "learning_rate": 2.823705786200484e-05,
      "loss": 0.9057,
      "num_input_tokens_seen": 39945584,
      "step": 68860
    },
    {
      "epoch": 10.256925826630921,
      "grad_norm": 11.8125,
      "learning_rate": 2.8233835767717642e-05,
      "loss": 1.1326,
      "num_input_tokens_seen": 39948368,
      "step": 68865
    },
    {
      "epoch": 10.25767053917188,
      "grad_norm": 15.6875,
      "learning_rate": 2.8230613618798086e-05,
      "loss": 1.2483,
      "num_input_tokens_seen": 39951056,
      "step": 68870
    },
    {
      "epoch": 10.258415251712838,
      "grad_norm": 11.75,
      "learning_rate": 2.822739141530059e-05,
      "loss": 0.9447,
      "num_input_tokens_seen": 39953872,
      "step": 68875
    },
    {
      "epoch": 10.259159964253797,
      "grad_norm": 8.6875,
      "learning_rate": 2.8224169157279597e-05,
      "loss": 0.8575,
      "num_input_tokens_seen": 39956976,
      "step": 68880
    },
    {
      "epoch": 10.259904676794758,
      "grad_norm": 10.9375,
      "learning_rate": 2.8220946844789535e-05,
      "loss": 0.97,
      "num_input_tokens_seen": 39959920,
      "step": 68885
    },
    {
      "epoch": 10.260649389335716,
      "grad_norm": 20.875,
      "learning_rate": 2.8217724477884854e-05,
      "loss": 1.3803,
      "num_input_tokens_seen": 39962672,
      "step": 68890
    },
    {
      "epoch": 10.261394101876675,
      "grad_norm": 22.125,
      "learning_rate": 2.821450205661999e-05,
      "loss": 1.355,
      "num_input_tokens_seen": 39965456,
      "step": 68895
    },
    {
      "epoch": 10.262138814417634,
      "grad_norm": 11.1875,
      "learning_rate": 2.8211279581049384e-05,
      "loss": 0.8835,
      "num_input_tokens_seen": 39968304,
      "step": 68900
    },
    {
      "epoch": 10.262883526958595,
      "grad_norm": 12.75,
      "learning_rate": 2.8208057051227473e-05,
      "loss": 1.0859,
      "num_input_tokens_seen": 39971088,
      "step": 68905
    },
    {
      "epoch": 10.263628239499553,
      "grad_norm": 9.6875,
      "learning_rate": 2.820483446720869e-05,
      "loss": 0.8445,
      "num_input_tokens_seen": 39974096,
      "step": 68910
    },
    {
      "epoch": 10.264372952040512,
      "grad_norm": 18.875,
      "learning_rate": 2.8201611829047498e-05,
      "loss": 1.2827,
      "num_input_tokens_seen": 39976912,
      "step": 68915
    },
    {
      "epoch": 10.26511766458147,
      "grad_norm": 8.75,
      "learning_rate": 2.819838913679832e-05,
      "loss": 1.2901,
      "num_input_tokens_seen": 39979824,
      "step": 68920
    },
    {
      "epoch": 10.265862377122431,
      "grad_norm": 8.6875,
      "learning_rate": 2.819516639051561e-05,
      "loss": 1.1808,
      "num_input_tokens_seen": 39982512,
      "step": 68925
    },
    {
      "epoch": 10.26660708966339,
      "grad_norm": 14.0,
      "learning_rate": 2.8191943590253806e-05,
      "loss": 1.371,
      "num_input_tokens_seen": 39985616,
      "step": 68930
    },
    {
      "epoch": 10.267351802204349,
      "grad_norm": 6.59375,
      "learning_rate": 2.8188720736067364e-05,
      "loss": 1.0287,
      "num_input_tokens_seen": 39988560,
      "step": 68935
    },
    {
      "epoch": 10.268096514745308,
      "grad_norm": 11.4375,
      "learning_rate": 2.818549782801073e-05,
      "loss": 0.8413,
      "num_input_tokens_seen": 39991184,
      "step": 68940
    },
    {
      "epoch": 10.268841227286268,
      "grad_norm": 13.9375,
      "learning_rate": 2.8182274866138343e-05,
      "loss": 1.2961,
      "num_input_tokens_seen": 39994032,
      "step": 68945
    },
    {
      "epoch": 10.269585939827227,
      "grad_norm": 6.03125,
      "learning_rate": 2.8179051850504656e-05,
      "loss": 0.9773,
      "num_input_tokens_seen": 39996784,
      "step": 68950
    },
    {
      "epoch": 10.270330652368186,
      "grad_norm": 13.4375,
      "learning_rate": 2.8175828781164127e-05,
      "loss": 1.1798,
      "num_input_tokens_seen": 39999568,
      "step": 68955
    },
    {
      "epoch": 10.271075364909144,
      "grad_norm": 13.0,
      "learning_rate": 2.8172605658171192e-05,
      "loss": 0.9154,
      "num_input_tokens_seen": 40002320,
      "step": 68960
    },
    {
      "epoch": 10.271820077450105,
      "grad_norm": 13.1875,
      "learning_rate": 2.8169382481580303e-05,
      "loss": 1.2386,
      "num_input_tokens_seen": 40005200,
      "step": 68965
    },
    {
      "epoch": 10.272564789991064,
      "grad_norm": 14.875,
      "learning_rate": 2.8166159251445928e-05,
      "loss": 1.0821,
      "num_input_tokens_seen": 40008016,
      "step": 68970
    },
    {
      "epoch": 10.273309502532022,
      "grad_norm": 20.0,
      "learning_rate": 2.8162935967822505e-05,
      "loss": 1.0942,
      "num_input_tokens_seen": 40010928,
      "step": 68975
    },
    {
      "epoch": 10.274054215072981,
      "grad_norm": 6.03125,
      "learning_rate": 2.8159712630764494e-05,
      "loss": 0.8241,
      "num_input_tokens_seen": 40013712,
      "step": 68980
    },
    {
      "epoch": 10.274798927613942,
      "grad_norm": 23.125,
      "learning_rate": 2.815648924032635e-05,
      "loss": 1.3923,
      "num_input_tokens_seen": 40016400,
      "step": 68985
    },
    {
      "epoch": 10.2755436401549,
      "grad_norm": 11.25,
      "learning_rate": 2.8153265796562528e-05,
      "loss": 0.8922,
      "num_input_tokens_seen": 40019216,
      "step": 68990
    },
    {
      "epoch": 10.27628835269586,
      "grad_norm": 7.6875,
      "learning_rate": 2.815004229952749e-05,
      "loss": 1.0523,
      "num_input_tokens_seen": 40021968,
      "step": 68995
    },
    {
      "epoch": 10.277033065236818,
      "grad_norm": 7.9375,
      "learning_rate": 2.8146818749275684e-05,
      "loss": 1.121,
      "num_input_tokens_seen": 40024880,
      "step": 69000
    },
    {
      "epoch": 10.277777777777779,
      "grad_norm": 16.5,
      "learning_rate": 2.814359514586158e-05,
      "loss": 1.0662,
      "num_input_tokens_seen": 40027504,
      "step": 69005
    },
    {
      "epoch": 10.278522490318737,
      "grad_norm": 11.625,
      "learning_rate": 2.8140371489339624e-05,
      "loss": 1.3213,
      "num_input_tokens_seen": 40030352,
      "step": 69010
    },
    {
      "epoch": 10.279267202859696,
      "grad_norm": 13.5625,
      "learning_rate": 2.8137147779764285e-05,
      "loss": 1.3312,
      "num_input_tokens_seen": 40033360,
      "step": 69015
    },
    {
      "epoch": 10.280011915400655,
      "grad_norm": 13.0625,
      "learning_rate": 2.8133924017190023e-05,
      "loss": 0.9711,
      "num_input_tokens_seen": 40036144,
      "step": 69020
    },
    {
      "epoch": 10.280756627941615,
      "grad_norm": 9.625,
      "learning_rate": 2.8130700201671296e-05,
      "loss": 0.8943,
      "num_input_tokens_seen": 40038832,
      "step": 69025
    },
    {
      "epoch": 10.281501340482574,
      "grad_norm": 11.375,
      "learning_rate": 2.812747633326257e-05,
      "loss": 1.0154,
      "num_input_tokens_seen": 40041680,
      "step": 69030
    },
    {
      "epoch": 10.282246053023533,
      "grad_norm": 16.125,
      "learning_rate": 2.812425241201832e-05,
      "loss": 0.9864,
      "num_input_tokens_seen": 40044304,
      "step": 69035
    },
    {
      "epoch": 10.282990765564492,
      "grad_norm": 10.625,
      "learning_rate": 2.8121028437993002e-05,
      "loss": 1.2675,
      "num_input_tokens_seen": 40047312,
      "step": 69040
    },
    {
      "epoch": 10.283735478105452,
      "grad_norm": 8.9375,
      "learning_rate": 2.8117804411241074e-05,
      "loss": 1.3491,
      "num_input_tokens_seen": 40050128,
      "step": 69045
    },
    {
      "epoch": 10.284480190646411,
      "grad_norm": 6.71875,
      "learning_rate": 2.8114580331817004e-05,
      "loss": 0.8817,
      "num_input_tokens_seen": 40053072,
      "step": 69050
    },
    {
      "epoch": 10.28522490318737,
      "grad_norm": 9.8125,
      "learning_rate": 2.8111356199775268e-05,
      "loss": 1.1168,
      "num_input_tokens_seen": 40055792,
      "step": 69055
    },
    {
      "epoch": 10.285969615728328,
      "grad_norm": 4.84375,
      "learning_rate": 2.8108132015170337e-05,
      "loss": 1.0815,
      "num_input_tokens_seen": 40058640,
      "step": 69060
    },
    {
      "epoch": 10.286714328269287,
      "grad_norm": 5.3125,
      "learning_rate": 2.8104907778056667e-05,
      "loss": 0.9554,
      "num_input_tokens_seen": 40061968,
      "step": 69065
    },
    {
      "epoch": 10.287459040810248,
      "grad_norm": 7.71875,
      "learning_rate": 2.8101683488488745e-05,
      "loss": 1.0128,
      "num_input_tokens_seen": 40064912,
      "step": 69070
    },
    {
      "epoch": 10.288203753351207,
      "grad_norm": 4.96875,
      "learning_rate": 2.8098459146521026e-05,
      "loss": 1.1554,
      "num_input_tokens_seen": 40067888,
      "step": 69075
    },
    {
      "epoch": 10.288948465892165,
      "grad_norm": 12.8125,
      "learning_rate": 2.8095234752207993e-05,
      "loss": 1.2382,
      "num_input_tokens_seen": 40070704,
      "step": 69080
    },
    {
      "epoch": 10.289693178433124,
      "grad_norm": 11.125,
      "learning_rate": 2.809201030560411e-05,
      "loss": 1.332,
      "num_input_tokens_seen": 40073392,
      "step": 69085
    },
    {
      "epoch": 10.290437890974085,
      "grad_norm": 10.3125,
      "learning_rate": 2.8088785806763856e-05,
      "loss": 1.2482,
      "num_input_tokens_seen": 40076304,
      "step": 69090
    },
    {
      "epoch": 10.291182603515043,
      "grad_norm": 9.75,
      "learning_rate": 2.8085561255741704e-05,
      "loss": 0.8806,
      "num_input_tokens_seen": 40079120,
      "step": 69095
    },
    {
      "epoch": 10.291927316056002,
      "grad_norm": 6.1875,
      "learning_rate": 2.8082336652592135e-05,
      "loss": 0.6805,
      "num_input_tokens_seen": 40082256,
      "step": 69100
    },
    {
      "epoch": 10.29267202859696,
      "grad_norm": 6.125,
      "learning_rate": 2.8079111997369624e-05,
      "loss": 1.2539,
      "num_input_tokens_seen": 40084848,
      "step": 69105
    },
    {
      "epoch": 10.293416741137921,
      "grad_norm": 5.65625,
      "learning_rate": 2.807588729012864e-05,
      "loss": 0.8483,
      "num_input_tokens_seen": 40087792,
      "step": 69110
    },
    {
      "epoch": 10.29416145367888,
      "grad_norm": 7.375,
      "learning_rate": 2.8072662530923666e-05,
      "loss": 1.1015,
      "num_input_tokens_seen": 40090832,
      "step": 69115
    },
    {
      "epoch": 10.294906166219839,
      "grad_norm": 6.09375,
      "learning_rate": 2.8069437719809182e-05,
      "loss": 0.9796,
      "num_input_tokens_seen": 40093872,
      "step": 69120
    },
    {
      "epoch": 10.295650878760798,
      "grad_norm": 11.1875,
      "learning_rate": 2.806621285683967e-05,
      "loss": 1.2453,
      "num_input_tokens_seen": 40096624,
      "step": 69125
    },
    {
      "epoch": 10.296395591301758,
      "grad_norm": 19.625,
      "learning_rate": 2.8062987942069603e-05,
      "loss": 1.1216,
      "num_input_tokens_seen": 40099728,
      "step": 69130
    },
    {
      "epoch": 10.297140303842717,
      "grad_norm": 24.25,
      "learning_rate": 2.8059762975553478e-05,
      "loss": 1.2714,
      "num_input_tokens_seen": 40102832,
      "step": 69135
    },
    {
      "epoch": 10.297885016383676,
      "grad_norm": 16.125,
      "learning_rate": 2.8056537957345757e-05,
      "loss": 1.1136,
      "num_input_tokens_seen": 40105616,
      "step": 69140
    },
    {
      "epoch": 10.298629728924634,
      "grad_norm": 6.75,
      "learning_rate": 2.8053312887500936e-05,
      "loss": 0.9869,
      "num_input_tokens_seen": 40108560,
      "step": 69145
    },
    {
      "epoch": 10.299374441465595,
      "grad_norm": 11.75,
      "learning_rate": 2.8050087766073496e-05,
      "loss": 1.2586,
      "num_input_tokens_seen": 40111440,
      "step": 69150
    },
    {
      "epoch": 10.300119154006554,
      "grad_norm": 11.125,
      "learning_rate": 2.804686259311792e-05,
      "loss": 1.2821,
      "num_input_tokens_seen": 40114480,
      "step": 69155
    },
    {
      "epoch": 10.300863866547513,
      "grad_norm": 8.5,
      "learning_rate": 2.8043637368688707e-05,
      "loss": 0.9831,
      "num_input_tokens_seen": 40117040,
      "step": 69160
    },
    {
      "epoch": 10.301608579088471,
      "grad_norm": 14.375,
      "learning_rate": 2.804041209284033e-05,
      "loss": 0.9018,
      "num_input_tokens_seen": 40120016,
      "step": 69165
    },
    {
      "epoch": 10.302353291629432,
      "grad_norm": 8.5,
      "learning_rate": 2.803718676562729e-05,
      "loss": 0.831,
      "num_input_tokens_seen": 40122864,
      "step": 69170
    },
    {
      "epoch": 10.30309800417039,
      "grad_norm": 8.8125,
      "learning_rate": 2.803396138710405e-05,
      "loss": 1.0703,
      "num_input_tokens_seen": 40125744,
      "step": 69175
    },
    {
      "epoch": 10.30384271671135,
      "grad_norm": 15.0625,
      "learning_rate": 2.8030735957325122e-05,
      "loss": 1.0511,
      "num_input_tokens_seen": 40128720,
      "step": 69180
    },
    {
      "epoch": 10.304587429252308,
      "grad_norm": 11.0625,
      "learning_rate": 2.8027510476344986e-05,
      "loss": 1.0351,
      "num_input_tokens_seen": 40131760,
      "step": 69185
    },
    {
      "epoch": 10.305332141793269,
      "grad_norm": 12.0,
      "learning_rate": 2.8024284944218145e-05,
      "loss": 1.207,
      "num_input_tokens_seen": 40134480,
      "step": 69190
    },
    {
      "epoch": 10.306076854334227,
      "grad_norm": 10.3125,
      "learning_rate": 2.802105936099908e-05,
      "loss": 0.9649,
      "num_input_tokens_seen": 40137488,
      "step": 69195
    },
    {
      "epoch": 10.306821566875186,
      "grad_norm": 5.28125,
      "learning_rate": 2.8017833726742293e-05,
      "loss": 1.2075,
      "num_input_tokens_seen": 40140496,
      "step": 69200
    },
    {
      "epoch": 10.307566279416145,
      "grad_norm": 11.0,
      "learning_rate": 2.8014608041502273e-05,
      "loss": 1.8301,
      "num_input_tokens_seen": 40143600,
      "step": 69205
    },
    {
      "epoch": 10.308310991957104,
      "grad_norm": 10.75,
      "learning_rate": 2.8011382305333505e-05,
      "loss": 1.5473,
      "num_input_tokens_seen": 40146800,
      "step": 69210
    },
    {
      "epoch": 10.309055704498064,
      "grad_norm": 18.375,
      "learning_rate": 2.8008156518290496e-05,
      "loss": 1.1614,
      "num_input_tokens_seen": 40149680,
      "step": 69215
    },
    {
      "epoch": 10.309800417039023,
      "grad_norm": 5.4375,
      "learning_rate": 2.8004930680427742e-05,
      "loss": 0.7576,
      "num_input_tokens_seen": 40152240,
      "step": 69220
    },
    {
      "epoch": 10.310545129579982,
      "grad_norm": 10.375,
      "learning_rate": 2.8001704791799732e-05,
      "loss": 0.9604,
      "num_input_tokens_seen": 40155280,
      "step": 69225
    },
    {
      "epoch": 10.31128984212094,
      "grad_norm": 11.1875,
      "learning_rate": 2.799847885246098e-05,
      "loss": 1.2956,
      "num_input_tokens_seen": 40158064,
      "step": 69230
    },
    {
      "epoch": 10.312034554661901,
      "grad_norm": 10.1875,
      "learning_rate": 2.799525286246597e-05,
      "loss": 1.2031,
      "num_input_tokens_seen": 40161040,
      "step": 69235
    },
    {
      "epoch": 10.31277926720286,
      "grad_norm": 11.75,
      "learning_rate": 2.7992026821869215e-05,
      "loss": 1.2273,
      "num_input_tokens_seen": 40163856,
      "step": 69240
    },
    {
      "epoch": 10.313523979743819,
      "grad_norm": 9.3125,
      "learning_rate": 2.7988800730725202e-05,
      "loss": 1.2899,
      "num_input_tokens_seen": 40167216,
      "step": 69245
    },
    {
      "epoch": 10.314268692284777,
      "grad_norm": 9.0625,
      "learning_rate": 2.7985574589088437e-05,
      "loss": 0.8823,
      "num_input_tokens_seen": 40170544,
      "step": 69250
    },
    {
      "epoch": 10.315013404825738,
      "grad_norm": 19.875,
      "learning_rate": 2.798234839701342e-05,
      "loss": 1.3144,
      "num_input_tokens_seen": 40173264,
      "step": 69255
    },
    {
      "epoch": 10.315758117366697,
      "grad_norm": 5.71875,
      "learning_rate": 2.797912215455466e-05,
      "loss": 1.0239,
      "num_input_tokens_seen": 40176368,
      "step": 69260
    },
    {
      "epoch": 10.316502829907655,
      "grad_norm": 14.5,
      "learning_rate": 2.797589586176666e-05,
      "loss": 1.0982,
      "num_input_tokens_seen": 40179376,
      "step": 69265
    },
    {
      "epoch": 10.317247542448614,
      "grad_norm": 15.1875,
      "learning_rate": 2.797266951870393e-05,
      "loss": 1.3216,
      "num_input_tokens_seen": 40181936,
      "step": 69270
    },
    {
      "epoch": 10.317992254989575,
      "grad_norm": 15.0625,
      "learning_rate": 2.7969443125420963e-05,
      "loss": 1.2752,
      "num_input_tokens_seen": 40184752,
      "step": 69275
    },
    {
      "epoch": 10.318736967530533,
      "grad_norm": 11.375,
      "learning_rate": 2.7966216681972278e-05,
      "loss": 0.9548,
      "num_input_tokens_seen": 40187536,
      "step": 69280
    },
    {
      "epoch": 10.319481680071492,
      "grad_norm": 7.125,
      "learning_rate": 2.7962990188412375e-05,
      "loss": 1.0265,
      "num_input_tokens_seen": 40190384,
      "step": 69285
    },
    {
      "epoch": 10.320226392612451,
      "grad_norm": 8.4375,
      "learning_rate": 2.7959763644795762e-05,
      "loss": 1.0783,
      "num_input_tokens_seen": 40193904,
      "step": 69290
    },
    {
      "epoch": 10.320971105153411,
      "grad_norm": 13.4375,
      "learning_rate": 2.7956537051176952e-05,
      "loss": 1.4952,
      "num_input_tokens_seen": 40196944,
      "step": 69295
    },
    {
      "epoch": 10.32171581769437,
      "grad_norm": 12.8125,
      "learning_rate": 2.7953310407610455e-05,
      "loss": 1.0119,
      "num_input_tokens_seen": 40199728,
      "step": 69300
    },
    {
      "epoch": 10.322460530235329,
      "grad_norm": 6.1875,
      "learning_rate": 2.7950083714150776e-05,
      "loss": 1.226,
      "num_input_tokens_seen": 40202352,
      "step": 69305
    },
    {
      "epoch": 10.323205242776288,
      "grad_norm": 9.875,
      "learning_rate": 2.794685697085243e-05,
      "loss": 0.8613,
      "num_input_tokens_seen": 40205072,
      "step": 69310
    },
    {
      "epoch": 10.323949955317248,
      "grad_norm": 7.78125,
      "learning_rate": 2.7943630177769932e-05,
      "loss": 1.1103,
      "num_input_tokens_seen": 40208016,
      "step": 69315
    },
    {
      "epoch": 10.324694667858207,
      "grad_norm": 9.375,
      "learning_rate": 2.79404033349578e-05,
      "loss": 1.0676,
      "num_input_tokens_seen": 40210704,
      "step": 69320
    },
    {
      "epoch": 10.325439380399166,
      "grad_norm": 7.0625,
      "learning_rate": 2.7937176442470535e-05,
      "loss": 0.6982,
      "num_input_tokens_seen": 40213232,
      "step": 69325
    },
    {
      "epoch": 10.326184092940125,
      "grad_norm": 4.65625,
      "learning_rate": 2.793394950036266e-05,
      "loss": 1.0293,
      "num_input_tokens_seen": 40216144,
      "step": 69330
    },
    {
      "epoch": 10.326928805481085,
      "grad_norm": 6.90625,
      "learning_rate": 2.7930722508688696e-05,
      "loss": 1.0839,
      "num_input_tokens_seen": 40218672,
      "step": 69335
    },
    {
      "epoch": 10.327673518022044,
      "grad_norm": 8.4375,
      "learning_rate": 2.792749546750315e-05,
      "loss": 1.3794,
      "num_input_tokens_seen": 40221424,
      "step": 69340
    },
    {
      "epoch": 10.328418230563003,
      "grad_norm": 7.65625,
      "learning_rate": 2.792426837686054e-05,
      "loss": 1.3219,
      "num_input_tokens_seen": 40224144,
      "step": 69345
    },
    {
      "epoch": 10.329162943103961,
      "grad_norm": 12.4375,
      "learning_rate": 2.7921041236815387e-05,
      "loss": 1.1615,
      "num_input_tokens_seen": 40226864,
      "step": 69350
    },
    {
      "epoch": 10.329907655644922,
      "grad_norm": 8.0625,
      "learning_rate": 2.7917814047422214e-05,
      "loss": 1.3525,
      "num_input_tokens_seen": 40229808,
      "step": 69355
    },
    {
      "epoch": 10.33065236818588,
      "grad_norm": 5.5,
      "learning_rate": 2.7914586808735542e-05,
      "loss": 1.0288,
      "num_input_tokens_seen": 40232432,
      "step": 69360
    },
    {
      "epoch": 10.33139708072684,
      "grad_norm": 19.125,
      "learning_rate": 2.7911359520809886e-05,
      "loss": 1.1553,
      "num_input_tokens_seen": 40235216,
      "step": 69365
    },
    {
      "epoch": 10.332141793267798,
      "grad_norm": 10.75,
      "learning_rate": 2.7908132183699775e-05,
      "loss": 0.8152,
      "num_input_tokens_seen": 40238448,
      "step": 69370
    },
    {
      "epoch": 10.332886505808759,
      "grad_norm": 18.375,
      "learning_rate": 2.790490479745972e-05,
      "loss": 1.113,
      "num_input_tokens_seen": 40241456,
      "step": 69375
    },
    {
      "epoch": 10.333631218349717,
      "grad_norm": 7.8125,
      "learning_rate": 2.7901677362144252e-05,
      "loss": 0.9426,
      "num_input_tokens_seen": 40244272,
      "step": 69380
    },
    {
      "epoch": 10.334375930890676,
      "grad_norm": 8.25,
      "learning_rate": 2.7898449877807885e-05,
      "loss": 1.2657,
      "num_input_tokens_seen": 40247664,
      "step": 69385
    },
    {
      "epoch": 10.335120643431635,
      "grad_norm": 16.5,
      "learning_rate": 2.7895222344505163e-05,
      "loss": 1.32,
      "num_input_tokens_seen": 40250352,
      "step": 69390
    },
    {
      "epoch": 10.335865355972594,
      "grad_norm": 11.3125,
      "learning_rate": 2.78919947622906e-05,
      "loss": 1.0354,
      "num_input_tokens_seen": 40253456,
      "step": 69395
    },
    {
      "epoch": 10.336610068513554,
      "grad_norm": 10.75,
      "learning_rate": 2.788876713121873e-05,
      "loss": 1.1889,
      "num_input_tokens_seen": 40256144,
      "step": 69400
    },
    {
      "epoch": 10.337354781054513,
      "grad_norm": 15.125,
      "learning_rate": 2.7885539451344077e-05,
      "loss": 1.1471,
      "num_input_tokens_seen": 40259152,
      "step": 69405
    },
    {
      "epoch": 10.338099493595472,
      "grad_norm": 18.875,
      "learning_rate": 2.788231172272116e-05,
      "loss": 1.2384,
      "num_input_tokens_seen": 40262032,
      "step": 69410
    },
    {
      "epoch": 10.33884420613643,
      "grad_norm": 9.1875,
      "learning_rate": 2.7879083945404517e-05,
      "loss": 0.9411,
      "num_input_tokens_seen": 40264560,
      "step": 69415
    },
    {
      "epoch": 10.339588918677391,
      "grad_norm": 5.84375,
      "learning_rate": 2.7875856119448672e-05,
      "loss": 1.0764,
      "num_input_tokens_seen": 40267376,
      "step": 69420
    },
    {
      "epoch": 10.34033363121835,
      "grad_norm": 10.5625,
      "learning_rate": 2.7872628244908167e-05,
      "loss": 1.0892,
      "num_input_tokens_seen": 40270384,
      "step": 69425
    },
    {
      "epoch": 10.341078343759309,
      "grad_norm": 6.21875,
      "learning_rate": 2.7869400321837525e-05,
      "loss": 0.8023,
      "num_input_tokens_seen": 40273072,
      "step": 69430
    },
    {
      "epoch": 10.341823056300267,
      "grad_norm": 11.4375,
      "learning_rate": 2.7866172350291286e-05,
      "loss": 1.2227,
      "num_input_tokens_seen": 40276112,
      "step": 69435
    },
    {
      "epoch": 10.342567768841228,
      "grad_norm": 9.8125,
      "learning_rate": 2.7862944330323982e-05,
      "loss": 0.8961,
      "num_input_tokens_seen": 40278832,
      "step": 69440
    },
    {
      "epoch": 10.343312481382187,
      "grad_norm": 6.40625,
      "learning_rate": 2.785971626199013e-05,
      "loss": 0.9678,
      "num_input_tokens_seen": 40282032,
      "step": 69445
    },
    {
      "epoch": 10.344057193923145,
      "grad_norm": 9.5,
      "learning_rate": 2.7856488145344285e-05,
      "loss": 1.1825,
      "num_input_tokens_seen": 40284816,
      "step": 69450
    },
    {
      "epoch": 10.344801906464104,
      "grad_norm": 15.3125,
      "learning_rate": 2.785325998044097e-05,
      "loss": 1.1858,
      "num_input_tokens_seen": 40287728,
      "step": 69455
    },
    {
      "epoch": 10.345546619005065,
      "grad_norm": 8.75,
      "learning_rate": 2.7850031767334734e-05,
      "loss": 1.0605,
      "num_input_tokens_seen": 40290576,
      "step": 69460
    },
    {
      "epoch": 10.346291331546023,
      "grad_norm": 5.9375,
      "learning_rate": 2.784680350608011e-05,
      "loss": 0.9356,
      "num_input_tokens_seen": 40293616,
      "step": 69465
    },
    {
      "epoch": 10.347036044086982,
      "grad_norm": 15.125,
      "learning_rate": 2.784357519673163e-05,
      "loss": 1.1754,
      "num_input_tokens_seen": 40296304,
      "step": 69470
    },
    {
      "epoch": 10.347780756627941,
      "grad_norm": 10.75,
      "learning_rate": 2.784034683934384e-05,
      "loss": 1.1512,
      "num_input_tokens_seen": 40299504,
      "step": 69475
    },
    {
      "epoch": 10.348525469168901,
      "grad_norm": 4.3125,
      "learning_rate": 2.7837118433971277e-05,
      "loss": 0.8711,
      "num_input_tokens_seen": 40302064,
      "step": 69480
    },
    {
      "epoch": 10.34927018170986,
      "grad_norm": 3.9375,
      "learning_rate": 2.7833889980668476e-05,
      "loss": 0.6957,
      "num_input_tokens_seen": 40305200,
      "step": 69485
    },
    {
      "epoch": 10.350014894250819,
      "grad_norm": 7.78125,
      "learning_rate": 2.7830661479489987e-05,
      "loss": 0.7893,
      "num_input_tokens_seen": 40308112,
      "step": 69490
    },
    {
      "epoch": 10.350759606791778,
      "grad_norm": 7.59375,
      "learning_rate": 2.782743293049035e-05,
      "loss": 0.8744,
      "num_input_tokens_seen": 40311312,
      "step": 69495
    },
    {
      "epoch": 10.351504319332738,
      "grad_norm": 15.0,
      "learning_rate": 2.7824204333724115e-05,
      "loss": 1.0761,
      "num_input_tokens_seen": 40314320,
      "step": 69500
    },
    {
      "epoch": 10.352249031873697,
      "grad_norm": 11.375,
      "learning_rate": 2.7820975689245805e-05,
      "loss": 0.8759,
      "num_input_tokens_seen": 40317392,
      "step": 69505
    },
    {
      "epoch": 10.352993744414656,
      "grad_norm": 5.84375,
      "learning_rate": 2.7817746997109983e-05,
      "loss": 0.6619,
      "num_input_tokens_seen": 40320048,
      "step": 69510
    },
    {
      "epoch": 10.353738456955615,
      "grad_norm": 9.75,
      "learning_rate": 2.7814518257371187e-05,
      "loss": 0.7921,
      "num_input_tokens_seen": 40323120,
      "step": 69515
    },
    {
      "epoch": 10.354483169496575,
      "grad_norm": 10.0,
      "learning_rate": 2.7811289470083972e-05,
      "loss": 1.1762,
      "num_input_tokens_seen": 40325808,
      "step": 69520
    },
    {
      "epoch": 10.355227882037534,
      "grad_norm": 9.5,
      "learning_rate": 2.7808060635302875e-05,
      "loss": 0.9309,
      "num_input_tokens_seen": 40328816,
      "step": 69525
    },
    {
      "epoch": 10.355972594578493,
      "grad_norm": 7.75,
      "learning_rate": 2.7804831753082445e-05,
      "loss": 1.1832,
      "num_input_tokens_seen": 40331632,
      "step": 69530
    },
    {
      "epoch": 10.356717307119451,
      "grad_norm": 8.0,
      "learning_rate": 2.7801602823477236e-05,
      "loss": 1.1175,
      "num_input_tokens_seen": 40334576,
      "step": 69535
    },
    {
      "epoch": 10.357462019660412,
      "grad_norm": 8.125,
      "learning_rate": 2.779837384654179e-05,
      "loss": 1.0331,
      "num_input_tokens_seen": 40337456,
      "step": 69540
    },
    {
      "epoch": 10.35820673220137,
      "grad_norm": 7.625,
      "learning_rate": 2.7795144822330673e-05,
      "loss": 0.908,
      "num_input_tokens_seen": 40340080,
      "step": 69545
    },
    {
      "epoch": 10.35895144474233,
      "grad_norm": 13.4375,
      "learning_rate": 2.7791915750898413e-05,
      "loss": 1.0019,
      "num_input_tokens_seen": 40342928,
      "step": 69550
    },
    {
      "epoch": 10.359696157283288,
      "grad_norm": 9.0,
      "learning_rate": 2.7788686632299577e-05,
      "loss": 1.1461,
      "num_input_tokens_seen": 40345648,
      "step": 69555
    },
    {
      "epoch": 10.360440869824249,
      "grad_norm": 8.3125,
      "learning_rate": 2.778545746658872e-05,
      "loss": 1.1322,
      "num_input_tokens_seen": 40348400,
      "step": 69560
    },
    {
      "epoch": 10.361185582365207,
      "grad_norm": 6.03125,
      "learning_rate": 2.7782228253820385e-05,
      "loss": 0.9597,
      "num_input_tokens_seen": 40351408,
      "step": 69565
    },
    {
      "epoch": 10.361930294906166,
      "grad_norm": 16.625,
      "learning_rate": 2.777899899404914e-05,
      "loss": 1.1963,
      "num_input_tokens_seen": 40354288,
      "step": 69570
    },
    {
      "epoch": 10.362675007447125,
      "grad_norm": 11.625,
      "learning_rate": 2.777576968732952e-05,
      "loss": 1.2716,
      "num_input_tokens_seen": 40357200,
      "step": 69575
    },
    {
      "epoch": 10.363419719988084,
      "grad_norm": 9.0625,
      "learning_rate": 2.7772540333716102e-05,
      "loss": 1.147,
      "num_input_tokens_seen": 40360176,
      "step": 69580
    },
    {
      "epoch": 10.364164432529044,
      "grad_norm": 7.875,
      "learning_rate": 2.7769310933263425e-05,
      "loss": 0.9265,
      "num_input_tokens_seen": 40363120,
      "step": 69585
    },
    {
      "epoch": 10.364909145070003,
      "grad_norm": 6.34375,
      "learning_rate": 2.776608148602605e-05,
      "loss": 0.8765,
      "num_input_tokens_seen": 40366128,
      "step": 69590
    },
    {
      "epoch": 10.365653857610962,
      "grad_norm": 8.5625,
      "learning_rate": 2.7762851992058548e-05,
      "loss": 0.9032,
      "num_input_tokens_seen": 40368816,
      "step": 69595
    },
    {
      "epoch": 10.36639857015192,
      "grad_norm": 10.9375,
      "learning_rate": 2.7759622451415473e-05,
      "loss": 1.1035,
      "num_input_tokens_seen": 40371536,
      "step": 69600
    },
    {
      "epoch": 10.367143282692881,
      "grad_norm": 14.4375,
      "learning_rate": 2.775639286415138e-05,
      "loss": 1.3737,
      "num_input_tokens_seen": 40374288,
      "step": 69605
    },
    {
      "epoch": 10.36788799523384,
      "grad_norm": 14.0625,
      "learning_rate": 2.7753163230320828e-05,
      "loss": 1.3618,
      "num_input_tokens_seen": 40377392,
      "step": 69610
    },
    {
      "epoch": 10.368632707774799,
      "grad_norm": 9.8125,
      "learning_rate": 2.774993354997838e-05,
      "loss": 1.1008,
      "num_input_tokens_seen": 40380304,
      "step": 69615
    },
    {
      "epoch": 10.369377420315757,
      "grad_norm": 9.9375,
      "learning_rate": 2.77467038231786e-05,
      "loss": 1.1711,
      "num_input_tokens_seen": 40383184,
      "step": 69620
    },
    {
      "epoch": 10.370122132856718,
      "grad_norm": 17.75,
      "learning_rate": 2.7743474049976054e-05,
      "loss": 1.1469,
      "num_input_tokens_seen": 40386096,
      "step": 69625
    },
    {
      "epoch": 10.370866845397677,
      "grad_norm": 11.25,
      "learning_rate": 2.77402442304253e-05,
      "loss": 0.9765,
      "num_input_tokens_seen": 40388720,
      "step": 69630
    },
    {
      "epoch": 10.371611557938635,
      "grad_norm": 7.84375,
      "learning_rate": 2.7737014364580904e-05,
      "loss": 0.9065,
      "num_input_tokens_seen": 40391536,
      "step": 69635
    },
    {
      "epoch": 10.372356270479594,
      "grad_norm": 9.5625,
      "learning_rate": 2.7733784452497436e-05,
      "loss": 0.832,
      "num_input_tokens_seen": 40394128,
      "step": 69640
    },
    {
      "epoch": 10.373100983020555,
      "grad_norm": 9.8125,
      "learning_rate": 2.7730554494229453e-05,
      "loss": 0.9346,
      "num_input_tokens_seen": 40397328,
      "step": 69645
    },
    {
      "epoch": 10.373845695561513,
      "grad_norm": 9.6875,
      "learning_rate": 2.772732448983153e-05,
      "loss": 1.1966,
      "num_input_tokens_seen": 40400208,
      "step": 69650
    },
    {
      "epoch": 10.374590408102472,
      "grad_norm": 7.125,
      "learning_rate": 2.7724094439358227e-05,
      "loss": 1.1094,
      "num_input_tokens_seen": 40402800,
      "step": 69655
    },
    {
      "epoch": 10.375335120643431,
      "grad_norm": 5.34375,
      "learning_rate": 2.7720864342864123e-05,
      "loss": 1.1677,
      "num_input_tokens_seen": 40405712,
      "step": 69660
    },
    {
      "epoch": 10.376079833184392,
      "grad_norm": 7.0625,
      "learning_rate": 2.771763420040378e-05,
      "loss": 1.0908,
      "num_input_tokens_seen": 40408720,
      "step": 69665
    },
    {
      "epoch": 10.37682454572535,
      "grad_norm": 8.5625,
      "learning_rate": 2.771440401203177e-05,
      "loss": 1.1772,
      "num_input_tokens_seen": 40411696,
      "step": 69670
    },
    {
      "epoch": 10.377569258266309,
      "grad_norm": 8.6875,
      "learning_rate": 2.7711173777802657e-05,
      "loss": 1.4278,
      "num_input_tokens_seen": 40414640,
      "step": 69675
    },
    {
      "epoch": 10.378313970807268,
      "grad_norm": 6.34375,
      "learning_rate": 2.770794349777102e-05,
      "loss": 1.1418,
      "num_input_tokens_seen": 40417680,
      "step": 69680
    },
    {
      "epoch": 10.379058683348228,
      "grad_norm": 7.46875,
      "learning_rate": 2.770471317199144e-05,
      "loss": 1.2465,
      "num_input_tokens_seen": 40420240,
      "step": 69685
    },
    {
      "epoch": 10.379803395889187,
      "grad_norm": 6.90625,
      "learning_rate": 2.7701482800518475e-05,
      "loss": 1.05,
      "num_input_tokens_seen": 40422960,
      "step": 69690
    },
    {
      "epoch": 10.380548108430146,
      "grad_norm": 4.28125,
      "learning_rate": 2.7698252383406696e-05,
      "loss": 1.0602,
      "num_input_tokens_seen": 40425616,
      "step": 69695
    },
    {
      "epoch": 10.381292820971105,
      "grad_norm": 12.4375,
      "learning_rate": 2.7695021920710694e-05,
      "loss": 1.2245,
      "num_input_tokens_seen": 40428528,
      "step": 69700
    },
    {
      "epoch": 10.382037533512065,
      "grad_norm": 12.5625,
      "learning_rate": 2.7691791412485035e-05,
      "loss": 1.0571,
      "num_input_tokens_seen": 40431280,
      "step": 69705
    },
    {
      "epoch": 10.382782246053024,
      "grad_norm": 16.875,
      "learning_rate": 2.768856085878429e-05,
      "loss": 1.1661,
      "num_input_tokens_seen": 40434480,
      "step": 69710
    },
    {
      "epoch": 10.383526958593983,
      "grad_norm": 12.0,
      "learning_rate": 2.7685330259663045e-05,
      "loss": 1.3326,
      "num_input_tokens_seen": 40437328,
      "step": 69715
    },
    {
      "epoch": 10.384271671134941,
      "grad_norm": 17.25,
      "learning_rate": 2.768209961517587e-05,
      "loss": 1.1267,
      "num_input_tokens_seen": 40440400,
      "step": 69720
    },
    {
      "epoch": 10.3850163836759,
      "grad_norm": 8.6875,
      "learning_rate": 2.7678868925377362e-05,
      "loss": 1.3664,
      "num_input_tokens_seen": 40443344,
      "step": 69725
    },
    {
      "epoch": 10.38576109621686,
      "grad_norm": 13.625,
      "learning_rate": 2.7675638190322073e-05,
      "loss": 0.8966,
      "num_input_tokens_seen": 40446832,
      "step": 69730
    },
    {
      "epoch": 10.38650580875782,
      "grad_norm": 11.4375,
      "learning_rate": 2.7672407410064603e-05,
      "loss": 1.2006,
      "num_input_tokens_seen": 40449968,
      "step": 69735
    },
    {
      "epoch": 10.387250521298778,
      "grad_norm": 8.0,
      "learning_rate": 2.7669176584659522e-05,
      "loss": 1.0673,
      "num_input_tokens_seen": 40452752,
      "step": 69740
    },
    {
      "epoch": 10.387995233839739,
      "grad_norm": 11.5,
      "learning_rate": 2.766594571416141e-05,
      "loss": 1.4347,
      "num_input_tokens_seen": 40455536,
      "step": 69745
    },
    {
      "epoch": 10.388739946380698,
      "grad_norm": 11.6875,
      "learning_rate": 2.7662714798624865e-05,
      "loss": 1.6637,
      "num_input_tokens_seen": 40458160,
      "step": 69750
    },
    {
      "epoch": 10.389484658921656,
      "grad_norm": 8.3125,
      "learning_rate": 2.7659483838104456e-05,
      "loss": 0.9213,
      "num_input_tokens_seen": 40461328,
      "step": 69755
    },
    {
      "epoch": 10.390229371462615,
      "grad_norm": 11.75,
      "learning_rate": 2.7656252832654766e-05,
      "loss": 1.4562,
      "num_input_tokens_seen": 40464080,
      "step": 69760
    },
    {
      "epoch": 10.390974084003574,
      "grad_norm": 7.25,
      "learning_rate": 2.765302178233039e-05,
      "loss": 1.1212,
      "num_input_tokens_seen": 40467024,
      "step": 69765
    },
    {
      "epoch": 10.391718796544534,
      "grad_norm": 7.96875,
      "learning_rate": 2.764979068718591e-05,
      "loss": 0.8772,
      "num_input_tokens_seen": 40469904,
      "step": 69770
    },
    {
      "epoch": 10.392463509085493,
      "grad_norm": 10.3125,
      "learning_rate": 2.7646559547275907e-05,
      "loss": 1.0999,
      "num_input_tokens_seen": 40472912,
      "step": 69775
    },
    {
      "epoch": 10.393208221626452,
      "grad_norm": 10.9375,
      "learning_rate": 2.7643328362654968e-05,
      "loss": 1.0574,
      "num_input_tokens_seen": 40475856,
      "step": 69780
    },
    {
      "epoch": 10.39395293416741,
      "grad_norm": 17.5,
      "learning_rate": 2.764009713337768e-05,
      "loss": 1.8011,
      "num_input_tokens_seen": 40478896,
      "step": 69785
    },
    {
      "epoch": 10.394697646708371,
      "grad_norm": 14.125,
      "learning_rate": 2.763686585949864e-05,
      "loss": 1.0767,
      "num_input_tokens_seen": 40481936,
      "step": 69790
    },
    {
      "epoch": 10.39544235924933,
      "grad_norm": 6.1875,
      "learning_rate": 2.7633634541072428e-05,
      "loss": 1.4019,
      "num_input_tokens_seen": 40485104,
      "step": 69795
    },
    {
      "epoch": 10.396187071790289,
      "grad_norm": 9.625,
      "learning_rate": 2.763040317815364e-05,
      "loss": 0.9679,
      "num_input_tokens_seen": 40487664,
      "step": 69800
    },
    {
      "epoch": 10.396931784331247,
      "grad_norm": 25.25,
      "learning_rate": 2.7627171770796868e-05,
      "loss": 1.6681,
      "num_input_tokens_seen": 40490832,
      "step": 69805
    },
    {
      "epoch": 10.397676496872208,
      "grad_norm": 8.625,
      "learning_rate": 2.762394031905669e-05,
      "loss": 0.9335,
      "num_input_tokens_seen": 40493552,
      "step": 69810
    },
    {
      "epoch": 10.398421209413167,
      "grad_norm": 4.0625,
      "learning_rate": 2.7620708822987708e-05,
      "loss": 0.8321,
      "num_input_tokens_seen": 40496240,
      "step": 69815
    },
    {
      "epoch": 10.399165921954125,
      "grad_norm": 10.6875,
      "learning_rate": 2.7617477282644515e-05,
      "loss": 1.049,
      "num_input_tokens_seen": 40499184,
      "step": 69820
    },
    {
      "epoch": 10.399910634495084,
      "grad_norm": 11.75,
      "learning_rate": 2.7614245698081702e-05,
      "loss": 0.9004,
      "num_input_tokens_seen": 40502032,
      "step": 69825
    },
    {
      "epoch": 10.400655347036045,
      "grad_norm": 12.3125,
      "learning_rate": 2.7611014069353874e-05,
      "loss": 0.9828,
      "num_input_tokens_seen": 40504880,
      "step": 69830
    },
    {
      "epoch": 10.401400059577004,
      "grad_norm": 14.9375,
      "learning_rate": 2.760778239651561e-05,
      "loss": 1.1511,
      "num_input_tokens_seen": 40507568,
      "step": 69835
    },
    {
      "epoch": 10.402144772117962,
      "grad_norm": 11.25,
      "learning_rate": 2.760455067962151e-05,
      "loss": 0.9157,
      "num_input_tokens_seen": 40510448,
      "step": 69840
    },
    {
      "epoch": 10.402889484658921,
      "grad_norm": 11.1875,
      "learning_rate": 2.7601318918726184e-05,
      "loss": 1.241,
      "num_input_tokens_seen": 40513488,
      "step": 69845
    },
    {
      "epoch": 10.403634197199882,
      "grad_norm": 7.71875,
      "learning_rate": 2.7598087113884207e-05,
      "loss": 0.9984,
      "num_input_tokens_seen": 40516304,
      "step": 69850
    },
    {
      "epoch": 10.40437890974084,
      "grad_norm": 9.5,
      "learning_rate": 2.7594855265150192e-05,
      "loss": 0.9475,
      "num_input_tokens_seen": 40518960,
      "step": 69855
    },
    {
      "epoch": 10.405123622281799,
      "grad_norm": 19.625,
      "learning_rate": 2.7591623372578736e-05,
      "loss": 1.1999,
      "num_input_tokens_seen": 40521872,
      "step": 69860
    },
    {
      "epoch": 10.405868334822758,
      "grad_norm": 8.625,
      "learning_rate": 2.758839143622444e-05,
      "loss": 1.4476,
      "num_input_tokens_seen": 40524912,
      "step": 69865
    },
    {
      "epoch": 10.406613047363718,
      "grad_norm": 18.125,
      "learning_rate": 2.7585159456141895e-05,
      "loss": 1.05,
      "num_input_tokens_seen": 40528048,
      "step": 69870
    },
    {
      "epoch": 10.407357759904677,
      "grad_norm": 11.75,
      "learning_rate": 2.7581927432385713e-05,
      "loss": 1.1367,
      "num_input_tokens_seen": 40531440,
      "step": 69875
    },
    {
      "epoch": 10.408102472445636,
      "grad_norm": 13.1875,
      "learning_rate": 2.757869536501049e-05,
      "loss": 0.8282,
      "num_input_tokens_seen": 40534224,
      "step": 69880
    },
    {
      "epoch": 10.408847184986595,
      "grad_norm": 9.875,
      "learning_rate": 2.7575463254070832e-05,
      "loss": 0.8551,
      "num_input_tokens_seen": 40537360,
      "step": 69885
    },
    {
      "epoch": 10.409591897527555,
      "grad_norm": 11.75,
      "learning_rate": 2.757223109962134e-05,
      "loss": 0.9859,
      "num_input_tokens_seen": 40540080,
      "step": 69890
    },
    {
      "epoch": 10.410336610068514,
      "grad_norm": 13.125,
      "learning_rate": 2.7568998901716613e-05,
      "loss": 1.1697,
      "num_input_tokens_seen": 40543120,
      "step": 69895
    },
    {
      "epoch": 10.411081322609473,
      "grad_norm": 7.28125,
      "learning_rate": 2.7565766660411263e-05,
      "loss": 1.0151,
      "num_input_tokens_seen": 40546000,
      "step": 69900
    },
    {
      "epoch": 10.411826035150431,
      "grad_norm": 5.03125,
      "learning_rate": 2.7562534375759898e-05,
      "loss": 1.0142,
      "num_input_tokens_seen": 40549008,
      "step": 69905
    },
    {
      "epoch": 10.41257074769139,
      "grad_norm": 9.9375,
      "learning_rate": 2.7559302047817108e-05,
      "loss": 0.8872,
      "num_input_tokens_seen": 40551856,
      "step": 69910
    },
    {
      "epoch": 10.41331546023235,
      "grad_norm": 7.71875,
      "learning_rate": 2.755606967663752e-05,
      "loss": 0.9311,
      "num_input_tokens_seen": 40554736,
      "step": 69915
    },
    {
      "epoch": 10.41406017277331,
      "grad_norm": 17.125,
      "learning_rate": 2.755283726227573e-05,
      "loss": 0.9483,
      "num_input_tokens_seen": 40557744,
      "step": 69920
    },
    {
      "epoch": 10.414804885314268,
      "grad_norm": 18.125,
      "learning_rate": 2.754960480478635e-05,
      "loss": 1.1909,
      "num_input_tokens_seen": 40560560,
      "step": 69925
    },
    {
      "epoch": 10.415549597855227,
      "grad_norm": 8.875,
      "learning_rate": 2.7546372304223983e-05,
      "loss": 0.8214,
      "num_input_tokens_seen": 40563408,
      "step": 69930
    },
    {
      "epoch": 10.416294310396188,
      "grad_norm": 10.625,
      "learning_rate": 2.7543139760643255e-05,
      "loss": 0.7392,
      "num_input_tokens_seen": 40566320,
      "step": 69935
    },
    {
      "epoch": 10.417039022937146,
      "grad_norm": 13.75,
      "learning_rate": 2.7539907174098755e-05,
      "loss": 0.966,
      "num_input_tokens_seen": 40569456,
      "step": 69940
    },
    {
      "epoch": 10.417783735478105,
      "grad_norm": 5.90625,
      "learning_rate": 2.7536674544645108e-05,
      "loss": 0.8373,
      "num_input_tokens_seen": 40572272,
      "step": 69945
    },
    {
      "epoch": 10.418528448019064,
      "grad_norm": 12.6875,
      "learning_rate": 2.7533441872336923e-05,
      "loss": 1.3499,
      "num_input_tokens_seen": 40575280,
      "step": 69950
    },
    {
      "epoch": 10.419273160560024,
      "grad_norm": 18.375,
      "learning_rate": 2.7530209157228808e-05,
      "loss": 1.1815,
      "num_input_tokens_seen": 40578448,
      "step": 69955
    },
    {
      "epoch": 10.420017873100983,
      "grad_norm": 7.40625,
      "learning_rate": 2.752697639937539e-05,
      "loss": 1.3032,
      "num_input_tokens_seen": 40581520,
      "step": 69960
    },
    {
      "epoch": 10.420762585641942,
      "grad_norm": 15.75,
      "learning_rate": 2.752374359883127e-05,
      "loss": 1.1271,
      "num_input_tokens_seen": 40584496,
      "step": 69965
    },
    {
      "epoch": 10.4215072981829,
      "grad_norm": 12.4375,
      "learning_rate": 2.7520510755651068e-05,
      "loss": 1.1083,
      "num_input_tokens_seen": 40587312,
      "step": 69970
    },
    {
      "epoch": 10.422252010723861,
      "grad_norm": 11.25,
      "learning_rate": 2.7517277869889395e-05,
      "loss": 1.0686,
      "num_input_tokens_seen": 40590352,
      "step": 69975
    },
    {
      "epoch": 10.42299672326482,
      "grad_norm": 11.375,
      "learning_rate": 2.7514044941600874e-05,
      "loss": 1.551,
      "num_input_tokens_seen": 40593104,
      "step": 69980
    },
    {
      "epoch": 10.423741435805779,
      "grad_norm": 13.625,
      "learning_rate": 2.7510811970840115e-05,
      "loss": 0.9547,
      "num_input_tokens_seen": 40596176,
      "step": 69985
    },
    {
      "epoch": 10.424486148346737,
      "grad_norm": 12.6875,
      "learning_rate": 2.7507578957661746e-05,
      "loss": 1.0975,
      "num_input_tokens_seen": 40598864,
      "step": 69990
    },
    {
      "epoch": 10.425230860887698,
      "grad_norm": 11.0625,
      "learning_rate": 2.7504345902120375e-05,
      "loss": 0.8161,
      "num_input_tokens_seen": 40602128,
      "step": 69995
    },
    {
      "epoch": 10.425975573428657,
      "grad_norm": 8.1875,
      "learning_rate": 2.7501112804270624e-05,
      "loss": 1.0886,
      "num_input_tokens_seen": 40605328,
      "step": 70000
    },
    {
      "epoch": 10.426720285969616,
      "grad_norm": 17.375,
      "learning_rate": 2.749787966416712e-05,
      "loss": 0.6646,
      "num_input_tokens_seen": 40608272,
      "step": 70005
    },
    {
      "epoch": 10.427464998510574,
      "grad_norm": 14.25,
      "learning_rate": 2.7494646481864472e-05,
      "loss": 1.2409,
      "num_input_tokens_seen": 40611472,
      "step": 70010
    },
    {
      "epoch": 10.428209711051535,
      "grad_norm": 7.875,
      "learning_rate": 2.749141325741731e-05,
      "loss": 1.1978,
      "num_input_tokens_seen": 40614160,
      "step": 70015
    },
    {
      "epoch": 10.428954423592494,
      "grad_norm": 14.0,
      "learning_rate": 2.7488179990880248e-05,
      "loss": 1.2064,
      "num_input_tokens_seen": 40617040,
      "step": 70020
    },
    {
      "epoch": 10.429699136133452,
      "grad_norm": 15.25,
      "learning_rate": 2.7484946682307917e-05,
      "loss": 1.3883,
      "num_input_tokens_seen": 40619888,
      "step": 70025
    },
    {
      "epoch": 10.430443848674411,
      "grad_norm": 11.625,
      "learning_rate": 2.7481713331754945e-05,
      "loss": 1.0058,
      "num_input_tokens_seen": 40622768,
      "step": 70030
    },
    {
      "epoch": 10.431188561215372,
      "grad_norm": 16.25,
      "learning_rate": 2.7478479939275937e-05,
      "loss": 1.3234,
      "num_input_tokens_seen": 40625552,
      "step": 70035
    },
    {
      "epoch": 10.43193327375633,
      "grad_norm": 9.125,
      "learning_rate": 2.7475246504925535e-05,
      "loss": 0.957,
      "num_input_tokens_seen": 40628272,
      "step": 70040
    },
    {
      "epoch": 10.43267798629729,
      "grad_norm": 13.8125,
      "learning_rate": 2.7472013028758364e-05,
      "loss": 1.4311,
      "num_input_tokens_seen": 40631248,
      "step": 70045
    },
    {
      "epoch": 10.433422698838248,
      "grad_norm": 7.78125,
      "learning_rate": 2.7468779510829036e-05,
      "loss": 1.6196,
      "num_input_tokens_seen": 40634128,
      "step": 70050
    },
    {
      "epoch": 10.434167411379208,
      "grad_norm": 10.375,
      "learning_rate": 2.746554595119219e-05,
      "loss": 1.3038,
      "num_input_tokens_seen": 40637104,
      "step": 70055
    },
    {
      "epoch": 10.434912123920167,
      "grad_norm": 16.875,
      "learning_rate": 2.7462312349902452e-05,
      "loss": 1.05,
      "num_input_tokens_seen": 40640112,
      "step": 70060
    },
    {
      "epoch": 10.435656836461126,
      "grad_norm": 6.78125,
      "learning_rate": 2.7459078707014453e-05,
      "loss": 0.9161,
      "num_input_tokens_seen": 40643088,
      "step": 70065
    },
    {
      "epoch": 10.436401549002085,
      "grad_norm": 11.9375,
      "learning_rate": 2.745584502258281e-05,
      "loss": 1.0153,
      "num_input_tokens_seen": 40645840,
      "step": 70070
    },
    {
      "epoch": 10.437146261543045,
      "grad_norm": 8.625,
      "learning_rate": 2.745261129666217e-05,
      "loss": 1.051,
      "num_input_tokens_seen": 40648912,
      "step": 70075
    },
    {
      "epoch": 10.437890974084004,
      "grad_norm": 6.9375,
      "learning_rate": 2.7449377529307147e-05,
      "loss": 0.9833,
      "num_input_tokens_seen": 40651792,
      "step": 70080
    },
    {
      "epoch": 10.438635686624963,
      "grad_norm": 16.375,
      "learning_rate": 2.7446143720572387e-05,
      "loss": 1.052,
      "num_input_tokens_seen": 40654800,
      "step": 70085
    },
    {
      "epoch": 10.439380399165922,
      "grad_norm": 7.59375,
      "learning_rate": 2.7442909870512513e-05,
      "loss": 0.6811,
      "num_input_tokens_seen": 40657776,
      "step": 70090
    },
    {
      "epoch": 10.44012511170688,
      "grad_norm": 9.8125,
      "learning_rate": 2.7439675979182155e-05,
      "loss": 1.1235,
      "num_input_tokens_seen": 40660688,
      "step": 70095
    },
    {
      "epoch": 10.44086982424784,
      "grad_norm": 9.875,
      "learning_rate": 2.7436442046635962e-05,
      "loss": 1.3086,
      "num_input_tokens_seen": 40663760,
      "step": 70100
    },
    {
      "epoch": 10.4416145367888,
      "grad_norm": 9.25,
      "learning_rate": 2.7433208072928546e-05,
      "loss": 1.0384,
      "num_input_tokens_seen": 40667024,
      "step": 70105
    },
    {
      "epoch": 10.442359249329758,
      "grad_norm": 8.25,
      "learning_rate": 2.7429974058114553e-05,
      "loss": 0.7127,
      "num_input_tokens_seen": 40669840,
      "step": 70110
    },
    {
      "epoch": 10.443103961870717,
      "grad_norm": 15.0625,
      "learning_rate": 2.7426740002248624e-05,
      "loss": 1.1383,
      "num_input_tokens_seen": 40672720,
      "step": 70115
    },
    {
      "epoch": 10.443848674411678,
      "grad_norm": 10.9375,
      "learning_rate": 2.7423505905385382e-05,
      "loss": 1.0265,
      "num_input_tokens_seen": 40675600,
      "step": 70120
    },
    {
      "epoch": 10.444593386952636,
      "grad_norm": 9.125,
      "learning_rate": 2.742027176757948e-05,
      "loss": 1.2616,
      "num_input_tokens_seen": 40678448,
      "step": 70125
    },
    {
      "epoch": 10.445338099493595,
      "grad_norm": 5.5,
      "learning_rate": 2.741703758888554e-05,
      "loss": 0.8931,
      "num_input_tokens_seen": 40681424,
      "step": 70130
    },
    {
      "epoch": 10.446082812034554,
      "grad_norm": 7.625,
      "learning_rate": 2.7413803369358217e-05,
      "loss": 0.9974,
      "num_input_tokens_seen": 40684304,
      "step": 70135
    },
    {
      "epoch": 10.446827524575514,
      "grad_norm": 9.4375,
      "learning_rate": 2.7410569109052124e-05,
      "loss": 1.2587,
      "num_input_tokens_seen": 40687216,
      "step": 70140
    },
    {
      "epoch": 10.447572237116473,
      "grad_norm": 10.375,
      "learning_rate": 2.7407334808021924e-05,
      "loss": 1.1573,
      "num_input_tokens_seen": 40690032,
      "step": 70145
    },
    {
      "epoch": 10.448316949657432,
      "grad_norm": 6.59375,
      "learning_rate": 2.740410046632224e-05,
      "loss": 1.1714,
      "num_input_tokens_seen": 40693040,
      "step": 70150
    },
    {
      "epoch": 10.44906166219839,
      "grad_norm": 9.1875,
      "learning_rate": 2.7400866084007732e-05,
      "loss": 1.056,
      "num_input_tokens_seen": 40696944,
      "step": 70155
    },
    {
      "epoch": 10.449806374739351,
      "grad_norm": 15.6875,
      "learning_rate": 2.7397631661133032e-05,
      "loss": 0.9825,
      "num_input_tokens_seen": 40699728,
      "step": 70160
    },
    {
      "epoch": 10.45055108728031,
      "grad_norm": 9.1875,
      "learning_rate": 2.7394397197752786e-05,
      "loss": 0.9409,
      "num_input_tokens_seen": 40703088,
      "step": 70165
    },
    {
      "epoch": 10.451295799821269,
      "grad_norm": 9.625,
      "learning_rate": 2.7391162693921624e-05,
      "loss": 0.9019,
      "num_input_tokens_seen": 40705904,
      "step": 70170
    },
    {
      "epoch": 10.452040512362228,
      "grad_norm": 10.5,
      "learning_rate": 2.7387928149694197e-05,
      "loss": 1.0816,
      "num_input_tokens_seen": 40709008,
      "step": 70175
    },
    {
      "epoch": 10.452785224903188,
      "grad_norm": 17.125,
      "learning_rate": 2.7384693565125153e-05,
      "loss": 1.0864,
      "num_input_tokens_seen": 40712080,
      "step": 70180
    },
    {
      "epoch": 10.453529937444147,
      "grad_norm": 9.625,
      "learning_rate": 2.7381458940269134e-05,
      "loss": 0.9336,
      "num_input_tokens_seen": 40714800,
      "step": 70185
    },
    {
      "epoch": 10.454274649985106,
      "grad_norm": 9.6875,
      "learning_rate": 2.737822427518079e-05,
      "loss": 0.8308,
      "num_input_tokens_seen": 40717680,
      "step": 70190
    },
    {
      "epoch": 10.455019362526064,
      "grad_norm": 13.0,
      "learning_rate": 2.7374989569914766e-05,
      "loss": 1.2363,
      "num_input_tokens_seen": 40720592,
      "step": 70195
    },
    {
      "epoch": 10.455764075067025,
      "grad_norm": 13.625,
      "learning_rate": 2.73717548245257e-05,
      "loss": 1.2622,
      "num_input_tokens_seen": 40723600,
      "step": 70200
    },
    {
      "epoch": 10.456508787607984,
      "grad_norm": 14.3125,
      "learning_rate": 2.736852003906826e-05,
      "loss": 1.2737,
      "num_input_tokens_seen": 40726576,
      "step": 70205
    },
    {
      "epoch": 10.457253500148942,
      "grad_norm": 10.875,
      "learning_rate": 2.736528521359707e-05,
      "loss": 0.8805,
      "num_input_tokens_seen": 40729264,
      "step": 70210
    },
    {
      "epoch": 10.457998212689901,
      "grad_norm": 11.0,
      "learning_rate": 2.736205034816679e-05,
      "loss": 1.2557,
      "num_input_tokens_seen": 40732464,
      "step": 70215
    },
    {
      "epoch": 10.458742925230862,
      "grad_norm": 5.5,
      "learning_rate": 2.735881544283207e-05,
      "loss": 0.9487,
      "num_input_tokens_seen": 40735312,
      "step": 70220
    },
    {
      "epoch": 10.45948763777182,
      "grad_norm": 13.5625,
      "learning_rate": 2.735558049764756e-05,
      "loss": 1.1984,
      "num_input_tokens_seen": 40738480,
      "step": 70225
    },
    {
      "epoch": 10.46023235031278,
      "grad_norm": 11.8125,
      "learning_rate": 2.735234551266792e-05,
      "loss": 1.3537,
      "num_input_tokens_seen": 40741424,
      "step": 70230
    },
    {
      "epoch": 10.460977062853738,
      "grad_norm": 7.6875,
      "learning_rate": 2.734911048794779e-05,
      "loss": 1.066,
      "num_input_tokens_seen": 40744080,
      "step": 70235
    },
    {
      "epoch": 10.461721775394698,
      "grad_norm": 9.5625,
      "learning_rate": 2.7345875423541817e-05,
      "loss": 1.0531,
      "num_input_tokens_seen": 40747024,
      "step": 70240
    },
    {
      "epoch": 10.462466487935657,
      "grad_norm": 9.125,
      "learning_rate": 2.7342640319504674e-05,
      "loss": 0.9652,
      "num_input_tokens_seen": 40750064,
      "step": 70245
    },
    {
      "epoch": 10.463211200476616,
      "grad_norm": 17.875,
      "learning_rate": 2.7339405175890998e-05,
      "loss": 0.8707,
      "num_input_tokens_seen": 40752784,
      "step": 70250
    },
    {
      "epoch": 10.463955913017575,
      "grad_norm": 8.8125,
      "learning_rate": 2.733616999275545e-05,
      "loss": 1.0541,
      "num_input_tokens_seen": 40755472,
      "step": 70255
    },
    {
      "epoch": 10.464700625558535,
      "grad_norm": 9.9375,
      "learning_rate": 2.7332934770152686e-05,
      "loss": 1.3862,
      "num_input_tokens_seen": 40758256,
      "step": 70260
    },
    {
      "epoch": 10.465445338099494,
      "grad_norm": 7.84375,
      "learning_rate": 2.7329699508137363e-05,
      "loss": 1.1154,
      "num_input_tokens_seen": 40761232,
      "step": 70265
    },
    {
      "epoch": 10.466190050640453,
      "grad_norm": 15.0,
      "learning_rate": 2.7326464206764125e-05,
      "loss": 1.24,
      "num_input_tokens_seen": 40764048,
      "step": 70270
    },
    {
      "epoch": 10.466934763181412,
      "grad_norm": 10.875,
      "learning_rate": 2.7323228866087647e-05,
      "loss": 1.0664,
      "num_input_tokens_seen": 40767024,
      "step": 70275
    },
    {
      "epoch": 10.46767947572237,
      "grad_norm": 15.0,
      "learning_rate": 2.731999348616257e-05,
      "loss": 0.9456,
      "num_input_tokens_seen": 40769936,
      "step": 70280
    },
    {
      "epoch": 10.46842418826333,
      "grad_norm": 10.125,
      "learning_rate": 2.731675806704357e-05,
      "loss": 0.9893,
      "num_input_tokens_seen": 40773040,
      "step": 70285
    },
    {
      "epoch": 10.46916890080429,
      "grad_norm": 7.3125,
      "learning_rate": 2.7313522608785295e-05,
      "loss": 0.7339,
      "num_input_tokens_seen": 40776240,
      "step": 70290
    },
    {
      "epoch": 10.469913613345248,
      "grad_norm": 13.4375,
      "learning_rate": 2.7310287111442407e-05,
      "loss": 1.0525,
      "num_input_tokens_seen": 40779344,
      "step": 70295
    },
    {
      "epoch": 10.470658325886207,
      "grad_norm": 11.8125,
      "learning_rate": 2.730705157506957e-05,
      "loss": 1.0824,
      "num_input_tokens_seen": 40782256,
      "step": 70300
    },
    {
      "epoch": 10.471403038427168,
      "grad_norm": 13.375,
      "learning_rate": 2.7303815999721433e-05,
      "loss": 1.333,
      "num_input_tokens_seen": 40785040,
      "step": 70305
    },
    {
      "epoch": 10.472147750968126,
      "grad_norm": 10.875,
      "learning_rate": 2.730058038545267e-05,
      "loss": 1.0423,
      "num_input_tokens_seen": 40788016,
      "step": 70310
    },
    {
      "epoch": 10.472892463509085,
      "grad_norm": 25.125,
      "learning_rate": 2.7297344732317938e-05,
      "loss": 0.9079,
      "num_input_tokens_seen": 40790768,
      "step": 70315
    },
    {
      "epoch": 10.473637176050044,
      "grad_norm": 8.5,
      "learning_rate": 2.7294109040371902e-05,
      "loss": 1.1542,
      "num_input_tokens_seen": 40793776,
      "step": 70320
    },
    {
      "epoch": 10.474381888591004,
      "grad_norm": 7.3125,
      "learning_rate": 2.729087330966923e-05,
      "loss": 1.0371,
      "num_input_tokens_seen": 40796784,
      "step": 70325
    },
    {
      "epoch": 10.475126601131963,
      "grad_norm": 10.625,
      "learning_rate": 2.7287637540264584e-05,
      "loss": 1.144,
      "num_input_tokens_seen": 40799824,
      "step": 70330
    },
    {
      "epoch": 10.475871313672922,
      "grad_norm": 8.25,
      "learning_rate": 2.7284401732212615e-05,
      "loss": 1.196,
      "num_input_tokens_seen": 40802544,
      "step": 70335
    },
    {
      "epoch": 10.47661602621388,
      "grad_norm": 21.375,
      "learning_rate": 2.7281165885568006e-05,
      "loss": 1.4641,
      "num_input_tokens_seen": 40805168,
      "step": 70340
    },
    {
      "epoch": 10.477360738754841,
      "grad_norm": 11.125,
      "learning_rate": 2.7277930000385414e-05,
      "loss": 1.2194,
      "num_input_tokens_seen": 40808080,
      "step": 70345
    },
    {
      "epoch": 10.4781054512958,
      "grad_norm": 11.8125,
      "learning_rate": 2.7274694076719513e-05,
      "loss": 1.1321,
      "num_input_tokens_seen": 40810800,
      "step": 70350
    },
    {
      "epoch": 10.478850163836759,
      "grad_norm": 9.9375,
      "learning_rate": 2.727145811462497e-05,
      "loss": 1.0477,
      "num_input_tokens_seen": 40813392,
      "step": 70355
    },
    {
      "epoch": 10.479594876377718,
      "grad_norm": 9.6875,
      "learning_rate": 2.7268222114156454e-05,
      "loss": 1.1562,
      "num_input_tokens_seen": 40816592,
      "step": 70360
    },
    {
      "epoch": 10.480339588918678,
      "grad_norm": 12.0625,
      "learning_rate": 2.7264986075368625e-05,
      "loss": 1.2658,
      "num_input_tokens_seen": 40819632,
      "step": 70365
    },
    {
      "epoch": 10.481084301459637,
      "grad_norm": 7.5625,
      "learning_rate": 2.726174999831616e-05,
      "loss": 0.7123,
      "num_input_tokens_seen": 40822736,
      "step": 70370
    },
    {
      "epoch": 10.481829014000596,
      "grad_norm": 12.6875,
      "learning_rate": 2.7258513883053727e-05,
      "loss": 1.7935,
      "num_input_tokens_seen": 40825552,
      "step": 70375
    },
    {
      "epoch": 10.482573726541554,
      "grad_norm": 8.75,
      "learning_rate": 2.7255277729635997e-05,
      "loss": 0.9676,
      "num_input_tokens_seen": 40828368,
      "step": 70380
    },
    {
      "epoch": 10.483318439082515,
      "grad_norm": 10.5,
      "learning_rate": 2.725204153811764e-05,
      "loss": 1.0909,
      "num_input_tokens_seen": 40831600,
      "step": 70385
    },
    {
      "epoch": 10.484063151623474,
      "grad_norm": 5.9375,
      "learning_rate": 2.7248805308553333e-05,
      "loss": 0.9079,
      "num_input_tokens_seen": 40834512,
      "step": 70390
    },
    {
      "epoch": 10.484807864164432,
      "grad_norm": 11.5625,
      "learning_rate": 2.7245569040997747e-05,
      "loss": 1.3077,
      "num_input_tokens_seen": 40837488,
      "step": 70395
    },
    {
      "epoch": 10.485552576705391,
      "grad_norm": 8.375,
      "learning_rate": 2.7242332735505555e-05,
      "loss": 0.9876,
      "num_input_tokens_seen": 40840368,
      "step": 70400
    },
    {
      "epoch": 10.486297289246352,
      "grad_norm": 10.625,
      "learning_rate": 2.7239096392131423e-05,
      "loss": 0.9286,
      "num_input_tokens_seen": 40843664,
      "step": 70405
    },
    {
      "epoch": 10.48704200178731,
      "grad_norm": 7.15625,
      "learning_rate": 2.723586001093004e-05,
      "loss": 1.0111,
      "num_input_tokens_seen": 40846896,
      "step": 70410
    },
    {
      "epoch": 10.48778671432827,
      "grad_norm": 10.0625,
      "learning_rate": 2.7232623591956074e-05,
      "loss": 1.0133,
      "num_input_tokens_seen": 40849840,
      "step": 70415
    },
    {
      "epoch": 10.488531426869228,
      "grad_norm": 9.1875,
      "learning_rate": 2.72293871352642e-05,
      "loss": 1.0346,
      "num_input_tokens_seen": 40852752,
      "step": 70420
    },
    {
      "epoch": 10.489276139410187,
      "grad_norm": 8.0625,
      "learning_rate": 2.7226150640909092e-05,
      "loss": 1.4335,
      "num_input_tokens_seen": 40855536,
      "step": 70425
    },
    {
      "epoch": 10.490020851951147,
      "grad_norm": 17.75,
      "learning_rate": 2.722291410894544e-05,
      "loss": 1.0947,
      "num_input_tokens_seen": 40858416,
      "step": 70430
    },
    {
      "epoch": 10.490765564492106,
      "grad_norm": 5.34375,
      "learning_rate": 2.721967753942791e-05,
      "loss": 0.9171,
      "num_input_tokens_seen": 40861072,
      "step": 70435
    },
    {
      "epoch": 10.491510277033065,
      "grad_norm": 8.875,
      "learning_rate": 2.721644093241118e-05,
      "loss": 1.0751,
      "num_input_tokens_seen": 40863632,
      "step": 70440
    },
    {
      "epoch": 10.492254989574024,
      "grad_norm": 11.4375,
      "learning_rate": 2.7213204287949938e-05,
      "loss": 1.2429,
      "num_input_tokens_seen": 40866576,
      "step": 70445
    },
    {
      "epoch": 10.492999702114984,
      "grad_norm": 13.125,
      "learning_rate": 2.7209967606098862e-05,
      "loss": 0.8835,
      "num_input_tokens_seen": 40869488,
      "step": 70450
    },
    {
      "epoch": 10.493744414655943,
      "grad_norm": 10.9375,
      "learning_rate": 2.7206730886912624e-05,
      "loss": 0.8854,
      "num_input_tokens_seen": 40872464,
      "step": 70455
    },
    {
      "epoch": 10.494489127196902,
      "grad_norm": 12.8125,
      "learning_rate": 2.7203494130445905e-05,
      "loss": 1.0798,
      "num_input_tokens_seen": 40875696,
      "step": 70460
    },
    {
      "epoch": 10.49523383973786,
      "grad_norm": 7.5625,
      "learning_rate": 2.7200257336753405e-05,
      "loss": 0.8636,
      "num_input_tokens_seen": 40878768,
      "step": 70465
    },
    {
      "epoch": 10.495978552278821,
      "grad_norm": 11.4375,
      "learning_rate": 2.7197020505889786e-05,
      "loss": 0.9962,
      "num_input_tokens_seen": 40881904,
      "step": 70470
    },
    {
      "epoch": 10.49672326481978,
      "grad_norm": 8.375,
      "learning_rate": 2.7193783637909736e-05,
      "loss": 0.8985,
      "num_input_tokens_seen": 40884368,
      "step": 70475
    },
    {
      "epoch": 10.497467977360738,
      "grad_norm": 10.625,
      "learning_rate": 2.7190546732867945e-05,
      "loss": 0.8542,
      "num_input_tokens_seen": 40887088,
      "step": 70480
    },
    {
      "epoch": 10.498212689901697,
      "grad_norm": 8.375,
      "learning_rate": 2.7187309790819092e-05,
      "loss": 1.0322,
      "num_input_tokens_seen": 40890000,
      "step": 70485
    },
    {
      "epoch": 10.498957402442658,
      "grad_norm": 12.0,
      "learning_rate": 2.7184072811817867e-05,
      "loss": 1.2973,
      "num_input_tokens_seen": 40893168,
      "step": 70490
    },
    {
      "epoch": 10.499702114983616,
      "grad_norm": 9.125,
      "learning_rate": 2.7180835795918952e-05,
      "loss": 0.995,
      "num_input_tokens_seen": 40895696,
      "step": 70495
    },
    {
      "epoch": 10.500446827524575,
      "grad_norm": 13.625,
      "learning_rate": 2.7177598743177028e-05,
      "loss": 1.0154,
      "num_input_tokens_seen": 40898640,
      "step": 70500
    },
    {
      "epoch": 10.501191540065534,
      "grad_norm": 7.09375,
      "learning_rate": 2.717436165364679e-05,
      "loss": 1.2183,
      "num_input_tokens_seen": 40901552,
      "step": 70505
    },
    {
      "epoch": 10.501936252606495,
      "grad_norm": 10.5,
      "learning_rate": 2.7171124527382917e-05,
      "loss": 1.0315,
      "num_input_tokens_seen": 40904432,
      "step": 70510
    },
    {
      "epoch": 10.502680965147453,
      "grad_norm": 11.4375,
      "learning_rate": 2.7167887364440102e-05,
      "loss": 0.8889,
      "num_input_tokens_seen": 40907152,
      "step": 70515
    },
    {
      "epoch": 10.503425677688412,
      "grad_norm": 7.125,
      "learning_rate": 2.7164650164873033e-05,
      "loss": 1.7234,
      "num_input_tokens_seen": 40910288,
      "step": 70520
    },
    {
      "epoch": 10.50417039022937,
      "grad_norm": 9.3125,
      "learning_rate": 2.7161412928736407e-05,
      "loss": 0.9068,
      "num_input_tokens_seen": 40913168,
      "step": 70525
    },
    {
      "epoch": 10.504915102770331,
      "grad_norm": 13.125,
      "learning_rate": 2.7158175656084906e-05,
      "loss": 1.0412,
      "num_input_tokens_seen": 40915984,
      "step": 70530
    },
    {
      "epoch": 10.50565981531129,
      "grad_norm": 22.75,
      "learning_rate": 2.7154938346973214e-05,
      "loss": 1.5122,
      "num_input_tokens_seen": 40918832,
      "step": 70535
    },
    {
      "epoch": 10.506404527852249,
      "grad_norm": 8.3125,
      "learning_rate": 2.715170100145603e-05,
      "loss": 0.591,
      "num_input_tokens_seen": 40921520,
      "step": 70540
    },
    {
      "epoch": 10.507149240393208,
      "grad_norm": 9.3125,
      "learning_rate": 2.7148463619588045e-05,
      "loss": 0.9464,
      "num_input_tokens_seen": 40924656,
      "step": 70545
    },
    {
      "epoch": 10.507893952934168,
      "grad_norm": 10.0,
      "learning_rate": 2.714522620142395e-05,
      "loss": 0.9142,
      "num_input_tokens_seen": 40927504,
      "step": 70550
    },
    {
      "epoch": 10.508638665475127,
      "grad_norm": 15.75,
      "learning_rate": 2.7141988747018437e-05,
      "loss": 1.2081,
      "num_input_tokens_seen": 40930224,
      "step": 70555
    },
    {
      "epoch": 10.509383378016086,
      "grad_norm": 12.125,
      "learning_rate": 2.713875125642621e-05,
      "loss": 0.9035,
      "num_input_tokens_seen": 40932912,
      "step": 70560
    },
    {
      "epoch": 10.510128090557044,
      "grad_norm": 7.65625,
      "learning_rate": 2.713551372970195e-05,
      "loss": 1.0957,
      "num_input_tokens_seen": 40935824,
      "step": 70565
    },
    {
      "epoch": 10.510872803098005,
      "grad_norm": 12.4375,
      "learning_rate": 2.7132276166900357e-05,
      "loss": 1.2345,
      "num_input_tokens_seen": 40938576,
      "step": 70570
    },
    {
      "epoch": 10.511617515638964,
      "grad_norm": 11.25,
      "learning_rate": 2.7129038568076122e-05,
      "loss": 1.2809,
      "num_input_tokens_seen": 40941360,
      "step": 70575
    },
    {
      "epoch": 10.512362228179922,
      "grad_norm": 12.125,
      "learning_rate": 2.712580093328394e-05,
      "loss": 0.9742,
      "num_input_tokens_seen": 40944144,
      "step": 70580
    },
    {
      "epoch": 10.513106940720881,
      "grad_norm": 9.4375,
      "learning_rate": 2.7122563262578515e-05,
      "loss": 1.2629,
      "num_input_tokens_seen": 40947024,
      "step": 70585
    },
    {
      "epoch": 10.513851653261842,
      "grad_norm": 6.3125,
      "learning_rate": 2.7119325556014546e-05,
      "loss": 1.1216,
      "num_input_tokens_seen": 40949968,
      "step": 70590
    },
    {
      "epoch": 10.5145963658028,
      "grad_norm": 11.75,
      "learning_rate": 2.7116087813646724e-05,
      "loss": 0.9785,
      "num_input_tokens_seen": 40953104,
      "step": 70595
    },
    {
      "epoch": 10.51534107834376,
      "grad_norm": 10.875,
      "learning_rate": 2.7112850035529748e-05,
      "loss": 1.1725,
      "num_input_tokens_seen": 40955824,
      "step": 70600
    },
    {
      "epoch": 10.516085790884718,
      "grad_norm": 19.625,
      "learning_rate": 2.7109612221718316e-05,
      "loss": 1.199,
      "num_input_tokens_seen": 40958544,
      "step": 70605
    },
    {
      "epoch": 10.516830503425677,
      "grad_norm": 4.34375,
      "learning_rate": 2.7106374372267136e-05,
      "loss": 0.9291,
      "num_input_tokens_seen": 40961328,
      "step": 70610
    },
    {
      "epoch": 10.517575215966637,
      "grad_norm": 6.75,
      "learning_rate": 2.7103136487230895e-05,
      "loss": 1.1109,
      "num_input_tokens_seen": 40963952,
      "step": 70615
    },
    {
      "epoch": 10.518319928507596,
      "grad_norm": 12.125,
      "learning_rate": 2.70998985666643e-05,
      "loss": 1.1246,
      "num_input_tokens_seen": 40967088,
      "step": 70620
    },
    {
      "epoch": 10.519064641048555,
      "grad_norm": 6.75,
      "learning_rate": 2.7096660610622055e-05,
      "loss": 1.449,
      "num_input_tokens_seen": 40969936,
      "step": 70625
    },
    {
      "epoch": 10.519809353589514,
      "grad_norm": 9.3125,
      "learning_rate": 2.7093422619158866e-05,
      "loss": 1.2614,
      "num_input_tokens_seen": 40972752,
      "step": 70630
    },
    {
      "epoch": 10.520554066130474,
      "grad_norm": 7.9375,
      "learning_rate": 2.709018459232942e-05,
      "loss": 1.0204,
      "num_input_tokens_seen": 40975792,
      "step": 70635
    },
    {
      "epoch": 10.521298778671433,
      "grad_norm": 8.25,
      "learning_rate": 2.7086946530188434e-05,
      "loss": 1.2726,
      "num_input_tokens_seen": 40978800,
      "step": 70640
    },
    {
      "epoch": 10.522043491212392,
      "grad_norm": 11.375,
      "learning_rate": 2.7083708432790605e-05,
      "loss": 0.9959,
      "num_input_tokens_seen": 40981936,
      "step": 70645
    },
    {
      "epoch": 10.52278820375335,
      "grad_norm": 6.625,
      "learning_rate": 2.7080470300190646e-05,
      "loss": 0.8121,
      "num_input_tokens_seen": 40984976,
      "step": 70650
    },
    {
      "epoch": 10.523532916294311,
      "grad_norm": 4.625,
      "learning_rate": 2.7077232132443247e-05,
      "loss": 0.8662,
      "num_input_tokens_seen": 40988112,
      "step": 70655
    },
    {
      "epoch": 10.52427762883527,
      "grad_norm": 11.75,
      "learning_rate": 2.7073993929603138e-05,
      "loss": 1.3688,
      "num_input_tokens_seen": 40991248,
      "step": 70660
    },
    {
      "epoch": 10.525022341376228,
      "grad_norm": 11.125,
      "learning_rate": 2.7070755691724993e-05,
      "loss": 1.1458,
      "num_input_tokens_seen": 40994096,
      "step": 70665
    },
    {
      "epoch": 10.525767053917187,
      "grad_norm": 9.6875,
      "learning_rate": 2.7067517418863543e-05,
      "loss": 1.1879,
      "num_input_tokens_seen": 40997040,
      "step": 70670
    },
    {
      "epoch": 10.526511766458148,
      "grad_norm": 6.3125,
      "learning_rate": 2.706427911107348e-05,
      "loss": 0.8692,
      "num_input_tokens_seen": 40999760,
      "step": 70675
    },
    {
      "epoch": 10.527256478999107,
      "grad_norm": 6.84375,
      "learning_rate": 2.7061040768409523e-05,
      "loss": 0.9164,
      "num_input_tokens_seen": 41002448,
      "step": 70680
    },
    {
      "epoch": 10.528001191540065,
      "grad_norm": 14.75,
      "learning_rate": 2.705780239092638e-05,
      "loss": 1.0831,
      "num_input_tokens_seen": 41005232,
      "step": 70685
    },
    {
      "epoch": 10.528745904081024,
      "grad_norm": 10.8125,
      "learning_rate": 2.705456397867876e-05,
      "loss": 0.7993,
      "num_input_tokens_seen": 41007888,
      "step": 70690
    },
    {
      "epoch": 10.529490616621985,
      "grad_norm": 8.3125,
      "learning_rate": 2.7051325531721366e-05,
      "loss": 0.9789,
      "num_input_tokens_seen": 41010672,
      "step": 70695
    },
    {
      "epoch": 10.530235329162943,
      "grad_norm": 8.625,
      "learning_rate": 2.704808705010891e-05,
      "loss": 1.3085,
      "num_input_tokens_seen": 41013520,
      "step": 70700
    },
    {
      "epoch": 10.530980041703902,
      "grad_norm": 11.5625,
      "learning_rate": 2.7044848533896105e-05,
      "loss": 1.0766,
      "num_input_tokens_seen": 41016432,
      "step": 70705
    },
    {
      "epoch": 10.53172475424486,
      "grad_norm": 13.125,
      "learning_rate": 2.704160998313766e-05,
      "loss": 1.1347,
      "num_input_tokens_seen": 41019120,
      "step": 70710
    },
    {
      "epoch": 10.532469466785821,
      "grad_norm": 6.1875,
      "learning_rate": 2.7038371397888295e-05,
      "loss": 1.3826,
      "num_input_tokens_seen": 41021840,
      "step": 70715
    },
    {
      "epoch": 10.53321417932678,
      "grad_norm": 5.625,
      "learning_rate": 2.7035132778202717e-05,
      "loss": 1.4161,
      "num_input_tokens_seen": 41024720,
      "step": 70720
    },
    {
      "epoch": 10.533958891867739,
      "grad_norm": 8.9375,
      "learning_rate": 2.7031894124135638e-05,
      "loss": 0.7848,
      "num_input_tokens_seen": 41027792,
      "step": 70725
    },
    {
      "epoch": 10.534703604408698,
      "grad_norm": 13.4375,
      "learning_rate": 2.7028655435741772e-05,
      "loss": 0.6886,
      "num_input_tokens_seen": 41030576,
      "step": 70730
    },
    {
      "epoch": 10.535448316949658,
      "grad_norm": 16.75,
      "learning_rate": 2.7025416713075836e-05,
      "loss": 1.3318,
      "num_input_tokens_seen": 41033392,
      "step": 70735
    },
    {
      "epoch": 10.536193029490617,
      "grad_norm": 6.6875,
      "learning_rate": 2.702217795619254e-05,
      "loss": 0.7829,
      "num_input_tokens_seen": 41036144,
      "step": 70740
    },
    {
      "epoch": 10.536937742031576,
      "grad_norm": 5.8125,
      "learning_rate": 2.7018939165146606e-05,
      "loss": 0.7839,
      "num_input_tokens_seen": 41039056,
      "step": 70745
    },
    {
      "epoch": 10.537682454572534,
      "grad_norm": 14.75,
      "learning_rate": 2.701570033999274e-05,
      "loss": 1.4031,
      "num_input_tokens_seen": 41042032,
      "step": 70750
    },
    {
      "epoch": 10.538427167113493,
      "grad_norm": 11.875,
      "learning_rate": 2.7012461480785668e-05,
      "loss": 1.406,
      "num_input_tokens_seen": 41045040,
      "step": 70755
    },
    {
      "epoch": 10.539171879654454,
      "grad_norm": 12.25,
      "learning_rate": 2.7009222587580114e-05,
      "loss": 0.9991,
      "num_input_tokens_seen": 41048080,
      "step": 70760
    },
    {
      "epoch": 10.539916592195413,
      "grad_norm": 11.8125,
      "learning_rate": 2.7005983660430778e-05,
      "loss": 1.5014,
      "num_input_tokens_seen": 41051024,
      "step": 70765
    },
    {
      "epoch": 10.540661304736371,
      "grad_norm": 7.90625,
      "learning_rate": 2.700274469939239e-05,
      "loss": 1.0964,
      "num_input_tokens_seen": 41054032,
      "step": 70770
    },
    {
      "epoch": 10.541406017277332,
      "grad_norm": 9.375,
      "learning_rate": 2.6999505704519662e-05,
      "loss": 0.9573,
      "num_input_tokens_seen": 41056720,
      "step": 70775
    },
    {
      "epoch": 10.54215072981829,
      "grad_norm": 9.4375,
      "learning_rate": 2.6996266675867322e-05,
      "loss": 0.9829,
      "num_input_tokens_seen": 41059920,
      "step": 70780
    },
    {
      "epoch": 10.54289544235925,
      "grad_norm": 10.75,
      "learning_rate": 2.6993027613490078e-05,
      "loss": 1.112,
      "num_input_tokens_seen": 41062896,
      "step": 70785
    },
    {
      "epoch": 10.543640154900208,
      "grad_norm": 10.5,
      "learning_rate": 2.698978851744266e-05,
      "loss": 1.1043,
      "num_input_tokens_seen": 41065904,
      "step": 70790
    },
    {
      "epoch": 10.544384867441167,
      "grad_norm": 15.9375,
      "learning_rate": 2.69865493877798e-05,
      "loss": 0.9928,
      "num_input_tokens_seen": 41068784,
      "step": 70795
    },
    {
      "epoch": 10.545129579982127,
      "grad_norm": 9.25,
      "learning_rate": 2.698331022455619e-05,
      "loss": 1.0026,
      "num_input_tokens_seen": 41071728,
      "step": 70800
    },
    {
      "epoch": 10.545874292523086,
      "grad_norm": 7.84375,
      "learning_rate": 2.6980071027826574e-05,
      "loss": 0.7436,
      "num_input_tokens_seen": 41074800,
      "step": 70805
    },
    {
      "epoch": 10.546619005064045,
      "grad_norm": 16.375,
      "learning_rate": 2.697683179764568e-05,
      "loss": 1.0062,
      "num_input_tokens_seen": 41077488,
      "step": 70810
    },
    {
      "epoch": 10.547363717605004,
      "grad_norm": 9.125,
      "learning_rate": 2.6973592534068216e-05,
      "loss": 0.9656,
      "num_input_tokens_seen": 41080368,
      "step": 70815
    },
    {
      "epoch": 10.548108430145964,
      "grad_norm": 9.0625,
      "learning_rate": 2.697035323714891e-05,
      "loss": 0.867,
      "num_input_tokens_seen": 41083152,
      "step": 70820
    },
    {
      "epoch": 10.548853142686923,
      "grad_norm": 8.1875,
      "learning_rate": 2.6967113906942494e-05,
      "loss": 1.3193,
      "num_input_tokens_seen": 41086000,
      "step": 70825
    },
    {
      "epoch": 10.549597855227882,
      "grad_norm": 7.9375,
      "learning_rate": 2.696387454350368e-05,
      "loss": 0.8517,
      "num_input_tokens_seen": 41088912,
      "step": 70830
    },
    {
      "epoch": 10.55034256776884,
      "grad_norm": 6.625,
      "learning_rate": 2.6960635146887202e-05,
      "loss": 1.0672,
      "num_input_tokens_seen": 41091952,
      "step": 70835
    },
    {
      "epoch": 10.551087280309801,
      "grad_norm": 10.9375,
      "learning_rate": 2.6957395717147794e-05,
      "loss": 0.7689,
      "num_input_tokens_seen": 41094832,
      "step": 70840
    },
    {
      "epoch": 10.55183199285076,
      "grad_norm": 10.0,
      "learning_rate": 2.695415625434017e-05,
      "loss": 1.2999,
      "num_input_tokens_seen": 41097680,
      "step": 70845
    },
    {
      "epoch": 10.552576705391719,
      "grad_norm": 8.4375,
      "learning_rate": 2.695091675851906e-05,
      "loss": 1.0545,
      "num_input_tokens_seen": 41100496,
      "step": 70850
    },
    {
      "epoch": 10.553321417932677,
      "grad_norm": 5.71875,
      "learning_rate": 2.6947677229739198e-05,
      "loss": 0.6669,
      "num_input_tokens_seen": 41103152,
      "step": 70855
    },
    {
      "epoch": 10.554066130473638,
      "grad_norm": 11.375,
      "learning_rate": 2.6944437668055313e-05,
      "loss": 1.0385,
      "num_input_tokens_seen": 41105936,
      "step": 70860
    },
    {
      "epoch": 10.554810843014597,
      "grad_norm": 9.0625,
      "learning_rate": 2.6941198073522118e-05,
      "loss": 1.2004,
      "num_input_tokens_seen": 41108912,
      "step": 70865
    },
    {
      "epoch": 10.555555555555555,
      "grad_norm": 9.9375,
      "learning_rate": 2.693795844619436e-05,
      "loss": 0.8868,
      "num_input_tokens_seen": 41111824,
      "step": 70870
    },
    {
      "epoch": 10.556300268096514,
      "grad_norm": 16.125,
      "learning_rate": 2.6934718786126763e-05,
      "loss": 1.157,
      "num_input_tokens_seen": 41114928,
      "step": 70875
    },
    {
      "epoch": 10.557044980637475,
      "grad_norm": 16.875,
      "learning_rate": 2.6931479093374056e-05,
      "loss": 1.1532,
      "num_input_tokens_seen": 41117936,
      "step": 70880
    },
    {
      "epoch": 10.557789693178433,
      "grad_norm": 8.3125,
      "learning_rate": 2.6928239367990974e-05,
      "loss": 1.3798,
      "num_input_tokens_seen": 41120880,
      "step": 70885
    },
    {
      "epoch": 10.558534405719392,
      "grad_norm": 3.96875,
      "learning_rate": 2.692499961003226e-05,
      "loss": 0.8712,
      "num_input_tokens_seen": 41123920,
      "step": 70890
    },
    {
      "epoch": 10.559279118260351,
      "grad_norm": 15.5,
      "learning_rate": 2.692175981955263e-05,
      "loss": 1.0616,
      "num_input_tokens_seen": 41126736,
      "step": 70895
    },
    {
      "epoch": 10.560023830801311,
      "grad_norm": 5.34375,
      "learning_rate": 2.691851999660681e-05,
      "loss": 1.2486,
      "num_input_tokens_seen": 41129392,
      "step": 70900
    },
    {
      "epoch": 10.56076854334227,
      "grad_norm": 8.9375,
      "learning_rate": 2.691528014124955e-05,
      "loss": 0.9883,
      "num_input_tokens_seen": 41132016,
      "step": 70905
    },
    {
      "epoch": 10.561513255883229,
      "grad_norm": 12.125,
      "learning_rate": 2.6912040253535574e-05,
      "loss": 1.0798,
      "num_input_tokens_seen": 41134928,
      "step": 70910
    },
    {
      "epoch": 10.562257968424188,
      "grad_norm": 8.1875,
      "learning_rate": 2.6908800333519625e-05,
      "loss": 0.8458,
      "num_input_tokens_seen": 41138064,
      "step": 70915
    },
    {
      "epoch": 10.563002680965148,
      "grad_norm": 9.1875,
      "learning_rate": 2.6905560381256434e-05,
      "loss": 1.4028,
      "num_input_tokens_seen": 41140752,
      "step": 70920
    },
    {
      "epoch": 10.563747393506107,
      "grad_norm": 9.8125,
      "learning_rate": 2.690232039680074e-05,
      "loss": 1.2495,
      "num_input_tokens_seen": 41143568,
      "step": 70925
    },
    {
      "epoch": 10.564492106047066,
      "grad_norm": 6.75,
      "learning_rate": 2.6899080380207276e-05,
      "loss": 1.0892,
      "num_input_tokens_seen": 41146384,
      "step": 70930
    },
    {
      "epoch": 10.565236818588025,
      "grad_norm": 9.5625,
      "learning_rate": 2.689584033153078e-05,
      "loss": 1.3145,
      "num_input_tokens_seen": 41149744,
      "step": 70935
    },
    {
      "epoch": 10.565981531128983,
      "grad_norm": 9.25,
      "learning_rate": 2.6892600250825982e-05,
      "loss": 1.3019,
      "num_input_tokens_seen": 41152464,
      "step": 70940
    },
    {
      "epoch": 10.566726243669944,
      "grad_norm": 14.3125,
      "learning_rate": 2.688936013814763e-05,
      "loss": 1.1522,
      "num_input_tokens_seen": 41155248,
      "step": 70945
    },
    {
      "epoch": 10.567470956210903,
      "grad_norm": 14.625,
      "learning_rate": 2.688611999355046e-05,
      "loss": 1.0304,
      "num_input_tokens_seen": 41158032,
      "step": 70950
    },
    {
      "epoch": 10.568215668751861,
      "grad_norm": 8.0625,
      "learning_rate": 2.6882879817089207e-05,
      "loss": 1.0412,
      "num_input_tokens_seen": 41160816,
      "step": 70955
    },
    {
      "epoch": 10.568960381292822,
      "grad_norm": 16.25,
      "learning_rate": 2.6879639608818618e-05,
      "loss": 1.117,
      "num_input_tokens_seen": 41163920,
      "step": 70960
    },
    {
      "epoch": 10.56970509383378,
      "grad_norm": 5.15625,
      "learning_rate": 2.6876399368793425e-05,
      "loss": 1.2026,
      "num_input_tokens_seen": 41166576,
      "step": 70965
    },
    {
      "epoch": 10.57044980637474,
      "grad_norm": 10.625,
      "learning_rate": 2.6873159097068366e-05,
      "loss": 1.1738,
      "num_input_tokens_seen": 41169168,
      "step": 70970
    },
    {
      "epoch": 10.571194518915698,
      "grad_norm": 15.125,
      "learning_rate": 2.68699187936982e-05,
      "loss": 1.3015,
      "num_input_tokens_seen": 41171920,
      "step": 70975
    },
    {
      "epoch": 10.571939231456657,
      "grad_norm": 9.1875,
      "learning_rate": 2.686667845873765e-05,
      "loss": 0.7284,
      "num_input_tokens_seen": 41174544,
      "step": 70980
    },
    {
      "epoch": 10.572683943997617,
      "grad_norm": 13.3125,
      "learning_rate": 2.686343809224146e-05,
      "loss": 1.164,
      "num_input_tokens_seen": 41177712,
      "step": 70985
    },
    {
      "epoch": 10.573428656538576,
      "grad_norm": 8.3125,
      "learning_rate": 2.6860197694264388e-05,
      "loss": 1.0824,
      "num_input_tokens_seen": 41180560,
      "step": 70990
    },
    {
      "epoch": 10.574173369079535,
      "grad_norm": 12.625,
      "learning_rate": 2.685695726486116e-05,
      "loss": 1.3892,
      "num_input_tokens_seen": 41183280,
      "step": 70995
    },
    {
      "epoch": 10.574918081620494,
      "grad_norm": 17.875,
      "learning_rate": 2.6853716804086527e-05,
      "loss": 0.9957,
      "num_input_tokens_seen": 41185968,
      "step": 71000
    },
    {
      "epoch": 10.575662794161454,
      "grad_norm": 12.25,
      "learning_rate": 2.6850476311995226e-05,
      "loss": 1.278,
      "num_input_tokens_seen": 41188752,
      "step": 71005
    },
    {
      "epoch": 10.576407506702413,
      "grad_norm": 6.625,
      "learning_rate": 2.6847235788642018e-05,
      "loss": 1.2877,
      "num_input_tokens_seen": 41191568,
      "step": 71010
    },
    {
      "epoch": 10.577152219243372,
      "grad_norm": 8.375,
      "learning_rate": 2.6843995234081636e-05,
      "loss": 0.9439,
      "num_input_tokens_seen": 41194576,
      "step": 71015
    },
    {
      "epoch": 10.57789693178433,
      "grad_norm": 15.75,
      "learning_rate": 2.6840754648368826e-05,
      "loss": 0.6638,
      "num_input_tokens_seen": 41197520,
      "step": 71020
    },
    {
      "epoch": 10.578641644325291,
      "grad_norm": 13.9375,
      "learning_rate": 2.6837514031558347e-05,
      "loss": 1.0669,
      "num_input_tokens_seen": 41200528,
      "step": 71025
    },
    {
      "epoch": 10.57938635686625,
      "grad_norm": 13.6875,
      "learning_rate": 2.6834273383704927e-05,
      "loss": 1.0847,
      "num_input_tokens_seen": 41203760,
      "step": 71030
    },
    {
      "epoch": 10.580131069407209,
      "grad_norm": 12.6875,
      "learning_rate": 2.6831032704863324e-05,
      "loss": 1.0322,
      "num_input_tokens_seen": 41206672,
      "step": 71035
    },
    {
      "epoch": 10.580875781948167,
      "grad_norm": 9.625,
      "learning_rate": 2.6827791995088282e-05,
      "loss": 1.0444,
      "num_input_tokens_seen": 41209488,
      "step": 71040
    },
    {
      "epoch": 10.581620494489128,
      "grad_norm": 8.5,
      "learning_rate": 2.6824551254434555e-05,
      "loss": 1.3308,
      "num_input_tokens_seen": 41212784,
      "step": 71045
    },
    {
      "epoch": 10.582365207030087,
      "grad_norm": 7.40625,
      "learning_rate": 2.6821310482956886e-05,
      "loss": 0.9574,
      "num_input_tokens_seen": 41215728,
      "step": 71050
    },
    {
      "epoch": 10.583109919571045,
      "grad_norm": 11.9375,
      "learning_rate": 2.6818069680710034e-05,
      "loss": 0.8308,
      "num_input_tokens_seen": 41218768,
      "step": 71055
    },
    {
      "epoch": 10.583854632112004,
      "grad_norm": 9.25,
      "learning_rate": 2.681482884774874e-05,
      "loss": 1.1055,
      "num_input_tokens_seen": 41221392,
      "step": 71060
    },
    {
      "epoch": 10.584599344652965,
      "grad_norm": 16.375,
      "learning_rate": 2.6811587984127758e-05,
      "loss": 1.3015,
      "num_input_tokens_seen": 41224144,
      "step": 71065
    },
    {
      "epoch": 10.585344057193923,
      "grad_norm": 13.125,
      "learning_rate": 2.680834708990183e-05,
      "loss": 0.7832,
      "num_input_tokens_seen": 41227152,
      "step": 71070
    },
    {
      "epoch": 10.586088769734882,
      "grad_norm": 9.875,
      "learning_rate": 2.680510616512572e-05,
      "loss": 1.104,
      "num_input_tokens_seen": 41229904,
      "step": 71075
    },
    {
      "epoch": 10.586833482275841,
      "grad_norm": 9.25,
      "learning_rate": 2.6801865209854177e-05,
      "loss": 1.0892,
      "num_input_tokens_seen": 41233040,
      "step": 71080
    },
    {
      "epoch": 10.587578194816802,
      "grad_norm": 10.125,
      "learning_rate": 2.6798624224141954e-05,
      "loss": 1.1809,
      "num_input_tokens_seen": 41235600,
      "step": 71085
    },
    {
      "epoch": 10.58832290735776,
      "grad_norm": 12.1875,
      "learning_rate": 2.6795383208043805e-05,
      "loss": 1.0928,
      "num_input_tokens_seen": 41238672,
      "step": 71090
    },
    {
      "epoch": 10.589067619898719,
      "grad_norm": 11.5,
      "learning_rate": 2.679214216161448e-05,
      "loss": 1.2411,
      "num_input_tokens_seen": 41241552,
      "step": 71095
    },
    {
      "epoch": 10.589812332439678,
      "grad_norm": 6.375,
      "learning_rate": 2.6788901084908734e-05,
      "loss": 0.9347,
      "num_input_tokens_seen": 41244304,
      "step": 71100
    },
    {
      "epoch": 10.590557044980638,
      "grad_norm": 8.25,
      "learning_rate": 2.678565997798132e-05,
      "loss": 1.0238,
      "num_input_tokens_seen": 41247408,
      "step": 71105
    },
    {
      "epoch": 10.591301757521597,
      "grad_norm": 11.875,
      "learning_rate": 2.6782418840886997e-05,
      "loss": 1.1635,
      "num_input_tokens_seen": 41250288,
      "step": 71110
    },
    {
      "epoch": 10.592046470062556,
      "grad_norm": 12.6875,
      "learning_rate": 2.6779177673680516e-05,
      "loss": 0.9711,
      "num_input_tokens_seen": 41253008,
      "step": 71115
    },
    {
      "epoch": 10.592791182603515,
      "grad_norm": 11.5,
      "learning_rate": 2.6775936476416636e-05,
      "loss": 1.1195,
      "num_input_tokens_seen": 41255824,
      "step": 71120
    },
    {
      "epoch": 10.593535895144473,
      "grad_norm": 13.875,
      "learning_rate": 2.6772695249150125e-05,
      "loss": 1.1995,
      "num_input_tokens_seen": 41258800,
      "step": 71125
    },
    {
      "epoch": 10.594280607685434,
      "grad_norm": 5.96875,
      "learning_rate": 2.6769453991935717e-05,
      "loss": 1.0638,
      "num_input_tokens_seen": 41261776,
      "step": 71130
    },
    {
      "epoch": 10.595025320226393,
      "grad_norm": 7.15625,
      "learning_rate": 2.676621270482819e-05,
      "loss": 1.2623,
      "num_input_tokens_seen": 41264496,
      "step": 71135
    },
    {
      "epoch": 10.595770032767351,
      "grad_norm": 23.0,
      "learning_rate": 2.6762971387882297e-05,
      "loss": 1.3239,
      "num_input_tokens_seen": 41267472,
      "step": 71140
    },
    {
      "epoch": 10.59651474530831,
      "grad_norm": 21.0,
      "learning_rate": 2.6759730041152787e-05,
      "loss": 1.4458,
      "num_input_tokens_seen": 41270480,
      "step": 71145
    },
    {
      "epoch": 10.59725945784927,
      "grad_norm": 8.625,
      "learning_rate": 2.6756488664694422e-05,
      "loss": 1.319,
      "num_input_tokens_seen": 41273104,
      "step": 71150
    },
    {
      "epoch": 10.59800417039023,
      "grad_norm": 13.9375,
      "learning_rate": 2.675324725856198e-05,
      "loss": 0.9403,
      "num_input_tokens_seen": 41276080,
      "step": 71155
    },
    {
      "epoch": 10.598748882931188,
      "grad_norm": 12.375,
      "learning_rate": 2.6750005822810197e-05,
      "loss": 1.2411,
      "num_input_tokens_seen": 41278672,
      "step": 71160
    },
    {
      "epoch": 10.599493595472147,
      "grad_norm": 26.875,
      "learning_rate": 2.6746764357493848e-05,
      "loss": 1.644,
      "num_input_tokens_seen": 41281328,
      "step": 71165
    },
    {
      "epoch": 10.600238308013108,
      "grad_norm": 13.4375,
      "learning_rate": 2.6743522862667687e-05,
      "loss": 1.2386,
      "num_input_tokens_seen": 41284048,
      "step": 71170
    },
    {
      "epoch": 10.600983020554066,
      "grad_norm": 6.8125,
      "learning_rate": 2.6740281338386484e-05,
      "loss": 1.0774,
      "num_input_tokens_seen": 41286928,
      "step": 71175
    },
    {
      "epoch": 10.601727733095025,
      "grad_norm": 9.5,
      "learning_rate": 2.6737039784705e-05,
      "loss": 1.2032,
      "num_input_tokens_seen": 41289968,
      "step": 71180
    },
    {
      "epoch": 10.602472445635984,
      "grad_norm": 6.9375,
      "learning_rate": 2.6733798201677985e-05,
      "loss": 0.6864,
      "num_input_tokens_seen": 41292656,
      "step": 71185
    },
    {
      "epoch": 10.603217158176944,
      "grad_norm": 12.5625,
      "learning_rate": 2.6730556589360216e-05,
      "loss": 1.188,
      "num_input_tokens_seen": 41295568,
      "step": 71190
    },
    {
      "epoch": 10.603961870717903,
      "grad_norm": 12.8125,
      "learning_rate": 2.672731494780645e-05,
      "loss": 1.1894,
      "num_input_tokens_seen": 41298352,
      "step": 71195
    },
    {
      "epoch": 10.604706583258862,
      "grad_norm": 9.125,
      "learning_rate": 2.672407327707146e-05,
      "loss": 1.1395,
      "num_input_tokens_seen": 41301040,
      "step": 71200
    },
    {
      "epoch": 10.60545129579982,
      "grad_norm": 17.875,
      "learning_rate": 2.6720831577209997e-05,
      "loss": 1.0738,
      "num_input_tokens_seen": 41303856,
      "step": 71205
    },
    {
      "epoch": 10.606196008340781,
      "grad_norm": 3.75,
      "learning_rate": 2.6717589848276835e-05,
      "loss": 0.665,
      "num_input_tokens_seen": 41306672,
      "step": 71210
    },
    {
      "epoch": 10.60694072088174,
      "grad_norm": 10.4375,
      "learning_rate": 2.671434809032674e-05,
      "loss": 1.2464,
      "num_input_tokens_seen": 41309552,
      "step": 71215
    },
    {
      "epoch": 10.607685433422699,
      "grad_norm": 5.53125,
      "learning_rate": 2.6711106303414478e-05,
      "loss": 1.2951,
      "num_input_tokens_seen": 41312976,
      "step": 71220
    },
    {
      "epoch": 10.608430145963657,
      "grad_norm": 10.0625,
      "learning_rate": 2.6707864487594815e-05,
      "loss": 1.1965,
      "num_input_tokens_seen": 41316304,
      "step": 71225
    },
    {
      "epoch": 10.609174858504618,
      "grad_norm": 12.5625,
      "learning_rate": 2.6704622642922512e-05,
      "loss": 0.7023,
      "num_input_tokens_seen": 41318992,
      "step": 71230
    },
    {
      "epoch": 10.609919571045577,
      "grad_norm": 12.0,
      "learning_rate": 2.6701380769452346e-05,
      "loss": 1.2663,
      "num_input_tokens_seen": 41322000,
      "step": 71235
    },
    {
      "epoch": 10.610664283586535,
      "grad_norm": 7.4375,
      "learning_rate": 2.6698138867239076e-05,
      "loss": 1.0908,
      "num_input_tokens_seen": 41325104,
      "step": 71240
    },
    {
      "epoch": 10.611408996127494,
      "grad_norm": 6.71875,
      "learning_rate": 2.6694896936337477e-05,
      "loss": 1.379,
      "num_input_tokens_seen": 41327888,
      "step": 71245
    },
    {
      "epoch": 10.612153708668455,
      "grad_norm": 9.875,
      "learning_rate": 2.6691654976802316e-05,
      "loss": 1.1164,
      "num_input_tokens_seen": 41330672,
      "step": 71250
    },
    {
      "epoch": 10.612898421209414,
      "grad_norm": 12.9375,
      "learning_rate": 2.6688412988688372e-05,
      "loss": 0.9324,
      "num_input_tokens_seen": 41333360,
      "step": 71255
    },
    {
      "epoch": 10.613643133750372,
      "grad_norm": 17.0,
      "learning_rate": 2.6685170972050404e-05,
      "loss": 1.3164,
      "num_input_tokens_seen": 41336400,
      "step": 71260
    },
    {
      "epoch": 10.614387846291331,
      "grad_norm": 14.6875,
      "learning_rate": 2.6681928926943183e-05,
      "loss": 1.1818,
      "num_input_tokens_seen": 41339120,
      "step": 71265
    },
    {
      "epoch": 10.615132558832292,
      "grad_norm": 10.625,
      "learning_rate": 2.667868685342148e-05,
      "loss": 0.7787,
      "num_input_tokens_seen": 41342096,
      "step": 71270
    },
    {
      "epoch": 10.61587727137325,
      "grad_norm": 7.9375,
      "learning_rate": 2.6675444751540068e-05,
      "loss": 1.0589,
      "num_input_tokens_seen": 41344944,
      "step": 71275
    },
    {
      "epoch": 10.616621983914209,
      "grad_norm": 6.0,
      "learning_rate": 2.667220262135372e-05,
      "loss": 1.0547,
      "num_input_tokens_seen": 41347600,
      "step": 71280
    },
    {
      "epoch": 10.617366696455168,
      "grad_norm": 15.25,
      "learning_rate": 2.666896046291721e-05,
      "loss": 0.7469,
      "num_input_tokens_seen": 41350640,
      "step": 71285
    },
    {
      "epoch": 10.618111408996128,
      "grad_norm": 10.75,
      "learning_rate": 2.6665718276285312e-05,
      "loss": 1.187,
      "num_input_tokens_seen": 41353392,
      "step": 71290
    },
    {
      "epoch": 10.618856121537087,
      "grad_norm": 11.0,
      "learning_rate": 2.66624760615128e-05,
      "loss": 1.1731,
      "num_input_tokens_seen": 41356560,
      "step": 71295
    },
    {
      "epoch": 10.619600834078046,
      "grad_norm": 10.1875,
      "learning_rate": 2.6659233818654434e-05,
      "loss": 0.8638,
      "num_input_tokens_seen": 41359440,
      "step": 71300
    },
    {
      "epoch": 10.620345546619005,
      "grad_norm": 7.96875,
      "learning_rate": 2.6655991547765e-05,
      "loss": 1.2441,
      "num_input_tokens_seen": 41362288,
      "step": 71305
    },
    {
      "epoch": 10.621090259159963,
      "grad_norm": 9.125,
      "learning_rate": 2.6652749248899277e-05,
      "loss": 1.004,
      "num_input_tokens_seen": 41365040,
      "step": 71310
    },
    {
      "epoch": 10.621834971700924,
      "grad_norm": 10.3125,
      "learning_rate": 2.6649506922112033e-05,
      "loss": 1.0451,
      "num_input_tokens_seen": 41367824,
      "step": 71315
    },
    {
      "epoch": 10.622579684241883,
      "grad_norm": 17.5,
      "learning_rate": 2.6646264567458052e-05,
      "loss": 1.1544,
      "num_input_tokens_seen": 41370640,
      "step": 71320
    },
    {
      "epoch": 10.623324396782841,
      "grad_norm": 9.6875,
      "learning_rate": 2.6643022184992096e-05,
      "loss": 1.0876,
      "num_input_tokens_seen": 41373392,
      "step": 71325
    },
    {
      "epoch": 10.6240691093238,
      "grad_norm": 12.75,
      "learning_rate": 2.6639779774768953e-05,
      "loss": 0.9505,
      "num_input_tokens_seen": 41376208,
      "step": 71330
    },
    {
      "epoch": 10.62481382186476,
      "grad_norm": 9.625,
      "learning_rate": 2.6636537336843396e-05,
      "loss": 1.2167,
      "num_input_tokens_seen": 41379216,
      "step": 71335
    },
    {
      "epoch": 10.62555853440572,
      "grad_norm": 6.25,
      "learning_rate": 2.663329487127021e-05,
      "loss": 1.261,
      "num_input_tokens_seen": 41381872,
      "step": 71340
    },
    {
      "epoch": 10.626303246946678,
      "grad_norm": 16.375,
      "learning_rate": 2.663005237810416e-05,
      "loss": 1.1362,
      "num_input_tokens_seen": 41384688,
      "step": 71345
    },
    {
      "epoch": 10.627047959487637,
      "grad_norm": 15.375,
      "learning_rate": 2.6626809857400033e-05,
      "loss": 1.2966,
      "num_input_tokens_seen": 41387824,
      "step": 71350
    },
    {
      "epoch": 10.627792672028598,
      "grad_norm": 6.75,
      "learning_rate": 2.662356730921261e-05,
      "loss": 1.1846,
      "num_input_tokens_seen": 41390640,
      "step": 71355
    },
    {
      "epoch": 10.628537384569556,
      "grad_norm": 4.78125,
      "learning_rate": 2.6620324733596664e-05,
      "loss": 1.1935,
      "num_input_tokens_seen": 41393584,
      "step": 71360
    },
    {
      "epoch": 10.629282097110515,
      "grad_norm": 8.375,
      "learning_rate": 2.661708213060698e-05,
      "loss": 1.1315,
      "num_input_tokens_seen": 41396432,
      "step": 71365
    },
    {
      "epoch": 10.630026809651474,
      "grad_norm": 8.25,
      "learning_rate": 2.661383950029834e-05,
      "loss": 1.2185,
      "num_input_tokens_seen": 41399184,
      "step": 71370
    },
    {
      "epoch": 10.630771522192434,
      "grad_norm": 10.1875,
      "learning_rate": 2.6610596842725522e-05,
      "loss": 1.0578,
      "num_input_tokens_seen": 41401904,
      "step": 71375
    },
    {
      "epoch": 10.631516234733393,
      "grad_norm": 6.375,
      "learning_rate": 2.66073541579433e-05,
      "loss": 1.1439,
      "num_input_tokens_seen": 41404752,
      "step": 71380
    },
    {
      "epoch": 10.632260947274352,
      "grad_norm": 8.8125,
      "learning_rate": 2.6604111446006464e-05,
      "loss": 1.1477,
      "num_input_tokens_seen": 41407984,
      "step": 71385
    },
    {
      "epoch": 10.63300565981531,
      "grad_norm": 11.5,
      "learning_rate": 2.6600868706969806e-05,
      "loss": 1.0845,
      "num_input_tokens_seen": 41410768,
      "step": 71390
    },
    {
      "epoch": 10.633750372356271,
      "grad_norm": 6.5625,
      "learning_rate": 2.6597625940888087e-05,
      "loss": 0.9967,
      "num_input_tokens_seen": 41413936,
      "step": 71395
    },
    {
      "epoch": 10.63449508489723,
      "grad_norm": 11.375,
      "learning_rate": 2.6594383147816103e-05,
      "loss": 0.9274,
      "num_input_tokens_seen": 41417072,
      "step": 71400
    },
    {
      "epoch": 10.635239797438189,
      "grad_norm": 14.125,
      "learning_rate": 2.659114032780864e-05,
      "loss": 1.3167,
      "num_input_tokens_seen": 41420048,
      "step": 71405
    },
    {
      "epoch": 10.635984509979147,
      "grad_norm": 4.03125,
      "learning_rate": 2.6587897480920478e-05,
      "loss": 0.8457,
      "num_input_tokens_seen": 41423152,
      "step": 71410
    },
    {
      "epoch": 10.636729222520108,
      "grad_norm": 15.625,
      "learning_rate": 2.6584654607206404e-05,
      "loss": 1.1937,
      "num_input_tokens_seen": 41425872,
      "step": 71415
    },
    {
      "epoch": 10.637473935061067,
      "grad_norm": 8.25,
      "learning_rate": 2.6581411706721194e-05,
      "loss": 1.0079,
      "num_input_tokens_seen": 41428688,
      "step": 71420
    },
    {
      "epoch": 10.638218647602026,
      "grad_norm": 5.40625,
      "learning_rate": 2.6578168779519652e-05,
      "loss": 1.0826,
      "num_input_tokens_seen": 41431472,
      "step": 71425
    },
    {
      "epoch": 10.638963360142984,
      "grad_norm": 14.75,
      "learning_rate": 2.657492582565654e-05,
      "loss": 1.5287,
      "num_input_tokens_seen": 41434352,
      "step": 71430
    },
    {
      "epoch": 10.639708072683945,
      "grad_norm": 14.5,
      "learning_rate": 2.6571682845186662e-05,
      "loss": 1.4006,
      "num_input_tokens_seen": 41437328,
      "step": 71435
    },
    {
      "epoch": 10.640452785224904,
      "grad_norm": 8.5,
      "learning_rate": 2.6568439838164798e-05,
      "loss": 1.0931,
      "num_input_tokens_seen": 41440368,
      "step": 71440
    },
    {
      "epoch": 10.641197497765862,
      "grad_norm": 8.1875,
      "learning_rate": 2.6565196804645738e-05,
      "loss": 0.9144,
      "num_input_tokens_seen": 41443312,
      "step": 71445
    },
    {
      "epoch": 10.641942210306821,
      "grad_norm": 9.5,
      "learning_rate": 2.6561953744684264e-05,
      "loss": 1.0784,
      "num_input_tokens_seen": 41446032,
      "step": 71450
    },
    {
      "epoch": 10.64268692284778,
      "grad_norm": 12.8125,
      "learning_rate": 2.655871065833518e-05,
      "loss": 1.6342,
      "num_input_tokens_seen": 41448944,
      "step": 71455
    },
    {
      "epoch": 10.64343163538874,
      "grad_norm": 3.203125,
      "learning_rate": 2.655546754565326e-05,
      "loss": 0.8749,
      "num_input_tokens_seen": 41451824,
      "step": 71460
    },
    {
      "epoch": 10.6441763479297,
      "grad_norm": 15.125,
      "learning_rate": 2.6552224406693293e-05,
      "loss": 0.9879,
      "num_input_tokens_seen": 41454768,
      "step": 71465
    },
    {
      "epoch": 10.644921060470658,
      "grad_norm": 9.5625,
      "learning_rate": 2.6548981241510073e-05,
      "loss": 0.6933,
      "num_input_tokens_seen": 41457552,
      "step": 71470
    },
    {
      "epoch": 10.645665773011618,
      "grad_norm": 7.34375,
      "learning_rate": 2.654573805015839e-05,
      "loss": 0.9499,
      "num_input_tokens_seen": 41460368,
      "step": 71475
    },
    {
      "epoch": 10.646410485552577,
      "grad_norm": 5.34375,
      "learning_rate": 2.654249483269303e-05,
      "loss": 1.0886,
      "num_input_tokens_seen": 41463248,
      "step": 71480
    },
    {
      "epoch": 10.647155198093536,
      "grad_norm": 13.25,
      "learning_rate": 2.65392515891688e-05,
      "loss": 0.9155,
      "num_input_tokens_seen": 41466096,
      "step": 71485
    },
    {
      "epoch": 10.647899910634495,
      "grad_norm": 13.0625,
      "learning_rate": 2.6536008319640466e-05,
      "loss": 1.221,
      "num_input_tokens_seen": 41469456,
      "step": 71490
    },
    {
      "epoch": 10.648644623175453,
      "grad_norm": 11.4375,
      "learning_rate": 2.6532765024162837e-05,
      "loss": 0.959,
      "num_input_tokens_seen": 41472176,
      "step": 71495
    },
    {
      "epoch": 10.649389335716414,
      "grad_norm": 5.65625,
      "learning_rate": 2.6529521702790705e-05,
      "loss": 1.0313,
      "num_input_tokens_seen": 41475152,
      "step": 71500
    },
    {
      "epoch": 10.650134048257373,
      "grad_norm": 7.40625,
      "learning_rate": 2.6526278355578848e-05,
      "loss": 0.7985,
      "num_input_tokens_seen": 41478032,
      "step": 71505
    },
    {
      "epoch": 10.650878760798332,
      "grad_norm": 9.625,
      "learning_rate": 2.6523034982582078e-05,
      "loss": 1.2549,
      "num_input_tokens_seen": 41481392,
      "step": 71510
    },
    {
      "epoch": 10.65162347333929,
      "grad_norm": 7.90625,
      "learning_rate": 2.6519791583855174e-05,
      "loss": 0.8671,
      "num_input_tokens_seen": 41484304,
      "step": 71515
    },
    {
      "epoch": 10.65236818588025,
      "grad_norm": 9.3125,
      "learning_rate": 2.6516548159452943e-05,
      "loss": 0.9728,
      "num_input_tokens_seen": 41487440,
      "step": 71520
    },
    {
      "epoch": 10.65311289842121,
      "grad_norm": 8.8125,
      "learning_rate": 2.651330470943017e-05,
      "loss": 1.2084,
      "num_input_tokens_seen": 41490544,
      "step": 71525
    },
    {
      "epoch": 10.653857610962168,
      "grad_norm": 9.5,
      "learning_rate": 2.651006123384165e-05,
      "loss": 0.9756,
      "num_input_tokens_seen": 41493584,
      "step": 71530
    },
    {
      "epoch": 10.654602323503127,
      "grad_norm": 5.84375,
      "learning_rate": 2.6506817732742173e-05,
      "loss": 1.2154,
      "num_input_tokens_seen": 41496432,
      "step": 71535
    },
    {
      "epoch": 10.655347036044088,
      "grad_norm": 11.375,
      "learning_rate": 2.6503574206186553e-05,
      "loss": 1.0465,
      "num_input_tokens_seen": 41499184,
      "step": 71540
    },
    {
      "epoch": 10.656091748585046,
      "grad_norm": 10.9375,
      "learning_rate": 2.6500330654229573e-05,
      "loss": 1.1913,
      "num_input_tokens_seen": 41502000,
      "step": 71545
    },
    {
      "epoch": 10.656836461126005,
      "grad_norm": 12.4375,
      "learning_rate": 2.649708707692603e-05,
      "loss": 0.9727,
      "num_input_tokens_seen": 41504752,
      "step": 71550
    },
    {
      "epoch": 10.657581173666964,
      "grad_norm": 8.625,
      "learning_rate": 2.6493843474330727e-05,
      "loss": 0.9042,
      "num_input_tokens_seen": 41507664,
      "step": 71555
    },
    {
      "epoch": 10.658325886207924,
      "grad_norm": 13.375,
      "learning_rate": 2.649059984649845e-05,
      "loss": 1.0386,
      "num_input_tokens_seen": 41510288,
      "step": 71560
    },
    {
      "epoch": 10.659070598748883,
      "grad_norm": 16.875,
      "learning_rate": 2.6487356193484002e-05,
      "loss": 1.0973,
      "num_input_tokens_seen": 41513456,
      "step": 71565
    },
    {
      "epoch": 10.659815311289842,
      "grad_norm": 10.25,
      "learning_rate": 2.6484112515342186e-05,
      "loss": 1.0483,
      "num_input_tokens_seen": 41516464,
      "step": 71570
    },
    {
      "epoch": 10.6605600238308,
      "grad_norm": 10.0625,
      "learning_rate": 2.6480868812127795e-05,
      "loss": 1.0147,
      "num_input_tokens_seen": 41519568,
      "step": 71575
    },
    {
      "epoch": 10.661304736371761,
      "grad_norm": 15.625,
      "learning_rate": 2.6477625083895636e-05,
      "loss": 0.9941,
      "num_input_tokens_seen": 41522192,
      "step": 71580
    },
    {
      "epoch": 10.66204944891272,
      "grad_norm": 7.0,
      "learning_rate": 2.6474381330700497e-05,
      "loss": 1.2177,
      "num_input_tokens_seen": 41524880,
      "step": 71585
    },
    {
      "epoch": 10.662794161453679,
      "grad_norm": 16.625,
      "learning_rate": 2.6471137552597193e-05,
      "loss": 1.0265,
      "num_input_tokens_seen": 41527664,
      "step": 71590
    },
    {
      "epoch": 10.663538873994638,
      "grad_norm": 10.75,
      "learning_rate": 2.646789374964051e-05,
      "loss": 1.2768,
      "num_input_tokens_seen": 41530704,
      "step": 71595
    },
    {
      "epoch": 10.664283586535598,
      "grad_norm": 9.8125,
      "learning_rate": 2.6464649921885247e-05,
      "loss": 0.8279,
      "num_input_tokens_seen": 41533520,
      "step": 71600
    },
    {
      "epoch": 10.665028299076557,
      "grad_norm": 12.75,
      "learning_rate": 2.646140606938622e-05,
      "loss": 1.2982,
      "num_input_tokens_seen": 41536880,
      "step": 71605
    },
    {
      "epoch": 10.665773011617516,
      "grad_norm": 12.625,
      "learning_rate": 2.6458162192198223e-05,
      "loss": 1.0187,
      "num_input_tokens_seen": 41539856,
      "step": 71610
    },
    {
      "epoch": 10.666517724158474,
      "grad_norm": 15.75,
      "learning_rate": 2.6454918290376053e-05,
      "loss": 1.0857,
      "num_input_tokens_seen": 41542928,
      "step": 71615
    },
    {
      "epoch": 10.667262436699435,
      "grad_norm": 15.1875,
      "learning_rate": 2.645167436397452e-05,
      "loss": 1.2762,
      "num_input_tokens_seen": 41546224,
      "step": 71620
    },
    {
      "epoch": 10.668007149240394,
      "grad_norm": 8.0625,
      "learning_rate": 2.644843041304843e-05,
      "loss": 1.1803,
      "num_input_tokens_seen": 41548880,
      "step": 71625
    },
    {
      "epoch": 10.668751861781352,
      "grad_norm": 6.90625,
      "learning_rate": 2.6445186437652577e-05,
      "loss": 0.7601,
      "num_input_tokens_seen": 41551856,
      "step": 71630
    },
    {
      "epoch": 10.669496574322311,
      "grad_norm": 7.40625,
      "learning_rate": 2.644194243784176e-05,
      "loss": 1.0449,
      "num_input_tokens_seen": 41555216,
      "step": 71635
    },
    {
      "epoch": 10.67024128686327,
      "grad_norm": 6.125,
      "learning_rate": 2.64386984136708e-05,
      "loss": 1.0821,
      "num_input_tokens_seen": 41558384,
      "step": 71640
    },
    {
      "epoch": 10.67098599940423,
      "grad_norm": 11.3125,
      "learning_rate": 2.6435454365194483e-05,
      "loss": 1.0208,
      "num_input_tokens_seen": 41561232,
      "step": 71645
    },
    {
      "epoch": 10.67173071194519,
      "grad_norm": 9.5625,
      "learning_rate": 2.6432210292467634e-05,
      "loss": 1.1804,
      "num_input_tokens_seen": 41564016,
      "step": 71650
    },
    {
      "epoch": 10.672475424486148,
      "grad_norm": 8.25,
      "learning_rate": 2.642896619554504e-05,
      "loss": 1.104,
      "num_input_tokens_seen": 41566832,
      "step": 71655
    },
    {
      "epoch": 10.673220137027108,
      "grad_norm": 15.3125,
      "learning_rate": 2.6425722074481525e-05,
      "loss": 0.9787,
      "num_input_tokens_seen": 41569616,
      "step": 71660
    },
    {
      "epoch": 10.673964849568067,
      "grad_norm": 6.625,
      "learning_rate": 2.6422477929331875e-05,
      "loss": 1.1755,
      "num_input_tokens_seen": 41572240,
      "step": 71665
    },
    {
      "epoch": 10.674709562109026,
      "grad_norm": 6.78125,
      "learning_rate": 2.6419233760150907e-05,
      "loss": 1.1168,
      "num_input_tokens_seen": 41575216,
      "step": 71670
    },
    {
      "epoch": 10.675454274649985,
      "grad_norm": 11.3125,
      "learning_rate": 2.6415989566993425e-05,
      "loss": 1.231,
      "num_input_tokens_seen": 41578032,
      "step": 71675
    },
    {
      "epoch": 10.676198987190944,
      "grad_norm": 13.0,
      "learning_rate": 2.6412745349914242e-05,
      "loss": 0.8164,
      "num_input_tokens_seen": 41580944,
      "step": 71680
    },
    {
      "epoch": 10.676943699731904,
      "grad_norm": 12.1875,
      "learning_rate": 2.6409501108968164e-05,
      "loss": 1.1738,
      "num_input_tokens_seen": 41583952,
      "step": 71685
    },
    {
      "epoch": 10.677688412272863,
      "grad_norm": 16.125,
      "learning_rate": 2.6406256844209998e-05,
      "loss": 0.9555,
      "num_input_tokens_seen": 41586608,
      "step": 71690
    },
    {
      "epoch": 10.678433124813822,
      "grad_norm": 14.375,
      "learning_rate": 2.6403012555694546e-05,
      "loss": 0.835,
      "num_input_tokens_seen": 41589520,
      "step": 71695
    },
    {
      "epoch": 10.67917783735478,
      "grad_norm": 5.125,
      "learning_rate": 2.639976824347663e-05,
      "loss": 1.045,
      "num_input_tokens_seen": 41592464,
      "step": 71700
    },
    {
      "epoch": 10.67992254989574,
      "grad_norm": 10.6875,
      "learning_rate": 2.639652390761105e-05,
      "loss": 0.9552,
      "num_input_tokens_seen": 41595376,
      "step": 71705
    },
    {
      "epoch": 10.6806672624367,
      "grad_norm": 5.875,
      "learning_rate": 2.639327954815261e-05,
      "loss": 1.2887,
      "num_input_tokens_seen": 41598256,
      "step": 71710
    },
    {
      "epoch": 10.681411974977658,
      "grad_norm": 13.75,
      "learning_rate": 2.6390035165156136e-05,
      "loss": 1.0901,
      "num_input_tokens_seen": 41601072,
      "step": 71715
    },
    {
      "epoch": 10.682156687518617,
      "grad_norm": 17.0,
      "learning_rate": 2.6386790758676432e-05,
      "loss": 0.9881,
      "num_input_tokens_seen": 41603920,
      "step": 71720
    },
    {
      "epoch": 10.682901400059578,
      "grad_norm": 8.375,
      "learning_rate": 2.6383546328768305e-05,
      "loss": 1.1713,
      "num_input_tokens_seen": 41607120,
      "step": 71725
    },
    {
      "epoch": 10.683646112600536,
      "grad_norm": 15.3125,
      "learning_rate": 2.6380301875486568e-05,
      "loss": 1.1216,
      "num_input_tokens_seen": 41609968,
      "step": 71730
    },
    {
      "epoch": 10.684390825141495,
      "grad_norm": 8.125,
      "learning_rate": 2.637705739888603e-05,
      "loss": 1.4357,
      "num_input_tokens_seen": 41612944,
      "step": 71735
    },
    {
      "epoch": 10.685135537682454,
      "grad_norm": 6.53125,
      "learning_rate": 2.6373812899021516e-05,
      "loss": 1.0601,
      "num_input_tokens_seen": 41615728,
      "step": 71740
    },
    {
      "epoch": 10.685880250223414,
      "grad_norm": 7.75,
      "learning_rate": 2.6370568375947825e-05,
      "loss": 1.056,
      "num_input_tokens_seen": 41618896,
      "step": 71745
    },
    {
      "epoch": 10.686624962764373,
      "grad_norm": 12.1875,
      "learning_rate": 2.636732382971977e-05,
      "loss": 0.9196,
      "num_input_tokens_seen": 41621872,
      "step": 71750
    },
    {
      "epoch": 10.687369675305332,
      "grad_norm": 7.28125,
      "learning_rate": 2.6364079260392178e-05,
      "loss": 1.6198,
      "num_input_tokens_seen": 41624880,
      "step": 71755
    },
    {
      "epoch": 10.68811438784629,
      "grad_norm": 8.25,
      "learning_rate": 2.6360834668019845e-05,
      "loss": 1.0375,
      "num_input_tokens_seen": 41627632,
      "step": 71760
    },
    {
      "epoch": 10.688859100387251,
      "grad_norm": 11.5,
      "learning_rate": 2.6357590052657595e-05,
      "loss": 1.0282,
      "num_input_tokens_seen": 41630672,
      "step": 71765
    },
    {
      "epoch": 10.68960381292821,
      "grad_norm": 9.0625,
      "learning_rate": 2.6354345414360236e-05,
      "loss": 1.1498,
      "num_input_tokens_seen": 41633584,
      "step": 71770
    },
    {
      "epoch": 10.690348525469169,
      "grad_norm": 20.875,
      "learning_rate": 2.6351100753182594e-05,
      "loss": 1.0642,
      "num_input_tokens_seen": 41636272,
      "step": 71775
    },
    {
      "epoch": 10.691093238010128,
      "grad_norm": 12.375,
      "learning_rate": 2.6347856069179483e-05,
      "loss": 1.3139,
      "num_input_tokens_seen": 41638992,
      "step": 71780
    },
    {
      "epoch": 10.691837950551088,
      "grad_norm": 4.375,
      "learning_rate": 2.6344611362405708e-05,
      "loss": 0.6457,
      "num_input_tokens_seen": 41642000,
      "step": 71785
    },
    {
      "epoch": 10.692582663092047,
      "grad_norm": 10.875,
      "learning_rate": 2.634136663291609e-05,
      "loss": 1.1179,
      "num_input_tokens_seen": 41645040,
      "step": 71790
    },
    {
      "epoch": 10.693327375633006,
      "grad_norm": 10.75,
      "learning_rate": 2.6338121880765447e-05,
      "loss": 1.0314,
      "num_input_tokens_seen": 41647856,
      "step": 71795
    },
    {
      "epoch": 10.694072088173964,
      "grad_norm": 18.375,
      "learning_rate": 2.6334877106008594e-05,
      "loss": 1.2575,
      "num_input_tokens_seen": 41650736,
      "step": 71800
    },
    {
      "epoch": 10.694816800714925,
      "grad_norm": 17.125,
      "learning_rate": 2.633163230870035e-05,
      "loss": 1.0542,
      "num_input_tokens_seen": 41653584,
      "step": 71805
    },
    {
      "epoch": 10.695561513255884,
      "grad_norm": 6.28125,
      "learning_rate": 2.632838748889553e-05,
      "loss": 1.1205,
      "num_input_tokens_seen": 41656368,
      "step": 71810
    },
    {
      "epoch": 10.696306225796842,
      "grad_norm": 18.0,
      "learning_rate": 2.6325142646648958e-05,
      "loss": 0.9606,
      "num_input_tokens_seen": 41659216,
      "step": 71815
    },
    {
      "epoch": 10.697050938337801,
      "grad_norm": 10.0625,
      "learning_rate": 2.632189778201544e-05,
      "loss": 0.9377,
      "num_input_tokens_seen": 41662288,
      "step": 71820
    },
    {
      "epoch": 10.69779565087876,
      "grad_norm": 5.96875,
      "learning_rate": 2.631865289504981e-05,
      "loss": 1.1696,
      "num_input_tokens_seen": 41665104,
      "step": 71825
    },
    {
      "epoch": 10.69854036341972,
      "grad_norm": 12.75,
      "learning_rate": 2.631540798580688e-05,
      "loss": 1.0349,
      "num_input_tokens_seen": 41667856,
      "step": 71830
    },
    {
      "epoch": 10.69928507596068,
      "grad_norm": 11.375,
      "learning_rate": 2.6312163054341464e-05,
      "loss": 1.1894,
      "num_input_tokens_seen": 41670704,
      "step": 71835
    },
    {
      "epoch": 10.700029788501638,
      "grad_norm": 13.4375,
      "learning_rate": 2.6308918100708386e-05,
      "loss": 1.2624,
      "num_input_tokens_seen": 41673424,
      "step": 71840
    },
    {
      "epoch": 10.700774501042597,
      "grad_norm": 6.25,
      "learning_rate": 2.6305673124962466e-05,
      "loss": 1.0537,
      "num_input_tokens_seen": 41676400,
      "step": 71845
    },
    {
      "epoch": 10.701519213583557,
      "grad_norm": 16.25,
      "learning_rate": 2.6302428127158535e-05,
      "loss": 1.1007,
      "num_input_tokens_seen": 41679088,
      "step": 71850
    },
    {
      "epoch": 10.702263926124516,
      "grad_norm": 8.1875,
      "learning_rate": 2.6299183107351395e-05,
      "loss": 0.9788,
      "num_input_tokens_seen": 41682096,
      "step": 71855
    },
    {
      "epoch": 10.703008638665475,
      "grad_norm": 10.0625,
      "learning_rate": 2.629593806559589e-05,
      "loss": 1.1035,
      "num_input_tokens_seen": 41684976,
      "step": 71860
    },
    {
      "epoch": 10.703753351206434,
      "grad_norm": 8.4375,
      "learning_rate": 2.629269300194681e-05,
      "loss": 1.2594,
      "num_input_tokens_seen": 41687856,
      "step": 71865
    },
    {
      "epoch": 10.704498063747394,
      "grad_norm": 5.40625,
      "learning_rate": 2.6289447916459005e-05,
      "loss": 0.8729,
      "num_input_tokens_seen": 41690864,
      "step": 71870
    },
    {
      "epoch": 10.705242776288353,
      "grad_norm": 7.6875,
      "learning_rate": 2.628620280918729e-05,
      "loss": 1.1537,
      "num_input_tokens_seen": 41693616,
      "step": 71875
    },
    {
      "epoch": 10.705987488829312,
      "grad_norm": 13.25,
      "learning_rate": 2.6282957680186476e-05,
      "loss": 1.1725,
      "num_input_tokens_seen": 41696624,
      "step": 71880
    },
    {
      "epoch": 10.70673220137027,
      "grad_norm": 22.75,
      "learning_rate": 2.6279712529511406e-05,
      "loss": 1.2713,
      "num_input_tokens_seen": 41699696,
      "step": 71885
    },
    {
      "epoch": 10.707476913911231,
      "grad_norm": 10.75,
      "learning_rate": 2.6276467357216895e-05,
      "loss": 1.2657,
      "num_input_tokens_seen": 41702800,
      "step": 71890
    },
    {
      "epoch": 10.70822162645219,
      "grad_norm": 8.1875,
      "learning_rate": 2.627322216335776e-05,
      "loss": 0.9499,
      "num_input_tokens_seen": 41705392,
      "step": 71895
    },
    {
      "epoch": 10.708966338993148,
      "grad_norm": 13.375,
      "learning_rate": 2.6269976947988834e-05,
      "loss": 0.8734,
      "num_input_tokens_seen": 41708464,
      "step": 71900
    },
    {
      "epoch": 10.709711051534107,
      "grad_norm": 11.0625,
      "learning_rate": 2.626673171116493e-05,
      "loss": 1.2652,
      "num_input_tokens_seen": 41711664,
      "step": 71905
    },
    {
      "epoch": 10.710455764075068,
      "grad_norm": 9.125,
      "learning_rate": 2.626348645294089e-05,
      "loss": 1.0934,
      "num_input_tokens_seen": 41714448,
      "step": 71910
    },
    {
      "epoch": 10.711200476616026,
      "grad_norm": 10.4375,
      "learning_rate": 2.6260241173371525e-05,
      "loss": 1.4234,
      "num_input_tokens_seen": 41717328,
      "step": 71915
    },
    {
      "epoch": 10.711945189156985,
      "grad_norm": 10.6875,
      "learning_rate": 2.625699587251167e-05,
      "loss": 1.1665,
      "num_input_tokens_seen": 41720144,
      "step": 71920
    },
    {
      "epoch": 10.712689901697944,
      "grad_norm": 16.375,
      "learning_rate": 2.6253750550416144e-05,
      "loss": 1.1598,
      "num_input_tokens_seen": 41723120,
      "step": 71925
    },
    {
      "epoch": 10.713434614238905,
      "grad_norm": 12.625,
      "learning_rate": 2.6250505207139782e-05,
      "loss": 1.1726,
      "num_input_tokens_seen": 41725680,
      "step": 71930
    },
    {
      "epoch": 10.714179326779863,
      "grad_norm": 7.5,
      "learning_rate": 2.62472598427374e-05,
      "loss": 1.166,
      "num_input_tokens_seen": 41728464,
      "step": 71935
    },
    {
      "epoch": 10.714924039320822,
      "grad_norm": 13.1875,
      "learning_rate": 2.624401445726383e-05,
      "loss": 0.8466,
      "num_input_tokens_seen": 41731280,
      "step": 71940
    },
    {
      "epoch": 10.71566875186178,
      "grad_norm": 9.75,
      "learning_rate": 2.6240769050773906e-05,
      "loss": 1.1657,
      "num_input_tokens_seen": 41734320,
      "step": 71945
    },
    {
      "epoch": 10.716413464402741,
      "grad_norm": 10.5625,
      "learning_rate": 2.6237523623322446e-05,
      "loss": 0.9344,
      "num_input_tokens_seen": 41737424,
      "step": 71950
    },
    {
      "epoch": 10.7171581769437,
      "grad_norm": 7.8125,
      "learning_rate": 2.6234278174964288e-05,
      "loss": 0.7757,
      "num_input_tokens_seen": 41740368,
      "step": 71955
    },
    {
      "epoch": 10.717902889484659,
      "grad_norm": 14.1875,
      "learning_rate": 2.6231032705754243e-05,
      "loss": 0.8379,
      "num_input_tokens_seen": 41743120,
      "step": 71960
    },
    {
      "epoch": 10.718647602025618,
      "grad_norm": 8.9375,
      "learning_rate": 2.6227787215747156e-05,
      "loss": 1.3585,
      "num_input_tokens_seen": 41745776,
      "step": 71965
    },
    {
      "epoch": 10.719392314566576,
      "grad_norm": 9.25,
      "learning_rate": 2.6224541704997856e-05,
      "loss": 1.0331,
      "num_input_tokens_seen": 41748944,
      "step": 71970
    },
    {
      "epoch": 10.720137027107537,
      "grad_norm": 4.96875,
      "learning_rate": 2.6221296173561166e-05,
      "loss": 1.0911,
      "num_input_tokens_seen": 41751984,
      "step": 71975
    },
    {
      "epoch": 10.720881739648496,
      "grad_norm": 8.5625,
      "learning_rate": 2.6218050621491925e-05,
      "loss": 1.2009,
      "num_input_tokens_seen": 41754704,
      "step": 71980
    },
    {
      "epoch": 10.721626452189454,
      "grad_norm": 10.0,
      "learning_rate": 2.6214805048844947e-05,
      "loss": 1.2375,
      "num_input_tokens_seen": 41757808,
      "step": 71985
    },
    {
      "epoch": 10.722371164730415,
      "grad_norm": 10.9375,
      "learning_rate": 2.621155945567508e-05,
      "loss": 1.0794,
      "num_input_tokens_seen": 41760816,
      "step": 71990
    },
    {
      "epoch": 10.723115877271374,
      "grad_norm": 19.375,
      "learning_rate": 2.620831384203714e-05,
      "loss": 1.1596,
      "num_input_tokens_seen": 41763472,
      "step": 71995
    },
    {
      "epoch": 10.723860589812332,
      "grad_norm": 10.0,
      "learning_rate": 2.6205068207985965e-05,
      "loss": 1.082,
      "num_input_tokens_seen": 41766416,
      "step": 72000
    },
    {
      "epoch": 10.724605302353291,
      "grad_norm": 13.5625,
      "learning_rate": 2.6201822553576394e-05,
      "loss": 0.756,
      "num_input_tokens_seen": 41769424,
      "step": 72005
    },
    {
      "epoch": 10.72535001489425,
      "grad_norm": 21.75,
      "learning_rate": 2.619857687886325e-05,
      "loss": 1.5158,
      "num_input_tokens_seen": 41772272,
      "step": 72010
    },
    {
      "epoch": 10.72609472743521,
      "grad_norm": 13.0625,
      "learning_rate": 2.6195331183901374e-05,
      "loss": 1.3925,
      "num_input_tokens_seen": 41776240,
      "step": 72015
    },
    {
      "epoch": 10.72683943997617,
      "grad_norm": 5.84375,
      "learning_rate": 2.6192085468745585e-05,
      "loss": 1.1749,
      "num_input_tokens_seen": 41779504,
      "step": 72020
    },
    {
      "epoch": 10.727584152517128,
      "grad_norm": 15.6875,
      "learning_rate": 2.6188839733450727e-05,
      "loss": 1.1021,
      "num_input_tokens_seen": 41782512,
      "step": 72025
    },
    {
      "epoch": 10.728328865058087,
      "grad_norm": 9.625,
      "learning_rate": 2.6185593978071627e-05,
      "loss": 1.223,
      "num_input_tokens_seen": 41785488,
      "step": 72030
    },
    {
      "epoch": 10.729073577599047,
      "grad_norm": 9.8125,
      "learning_rate": 2.6182348202663122e-05,
      "loss": 1.0571,
      "num_input_tokens_seen": 41788272,
      "step": 72035
    },
    {
      "epoch": 10.729818290140006,
      "grad_norm": 11.25,
      "learning_rate": 2.617910240728004e-05,
      "loss": 1.2776,
      "num_input_tokens_seen": 41791184,
      "step": 72040
    },
    {
      "epoch": 10.730563002680965,
      "grad_norm": 13.3125,
      "learning_rate": 2.6175856591977226e-05,
      "loss": 1.1824,
      "num_input_tokens_seen": 41793904,
      "step": 72045
    },
    {
      "epoch": 10.731307715221924,
      "grad_norm": 9.125,
      "learning_rate": 2.6172610756809517e-05,
      "loss": 1.4145,
      "num_input_tokens_seen": 41796592,
      "step": 72050
    },
    {
      "epoch": 10.732052427762884,
      "grad_norm": 12.625,
      "learning_rate": 2.6169364901831732e-05,
      "loss": 1.0982,
      "num_input_tokens_seen": 41799536,
      "step": 72055
    },
    {
      "epoch": 10.732797140303843,
      "grad_norm": 8.6875,
      "learning_rate": 2.6166119027098724e-05,
      "loss": 0.7678,
      "num_input_tokens_seen": 41802288,
      "step": 72060
    },
    {
      "epoch": 10.733541852844802,
      "grad_norm": 13.4375,
      "learning_rate": 2.6162873132665315e-05,
      "loss": 1.2049,
      "num_input_tokens_seen": 41804976,
      "step": 72065
    },
    {
      "epoch": 10.73428656538576,
      "grad_norm": 5.625,
      "learning_rate": 2.6159627218586345e-05,
      "loss": 1.3039,
      "num_input_tokens_seen": 41808144,
      "step": 72070
    },
    {
      "epoch": 10.735031277926721,
      "grad_norm": 10.0625,
      "learning_rate": 2.6156381284916653e-05,
      "loss": 1.1612,
      "num_input_tokens_seen": 41811088,
      "step": 72075
    },
    {
      "epoch": 10.73577599046768,
      "grad_norm": 16.375,
      "learning_rate": 2.615313533171107e-05,
      "loss": 1.0807,
      "num_input_tokens_seen": 41813776,
      "step": 72080
    },
    {
      "epoch": 10.736520703008638,
      "grad_norm": 15.625,
      "learning_rate": 2.6149889359024447e-05,
      "loss": 0.9325,
      "num_input_tokens_seen": 41816592,
      "step": 72085
    },
    {
      "epoch": 10.737265415549597,
      "grad_norm": 12.25,
      "learning_rate": 2.6146643366911612e-05,
      "loss": 0.9082,
      "num_input_tokens_seen": 41819824,
      "step": 72090
    },
    {
      "epoch": 10.738010128090558,
      "grad_norm": 13.9375,
      "learning_rate": 2.614339735542739e-05,
      "loss": 1.1142,
      "num_input_tokens_seen": 41822704,
      "step": 72095
    },
    {
      "epoch": 10.738754840631517,
      "grad_norm": 9.625,
      "learning_rate": 2.6140151324626644e-05,
      "loss": 0.8834,
      "num_input_tokens_seen": 41825328,
      "step": 72100
    },
    {
      "epoch": 10.739499553172475,
      "grad_norm": 13.3125,
      "learning_rate": 2.61369052745642e-05,
      "loss": 1.2322,
      "num_input_tokens_seen": 41828240,
      "step": 72105
    },
    {
      "epoch": 10.740244265713434,
      "grad_norm": 14.125,
      "learning_rate": 2.6133659205294892e-05,
      "loss": 0.9159,
      "num_input_tokens_seen": 41830928,
      "step": 72110
    },
    {
      "epoch": 10.740988978254395,
      "grad_norm": 6.6875,
      "learning_rate": 2.6130413116873557e-05,
      "loss": 1.1274,
      "num_input_tokens_seen": 41834256,
      "step": 72115
    },
    {
      "epoch": 10.741733690795353,
      "grad_norm": 5.71875,
      "learning_rate": 2.6127167009355058e-05,
      "loss": 0.7179,
      "num_input_tokens_seen": 41837040,
      "step": 72120
    },
    {
      "epoch": 10.742478403336312,
      "grad_norm": 19.125,
      "learning_rate": 2.6123920882794208e-05,
      "loss": 1.5153,
      "num_input_tokens_seen": 41839984,
      "step": 72125
    },
    {
      "epoch": 10.74322311587727,
      "grad_norm": 5.34375,
      "learning_rate": 2.6120674737245854e-05,
      "loss": 0.7916,
      "num_input_tokens_seen": 41843024,
      "step": 72130
    },
    {
      "epoch": 10.743967828418231,
      "grad_norm": 7.5625,
      "learning_rate": 2.611742857276484e-05,
      "loss": 1.1339,
      "num_input_tokens_seen": 41846064,
      "step": 72135
    },
    {
      "epoch": 10.74471254095919,
      "grad_norm": 6.46875,
      "learning_rate": 2.6114182389406012e-05,
      "loss": 0.8127,
      "num_input_tokens_seen": 41848848,
      "step": 72140
    },
    {
      "epoch": 10.745457253500149,
      "grad_norm": 13.0625,
      "learning_rate": 2.6110936187224205e-05,
      "loss": 1.5712,
      "num_input_tokens_seen": 41851952,
      "step": 72145
    },
    {
      "epoch": 10.746201966041108,
      "grad_norm": 9.9375,
      "learning_rate": 2.610768996627426e-05,
      "loss": 1.2302,
      "num_input_tokens_seen": 41854800,
      "step": 72150
    },
    {
      "epoch": 10.746946678582066,
      "grad_norm": 11.125,
      "learning_rate": 2.6104443726611016e-05,
      "loss": 0.86,
      "num_input_tokens_seen": 41857680,
      "step": 72155
    },
    {
      "epoch": 10.747691391123027,
      "grad_norm": 6.3125,
      "learning_rate": 2.610119746828932e-05,
      "loss": 0.9792,
      "num_input_tokens_seen": 41860720,
      "step": 72160
    },
    {
      "epoch": 10.748436103663986,
      "grad_norm": 20.5,
      "learning_rate": 2.6097951191364007e-05,
      "loss": 1.0067,
      "num_input_tokens_seen": 41863696,
      "step": 72165
    },
    {
      "epoch": 10.749180816204944,
      "grad_norm": 12.75,
      "learning_rate": 2.6094704895889927e-05,
      "loss": 1.0422,
      "num_input_tokens_seen": 41866480,
      "step": 72170
    },
    {
      "epoch": 10.749925528745905,
      "grad_norm": 16.25,
      "learning_rate": 2.609145858192192e-05,
      "loss": 1.2685,
      "num_input_tokens_seen": 41869232,
      "step": 72175
    },
    {
      "epoch": 10.750670241286864,
      "grad_norm": 12.375,
      "learning_rate": 2.608821224951483e-05,
      "loss": 1.0468,
      "num_input_tokens_seen": 41872208,
      "step": 72180
    },
    {
      "epoch": 10.751414953827823,
      "grad_norm": 6.09375,
      "learning_rate": 2.608496589872351e-05,
      "loss": 1.0593,
      "num_input_tokens_seen": 41875184,
      "step": 72185
    },
    {
      "epoch": 10.752159666368781,
      "grad_norm": 10.0625,
      "learning_rate": 2.6081719529602776e-05,
      "loss": 1.2134,
      "num_input_tokens_seen": 41878000,
      "step": 72190
    },
    {
      "epoch": 10.75290437890974,
      "grad_norm": 15.75,
      "learning_rate": 2.6078473142207498e-05,
      "loss": 0.8052,
      "num_input_tokens_seen": 41881008,
      "step": 72195
    },
    {
      "epoch": 10.7536490914507,
      "grad_norm": 8.625,
      "learning_rate": 2.607522673659251e-05,
      "loss": 1.3798,
      "num_input_tokens_seen": 41883696,
      "step": 72200
    },
    {
      "epoch": 10.75439380399166,
      "grad_norm": 11.125,
      "learning_rate": 2.6071980312812665e-05,
      "loss": 1.0436,
      "num_input_tokens_seen": 41886928,
      "step": 72205
    },
    {
      "epoch": 10.755138516532618,
      "grad_norm": 13.3125,
      "learning_rate": 2.6068733870922797e-05,
      "loss": 0.8169,
      "num_input_tokens_seen": 41889776,
      "step": 72210
    },
    {
      "epoch": 10.755883229073577,
      "grad_norm": 26.625,
      "learning_rate": 2.606548741097776e-05,
      "loss": 1.5283,
      "num_input_tokens_seen": 41892528,
      "step": 72215
    },
    {
      "epoch": 10.756627941614537,
      "grad_norm": 11.25,
      "learning_rate": 2.6062240933032394e-05,
      "loss": 1.2185,
      "num_input_tokens_seen": 41895504,
      "step": 72220
    },
    {
      "epoch": 10.757372654155496,
      "grad_norm": 8.625,
      "learning_rate": 2.6058994437141554e-05,
      "loss": 1.1072,
      "num_input_tokens_seen": 41898256,
      "step": 72225
    },
    {
      "epoch": 10.758117366696455,
      "grad_norm": 19.125,
      "learning_rate": 2.605574792336007e-05,
      "loss": 0.829,
      "num_input_tokens_seen": 41901328,
      "step": 72230
    },
    {
      "epoch": 10.758862079237414,
      "grad_norm": 6.90625,
      "learning_rate": 2.6052501391742802e-05,
      "loss": 1.0554,
      "num_input_tokens_seen": 41904112,
      "step": 72235
    },
    {
      "epoch": 10.759606791778374,
      "grad_norm": 10.8125,
      "learning_rate": 2.604925484234459e-05,
      "loss": 1.1712,
      "num_input_tokens_seen": 41907120,
      "step": 72240
    },
    {
      "epoch": 10.760351504319333,
      "grad_norm": 9.375,
      "learning_rate": 2.6046008275220286e-05,
      "loss": 1.2271,
      "num_input_tokens_seen": 41909872,
      "step": 72245
    },
    {
      "epoch": 10.761096216860292,
      "grad_norm": 16.625,
      "learning_rate": 2.604276169042473e-05,
      "loss": 1.1692,
      "num_input_tokens_seen": 41912976,
      "step": 72250
    },
    {
      "epoch": 10.76184092940125,
      "grad_norm": 9.3125,
      "learning_rate": 2.6039515088012783e-05,
      "loss": 1.2023,
      "num_input_tokens_seen": 41915632,
      "step": 72255
    },
    {
      "epoch": 10.762585641942211,
      "grad_norm": 15.75,
      "learning_rate": 2.6036268468039282e-05,
      "loss": 0.929,
      "num_input_tokens_seen": 41918352,
      "step": 72260
    },
    {
      "epoch": 10.76333035448317,
      "grad_norm": 10.3125,
      "learning_rate": 2.603302183055908e-05,
      "loss": 1.2974,
      "num_input_tokens_seen": 41921072,
      "step": 72265
    },
    {
      "epoch": 10.764075067024129,
      "grad_norm": 10.0625,
      "learning_rate": 2.6029775175627024e-05,
      "loss": 0.8192,
      "num_input_tokens_seen": 41923728,
      "step": 72270
    },
    {
      "epoch": 10.764819779565087,
      "grad_norm": 6.25,
      "learning_rate": 2.602652850329796e-05,
      "loss": 1.0911,
      "num_input_tokens_seen": 41926768,
      "step": 72275
    },
    {
      "epoch": 10.765564492106048,
      "grad_norm": 6.625,
      "learning_rate": 2.6023281813626737e-05,
      "loss": 0.9855,
      "num_input_tokens_seen": 41929488,
      "step": 72280
    },
    {
      "epoch": 10.766309204647007,
      "grad_norm": 8.25,
      "learning_rate": 2.602003510666822e-05,
      "loss": 1.2753,
      "num_input_tokens_seen": 41932432,
      "step": 72285
    },
    {
      "epoch": 10.767053917187965,
      "grad_norm": 10.125,
      "learning_rate": 2.6016788382477238e-05,
      "loss": 1.1846,
      "num_input_tokens_seen": 41935472,
      "step": 72290
    },
    {
      "epoch": 10.767798629728924,
      "grad_norm": 10.875,
      "learning_rate": 2.6013541641108646e-05,
      "loss": 0.9663,
      "num_input_tokens_seen": 41938288,
      "step": 72295
    },
    {
      "epoch": 10.768543342269885,
      "grad_norm": 4.5,
      "learning_rate": 2.6010294882617304e-05,
      "loss": 1.0276,
      "num_input_tokens_seen": 41941104,
      "step": 72300
    },
    {
      "epoch": 10.769288054810843,
      "grad_norm": 5.46875,
      "learning_rate": 2.6007048107058058e-05,
      "loss": 1.1645,
      "num_input_tokens_seen": 41943632,
      "step": 72305
    },
    {
      "epoch": 10.770032767351802,
      "grad_norm": 11.125,
      "learning_rate": 2.6003801314485755e-05,
      "loss": 1.0753,
      "num_input_tokens_seen": 41946384,
      "step": 72310
    },
    {
      "epoch": 10.770777479892761,
      "grad_norm": 8.1875,
      "learning_rate": 2.600055450495525e-05,
      "loss": 1.0878,
      "num_input_tokens_seen": 41949456,
      "step": 72315
    },
    {
      "epoch": 10.771522192433721,
      "grad_norm": 5.0,
      "learning_rate": 2.5997307678521392e-05,
      "loss": 0.9317,
      "num_input_tokens_seen": 41952784,
      "step": 72320
    },
    {
      "epoch": 10.77226690497468,
      "grad_norm": 9.8125,
      "learning_rate": 2.5994060835239036e-05,
      "loss": 0.7447,
      "num_input_tokens_seen": 41955600,
      "step": 72325
    },
    {
      "epoch": 10.773011617515639,
      "grad_norm": 20.25,
      "learning_rate": 2.5990813975163032e-05,
      "loss": 1.1394,
      "num_input_tokens_seen": 41958576,
      "step": 72330
    },
    {
      "epoch": 10.773756330056598,
      "grad_norm": 6.5,
      "learning_rate": 2.598756709834823e-05,
      "loss": 0.9689,
      "num_input_tokens_seen": 41961552,
      "step": 72335
    },
    {
      "epoch": 10.774501042597556,
      "grad_norm": 6.5625,
      "learning_rate": 2.598432020484949e-05,
      "loss": 1.0422,
      "num_input_tokens_seen": 41964528,
      "step": 72340
    },
    {
      "epoch": 10.775245755138517,
      "grad_norm": 8.9375,
      "learning_rate": 2.598107329472166e-05,
      "loss": 1.1864,
      "num_input_tokens_seen": 41967088,
      "step": 72345
    },
    {
      "epoch": 10.775990467679476,
      "grad_norm": 6.40625,
      "learning_rate": 2.5977826368019598e-05,
      "loss": 1.1742,
      "num_input_tokens_seen": 41970000,
      "step": 72350
    },
    {
      "epoch": 10.776735180220435,
      "grad_norm": 8.4375,
      "learning_rate": 2.5974579424798146e-05,
      "loss": 1.1295,
      "num_input_tokens_seen": 41972656,
      "step": 72355
    },
    {
      "epoch": 10.777479892761393,
      "grad_norm": 9.0625,
      "learning_rate": 2.5971332465112165e-05,
      "loss": 1.0472,
      "num_input_tokens_seen": 41975504,
      "step": 72360
    },
    {
      "epoch": 10.778224605302354,
      "grad_norm": 11.8125,
      "learning_rate": 2.5968085489016507e-05,
      "loss": 1.0468,
      "num_input_tokens_seen": 41978320,
      "step": 72365
    },
    {
      "epoch": 10.778969317843313,
      "grad_norm": 7.875,
      "learning_rate": 2.5964838496566035e-05,
      "loss": 1.0056,
      "num_input_tokens_seen": 41981200,
      "step": 72370
    },
    {
      "epoch": 10.779714030384271,
      "grad_norm": 9.6875,
      "learning_rate": 2.596159148781559e-05,
      "loss": 1.0626,
      "num_input_tokens_seen": 41984080,
      "step": 72375
    },
    {
      "epoch": 10.78045874292523,
      "grad_norm": 11.25,
      "learning_rate": 2.5958344462820045e-05,
      "loss": 1.0256,
      "num_input_tokens_seen": 41986896,
      "step": 72380
    },
    {
      "epoch": 10.78120345546619,
      "grad_norm": 17.875,
      "learning_rate": 2.5955097421634244e-05,
      "loss": 1.307,
      "num_input_tokens_seen": 41989872,
      "step": 72385
    },
    {
      "epoch": 10.78194816800715,
      "grad_norm": 12.1875,
      "learning_rate": 2.5951850364313036e-05,
      "loss": 1.1702,
      "num_input_tokens_seen": 41992752,
      "step": 72390
    },
    {
      "epoch": 10.782692880548108,
      "grad_norm": 18.25,
      "learning_rate": 2.5948603290911283e-05,
      "loss": 1.0216,
      "num_input_tokens_seen": 41995408,
      "step": 72395
    },
    {
      "epoch": 10.783437593089067,
      "grad_norm": 19.0,
      "learning_rate": 2.594535620148384e-05,
      "loss": 1.179,
      "num_input_tokens_seen": 41998224,
      "step": 72400
    },
    {
      "epoch": 10.784182305630027,
      "grad_norm": 8.6875,
      "learning_rate": 2.5942109096085566e-05,
      "loss": 0.8579,
      "num_input_tokens_seen": 42001264,
      "step": 72405
    },
    {
      "epoch": 10.784927018170986,
      "grad_norm": 10.1875,
      "learning_rate": 2.5938861974771316e-05,
      "loss": 1.1079,
      "num_input_tokens_seen": 42004560,
      "step": 72410
    },
    {
      "epoch": 10.785671730711945,
      "grad_norm": 7.21875,
      "learning_rate": 2.5935614837595958e-05,
      "loss": 0.8499,
      "num_input_tokens_seen": 42007696,
      "step": 72415
    },
    {
      "epoch": 10.786416443252904,
      "grad_norm": 5.875,
      "learning_rate": 2.5932367684614328e-05,
      "loss": 1.1215,
      "num_input_tokens_seen": 42010608,
      "step": 72420
    },
    {
      "epoch": 10.787161155793864,
      "grad_norm": 14.875,
      "learning_rate": 2.59291205158813e-05,
      "loss": 1.1648,
      "num_input_tokens_seen": 42013360,
      "step": 72425
    },
    {
      "epoch": 10.787905868334823,
      "grad_norm": 13.6875,
      "learning_rate": 2.5925873331451715e-05,
      "loss": 1.3616,
      "num_input_tokens_seen": 42016112,
      "step": 72430
    },
    {
      "epoch": 10.788650580875782,
      "grad_norm": 12.5625,
      "learning_rate": 2.5922626131380444e-05,
      "loss": 1.2236,
      "num_input_tokens_seen": 42018928,
      "step": 72435
    },
    {
      "epoch": 10.78939529341674,
      "grad_norm": 15.1875,
      "learning_rate": 2.5919378915722347e-05,
      "loss": 1.1059,
      "num_input_tokens_seen": 42021808,
      "step": 72440
    },
    {
      "epoch": 10.790140005957701,
      "grad_norm": 7.34375,
      "learning_rate": 2.5916131684532274e-05,
      "loss": 1.0165,
      "num_input_tokens_seen": 42025008,
      "step": 72445
    },
    {
      "epoch": 10.79088471849866,
      "grad_norm": 10.4375,
      "learning_rate": 2.5912884437865093e-05,
      "loss": 1.1066,
      "num_input_tokens_seen": 42027824,
      "step": 72450
    },
    {
      "epoch": 10.791629431039619,
      "grad_norm": 8.25,
      "learning_rate": 2.5909637175775652e-05,
      "loss": 1.1084,
      "num_input_tokens_seen": 42030512,
      "step": 72455
    },
    {
      "epoch": 10.792374143580577,
      "grad_norm": 9.3125,
      "learning_rate": 2.5906389898318817e-05,
      "loss": 1.1065,
      "num_input_tokens_seen": 42033392,
      "step": 72460
    },
    {
      "epoch": 10.793118856121538,
      "grad_norm": 8.75,
      "learning_rate": 2.5903142605549445e-05,
      "loss": 0.9573,
      "num_input_tokens_seen": 42036208,
      "step": 72465
    },
    {
      "epoch": 10.793863568662497,
      "grad_norm": 12.6875,
      "learning_rate": 2.58998952975224e-05,
      "loss": 0.9219,
      "num_input_tokens_seen": 42039504,
      "step": 72470
    },
    {
      "epoch": 10.794608281203455,
      "grad_norm": 6.96875,
      "learning_rate": 2.5896647974292533e-05,
      "loss": 0.9967,
      "num_input_tokens_seen": 42042544,
      "step": 72475
    },
    {
      "epoch": 10.795352993744414,
      "grad_norm": 10.9375,
      "learning_rate": 2.589340063591471e-05,
      "loss": 0.9524,
      "num_input_tokens_seen": 42045296,
      "step": 72480
    },
    {
      "epoch": 10.796097706285373,
      "grad_norm": 5.21875,
      "learning_rate": 2.5890153282443797e-05,
      "loss": 1.0263,
      "num_input_tokens_seen": 42048016,
      "step": 72485
    },
    {
      "epoch": 10.796842418826333,
      "grad_norm": 9.6875,
      "learning_rate": 2.5886905913934643e-05,
      "loss": 1.2068,
      "num_input_tokens_seen": 42050928,
      "step": 72490
    },
    {
      "epoch": 10.797587131367292,
      "grad_norm": 5.0625,
      "learning_rate": 2.5883658530442117e-05,
      "loss": 1.1357,
      "num_input_tokens_seen": 42053872,
      "step": 72495
    },
    {
      "epoch": 10.798331843908251,
      "grad_norm": 17.125,
      "learning_rate": 2.5880411132021083e-05,
      "loss": 1.1873,
      "num_input_tokens_seen": 42056720,
      "step": 72500
    },
    {
      "epoch": 10.799076556449211,
      "grad_norm": 17.375,
      "learning_rate": 2.5877163718726394e-05,
      "loss": 1.1791,
      "num_input_tokens_seen": 42059568,
      "step": 72505
    },
    {
      "epoch": 10.79982126899017,
      "grad_norm": 8.625,
      "learning_rate": 2.5873916290612915e-05,
      "loss": 0.9114,
      "num_input_tokens_seen": 42062320,
      "step": 72510
    },
    {
      "epoch": 10.800565981531129,
      "grad_norm": 9.1875,
      "learning_rate": 2.5870668847735512e-05,
      "loss": 1.0785,
      "num_input_tokens_seen": 42065104,
      "step": 72515
    },
    {
      "epoch": 10.801310694072088,
      "grad_norm": 11.375,
      "learning_rate": 2.5867421390149037e-05,
      "loss": 1.0615,
      "num_input_tokens_seen": 42067760,
      "step": 72520
    },
    {
      "epoch": 10.802055406613047,
      "grad_norm": 16.875,
      "learning_rate": 2.5864173917908363e-05,
      "loss": 1.3242,
      "num_input_tokens_seen": 42070608,
      "step": 72525
    },
    {
      "epoch": 10.802800119154007,
      "grad_norm": 12.6875,
      "learning_rate": 2.5860926431068344e-05,
      "loss": 1.3127,
      "num_input_tokens_seen": 42073584,
      "step": 72530
    },
    {
      "epoch": 10.803544831694966,
      "grad_norm": 11.25,
      "learning_rate": 2.5857678929683855e-05,
      "loss": 1.2507,
      "num_input_tokens_seen": 42076496,
      "step": 72535
    },
    {
      "epoch": 10.804289544235925,
      "grad_norm": 8.875,
      "learning_rate": 2.585443141380975e-05,
      "loss": 1.1125,
      "num_input_tokens_seen": 42079696,
      "step": 72540
    },
    {
      "epoch": 10.805034256776883,
      "grad_norm": 13.125,
      "learning_rate": 2.5851183883500895e-05,
      "loss": 1.3172,
      "num_input_tokens_seen": 42082640,
      "step": 72545
    },
    {
      "epoch": 10.805778969317844,
      "grad_norm": 8.8125,
      "learning_rate": 2.5847936338812158e-05,
      "loss": 1.0909,
      "num_input_tokens_seen": 42085328,
      "step": 72550
    },
    {
      "epoch": 10.806523681858803,
      "grad_norm": 7.09375,
      "learning_rate": 2.5844688779798393e-05,
      "loss": 1.0028,
      "num_input_tokens_seen": 42088144,
      "step": 72555
    },
    {
      "epoch": 10.807268394399761,
      "grad_norm": 10.5,
      "learning_rate": 2.5841441206514468e-05,
      "loss": 1.3162,
      "num_input_tokens_seen": 42090992,
      "step": 72560
    },
    {
      "epoch": 10.80801310694072,
      "grad_norm": 7.0,
      "learning_rate": 2.583819361901525e-05,
      "loss": 0.9672,
      "num_input_tokens_seen": 42093904,
      "step": 72565
    },
    {
      "epoch": 10.80875781948168,
      "grad_norm": 12.1875,
      "learning_rate": 2.5834946017355598e-05,
      "loss": 0.8415,
      "num_input_tokens_seen": 42096912,
      "step": 72570
    },
    {
      "epoch": 10.80950253202264,
      "grad_norm": 7.25,
      "learning_rate": 2.583169840159039e-05,
      "loss": 1.0916,
      "num_input_tokens_seen": 42099760,
      "step": 72575
    },
    {
      "epoch": 10.810247244563598,
      "grad_norm": 8.1875,
      "learning_rate": 2.582845077177448e-05,
      "loss": 1.0193,
      "num_input_tokens_seen": 42102992,
      "step": 72580
    },
    {
      "epoch": 10.810991957104557,
      "grad_norm": 12.4375,
      "learning_rate": 2.5825203127962737e-05,
      "loss": 1.0519,
      "num_input_tokens_seen": 42106096,
      "step": 72585
    },
    {
      "epoch": 10.811736669645517,
      "grad_norm": 9.875,
      "learning_rate": 2.582195547021003e-05,
      "loss": 1.1381,
      "num_input_tokens_seen": 42108912,
      "step": 72590
    },
    {
      "epoch": 10.812481382186476,
      "grad_norm": 5.40625,
      "learning_rate": 2.581870779857121e-05,
      "loss": 1.0753,
      "num_input_tokens_seen": 42112112,
      "step": 72595
    },
    {
      "epoch": 10.813226094727435,
      "grad_norm": 6.40625,
      "learning_rate": 2.581546011310116e-05,
      "loss": 1.2893,
      "num_input_tokens_seen": 42114800,
      "step": 72600
    },
    {
      "epoch": 10.813970807268394,
      "grad_norm": 10.8125,
      "learning_rate": 2.5812212413854738e-05,
      "loss": 1.1933,
      "num_input_tokens_seen": 42117776,
      "step": 72605
    },
    {
      "epoch": 10.814715519809354,
      "grad_norm": 10.1875,
      "learning_rate": 2.5808964700886812e-05,
      "loss": 0.8397,
      "num_input_tokens_seen": 42120784,
      "step": 72610
    },
    {
      "epoch": 10.815460232350313,
      "grad_norm": 12.75,
      "learning_rate": 2.5805716974252257e-05,
      "loss": 1.1259,
      "num_input_tokens_seen": 42123472,
      "step": 72615
    },
    {
      "epoch": 10.816204944891272,
      "grad_norm": 11.6875,
      "learning_rate": 2.5802469234005927e-05,
      "loss": 1.2655,
      "num_input_tokens_seen": 42126576,
      "step": 72620
    },
    {
      "epoch": 10.81694965743223,
      "grad_norm": 8.5,
      "learning_rate": 2.57992214802027e-05,
      "loss": 1.2782,
      "num_input_tokens_seen": 42129296,
      "step": 72625
    },
    {
      "epoch": 10.817694369973191,
      "grad_norm": 10.5625,
      "learning_rate": 2.579597371289743e-05,
      "loss": 0.9168,
      "num_input_tokens_seen": 42132112,
      "step": 72630
    },
    {
      "epoch": 10.81843908251415,
      "grad_norm": 7.59375,
      "learning_rate": 2.5792725932144996e-05,
      "loss": 0.9649,
      "num_input_tokens_seen": 42135120,
      "step": 72635
    },
    {
      "epoch": 10.819183795055109,
      "grad_norm": 29.25,
      "learning_rate": 2.5789478138000262e-05,
      "loss": 1.1326,
      "num_input_tokens_seen": 42137968,
      "step": 72640
    },
    {
      "epoch": 10.819928507596067,
      "grad_norm": 8.75,
      "learning_rate": 2.5786230330518096e-05,
      "loss": 1.1807,
      "num_input_tokens_seen": 42141008,
      "step": 72645
    },
    {
      "epoch": 10.820673220137028,
      "grad_norm": 20.625,
      "learning_rate": 2.5782982509753377e-05,
      "loss": 0.8906,
      "num_input_tokens_seen": 42143664,
      "step": 72650
    },
    {
      "epoch": 10.821417932677987,
      "grad_norm": 13.8125,
      "learning_rate": 2.5779734675760957e-05,
      "loss": 1.4301,
      "num_input_tokens_seen": 42146448,
      "step": 72655
    },
    {
      "epoch": 10.822162645218945,
      "grad_norm": 6.1875,
      "learning_rate": 2.5776486828595715e-05,
      "loss": 1.0744,
      "num_input_tokens_seen": 42149456,
      "step": 72660
    },
    {
      "epoch": 10.822907357759904,
      "grad_norm": 9.75,
      "learning_rate": 2.5773238968312514e-05,
      "loss": 0.9831,
      "num_input_tokens_seen": 42152688,
      "step": 72665
    },
    {
      "epoch": 10.823652070300863,
      "grad_norm": 8.75,
      "learning_rate": 2.5769991094966228e-05,
      "loss": 1.3954,
      "num_input_tokens_seen": 42155824,
      "step": 72670
    },
    {
      "epoch": 10.824396782841823,
      "grad_norm": 7.09375,
      "learning_rate": 2.5766743208611726e-05,
      "loss": 1.3704,
      "num_input_tokens_seen": 42158576,
      "step": 72675
    },
    {
      "epoch": 10.825141495382782,
      "grad_norm": 9.9375,
      "learning_rate": 2.576349530930388e-05,
      "loss": 1.1249,
      "num_input_tokens_seen": 42161424,
      "step": 72680
    },
    {
      "epoch": 10.825886207923741,
      "grad_norm": 8.1875,
      "learning_rate": 2.5760247397097553e-05,
      "loss": 1.3423,
      "num_input_tokens_seen": 42164240,
      "step": 72685
    },
    {
      "epoch": 10.826630920464702,
      "grad_norm": 16.625,
      "learning_rate": 2.5756999472047617e-05,
      "loss": 0.9218,
      "num_input_tokens_seen": 42167888,
      "step": 72690
    },
    {
      "epoch": 10.82737563300566,
      "grad_norm": 8.25,
      "learning_rate": 2.575375153420894e-05,
      "loss": 0.8755,
      "num_input_tokens_seen": 42171024,
      "step": 72695
    },
    {
      "epoch": 10.828120345546619,
      "grad_norm": 4.90625,
      "learning_rate": 2.5750503583636402e-05,
      "loss": 0.978,
      "num_input_tokens_seen": 42174096,
      "step": 72700
    },
    {
      "epoch": 10.828865058087578,
      "grad_norm": 5.5625,
      "learning_rate": 2.5747255620384868e-05,
      "loss": 1.2074,
      "num_input_tokens_seen": 42176880,
      "step": 72705
    },
    {
      "epoch": 10.829609770628537,
      "grad_norm": 8.5,
      "learning_rate": 2.5744007644509215e-05,
      "loss": 0.8939,
      "num_input_tokens_seen": 42179664,
      "step": 72710
    },
    {
      "epoch": 10.830354483169497,
      "grad_norm": 9.0,
      "learning_rate": 2.574075965606431e-05,
      "loss": 1.0665,
      "num_input_tokens_seen": 42182512,
      "step": 72715
    },
    {
      "epoch": 10.831099195710456,
      "grad_norm": 15.9375,
      "learning_rate": 2.5737511655105018e-05,
      "loss": 0.9654,
      "num_input_tokens_seen": 42185232,
      "step": 72720
    },
    {
      "epoch": 10.831843908251415,
      "grad_norm": 23.125,
      "learning_rate": 2.573426364168622e-05,
      "loss": 1.3491,
      "num_input_tokens_seen": 42188304,
      "step": 72725
    },
    {
      "epoch": 10.832588620792373,
      "grad_norm": 10.5,
      "learning_rate": 2.5731015615862774e-05,
      "loss": 1.2657,
      "num_input_tokens_seen": 42191088,
      "step": 72730
    },
    {
      "epoch": 10.833333333333334,
      "grad_norm": 10.0625,
      "learning_rate": 2.572776757768957e-05,
      "loss": 1.1218,
      "num_input_tokens_seen": 42193840,
      "step": 72735
    },
    {
      "epoch": 10.834078045874293,
      "grad_norm": 7.96875,
      "learning_rate": 2.5724519527221468e-05,
      "loss": 1.0526,
      "num_input_tokens_seen": 42196528,
      "step": 72740
    },
    {
      "epoch": 10.834822758415251,
      "grad_norm": 6.15625,
      "learning_rate": 2.5721271464513354e-05,
      "loss": 0.6345,
      "num_input_tokens_seen": 42199248,
      "step": 72745
    },
    {
      "epoch": 10.83556747095621,
      "grad_norm": 7.4375,
      "learning_rate": 2.571802338962009e-05,
      "loss": 0.932,
      "num_input_tokens_seen": 42202480,
      "step": 72750
    },
    {
      "epoch": 10.83631218349717,
      "grad_norm": 9.0625,
      "learning_rate": 2.5714775302596545e-05,
      "loss": 0.9606,
      "num_input_tokens_seen": 42205136,
      "step": 72755
    },
    {
      "epoch": 10.83705689603813,
      "grad_norm": 10.1875,
      "learning_rate": 2.57115272034976e-05,
      "loss": 1.4406,
      "num_input_tokens_seen": 42208304,
      "step": 72760
    },
    {
      "epoch": 10.837801608579088,
      "grad_norm": 16.875,
      "learning_rate": 2.5708279092378123e-05,
      "loss": 1.1368,
      "num_input_tokens_seen": 42211120,
      "step": 72765
    },
    {
      "epoch": 10.838546321120047,
      "grad_norm": 7.25,
      "learning_rate": 2.5705030969292992e-05,
      "loss": 0.941,
      "num_input_tokens_seen": 42213904,
      "step": 72770
    },
    {
      "epoch": 10.839291033661008,
      "grad_norm": 3.34375,
      "learning_rate": 2.5701782834297078e-05,
      "loss": 1.0342,
      "num_input_tokens_seen": 42216848,
      "step": 72775
    },
    {
      "epoch": 10.840035746201966,
      "grad_norm": 10.625,
      "learning_rate": 2.5698534687445263e-05,
      "loss": 1.2694,
      "num_input_tokens_seen": 42219728,
      "step": 72780
    },
    {
      "epoch": 10.840780458742925,
      "grad_norm": 10.75,
      "learning_rate": 2.5695286528792413e-05,
      "loss": 0.9258,
      "num_input_tokens_seen": 42222320,
      "step": 72785
    },
    {
      "epoch": 10.841525171283884,
      "grad_norm": 13.0,
      "learning_rate": 2.56920383583934e-05,
      "loss": 1.1894,
      "num_input_tokens_seen": 42224976,
      "step": 72790
    },
    {
      "epoch": 10.842269883824844,
      "grad_norm": 15.875,
      "learning_rate": 2.56887901763031e-05,
      "loss": 1.1791,
      "num_input_tokens_seen": 42227888,
      "step": 72795
    },
    {
      "epoch": 10.843014596365803,
      "grad_norm": 5.21875,
      "learning_rate": 2.5685541982576395e-05,
      "loss": 0.7854,
      "num_input_tokens_seen": 42230960,
      "step": 72800
    },
    {
      "epoch": 10.843759308906762,
      "grad_norm": 5.875,
      "learning_rate": 2.5682293777268153e-05,
      "loss": 0.9757,
      "num_input_tokens_seen": 42233680,
      "step": 72805
    },
    {
      "epoch": 10.84450402144772,
      "grad_norm": 11.5625,
      "learning_rate": 2.567904556043325e-05,
      "loss": 1.0038,
      "num_input_tokens_seen": 42236560,
      "step": 72810
    },
    {
      "epoch": 10.845248733988681,
      "grad_norm": 7.46875,
      "learning_rate": 2.5675797332126566e-05,
      "loss": 1.0562,
      "num_input_tokens_seen": 42239760,
      "step": 72815
    },
    {
      "epoch": 10.84599344652964,
      "grad_norm": 12.0,
      "learning_rate": 2.567254909240297e-05,
      "loss": 1.1122,
      "num_input_tokens_seen": 42242640,
      "step": 72820
    },
    {
      "epoch": 10.846738159070599,
      "grad_norm": 11.75,
      "learning_rate": 2.566930084131734e-05,
      "loss": 1.0863,
      "num_input_tokens_seen": 42245328,
      "step": 72825
    },
    {
      "epoch": 10.847482871611557,
      "grad_norm": 12.75,
      "learning_rate": 2.566605257892456e-05,
      "loss": 1.4837,
      "num_input_tokens_seen": 42248176,
      "step": 72830
    },
    {
      "epoch": 10.848227584152518,
      "grad_norm": 11.625,
      "learning_rate": 2.5662804305279485e-05,
      "loss": 1.3279,
      "num_input_tokens_seen": 42251088,
      "step": 72835
    },
    {
      "epoch": 10.848972296693477,
      "grad_norm": 8.375,
      "learning_rate": 2.5659556020437015e-05,
      "loss": 0.936,
      "num_input_tokens_seen": 42254000,
      "step": 72840
    },
    {
      "epoch": 10.849717009234435,
      "grad_norm": 7.15625,
      "learning_rate": 2.5656307724452016e-05,
      "loss": 0.6974,
      "num_input_tokens_seen": 42256688,
      "step": 72845
    },
    {
      "epoch": 10.850461721775394,
      "grad_norm": 9.8125,
      "learning_rate": 2.565305941737936e-05,
      "loss": 0.9726,
      "num_input_tokens_seen": 42259248,
      "step": 72850
    },
    {
      "epoch": 10.851206434316353,
      "grad_norm": 13.875,
      "learning_rate": 2.5649811099273935e-05,
      "loss": 1.3917,
      "num_input_tokens_seen": 42262416,
      "step": 72855
    },
    {
      "epoch": 10.851951146857314,
      "grad_norm": 13.5625,
      "learning_rate": 2.564656277019061e-05,
      "loss": 1.5555,
      "num_input_tokens_seen": 42265328,
      "step": 72860
    },
    {
      "epoch": 10.852695859398272,
      "grad_norm": 9.0,
      "learning_rate": 2.5643314430184257e-05,
      "loss": 1.237,
      "num_input_tokens_seen": 42268272,
      "step": 72865
    },
    {
      "epoch": 10.853440571939231,
      "grad_norm": 6.65625,
      "learning_rate": 2.564006607930977e-05,
      "loss": 1.1152,
      "num_input_tokens_seen": 42271184,
      "step": 72870
    },
    {
      "epoch": 10.85418528448019,
      "grad_norm": 6.8125,
      "learning_rate": 2.5636817717622015e-05,
      "loss": 1.0349,
      "num_input_tokens_seen": 42273744,
      "step": 72875
    },
    {
      "epoch": 10.85492999702115,
      "grad_norm": 13.6875,
      "learning_rate": 2.5633569345175873e-05,
      "loss": 1.0798,
      "num_input_tokens_seen": 42276176,
      "step": 72880
    },
    {
      "epoch": 10.85567470956211,
      "grad_norm": 11.0625,
      "learning_rate": 2.5630320962026217e-05,
      "loss": 1.0731,
      "num_input_tokens_seen": 42278864,
      "step": 72885
    },
    {
      "epoch": 10.856419422103068,
      "grad_norm": 12.5,
      "learning_rate": 2.5627072568227927e-05,
      "loss": 0.9436,
      "num_input_tokens_seen": 42281840,
      "step": 72890
    },
    {
      "epoch": 10.857164134644027,
      "grad_norm": 9.3125,
      "learning_rate": 2.5623824163835887e-05,
      "loss": 1.2365,
      "num_input_tokens_seen": 42285104,
      "step": 72895
    },
    {
      "epoch": 10.857908847184987,
      "grad_norm": 12.6875,
      "learning_rate": 2.5620575748904968e-05,
      "loss": 0.7863,
      "num_input_tokens_seen": 42288240,
      "step": 72900
    },
    {
      "epoch": 10.858653559725946,
      "grad_norm": 7.9375,
      "learning_rate": 2.5617327323490055e-05,
      "loss": 0.9758,
      "num_input_tokens_seen": 42291024,
      "step": 72905
    },
    {
      "epoch": 10.859398272266905,
      "grad_norm": 9.125,
      "learning_rate": 2.5614078887646025e-05,
      "loss": 1.0251,
      "num_input_tokens_seen": 42294000,
      "step": 72910
    },
    {
      "epoch": 10.860142984807863,
      "grad_norm": 8.0625,
      "learning_rate": 2.5610830441427762e-05,
      "loss": 1.1655,
      "num_input_tokens_seen": 42296912,
      "step": 72915
    },
    {
      "epoch": 10.860887697348824,
      "grad_norm": 8.75,
      "learning_rate": 2.5607581984890134e-05,
      "loss": 1.0126,
      "num_input_tokens_seen": 42300048,
      "step": 72920
    },
    {
      "epoch": 10.861632409889783,
      "grad_norm": 10.5,
      "learning_rate": 2.5604333518088026e-05,
      "loss": 1.0925,
      "num_input_tokens_seen": 42302928,
      "step": 72925
    },
    {
      "epoch": 10.862377122430741,
      "grad_norm": 15.75,
      "learning_rate": 2.560108504107631e-05,
      "loss": 1.1263,
      "num_input_tokens_seen": 42305808,
      "step": 72930
    },
    {
      "epoch": 10.8631218349717,
      "grad_norm": 18.0,
      "learning_rate": 2.5597836553909884e-05,
      "loss": 1.1263,
      "num_input_tokens_seen": 42308720,
      "step": 72935
    },
    {
      "epoch": 10.86386654751266,
      "grad_norm": 16.0,
      "learning_rate": 2.5594588056643608e-05,
      "loss": 1.3407,
      "num_input_tokens_seen": 42311696,
      "step": 72940
    },
    {
      "epoch": 10.86461126005362,
      "grad_norm": 8.25,
      "learning_rate": 2.5591339549332383e-05,
      "loss": 1.0372,
      "num_input_tokens_seen": 42314384,
      "step": 72945
    },
    {
      "epoch": 10.865355972594578,
      "grad_norm": 13.75,
      "learning_rate": 2.5588091032031075e-05,
      "loss": 0.9226,
      "num_input_tokens_seen": 42316848,
      "step": 72950
    },
    {
      "epoch": 10.866100685135537,
      "grad_norm": 5.96875,
      "learning_rate": 2.5584842504794558e-05,
      "loss": 0.8771,
      "num_input_tokens_seen": 42319632,
      "step": 72955
    },
    {
      "epoch": 10.866845397676498,
      "grad_norm": 10.5625,
      "learning_rate": 2.5581593967677724e-05,
      "loss": 1.1872,
      "num_input_tokens_seen": 42322384,
      "step": 72960
    },
    {
      "epoch": 10.867590110217456,
      "grad_norm": 8.625,
      "learning_rate": 2.557834542073545e-05,
      "loss": 1.0974,
      "num_input_tokens_seen": 42325072,
      "step": 72965
    },
    {
      "epoch": 10.868334822758415,
      "grad_norm": 11.4375,
      "learning_rate": 2.557509686402262e-05,
      "loss": 1.5585,
      "num_input_tokens_seen": 42327984,
      "step": 72970
    },
    {
      "epoch": 10.869079535299374,
      "grad_norm": 13.9375,
      "learning_rate": 2.5571848297594116e-05,
      "loss": 1.1006,
      "num_input_tokens_seen": 42330608,
      "step": 72975
    },
    {
      "epoch": 10.869824247840334,
      "grad_norm": 12.625,
      "learning_rate": 2.5568599721504814e-05,
      "loss": 0.7563,
      "num_input_tokens_seen": 42333424,
      "step": 72980
    },
    {
      "epoch": 10.870568960381293,
      "grad_norm": 13.875,
      "learning_rate": 2.5565351135809597e-05,
      "loss": 1.1142,
      "num_input_tokens_seen": 42336400,
      "step": 72985
    },
    {
      "epoch": 10.871313672922252,
      "grad_norm": 15.3125,
      "learning_rate": 2.5562102540563355e-05,
      "loss": 0.8661,
      "num_input_tokens_seen": 42339568,
      "step": 72990
    },
    {
      "epoch": 10.87205838546321,
      "grad_norm": 5.3125,
      "learning_rate": 2.5558853935820948e-05,
      "loss": 1.1559,
      "num_input_tokens_seen": 42342320,
      "step": 72995
    },
    {
      "epoch": 10.872803098004171,
      "grad_norm": 8.6875,
      "learning_rate": 2.555560532163728e-05,
      "loss": 1.1341,
      "num_input_tokens_seen": 42345104,
      "step": 73000
    },
    {
      "epoch": 10.87354781054513,
      "grad_norm": 13.9375,
      "learning_rate": 2.555235669806722e-05,
      "loss": 1.4659,
      "num_input_tokens_seen": 42348048,
      "step": 73005
    },
    {
      "epoch": 10.874292523086089,
      "grad_norm": 9.625,
      "learning_rate": 2.554910806516566e-05,
      "loss": 1.2219,
      "num_input_tokens_seen": 42350992,
      "step": 73010
    },
    {
      "epoch": 10.875037235627047,
      "grad_norm": 9.4375,
      "learning_rate": 2.5545859422987478e-05,
      "loss": 0.8619,
      "num_input_tokens_seen": 42354352,
      "step": 73015
    },
    {
      "epoch": 10.875781948168008,
      "grad_norm": 6.59375,
      "learning_rate": 2.554261077158755e-05,
      "loss": 1.0511,
      "num_input_tokens_seen": 42357392,
      "step": 73020
    },
    {
      "epoch": 10.876526660708967,
      "grad_norm": 17.5,
      "learning_rate": 2.5539362111020765e-05,
      "loss": 0.883,
      "num_input_tokens_seen": 42360400,
      "step": 73025
    },
    {
      "epoch": 10.877271373249926,
      "grad_norm": 12.625,
      "learning_rate": 2.5536113441342014e-05,
      "loss": 1.0593,
      "num_input_tokens_seen": 42363344,
      "step": 73030
    },
    {
      "epoch": 10.878016085790884,
      "grad_norm": 6.90625,
      "learning_rate": 2.5532864762606164e-05,
      "loss": 0.6176,
      "num_input_tokens_seen": 42366192,
      "step": 73035
    },
    {
      "epoch": 10.878760798331843,
      "grad_norm": 10.625,
      "learning_rate": 2.55296160748681e-05,
      "loss": 1.0184,
      "num_input_tokens_seen": 42369168,
      "step": 73040
    },
    {
      "epoch": 10.879505510872804,
      "grad_norm": 8.25,
      "learning_rate": 2.5526367378182725e-05,
      "loss": 0.9131,
      "num_input_tokens_seen": 42372176,
      "step": 73045
    },
    {
      "epoch": 10.880250223413762,
      "grad_norm": 6.375,
      "learning_rate": 2.55231186726049e-05,
      "loss": 1.0306,
      "num_input_tokens_seen": 42375024,
      "step": 73050
    },
    {
      "epoch": 10.880994935954721,
      "grad_norm": 10.8125,
      "learning_rate": 2.5519869958189513e-05,
      "loss": 1.1025,
      "num_input_tokens_seen": 42377456,
      "step": 73055
    },
    {
      "epoch": 10.88173964849568,
      "grad_norm": 7.65625,
      "learning_rate": 2.5516621234991456e-05,
      "loss": 1.1022,
      "num_input_tokens_seen": 42380336,
      "step": 73060
    },
    {
      "epoch": 10.88248436103664,
      "grad_norm": 10.5625,
      "learning_rate": 2.551337250306561e-05,
      "loss": 1.0283,
      "num_input_tokens_seen": 42383056,
      "step": 73065
    },
    {
      "epoch": 10.8832290735776,
      "grad_norm": 6.125,
      "learning_rate": 2.5510123762466853e-05,
      "loss": 0.9353,
      "num_input_tokens_seen": 42385808,
      "step": 73070
    },
    {
      "epoch": 10.883973786118558,
      "grad_norm": 16.25,
      "learning_rate": 2.5506875013250075e-05,
      "loss": 1.2808,
      "num_input_tokens_seen": 42388688,
      "step": 73075
    },
    {
      "epoch": 10.884718498659517,
      "grad_norm": 4.75,
      "learning_rate": 2.5503626255470164e-05,
      "loss": 0.8991,
      "num_input_tokens_seen": 42391440,
      "step": 73080
    },
    {
      "epoch": 10.885463211200477,
      "grad_norm": 10.5,
      "learning_rate": 2.5500377489181992e-05,
      "loss": 0.9164,
      "num_input_tokens_seen": 42394544,
      "step": 73085
    },
    {
      "epoch": 10.886207923741436,
      "grad_norm": 6.28125,
      "learning_rate": 2.5497128714440456e-05,
      "loss": 0.9159,
      "num_input_tokens_seen": 42397264,
      "step": 73090
    },
    {
      "epoch": 10.886952636282395,
      "grad_norm": 8.8125,
      "learning_rate": 2.549387993130043e-05,
      "loss": 1.2163,
      "num_input_tokens_seen": 42400368,
      "step": 73095
    },
    {
      "epoch": 10.887697348823353,
      "grad_norm": 8.8125,
      "learning_rate": 2.5490631139816806e-05,
      "loss": 0.8613,
      "num_input_tokens_seen": 42403216,
      "step": 73100
    },
    {
      "epoch": 10.888442061364314,
      "grad_norm": 7.0625,
      "learning_rate": 2.548738234004447e-05,
      "loss": 1.0161,
      "num_input_tokens_seen": 42406480,
      "step": 73105
    },
    {
      "epoch": 10.889186773905273,
      "grad_norm": 9.1875,
      "learning_rate": 2.5484133532038307e-05,
      "loss": 0.9599,
      "num_input_tokens_seen": 42409328,
      "step": 73110
    },
    {
      "epoch": 10.889931486446232,
      "grad_norm": 12.0,
      "learning_rate": 2.5480884715853197e-05,
      "loss": 0.9487,
      "num_input_tokens_seen": 42412304,
      "step": 73115
    },
    {
      "epoch": 10.89067619898719,
      "grad_norm": 6.28125,
      "learning_rate": 2.547763589154403e-05,
      "loss": 0.915,
      "num_input_tokens_seen": 42415152,
      "step": 73120
    },
    {
      "epoch": 10.89142091152815,
      "grad_norm": 6.90625,
      "learning_rate": 2.5474387059165687e-05,
      "loss": 1.2454,
      "num_input_tokens_seen": 42418288,
      "step": 73125
    },
    {
      "epoch": 10.89216562406911,
      "grad_norm": 7.8125,
      "learning_rate": 2.547113821877306e-05,
      "loss": 1.0462,
      "num_input_tokens_seen": 42421712,
      "step": 73130
    },
    {
      "epoch": 10.892910336610068,
      "grad_norm": 12.0625,
      "learning_rate": 2.5467889370421027e-05,
      "loss": 1.1376,
      "num_input_tokens_seen": 42424400,
      "step": 73135
    },
    {
      "epoch": 10.893655049151027,
      "grad_norm": 8.25,
      "learning_rate": 2.546464051416448e-05,
      "loss": 0.9575,
      "num_input_tokens_seen": 42427344,
      "step": 73140
    },
    {
      "epoch": 10.894399761691988,
      "grad_norm": 6.5,
      "learning_rate": 2.5461391650058307e-05,
      "loss": 0.834,
      "num_input_tokens_seen": 42430064,
      "step": 73145
    },
    {
      "epoch": 10.895144474232946,
      "grad_norm": 5.3125,
      "learning_rate": 2.5458142778157396e-05,
      "loss": 1.0065,
      "num_input_tokens_seen": 42432816,
      "step": 73150
    },
    {
      "epoch": 10.895889186773905,
      "grad_norm": 7.65625,
      "learning_rate": 2.545489389851662e-05,
      "loss": 0.7212,
      "num_input_tokens_seen": 42435664,
      "step": 73155
    },
    {
      "epoch": 10.896633899314864,
      "grad_norm": 11.5625,
      "learning_rate": 2.5451645011190872e-05,
      "loss": 1.1261,
      "num_input_tokens_seen": 42438704,
      "step": 73160
    },
    {
      "epoch": 10.897378611855824,
      "grad_norm": 10.4375,
      "learning_rate": 2.5448396116235046e-05,
      "loss": 0.8418,
      "num_input_tokens_seen": 42441776,
      "step": 73165
    },
    {
      "epoch": 10.898123324396783,
      "grad_norm": 10.9375,
      "learning_rate": 2.5445147213704017e-05,
      "loss": 0.9842,
      "num_input_tokens_seen": 42444528,
      "step": 73170
    },
    {
      "epoch": 10.898868036937742,
      "grad_norm": 11.375,
      "learning_rate": 2.5441898303652688e-05,
      "loss": 0.9704,
      "num_input_tokens_seen": 42447248,
      "step": 73175
    },
    {
      "epoch": 10.8996127494787,
      "grad_norm": 13.1875,
      "learning_rate": 2.5438649386135932e-05,
      "loss": 1.0927,
      "num_input_tokens_seen": 42450000,
      "step": 73180
    },
    {
      "epoch": 10.90035746201966,
      "grad_norm": 19.75,
      "learning_rate": 2.5435400461208637e-05,
      "loss": 0.9994,
      "num_input_tokens_seen": 42453008,
      "step": 73185
    },
    {
      "epoch": 10.90110217456062,
      "grad_norm": 17.375,
      "learning_rate": 2.5432151528925702e-05,
      "loss": 1.2618,
      "num_input_tokens_seen": 42456336,
      "step": 73190
    },
    {
      "epoch": 10.901846887101579,
      "grad_norm": 5.78125,
      "learning_rate": 2.5428902589341996e-05,
      "loss": 1.0926,
      "num_input_tokens_seen": 42459120,
      "step": 73195
    },
    {
      "epoch": 10.902591599642538,
      "grad_norm": 7.875,
      "learning_rate": 2.542565364251242e-05,
      "loss": 1.281,
      "num_input_tokens_seen": 42462192,
      "step": 73200
    },
    {
      "epoch": 10.903336312183498,
      "grad_norm": 7.5625,
      "learning_rate": 2.542240468849186e-05,
      "loss": 0.9881,
      "num_input_tokens_seen": 42465072,
      "step": 73205
    },
    {
      "epoch": 10.904081024724457,
      "grad_norm": 10.9375,
      "learning_rate": 2.5419155727335204e-05,
      "loss": 0.9017,
      "num_input_tokens_seen": 42468016,
      "step": 73210
    },
    {
      "epoch": 10.904825737265416,
      "grad_norm": 11.8125,
      "learning_rate": 2.5415906759097336e-05,
      "loss": 1.0181,
      "num_input_tokens_seen": 42470832,
      "step": 73215
    },
    {
      "epoch": 10.905570449806374,
      "grad_norm": 10.0625,
      "learning_rate": 2.5412657783833143e-05,
      "loss": 1.2446,
      "num_input_tokens_seen": 42473776,
      "step": 73220
    },
    {
      "epoch": 10.906315162347333,
      "grad_norm": 8.8125,
      "learning_rate": 2.5409408801597517e-05,
      "loss": 0.9802,
      "num_input_tokens_seen": 42476592,
      "step": 73225
    },
    {
      "epoch": 10.907059874888294,
      "grad_norm": 8.9375,
      "learning_rate": 2.540615981244535e-05,
      "loss": 1.1254,
      "num_input_tokens_seen": 42479408,
      "step": 73230
    },
    {
      "epoch": 10.907804587429252,
      "grad_norm": 7.40625,
      "learning_rate": 2.5402910816431525e-05,
      "loss": 1.1333,
      "num_input_tokens_seen": 42482384,
      "step": 73235
    },
    {
      "epoch": 10.908549299970211,
      "grad_norm": 11.625,
      "learning_rate": 2.5399661813610925e-05,
      "loss": 1.1192,
      "num_input_tokens_seen": 42485424,
      "step": 73240
    },
    {
      "epoch": 10.90929401251117,
      "grad_norm": 11.8125,
      "learning_rate": 2.5396412804038455e-05,
      "loss": 1.1449,
      "num_input_tokens_seen": 42488144,
      "step": 73245
    },
    {
      "epoch": 10.91003872505213,
      "grad_norm": 8.3125,
      "learning_rate": 2.5393163787768988e-05,
      "loss": 0.9941,
      "num_input_tokens_seen": 42491056,
      "step": 73250
    },
    {
      "epoch": 10.91078343759309,
      "grad_norm": 5.125,
      "learning_rate": 2.5389914764857413e-05,
      "loss": 1.5085,
      "num_input_tokens_seen": 42494064,
      "step": 73255
    },
    {
      "epoch": 10.911528150134048,
      "grad_norm": 7.03125,
      "learning_rate": 2.538666573535863e-05,
      "loss": 0.9234,
      "num_input_tokens_seen": 42496880,
      "step": 73260
    },
    {
      "epoch": 10.912272862675007,
      "grad_norm": 8.1875,
      "learning_rate": 2.5383416699327524e-05,
      "loss": 1.0201,
      "num_input_tokens_seen": 42499632,
      "step": 73265
    },
    {
      "epoch": 10.913017575215967,
      "grad_norm": 15.0,
      "learning_rate": 2.5380167656818978e-05,
      "loss": 1.1341,
      "num_input_tokens_seen": 42502576,
      "step": 73270
    },
    {
      "epoch": 10.913762287756926,
      "grad_norm": 12.5,
      "learning_rate": 2.537691860788789e-05,
      "loss": 1.055,
      "num_input_tokens_seen": 42505616,
      "step": 73275
    },
    {
      "epoch": 10.914507000297885,
      "grad_norm": 9.5625,
      "learning_rate": 2.5373669552589146e-05,
      "loss": 0.9907,
      "num_input_tokens_seen": 42508432,
      "step": 73280
    },
    {
      "epoch": 10.915251712838844,
      "grad_norm": 12.5625,
      "learning_rate": 2.537042049097763e-05,
      "loss": 0.9945,
      "num_input_tokens_seen": 42511344,
      "step": 73285
    },
    {
      "epoch": 10.915996425379804,
      "grad_norm": 12.0625,
      "learning_rate": 2.5367171423108238e-05,
      "loss": 1.0125,
      "num_input_tokens_seen": 42514224,
      "step": 73290
    },
    {
      "epoch": 10.916741137920763,
      "grad_norm": 16.375,
      "learning_rate": 2.5363922349035857e-05,
      "loss": 1.2288,
      "num_input_tokens_seen": 42516848,
      "step": 73295
    },
    {
      "epoch": 10.917485850461722,
      "grad_norm": 18.625,
      "learning_rate": 2.5360673268815378e-05,
      "loss": 0.997,
      "num_input_tokens_seen": 42520368,
      "step": 73300
    },
    {
      "epoch": 10.91823056300268,
      "grad_norm": 6.1875,
      "learning_rate": 2.535742418250169e-05,
      "loss": 1.089,
      "num_input_tokens_seen": 42523344,
      "step": 73305
    },
    {
      "epoch": 10.918975275543641,
      "grad_norm": 8.4375,
      "learning_rate": 2.535417509014969e-05,
      "loss": 1.0355,
      "num_input_tokens_seen": 42526352,
      "step": 73310
    },
    {
      "epoch": 10.9197199880846,
      "grad_norm": 8.8125,
      "learning_rate": 2.5350925991814263e-05,
      "loss": 1.2907,
      "num_input_tokens_seen": 42529584,
      "step": 73315
    },
    {
      "epoch": 10.920464700625558,
      "grad_norm": 11.4375,
      "learning_rate": 2.5347676887550286e-05,
      "loss": 1.2149,
      "num_input_tokens_seen": 42532560,
      "step": 73320
    },
    {
      "epoch": 10.921209413166517,
      "grad_norm": 6.9375,
      "learning_rate": 2.534442777741267e-05,
      "loss": 1.3304,
      "num_input_tokens_seen": 42535632,
      "step": 73325
    },
    {
      "epoch": 10.921954125707478,
      "grad_norm": 16.25,
      "learning_rate": 2.5341178661456293e-05,
      "loss": 0.8616,
      "num_input_tokens_seen": 42538800,
      "step": 73330
    },
    {
      "epoch": 10.922698838248436,
      "grad_norm": 13.625,
      "learning_rate": 2.533792953973605e-05,
      "loss": 1.3348,
      "num_input_tokens_seen": 42541840,
      "step": 73335
    },
    {
      "epoch": 10.923443550789395,
      "grad_norm": 10.75,
      "learning_rate": 2.533468041230683e-05,
      "loss": 1.4179,
      "num_input_tokens_seen": 42544752,
      "step": 73340
    },
    {
      "epoch": 10.924188263330354,
      "grad_norm": 12.5,
      "learning_rate": 2.5331431279223528e-05,
      "loss": 0.9598,
      "num_input_tokens_seen": 42547536,
      "step": 73345
    },
    {
      "epoch": 10.924932975871315,
      "grad_norm": 9.8125,
      "learning_rate": 2.5328182140541028e-05,
      "loss": 1.0814,
      "num_input_tokens_seen": 42550416,
      "step": 73350
    },
    {
      "epoch": 10.925677688412273,
      "grad_norm": 12.5625,
      "learning_rate": 2.5324932996314233e-05,
      "loss": 1.1731,
      "num_input_tokens_seen": 42553360,
      "step": 73355
    },
    {
      "epoch": 10.926422400953232,
      "grad_norm": 10.9375,
      "learning_rate": 2.5321683846598015e-05,
      "loss": 1.1508,
      "num_input_tokens_seen": 42556368,
      "step": 73360
    },
    {
      "epoch": 10.92716711349419,
      "grad_norm": 12.75,
      "learning_rate": 2.531843469144728e-05,
      "loss": 1.0051,
      "num_input_tokens_seen": 42559280,
      "step": 73365
    },
    {
      "epoch": 10.92791182603515,
      "grad_norm": 10.9375,
      "learning_rate": 2.5315185530916907e-05,
      "loss": 1.6711,
      "num_input_tokens_seen": 42562096,
      "step": 73370
    },
    {
      "epoch": 10.92865653857611,
      "grad_norm": 6.3125,
      "learning_rate": 2.5311936365061804e-05,
      "loss": 0.7431,
      "num_input_tokens_seen": 42564880,
      "step": 73375
    },
    {
      "epoch": 10.929401251117069,
      "grad_norm": 18.75,
      "learning_rate": 2.530868719393685e-05,
      "loss": 1.1811,
      "num_input_tokens_seen": 42567696,
      "step": 73380
    },
    {
      "epoch": 10.930145963658028,
      "grad_norm": 17.375,
      "learning_rate": 2.5305438017596937e-05,
      "loss": 1.2466,
      "num_input_tokens_seen": 42570608,
      "step": 73385
    },
    {
      "epoch": 10.930890676198988,
      "grad_norm": 9.5,
      "learning_rate": 2.5302188836096963e-05,
      "loss": 0.9868,
      "num_input_tokens_seen": 42573392,
      "step": 73390
    },
    {
      "epoch": 10.931635388739947,
      "grad_norm": 11.6875,
      "learning_rate": 2.5298939649491816e-05,
      "loss": 0.95,
      "num_input_tokens_seen": 42576560,
      "step": 73395
    },
    {
      "epoch": 10.932380101280906,
      "grad_norm": 7.4375,
      "learning_rate": 2.5295690457836384e-05,
      "loss": 0.7772,
      "num_input_tokens_seen": 42579600,
      "step": 73400
    },
    {
      "epoch": 10.933124813821864,
      "grad_norm": 12.25,
      "learning_rate": 2.529244126118556e-05,
      "loss": 1.2445,
      "num_input_tokens_seen": 42582480,
      "step": 73405
    },
    {
      "epoch": 10.933869526362823,
      "grad_norm": 13.375,
      "learning_rate": 2.5289192059594253e-05,
      "loss": 0.8446,
      "num_input_tokens_seen": 42585424,
      "step": 73410
    },
    {
      "epoch": 10.934614238903784,
      "grad_norm": 8.625,
      "learning_rate": 2.5285942853117327e-05,
      "loss": 1.2916,
      "num_input_tokens_seen": 42588368,
      "step": 73415
    },
    {
      "epoch": 10.935358951444742,
      "grad_norm": 7.03125,
      "learning_rate": 2.5282693641809683e-05,
      "loss": 1.2883,
      "num_input_tokens_seen": 42591216,
      "step": 73420
    },
    {
      "epoch": 10.936103663985701,
      "grad_norm": 13.9375,
      "learning_rate": 2.5279444425726228e-05,
      "loss": 0.9274,
      "num_input_tokens_seen": 42594512,
      "step": 73425
    },
    {
      "epoch": 10.93684837652666,
      "grad_norm": 13.25,
      "learning_rate": 2.5276195204921837e-05,
      "loss": 1.1084,
      "num_input_tokens_seen": 42597488,
      "step": 73430
    },
    {
      "epoch": 10.93759308906762,
      "grad_norm": 11.3125,
      "learning_rate": 2.5272945979451413e-05,
      "loss": 1.3142,
      "num_input_tokens_seen": 42600528,
      "step": 73435
    },
    {
      "epoch": 10.93833780160858,
      "grad_norm": 8.9375,
      "learning_rate": 2.5269696749369844e-05,
      "loss": 0.8716,
      "num_input_tokens_seen": 42603248,
      "step": 73440
    },
    {
      "epoch": 10.939082514149538,
      "grad_norm": 9.5625,
      "learning_rate": 2.5266447514732023e-05,
      "loss": 1.1276,
      "num_input_tokens_seen": 42606096,
      "step": 73445
    },
    {
      "epoch": 10.939827226690497,
      "grad_norm": 10.0,
      "learning_rate": 2.5263198275592835e-05,
      "loss": 1.0046,
      "num_input_tokens_seen": 42609008,
      "step": 73450
    },
    {
      "epoch": 10.940571939231457,
      "grad_norm": 8.0,
      "learning_rate": 2.5259949032007186e-05,
      "loss": 0.7443,
      "num_input_tokens_seen": 42611760,
      "step": 73455
    },
    {
      "epoch": 10.941316651772416,
      "grad_norm": 12.0,
      "learning_rate": 2.5256699784029958e-05,
      "loss": 1.0357,
      "num_input_tokens_seen": 42614768,
      "step": 73460
    },
    {
      "epoch": 10.942061364313375,
      "grad_norm": 11.9375,
      "learning_rate": 2.525345053171605e-05,
      "loss": 0.9022,
      "num_input_tokens_seen": 42617552,
      "step": 73465
    },
    {
      "epoch": 10.942806076854334,
      "grad_norm": 8.5625,
      "learning_rate": 2.525020127512035e-05,
      "loss": 0.9272,
      "num_input_tokens_seen": 42620336,
      "step": 73470
    },
    {
      "epoch": 10.943550789395294,
      "grad_norm": 8.8125,
      "learning_rate": 2.524695201429776e-05,
      "loss": 0.9471,
      "num_input_tokens_seen": 42623152,
      "step": 73475
    },
    {
      "epoch": 10.944295501936253,
      "grad_norm": 11.25,
      "learning_rate": 2.5243702749303173e-05,
      "loss": 0.8897,
      "num_input_tokens_seen": 42625968,
      "step": 73480
    },
    {
      "epoch": 10.945040214477212,
      "grad_norm": 9.3125,
      "learning_rate": 2.5240453480191463e-05,
      "loss": 1.1026,
      "num_input_tokens_seen": 42628816,
      "step": 73485
    },
    {
      "epoch": 10.94578492701817,
      "grad_norm": 6.78125,
      "learning_rate": 2.5237204207017533e-05,
      "loss": 1.1634,
      "num_input_tokens_seen": 42631792,
      "step": 73490
    },
    {
      "epoch": 10.946529639559131,
      "grad_norm": 10.5625,
      "learning_rate": 2.523395492983629e-05,
      "loss": 1.194,
      "num_input_tokens_seen": 42634544,
      "step": 73495
    },
    {
      "epoch": 10.94727435210009,
      "grad_norm": 14.8125,
      "learning_rate": 2.5230705648702608e-05,
      "loss": 1.4753,
      "num_input_tokens_seen": 42637488,
      "step": 73500
    },
    {
      "epoch": 10.948019064641048,
      "grad_norm": 9.25,
      "learning_rate": 2.52274563636714e-05,
      "loss": 1.1812,
      "num_input_tokens_seen": 42640176,
      "step": 73505
    },
    {
      "epoch": 10.948763777182007,
      "grad_norm": 5.96875,
      "learning_rate": 2.5224207074797533e-05,
      "loss": 1.1911,
      "num_input_tokens_seen": 42642960,
      "step": 73510
    },
    {
      "epoch": 10.949508489722968,
      "grad_norm": 3.34375,
      "learning_rate": 2.522095778213593e-05,
      "loss": 0.8971,
      "num_input_tokens_seen": 42646192,
      "step": 73515
    },
    {
      "epoch": 10.950253202263927,
      "grad_norm": 5.875,
      "learning_rate": 2.5217708485741458e-05,
      "loss": 0.8212,
      "num_input_tokens_seen": 42649424,
      "step": 73520
    },
    {
      "epoch": 10.950997914804885,
      "grad_norm": 6.90625,
      "learning_rate": 2.5214459185669028e-05,
      "loss": 1.0704,
      "num_input_tokens_seen": 42652784,
      "step": 73525
    },
    {
      "epoch": 10.951742627345844,
      "grad_norm": 11.625,
      "learning_rate": 2.5211209881973525e-05,
      "loss": 1.1049,
      "num_input_tokens_seen": 42655856,
      "step": 73530
    },
    {
      "epoch": 10.952487339886805,
      "grad_norm": 5.875,
      "learning_rate": 2.5207960574709843e-05,
      "loss": 0.9286,
      "num_input_tokens_seen": 42658960,
      "step": 73535
    },
    {
      "epoch": 10.953232052427763,
      "grad_norm": 5.46875,
      "learning_rate": 2.520471126393289e-05,
      "loss": 1.2445,
      "num_input_tokens_seen": 42661808,
      "step": 73540
    },
    {
      "epoch": 10.953976764968722,
      "grad_norm": 9.0,
      "learning_rate": 2.5201461949697534e-05,
      "loss": 1.1414,
      "num_input_tokens_seen": 42664656,
      "step": 73545
    },
    {
      "epoch": 10.95472147750968,
      "grad_norm": 10.8125,
      "learning_rate": 2.5198212632058694e-05,
      "loss": 1.2142,
      "num_input_tokens_seen": 42667728,
      "step": 73550
    },
    {
      "epoch": 10.95546619005064,
      "grad_norm": 7.4375,
      "learning_rate": 2.519496331107125e-05,
      "loss": 1.0208,
      "num_input_tokens_seen": 42670480,
      "step": 73555
    },
    {
      "epoch": 10.9562109025916,
      "grad_norm": 5.25,
      "learning_rate": 2.51917139867901e-05,
      "loss": 1.1642,
      "num_input_tokens_seen": 42673008,
      "step": 73560
    },
    {
      "epoch": 10.956955615132559,
      "grad_norm": 5.09375,
      "learning_rate": 2.5188464659270133e-05,
      "loss": 0.9855,
      "num_input_tokens_seen": 42676016,
      "step": 73565
    },
    {
      "epoch": 10.957700327673518,
      "grad_norm": 14.625,
      "learning_rate": 2.5185215328566247e-05,
      "loss": 1.2003,
      "num_input_tokens_seen": 42678928,
      "step": 73570
    },
    {
      "epoch": 10.958445040214476,
      "grad_norm": 10.5,
      "learning_rate": 2.5181965994733343e-05,
      "loss": 1.2429,
      "num_input_tokens_seen": 42681872,
      "step": 73575
    },
    {
      "epoch": 10.959189752755437,
      "grad_norm": 14.625,
      "learning_rate": 2.5178716657826302e-05,
      "loss": 1.0778,
      "num_input_tokens_seen": 42684528,
      "step": 73580
    },
    {
      "epoch": 10.959934465296396,
      "grad_norm": 11.4375,
      "learning_rate": 2.5175467317900026e-05,
      "loss": 1.2464,
      "num_input_tokens_seen": 42687216,
      "step": 73585
    },
    {
      "epoch": 10.960679177837354,
      "grad_norm": 7.03125,
      "learning_rate": 2.517221797500941e-05,
      "loss": 0.8282,
      "num_input_tokens_seen": 42690160,
      "step": 73590
    },
    {
      "epoch": 10.961423890378313,
      "grad_norm": 16.125,
      "learning_rate": 2.516896862920935e-05,
      "loss": 1.2544,
      "num_input_tokens_seen": 42693200,
      "step": 73595
    },
    {
      "epoch": 10.962168602919274,
      "grad_norm": 11.75,
      "learning_rate": 2.5165719280554728e-05,
      "loss": 0.8181,
      "num_input_tokens_seen": 42695792,
      "step": 73600
    },
    {
      "epoch": 10.962913315460233,
      "grad_norm": 13.8125,
      "learning_rate": 2.5162469929100452e-05,
      "loss": 1.3724,
      "num_input_tokens_seen": 42698832,
      "step": 73605
    },
    {
      "epoch": 10.963658028001191,
      "grad_norm": 7.46875,
      "learning_rate": 2.5159220574901417e-05,
      "loss": 1.0089,
      "num_input_tokens_seen": 42702000,
      "step": 73610
    },
    {
      "epoch": 10.96440274054215,
      "grad_norm": 9.0625,
      "learning_rate": 2.5155971218012503e-05,
      "loss": 1.144,
      "num_input_tokens_seen": 42704880,
      "step": 73615
    },
    {
      "epoch": 10.96514745308311,
      "grad_norm": 10.4375,
      "learning_rate": 2.5152721858488615e-05,
      "loss": 1.216,
      "num_input_tokens_seen": 42707568,
      "step": 73620
    },
    {
      "epoch": 10.96589216562407,
      "grad_norm": 21.375,
      "learning_rate": 2.5149472496384645e-05,
      "loss": 1.3683,
      "num_input_tokens_seen": 42710320,
      "step": 73625
    },
    {
      "epoch": 10.966636878165028,
      "grad_norm": 17.25,
      "learning_rate": 2.5146223131755493e-05,
      "loss": 1.1543,
      "num_input_tokens_seen": 42712912,
      "step": 73630
    },
    {
      "epoch": 10.967381590705987,
      "grad_norm": 6.6875,
      "learning_rate": 2.514297376465605e-05,
      "loss": 1.0434,
      "num_input_tokens_seen": 42716208,
      "step": 73635
    },
    {
      "epoch": 10.968126303246947,
      "grad_norm": 7.90625,
      "learning_rate": 2.5139724395141207e-05,
      "loss": 1.03,
      "num_input_tokens_seen": 42719248,
      "step": 73640
    },
    {
      "epoch": 10.968871015787906,
      "grad_norm": 7.8125,
      "learning_rate": 2.513647502326587e-05,
      "loss": 0.8465,
      "num_input_tokens_seen": 42722288,
      "step": 73645
    },
    {
      "epoch": 10.969615728328865,
      "grad_norm": 13.3125,
      "learning_rate": 2.513322564908492e-05,
      "loss": 1.0173,
      "num_input_tokens_seen": 42725040,
      "step": 73650
    },
    {
      "epoch": 10.970360440869824,
      "grad_norm": 16.625,
      "learning_rate": 2.512997627265326e-05,
      "loss": 0.9688,
      "num_input_tokens_seen": 42728080,
      "step": 73655
    },
    {
      "epoch": 10.971105153410784,
      "grad_norm": 14.75,
      "learning_rate": 2.5126726894025782e-05,
      "loss": 0.8965,
      "num_input_tokens_seen": 42731312,
      "step": 73660
    },
    {
      "epoch": 10.971849865951743,
      "grad_norm": 23.0,
      "learning_rate": 2.5123477513257376e-05,
      "loss": 1.4377,
      "num_input_tokens_seen": 42734032,
      "step": 73665
    },
    {
      "epoch": 10.972594578492702,
      "grad_norm": 12.125,
      "learning_rate": 2.5120228130402955e-05,
      "loss": 0.8554,
      "num_input_tokens_seen": 42737104,
      "step": 73670
    },
    {
      "epoch": 10.97333929103366,
      "grad_norm": 18.125,
      "learning_rate": 2.5116978745517394e-05,
      "loss": 1.0923,
      "num_input_tokens_seen": 42740016,
      "step": 73675
    },
    {
      "epoch": 10.974084003574621,
      "grad_norm": 15.0,
      "learning_rate": 2.5113729358655602e-05,
      "loss": 0.932,
      "num_input_tokens_seen": 42742896,
      "step": 73680
    },
    {
      "epoch": 10.97482871611558,
      "grad_norm": 11.9375,
      "learning_rate": 2.5110479969872463e-05,
      "loss": 1.0622,
      "num_input_tokens_seen": 42746096,
      "step": 73685
    },
    {
      "epoch": 10.975573428656539,
      "grad_norm": 7.90625,
      "learning_rate": 2.510723057922288e-05,
      "loss": 1.1752,
      "num_input_tokens_seen": 42748944,
      "step": 73690
    },
    {
      "epoch": 10.976318141197497,
      "grad_norm": 7.5,
      "learning_rate": 2.510398118676174e-05,
      "loss": 0.9213,
      "num_input_tokens_seen": 42751664,
      "step": 73695
    },
    {
      "epoch": 10.977062853738456,
      "grad_norm": 11.0625,
      "learning_rate": 2.5100731792543948e-05,
      "loss": 1.1193,
      "num_input_tokens_seen": 42754640,
      "step": 73700
    },
    {
      "epoch": 10.977807566279417,
      "grad_norm": 20.125,
      "learning_rate": 2.5097482396624393e-05,
      "loss": 1.2238,
      "num_input_tokens_seen": 42758256,
      "step": 73705
    },
    {
      "epoch": 10.978552278820375,
      "grad_norm": 11.25,
      "learning_rate": 2.5094232999057975e-05,
      "loss": 1.2,
      "num_input_tokens_seen": 42761168,
      "step": 73710
    },
    {
      "epoch": 10.979296991361334,
      "grad_norm": 7.96875,
      "learning_rate": 2.5090983599899587e-05,
      "loss": 0.8258,
      "num_input_tokens_seen": 42763824,
      "step": 73715
    },
    {
      "epoch": 10.980041703902295,
      "grad_norm": 13.9375,
      "learning_rate": 2.508773419920412e-05,
      "loss": 0.9783,
      "num_input_tokens_seen": 42766864,
      "step": 73720
    },
    {
      "epoch": 10.980786416443253,
      "grad_norm": 5.5625,
      "learning_rate": 2.508448479702647e-05,
      "loss": 0.8153,
      "num_input_tokens_seen": 42770000,
      "step": 73725
    },
    {
      "epoch": 10.981531128984212,
      "grad_norm": 11.5,
      "learning_rate": 2.5081235393421537e-05,
      "loss": 0.9166,
      "num_input_tokens_seen": 42772656,
      "step": 73730
    },
    {
      "epoch": 10.982275841525171,
      "grad_norm": 7.625,
      "learning_rate": 2.507798598844422e-05,
      "loss": 1.1901,
      "num_input_tokens_seen": 42775504,
      "step": 73735
    },
    {
      "epoch": 10.98302055406613,
      "grad_norm": 10.8125,
      "learning_rate": 2.5074736582149405e-05,
      "loss": 0.9856,
      "num_input_tokens_seen": 42778256,
      "step": 73740
    },
    {
      "epoch": 10.98376526660709,
      "grad_norm": 12.4375,
      "learning_rate": 2.507148717459199e-05,
      "loss": 1.1087,
      "num_input_tokens_seen": 42781264,
      "step": 73745
    },
    {
      "epoch": 10.984509979148049,
      "grad_norm": 22.25,
      "learning_rate": 2.5068237765826875e-05,
      "loss": 1.2928,
      "num_input_tokens_seen": 42784080,
      "step": 73750
    },
    {
      "epoch": 10.985254691689008,
      "grad_norm": 8.625,
      "learning_rate": 2.5064988355908952e-05,
      "loss": 1.1552,
      "num_input_tokens_seen": 42786800,
      "step": 73755
    },
    {
      "epoch": 10.985999404229966,
      "grad_norm": 8.5625,
      "learning_rate": 2.5061738944893115e-05,
      "loss": 1.0778,
      "num_input_tokens_seen": 42789776,
      "step": 73760
    },
    {
      "epoch": 10.986744116770927,
      "grad_norm": 8.625,
      "learning_rate": 2.5058489532834262e-05,
      "loss": 1.2642,
      "num_input_tokens_seen": 42792848,
      "step": 73765
    },
    {
      "epoch": 10.987488829311886,
      "grad_norm": 5.75,
      "learning_rate": 2.5055240119787287e-05,
      "loss": 1.0551,
      "num_input_tokens_seen": 42795888,
      "step": 73770
    },
    {
      "epoch": 10.988233541852845,
      "grad_norm": 9.9375,
      "learning_rate": 2.5051990705807092e-05,
      "loss": 1.5057,
      "num_input_tokens_seen": 42799184,
      "step": 73775
    },
    {
      "epoch": 10.988978254393803,
      "grad_norm": 7.4375,
      "learning_rate": 2.504874129094856e-05,
      "loss": 1.0092,
      "num_input_tokens_seen": 42802064,
      "step": 73780
    },
    {
      "epoch": 10.989722966934764,
      "grad_norm": 7.71875,
      "learning_rate": 2.504549187526659e-05,
      "loss": 1.1384,
      "num_input_tokens_seen": 42804784,
      "step": 73785
    },
    {
      "epoch": 10.990467679475723,
      "grad_norm": 7.125,
      "learning_rate": 2.504224245881609e-05,
      "loss": 1.1561,
      "num_input_tokens_seen": 42807472,
      "step": 73790
    },
    {
      "epoch": 10.991212392016681,
      "grad_norm": 14.5625,
      "learning_rate": 2.5038993041651947e-05,
      "loss": 1.4924,
      "num_input_tokens_seen": 42810640,
      "step": 73795
    },
    {
      "epoch": 10.99195710455764,
      "grad_norm": 13.1875,
      "learning_rate": 2.503574362382905e-05,
      "loss": 1.0524,
      "num_input_tokens_seen": 42813456,
      "step": 73800
    },
    {
      "epoch": 10.9927018170986,
      "grad_norm": 13.625,
      "learning_rate": 2.5032494205402303e-05,
      "loss": 0.7573,
      "num_input_tokens_seen": 42816144,
      "step": 73805
    },
    {
      "epoch": 10.99344652963956,
      "grad_norm": 6.46875,
      "learning_rate": 2.5029244786426603e-05,
      "loss": 0.9502,
      "num_input_tokens_seen": 42818992,
      "step": 73810
    },
    {
      "epoch": 10.994191242180518,
      "grad_norm": 11.3125,
      "learning_rate": 2.5025995366956835e-05,
      "loss": 1.0884,
      "num_input_tokens_seen": 42821776,
      "step": 73815
    },
    {
      "epoch": 10.994935954721477,
      "grad_norm": 18.375,
      "learning_rate": 2.5022745947047904e-05,
      "loss": 1.675,
      "num_input_tokens_seen": 42824752,
      "step": 73820
    },
    {
      "epoch": 10.995680667262437,
      "grad_norm": 7.5625,
      "learning_rate": 2.5019496526754705e-05,
      "loss": 1.2601,
      "num_input_tokens_seen": 42827664,
      "step": 73825
    },
    {
      "epoch": 10.996425379803396,
      "grad_norm": 10.875,
      "learning_rate": 2.501624710613213e-05,
      "loss": 1.0728,
      "num_input_tokens_seen": 42830704,
      "step": 73830
    },
    {
      "epoch": 10.997170092344355,
      "grad_norm": 6.28125,
      "learning_rate": 2.501299768523508e-05,
      "loss": 0.9073,
      "num_input_tokens_seen": 42833424,
      "step": 73835
    },
    {
      "epoch": 10.997914804885314,
      "grad_norm": 9.0625,
      "learning_rate": 2.5009748264118442e-05,
      "loss": 1.1419,
      "num_input_tokens_seen": 42836496,
      "step": 73840
    },
    {
      "epoch": 10.998659517426274,
      "grad_norm": 16.875,
      "learning_rate": 2.500649884283713e-05,
      "loss": 1.0081,
      "num_input_tokens_seen": 42839184,
      "step": 73845
    },
    {
      "epoch": 10.999404229967233,
      "grad_norm": 6.59375,
      "learning_rate": 2.5003249421446012e-05,
      "loss": 1.106,
      "num_input_tokens_seen": 42842416,
      "step": 73850
    },
    {
      "epoch": 11.0,
      "eval_loss": 1.0793185234069824,
      "eval_runtime": 43.7272,
      "eval_samples_per_second": 68.241,
      "eval_steps_per_second": 17.06,
      "num_input_tokens_seen": 42844400,
      "step": 73854
    },
    {
      "epoch": 11.000148942508192,
      "grad_norm": 14.5625,
      "learning_rate": 2.5e-05,
      "loss": 1.0388,
      "num_input_tokens_seen": 42845104,
      "step": 73855
    },
    {
      "epoch": 11.00089365504915,
      "grad_norm": 7.40625,
      "learning_rate": 2.4996750578553997e-05,
      "loss": 1.0437,
      "num_input_tokens_seen": 42847952,
      "step": 73860
    },
    {
      "epoch": 11.001638367590111,
      "grad_norm": 12.25,
      "learning_rate": 2.499350115716288e-05,
      "loss": 1.287,
      "num_input_tokens_seen": 42850800,
      "step": 73865
    },
    {
      "epoch": 11.00238308013107,
      "grad_norm": 15.6875,
      "learning_rate": 2.4990251735881563e-05,
      "loss": 1.412,
      "num_input_tokens_seen": 42853648,
      "step": 73870
    },
    {
      "epoch": 11.003127792672029,
      "grad_norm": 14.125,
      "learning_rate": 2.4987002314764926e-05,
      "loss": 1.3657,
      "num_input_tokens_seen": 42856656,
      "step": 73875
    },
    {
      "epoch": 11.003872505212987,
      "grad_norm": 11.5,
      "learning_rate": 2.4983752893867877e-05,
      "loss": 0.7928,
      "num_input_tokens_seen": 42859504,
      "step": 73880
    },
    {
      "epoch": 11.004617217753948,
      "grad_norm": 11.5,
      "learning_rate": 2.4980503473245298e-05,
      "loss": 1.1828,
      "num_input_tokens_seen": 42862352,
      "step": 73885
    },
    {
      "epoch": 11.005361930294907,
      "grad_norm": 13.5625,
      "learning_rate": 2.4977254052952102e-05,
      "loss": 1.1705,
      "num_input_tokens_seen": 42865104,
      "step": 73890
    },
    {
      "epoch": 11.006106642835865,
      "grad_norm": 17.0,
      "learning_rate": 2.4974004633043168e-05,
      "loss": 0.8152,
      "num_input_tokens_seen": 42867792,
      "step": 73895
    },
    {
      "epoch": 11.006851355376824,
      "grad_norm": 20.625,
      "learning_rate": 2.4970755213573403e-05,
      "loss": 1.3084,
      "num_input_tokens_seen": 42870576,
      "step": 73900
    },
    {
      "epoch": 11.007596067917783,
      "grad_norm": 9.375,
      "learning_rate": 2.4967505794597703e-05,
      "loss": 1.012,
      "num_input_tokens_seen": 42873360,
      "step": 73905
    },
    {
      "epoch": 11.008340780458743,
      "grad_norm": 7.03125,
      "learning_rate": 2.4964256376170954e-05,
      "loss": 0.8714,
      "num_input_tokens_seen": 42876304,
      "step": 73910
    },
    {
      "epoch": 11.009085492999702,
      "grad_norm": 6.5625,
      "learning_rate": 2.4961006958348066e-05,
      "loss": 1.1072,
      "num_input_tokens_seen": 42879024,
      "step": 73915
    },
    {
      "epoch": 11.009830205540661,
      "grad_norm": 8.75,
      "learning_rate": 2.495775754118391e-05,
      "loss": 1.4265,
      "num_input_tokens_seen": 42882064,
      "step": 73920
    },
    {
      "epoch": 11.01057491808162,
      "grad_norm": 7.03125,
      "learning_rate": 2.4954508124733413e-05,
      "loss": 0.8171,
      "num_input_tokens_seen": 42885008,
      "step": 73925
    },
    {
      "epoch": 11.01131963062258,
      "grad_norm": 13.1875,
      "learning_rate": 2.495125870905144e-05,
      "loss": 1.1417,
      "num_input_tokens_seen": 42887888,
      "step": 73930
    },
    {
      "epoch": 11.012064343163539,
      "grad_norm": 7.71875,
      "learning_rate": 2.4948009294192913e-05,
      "loss": 0.896,
      "num_input_tokens_seen": 42890800,
      "step": 73935
    },
    {
      "epoch": 11.012809055704498,
      "grad_norm": 9.625,
      "learning_rate": 2.494475988021272e-05,
      "loss": 1.0942,
      "num_input_tokens_seen": 42893776,
      "step": 73940
    },
    {
      "epoch": 11.013553768245457,
      "grad_norm": 11.6875,
      "learning_rate": 2.4941510467165744e-05,
      "loss": 1.0314,
      "num_input_tokens_seen": 42896560,
      "step": 73945
    },
    {
      "epoch": 11.014298480786417,
      "grad_norm": 12.5,
      "learning_rate": 2.4938261055106894e-05,
      "loss": 0.7914,
      "num_input_tokens_seen": 42899216,
      "step": 73950
    },
    {
      "epoch": 11.015043193327376,
      "grad_norm": 11.875,
      "learning_rate": 2.493501164409105e-05,
      "loss": 0.9968,
      "num_input_tokens_seen": 42902000,
      "step": 73955
    },
    {
      "epoch": 11.015787905868335,
      "grad_norm": 18.625,
      "learning_rate": 2.493176223417313e-05,
      "loss": 1.512,
      "num_input_tokens_seen": 42904912,
      "step": 73960
    },
    {
      "epoch": 11.016532618409293,
      "grad_norm": 6.78125,
      "learning_rate": 2.4928512825408006e-05,
      "loss": 0.8552,
      "num_input_tokens_seen": 42907760,
      "step": 73965
    },
    {
      "epoch": 11.017277330950254,
      "grad_norm": 13.8125,
      "learning_rate": 2.4925263417850598e-05,
      "loss": 1.0903,
      "num_input_tokens_seen": 42910512,
      "step": 73970
    },
    {
      "epoch": 11.018022043491213,
      "grad_norm": 7.15625,
      "learning_rate": 2.4922014011555784e-05,
      "loss": 1.075,
      "num_input_tokens_seen": 42913456,
      "step": 73975
    },
    {
      "epoch": 11.018766756032171,
      "grad_norm": 11.875,
      "learning_rate": 2.4918764606578465e-05,
      "loss": 1.0408,
      "num_input_tokens_seen": 42916368,
      "step": 73980
    },
    {
      "epoch": 11.01951146857313,
      "grad_norm": 17.875,
      "learning_rate": 2.491551520297354e-05,
      "loss": 1.3248,
      "num_input_tokens_seen": 42919152,
      "step": 73985
    },
    {
      "epoch": 11.02025618111409,
      "grad_norm": 12.0625,
      "learning_rate": 2.4912265800795885e-05,
      "loss": 1.1351,
      "num_input_tokens_seen": 42922032,
      "step": 73990
    },
    {
      "epoch": 11.02100089365505,
      "grad_norm": 8.1875,
      "learning_rate": 2.4909016400100423e-05,
      "loss": 0.7475,
      "num_input_tokens_seen": 42925008,
      "step": 73995
    },
    {
      "epoch": 11.021745606196008,
      "grad_norm": 11.8125,
      "learning_rate": 2.490576700094203e-05,
      "loss": 1.1331,
      "num_input_tokens_seen": 42927920,
      "step": 74000
    },
    {
      "epoch": 11.022490318736967,
      "grad_norm": 12.0,
      "learning_rate": 2.490251760337561e-05,
      "loss": 1.3938,
      "num_input_tokens_seen": 42931120,
      "step": 74005
    },
    {
      "epoch": 11.023235031277927,
      "grad_norm": 9.6875,
      "learning_rate": 2.4899268207456055e-05,
      "loss": 1.0249,
      "num_input_tokens_seen": 42933808,
      "step": 74010
    },
    {
      "epoch": 11.023979743818886,
      "grad_norm": 8.6875,
      "learning_rate": 2.4896018813238263e-05,
      "loss": 1.0701,
      "num_input_tokens_seen": 42936720,
      "step": 74015
    },
    {
      "epoch": 11.024724456359845,
      "grad_norm": 6.53125,
      "learning_rate": 2.4892769420777134e-05,
      "loss": 1.2784,
      "num_input_tokens_seen": 42939952,
      "step": 74020
    },
    {
      "epoch": 11.025469168900804,
      "grad_norm": 9.3125,
      "learning_rate": 2.4889520030127543e-05,
      "loss": 1.1085,
      "num_input_tokens_seen": 42942864,
      "step": 74025
    },
    {
      "epoch": 11.026213881441764,
      "grad_norm": 10.5,
      "learning_rate": 2.488627064134441e-05,
      "loss": 0.918,
      "num_input_tokens_seen": 42945744,
      "step": 74030
    },
    {
      "epoch": 11.026958593982723,
      "grad_norm": 21.25,
      "learning_rate": 2.4883021254482612e-05,
      "loss": 1.3027,
      "num_input_tokens_seen": 42948688,
      "step": 74035
    },
    {
      "epoch": 11.027703306523682,
      "grad_norm": 13.5625,
      "learning_rate": 2.487977186959705e-05,
      "loss": 1.381,
      "num_input_tokens_seen": 42951344,
      "step": 74040
    },
    {
      "epoch": 11.02844801906464,
      "grad_norm": 10.0625,
      "learning_rate": 2.487652248674262e-05,
      "loss": 0.8674,
      "num_input_tokens_seen": 42953968,
      "step": 74045
    },
    {
      "epoch": 11.029192731605601,
      "grad_norm": 3.828125,
      "learning_rate": 2.4873273105974227e-05,
      "loss": 0.6671,
      "num_input_tokens_seen": 42956944,
      "step": 74050
    },
    {
      "epoch": 11.02993744414656,
      "grad_norm": 15.9375,
      "learning_rate": 2.487002372734674e-05,
      "loss": 1.2814,
      "num_input_tokens_seen": 42960016,
      "step": 74055
    },
    {
      "epoch": 11.030682156687519,
      "grad_norm": 11.3125,
      "learning_rate": 2.4866774350915084e-05,
      "loss": 1.1586,
      "num_input_tokens_seen": 42963088,
      "step": 74060
    },
    {
      "epoch": 11.031426869228477,
      "grad_norm": 15.0625,
      "learning_rate": 2.486352497673414e-05,
      "loss": 1.0162,
      "num_input_tokens_seen": 42965808,
      "step": 74065
    },
    {
      "epoch": 11.032171581769436,
      "grad_norm": 12.0,
      "learning_rate": 2.4860275604858796e-05,
      "loss": 0.9345,
      "num_input_tokens_seen": 42968912,
      "step": 74070
    },
    {
      "epoch": 11.032916294310397,
      "grad_norm": 20.25,
      "learning_rate": 2.485702623534396e-05,
      "loss": 1.0872,
      "num_input_tokens_seen": 42971888,
      "step": 74075
    },
    {
      "epoch": 11.033661006851355,
      "grad_norm": 6.03125,
      "learning_rate": 2.485377686824451e-05,
      "loss": 1.1127,
      "num_input_tokens_seen": 42975056,
      "step": 74080
    },
    {
      "epoch": 11.034405719392314,
      "grad_norm": 6.53125,
      "learning_rate": 2.485052750361536e-05,
      "loss": 0.7939,
      "num_input_tokens_seen": 42977904,
      "step": 74085
    },
    {
      "epoch": 11.035150431933273,
      "grad_norm": 11.75,
      "learning_rate": 2.4847278141511387e-05,
      "loss": 0.8314,
      "num_input_tokens_seen": 42980496,
      "step": 74090
    },
    {
      "epoch": 11.035895144474233,
      "grad_norm": 9.75,
      "learning_rate": 2.4844028781987506e-05,
      "loss": 1.1265,
      "num_input_tokens_seen": 42983440,
      "step": 74095
    },
    {
      "epoch": 11.036639857015192,
      "grad_norm": 10.1875,
      "learning_rate": 2.48407794250986e-05,
      "loss": 0.9238,
      "num_input_tokens_seen": 42986160,
      "step": 74100
    },
    {
      "epoch": 11.037384569556151,
      "grad_norm": 22.125,
      "learning_rate": 2.4837530070899557e-05,
      "loss": 1.2821,
      "num_input_tokens_seen": 42988720,
      "step": 74105
    },
    {
      "epoch": 11.03812928209711,
      "grad_norm": 12.125,
      "learning_rate": 2.483428071944528e-05,
      "loss": 0.9951,
      "num_input_tokens_seen": 42991440,
      "step": 74110
    },
    {
      "epoch": 11.03887399463807,
      "grad_norm": 9.875,
      "learning_rate": 2.483103137079066e-05,
      "loss": 1.0794,
      "num_input_tokens_seen": 42994352,
      "step": 74115
    },
    {
      "epoch": 11.039618707179029,
      "grad_norm": 7.125,
      "learning_rate": 2.4827782024990596e-05,
      "loss": 1.0959,
      "num_input_tokens_seen": 42997456,
      "step": 74120
    },
    {
      "epoch": 11.040363419719988,
      "grad_norm": 6.3125,
      "learning_rate": 2.4824532682099973e-05,
      "loss": 0.9867,
      "num_input_tokens_seen": 43000240,
      "step": 74125
    },
    {
      "epoch": 11.041108132260947,
      "grad_norm": 7.3125,
      "learning_rate": 2.48212833421737e-05,
      "loss": 0.9606,
      "num_input_tokens_seen": 43003312,
      "step": 74130
    },
    {
      "epoch": 11.041852844801907,
      "grad_norm": 10.5625,
      "learning_rate": 2.4818034005266663e-05,
      "loss": 1.1891,
      "num_input_tokens_seen": 43006096,
      "step": 74135
    },
    {
      "epoch": 11.042597557342866,
      "grad_norm": 15.1875,
      "learning_rate": 2.481478467143376e-05,
      "loss": 1.2165,
      "num_input_tokens_seen": 43009232,
      "step": 74140
    },
    {
      "epoch": 11.043342269883825,
      "grad_norm": 8.3125,
      "learning_rate": 2.4811535340729876e-05,
      "loss": 1.1669,
      "num_input_tokens_seen": 43012496,
      "step": 74145
    },
    {
      "epoch": 11.044086982424783,
      "grad_norm": 13.75,
      "learning_rate": 2.4808286013209905e-05,
      "loss": 1.2633,
      "num_input_tokens_seen": 43015312,
      "step": 74150
    },
    {
      "epoch": 11.044831694965744,
      "grad_norm": 24.625,
      "learning_rate": 2.4805036688928758e-05,
      "loss": 1.1938,
      "num_input_tokens_seen": 43018192,
      "step": 74155
    },
    {
      "epoch": 11.045576407506703,
      "grad_norm": 11.5,
      "learning_rate": 2.4801787367941305e-05,
      "loss": 1.2968,
      "num_input_tokens_seen": 43021008,
      "step": 74160
    },
    {
      "epoch": 11.046321120047661,
      "grad_norm": 9.0,
      "learning_rate": 2.4798538050302468e-05,
      "loss": 1.1863,
      "num_input_tokens_seen": 43024112,
      "step": 74165
    },
    {
      "epoch": 11.04706583258862,
      "grad_norm": 21.5,
      "learning_rate": 2.4795288736067118e-05,
      "loss": 1.2254,
      "num_input_tokens_seen": 43026992,
      "step": 74170
    },
    {
      "epoch": 11.04781054512958,
      "grad_norm": 12.0625,
      "learning_rate": 2.4792039425290163e-05,
      "loss": 1.4607,
      "num_input_tokens_seen": 43029744,
      "step": 74175
    },
    {
      "epoch": 11.04855525767054,
      "grad_norm": 20.875,
      "learning_rate": 2.4788790118026487e-05,
      "loss": 0.9862,
      "num_input_tokens_seen": 43032496,
      "step": 74180
    },
    {
      "epoch": 11.049299970211498,
      "grad_norm": 5.5,
      "learning_rate": 2.4785540814330978e-05,
      "loss": 0.8113,
      "num_input_tokens_seen": 43035408,
      "step": 74185
    },
    {
      "epoch": 11.050044682752457,
      "grad_norm": 11.75,
      "learning_rate": 2.478229151425855e-05,
      "loss": 1.5162,
      "num_input_tokens_seen": 43038352,
      "step": 74190
    },
    {
      "epoch": 11.050789395293418,
      "grad_norm": 14.4375,
      "learning_rate": 2.4779042217864077e-05,
      "loss": 1.4768,
      "num_input_tokens_seen": 43041264,
      "step": 74195
    },
    {
      "epoch": 11.051534107834376,
      "grad_norm": 6.1875,
      "learning_rate": 2.477579292520247e-05,
      "loss": 0.7337,
      "num_input_tokens_seen": 43043952,
      "step": 74200
    },
    {
      "epoch": 11.052278820375335,
      "grad_norm": 7.78125,
      "learning_rate": 2.477254363632861e-05,
      "loss": 0.911,
      "num_input_tokens_seen": 43047216,
      "step": 74205
    },
    {
      "epoch": 11.053023532916294,
      "grad_norm": 6.0625,
      "learning_rate": 2.4769294351297398e-05,
      "loss": 0.8718,
      "num_input_tokens_seen": 43049936,
      "step": 74210
    },
    {
      "epoch": 11.053768245457254,
      "grad_norm": 9.8125,
      "learning_rate": 2.4766045070163713e-05,
      "loss": 0.7661,
      "num_input_tokens_seen": 43053104,
      "step": 74215
    },
    {
      "epoch": 11.054512957998213,
      "grad_norm": 9.0,
      "learning_rate": 2.476279579298247e-05,
      "loss": 1.0451,
      "num_input_tokens_seen": 43055856,
      "step": 74220
    },
    {
      "epoch": 11.055257670539172,
      "grad_norm": 11.0,
      "learning_rate": 2.475954651980855e-05,
      "loss": 1.0142,
      "num_input_tokens_seen": 43058704,
      "step": 74225
    },
    {
      "epoch": 11.05600238308013,
      "grad_norm": 13.0,
      "learning_rate": 2.4756297250696837e-05,
      "loss": 1.5759,
      "num_input_tokens_seen": 43061776,
      "step": 74230
    },
    {
      "epoch": 11.056747095621091,
      "grad_norm": 11.875,
      "learning_rate": 2.4753047985702243e-05,
      "loss": 1.1961,
      "num_input_tokens_seen": 43064816,
      "step": 74235
    },
    {
      "epoch": 11.05749180816205,
      "grad_norm": 7.78125,
      "learning_rate": 2.474979872487965e-05,
      "loss": 1.3872,
      "num_input_tokens_seen": 43067664,
      "step": 74240
    },
    {
      "epoch": 11.058236520703009,
      "grad_norm": 8.375,
      "learning_rate": 2.474654946828396e-05,
      "loss": 1.3604,
      "num_input_tokens_seen": 43070640,
      "step": 74245
    },
    {
      "epoch": 11.058981233243967,
      "grad_norm": 14.625,
      "learning_rate": 2.474330021597004e-05,
      "loss": 0.8942,
      "num_input_tokens_seen": 43073360,
      "step": 74250
    },
    {
      "epoch": 11.059725945784926,
      "grad_norm": 6.65625,
      "learning_rate": 2.474005096799282e-05,
      "loss": 0.6358,
      "num_input_tokens_seen": 43076144,
      "step": 74255
    },
    {
      "epoch": 11.060470658325887,
      "grad_norm": 9.0,
      "learning_rate": 2.4736801724407174e-05,
      "loss": 1.5426,
      "num_input_tokens_seen": 43079056,
      "step": 74260
    },
    {
      "epoch": 11.061215370866845,
      "grad_norm": 17.25,
      "learning_rate": 2.4733552485267983e-05,
      "loss": 1.1227,
      "num_input_tokens_seen": 43081936,
      "step": 74265
    },
    {
      "epoch": 11.061960083407804,
      "grad_norm": 7.15625,
      "learning_rate": 2.4730303250630165e-05,
      "loss": 1.2787,
      "num_input_tokens_seen": 43084880,
      "step": 74270
    },
    {
      "epoch": 11.062704795948763,
      "grad_norm": 11.75,
      "learning_rate": 2.4727054020548592e-05,
      "loss": 1.1249,
      "num_input_tokens_seen": 43087856,
      "step": 74275
    },
    {
      "epoch": 11.063449508489724,
      "grad_norm": 14.375,
      "learning_rate": 2.4723804795078172e-05,
      "loss": 0.9812,
      "num_input_tokens_seen": 43090928,
      "step": 74280
    },
    {
      "epoch": 11.064194221030682,
      "grad_norm": 6.34375,
      "learning_rate": 2.4720555574273775e-05,
      "loss": 0.9197,
      "num_input_tokens_seen": 43094000,
      "step": 74285
    },
    {
      "epoch": 11.064938933571641,
      "grad_norm": 4.84375,
      "learning_rate": 2.471730635819032e-05,
      "loss": 1.0278,
      "num_input_tokens_seen": 43097040,
      "step": 74290
    },
    {
      "epoch": 11.0656836461126,
      "grad_norm": 14.0625,
      "learning_rate": 2.4714057146882676e-05,
      "loss": 1.0881,
      "num_input_tokens_seen": 43099952,
      "step": 74295
    },
    {
      "epoch": 11.06642835865356,
      "grad_norm": 8.125,
      "learning_rate": 2.4710807940405756e-05,
      "loss": 1.1981,
      "num_input_tokens_seen": 43102768,
      "step": 74300
    },
    {
      "epoch": 11.067173071194519,
      "grad_norm": 15.5,
      "learning_rate": 2.470755873881444e-05,
      "loss": 1.018,
      "num_input_tokens_seen": 43105648,
      "step": 74305
    },
    {
      "epoch": 11.067917783735478,
      "grad_norm": 10.0,
      "learning_rate": 2.470430954216362e-05,
      "loss": 0.7545,
      "num_input_tokens_seen": 43108432,
      "step": 74310
    },
    {
      "epoch": 11.068662496276437,
      "grad_norm": 7.875,
      "learning_rate": 2.4701060350508194e-05,
      "loss": 0.8599,
      "num_input_tokens_seen": 43111472,
      "step": 74315
    },
    {
      "epoch": 11.069407208817397,
      "grad_norm": 11.6875,
      "learning_rate": 2.4697811163903036e-05,
      "loss": 0.7079,
      "num_input_tokens_seen": 43114192,
      "step": 74320
    },
    {
      "epoch": 11.070151921358356,
      "grad_norm": 10.4375,
      "learning_rate": 2.4694561982403065e-05,
      "loss": 0.7781,
      "num_input_tokens_seen": 43117072,
      "step": 74325
    },
    {
      "epoch": 11.070896633899315,
      "grad_norm": 11.25,
      "learning_rate": 2.4691312806063154e-05,
      "loss": 1.011,
      "num_input_tokens_seen": 43119984,
      "step": 74330
    },
    {
      "epoch": 11.071641346440273,
      "grad_norm": 11.0625,
      "learning_rate": 2.4688063634938198e-05,
      "loss": 1.1283,
      "num_input_tokens_seen": 43122608,
      "step": 74335
    },
    {
      "epoch": 11.072386058981234,
      "grad_norm": 5.5625,
      "learning_rate": 2.46848144690831e-05,
      "loss": 1.2657,
      "num_input_tokens_seen": 43125424,
      "step": 74340
    },
    {
      "epoch": 11.073130771522193,
      "grad_norm": 10.5,
      "learning_rate": 2.468156530855273e-05,
      "loss": 0.9291,
      "num_input_tokens_seen": 43128080,
      "step": 74345
    },
    {
      "epoch": 11.073875484063151,
      "grad_norm": 18.25,
      "learning_rate": 2.4678316153401994e-05,
      "loss": 1.3375,
      "num_input_tokens_seen": 43130768,
      "step": 74350
    },
    {
      "epoch": 11.07462019660411,
      "grad_norm": 7.15625,
      "learning_rate": 2.4675067003685776e-05,
      "loss": 0.9496,
      "num_input_tokens_seen": 43133616,
      "step": 74355
    },
    {
      "epoch": 11.07536490914507,
      "grad_norm": 5.34375,
      "learning_rate": 2.4671817859458974e-05,
      "loss": 0.8057,
      "num_input_tokens_seen": 43136560,
      "step": 74360
    },
    {
      "epoch": 11.07610962168603,
      "grad_norm": 11.5625,
      "learning_rate": 2.4668568720776478e-05,
      "loss": 1.3416,
      "num_input_tokens_seen": 43139280,
      "step": 74365
    },
    {
      "epoch": 11.076854334226988,
      "grad_norm": 6.46875,
      "learning_rate": 2.466531958769317e-05,
      "loss": 0.9487,
      "num_input_tokens_seen": 43142256,
      "step": 74370
    },
    {
      "epoch": 11.077599046767947,
      "grad_norm": 5.5625,
      "learning_rate": 2.466207046026395e-05,
      "loss": 0.7974,
      "num_input_tokens_seen": 43145040,
      "step": 74375
    },
    {
      "epoch": 11.078343759308908,
      "grad_norm": 7.5,
      "learning_rate": 2.4658821338543713e-05,
      "loss": 0.8578,
      "num_input_tokens_seen": 43147952,
      "step": 74380
    },
    {
      "epoch": 11.079088471849866,
      "grad_norm": 13.75,
      "learning_rate": 2.465557222258734e-05,
      "loss": 1.1073,
      "num_input_tokens_seen": 43150704,
      "step": 74385
    },
    {
      "epoch": 11.079833184390825,
      "grad_norm": 7.21875,
      "learning_rate": 2.4652323112449716e-05,
      "loss": 0.7555,
      "num_input_tokens_seen": 43153328,
      "step": 74390
    },
    {
      "epoch": 11.080577896931784,
      "grad_norm": 8.125,
      "learning_rate": 2.464907400818575e-05,
      "loss": 1.187,
      "num_input_tokens_seen": 43156336,
      "step": 74395
    },
    {
      "epoch": 11.081322609472744,
      "grad_norm": 8.5,
      "learning_rate": 2.4645824909850316e-05,
      "loss": 0.8066,
      "num_input_tokens_seen": 43159408,
      "step": 74400
    },
    {
      "epoch": 11.082067322013703,
      "grad_norm": 9.125,
      "learning_rate": 2.4642575817498313e-05,
      "loss": 0.8882,
      "num_input_tokens_seen": 43161936,
      "step": 74405
    },
    {
      "epoch": 11.082812034554662,
      "grad_norm": 10.25,
      "learning_rate": 2.463932673118462e-05,
      "loss": 1.0422,
      "num_input_tokens_seen": 43164624,
      "step": 74410
    },
    {
      "epoch": 11.08355674709562,
      "grad_norm": 8.1875,
      "learning_rate": 2.463607765096415e-05,
      "loss": 1.0075,
      "num_input_tokens_seen": 43167856,
      "step": 74415
    },
    {
      "epoch": 11.08430145963658,
      "grad_norm": 7.78125,
      "learning_rate": 2.4632828576891774e-05,
      "loss": 0.9937,
      "num_input_tokens_seen": 43170576,
      "step": 74420
    },
    {
      "epoch": 11.08504617217754,
      "grad_norm": 4.125,
      "learning_rate": 2.4629579509022374e-05,
      "loss": 1.1217,
      "num_input_tokens_seen": 43173424,
      "step": 74425
    },
    {
      "epoch": 11.085790884718499,
      "grad_norm": 5.1875,
      "learning_rate": 2.4626330447410864e-05,
      "loss": 0.7907,
      "num_input_tokens_seen": 43176560,
      "step": 74430
    },
    {
      "epoch": 11.086535597259457,
      "grad_norm": 5.375,
      "learning_rate": 2.4623081392112117e-05,
      "loss": 1.2993,
      "num_input_tokens_seen": 43179696,
      "step": 74435
    },
    {
      "epoch": 11.087280309800416,
      "grad_norm": 10.1875,
      "learning_rate": 2.4619832343181028e-05,
      "loss": 1.0751,
      "num_input_tokens_seen": 43182576,
      "step": 74440
    },
    {
      "epoch": 11.088025022341377,
      "grad_norm": 7.40625,
      "learning_rate": 2.461658330067248e-05,
      "loss": 1.2749,
      "num_input_tokens_seen": 43185680,
      "step": 74445
    },
    {
      "epoch": 11.088769734882336,
      "grad_norm": 9.1875,
      "learning_rate": 2.4613334264641373e-05,
      "loss": 1.3401,
      "num_input_tokens_seen": 43188496,
      "step": 74450
    },
    {
      "epoch": 11.089514447423294,
      "grad_norm": 11.0,
      "learning_rate": 2.4610085235142586e-05,
      "loss": 1.323,
      "num_input_tokens_seen": 43191184,
      "step": 74455
    },
    {
      "epoch": 11.090259159964253,
      "grad_norm": 11.375,
      "learning_rate": 2.4606836212231018e-05,
      "loss": 1.0263,
      "num_input_tokens_seen": 43193872,
      "step": 74460
    },
    {
      "epoch": 11.091003872505214,
      "grad_norm": 5.34375,
      "learning_rate": 2.4603587195961554e-05,
      "loss": 1.4588,
      "num_input_tokens_seen": 43196688,
      "step": 74465
    },
    {
      "epoch": 11.091748585046172,
      "grad_norm": 12.1875,
      "learning_rate": 2.460033818638908e-05,
      "loss": 1.295,
      "num_input_tokens_seen": 43199376,
      "step": 74470
    },
    {
      "epoch": 11.092493297587131,
      "grad_norm": 4.40625,
      "learning_rate": 2.4597089183568488e-05,
      "loss": 0.9177,
      "num_input_tokens_seen": 43202064,
      "step": 74475
    },
    {
      "epoch": 11.09323801012809,
      "grad_norm": 10.9375,
      "learning_rate": 2.4593840187554654e-05,
      "loss": 1.1598,
      "num_input_tokens_seen": 43204912,
      "step": 74480
    },
    {
      "epoch": 11.09398272266905,
      "grad_norm": 8.6875,
      "learning_rate": 2.459059119840249e-05,
      "loss": 0.7664,
      "num_input_tokens_seen": 43207600,
      "step": 74485
    },
    {
      "epoch": 11.09472743521001,
      "grad_norm": 7.625,
      "learning_rate": 2.4587342216166856e-05,
      "loss": 1.0819,
      "num_input_tokens_seen": 43210704,
      "step": 74490
    },
    {
      "epoch": 11.095472147750968,
      "grad_norm": 7.03125,
      "learning_rate": 2.458409324090267e-05,
      "loss": 0.9821,
      "num_input_tokens_seen": 43213680,
      "step": 74495
    },
    {
      "epoch": 11.096216860291927,
      "grad_norm": 9.5,
      "learning_rate": 2.45808442726648e-05,
      "loss": 0.9509,
      "num_input_tokens_seen": 43216656,
      "step": 74500
    },
    {
      "epoch": 11.096961572832887,
      "grad_norm": 12.5625,
      "learning_rate": 2.4577595311508143e-05,
      "loss": 1.0887,
      "num_input_tokens_seen": 43219568,
      "step": 74505
    },
    {
      "epoch": 11.097706285373846,
      "grad_norm": 7.5625,
      "learning_rate": 2.4574346357487588e-05,
      "loss": 0.9386,
      "num_input_tokens_seen": 43222288,
      "step": 74510
    },
    {
      "epoch": 11.098450997914805,
      "grad_norm": 7.46875,
      "learning_rate": 2.4571097410658006e-05,
      "loss": 1.0464,
      "num_input_tokens_seen": 43225200,
      "step": 74515
    },
    {
      "epoch": 11.099195710455763,
      "grad_norm": 17.125,
      "learning_rate": 2.4567848471074307e-05,
      "loss": 1.2323,
      "num_input_tokens_seen": 43228112,
      "step": 74520
    },
    {
      "epoch": 11.099940422996724,
      "grad_norm": 11.125,
      "learning_rate": 2.4564599538791362e-05,
      "loss": 1.1667,
      "num_input_tokens_seen": 43231088,
      "step": 74525
    },
    {
      "epoch": 11.100685135537683,
      "grad_norm": 6.71875,
      "learning_rate": 2.4561350613864074e-05,
      "loss": 0.9262,
      "num_input_tokens_seen": 43233968,
      "step": 74530
    },
    {
      "epoch": 11.101429848078642,
      "grad_norm": 5.03125,
      "learning_rate": 2.4558101696347315e-05,
      "loss": 0.9042,
      "num_input_tokens_seen": 43236816,
      "step": 74535
    },
    {
      "epoch": 11.1021745606196,
      "grad_norm": 14.875,
      "learning_rate": 2.4554852786295985e-05,
      "loss": 0.9394,
      "num_input_tokens_seen": 43239568,
      "step": 74540
    },
    {
      "epoch": 11.10291927316056,
      "grad_norm": 12.25,
      "learning_rate": 2.4551603883764963e-05,
      "loss": 0.9972,
      "num_input_tokens_seen": 43242480,
      "step": 74545
    },
    {
      "epoch": 11.10366398570152,
      "grad_norm": 4.3125,
      "learning_rate": 2.454835498880913e-05,
      "loss": 0.8716,
      "num_input_tokens_seen": 43245904,
      "step": 74550
    },
    {
      "epoch": 11.104408698242478,
      "grad_norm": 12.8125,
      "learning_rate": 2.454510610148339e-05,
      "loss": 1.3393,
      "num_input_tokens_seen": 43248816,
      "step": 74555
    },
    {
      "epoch": 11.105153410783437,
      "grad_norm": 5.6875,
      "learning_rate": 2.454185722184261e-05,
      "loss": 1.2647,
      "num_input_tokens_seen": 43251824,
      "step": 74560
    },
    {
      "epoch": 11.105898123324398,
      "grad_norm": 8.125,
      "learning_rate": 2.4538608349941695e-05,
      "loss": 1.2982,
      "num_input_tokens_seen": 43254576,
      "step": 74565
    },
    {
      "epoch": 11.106642835865356,
      "grad_norm": 12.1875,
      "learning_rate": 2.453535948583552e-05,
      "loss": 0.9637,
      "num_input_tokens_seen": 43257360,
      "step": 74570
    },
    {
      "epoch": 11.107387548406315,
      "grad_norm": 21.0,
      "learning_rate": 2.453211062957898e-05,
      "loss": 1.2583,
      "num_input_tokens_seen": 43260240,
      "step": 74575
    },
    {
      "epoch": 11.108132260947274,
      "grad_norm": 11.875,
      "learning_rate": 2.4528861781226942e-05,
      "loss": 1.1516,
      "num_input_tokens_seen": 43263280,
      "step": 74580
    },
    {
      "epoch": 11.108876973488233,
      "grad_norm": 13.5625,
      "learning_rate": 2.4525612940834315e-05,
      "loss": 1.1803,
      "num_input_tokens_seen": 43266128,
      "step": 74585
    },
    {
      "epoch": 11.109621686029193,
      "grad_norm": 16.25,
      "learning_rate": 2.452236410845598e-05,
      "loss": 1.3474,
      "num_input_tokens_seen": 43268944,
      "step": 74590
    },
    {
      "epoch": 11.110366398570152,
      "grad_norm": 10.5,
      "learning_rate": 2.4519115284146806e-05,
      "loss": 1.1744,
      "num_input_tokens_seen": 43271760,
      "step": 74595
    },
    {
      "epoch": 11.11111111111111,
      "grad_norm": 7.625,
      "learning_rate": 2.45158664679617e-05,
      "loss": 1.3274,
      "num_input_tokens_seen": 43274576,
      "step": 74600
    },
    {
      "epoch": 11.11185582365207,
      "grad_norm": 11.0,
      "learning_rate": 2.4512617659955532e-05,
      "loss": 0.9673,
      "num_input_tokens_seen": 43277424,
      "step": 74605
    },
    {
      "epoch": 11.11260053619303,
      "grad_norm": 7.96875,
      "learning_rate": 2.45093688601832e-05,
      "loss": 1.2333,
      "num_input_tokens_seen": 43280144,
      "step": 74610
    },
    {
      "epoch": 11.113345248733989,
      "grad_norm": 9.625,
      "learning_rate": 2.450612006869957e-05,
      "loss": 1.3011,
      "num_input_tokens_seen": 43283312,
      "step": 74615
    },
    {
      "epoch": 11.114089961274948,
      "grad_norm": 11.8125,
      "learning_rate": 2.450287128555955e-05,
      "loss": 1.1814,
      "num_input_tokens_seen": 43286320,
      "step": 74620
    },
    {
      "epoch": 11.114834673815906,
      "grad_norm": 10.5625,
      "learning_rate": 2.4499622510818017e-05,
      "loss": 1.096,
      "num_input_tokens_seen": 43289040,
      "step": 74625
    },
    {
      "epoch": 11.115579386356867,
      "grad_norm": 4.71875,
      "learning_rate": 2.4496373744529842e-05,
      "loss": 1.1601,
      "num_input_tokens_seen": 43291824,
      "step": 74630
    },
    {
      "epoch": 11.116324098897826,
      "grad_norm": 19.0,
      "learning_rate": 2.449312498674993e-05,
      "loss": 1.2486,
      "num_input_tokens_seen": 43294608,
      "step": 74635
    },
    {
      "epoch": 11.117068811438784,
      "grad_norm": 14.3125,
      "learning_rate": 2.448987623753315e-05,
      "loss": 1.1195,
      "num_input_tokens_seen": 43297808,
      "step": 74640
    },
    {
      "epoch": 11.117813523979743,
      "grad_norm": 11.125,
      "learning_rate": 2.44866274969344e-05,
      "loss": 1.1967,
      "num_input_tokens_seen": 43300464,
      "step": 74645
    },
    {
      "epoch": 11.118558236520704,
      "grad_norm": 8.4375,
      "learning_rate": 2.4483378765008543e-05,
      "loss": 1.3727,
      "num_input_tokens_seen": 43303344,
      "step": 74650
    },
    {
      "epoch": 11.119302949061662,
      "grad_norm": 5.0,
      "learning_rate": 2.4480130041810493e-05,
      "loss": 0.9,
      "num_input_tokens_seen": 43306256,
      "step": 74655
    },
    {
      "epoch": 11.120047661602621,
      "grad_norm": 10.5,
      "learning_rate": 2.4476881327395108e-05,
      "loss": 1.0281,
      "num_input_tokens_seen": 43309136,
      "step": 74660
    },
    {
      "epoch": 11.12079237414358,
      "grad_norm": 6.375,
      "learning_rate": 2.447363262181728e-05,
      "loss": 0.9935,
      "num_input_tokens_seen": 43312016,
      "step": 74665
    },
    {
      "epoch": 11.12153708668454,
      "grad_norm": 6.5625,
      "learning_rate": 2.44703839251319e-05,
      "loss": 1.1061,
      "num_input_tokens_seen": 43314800,
      "step": 74670
    },
    {
      "epoch": 11.1222817992255,
      "grad_norm": 7.8125,
      "learning_rate": 2.4467135237393842e-05,
      "loss": 0.9169,
      "num_input_tokens_seen": 43317424,
      "step": 74675
    },
    {
      "epoch": 11.123026511766458,
      "grad_norm": 7.90625,
      "learning_rate": 2.4463886558658e-05,
      "loss": 0.8146,
      "num_input_tokens_seen": 43320240,
      "step": 74680
    },
    {
      "epoch": 11.123771224307417,
      "grad_norm": 14.0,
      "learning_rate": 2.4460637888979234e-05,
      "loss": 0.9469,
      "num_input_tokens_seen": 43323280,
      "step": 74685
    },
    {
      "epoch": 11.124515936848377,
      "grad_norm": 16.375,
      "learning_rate": 2.4457389228412457e-05,
      "loss": 1.2018,
      "num_input_tokens_seen": 43326192,
      "step": 74690
    },
    {
      "epoch": 11.125260649389336,
      "grad_norm": 17.0,
      "learning_rate": 2.445414057701253e-05,
      "loss": 1.2489,
      "num_input_tokens_seen": 43329008,
      "step": 74695
    },
    {
      "epoch": 11.126005361930295,
      "grad_norm": 14.1875,
      "learning_rate": 2.4450891934834345e-05,
      "loss": 1.2928,
      "num_input_tokens_seen": 43331568,
      "step": 74700
    },
    {
      "epoch": 11.126750074471254,
      "grad_norm": 6.75,
      "learning_rate": 2.4447643301932785e-05,
      "loss": 0.9391,
      "num_input_tokens_seen": 43334512,
      "step": 74705
    },
    {
      "epoch": 11.127494787012214,
      "grad_norm": 8.3125,
      "learning_rate": 2.4444394678362727e-05,
      "loss": 0.9341,
      "num_input_tokens_seen": 43337392,
      "step": 74710
    },
    {
      "epoch": 11.128239499553173,
      "grad_norm": 17.5,
      "learning_rate": 2.444114606417906e-05,
      "loss": 1.4507,
      "num_input_tokens_seen": 43340144,
      "step": 74715
    },
    {
      "epoch": 11.128984212094132,
      "grad_norm": 14.9375,
      "learning_rate": 2.4437897459436654e-05,
      "loss": 1.2389,
      "num_input_tokens_seen": 43342960,
      "step": 74720
    },
    {
      "epoch": 11.12972892463509,
      "grad_norm": 6.6875,
      "learning_rate": 2.4434648864190405e-05,
      "loss": 1.1149,
      "num_input_tokens_seen": 43346160,
      "step": 74725
    },
    {
      "epoch": 11.13047363717605,
      "grad_norm": 9.625,
      "learning_rate": 2.4431400278495188e-05,
      "loss": 1.1068,
      "num_input_tokens_seen": 43348880,
      "step": 74730
    },
    {
      "epoch": 11.13121834971701,
      "grad_norm": 5.53125,
      "learning_rate": 2.442815170240589e-05,
      "loss": 1.0541,
      "num_input_tokens_seen": 43351504,
      "step": 74735
    },
    {
      "epoch": 11.131963062257968,
      "grad_norm": 13.125,
      "learning_rate": 2.442490313597738e-05,
      "loss": 1.3729,
      "num_input_tokens_seen": 43354160,
      "step": 74740
    },
    {
      "epoch": 11.132707774798927,
      "grad_norm": 19.375,
      "learning_rate": 2.4421654579264553e-05,
      "loss": 1.4518,
      "num_input_tokens_seen": 43356784,
      "step": 74745
    },
    {
      "epoch": 11.133452487339888,
      "grad_norm": 11.1875,
      "learning_rate": 2.4418406032322286e-05,
      "loss": 0.9042,
      "num_input_tokens_seen": 43359440,
      "step": 74750
    },
    {
      "epoch": 11.134197199880846,
      "grad_norm": 12.6875,
      "learning_rate": 2.4415157495205445e-05,
      "loss": 0.9874,
      "num_input_tokens_seen": 43361968,
      "step": 74755
    },
    {
      "epoch": 11.134941912421805,
      "grad_norm": 10.125,
      "learning_rate": 2.4411908967968938e-05,
      "loss": 0.8364,
      "num_input_tokens_seen": 43364656,
      "step": 74760
    },
    {
      "epoch": 11.135686624962764,
      "grad_norm": 12.8125,
      "learning_rate": 2.4408660450667626e-05,
      "loss": 1.4532,
      "num_input_tokens_seen": 43367440,
      "step": 74765
    },
    {
      "epoch": 11.136431337503723,
      "grad_norm": 7.53125,
      "learning_rate": 2.4405411943356398e-05,
      "loss": 0.8766,
      "num_input_tokens_seen": 43370544,
      "step": 74770
    },
    {
      "epoch": 11.137176050044683,
      "grad_norm": 16.25,
      "learning_rate": 2.440216344609012e-05,
      "loss": 1.2921,
      "num_input_tokens_seen": 43373456,
      "step": 74775
    },
    {
      "epoch": 11.137920762585642,
      "grad_norm": 10.0625,
      "learning_rate": 2.4398914958923695e-05,
      "loss": 0.9821,
      "num_input_tokens_seen": 43376336,
      "step": 74780
    },
    {
      "epoch": 11.1386654751266,
      "grad_norm": 10.125,
      "learning_rate": 2.439566648191199e-05,
      "loss": 0.9584,
      "num_input_tokens_seen": 43379120,
      "step": 74785
    },
    {
      "epoch": 11.13941018766756,
      "grad_norm": 15.1875,
      "learning_rate": 2.439241801510987e-05,
      "loss": 1.1647,
      "num_input_tokens_seen": 43382224,
      "step": 74790
    },
    {
      "epoch": 11.14015490020852,
      "grad_norm": 17.75,
      "learning_rate": 2.4389169558572247e-05,
      "loss": 1.0303,
      "num_input_tokens_seen": 43384880,
      "step": 74795
    },
    {
      "epoch": 11.140899612749479,
      "grad_norm": 16.875,
      "learning_rate": 2.4385921112353978e-05,
      "loss": 0.8694,
      "num_input_tokens_seen": 43387536,
      "step": 74800
    },
    {
      "epoch": 11.141644325290438,
      "grad_norm": 6.40625,
      "learning_rate": 2.438267267650995e-05,
      "loss": 0.8325,
      "num_input_tokens_seen": 43390608,
      "step": 74805
    },
    {
      "epoch": 11.142389037831396,
      "grad_norm": 12.0625,
      "learning_rate": 2.4379424251095034e-05,
      "loss": 0.9261,
      "num_input_tokens_seen": 43393584,
      "step": 74810
    },
    {
      "epoch": 11.143133750372357,
      "grad_norm": 7.4375,
      "learning_rate": 2.4376175836164122e-05,
      "loss": 1.3172,
      "num_input_tokens_seen": 43396368,
      "step": 74815
    },
    {
      "epoch": 11.143878462913316,
      "grad_norm": 10.3125,
      "learning_rate": 2.4372927431772076e-05,
      "loss": 1.0659,
      "num_input_tokens_seen": 43399152,
      "step": 74820
    },
    {
      "epoch": 11.144623175454274,
      "grad_norm": 10.3125,
      "learning_rate": 2.4369679037973792e-05,
      "loss": 1.0235,
      "num_input_tokens_seen": 43402448,
      "step": 74825
    },
    {
      "epoch": 11.145367887995233,
      "grad_norm": 6.5625,
      "learning_rate": 2.436643065482414e-05,
      "loss": 1.2597,
      "num_input_tokens_seen": 43405360,
      "step": 74830
    },
    {
      "epoch": 11.146112600536194,
      "grad_norm": 8.5,
      "learning_rate": 2.4363182282377994e-05,
      "loss": 0.9758,
      "num_input_tokens_seen": 43408592,
      "step": 74835
    },
    {
      "epoch": 11.146857313077152,
      "grad_norm": 14.5625,
      "learning_rate": 2.4359933920690242e-05,
      "loss": 1.4729,
      "num_input_tokens_seen": 43411856,
      "step": 74840
    },
    {
      "epoch": 11.147602025618111,
      "grad_norm": 13.0,
      "learning_rate": 2.4356685569815742e-05,
      "loss": 0.9278,
      "num_input_tokens_seen": 43414704,
      "step": 74845
    },
    {
      "epoch": 11.14834673815907,
      "grad_norm": 6.90625,
      "learning_rate": 2.43534372298094e-05,
      "loss": 0.8394,
      "num_input_tokens_seen": 43417776,
      "step": 74850
    },
    {
      "epoch": 11.14909145070003,
      "grad_norm": 12.5,
      "learning_rate": 2.4350188900726068e-05,
      "loss": 1.1596,
      "num_input_tokens_seen": 43421104,
      "step": 74855
    },
    {
      "epoch": 11.14983616324099,
      "grad_norm": 8.375,
      "learning_rate": 2.4346940582620644e-05,
      "loss": 0.9405,
      "num_input_tokens_seen": 43424016,
      "step": 74860
    },
    {
      "epoch": 11.150580875781948,
      "grad_norm": 7.875,
      "learning_rate": 2.4343692275547993e-05,
      "loss": 1.0054,
      "num_input_tokens_seen": 43426864,
      "step": 74865
    },
    {
      "epoch": 11.151325588322907,
      "grad_norm": 13.1875,
      "learning_rate": 2.434044397956299e-05,
      "loss": 1.1195,
      "num_input_tokens_seen": 43429776,
      "step": 74870
    },
    {
      "epoch": 11.152070300863867,
      "grad_norm": 11.625,
      "learning_rate": 2.433719569472052e-05,
      "loss": 1.0997,
      "num_input_tokens_seen": 43432528,
      "step": 74875
    },
    {
      "epoch": 11.152815013404826,
      "grad_norm": 18.25,
      "learning_rate": 2.433394742107545e-05,
      "loss": 0.9534,
      "num_input_tokens_seen": 43435984,
      "step": 74880
    },
    {
      "epoch": 11.153559725945785,
      "grad_norm": 14.375,
      "learning_rate": 2.4330699158682666e-05,
      "loss": 1.0314,
      "num_input_tokens_seen": 43438800,
      "step": 74885
    },
    {
      "epoch": 11.154304438486744,
      "grad_norm": 6.625,
      "learning_rate": 2.432745090759703e-05,
      "loss": 1.0837,
      "num_input_tokens_seen": 43441968,
      "step": 74890
    },
    {
      "epoch": 11.155049151027704,
      "grad_norm": 11.5,
      "learning_rate": 2.432420266787344e-05,
      "loss": 1.402,
      "num_input_tokens_seen": 43444880,
      "step": 74895
    },
    {
      "epoch": 11.155793863568663,
      "grad_norm": 6.78125,
      "learning_rate": 2.4320954439566752e-05,
      "loss": 1.122,
      "num_input_tokens_seen": 43447792,
      "step": 74900
    },
    {
      "epoch": 11.156538576109622,
      "grad_norm": 14.8125,
      "learning_rate": 2.4317706222731853e-05,
      "loss": 1.481,
      "num_input_tokens_seen": 43450704,
      "step": 74905
    },
    {
      "epoch": 11.15728328865058,
      "grad_norm": 12.375,
      "learning_rate": 2.4314458017423618e-05,
      "loss": 1.051,
      "num_input_tokens_seen": 43453456,
      "step": 74910
    },
    {
      "epoch": 11.158028001191541,
      "grad_norm": 8.3125,
      "learning_rate": 2.43112098236969e-05,
      "loss": 1.095,
      "num_input_tokens_seen": 43456304,
      "step": 74915
    },
    {
      "epoch": 11.1587727137325,
      "grad_norm": 12.375,
      "learning_rate": 2.430796164160661e-05,
      "loss": 1.1281,
      "num_input_tokens_seen": 43459184,
      "step": 74920
    },
    {
      "epoch": 11.159517426273458,
      "grad_norm": 8.75,
      "learning_rate": 2.430471347120759e-05,
      "loss": 0.8842,
      "num_input_tokens_seen": 43461936,
      "step": 74925
    },
    {
      "epoch": 11.160262138814417,
      "grad_norm": 6.0,
      "learning_rate": 2.4301465312554743e-05,
      "loss": 1.1983,
      "num_input_tokens_seen": 43464656,
      "step": 74930
    },
    {
      "epoch": 11.161006851355376,
      "grad_norm": 5.25,
      "learning_rate": 2.429821716570292e-05,
      "loss": 0.8816,
      "num_input_tokens_seen": 43467632,
      "step": 74935
    },
    {
      "epoch": 11.161751563896336,
      "grad_norm": 17.875,
      "learning_rate": 2.4294969030707013e-05,
      "loss": 1.5624,
      "num_input_tokens_seen": 43470352,
      "step": 74940
    },
    {
      "epoch": 11.162496276437295,
      "grad_norm": 6.59375,
      "learning_rate": 2.4291720907621886e-05,
      "loss": 0.8208,
      "num_input_tokens_seen": 43473328,
      "step": 74945
    },
    {
      "epoch": 11.163240988978254,
      "grad_norm": 12.625,
      "learning_rate": 2.4288472796502407e-05,
      "loss": 1.0506,
      "num_input_tokens_seen": 43476176,
      "step": 74950
    },
    {
      "epoch": 11.163985701519213,
      "grad_norm": 8.625,
      "learning_rate": 2.4285224697403464e-05,
      "loss": 0.7342,
      "num_input_tokens_seen": 43479088,
      "step": 74955
    },
    {
      "epoch": 11.164730414060173,
      "grad_norm": 15.25,
      "learning_rate": 2.4281976610379914e-05,
      "loss": 1.0606,
      "num_input_tokens_seen": 43482256,
      "step": 74960
    },
    {
      "epoch": 11.165475126601132,
      "grad_norm": 6.59375,
      "learning_rate": 2.427872853548665e-05,
      "loss": 1.1219,
      "num_input_tokens_seen": 43484880,
      "step": 74965
    },
    {
      "epoch": 11.16621983914209,
      "grad_norm": 12.3125,
      "learning_rate": 2.427548047277853e-05,
      "loss": 1.2749,
      "num_input_tokens_seen": 43487696,
      "step": 74970
    },
    {
      "epoch": 11.16696455168305,
      "grad_norm": 13.125,
      "learning_rate": 2.4272232422310436e-05,
      "loss": 1.0495,
      "num_input_tokens_seen": 43490480,
      "step": 74975
    },
    {
      "epoch": 11.16770926422401,
      "grad_norm": 13.9375,
      "learning_rate": 2.4268984384137225e-05,
      "loss": 1.154,
      "num_input_tokens_seen": 43493200,
      "step": 74980
    },
    {
      "epoch": 11.168453976764969,
      "grad_norm": 11.25,
      "learning_rate": 2.426573635831379e-05,
      "loss": 1.1047,
      "num_input_tokens_seen": 43496016,
      "step": 74985
    },
    {
      "epoch": 11.169198689305928,
      "grad_norm": 7.40625,
      "learning_rate": 2.426248834489499e-05,
      "loss": 0.9997,
      "num_input_tokens_seen": 43498896,
      "step": 74990
    },
    {
      "epoch": 11.169943401846886,
      "grad_norm": 11.9375,
      "learning_rate": 2.4259240343935695e-05,
      "loss": 1.5728,
      "num_input_tokens_seen": 43501968,
      "step": 74995
    },
    {
      "epoch": 11.170688114387847,
      "grad_norm": 14.5,
      "learning_rate": 2.4255992355490788e-05,
      "loss": 1.1146,
      "num_input_tokens_seen": 43504496,
      "step": 75000
    },
    {
      "epoch": 11.171432826928806,
      "grad_norm": 10.25,
      "learning_rate": 2.425274437961513e-05,
      "loss": 0.9009,
      "num_input_tokens_seen": 43507600,
      "step": 75005
    },
    {
      "epoch": 11.172177539469764,
      "grad_norm": 11.375,
      "learning_rate": 2.42494964163636e-05,
      "loss": 1.1652,
      "num_input_tokens_seen": 43510704,
      "step": 75010
    },
    {
      "epoch": 11.172922252010723,
      "grad_norm": 6.90625,
      "learning_rate": 2.4246248465791058e-05,
      "loss": 1.0338,
      "num_input_tokens_seen": 43513584,
      "step": 75015
    },
    {
      "epoch": 11.173666964551684,
      "grad_norm": 14.625,
      "learning_rate": 2.4243000527952388e-05,
      "loss": 1.0543,
      "num_input_tokens_seen": 43516432,
      "step": 75020
    },
    {
      "epoch": 11.174411677092642,
      "grad_norm": 10.8125,
      "learning_rate": 2.423975260290246e-05,
      "loss": 1.1918,
      "num_input_tokens_seen": 43519312,
      "step": 75025
    },
    {
      "epoch": 11.175156389633601,
      "grad_norm": 5.25,
      "learning_rate": 2.4236504690696125e-05,
      "loss": 1.0506,
      "num_input_tokens_seen": 43522224,
      "step": 75030
    },
    {
      "epoch": 11.17590110217456,
      "grad_norm": 9.3125,
      "learning_rate": 2.423325679138828e-05,
      "loss": 1.1675,
      "num_input_tokens_seen": 43524944,
      "step": 75035
    },
    {
      "epoch": 11.17664581471552,
      "grad_norm": 13.9375,
      "learning_rate": 2.4230008905033774e-05,
      "loss": 1.2165,
      "num_input_tokens_seen": 43527536,
      "step": 75040
    },
    {
      "epoch": 11.17739052725648,
      "grad_norm": 7.1875,
      "learning_rate": 2.4226761031687496e-05,
      "loss": 1.0673,
      "num_input_tokens_seen": 43530160,
      "step": 75045
    },
    {
      "epoch": 11.178135239797438,
      "grad_norm": 4.71875,
      "learning_rate": 2.4223513171404288e-05,
      "loss": 1.1031,
      "num_input_tokens_seen": 43533168,
      "step": 75050
    },
    {
      "epoch": 11.178879952338397,
      "grad_norm": 9.3125,
      "learning_rate": 2.4220265324239045e-05,
      "loss": 0.8002,
      "num_input_tokens_seen": 43536016,
      "step": 75055
    },
    {
      "epoch": 11.179624664879357,
      "grad_norm": 12.9375,
      "learning_rate": 2.4217017490246626e-05,
      "loss": 1.194,
      "num_input_tokens_seen": 43539312,
      "step": 75060
    },
    {
      "epoch": 11.180369377420316,
      "grad_norm": 11.125,
      "learning_rate": 2.4213769669481906e-05,
      "loss": 0.9152,
      "num_input_tokens_seen": 43542128,
      "step": 75065
    },
    {
      "epoch": 11.181114089961275,
      "grad_norm": 16.75,
      "learning_rate": 2.421052186199974e-05,
      "loss": 1.1859,
      "num_input_tokens_seen": 43545200,
      "step": 75070
    },
    {
      "epoch": 11.181858802502234,
      "grad_norm": 13.375,
      "learning_rate": 2.420727406785501e-05,
      "loss": 0.9414,
      "num_input_tokens_seen": 43548048,
      "step": 75075
    },
    {
      "epoch": 11.182603515043194,
      "grad_norm": 7.8125,
      "learning_rate": 2.4204026287102578e-05,
      "loss": 0.9475,
      "num_input_tokens_seen": 43550768,
      "step": 75080
    },
    {
      "epoch": 11.183348227584153,
      "grad_norm": 13.25,
      "learning_rate": 2.4200778519797306e-05,
      "loss": 1.4349,
      "num_input_tokens_seen": 43553840,
      "step": 75085
    },
    {
      "epoch": 11.184092940125112,
      "grad_norm": 8.5,
      "learning_rate": 2.4197530765994082e-05,
      "loss": 1.2982,
      "num_input_tokens_seen": 43557232,
      "step": 75090
    },
    {
      "epoch": 11.18483765266607,
      "grad_norm": 14.375,
      "learning_rate": 2.4194283025747748e-05,
      "loss": 1.214,
      "num_input_tokens_seen": 43560240,
      "step": 75095
    },
    {
      "epoch": 11.18558236520703,
      "grad_norm": 6.65625,
      "learning_rate": 2.4191035299113194e-05,
      "loss": 1.029,
      "num_input_tokens_seen": 43563056,
      "step": 75100
    },
    {
      "epoch": 11.18632707774799,
      "grad_norm": 7.78125,
      "learning_rate": 2.418778758614526e-05,
      "loss": 1.0652,
      "num_input_tokens_seen": 43566000,
      "step": 75105
    },
    {
      "epoch": 11.187071790288948,
      "grad_norm": 9.875,
      "learning_rate": 2.4184539886898846e-05,
      "loss": 0.986,
      "num_input_tokens_seen": 43568944,
      "step": 75110
    },
    {
      "epoch": 11.187816502829907,
      "grad_norm": 20.5,
      "learning_rate": 2.41812922014288e-05,
      "loss": 1.4956,
      "num_input_tokens_seen": 43571664,
      "step": 75115
    },
    {
      "epoch": 11.188561215370866,
      "grad_norm": 14.8125,
      "learning_rate": 2.417804452978998e-05,
      "loss": 1.3936,
      "num_input_tokens_seen": 43575184,
      "step": 75120
    },
    {
      "epoch": 11.189305927911827,
      "grad_norm": 15.1875,
      "learning_rate": 2.417479687203727e-05,
      "loss": 1.0747,
      "num_input_tokens_seen": 43577936,
      "step": 75125
    },
    {
      "epoch": 11.190050640452785,
      "grad_norm": 5.15625,
      "learning_rate": 2.4171549228225524e-05,
      "loss": 0.9798,
      "num_input_tokens_seen": 43580912,
      "step": 75130
    },
    {
      "epoch": 11.190795352993744,
      "grad_norm": 13.125,
      "learning_rate": 2.4168301598409617e-05,
      "loss": 0.9535,
      "num_input_tokens_seen": 43583984,
      "step": 75135
    },
    {
      "epoch": 11.191540065534703,
      "grad_norm": 10.75,
      "learning_rate": 2.41650539826444e-05,
      "loss": 1.5194,
      "num_input_tokens_seen": 43587024,
      "step": 75140
    },
    {
      "epoch": 11.192284778075663,
      "grad_norm": 11.5625,
      "learning_rate": 2.416180638098476e-05,
      "loss": 1.1902,
      "num_input_tokens_seen": 43589872,
      "step": 75145
    },
    {
      "epoch": 11.193029490616622,
      "grad_norm": 8.1875,
      "learning_rate": 2.4158558793485545e-05,
      "loss": 1.0329,
      "num_input_tokens_seen": 43592976,
      "step": 75150
    },
    {
      "epoch": 11.19377420315758,
      "grad_norm": 9.25,
      "learning_rate": 2.4155311220201616e-05,
      "loss": 0.9378,
      "num_input_tokens_seen": 43595696,
      "step": 75155
    },
    {
      "epoch": 11.19451891569854,
      "grad_norm": 4.96875,
      "learning_rate": 2.4152063661187855e-05,
      "loss": 1.2295,
      "num_input_tokens_seen": 43598448,
      "step": 75160
    },
    {
      "epoch": 11.1952636282395,
      "grad_norm": 12.5625,
      "learning_rate": 2.414881611649911e-05,
      "loss": 1.037,
      "num_input_tokens_seen": 43601712,
      "step": 75165
    },
    {
      "epoch": 11.196008340780459,
      "grad_norm": 7.03125,
      "learning_rate": 2.414556858619026e-05,
      "loss": 1.0931,
      "num_input_tokens_seen": 43604752,
      "step": 75170
    },
    {
      "epoch": 11.196753053321418,
      "grad_norm": 9.5,
      "learning_rate": 2.4142321070316147e-05,
      "loss": 1.0243,
      "num_input_tokens_seen": 43607536,
      "step": 75175
    },
    {
      "epoch": 11.197497765862376,
      "grad_norm": 9.125,
      "learning_rate": 2.4139073568931658e-05,
      "loss": 1.0146,
      "num_input_tokens_seen": 43610544,
      "step": 75180
    },
    {
      "epoch": 11.198242478403337,
      "grad_norm": 7.59375,
      "learning_rate": 2.4135826082091636e-05,
      "loss": 0.9815,
      "num_input_tokens_seen": 43613232,
      "step": 75185
    },
    {
      "epoch": 11.198987190944296,
      "grad_norm": 15.9375,
      "learning_rate": 2.4132578609850966e-05,
      "loss": 0.8163,
      "num_input_tokens_seen": 43616272,
      "step": 75190
    },
    {
      "epoch": 11.199731903485254,
      "grad_norm": 7.34375,
      "learning_rate": 2.41293311522645e-05,
      "loss": 0.8925,
      "num_input_tokens_seen": 43619376,
      "step": 75195
    },
    {
      "epoch": 11.200476616026213,
      "grad_norm": 11.1875,
      "learning_rate": 2.4126083709387094e-05,
      "loss": 1.4398,
      "num_input_tokens_seen": 43622416,
      "step": 75200
    },
    {
      "epoch": 11.201221328567174,
      "grad_norm": 16.5,
      "learning_rate": 2.4122836281273618e-05,
      "loss": 1.0428,
      "num_input_tokens_seen": 43625488,
      "step": 75205
    },
    {
      "epoch": 11.201966041108133,
      "grad_norm": 8.0,
      "learning_rate": 2.411958886797892e-05,
      "loss": 1.1059,
      "num_input_tokens_seen": 43628464,
      "step": 75210
    },
    {
      "epoch": 11.202710753649091,
      "grad_norm": 9.8125,
      "learning_rate": 2.4116341469557888e-05,
      "loss": 1.2757,
      "num_input_tokens_seen": 43631504,
      "step": 75215
    },
    {
      "epoch": 11.20345546619005,
      "grad_norm": 7.09375,
      "learning_rate": 2.4113094086065356e-05,
      "loss": 1.2457,
      "num_input_tokens_seen": 43634448,
      "step": 75220
    },
    {
      "epoch": 11.20420017873101,
      "grad_norm": 12.4375,
      "learning_rate": 2.4109846717556206e-05,
      "loss": 1.133,
      "num_input_tokens_seen": 43637072,
      "step": 75225
    },
    {
      "epoch": 11.20494489127197,
      "grad_norm": 6.6875,
      "learning_rate": 2.4106599364085296e-05,
      "loss": 1.335,
      "num_input_tokens_seen": 43639920,
      "step": 75230
    },
    {
      "epoch": 11.205689603812928,
      "grad_norm": 12.6875,
      "learning_rate": 2.4103352025707473e-05,
      "loss": 1.1729,
      "num_input_tokens_seen": 43642736,
      "step": 75235
    },
    {
      "epoch": 11.206434316353887,
      "grad_norm": 4.78125,
      "learning_rate": 2.4100104702477614e-05,
      "loss": 1.1208,
      "num_input_tokens_seen": 43645360,
      "step": 75240
    },
    {
      "epoch": 11.207179028894847,
      "grad_norm": 7.5,
      "learning_rate": 2.4096857394450557e-05,
      "loss": 0.9811,
      "num_input_tokens_seen": 43648368,
      "step": 75245
    },
    {
      "epoch": 11.207923741435806,
      "grad_norm": 12.0,
      "learning_rate": 2.4093610101681192e-05,
      "loss": 1.1695,
      "num_input_tokens_seen": 43651472,
      "step": 75250
    },
    {
      "epoch": 11.208668453976765,
      "grad_norm": 13.1875,
      "learning_rate": 2.4090362824224347e-05,
      "loss": 1.2568,
      "num_input_tokens_seen": 43654256,
      "step": 75255
    },
    {
      "epoch": 11.209413166517724,
      "grad_norm": 8.875,
      "learning_rate": 2.4087115562134913e-05,
      "loss": 0.9929,
      "num_input_tokens_seen": 43657616,
      "step": 75260
    },
    {
      "epoch": 11.210157879058684,
      "grad_norm": 10.8125,
      "learning_rate": 2.4083868315467725e-05,
      "loss": 1.4458,
      "num_input_tokens_seen": 43660880,
      "step": 75265
    },
    {
      "epoch": 11.210902591599643,
      "grad_norm": 9.5,
      "learning_rate": 2.4080621084277656e-05,
      "loss": 1.1734,
      "num_input_tokens_seen": 43663824,
      "step": 75270
    },
    {
      "epoch": 11.211647304140602,
      "grad_norm": 3.5,
      "learning_rate": 2.4077373868619562e-05,
      "loss": 1.1287,
      "num_input_tokens_seen": 43666608,
      "step": 75275
    },
    {
      "epoch": 11.21239201668156,
      "grad_norm": 9.25,
      "learning_rate": 2.4074126668548287e-05,
      "loss": 1.1032,
      "num_input_tokens_seen": 43669232,
      "step": 75280
    },
    {
      "epoch": 11.21313672922252,
      "grad_norm": 17.125,
      "learning_rate": 2.4070879484118712e-05,
      "loss": 1.0791,
      "num_input_tokens_seen": 43671888,
      "step": 75285
    },
    {
      "epoch": 11.21388144176348,
      "grad_norm": 10.6875,
      "learning_rate": 2.4067632315385675e-05,
      "loss": 1.2262,
      "num_input_tokens_seen": 43674768,
      "step": 75290
    },
    {
      "epoch": 11.214626154304439,
      "grad_norm": 6.0,
      "learning_rate": 2.4064385162404048e-05,
      "loss": 1.0733,
      "num_input_tokens_seen": 43677552,
      "step": 75295
    },
    {
      "epoch": 11.215370866845397,
      "grad_norm": 5.90625,
      "learning_rate": 2.406113802522868e-05,
      "loss": 1.0262,
      "num_input_tokens_seen": 43680464,
      "step": 75300
    },
    {
      "epoch": 11.216115579386356,
      "grad_norm": 10.5,
      "learning_rate": 2.4057890903914437e-05,
      "loss": 1.1636,
      "num_input_tokens_seen": 43683376,
      "step": 75305
    },
    {
      "epoch": 11.216860291927317,
      "grad_norm": 9.5625,
      "learning_rate": 2.405464379851617e-05,
      "loss": 1.3344,
      "num_input_tokens_seen": 43686288,
      "step": 75310
    },
    {
      "epoch": 11.217605004468275,
      "grad_norm": 12.75,
      "learning_rate": 2.4051396709088726e-05,
      "loss": 1.1378,
      "num_input_tokens_seen": 43689040,
      "step": 75315
    },
    {
      "epoch": 11.218349717009234,
      "grad_norm": 11.9375,
      "learning_rate": 2.4048149635686977e-05,
      "loss": 0.9164,
      "num_input_tokens_seen": 43691856,
      "step": 75320
    },
    {
      "epoch": 11.219094429550193,
      "grad_norm": 9.125,
      "learning_rate": 2.4044902578365765e-05,
      "loss": 1.2616,
      "num_input_tokens_seen": 43694672,
      "step": 75325
    },
    {
      "epoch": 11.219839142091153,
      "grad_norm": 12.5,
      "learning_rate": 2.404165553717996e-05,
      "loss": 1.1792,
      "num_input_tokens_seen": 43697776,
      "step": 75330
    },
    {
      "epoch": 11.220583854632112,
      "grad_norm": 10.9375,
      "learning_rate": 2.4038408512184406e-05,
      "loss": 1.1305,
      "num_input_tokens_seen": 43700656,
      "step": 75335
    },
    {
      "epoch": 11.221328567173071,
      "grad_norm": 12.0625,
      "learning_rate": 2.4035161503433974e-05,
      "loss": 0.8788,
      "num_input_tokens_seen": 43703504,
      "step": 75340
    },
    {
      "epoch": 11.22207327971403,
      "grad_norm": 8.1875,
      "learning_rate": 2.4031914510983492e-05,
      "loss": 0.9574,
      "num_input_tokens_seen": 43706544,
      "step": 75345
    },
    {
      "epoch": 11.22281799225499,
      "grad_norm": 8.4375,
      "learning_rate": 2.402866753488784e-05,
      "loss": 0.9073,
      "num_input_tokens_seen": 43709424,
      "step": 75350
    },
    {
      "epoch": 11.223562704795949,
      "grad_norm": 7.5,
      "learning_rate": 2.4025420575201867e-05,
      "loss": 1.2887,
      "num_input_tokens_seen": 43712336,
      "step": 75355
    },
    {
      "epoch": 11.224307417336908,
      "grad_norm": 13.625,
      "learning_rate": 2.402217363198041e-05,
      "loss": 0.9749,
      "num_input_tokens_seen": 43715152,
      "step": 75360
    },
    {
      "epoch": 11.225052129877866,
      "grad_norm": 12.0,
      "learning_rate": 2.4018926705278347e-05,
      "loss": 1.1549,
      "num_input_tokens_seen": 43717936,
      "step": 75365
    },
    {
      "epoch": 11.225796842418827,
      "grad_norm": 9.8125,
      "learning_rate": 2.4015679795150513e-05,
      "loss": 0.7574,
      "num_input_tokens_seen": 43720528,
      "step": 75370
    },
    {
      "epoch": 11.226541554959786,
      "grad_norm": 15.875,
      "learning_rate": 2.4012432901651778e-05,
      "loss": 1.3854,
      "num_input_tokens_seen": 43723536,
      "step": 75375
    },
    {
      "epoch": 11.227286267500745,
      "grad_norm": 9.4375,
      "learning_rate": 2.400918602483697e-05,
      "loss": 1.3358,
      "num_input_tokens_seen": 43726992,
      "step": 75380
    },
    {
      "epoch": 11.228030980041703,
      "grad_norm": 12.25,
      "learning_rate": 2.400593916476097e-05,
      "loss": 1.119,
      "num_input_tokens_seen": 43729776,
      "step": 75385
    },
    {
      "epoch": 11.228775692582664,
      "grad_norm": 6.3125,
      "learning_rate": 2.4002692321478617e-05,
      "loss": 0.7082,
      "num_input_tokens_seen": 43732464,
      "step": 75390
    },
    {
      "epoch": 11.229520405123623,
      "grad_norm": 12.4375,
      "learning_rate": 2.399944549504476e-05,
      "loss": 0.977,
      "num_input_tokens_seen": 43735120,
      "step": 75395
    },
    {
      "epoch": 11.230265117664581,
      "grad_norm": 7.03125,
      "learning_rate": 2.399619868551425e-05,
      "loss": 1.0186,
      "num_input_tokens_seen": 43737872,
      "step": 75400
    },
    {
      "epoch": 11.23100983020554,
      "grad_norm": 11.0,
      "learning_rate": 2.3992951892941948e-05,
      "loss": 1.3304,
      "num_input_tokens_seen": 43740944,
      "step": 75405
    },
    {
      "epoch": 11.2317545427465,
      "grad_norm": 8.4375,
      "learning_rate": 2.3989705117382705e-05,
      "loss": 1.307,
      "num_input_tokens_seen": 43743920,
      "step": 75410
    },
    {
      "epoch": 11.23249925528746,
      "grad_norm": 9.0,
      "learning_rate": 2.3986458358891353e-05,
      "loss": 1.0036,
      "num_input_tokens_seen": 43746864,
      "step": 75415
    },
    {
      "epoch": 11.233243967828418,
      "grad_norm": 5.71875,
      "learning_rate": 2.3983211617522768e-05,
      "loss": 0.9589,
      "num_input_tokens_seen": 43749808,
      "step": 75420
    },
    {
      "epoch": 11.233988680369377,
      "grad_norm": 21.875,
      "learning_rate": 2.3979964893331787e-05,
      "loss": 1.3741,
      "num_input_tokens_seen": 43752688,
      "step": 75425
    },
    {
      "epoch": 11.234733392910337,
      "grad_norm": 8.4375,
      "learning_rate": 2.397671818637327e-05,
      "loss": 1.0273,
      "num_input_tokens_seen": 43755376,
      "step": 75430
    },
    {
      "epoch": 11.235478105451296,
      "grad_norm": 8.4375,
      "learning_rate": 2.3973471496702052e-05,
      "loss": 1.0514,
      "num_input_tokens_seen": 43758416,
      "step": 75435
    },
    {
      "epoch": 11.236222817992255,
      "grad_norm": 15.375,
      "learning_rate": 2.397022482437298e-05,
      "loss": 0.8521,
      "num_input_tokens_seen": 43761552,
      "step": 75440
    },
    {
      "epoch": 11.236967530533214,
      "grad_norm": 17.5,
      "learning_rate": 2.3966978169440927e-05,
      "loss": 1.1583,
      "num_input_tokens_seen": 43764432,
      "step": 75445
    },
    {
      "epoch": 11.237712243074174,
      "grad_norm": 8.1875,
      "learning_rate": 2.396373153196072e-05,
      "loss": 1.2966,
      "num_input_tokens_seen": 43767440,
      "step": 75450
    },
    {
      "epoch": 11.238456955615133,
      "grad_norm": 8.375,
      "learning_rate": 2.3960484911987223e-05,
      "loss": 0.9179,
      "num_input_tokens_seen": 43770288,
      "step": 75455
    },
    {
      "epoch": 11.239201668156092,
      "grad_norm": 7.75,
      "learning_rate": 2.395723830957527e-05,
      "loss": 1.097,
      "num_input_tokens_seen": 43773296,
      "step": 75460
    },
    {
      "epoch": 11.23994638069705,
      "grad_norm": 17.375,
      "learning_rate": 2.3953991724779723e-05,
      "loss": 1.1717,
      "num_input_tokens_seen": 43776304,
      "step": 75465
    },
    {
      "epoch": 11.24069109323801,
      "grad_norm": 10.125,
      "learning_rate": 2.395074515765542e-05,
      "loss": 1.2119,
      "num_input_tokens_seen": 43779056,
      "step": 75470
    },
    {
      "epoch": 11.24143580577897,
      "grad_norm": 8.125,
      "learning_rate": 2.3947498608257204e-05,
      "loss": 1.0289,
      "num_input_tokens_seen": 43782032,
      "step": 75475
    },
    {
      "epoch": 11.242180518319929,
      "grad_norm": 9.625,
      "learning_rate": 2.394425207663994e-05,
      "loss": 0.9088,
      "num_input_tokens_seen": 43784976,
      "step": 75480
    },
    {
      "epoch": 11.242925230860887,
      "grad_norm": 12.125,
      "learning_rate": 2.3941005562858452e-05,
      "loss": 1.2199,
      "num_input_tokens_seen": 43788080,
      "step": 75485
    },
    {
      "epoch": 11.243669943401846,
      "grad_norm": 6.75,
      "learning_rate": 2.393775906696761e-05,
      "loss": 1.1643,
      "num_input_tokens_seen": 43791184,
      "step": 75490
    },
    {
      "epoch": 11.244414655942807,
      "grad_norm": 7.5,
      "learning_rate": 2.393451258902224e-05,
      "loss": 0.9087,
      "num_input_tokens_seen": 43794096,
      "step": 75495
    },
    {
      "epoch": 11.245159368483765,
      "grad_norm": 5.53125,
      "learning_rate": 2.3931266129077206e-05,
      "loss": 0.8558,
      "num_input_tokens_seen": 43796976,
      "step": 75500
    },
    {
      "epoch": 11.245904081024724,
      "grad_norm": 9.25,
      "learning_rate": 2.3928019687187338e-05,
      "loss": 1.2318,
      "num_input_tokens_seen": 43800752,
      "step": 75505
    },
    {
      "epoch": 11.246648793565683,
      "grad_norm": 7.40625,
      "learning_rate": 2.392477326340749e-05,
      "loss": 1.1858,
      "num_input_tokens_seen": 43803696,
      "step": 75510
    },
    {
      "epoch": 11.247393506106643,
      "grad_norm": 15.875,
      "learning_rate": 2.3921526857792508e-05,
      "loss": 0.9821,
      "num_input_tokens_seen": 43806800,
      "step": 75515
    },
    {
      "epoch": 11.248138218647602,
      "grad_norm": 14.25,
      "learning_rate": 2.3918280470397226e-05,
      "loss": 1.0725,
      "num_input_tokens_seen": 43809680,
      "step": 75520
    },
    {
      "epoch": 11.248882931188561,
      "grad_norm": 10.8125,
      "learning_rate": 2.3915034101276504e-05,
      "loss": 1.0614,
      "num_input_tokens_seen": 43812720,
      "step": 75525
    },
    {
      "epoch": 11.24962764372952,
      "grad_norm": 19.875,
      "learning_rate": 2.3911787750485172e-05,
      "loss": 1.2186,
      "num_input_tokens_seen": 43815632,
      "step": 75530
    },
    {
      "epoch": 11.25037235627048,
      "grad_norm": 11.4375,
      "learning_rate": 2.3908541418078087e-05,
      "loss": 1.0974,
      "num_input_tokens_seen": 43819184,
      "step": 75535
    },
    {
      "epoch": 11.251117068811439,
      "grad_norm": 5.6875,
      "learning_rate": 2.3905295104110076e-05,
      "loss": 0.5671,
      "num_input_tokens_seen": 43821808,
      "step": 75540
    },
    {
      "epoch": 11.251861781352398,
      "grad_norm": 5.4375,
      "learning_rate": 2.3902048808636e-05,
      "loss": 0.9715,
      "num_input_tokens_seen": 43824848,
      "step": 75545
    },
    {
      "epoch": 11.252606493893357,
      "grad_norm": 4.5,
      "learning_rate": 2.3898802531710693e-05,
      "loss": 0.7083,
      "num_input_tokens_seen": 43827792,
      "step": 75550
    },
    {
      "epoch": 11.253351206434317,
      "grad_norm": 7.9375,
      "learning_rate": 2.389555627338899e-05,
      "loss": 0.7425,
      "num_input_tokens_seen": 43830736,
      "step": 75555
    },
    {
      "epoch": 11.254095918975276,
      "grad_norm": 6.84375,
      "learning_rate": 2.3892310033725747e-05,
      "loss": 1.1945,
      "num_input_tokens_seen": 43833904,
      "step": 75560
    },
    {
      "epoch": 11.254840631516235,
      "grad_norm": 14.6875,
      "learning_rate": 2.38890638127758e-05,
      "loss": 1.1019,
      "num_input_tokens_seen": 43836592,
      "step": 75565
    },
    {
      "epoch": 11.255585344057193,
      "grad_norm": 3.84375,
      "learning_rate": 2.3885817610593994e-05,
      "loss": 0.8499,
      "num_input_tokens_seen": 43839728,
      "step": 75570
    },
    {
      "epoch": 11.256330056598154,
      "grad_norm": 9.9375,
      "learning_rate": 2.3882571427235156e-05,
      "loss": 1.0701,
      "num_input_tokens_seen": 43843120,
      "step": 75575
    },
    {
      "epoch": 11.257074769139113,
      "grad_norm": 10.9375,
      "learning_rate": 2.3879325262754152e-05,
      "loss": 1.0009,
      "num_input_tokens_seen": 43846000,
      "step": 75580
    },
    {
      "epoch": 11.257819481680071,
      "grad_norm": 11.5625,
      "learning_rate": 2.3876079117205795e-05,
      "loss": 1.2903,
      "num_input_tokens_seen": 43848912,
      "step": 75585
    },
    {
      "epoch": 11.25856419422103,
      "grad_norm": 7.25,
      "learning_rate": 2.387283299064495e-05,
      "loss": 1.0371,
      "num_input_tokens_seen": 43851856,
      "step": 75590
    },
    {
      "epoch": 11.25930890676199,
      "grad_norm": 15.1875,
      "learning_rate": 2.3869586883126445e-05,
      "loss": 1.0509,
      "num_input_tokens_seen": 43854992,
      "step": 75595
    },
    {
      "epoch": 11.26005361930295,
      "grad_norm": 11.6875,
      "learning_rate": 2.3866340794705117e-05,
      "loss": 1.1914,
      "num_input_tokens_seen": 43857936,
      "step": 75600
    },
    {
      "epoch": 11.260798331843908,
      "grad_norm": 9.0625,
      "learning_rate": 2.3863094725435813e-05,
      "loss": 0.8463,
      "num_input_tokens_seen": 43860688,
      "step": 75605
    },
    {
      "epoch": 11.261543044384867,
      "grad_norm": 10.625,
      "learning_rate": 2.3859848675373358e-05,
      "loss": 1.3245,
      "num_input_tokens_seen": 43863696,
      "step": 75610
    },
    {
      "epoch": 11.262287756925826,
      "grad_norm": 8.5,
      "learning_rate": 2.385660264457261e-05,
      "loss": 0.7108,
      "num_input_tokens_seen": 43866768,
      "step": 75615
    },
    {
      "epoch": 11.263032469466786,
      "grad_norm": 7.65625,
      "learning_rate": 2.385335663308839e-05,
      "loss": 0.9962,
      "num_input_tokens_seen": 43869648,
      "step": 75620
    },
    {
      "epoch": 11.263777182007745,
      "grad_norm": 8.3125,
      "learning_rate": 2.3850110640975555e-05,
      "loss": 1.1699,
      "num_input_tokens_seen": 43872528,
      "step": 75625
    },
    {
      "epoch": 11.264521894548704,
      "grad_norm": 17.125,
      "learning_rate": 2.3846864668288933e-05,
      "loss": 1.0753,
      "num_input_tokens_seen": 43875632,
      "step": 75630
    },
    {
      "epoch": 11.265266607089663,
      "grad_norm": 10.0625,
      "learning_rate": 2.3843618715083353e-05,
      "loss": 1.1934,
      "num_input_tokens_seen": 43878608,
      "step": 75635
    },
    {
      "epoch": 11.266011319630623,
      "grad_norm": 11.125,
      "learning_rate": 2.3840372781413667e-05,
      "loss": 1.1788,
      "num_input_tokens_seen": 43881840,
      "step": 75640
    },
    {
      "epoch": 11.266756032171582,
      "grad_norm": 8.1875,
      "learning_rate": 2.3837126867334687e-05,
      "loss": 0.9839,
      "num_input_tokens_seen": 43884944,
      "step": 75645
    },
    {
      "epoch": 11.26750074471254,
      "grad_norm": 18.0,
      "learning_rate": 2.3833880972901285e-05,
      "loss": 1.2836,
      "num_input_tokens_seen": 43888208,
      "step": 75650
    },
    {
      "epoch": 11.2682454572535,
      "grad_norm": 8.125,
      "learning_rate": 2.3830635098168267e-05,
      "loss": 1.119,
      "num_input_tokens_seen": 43891152,
      "step": 75655
    },
    {
      "epoch": 11.26899016979446,
      "grad_norm": 10.9375,
      "learning_rate": 2.3827389243190486e-05,
      "loss": 1.0564,
      "num_input_tokens_seen": 43893840,
      "step": 75660
    },
    {
      "epoch": 11.269734882335419,
      "grad_norm": 10.25,
      "learning_rate": 2.3824143408022773e-05,
      "loss": 1.3619,
      "num_input_tokens_seen": 43896656,
      "step": 75665
    },
    {
      "epoch": 11.270479594876377,
      "grad_norm": 3.90625,
      "learning_rate": 2.3820897592719964e-05,
      "loss": 1.0507,
      "num_input_tokens_seen": 43899632,
      "step": 75670
    },
    {
      "epoch": 11.271224307417336,
      "grad_norm": 10.3125,
      "learning_rate": 2.3817651797336894e-05,
      "loss": 1.1435,
      "num_input_tokens_seen": 43902256,
      "step": 75675
    },
    {
      "epoch": 11.271969019958297,
      "grad_norm": 8.875,
      "learning_rate": 2.3814406021928382e-05,
      "loss": 1.3088,
      "num_input_tokens_seen": 43905456,
      "step": 75680
    },
    {
      "epoch": 11.272713732499255,
      "grad_norm": 10.0,
      "learning_rate": 2.381116026654929e-05,
      "loss": 1.2537,
      "num_input_tokens_seen": 43908496,
      "step": 75685
    },
    {
      "epoch": 11.273458445040214,
      "grad_norm": 7.3125,
      "learning_rate": 2.3807914531254417e-05,
      "loss": 0.8506,
      "num_input_tokens_seen": 43911408,
      "step": 75690
    },
    {
      "epoch": 11.274203157581173,
      "grad_norm": 12.1875,
      "learning_rate": 2.3804668816098635e-05,
      "loss": 1.1585,
      "num_input_tokens_seen": 43914416,
      "step": 75695
    },
    {
      "epoch": 11.274947870122134,
      "grad_norm": 12.5,
      "learning_rate": 2.3801423121136752e-05,
      "loss": 0.9882,
      "num_input_tokens_seen": 43917360,
      "step": 75700
    },
    {
      "epoch": 11.275692582663092,
      "grad_norm": 11.0,
      "learning_rate": 2.379817744642361e-05,
      "loss": 1.2835,
      "num_input_tokens_seen": 43920560,
      "step": 75705
    },
    {
      "epoch": 11.276437295204051,
      "grad_norm": 13.3125,
      "learning_rate": 2.379493179201403e-05,
      "loss": 0.9773,
      "num_input_tokens_seen": 43923504,
      "step": 75710
    },
    {
      "epoch": 11.27718200774501,
      "grad_norm": 8.1875,
      "learning_rate": 2.3791686157962866e-05,
      "loss": 1.1902,
      "num_input_tokens_seen": 43926288,
      "step": 75715
    },
    {
      "epoch": 11.27792672028597,
      "grad_norm": 6.40625,
      "learning_rate": 2.378844054432493e-05,
      "loss": 0.8979,
      "num_input_tokens_seen": 43929264,
      "step": 75720
    },
    {
      "epoch": 11.278671432826929,
      "grad_norm": 15.0,
      "learning_rate": 2.378519495115506e-05,
      "loss": 0.926,
      "num_input_tokens_seen": 43931888,
      "step": 75725
    },
    {
      "epoch": 11.279416145367888,
      "grad_norm": 13.8125,
      "learning_rate": 2.3781949378508085e-05,
      "loss": 1.1416,
      "num_input_tokens_seen": 43934928,
      "step": 75730
    },
    {
      "epoch": 11.280160857908847,
      "grad_norm": 11.25,
      "learning_rate": 2.3778703826438833e-05,
      "loss": 1.3847,
      "num_input_tokens_seen": 43937616,
      "step": 75735
    },
    {
      "epoch": 11.280905570449807,
      "grad_norm": 6.90625,
      "learning_rate": 2.377545829500215e-05,
      "loss": 1.1172,
      "num_input_tokens_seen": 43940368,
      "step": 75740
    },
    {
      "epoch": 11.281650282990766,
      "grad_norm": 12.0625,
      "learning_rate": 2.377221278425284e-05,
      "loss": 0.8327,
      "num_input_tokens_seen": 43943024,
      "step": 75745
    },
    {
      "epoch": 11.282394995531725,
      "grad_norm": 12.625,
      "learning_rate": 2.376896729424576e-05,
      "loss": 1.6197,
      "num_input_tokens_seen": 43946128,
      "step": 75750
    },
    {
      "epoch": 11.283139708072683,
      "grad_norm": 12.25,
      "learning_rate": 2.3765721825035724e-05,
      "loss": 1.0168,
      "num_input_tokens_seen": 43948944,
      "step": 75755
    },
    {
      "epoch": 11.283884420613644,
      "grad_norm": 8.5,
      "learning_rate": 2.376247637667756e-05,
      "loss": 1.1008,
      "num_input_tokens_seen": 43951632,
      "step": 75760
    },
    {
      "epoch": 11.284629133154603,
      "grad_norm": 8.75,
      "learning_rate": 2.3759230949226103e-05,
      "loss": 0.9623,
      "num_input_tokens_seen": 43954640,
      "step": 75765
    },
    {
      "epoch": 11.285373845695561,
      "grad_norm": 8.25,
      "learning_rate": 2.375598554273617e-05,
      "loss": 1.1318,
      "num_input_tokens_seen": 43957424,
      "step": 75770
    },
    {
      "epoch": 11.28611855823652,
      "grad_norm": 6.5625,
      "learning_rate": 2.3752740157262607e-05,
      "loss": 1.0212,
      "num_input_tokens_seen": 43960336,
      "step": 75775
    },
    {
      "epoch": 11.28686327077748,
      "grad_norm": 11.75,
      "learning_rate": 2.374949479286022e-05,
      "loss": 1.3571,
      "num_input_tokens_seen": 43963376,
      "step": 75780
    },
    {
      "epoch": 11.28760798331844,
      "grad_norm": 11.9375,
      "learning_rate": 2.374624944958386e-05,
      "loss": 0.9818,
      "num_input_tokens_seen": 43966096,
      "step": 75785
    },
    {
      "epoch": 11.288352695859398,
      "grad_norm": 4.90625,
      "learning_rate": 2.3743004127488332e-05,
      "loss": 0.8652,
      "num_input_tokens_seen": 43969136,
      "step": 75790
    },
    {
      "epoch": 11.289097408400357,
      "grad_norm": 8.625,
      "learning_rate": 2.373975882662848e-05,
      "loss": 1.0292,
      "num_input_tokens_seen": 43972176,
      "step": 75795
    },
    {
      "epoch": 11.289842120941316,
      "grad_norm": 9.9375,
      "learning_rate": 2.3736513547059124e-05,
      "loss": 1.4179,
      "num_input_tokens_seen": 43975056,
      "step": 75800
    },
    {
      "epoch": 11.290586833482276,
      "grad_norm": 9.125,
      "learning_rate": 2.373326828883507e-05,
      "loss": 0.9775,
      "num_input_tokens_seen": 43977872,
      "step": 75805
    },
    {
      "epoch": 11.291331546023235,
      "grad_norm": 11.4375,
      "learning_rate": 2.3730023052011178e-05,
      "loss": 1.1879,
      "num_input_tokens_seen": 43980656,
      "step": 75810
    },
    {
      "epoch": 11.292076258564194,
      "grad_norm": 11.875,
      "learning_rate": 2.3726777836642243e-05,
      "loss": 1.1462,
      "num_input_tokens_seen": 43983920,
      "step": 75815
    },
    {
      "epoch": 11.292820971105153,
      "grad_norm": 19.5,
      "learning_rate": 2.3723532642783114e-05,
      "loss": 1.2925,
      "num_input_tokens_seen": 43986864,
      "step": 75820
    },
    {
      "epoch": 11.293565683646113,
      "grad_norm": 12.1875,
      "learning_rate": 2.3720287470488596e-05,
      "loss": 0.8823,
      "num_input_tokens_seen": 43989712,
      "step": 75825
    },
    {
      "epoch": 11.294310396187072,
      "grad_norm": 10.75,
      "learning_rate": 2.3717042319813527e-05,
      "loss": 1.284,
      "num_input_tokens_seen": 43993136,
      "step": 75830
    },
    {
      "epoch": 11.29505510872803,
      "grad_norm": 19.0,
      "learning_rate": 2.3713797190812726e-05,
      "loss": 1.1671,
      "num_input_tokens_seen": 43995920,
      "step": 75835
    },
    {
      "epoch": 11.29579982126899,
      "grad_norm": 8.875,
      "learning_rate": 2.3710552083540998e-05,
      "loss": 1.0453,
      "num_input_tokens_seen": 43998608,
      "step": 75840
    },
    {
      "epoch": 11.29654453380995,
      "grad_norm": 14.4375,
      "learning_rate": 2.3707306998053198e-05,
      "loss": 1.4262,
      "num_input_tokens_seen": 44001456,
      "step": 75845
    },
    {
      "epoch": 11.297289246350909,
      "grad_norm": 5.375,
      "learning_rate": 2.370406193440412e-05,
      "loss": 0.8626,
      "num_input_tokens_seen": 44004432,
      "step": 75850
    },
    {
      "epoch": 11.298033958891867,
      "grad_norm": 13.9375,
      "learning_rate": 2.3700816892648608e-05,
      "loss": 1.0182,
      "num_input_tokens_seen": 44007280,
      "step": 75855
    },
    {
      "epoch": 11.298778671432826,
      "grad_norm": 4.625,
      "learning_rate": 2.369757187284147e-05,
      "loss": 1.2466,
      "num_input_tokens_seen": 44010224,
      "step": 75860
    },
    {
      "epoch": 11.299523383973787,
      "grad_norm": 4.71875,
      "learning_rate": 2.3694326875037536e-05,
      "loss": 1.0834,
      "num_input_tokens_seen": 44013360,
      "step": 75865
    },
    {
      "epoch": 11.300268096514746,
      "grad_norm": 6.65625,
      "learning_rate": 2.3691081899291613e-05,
      "loss": 0.9774,
      "num_input_tokens_seen": 44016176,
      "step": 75870
    },
    {
      "epoch": 11.301012809055704,
      "grad_norm": 8.9375,
      "learning_rate": 2.3687836945658542e-05,
      "loss": 1.0896,
      "num_input_tokens_seen": 44019088,
      "step": 75875
    },
    {
      "epoch": 11.301757521596663,
      "grad_norm": 10.0625,
      "learning_rate": 2.3684592014193133e-05,
      "loss": 1.23,
      "num_input_tokens_seen": 44021936,
      "step": 75880
    },
    {
      "epoch": 11.302502234137624,
      "grad_norm": 13.375,
      "learning_rate": 2.3681347104950193e-05,
      "loss": 0.9508,
      "num_input_tokens_seen": 44024784,
      "step": 75885
    },
    {
      "epoch": 11.303246946678582,
      "grad_norm": 10.9375,
      "learning_rate": 2.3678102217984564e-05,
      "loss": 1.2541,
      "num_input_tokens_seen": 44027632,
      "step": 75890
    },
    {
      "epoch": 11.303991659219541,
      "grad_norm": 7.5625,
      "learning_rate": 2.3674857353351048e-05,
      "loss": 1.3097,
      "num_input_tokens_seen": 44030416,
      "step": 75895
    },
    {
      "epoch": 11.3047363717605,
      "grad_norm": 11.1875,
      "learning_rate": 2.3671612511104476e-05,
      "loss": 1.1049,
      "num_input_tokens_seen": 44033424,
      "step": 75900
    },
    {
      "epoch": 11.30548108430146,
      "grad_norm": 9.1875,
      "learning_rate": 2.366836769129965e-05,
      "loss": 1.288,
      "num_input_tokens_seen": 44036112,
      "step": 75905
    },
    {
      "epoch": 11.30622579684242,
      "grad_norm": 8.6875,
      "learning_rate": 2.366512289399141e-05,
      "loss": 1.0893,
      "num_input_tokens_seen": 44038992,
      "step": 75910
    },
    {
      "epoch": 11.306970509383378,
      "grad_norm": 13.25,
      "learning_rate": 2.3661878119234562e-05,
      "loss": 1.0418,
      "num_input_tokens_seen": 44041552,
      "step": 75915
    },
    {
      "epoch": 11.307715221924337,
      "grad_norm": 10.25,
      "learning_rate": 2.3658633367083914e-05,
      "loss": 1.1257,
      "num_input_tokens_seen": 44044272,
      "step": 75920
    },
    {
      "epoch": 11.308459934465297,
      "grad_norm": 10.5625,
      "learning_rate": 2.3655388637594298e-05,
      "loss": 1.2382,
      "num_input_tokens_seen": 44046960,
      "step": 75925
    },
    {
      "epoch": 11.309204647006256,
      "grad_norm": 13.5,
      "learning_rate": 2.3652143930820523e-05,
      "loss": 1.3977,
      "num_input_tokens_seen": 44049872,
      "step": 75930
    },
    {
      "epoch": 11.309949359547215,
      "grad_norm": 11.0625,
      "learning_rate": 2.364889924681741e-05,
      "loss": 0.9585,
      "num_input_tokens_seen": 44052432,
      "step": 75935
    },
    {
      "epoch": 11.310694072088173,
      "grad_norm": 10.3125,
      "learning_rate": 2.364565458563976e-05,
      "loss": 1.0673,
      "num_input_tokens_seen": 44055760,
      "step": 75940
    },
    {
      "epoch": 11.311438784629134,
      "grad_norm": 15.625,
      "learning_rate": 2.364240994734241e-05,
      "loss": 0.7592,
      "num_input_tokens_seen": 44058800,
      "step": 75945
    },
    {
      "epoch": 11.312183497170093,
      "grad_norm": 17.375,
      "learning_rate": 2.3639165331980157e-05,
      "loss": 1.0088,
      "num_input_tokens_seen": 44061616,
      "step": 75950
    },
    {
      "epoch": 11.312928209711052,
      "grad_norm": 11.4375,
      "learning_rate": 2.3635920739607828e-05,
      "loss": 1.0895,
      "num_input_tokens_seen": 44064720,
      "step": 75955
    },
    {
      "epoch": 11.31367292225201,
      "grad_norm": 12.0,
      "learning_rate": 2.3632676170280235e-05,
      "loss": 1.0999,
      "num_input_tokens_seen": 44067856,
      "step": 75960
    },
    {
      "epoch": 11.31441763479297,
      "grad_norm": 13.4375,
      "learning_rate": 2.362943162405218e-05,
      "loss": 0.9804,
      "num_input_tokens_seen": 44070736,
      "step": 75965
    },
    {
      "epoch": 11.31516234733393,
      "grad_norm": 10.75,
      "learning_rate": 2.3626187100978496e-05,
      "loss": 1.4138,
      "num_input_tokens_seen": 44073648,
      "step": 75970
    },
    {
      "epoch": 11.315907059874888,
      "grad_norm": 15.875,
      "learning_rate": 2.362294260111397e-05,
      "loss": 1.2657,
      "num_input_tokens_seen": 44076112,
      "step": 75975
    },
    {
      "epoch": 11.316651772415847,
      "grad_norm": 11.0625,
      "learning_rate": 2.3619698124513438e-05,
      "loss": 0.6294,
      "num_input_tokens_seen": 44078992,
      "step": 75980
    },
    {
      "epoch": 11.317396484956806,
      "grad_norm": 10.25,
      "learning_rate": 2.3616453671231694e-05,
      "loss": 1.1005,
      "num_input_tokens_seen": 44081936,
      "step": 75985
    },
    {
      "epoch": 11.318141197497766,
      "grad_norm": 8.8125,
      "learning_rate": 2.3613209241323574e-05,
      "loss": 1.1935,
      "num_input_tokens_seen": 44085136,
      "step": 75990
    },
    {
      "epoch": 11.318885910038725,
      "grad_norm": 7.4375,
      "learning_rate": 2.360996483484387e-05,
      "loss": 1.0023,
      "num_input_tokens_seen": 44087888,
      "step": 75995
    },
    {
      "epoch": 11.319630622579684,
      "grad_norm": 10.6875,
      "learning_rate": 2.360672045184739e-05,
      "loss": 1.0066,
      "num_input_tokens_seen": 44090736,
      "step": 76000
    },
    {
      "epoch": 11.320375335120643,
      "grad_norm": 14.5625,
      "learning_rate": 2.3603476092388963e-05,
      "loss": 0.9026,
      "num_input_tokens_seen": 44093776,
      "step": 76005
    },
    {
      "epoch": 11.321120047661603,
      "grad_norm": 6.4375,
      "learning_rate": 2.3600231756523373e-05,
      "loss": 0.7837,
      "num_input_tokens_seen": 44096720,
      "step": 76010
    },
    {
      "epoch": 11.321864760202562,
      "grad_norm": 22.875,
      "learning_rate": 2.3596987444305456e-05,
      "loss": 1.2066,
      "num_input_tokens_seen": 44099536,
      "step": 76015
    },
    {
      "epoch": 11.32260947274352,
      "grad_norm": 15.625,
      "learning_rate": 2.359374315579e-05,
      "loss": 1.1524,
      "num_input_tokens_seen": 44102320,
      "step": 76020
    },
    {
      "epoch": 11.32335418528448,
      "grad_norm": 9.1875,
      "learning_rate": 2.3590498891031838e-05,
      "loss": 1.2588,
      "num_input_tokens_seen": 44105136,
      "step": 76025
    },
    {
      "epoch": 11.32409889782544,
      "grad_norm": 10.0,
      "learning_rate": 2.3587254650085757e-05,
      "loss": 1.2429,
      "num_input_tokens_seen": 44108080,
      "step": 76030
    },
    {
      "epoch": 11.324843610366399,
      "grad_norm": 15.0625,
      "learning_rate": 2.3584010433006577e-05,
      "loss": 0.932,
      "num_input_tokens_seen": 44111120,
      "step": 76035
    },
    {
      "epoch": 11.325588322907358,
      "grad_norm": 16.5,
      "learning_rate": 2.3580766239849102e-05,
      "loss": 1.3281,
      "num_input_tokens_seen": 44113840,
      "step": 76040
    },
    {
      "epoch": 11.326333035448316,
      "grad_norm": 10.0625,
      "learning_rate": 2.3577522070668128e-05,
      "loss": 1.2291,
      "num_input_tokens_seen": 44116752,
      "step": 76045
    },
    {
      "epoch": 11.327077747989277,
      "grad_norm": 13.0625,
      "learning_rate": 2.3574277925518488e-05,
      "loss": 1.0692,
      "num_input_tokens_seen": 44119568,
      "step": 76050
    },
    {
      "epoch": 11.327822460530236,
      "grad_norm": 9.4375,
      "learning_rate": 2.357103380445496e-05,
      "loss": 0.9021,
      "num_input_tokens_seen": 44122576,
      "step": 76055
    },
    {
      "epoch": 11.328567173071194,
      "grad_norm": 9.0625,
      "learning_rate": 2.356778970753237e-05,
      "loss": 1.2456,
      "num_input_tokens_seen": 44125424,
      "step": 76060
    },
    {
      "epoch": 11.329311885612153,
      "grad_norm": 13.0,
      "learning_rate": 2.3564545634805516e-05,
      "loss": 1.0215,
      "num_input_tokens_seen": 44128528,
      "step": 76065
    },
    {
      "epoch": 11.330056598153114,
      "grad_norm": 11.875,
      "learning_rate": 2.356130158632921e-05,
      "loss": 1.2776,
      "num_input_tokens_seen": 44131472,
      "step": 76070
    },
    {
      "epoch": 11.330801310694072,
      "grad_norm": 16.0,
      "learning_rate": 2.3558057562158247e-05,
      "loss": 1.5026,
      "num_input_tokens_seen": 44134512,
      "step": 76075
    },
    {
      "epoch": 11.331546023235031,
      "grad_norm": 10.5,
      "learning_rate": 2.355481356234743e-05,
      "loss": 0.9243,
      "num_input_tokens_seen": 44137264,
      "step": 76080
    },
    {
      "epoch": 11.33229073577599,
      "grad_norm": 11.625,
      "learning_rate": 2.355156958695158e-05,
      "loss": 1.158,
      "num_input_tokens_seen": 44140496,
      "step": 76085
    },
    {
      "epoch": 11.33303544831695,
      "grad_norm": 7.0625,
      "learning_rate": 2.354832563602548e-05,
      "loss": 0.8858,
      "num_input_tokens_seen": 44143184,
      "step": 76090
    },
    {
      "epoch": 11.33378016085791,
      "grad_norm": 21.875,
      "learning_rate": 2.3545081709623953e-05,
      "loss": 1.0425,
      "num_input_tokens_seen": 44146000,
      "step": 76095
    },
    {
      "epoch": 11.334524873398868,
      "grad_norm": 14.0,
      "learning_rate": 2.354183780780178e-05,
      "loss": 1.1504,
      "num_input_tokens_seen": 44148752,
      "step": 76100
    },
    {
      "epoch": 11.335269585939827,
      "grad_norm": 8.9375,
      "learning_rate": 2.3538593930613784e-05,
      "loss": 0.9357,
      "num_input_tokens_seen": 44151920,
      "step": 76105
    },
    {
      "epoch": 11.336014298480787,
      "grad_norm": 13.0625,
      "learning_rate": 2.353535007811475e-05,
      "loss": 1.0064,
      "num_input_tokens_seen": 44154832,
      "step": 76110
    },
    {
      "epoch": 11.336759011021746,
      "grad_norm": 16.375,
      "learning_rate": 2.3532106250359498e-05,
      "loss": 1.2156,
      "num_input_tokens_seen": 44157808,
      "step": 76115
    },
    {
      "epoch": 11.337503723562705,
      "grad_norm": 31.0,
      "learning_rate": 2.3528862447402817e-05,
      "loss": 1.0183,
      "num_input_tokens_seen": 44160592,
      "step": 76120
    },
    {
      "epoch": 11.338248436103664,
      "grad_norm": 12.9375,
      "learning_rate": 2.3525618669299505e-05,
      "loss": 1.3081,
      "num_input_tokens_seen": 44163728,
      "step": 76125
    },
    {
      "epoch": 11.338993148644622,
      "grad_norm": 12.75,
      "learning_rate": 2.3522374916104377e-05,
      "loss": 0.8619,
      "num_input_tokens_seen": 44166224,
      "step": 76130
    },
    {
      "epoch": 11.339737861185583,
      "grad_norm": 5.46875,
      "learning_rate": 2.3519131187872207e-05,
      "loss": 1.0696,
      "num_input_tokens_seen": 44169136,
      "step": 76135
    },
    {
      "epoch": 11.340482573726542,
      "grad_norm": 6.25,
      "learning_rate": 2.3515887484657823e-05,
      "loss": 1.2197,
      "num_input_tokens_seen": 44171920,
      "step": 76140
    },
    {
      "epoch": 11.3412272862675,
      "grad_norm": 10.375,
      "learning_rate": 2.3512643806516e-05,
      "loss": 0.834,
      "num_input_tokens_seen": 44174800,
      "step": 76145
    },
    {
      "epoch": 11.341971998808459,
      "grad_norm": 13.8125,
      "learning_rate": 2.3509400153501556e-05,
      "loss": 0.9531,
      "num_input_tokens_seen": 44177616,
      "step": 76150
    },
    {
      "epoch": 11.34271671134942,
      "grad_norm": 13.8125,
      "learning_rate": 2.3506156525669286e-05,
      "loss": 1.2075,
      "num_input_tokens_seen": 44180432,
      "step": 76155
    },
    {
      "epoch": 11.343461423890378,
      "grad_norm": 10.25,
      "learning_rate": 2.3502912923073976e-05,
      "loss": 0.8381,
      "num_input_tokens_seen": 44183280,
      "step": 76160
    },
    {
      "epoch": 11.344206136431337,
      "grad_norm": 12.8125,
      "learning_rate": 2.349966934577044e-05,
      "loss": 1.1128,
      "num_input_tokens_seen": 44186352,
      "step": 76165
    },
    {
      "epoch": 11.344950848972296,
      "grad_norm": 8.5625,
      "learning_rate": 2.349642579381345e-05,
      "loss": 1.3309,
      "num_input_tokens_seen": 44189200,
      "step": 76170
    },
    {
      "epoch": 11.345695561513256,
      "grad_norm": 12.1875,
      "learning_rate": 2.349318226725783e-05,
      "loss": 1.255,
      "num_input_tokens_seen": 44192144,
      "step": 76175
    },
    {
      "epoch": 11.346440274054215,
      "grad_norm": 7.53125,
      "learning_rate": 2.3489938766158354e-05,
      "loss": 1.1568,
      "num_input_tokens_seen": 44195184,
      "step": 76180
    },
    {
      "epoch": 11.347184986595174,
      "grad_norm": 10.4375,
      "learning_rate": 2.3486695290569838e-05,
      "loss": 0.943,
      "num_input_tokens_seen": 44198064,
      "step": 76185
    },
    {
      "epoch": 11.347929699136133,
      "grad_norm": 7.53125,
      "learning_rate": 2.348345184054706e-05,
      "loss": 1.1769,
      "num_input_tokens_seen": 44200944,
      "step": 76190
    },
    {
      "epoch": 11.348674411677093,
      "grad_norm": 9.75,
      "learning_rate": 2.3480208416144832e-05,
      "loss": 1.4313,
      "num_input_tokens_seen": 44203664,
      "step": 76195
    },
    {
      "epoch": 11.349419124218052,
      "grad_norm": 10.125,
      "learning_rate": 2.3476965017417935e-05,
      "loss": 0.9408,
      "num_input_tokens_seen": 44206416,
      "step": 76200
    },
    {
      "epoch": 11.35016383675901,
      "grad_norm": 7.78125,
      "learning_rate": 2.3473721644421155e-05,
      "loss": 1.0216,
      "num_input_tokens_seen": 44209552,
      "step": 76205
    },
    {
      "epoch": 11.35090854929997,
      "grad_norm": 18.0,
      "learning_rate": 2.3470478297209307e-05,
      "loss": 1.0001,
      "num_input_tokens_seen": 44212432,
      "step": 76210
    },
    {
      "epoch": 11.35165326184093,
      "grad_norm": 7.78125,
      "learning_rate": 2.3467234975837162e-05,
      "loss": 1.0713,
      "num_input_tokens_seen": 44215408,
      "step": 76215
    },
    {
      "epoch": 11.352397974381889,
      "grad_norm": 12.375,
      "learning_rate": 2.3463991680359536e-05,
      "loss": 1.0687,
      "num_input_tokens_seen": 44218320,
      "step": 76220
    },
    {
      "epoch": 11.353142686922848,
      "grad_norm": 16.875,
      "learning_rate": 2.346074841083121e-05,
      "loss": 1.2913,
      "num_input_tokens_seen": 44221072,
      "step": 76225
    },
    {
      "epoch": 11.353887399463806,
      "grad_norm": 8.125,
      "learning_rate": 2.345750516730697e-05,
      "loss": 1.1866,
      "num_input_tokens_seen": 44224144,
      "step": 76230
    },
    {
      "epoch": 11.354632112004767,
      "grad_norm": 11.75,
      "learning_rate": 2.3454261949841622e-05,
      "loss": 1.1465,
      "num_input_tokens_seen": 44226768,
      "step": 76235
    },
    {
      "epoch": 11.355376824545726,
      "grad_norm": 17.125,
      "learning_rate": 2.3451018758489932e-05,
      "loss": 1.436,
      "num_input_tokens_seen": 44229552,
      "step": 76240
    },
    {
      "epoch": 11.356121537086684,
      "grad_norm": 8.0,
      "learning_rate": 2.3447775593306716e-05,
      "loss": 0.976,
      "num_input_tokens_seen": 44232752,
      "step": 76245
    },
    {
      "epoch": 11.356866249627643,
      "grad_norm": 8.5,
      "learning_rate": 2.3444532454346745e-05,
      "loss": 1.1939,
      "num_input_tokens_seen": 44235760,
      "step": 76250
    },
    {
      "epoch": 11.357610962168604,
      "grad_norm": 5.5625,
      "learning_rate": 2.3441289341664822e-05,
      "loss": 0.7661,
      "num_input_tokens_seen": 44238928,
      "step": 76255
    },
    {
      "epoch": 11.358355674709562,
      "grad_norm": 8.9375,
      "learning_rate": 2.3438046255315735e-05,
      "loss": 0.9403,
      "num_input_tokens_seen": 44242032,
      "step": 76260
    },
    {
      "epoch": 11.359100387250521,
      "grad_norm": 12.1875,
      "learning_rate": 2.3434803195354268e-05,
      "loss": 1.5099,
      "num_input_tokens_seen": 44244720,
      "step": 76265
    },
    {
      "epoch": 11.35984509979148,
      "grad_norm": 10.5625,
      "learning_rate": 2.3431560161835204e-05,
      "loss": 1.5687,
      "num_input_tokens_seen": 44247504,
      "step": 76270
    },
    {
      "epoch": 11.36058981233244,
      "grad_norm": 13.25,
      "learning_rate": 2.3428317154813344e-05,
      "loss": 0.994,
      "num_input_tokens_seen": 44250544,
      "step": 76275
    },
    {
      "epoch": 11.3613345248734,
      "grad_norm": 6.4375,
      "learning_rate": 2.342507417434347e-05,
      "loss": 1.2195,
      "num_input_tokens_seen": 44253584,
      "step": 76280
    },
    {
      "epoch": 11.362079237414358,
      "grad_norm": 5.5,
      "learning_rate": 2.3421831220480357e-05,
      "loss": 0.9502,
      "num_input_tokens_seen": 44256528,
      "step": 76285
    },
    {
      "epoch": 11.362823949955317,
      "grad_norm": 9.875,
      "learning_rate": 2.341858829327881e-05,
      "loss": 1.1693,
      "num_input_tokens_seen": 44259408,
      "step": 76290
    },
    {
      "epoch": 11.363568662496277,
      "grad_norm": 9.375,
      "learning_rate": 2.34153453927936e-05,
      "loss": 1.1278,
      "num_input_tokens_seen": 44262256,
      "step": 76295
    },
    {
      "epoch": 11.364313375037236,
      "grad_norm": 11.3125,
      "learning_rate": 2.341210251907953e-05,
      "loss": 1.1772,
      "num_input_tokens_seen": 44265104,
      "step": 76300
    },
    {
      "epoch": 11.365058087578195,
      "grad_norm": 10.5,
      "learning_rate": 2.340885967219136e-05,
      "loss": 0.9195,
      "num_input_tokens_seen": 44267824,
      "step": 76305
    },
    {
      "epoch": 11.365802800119154,
      "grad_norm": 13.3125,
      "learning_rate": 2.3405616852183902e-05,
      "loss": 1.0955,
      "num_input_tokens_seen": 44270928,
      "step": 76310
    },
    {
      "epoch": 11.366547512660112,
      "grad_norm": 10.5,
      "learning_rate": 2.3402374059111912e-05,
      "loss": 1.0911,
      "num_input_tokens_seen": 44273712,
      "step": 76315
    },
    {
      "epoch": 11.367292225201073,
      "grad_norm": 12.4375,
      "learning_rate": 2.3399131293030204e-05,
      "loss": 1.4082,
      "num_input_tokens_seen": 44276304,
      "step": 76320
    },
    {
      "epoch": 11.368036937742032,
      "grad_norm": 8.0,
      "learning_rate": 2.339588855399354e-05,
      "loss": 0.9292,
      "num_input_tokens_seen": 44279056,
      "step": 76325
    },
    {
      "epoch": 11.36878165028299,
      "grad_norm": 8.9375,
      "learning_rate": 2.3392645842056707e-05,
      "loss": 0.77,
      "num_input_tokens_seen": 44282000,
      "step": 76330
    },
    {
      "epoch": 11.36952636282395,
      "grad_norm": 11.625,
      "learning_rate": 2.338940315727449e-05,
      "loss": 1.112,
      "num_input_tokens_seen": 44284944,
      "step": 76335
    },
    {
      "epoch": 11.37027107536491,
      "grad_norm": 7.125,
      "learning_rate": 2.3386160499701663e-05,
      "loss": 1.1951,
      "num_input_tokens_seen": 44288048,
      "step": 76340
    },
    {
      "epoch": 11.371015787905868,
      "grad_norm": 3.765625,
      "learning_rate": 2.3382917869393027e-05,
      "loss": 1.0953,
      "num_input_tokens_seen": 44290896,
      "step": 76345
    },
    {
      "epoch": 11.371760500446827,
      "grad_norm": 14.625,
      "learning_rate": 2.3379675266403335e-05,
      "loss": 1.2646,
      "num_input_tokens_seen": 44294192,
      "step": 76350
    },
    {
      "epoch": 11.372505212987786,
      "grad_norm": 15.3125,
      "learning_rate": 2.3376432690787396e-05,
      "loss": 0.9398,
      "num_input_tokens_seen": 44297296,
      "step": 76355
    },
    {
      "epoch": 11.373249925528746,
      "grad_norm": 8.0,
      "learning_rate": 2.3373190142599973e-05,
      "loss": 0.9872,
      "num_input_tokens_seen": 44300016,
      "step": 76360
    },
    {
      "epoch": 11.373994638069705,
      "grad_norm": 12.375,
      "learning_rate": 2.3369947621895845e-05,
      "loss": 1.2816,
      "num_input_tokens_seen": 44302864,
      "step": 76365
    },
    {
      "epoch": 11.374739350610664,
      "grad_norm": 5.9375,
      "learning_rate": 2.3366705128729805e-05,
      "loss": 0.8165,
      "num_input_tokens_seen": 44305616,
      "step": 76370
    },
    {
      "epoch": 11.375484063151623,
      "grad_norm": 9.9375,
      "learning_rate": 2.3363462663156606e-05,
      "loss": 0.9157,
      "num_input_tokens_seen": 44308816,
      "step": 76375
    },
    {
      "epoch": 11.376228775692583,
      "grad_norm": 16.0,
      "learning_rate": 2.3360220225231057e-05,
      "loss": 0.9939,
      "num_input_tokens_seen": 44311824,
      "step": 76380
    },
    {
      "epoch": 11.376973488233542,
      "grad_norm": 8.5625,
      "learning_rate": 2.335697781500791e-05,
      "loss": 1.0622,
      "num_input_tokens_seen": 44314864,
      "step": 76385
    },
    {
      "epoch": 11.3777182007745,
      "grad_norm": 10.75,
      "learning_rate": 2.3353735432541957e-05,
      "loss": 1.2934,
      "num_input_tokens_seen": 44317776,
      "step": 76390
    },
    {
      "epoch": 11.37846291331546,
      "grad_norm": 16.0,
      "learning_rate": 2.335049307788797e-05,
      "loss": 1.2759,
      "num_input_tokens_seen": 44320688,
      "step": 76395
    },
    {
      "epoch": 11.37920762585642,
      "grad_norm": 9.6875,
      "learning_rate": 2.334725075110073e-05,
      "loss": 1.3318,
      "num_input_tokens_seen": 44323632,
      "step": 76400
    },
    {
      "epoch": 11.379952338397379,
      "grad_norm": 9.375,
      "learning_rate": 2.3344008452235008e-05,
      "loss": 1.0389,
      "num_input_tokens_seen": 44326416,
      "step": 76405
    },
    {
      "epoch": 11.380697050938338,
      "grad_norm": 8.25,
      "learning_rate": 2.3340766181345572e-05,
      "loss": 0.6842,
      "num_input_tokens_seen": 44329328,
      "step": 76410
    },
    {
      "epoch": 11.381441763479296,
      "grad_norm": 7.03125,
      "learning_rate": 2.3337523938487214e-05,
      "loss": 0.9834,
      "num_input_tokens_seen": 44332176,
      "step": 76415
    },
    {
      "epoch": 11.382186476020257,
      "grad_norm": 5.84375,
      "learning_rate": 2.3334281723714694e-05,
      "loss": 0.7944,
      "num_input_tokens_seen": 44335088,
      "step": 76420
    },
    {
      "epoch": 11.382931188561216,
      "grad_norm": 11.625,
      "learning_rate": 2.3331039537082796e-05,
      "loss": 1.1241,
      "num_input_tokens_seen": 44338096,
      "step": 76425
    },
    {
      "epoch": 11.383675901102174,
      "grad_norm": 12.0625,
      "learning_rate": 2.332779737864628e-05,
      "loss": 1.0658,
      "num_input_tokens_seen": 44341040,
      "step": 76430
    },
    {
      "epoch": 11.384420613643133,
      "grad_norm": 9.8125,
      "learning_rate": 2.3324555248459938e-05,
      "loss": 1.3637,
      "num_input_tokens_seen": 44344240,
      "step": 76435
    },
    {
      "epoch": 11.385165326184094,
      "grad_norm": 6.0625,
      "learning_rate": 2.3321313146578532e-05,
      "loss": 1.1823,
      "num_input_tokens_seen": 44346928,
      "step": 76440
    },
    {
      "epoch": 11.385910038725052,
      "grad_norm": 5.21875,
      "learning_rate": 2.3318071073056826e-05,
      "loss": 0.9154,
      "num_input_tokens_seen": 44349872,
      "step": 76445
    },
    {
      "epoch": 11.386654751266011,
      "grad_norm": 9.375,
      "learning_rate": 2.3314829027949606e-05,
      "loss": 1.08,
      "num_input_tokens_seen": 44352624,
      "step": 76450
    },
    {
      "epoch": 11.38739946380697,
      "grad_norm": 10.625,
      "learning_rate": 2.3311587011311634e-05,
      "loss": 0.933,
      "num_input_tokens_seen": 44355600,
      "step": 76455
    },
    {
      "epoch": 11.38814417634793,
      "grad_norm": 11.1875,
      "learning_rate": 2.330834502319769e-05,
      "loss": 1.2559,
      "num_input_tokens_seen": 44358384,
      "step": 76460
    },
    {
      "epoch": 11.38888888888889,
      "grad_norm": 8.5,
      "learning_rate": 2.3305103063662522e-05,
      "loss": 1.2334,
      "num_input_tokens_seen": 44361584,
      "step": 76465
    },
    {
      "epoch": 11.389633601429848,
      "grad_norm": 11.5,
      "learning_rate": 2.330186113276093e-05,
      "loss": 1.0757,
      "num_input_tokens_seen": 44364976,
      "step": 76470
    },
    {
      "epoch": 11.390378313970807,
      "grad_norm": 8.0625,
      "learning_rate": 2.3298619230547656e-05,
      "loss": 0.7809,
      "num_input_tokens_seen": 44367856,
      "step": 76475
    },
    {
      "epoch": 11.391123026511767,
      "grad_norm": 9.6875,
      "learning_rate": 2.329537735707749e-05,
      "loss": 1.2721,
      "num_input_tokens_seen": 44370608,
      "step": 76480
    },
    {
      "epoch": 11.391867739052726,
      "grad_norm": 6.46875,
      "learning_rate": 2.3292135512405198e-05,
      "loss": 1.0215,
      "num_input_tokens_seen": 44373456,
      "step": 76485
    },
    {
      "epoch": 11.392612451593685,
      "grad_norm": 9.1875,
      "learning_rate": 2.3288893696585528e-05,
      "loss": 1.0614,
      "num_input_tokens_seen": 44376368,
      "step": 76490
    },
    {
      "epoch": 11.393357164134644,
      "grad_norm": 13.25,
      "learning_rate": 2.328565190967327e-05,
      "loss": 1.2693,
      "num_input_tokens_seen": 44379504,
      "step": 76495
    },
    {
      "epoch": 11.394101876675602,
      "grad_norm": 7.25,
      "learning_rate": 2.3282410151723167e-05,
      "loss": 0.7706,
      "num_input_tokens_seen": 44382256,
      "step": 76500
    },
    {
      "epoch": 11.394846589216563,
      "grad_norm": 7.28125,
      "learning_rate": 2.327916842279001e-05,
      "loss": 0.9959,
      "num_input_tokens_seen": 44385264,
      "step": 76505
    },
    {
      "epoch": 11.395591301757522,
      "grad_norm": 13.0,
      "learning_rate": 2.3275926722928542e-05,
      "loss": 1.0229,
      "num_input_tokens_seen": 44387920,
      "step": 76510
    },
    {
      "epoch": 11.39633601429848,
      "grad_norm": 16.0,
      "learning_rate": 2.327268505219355e-05,
      "loss": 1.0558,
      "num_input_tokens_seen": 44390704,
      "step": 76515
    },
    {
      "epoch": 11.39708072683944,
      "grad_norm": 7.59375,
      "learning_rate": 2.326944341063979e-05,
      "loss": 1.2339,
      "num_input_tokens_seen": 44393840,
      "step": 76520
    },
    {
      "epoch": 11.3978254393804,
      "grad_norm": 16.75,
      "learning_rate": 2.326620179832202e-05,
      "loss": 1.1921,
      "num_input_tokens_seen": 44397040,
      "step": 76525
    },
    {
      "epoch": 11.398570151921358,
      "grad_norm": 7.71875,
      "learning_rate": 2.3262960215295014e-05,
      "loss": 1.2656,
      "num_input_tokens_seen": 44400304,
      "step": 76530
    },
    {
      "epoch": 11.399314864462317,
      "grad_norm": 15.4375,
      "learning_rate": 2.3259718661613518e-05,
      "loss": 1.0278,
      "num_input_tokens_seen": 44403024,
      "step": 76535
    },
    {
      "epoch": 11.400059577003276,
      "grad_norm": 8.625,
      "learning_rate": 2.3256477137332315e-05,
      "loss": 1.2451,
      "num_input_tokens_seen": 44406320,
      "step": 76540
    },
    {
      "epoch": 11.400804289544237,
      "grad_norm": 7.1875,
      "learning_rate": 2.325323564250615e-05,
      "loss": 0.797,
      "num_input_tokens_seen": 44409488,
      "step": 76545
    },
    {
      "epoch": 11.401549002085195,
      "grad_norm": 17.0,
      "learning_rate": 2.324999417718981e-05,
      "loss": 0.9016,
      "num_input_tokens_seen": 44412368,
      "step": 76550
    },
    {
      "epoch": 11.402293714626154,
      "grad_norm": 10.9375,
      "learning_rate": 2.3246752741438026e-05,
      "loss": 1.1177,
      "num_input_tokens_seen": 44415344,
      "step": 76555
    },
    {
      "epoch": 11.403038427167113,
      "grad_norm": 5.96875,
      "learning_rate": 2.324351133530558e-05,
      "loss": 1.065,
      "num_input_tokens_seen": 44418160,
      "step": 76560
    },
    {
      "epoch": 11.403783139708073,
      "grad_norm": 7.25,
      "learning_rate": 2.3240269958847226e-05,
      "loss": 0.9703,
      "num_input_tokens_seen": 44421200,
      "step": 76565
    },
    {
      "epoch": 11.404527852249032,
      "grad_norm": 9.75,
      "learning_rate": 2.3237028612117712e-05,
      "loss": 1.0297,
      "num_input_tokens_seen": 44424048,
      "step": 76570
    },
    {
      "epoch": 11.40527256478999,
      "grad_norm": 11.9375,
      "learning_rate": 2.3233787295171818e-05,
      "loss": 1.0688,
      "num_input_tokens_seen": 44426736,
      "step": 76575
    },
    {
      "epoch": 11.40601727733095,
      "grad_norm": 7.1875,
      "learning_rate": 2.323054600806428e-05,
      "loss": 1.31,
      "num_input_tokens_seen": 44429712,
      "step": 76580
    },
    {
      "epoch": 11.40676198987191,
      "grad_norm": 13.0,
      "learning_rate": 2.322730475084988e-05,
      "loss": 0.9244,
      "num_input_tokens_seen": 44432656,
      "step": 76585
    },
    {
      "epoch": 11.407506702412869,
      "grad_norm": 8.8125,
      "learning_rate": 2.3224063523583363e-05,
      "loss": 1.1813,
      "num_input_tokens_seen": 44435472,
      "step": 76590
    },
    {
      "epoch": 11.408251414953828,
      "grad_norm": 9.625,
      "learning_rate": 2.322082232631949e-05,
      "loss": 1.098,
      "num_input_tokens_seen": 44438128,
      "step": 76595
    },
    {
      "epoch": 11.408996127494786,
      "grad_norm": 14.0,
      "learning_rate": 2.3217581159113016e-05,
      "loss": 1.4914,
      "num_input_tokens_seen": 44440848,
      "step": 76600
    },
    {
      "epoch": 11.409740840035747,
      "grad_norm": 14.9375,
      "learning_rate": 2.3214340022018688e-05,
      "loss": 1.0898,
      "num_input_tokens_seen": 44443792,
      "step": 76605
    },
    {
      "epoch": 11.410485552576706,
      "grad_norm": 8.1875,
      "learning_rate": 2.321109891509128e-05,
      "loss": 0.9976,
      "num_input_tokens_seen": 44446416,
      "step": 76610
    },
    {
      "epoch": 11.411230265117664,
      "grad_norm": 6.65625,
      "learning_rate": 2.3207857838385524e-05,
      "loss": 1.1363,
      "num_input_tokens_seen": 44448944,
      "step": 76615
    },
    {
      "epoch": 11.411974977658623,
      "grad_norm": 13.0,
      "learning_rate": 2.32046167919562e-05,
      "loss": 1.2604,
      "num_input_tokens_seen": 44452304,
      "step": 76620
    },
    {
      "epoch": 11.412719690199584,
      "grad_norm": 7.5625,
      "learning_rate": 2.320137577585805e-05,
      "loss": 1.388,
      "num_input_tokens_seen": 44455184,
      "step": 76625
    },
    {
      "epoch": 11.413464402740543,
      "grad_norm": 10.8125,
      "learning_rate": 2.319813479014583e-05,
      "loss": 0.8981,
      "num_input_tokens_seen": 44458224,
      "step": 76630
    },
    {
      "epoch": 11.414209115281501,
      "grad_norm": 15.125,
      "learning_rate": 2.319489383487428e-05,
      "loss": 1.0321,
      "num_input_tokens_seen": 44461072,
      "step": 76635
    },
    {
      "epoch": 11.41495382782246,
      "grad_norm": 10.0625,
      "learning_rate": 2.3191652910098174e-05,
      "loss": 1.2826,
      "num_input_tokens_seen": 44463920,
      "step": 76640
    },
    {
      "epoch": 11.41569854036342,
      "grad_norm": 12.9375,
      "learning_rate": 2.3188412015872258e-05,
      "loss": 1.0374,
      "num_input_tokens_seen": 44466800,
      "step": 76645
    },
    {
      "epoch": 11.41644325290438,
      "grad_norm": 11.6875,
      "learning_rate": 2.3185171152251265e-05,
      "loss": 1.2529,
      "num_input_tokens_seen": 44469648,
      "step": 76650
    },
    {
      "epoch": 11.417187965445338,
      "grad_norm": 11.4375,
      "learning_rate": 2.3181930319289975e-05,
      "loss": 0.8876,
      "num_input_tokens_seen": 44473072,
      "step": 76655
    },
    {
      "epoch": 11.417932677986297,
      "grad_norm": 5.9375,
      "learning_rate": 2.3178689517043116e-05,
      "loss": 1.255,
      "num_input_tokens_seen": 44476176,
      "step": 76660
    },
    {
      "epoch": 11.418677390527257,
      "grad_norm": 7.4375,
      "learning_rate": 2.3175448745565454e-05,
      "loss": 1.3178,
      "num_input_tokens_seen": 44479024,
      "step": 76665
    },
    {
      "epoch": 11.419422103068216,
      "grad_norm": 9.375,
      "learning_rate": 2.317220800491172e-05,
      "loss": 0.9459,
      "num_input_tokens_seen": 44482064,
      "step": 76670
    },
    {
      "epoch": 11.420166815609175,
      "grad_norm": 13.8125,
      "learning_rate": 2.3168967295136685e-05,
      "loss": 1.3147,
      "num_input_tokens_seen": 44485264,
      "step": 76675
    },
    {
      "epoch": 11.420911528150134,
      "grad_norm": 14.75,
      "learning_rate": 2.3165726616295083e-05,
      "loss": 1.132,
      "num_input_tokens_seen": 44488080,
      "step": 76680
    },
    {
      "epoch": 11.421656240691092,
      "grad_norm": 8.375,
      "learning_rate": 2.316248596844166e-05,
      "loss": 0.7669,
      "num_input_tokens_seen": 44490736,
      "step": 76685
    },
    {
      "epoch": 11.422400953232053,
      "grad_norm": 8.25,
      "learning_rate": 2.3159245351631176e-05,
      "loss": 0.9557,
      "num_input_tokens_seen": 44493968,
      "step": 76690
    },
    {
      "epoch": 11.423145665773012,
      "grad_norm": 9.9375,
      "learning_rate": 2.315600476591837e-05,
      "loss": 1.0403,
      "num_input_tokens_seen": 44497168,
      "step": 76695
    },
    {
      "epoch": 11.42389037831397,
      "grad_norm": 14.8125,
      "learning_rate": 2.3152764211357988e-05,
      "loss": 1.2258,
      "num_input_tokens_seen": 44499824,
      "step": 76700
    },
    {
      "epoch": 11.42463509085493,
      "grad_norm": 9.8125,
      "learning_rate": 2.314952368800477e-05,
      "loss": 0.8479,
      "num_input_tokens_seen": 44502768,
      "step": 76705
    },
    {
      "epoch": 11.42537980339589,
      "grad_norm": 17.5,
      "learning_rate": 2.3146283195913482e-05,
      "loss": 1.1821,
      "num_input_tokens_seen": 44505456,
      "step": 76710
    },
    {
      "epoch": 11.426124515936849,
      "grad_norm": 6.6875,
      "learning_rate": 2.3143042735138848e-05,
      "loss": 1.1217,
      "num_input_tokens_seen": 44508144,
      "step": 76715
    },
    {
      "epoch": 11.426869228477807,
      "grad_norm": 6.90625,
      "learning_rate": 2.3139802305735618e-05,
      "loss": 0.7048,
      "num_input_tokens_seen": 44511056,
      "step": 76720
    },
    {
      "epoch": 11.427613941018766,
      "grad_norm": 4.03125,
      "learning_rate": 2.3136561907758543e-05,
      "loss": 1.1095,
      "num_input_tokens_seen": 44514000,
      "step": 76725
    },
    {
      "epoch": 11.428358653559727,
      "grad_norm": 11.8125,
      "learning_rate": 2.3133321541262356e-05,
      "loss": 1.0685,
      "num_input_tokens_seen": 44517104,
      "step": 76730
    },
    {
      "epoch": 11.429103366100685,
      "grad_norm": 10.8125,
      "learning_rate": 2.3130081206301812e-05,
      "loss": 0.9453,
      "num_input_tokens_seen": 44519696,
      "step": 76735
    },
    {
      "epoch": 11.429848078641644,
      "grad_norm": 11.5,
      "learning_rate": 2.3126840902931633e-05,
      "loss": 1.2134,
      "num_input_tokens_seen": 44522384,
      "step": 76740
    },
    {
      "epoch": 11.430592791182603,
      "grad_norm": 6.75,
      "learning_rate": 2.312360063120658e-05,
      "loss": 1.1212,
      "num_input_tokens_seen": 44525232,
      "step": 76745
    },
    {
      "epoch": 11.431337503723563,
      "grad_norm": 20.5,
      "learning_rate": 2.3120360391181388e-05,
      "loss": 1.2287,
      "num_input_tokens_seen": 44528176,
      "step": 76750
    },
    {
      "epoch": 11.432082216264522,
      "grad_norm": 13.1875,
      "learning_rate": 2.31171201829108e-05,
      "loss": 1.0237,
      "num_input_tokens_seen": 44531152,
      "step": 76755
    },
    {
      "epoch": 11.432826928805481,
      "grad_norm": 9.25,
      "learning_rate": 2.3113880006449547e-05,
      "loss": 0.8943,
      "num_input_tokens_seen": 44534160,
      "step": 76760
    },
    {
      "epoch": 11.43357164134644,
      "grad_norm": 10.0625,
      "learning_rate": 2.3110639861852373e-05,
      "loss": 1.2505,
      "num_input_tokens_seen": 44536688,
      "step": 76765
    },
    {
      "epoch": 11.4343163538874,
      "grad_norm": 10.0625,
      "learning_rate": 2.3107399749174027e-05,
      "loss": 1.1768,
      "num_input_tokens_seen": 44539536,
      "step": 76770
    },
    {
      "epoch": 11.435061066428359,
      "grad_norm": 7.9375,
      "learning_rate": 2.3104159668469226e-05,
      "loss": 1.4208,
      "num_input_tokens_seen": 44542320,
      "step": 76775
    },
    {
      "epoch": 11.435805778969318,
      "grad_norm": 10.375,
      "learning_rate": 2.3100919619792733e-05,
      "loss": 0.887,
      "num_input_tokens_seen": 44545104,
      "step": 76780
    },
    {
      "epoch": 11.436550491510276,
      "grad_norm": 10.5625,
      "learning_rate": 2.3097679603199267e-05,
      "loss": 1.2953,
      "num_input_tokens_seen": 44548080,
      "step": 76785
    },
    {
      "epoch": 11.437295204051237,
      "grad_norm": 8.6875,
      "learning_rate": 2.3094439618743572e-05,
      "loss": 0.844,
      "num_input_tokens_seen": 44551312,
      "step": 76790
    },
    {
      "epoch": 11.438039916592196,
      "grad_norm": 9.5625,
      "learning_rate": 2.3091199666480377e-05,
      "loss": 0.8623,
      "num_input_tokens_seen": 44554480,
      "step": 76795
    },
    {
      "epoch": 11.438784629133155,
      "grad_norm": 9.6875,
      "learning_rate": 2.3087959746464432e-05,
      "loss": 0.6992,
      "num_input_tokens_seen": 44557296,
      "step": 76800
    },
    {
      "epoch": 11.439529341674113,
      "grad_norm": 20.0,
      "learning_rate": 2.3084719858750464e-05,
      "loss": 1.7072,
      "num_input_tokens_seen": 44560208,
      "step": 76805
    },
    {
      "epoch": 11.440274054215074,
      "grad_norm": 15.125,
      "learning_rate": 2.3081480003393198e-05,
      "loss": 1.2324,
      "num_input_tokens_seen": 44563152,
      "step": 76810
    },
    {
      "epoch": 11.441018766756033,
      "grad_norm": 12.3125,
      "learning_rate": 2.3078240180447384e-05,
      "loss": 1.1498,
      "num_input_tokens_seen": 44565968,
      "step": 76815
    },
    {
      "epoch": 11.441763479296991,
      "grad_norm": 27.0,
      "learning_rate": 2.307500038996775e-05,
      "loss": 1.6111,
      "num_input_tokens_seen": 44568560,
      "step": 76820
    },
    {
      "epoch": 11.44250819183795,
      "grad_norm": 9.8125,
      "learning_rate": 2.3071760632009028e-05,
      "loss": 0.931,
      "num_input_tokens_seen": 44571312,
      "step": 76825
    },
    {
      "epoch": 11.443252904378909,
      "grad_norm": 10.75,
      "learning_rate": 2.3068520906625943e-05,
      "loss": 1.0361,
      "num_input_tokens_seen": 44573936,
      "step": 76830
    },
    {
      "epoch": 11.44399761691987,
      "grad_norm": 13.3125,
      "learning_rate": 2.306528121387324e-05,
      "loss": 1.1079,
      "num_input_tokens_seen": 44577008,
      "step": 76835
    },
    {
      "epoch": 11.444742329460828,
      "grad_norm": 7.03125,
      "learning_rate": 2.306204155380565e-05,
      "loss": 0.8635,
      "num_input_tokens_seen": 44579856,
      "step": 76840
    },
    {
      "epoch": 11.445487042001787,
      "grad_norm": 9.0625,
      "learning_rate": 2.3058801926477885e-05,
      "loss": 0.9742,
      "num_input_tokens_seen": 44582992,
      "step": 76845
    },
    {
      "epoch": 11.446231754542746,
      "grad_norm": 9.25,
      "learning_rate": 2.3055562331944703e-05,
      "loss": 1.1888,
      "num_input_tokens_seen": 44585904,
      "step": 76850
    },
    {
      "epoch": 11.446976467083706,
      "grad_norm": 20.625,
      "learning_rate": 2.3052322770260808e-05,
      "loss": 1.3733,
      "num_input_tokens_seen": 44588976,
      "step": 76855
    },
    {
      "epoch": 11.447721179624665,
      "grad_norm": 6.75,
      "learning_rate": 2.3049083241480948e-05,
      "loss": 0.8176,
      "num_input_tokens_seen": 44591664,
      "step": 76860
    },
    {
      "epoch": 11.448465892165624,
      "grad_norm": 10.0,
      "learning_rate": 2.3045843745659834e-05,
      "loss": 1.0859,
      "num_input_tokens_seen": 44594576,
      "step": 76865
    },
    {
      "epoch": 11.449210604706582,
      "grad_norm": 9.0,
      "learning_rate": 2.3042604282852215e-05,
      "loss": 0.7188,
      "num_input_tokens_seen": 44597552,
      "step": 76870
    },
    {
      "epoch": 11.449955317247543,
      "grad_norm": 10.4375,
      "learning_rate": 2.3039364853112794e-05,
      "loss": 1.0685,
      "num_input_tokens_seen": 44600368,
      "step": 76875
    },
    {
      "epoch": 11.450700029788502,
      "grad_norm": 13.6875,
      "learning_rate": 2.3036125456496324e-05,
      "loss": 1.3442,
      "num_input_tokens_seen": 44603440,
      "step": 76880
    },
    {
      "epoch": 11.45144474232946,
      "grad_norm": 8.125,
      "learning_rate": 2.303288609305752e-05,
      "loss": 1.0186,
      "num_input_tokens_seen": 44606288,
      "step": 76885
    },
    {
      "epoch": 11.45218945487042,
      "grad_norm": 18.375,
      "learning_rate": 2.3029646762851096e-05,
      "loss": 0.9865,
      "num_input_tokens_seen": 44609040,
      "step": 76890
    },
    {
      "epoch": 11.45293416741138,
      "grad_norm": 7.625,
      "learning_rate": 2.3026407465931797e-05,
      "loss": 0.9206,
      "num_input_tokens_seen": 44611984,
      "step": 76895
    },
    {
      "epoch": 11.453678879952339,
      "grad_norm": 5.375,
      "learning_rate": 2.3023168202354324e-05,
      "loss": 1.207,
      "num_input_tokens_seen": 44615152,
      "step": 76900
    },
    {
      "epoch": 11.454423592493297,
      "grad_norm": 14.9375,
      "learning_rate": 2.301992897217343e-05,
      "loss": 1.3212,
      "num_input_tokens_seen": 44618064,
      "step": 76905
    },
    {
      "epoch": 11.455168305034256,
      "grad_norm": 13.375,
      "learning_rate": 2.3016689775443806e-05,
      "loss": 1.0113,
      "num_input_tokens_seen": 44620976,
      "step": 76910
    },
    {
      "epoch": 11.455913017575217,
      "grad_norm": 10.0,
      "learning_rate": 2.3013450612220207e-05,
      "loss": 0.8613,
      "num_input_tokens_seen": 44623664,
      "step": 76915
    },
    {
      "epoch": 11.456657730116175,
      "grad_norm": 7.5,
      "learning_rate": 2.3010211482557335e-05,
      "loss": 1.0535,
      "num_input_tokens_seen": 44626544,
      "step": 76920
    },
    {
      "epoch": 11.457402442657134,
      "grad_norm": 6.71875,
      "learning_rate": 2.3006972386509925e-05,
      "loss": 1.1145,
      "num_input_tokens_seen": 44629584,
      "step": 76925
    },
    {
      "epoch": 11.458147155198093,
      "grad_norm": 13.625,
      "learning_rate": 2.3003733324132693e-05,
      "loss": 1.3875,
      "num_input_tokens_seen": 44632592,
      "step": 76930
    },
    {
      "epoch": 11.458891867739053,
      "grad_norm": 16.125,
      "learning_rate": 2.300049429548034e-05,
      "loss": 1.4101,
      "num_input_tokens_seen": 44635344,
      "step": 76935
    },
    {
      "epoch": 11.459636580280012,
      "grad_norm": 9.4375,
      "learning_rate": 2.299725530060762e-05,
      "loss": 1.0609,
      "num_input_tokens_seen": 44638160,
      "step": 76940
    },
    {
      "epoch": 11.460381292820971,
      "grad_norm": 12.9375,
      "learning_rate": 2.2994016339569224e-05,
      "loss": 1.0288,
      "num_input_tokens_seen": 44641072,
      "step": 76945
    },
    {
      "epoch": 11.46112600536193,
      "grad_norm": 7.25,
      "learning_rate": 2.2990777412419892e-05,
      "loss": 0.908,
      "num_input_tokens_seen": 44643856,
      "step": 76950
    },
    {
      "epoch": 11.46187071790289,
      "grad_norm": 8.5625,
      "learning_rate": 2.298753851921433e-05,
      "loss": 0.9905,
      "num_input_tokens_seen": 44647056,
      "step": 76955
    },
    {
      "epoch": 11.462615430443849,
      "grad_norm": 12.125,
      "learning_rate": 2.2984299660007263e-05,
      "loss": 1.1838,
      "num_input_tokens_seen": 44650096,
      "step": 76960
    },
    {
      "epoch": 11.463360142984808,
      "grad_norm": 9.5625,
      "learning_rate": 2.2981060834853406e-05,
      "loss": 1.1954,
      "num_input_tokens_seen": 44653168,
      "step": 76965
    },
    {
      "epoch": 11.464104855525767,
      "grad_norm": 26.125,
      "learning_rate": 2.2977822043807466e-05,
      "loss": 1.0315,
      "num_input_tokens_seen": 44656272,
      "step": 76970
    },
    {
      "epoch": 11.464849568066727,
      "grad_norm": 17.625,
      "learning_rate": 2.2974583286924176e-05,
      "loss": 1.3583,
      "num_input_tokens_seen": 44659088,
      "step": 76975
    },
    {
      "epoch": 11.465594280607686,
      "grad_norm": 12.25,
      "learning_rate": 2.297134456425823e-05,
      "loss": 0.9416,
      "num_input_tokens_seen": 44662224,
      "step": 76980
    },
    {
      "epoch": 11.466338993148645,
      "grad_norm": 8.75,
      "learning_rate": 2.2968105875864368e-05,
      "loss": 0.7746,
      "num_input_tokens_seen": 44664848,
      "step": 76985
    },
    {
      "epoch": 11.467083705689603,
      "grad_norm": 7.28125,
      "learning_rate": 2.2964867221797286e-05,
      "loss": 1.0059,
      "num_input_tokens_seen": 44667888,
      "step": 76990
    },
    {
      "epoch": 11.467828418230564,
      "grad_norm": 18.5,
      "learning_rate": 2.296162860211171e-05,
      "loss": 1.2586,
      "num_input_tokens_seen": 44670768,
      "step": 76995
    },
    {
      "epoch": 11.468573130771523,
      "grad_norm": 8.0,
      "learning_rate": 2.2958390016862335e-05,
      "loss": 1.1821,
      "num_input_tokens_seen": 44673520,
      "step": 77000
    },
    {
      "epoch": 11.469317843312481,
      "grad_norm": 13.6875,
      "learning_rate": 2.29551514661039e-05,
      "loss": 0.8617,
      "num_input_tokens_seen": 44676528,
      "step": 77005
    },
    {
      "epoch": 11.47006255585344,
      "grad_norm": 8.875,
      "learning_rate": 2.2951912949891098e-05,
      "loss": 0.9218,
      "num_input_tokens_seen": 44679536,
      "step": 77010
    },
    {
      "epoch": 11.470807268394399,
      "grad_norm": 9.25,
      "learning_rate": 2.294867446827864e-05,
      "loss": 0.893,
      "num_input_tokens_seen": 44682320,
      "step": 77015
    },
    {
      "epoch": 11.47155198093536,
      "grad_norm": 12.3125,
      "learning_rate": 2.294543602132125e-05,
      "loss": 0.8729,
      "num_input_tokens_seen": 44685232,
      "step": 77020
    },
    {
      "epoch": 11.472296693476318,
      "grad_norm": 11.125,
      "learning_rate": 2.2942197609073624e-05,
      "loss": 1.005,
      "num_input_tokens_seen": 44687856,
      "step": 77025
    },
    {
      "epoch": 11.473041406017277,
      "grad_norm": 5.03125,
      "learning_rate": 2.2938959231590483e-05,
      "loss": 0.8563,
      "num_input_tokens_seen": 44691056,
      "step": 77030
    },
    {
      "epoch": 11.473786118558236,
      "grad_norm": 8.3125,
      "learning_rate": 2.2935720888926522e-05,
      "loss": 1.0779,
      "num_input_tokens_seen": 44693808,
      "step": 77035
    },
    {
      "epoch": 11.474530831099196,
      "grad_norm": 10.4375,
      "learning_rate": 2.2932482581136466e-05,
      "loss": 0.9286,
      "num_input_tokens_seen": 44696496,
      "step": 77040
    },
    {
      "epoch": 11.475275543640155,
      "grad_norm": 9.0625,
      "learning_rate": 2.292924430827502e-05,
      "loss": 0.9677,
      "num_input_tokens_seen": 44699376,
      "step": 77045
    },
    {
      "epoch": 11.476020256181114,
      "grad_norm": 8.5625,
      "learning_rate": 2.292600607039687e-05,
      "loss": 1.1432,
      "num_input_tokens_seen": 44702256,
      "step": 77050
    },
    {
      "epoch": 11.476764968722073,
      "grad_norm": 10.25,
      "learning_rate": 2.2922767867556755e-05,
      "loss": 1.2779,
      "num_input_tokens_seen": 44704944,
      "step": 77055
    },
    {
      "epoch": 11.477509681263033,
      "grad_norm": 12.4375,
      "learning_rate": 2.291952969980936e-05,
      "loss": 0.9845,
      "num_input_tokens_seen": 44707888,
      "step": 77060
    },
    {
      "epoch": 11.478254393803992,
      "grad_norm": 14.3125,
      "learning_rate": 2.29162915672094e-05,
      "loss": 1.8236,
      "num_input_tokens_seen": 44710448,
      "step": 77065
    },
    {
      "epoch": 11.47899910634495,
      "grad_norm": 9.6875,
      "learning_rate": 2.2913053469811568e-05,
      "loss": 1.0287,
      "num_input_tokens_seen": 44713360,
      "step": 77070
    },
    {
      "epoch": 11.47974381888591,
      "grad_norm": 12.4375,
      "learning_rate": 2.2909815407670584e-05,
      "loss": 1.4231,
      "num_input_tokens_seen": 44716080,
      "step": 77075
    },
    {
      "epoch": 11.48048853142687,
      "grad_norm": 20.375,
      "learning_rate": 2.2906577380841143e-05,
      "loss": 1.1192,
      "num_input_tokens_seen": 44718928,
      "step": 77080
    },
    {
      "epoch": 11.481233243967829,
      "grad_norm": 6.71875,
      "learning_rate": 2.290333938937795e-05,
      "loss": 1.0047,
      "num_input_tokens_seen": 44721776,
      "step": 77085
    },
    {
      "epoch": 11.481977956508787,
      "grad_norm": 12.125,
      "learning_rate": 2.2900101433335704e-05,
      "loss": 1.2141,
      "num_input_tokens_seen": 44724560,
      "step": 77090
    },
    {
      "epoch": 11.482722669049746,
      "grad_norm": 16.25,
      "learning_rate": 2.289686351276911e-05,
      "loss": 1.3321,
      "num_input_tokens_seen": 44727312,
      "step": 77095
    },
    {
      "epoch": 11.483467381590707,
      "grad_norm": 8.8125,
      "learning_rate": 2.2893625627732877e-05,
      "loss": 1.2151,
      "num_input_tokens_seen": 44730128,
      "step": 77100
    },
    {
      "epoch": 11.484212094131665,
      "grad_norm": 10.8125,
      "learning_rate": 2.2890387778281686e-05,
      "loss": 1.151,
      "num_input_tokens_seen": 44733040,
      "step": 77105
    },
    {
      "epoch": 11.484956806672624,
      "grad_norm": 11.6875,
      "learning_rate": 2.2887149964470258e-05,
      "loss": 1.238,
      "num_input_tokens_seen": 44735984,
      "step": 77110
    },
    {
      "epoch": 11.485701519213583,
      "grad_norm": 7.375,
      "learning_rate": 2.2883912186353282e-05,
      "loss": 0.9881,
      "num_input_tokens_seen": 44738672,
      "step": 77115
    },
    {
      "epoch": 11.486446231754543,
      "grad_norm": 14.125,
      "learning_rate": 2.288067444398546e-05,
      "loss": 0.6889,
      "num_input_tokens_seen": 44741648,
      "step": 77120
    },
    {
      "epoch": 11.487190944295502,
      "grad_norm": 8.75,
      "learning_rate": 2.2877436737421494e-05,
      "loss": 1.1378,
      "num_input_tokens_seen": 44744432,
      "step": 77125
    },
    {
      "epoch": 11.487935656836461,
      "grad_norm": 10.375,
      "learning_rate": 2.287419906671606e-05,
      "loss": 1.2763,
      "num_input_tokens_seen": 44747280,
      "step": 77130
    },
    {
      "epoch": 11.48868036937742,
      "grad_norm": 7.8125,
      "learning_rate": 2.287096143192389e-05,
      "loss": 1.1079,
      "num_input_tokens_seen": 44750032,
      "step": 77135
    },
    {
      "epoch": 11.48942508191838,
      "grad_norm": 7.46875,
      "learning_rate": 2.286772383309965e-05,
      "loss": 1.0913,
      "num_input_tokens_seen": 44752944,
      "step": 77140
    },
    {
      "epoch": 11.490169794459339,
      "grad_norm": 13.8125,
      "learning_rate": 2.286448627029806e-05,
      "loss": 1.4635,
      "num_input_tokens_seen": 44756016,
      "step": 77145
    },
    {
      "epoch": 11.490914507000298,
      "grad_norm": 6.3125,
      "learning_rate": 2.2861248743573794e-05,
      "loss": 1.0008,
      "num_input_tokens_seen": 44758768,
      "step": 77150
    },
    {
      "epoch": 11.491659219541257,
      "grad_norm": 5.375,
      "learning_rate": 2.2858011252981566e-05,
      "loss": 0.9147,
      "num_input_tokens_seen": 44761552,
      "step": 77155
    },
    {
      "epoch": 11.492403932082217,
      "grad_norm": 7.15625,
      "learning_rate": 2.285477379857605e-05,
      "loss": 1.2213,
      "num_input_tokens_seen": 44764624,
      "step": 77160
    },
    {
      "epoch": 11.493148644623176,
      "grad_norm": 11.25,
      "learning_rate": 2.2851536380411958e-05,
      "loss": 0.8682,
      "num_input_tokens_seen": 44768112,
      "step": 77165
    },
    {
      "epoch": 11.493893357164135,
      "grad_norm": 14.125,
      "learning_rate": 2.284829899854398e-05,
      "loss": 1.0077,
      "num_input_tokens_seen": 44770800,
      "step": 77170
    },
    {
      "epoch": 11.494638069705093,
      "grad_norm": 18.5,
      "learning_rate": 2.284506165302679e-05,
      "loss": 0.7851,
      "num_input_tokens_seen": 44773808,
      "step": 77175
    },
    {
      "epoch": 11.495382782246054,
      "grad_norm": 12.375,
      "learning_rate": 2.2841824343915103e-05,
      "loss": 0.9671,
      "num_input_tokens_seen": 44776464,
      "step": 77180
    },
    {
      "epoch": 11.496127494787013,
      "grad_norm": 14.375,
      "learning_rate": 2.28385870712636e-05,
      "loss": 1.1207,
      "num_input_tokens_seen": 44779216,
      "step": 77185
    },
    {
      "epoch": 11.496872207327971,
      "grad_norm": 8.4375,
      "learning_rate": 2.283534983512697e-05,
      "loss": 0.9114,
      "num_input_tokens_seen": 44781968,
      "step": 77190
    },
    {
      "epoch": 11.49761691986893,
      "grad_norm": 9.375,
      "learning_rate": 2.2832112635559897e-05,
      "loss": 1.0556,
      "num_input_tokens_seen": 44784816,
      "step": 77195
    },
    {
      "epoch": 11.498361632409889,
      "grad_norm": 6.3125,
      "learning_rate": 2.282887547261709e-05,
      "loss": 1.1481,
      "num_input_tokens_seen": 44787952,
      "step": 77200
    },
    {
      "epoch": 11.49910634495085,
      "grad_norm": 13.1875,
      "learning_rate": 2.2825638346353223e-05,
      "loss": 1.2295,
      "num_input_tokens_seen": 44791184,
      "step": 77205
    },
    {
      "epoch": 11.499851057491808,
      "grad_norm": 3.5,
      "learning_rate": 2.2822401256822974e-05,
      "loss": 0.9147,
      "num_input_tokens_seen": 44794160,
      "step": 77210
    },
    {
      "epoch": 11.500595770032767,
      "grad_norm": 17.125,
      "learning_rate": 2.2819164204081057e-05,
      "loss": 0.8439,
      "num_input_tokens_seen": 44796784,
      "step": 77215
    },
    {
      "epoch": 11.501340482573726,
      "grad_norm": 10.125,
      "learning_rate": 2.281592718818214e-05,
      "loss": 1.0567,
      "num_input_tokens_seen": 44799600,
      "step": 77220
    },
    {
      "epoch": 11.502085195114686,
      "grad_norm": 5.84375,
      "learning_rate": 2.2812690209180914e-05,
      "loss": 1.0155,
      "num_input_tokens_seen": 44802256,
      "step": 77225
    },
    {
      "epoch": 11.502829907655645,
      "grad_norm": 6.1875,
      "learning_rate": 2.2809453267132054e-05,
      "loss": 1.4148,
      "num_input_tokens_seen": 44805136,
      "step": 77230
    },
    {
      "epoch": 11.503574620196604,
      "grad_norm": 9.3125,
      "learning_rate": 2.2806216362090267e-05,
      "loss": 0.8285,
      "num_input_tokens_seen": 44807952,
      "step": 77235
    },
    {
      "epoch": 11.504319332737563,
      "grad_norm": 15.1875,
      "learning_rate": 2.2802979494110213e-05,
      "loss": 1.2009,
      "num_input_tokens_seen": 44810992,
      "step": 77240
    },
    {
      "epoch": 11.505064045278523,
      "grad_norm": 12.75,
      "learning_rate": 2.27997426632466e-05,
      "loss": 1.0475,
      "num_input_tokens_seen": 44814064,
      "step": 77245
    },
    {
      "epoch": 11.505808757819482,
      "grad_norm": 9.3125,
      "learning_rate": 2.2796505869554098e-05,
      "loss": 0.6769,
      "num_input_tokens_seen": 44816976,
      "step": 77250
    },
    {
      "epoch": 11.50655347036044,
      "grad_norm": 14.125,
      "learning_rate": 2.2793269113087385e-05,
      "loss": 0.6049,
      "num_input_tokens_seen": 44819856,
      "step": 77255
    },
    {
      "epoch": 11.5072981829014,
      "grad_norm": 17.875,
      "learning_rate": 2.279003239390115e-05,
      "loss": 0.9851,
      "num_input_tokens_seen": 44822608,
      "step": 77260
    },
    {
      "epoch": 11.50804289544236,
      "grad_norm": 7.4375,
      "learning_rate": 2.2786795712050065e-05,
      "loss": 0.7411,
      "num_input_tokens_seen": 44825360,
      "step": 77265
    },
    {
      "epoch": 11.508787607983319,
      "grad_norm": 8.25,
      "learning_rate": 2.2783559067588822e-05,
      "loss": 1.2664,
      "num_input_tokens_seen": 44828080,
      "step": 77270
    },
    {
      "epoch": 11.509532320524277,
      "grad_norm": 20.25,
      "learning_rate": 2.278032246057209e-05,
      "loss": 0.8983,
      "num_input_tokens_seen": 44831120,
      "step": 77275
    },
    {
      "epoch": 11.510277033065236,
      "grad_norm": 9.375,
      "learning_rate": 2.2777085891054566e-05,
      "loss": 0.9589,
      "num_input_tokens_seen": 44834160,
      "step": 77280
    },
    {
      "epoch": 11.511021745606197,
      "grad_norm": 20.625,
      "learning_rate": 2.277384935909091e-05,
      "loss": 1.029,
      "num_input_tokens_seen": 44836912,
      "step": 77285
    },
    {
      "epoch": 11.511766458147155,
      "grad_norm": 10.5,
      "learning_rate": 2.277061286473581e-05,
      "loss": 0.9391,
      "num_input_tokens_seen": 44839856,
      "step": 77290
    },
    {
      "epoch": 11.512511170688114,
      "grad_norm": 14.8125,
      "learning_rate": 2.2767376408043935e-05,
      "loss": 0.9614,
      "num_input_tokens_seen": 44842768,
      "step": 77295
    },
    {
      "epoch": 11.513255883229073,
      "grad_norm": 8.0625,
      "learning_rate": 2.2764139989069962e-05,
      "loss": 1.172,
      "num_input_tokens_seen": 44845392,
      "step": 77300
    },
    {
      "epoch": 11.514000595770034,
      "grad_norm": 9.0,
      "learning_rate": 2.276090360786858e-05,
      "loss": 1.106,
      "num_input_tokens_seen": 44848432,
      "step": 77305
    },
    {
      "epoch": 11.514745308310992,
      "grad_norm": 17.875,
      "learning_rate": 2.2757667264494448e-05,
      "loss": 1.1383,
      "num_input_tokens_seen": 44851152,
      "step": 77310
    },
    {
      "epoch": 11.515490020851951,
      "grad_norm": 4.84375,
      "learning_rate": 2.275443095900226e-05,
      "loss": 0.8454,
      "num_input_tokens_seen": 44854224,
      "step": 77315
    },
    {
      "epoch": 11.51623473339291,
      "grad_norm": 12.875,
      "learning_rate": 2.2751194691446666e-05,
      "loss": 0.8816,
      "num_input_tokens_seen": 44857136,
      "step": 77320
    },
    {
      "epoch": 11.51697944593387,
      "grad_norm": 18.375,
      "learning_rate": 2.2747958461882365e-05,
      "loss": 1.0166,
      "num_input_tokens_seen": 44860048,
      "step": 77325
    },
    {
      "epoch": 11.517724158474829,
      "grad_norm": 6.71875,
      "learning_rate": 2.2744722270364012e-05,
      "loss": 1.0487,
      "num_input_tokens_seen": 44862960,
      "step": 77330
    },
    {
      "epoch": 11.518468871015788,
      "grad_norm": 8.0625,
      "learning_rate": 2.274148611694628e-05,
      "loss": 0.9942,
      "num_input_tokens_seen": 44865904,
      "step": 77335
    },
    {
      "epoch": 11.519213583556747,
      "grad_norm": 6.0,
      "learning_rate": 2.2738250001683846e-05,
      "loss": 0.9445,
      "num_input_tokens_seen": 44868816,
      "step": 77340
    },
    {
      "epoch": 11.519958296097705,
      "grad_norm": 9.0625,
      "learning_rate": 2.2735013924631378e-05,
      "loss": 1.1971,
      "num_input_tokens_seen": 44871984,
      "step": 77345
    },
    {
      "epoch": 11.520703008638666,
      "grad_norm": 13.4375,
      "learning_rate": 2.273177788584355e-05,
      "loss": 1.4652,
      "num_input_tokens_seen": 44875088,
      "step": 77350
    },
    {
      "epoch": 11.521447721179625,
      "grad_norm": 10.875,
      "learning_rate": 2.272854188537503e-05,
      "loss": 0.9803,
      "num_input_tokens_seen": 44877872,
      "step": 77355
    },
    {
      "epoch": 11.522192433720583,
      "grad_norm": 8.875,
      "learning_rate": 2.272530592328049e-05,
      "loss": 0.8916,
      "num_input_tokens_seen": 44880912,
      "step": 77360
    },
    {
      "epoch": 11.522937146261544,
      "grad_norm": 8.875,
      "learning_rate": 2.272206999961459e-05,
      "loss": 0.8701,
      "num_input_tokens_seen": 44883696,
      "step": 77365
    },
    {
      "epoch": 11.523681858802503,
      "grad_norm": 9.875,
      "learning_rate": 2.2718834114432e-05,
      "loss": 1.0908,
      "num_input_tokens_seen": 44886352,
      "step": 77370
    },
    {
      "epoch": 11.524426571343461,
      "grad_norm": 7.4375,
      "learning_rate": 2.2715598267787394e-05,
      "loss": 0.9981,
      "num_input_tokens_seen": 44888912,
      "step": 77375
    },
    {
      "epoch": 11.52517128388442,
      "grad_norm": 13.75,
      "learning_rate": 2.2712362459735425e-05,
      "loss": 1.3107,
      "num_input_tokens_seen": 44891632,
      "step": 77380
    },
    {
      "epoch": 11.525915996425379,
      "grad_norm": 14.8125,
      "learning_rate": 2.2709126690330778e-05,
      "loss": 1.0604,
      "num_input_tokens_seen": 44894608,
      "step": 77385
    },
    {
      "epoch": 11.52666070896634,
      "grad_norm": 13.375,
      "learning_rate": 2.27058909596281e-05,
      "loss": 1.1886,
      "num_input_tokens_seen": 44897744,
      "step": 77390
    },
    {
      "epoch": 11.527405421507298,
      "grad_norm": 8.8125,
      "learning_rate": 2.2702655267682068e-05,
      "loss": 1.0425,
      "num_input_tokens_seen": 44900656,
      "step": 77395
    },
    {
      "epoch": 11.528150134048257,
      "grad_norm": 7.875,
      "learning_rate": 2.2699419614547333e-05,
      "loss": 1.176,
      "num_input_tokens_seen": 44903600,
      "step": 77400
    },
    {
      "epoch": 11.528894846589216,
      "grad_norm": 10.8125,
      "learning_rate": 2.2696184000278573e-05,
      "loss": 1.0616,
      "num_input_tokens_seen": 44906384,
      "step": 77405
    },
    {
      "epoch": 11.529639559130176,
      "grad_norm": 17.875,
      "learning_rate": 2.2692948424930445e-05,
      "loss": 1.5337,
      "num_input_tokens_seen": 44909328,
      "step": 77410
    },
    {
      "epoch": 11.530384271671135,
      "grad_norm": 14.5625,
      "learning_rate": 2.2689712888557603e-05,
      "loss": 0.775,
      "num_input_tokens_seen": 44912624,
      "step": 77415
    },
    {
      "epoch": 11.531128984212094,
      "grad_norm": 4.8125,
      "learning_rate": 2.268647739121471e-05,
      "loss": 0.9871,
      "num_input_tokens_seen": 44915760,
      "step": 77420
    },
    {
      "epoch": 11.531873696753053,
      "grad_norm": 6.59375,
      "learning_rate": 2.2683241932956432e-05,
      "loss": 0.9054,
      "num_input_tokens_seen": 44918704,
      "step": 77425
    },
    {
      "epoch": 11.532618409294013,
      "grad_norm": 17.125,
      "learning_rate": 2.2680006513837436e-05,
      "loss": 0.907,
      "num_input_tokens_seen": 44921456,
      "step": 77430
    },
    {
      "epoch": 11.533363121834972,
      "grad_norm": 8.75,
      "learning_rate": 2.2676771133912355e-05,
      "loss": 1.1649,
      "num_input_tokens_seen": 44924592,
      "step": 77435
    },
    {
      "epoch": 11.53410783437593,
      "grad_norm": 10.0625,
      "learning_rate": 2.2673535793235877e-05,
      "loss": 0.8535,
      "num_input_tokens_seen": 44927344,
      "step": 77440
    },
    {
      "epoch": 11.53485254691689,
      "grad_norm": 12.6875,
      "learning_rate": 2.2670300491862646e-05,
      "loss": 1.1027,
      "num_input_tokens_seen": 44930640,
      "step": 77445
    },
    {
      "epoch": 11.53559725945785,
      "grad_norm": 13.125,
      "learning_rate": 2.2667065229847323e-05,
      "loss": 1.13,
      "num_input_tokens_seen": 44933328,
      "step": 77450
    },
    {
      "epoch": 11.536341971998809,
      "grad_norm": 8.0,
      "learning_rate": 2.266383000724456e-05,
      "loss": 1.04,
      "num_input_tokens_seen": 44936560,
      "step": 77455
    },
    {
      "epoch": 11.537086684539767,
      "grad_norm": 7.28125,
      "learning_rate": 2.2660594824109008e-05,
      "loss": 1.4097,
      "num_input_tokens_seen": 44939248,
      "step": 77460
    },
    {
      "epoch": 11.537831397080726,
      "grad_norm": 17.375,
      "learning_rate": 2.2657359680495335e-05,
      "loss": 1.173,
      "num_input_tokens_seen": 44941808,
      "step": 77465
    },
    {
      "epoch": 11.538576109621687,
      "grad_norm": 10.3125,
      "learning_rate": 2.2654124576458182e-05,
      "loss": 1.3272,
      "num_input_tokens_seen": 44944400,
      "step": 77470
    },
    {
      "epoch": 11.539320822162646,
      "grad_norm": 11.5,
      "learning_rate": 2.265088951205222e-05,
      "loss": 1.4986,
      "num_input_tokens_seen": 44947472,
      "step": 77475
    },
    {
      "epoch": 11.540065534703604,
      "grad_norm": 9.25,
      "learning_rate": 2.2647654487332086e-05,
      "loss": 1.1007,
      "num_input_tokens_seen": 44950544,
      "step": 77480
    },
    {
      "epoch": 11.540810247244563,
      "grad_norm": 7.3125,
      "learning_rate": 2.2644419502352444e-05,
      "loss": 0.7781,
      "num_input_tokens_seen": 44953200,
      "step": 77485
    },
    {
      "epoch": 11.541554959785524,
      "grad_norm": 8.1875,
      "learning_rate": 2.264118455716794e-05,
      "loss": 1.0067,
      "num_input_tokens_seen": 44956176,
      "step": 77490
    },
    {
      "epoch": 11.542299672326482,
      "grad_norm": 10.9375,
      "learning_rate": 2.2637949651833218e-05,
      "loss": 1.2328,
      "num_input_tokens_seen": 44959440,
      "step": 77495
    },
    {
      "epoch": 11.543044384867441,
      "grad_norm": 8.8125,
      "learning_rate": 2.2634714786402942e-05,
      "loss": 0.9184,
      "num_input_tokens_seen": 44962384,
      "step": 77500
    },
    {
      "epoch": 11.5437890974084,
      "grad_norm": 13.875,
      "learning_rate": 2.2631479960931747e-05,
      "loss": 1.2617,
      "num_input_tokens_seen": 44965360,
      "step": 77505
    },
    {
      "epoch": 11.54453380994936,
      "grad_norm": 6.1875,
      "learning_rate": 2.26282451754743e-05,
      "loss": 0.9315,
      "num_input_tokens_seen": 44968336,
      "step": 77510
    },
    {
      "epoch": 11.54527852249032,
      "grad_norm": 8.4375,
      "learning_rate": 2.262501043008524e-05,
      "loss": 1.1441,
      "num_input_tokens_seen": 44971216,
      "step": 77515
    },
    {
      "epoch": 11.546023235031278,
      "grad_norm": 15.875,
      "learning_rate": 2.2621775724819218e-05,
      "loss": 1.159,
      "num_input_tokens_seen": 44974128,
      "step": 77520
    },
    {
      "epoch": 11.546767947572237,
      "grad_norm": 6.96875,
      "learning_rate": 2.2618541059730862e-05,
      "loss": 0.8935,
      "num_input_tokens_seen": 44976784,
      "step": 77525
    },
    {
      "epoch": 11.547512660113195,
      "grad_norm": 6.625,
      "learning_rate": 2.2615306434874853e-05,
      "loss": 0.7975,
      "num_input_tokens_seen": 44980080,
      "step": 77530
    },
    {
      "epoch": 11.548257372654156,
      "grad_norm": 10.6875,
      "learning_rate": 2.2612071850305812e-05,
      "loss": 1.0109,
      "num_input_tokens_seen": 44983024,
      "step": 77535
    },
    {
      "epoch": 11.549002085195115,
      "grad_norm": 10.875,
      "learning_rate": 2.2608837306078385e-05,
      "loss": 1.1173,
      "num_input_tokens_seen": 44986096,
      "step": 77540
    },
    {
      "epoch": 11.549746797736073,
      "grad_norm": 8.625,
      "learning_rate": 2.2605602802247227e-05,
      "loss": 0.6983,
      "num_input_tokens_seen": 44989008,
      "step": 77545
    },
    {
      "epoch": 11.550491510277032,
      "grad_norm": 10.9375,
      "learning_rate": 2.2602368338866974e-05,
      "loss": 0.9729,
      "num_input_tokens_seen": 44991568,
      "step": 77550
    },
    {
      "epoch": 11.551236222817993,
      "grad_norm": 16.125,
      "learning_rate": 2.2599133915992273e-05,
      "loss": 1.3689,
      "num_input_tokens_seen": 44994576,
      "step": 77555
    },
    {
      "epoch": 11.551980935358952,
      "grad_norm": 17.25,
      "learning_rate": 2.2595899533677756e-05,
      "loss": 1.3518,
      "num_input_tokens_seen": 44997488,
      "step": 77560
    },
    {
      "epoch": 11.55272564789991,
      "grad_norm": 5.84375,
      "learning_rate": 2.2592665191978085e-05,
      "loss": 0.7974,
      "num_input_tokens_seen": 45000304,
      "step": 77565
    },
    {
      "epoch": 11.553470360440869,
      "grad_norm": 6.84375,
      "learning_rate": 2.2589430890947885e-05,
      "loss": 1.0307,
      "num_input_tokens_seen": 45002992,
      "step": 77570
    },
    {
      "epoch": 11.55421507298183,
      "grad_norm": 10.5,
      "learning_rate": 2.2586196630641792e-05,
      "loss": 0.9704,
      "num_input_tokens_seen": 45005872,
      "step": 77575
    },
    {
      "epoch": 11.554959785522788,
      "grad_norm": 13.875,
      "learning_rate": 2.2582962411114464e-05,
      "loss": 1.1633,
      "num_input_tokens_seen": 45008464,
      "step": 77580
    },
    {
      "epoch": 11.555704498063747,
      "grad_norm": 15.625,
      "learning_rate": 2.2579728232420525e-05,
      "loss": 1.2282,
      "num_input_tokens_seen": 45011408,
      "step": 77585
    },
    {
      "epoch": 11.556449210604706,
      "grad_norm": 13.25,
      "learning_rate": 2.2576494094614624e-05,
      "loss": 1.392,
      "num_input_tokens_seen": 45014576,
      "step": 77590
    },
    {
      "epoch": 11.557193923145666,
      "grad_norm": 8.6875,
      "learning_rate": 2.257325999775138e-05,
      "loss": 0.9528,
      "num_input_tokens_seen": 45017264,
      "step": 77595
    },
    {
      "epoch": 11.557938635686625,
      "grad_norm": 10.625,
      "learning_rate": 2.257002594188545e-05,
      "loss": 1.0679,
      "num_input_tokens_seen": 45020176,
      "step": 77600
    },
    {
      "epoch": 11.558683348227584,
      "grad_norm": 5.6875,
      "learning_rate": 2.2566791927071453e-05,
      "loss": 0.8994,
      "num_input_tokens_seen": 45023088,
      "step": 77605
    },
    {
      "epoch": 11.559428060768543,
      "grad_norm": 7.4375,
      "learning_rate": 2.2563557953364043e-05,
      "loss": 0.884,
      "num_input_tokens_seen": 45026160,
      "step": 77610
    },
    {
      "epoch": 11.560172773309503,
      "grad_norm": 9.6875,
      "learning_rate": 2.256032402081785e-05,
      "loss": 0.9041,
      "num_input_tokens_seen": 45028912,
      "step": 77615
    },
    {
      "epoch": 11.560917485850462,
      "grad_norm": 12.875,
      "learning_rate": 2.2557090129487493e-05,
      "loss": 1.3164,
      "num_input_tokens_seen": 45031600,
      "step": 77620
    },
    {
      "epoch": 11.56166219839142,
      "grad_norm": 9.6875,
      "learning_rate": 2.2553856279427625e-05,
      "loss": 0.9291,
      "num_input_tokens_seen": 45034992,
      "step": 77625
    },
    {
      "epoch": 11.56240691093238,
      "grad_norm": 9.5625,
      "learning_rate": 2.2550622470692852e-05,
      "loss": 1.3379,
      "num_input_tokens_seen": 45038224,
      "step": 77630
    },
    {
      "epoch": 11.56315162347334,
      "grad_norm": 12.375,
      "learning_rate": 2.2547388703337837e-05,
      "loss": 1.1262,
      "num_input_tokens_seen": 45041264,
      "step": 77635
    },
    {
      "epoch": 11.563896336014299,
      "grad_norm": 9.3125,
      "learning_rate": 2.2544154977417187e-05,
      "loss": 1.1005,
      "num_input_tokens_seen": 45044048,
      "step": 77640
    },
    {
      "epoch": 11.564641048555258,
      "grad_norm": 7.0,
      "learning_rate": 2.2540921292985553e-05,
      "loss": 0.872,
      "num_input_tokens_seen": 45047024,
      "step": 77645
    },
    {
      "epoch": 11.565385761096216,
      "grad_norm": 11.875,
      "learning_rate": 2.2537687650097554e-05,
      "loss": 1.3826,
      "num_input_tokens_seen": 45050064,
      "step": 77650
    },
    {
      "epoch": 11.566130473637177,
      "grad_norm": 6.65625,
      "learning_rate": 2.2534454048807814e-05,
      "loss": 0.8465,
      "num_input_tokens_seen": 45052976,
      "step": 77655
    },
    {
      "epoch": 11.566875186178136,
      "grad_norm": 7.65625,
      "learning_rate": 2.2531220489170977e-05,
      "loss": 1.0249,
      "num_input_tokens_seen": 45055504,
      "step": 77660
    },
    {
      "epoch": 11.567619898719094,
      "grad_norm": 9.5625,
      "learning_rate": 2.2527986971241642e-05,
      "loss": 0.9109,
      "num_input_tokens_seen": 45058288,
      "step": 77665
    },
    {
      "epoch": 11.568364611260053,
      "grad_norm": 14.4375,
      "learning_rate": 2.252475349507447e-05,
      "loss": 1.0891,
      "num_input_tokens_seen": 45061008,
      "step": 77670
    },
    {
      "epoch": 11.569109323801012,
      "grad_norm": 11.375,
      "learning_rate": 2.2521520060724062e-05,
      "loss": 0.8048,
      "num_input_tokens_seen": 45063792,
      "step": 77675
    },
    {
      "epoch": 11.569854036341972,
      "grad_norm": 15.9375,
      "learning_rate": 2.251828666824506e-05,
      "loss": 1.1096,
      "num_input_tokens_seen": 45066736,
      "step": 77680
    },
    {
      "epoch": 11.570598748882931,
      "grad_norm": 9.5625,
      "learning_rate": 2.2515053317692082e-05,
      "loss": 0.9977,
      "num_input_tokens_seen": 45069840,
      "step": 77685
    },
    {
      "epoch": 11.57134346142389,
      "grad_norm": 6.3125,
      "learning_rate": 2.2511820009119755e-05,
      "loss": 1.0693,
      "num_input_tokens_seen": 45072688,
      "step": 77690
    },
    {
      "epoch": 11.57208817396485,
      "grad_norm": 20.5,
      "learning_rate": 2.25085867425827e-05,
      "loss": 1.4173,
      "num_input_tokens_seen": 45075440,
      "step": 77695
    },
    {
      "epoch": 11.57283288650581,
      "grad_norm": 13.9375,
      "learning_rate": 2.2505353518135534e-05,
      "loss": 1.2394,
      "num_input_tokens_seen": 45078192,
      "step": 77700
    },
    {
      "epoch": 11.573577599046768,
      "grad_norm": 11.3125,
      "learning_rate": 2.250212033583289e-05,
      "loss": 1.1722,
      "num_input_tokens_seen": 45080720,
      "step": 77705
    },
    {
      "epoch": 11.574322311587727,
      "grad_norm": 8.125,
      "learning_rate": 2.2498887195729375e-05,
      "loss": 1.1619,
      "num_input_tokens_seen": 45083568,
      "step": 77710
    },
    {
      "epoch": 11.575067024128685,
      "grad_norm": 8.6875,
      "learning_rate": 2.2495654097879627e-05,
      "loss": 1.2288,
      "num_input_tokens_seen": 45086480,
      "step": 77715
    },
    {
      "epoch": 11.575811736669646,
      "grad_norm": 9.6875,
      "learning_rate": 2.2492421042338257e-05,
      "loss": 1.1801,
      "num_input_tokens_seen": 45089392,
      "step": 77720
    },
    {
      "epoch": 11.576556449210605,
      "grad_norm": 7.5,
      "learning_rate": 2.2489188029159887e-05,
      "loss": 0.8589,
      "num_input_tokens_seen": 45092272,
      "step": 77725
    },
    {
      "epoch": 11.577301161751564,
      "grad_norm": 9.0625,
      "learning_rate": 2.2485955058399135e-05,
      "loss": 1.087,
      "num_input_tokens_seen": 45095248,
      "step": 77730
    },
    {
      "epoch": 11.578045874292522,
      "grad_norm": 17.625,
      "learning_rate": 2.2482722130110608e-05,
      "loss": 1.2917,
      "num_input_tokens_seen": 45098064,
      "step": 77735
    },
    {
      "epoch": 11.578790586833483,
      "grad_norm": 14.375,
      "learning_rate": 2.2479489244348938e-05,
      "loss": 0.8888,
      "num_input_tokens_seen": 45101104,
      "step": 77740
    },
    {
      "epoch": 11.579535299374442,
      "grad_norm": 7.15625,
      "learning_rate": 2.2476256401168736e-05,
      "loss": 1.0755,
      "num_input_tokens_seen": 45104272,
      "step": 77745
    },
    {
      "epoch": 11.5802800119154,
      "grad_norm": 7.6875,
      "learning_rate": 2.247302360062461e-05,
      "loss": 0.7296,
      "num_input_tokens_seen": 45107056,
      "step": 77750
    },
    {
      "epoch": 11.581024724456359,
      "grad_norm": 12.8125,
      "learning_rate": 2.246979084277119e-05,
      "loss": 0.6608,
      "num_input_tokens_seen": 45109872,
      "step": 77755
    },
    {
      "epoch": 11.58176943699732,
      "grad_norm": 11.3125,
      "learning_rate": 2.2466558127663086e-05,
      "loss": 1.1088,
      "num_input_tokens_seen": 45112752,
      "step": 77760
    },
    {
      "epoch": 11.582514149538278,
      "grad_norm": 11.8125,
      "learning_rate": 2.246332545535489e-05,
      "loss": 1.0411,
      "num_input_tokens_seen": 45115440,
      "step": 77765
    },
    {
      "epoch": 11.583258862079237,
      "grad_norm": 5.125,
      "learning_rate": 2.246009282590125e-05,
      "loss": 1.017,
      "num_input_tokens_seen": 45118288,
      "step": 77770
    },
    {
      "epoch": 11.584003574620196,
      "grad_norm": 10.3125,
      "learning_rate": 2.2456860239356755e-05,
      "loss": 1.0492,
      "num_input_tokens_seen": 45121008,
      "step": 77775
    },
    {
      "epoch": 11.584748287161156,
      "grad_norm": 13.0,
      "learning_rate": 2.245362769577602e-05,
      "loss": 1.1837,
      "num_input_tokens_seen": 45123760,
      "step": 77780
    },
    {
      "epoch": 11.585492999702115,
      "grad_norm": 6.5,
      "learning_rate": 2.245039519521366e-05,
      "loss": 0.9786,
      "num_input_tokens_seen": 45126800,
      "step": 77785
    },
    {
      "epoch": 11.586237712243074,
      "grad_norm": 13.0625,
      "learning_rate": 2.2447162737724274e-05,
      "loss": 1.1734,
      "num_input_tokens_seen": 45129712,
      "step": 77790
    },
    {
      "epoch": 11.586982424784033,
      "grad_norm": 9.9375,
      "learning_rate": 2.2443930323362487e-05,
      "loss": 0.8116,
      "num_input_tokens_seen": 45132720,
      "step": 77795
    },
    {
      "epoch": 11.587727137324993,
      "grad_norm": 5.9375,
      "learning_rate": 2.244069795218289e-05,
      "loss": 1.008,
      "num_input_tokens_seen": 45135408,
      "step": 77800
    },
    {
      "epoch": 11.588471849865952,
      "grad_norm": 17.5,
      "learning_rate": 2.243746562424011e-05,
      "loss": 1.2449,
      "num_input_tokens_seen": 45138000,
      "step": 77805
    },
    {
      "epoch": 11.58921656240691,
      "grad_norm": 16.0,
      "learning_rate": 2.2434233339588746e-05,
      "loss": 1.0145,
      "num_input_tokens_seen": 45141040,
      "step": 77810
    },
    {
      "epoch": 11.58996127494787,
      "grad_norm": 16.125,
      "learning_rate": 2.2431001098283393e-05,
      "loss": 1.2082,
      "num_input_tokens_seen": 45143792,
      "step": 77815
    },
    {
      "epoch": 11.59070598748883,
      "grad_norm": 11.8125,
      "learning_rate": 2.2427768900378674e-05,
      "loss": 1.1148,
      "num_input_tokens_seen": 45146672,
      "step": 77820
    },
    {
      "epoch": 11.591450700029789,
      "grad_norm": 5.3125,
      "learning_rate": 2.2424536745929174e-05,
      "loss": 0.9956,
      "num_input_tokens_seen": 45149328,
      "step": 77825
    },
    {
      "epoch": 11.592195412570748,
      "grad_norm": 9.4375,
      "learning_rate": 2.2421304634989517e-05,
      "loss": 1.1357,
      "num_input_tokens_seen": 45152144,
      "step": 77830
    },
    {
      "epoch": 11.592940125111706,
      "grad_norm": 9.0,
      "learning_rate": 2.2418072567614286e-05,
      "loss": 1.2568,
      "num_input_tokens_seen": 45154928,
      "step": 77835
    },
    {
      "epoch": 11.593684837652667,
      "grad_norm": 7.625,
      "learning_rate": 2.241484054385811e-05,
      "loss": 1.0609,
      "num_input_tokens_seen": 45157744,
      "step": 77840
    },
    {
      "epoch": 11.594429550193626,
      "grad_norm": 8.0625,
      "learning_rate": 2.2411608563775564e-05,
      "loss": 1.5314,
      "num_input_tokens_seen": 45160848,
      "step": 77845
    },
    {
      "epoch": 11.595174262734584,
      "grad_norm": 6.4375,
      "learning_rate": 2.240837662742127e-05,
      "loss": 1.1306,
      "num_input_tokens_seen": 45163728,
      "step": 77850
    },
    {
      "epoch": 11.595918975275543,
      "grad_norm": 8.4375,
      "learning_rate": 2.240514473484982e-05,
      "loss": 1.2329,
      "num_input_tokens_seen": 45166672,
      "step": 77855
    },
    {
      "epoch": 11.596663687816502,
      "grad_norm": 7.9375,
      "learning_rate": 2.24019128861158e-05,
      "loss": 1.0189,
      "num_input_tokens_seen": 45169744,
      "step": 77860
    },
    {
      "epoch": 11.597408400357462,
      "grad_norm": 8.3125,
      "learning_rate": 2.2398681081273832e-05,
      "loss": 1.0037,
      "num_input_tokens_seen": 45172368,
      "step": 77865
    },
    {
      "epoch": 11.598153112898421,
      "grad_norm": 13.125,
      "learning_rate": 2.239544932037849e-05,
      "loss": 0.9226,
      "num_input_tokens_seen": 45175472,
      "step": 77870
    },
    {
      "epoch": 11.59889782543938,
      "grad_norm": 7.4375,
      "learning_rate": 2.2392217603484397e-05,
      "loss": 0.9217,
      "num_input_tokens_seen": 45178480,
      "step": 77875
    },
    {
      "epoch": 11.59964253798034,
      "grad_norm": 23.125,
      "learning_rate": 2.2388985930646135e-05,
      "loss": 1.0937,
      "num_input_tokens_seen": 45181328,
      "step": 77880
    },
    {
      "epoch": 11.6003872505213,
      "grad_norm": 5.8125,
      "learning_rate": 2.2385754301918303e-05,
      "loss": 1.1751,
      "num_input_tokens_seen": 45184496,
      "step": 77885
    },
    {
      "epoch": 11.601131963062258,
      "grad_norm": 6.90625,
      "learning_rate": 2.2382522717355498e-05,
      "loss": 1.4843,
      "num_input_tokens_seen": 45187248,
      "step": 77890
    },
    {
      "epoch": 11.601876675603217,
      "grad_norm": 11.0,
      "learning_rate": 2.2379291177012295e-05,
      "loss": 0.9633,
      "num_input_tokens_seen": 45190128,
      "step": 77895
    },
    {
      "epoch": 11.602621388144176,
      "grad_norm": 11.625,
      "learning_rate": 2.2376059680943324e-05,
      "loss": 1.2331,
      "num_input_tokens_seen": 45193104,
      "step": 77900
    },
    {
      "epoch": 11.603366100685136,
      "grad_norm": 13.375,
      "learning_rate": 2.237282822920314e-05,
      "loss": 1.2782,
      "num_input_tokens_seen": 45196112,
      "step": 77905
    },
    {
      "epoch": 11.604110813226095,
      "grad_norm": 9.6875,
      "learning_rate": 2.2369596821846367e-05,
      "loss": 1.0651,
      "num_input_tokens_seen": 45198992,
      "step": 77910
    },
    {
      "epoch": 11.604855525767054,
      "grad_norm": 9.625,
      "learning_rate": 2.2366365458927574e-05,
      "loss": 0.8443,
      "num_input_tokens_seen": 45202352,
      "step": 77915
    },
    {
      "epoch": 11.605600238308012,
      "grad_norm": 8.875,
      "learning_rate": 2.236313414050137e-05,
      "loss": 1.3741,
      "num_input_tokens_seen": 45205424,
      "step": 77920
    },
    {
      "epoch": 11.606344950848973,
      "grad_norm": 22.25,
      "learning_rate": 2.2359902866622317e-05,
      "loss": 1.2819,
      "num_input_tokens_seen": 45208240,
      "step": 77925
    },
    {
      "epoch": 11.607089663389932,
      "grad_norm": 3.59375,
      "learning_rate": 2.2356671637345038e-05,
      "loss": 1.0494,
      "num_input_tokens_seen": 45211088,
      "step": 77930
    },
    {
      "epoch": 11.60783437593089,
      "grad_norm": 7.65625,
      "learning_rate": 2.2353440452724102e-05,
      "loss": 1.1359,
      "num_input_tokens_seen": 45214224,
      "step": 77935
    },
    {
      "epoch": 11.60857908847185,
      "grad_norm": 8.9375,
      "learning_rate": 2.235020931281409e-05,
      "loss": 1.1996,
      "num_input_tokens_seen": 45217296,
      "step": 77940
    },
    {
      "epoch": 11.60932380101281,
      "grad_norm": 6.34375,
      "learning_rate": 2.2346978217669613e-05,
      "loss": 0.8901,
      "num_input_tokens_seen": 45220304,
      "step": 77945
    },
    {
      "epoch": 11.610068513553768,
      "grad_norm": 18.75,
      "learning_rate": 2.2343747167345233e-05,
      "loss": 0.9916,
      "num_input_tokens_seen": 45223184,
      "step": 77950
    },
    {
      "epoch": 11.610813226094727,
      "grad_norm": 7.4375,
      "learning_rate": 2.2340516161895553e-05,
      "loss": 1.2192,
      "num_input_tokens_seen": 45226480,
      "step": 77955
    },
    {
      "epoch": 11.611557938635686,
      "grad_norm": 5.59375,
      "learning_rate": 2.2337285201375137e-05,
      "loss": 0.8317,
      "num_input_tokens_seen": 45229360,
      "step": 77960
    },
    {
      "epoch": 11.612302651176647,
      "grad_norm": 18.0,
      "learning_rate": 2.233405428583859e-05,
      "loss": 0.9742,
      "num_input_tokens_seen": 45231920,
      "step": 77965
    },
    {
      "epoch": 11.613047363717605,
      "grad_norm": 11.6875,
      "learning_rate": 2.233082341534049e-05,
      "loss": 1.2533,
      "num_input_tokens_seen": 45234768,
      "step": 77970
    },
    {
      "epoch": 11.613792076258564,
      "grad_norm": 21.625,
      "learning_rate": 2.2327592589935403e-05,
      "loss": 1.2985,
      "num_input_tokens_seen": 45237712,
      "step": 77975
    },
    {
      "epoch": 11.614536788799523,
      "grad_norm": 8.4375,
      "learning_rate": 2.2324361809677933e-05,
      "loss": 0.9599,
      "num_input_tokens_seen": 45240560,
      "step": 77980
    },
    {
      "epoch": 11.615281501340483,
      "grad_norm": 5.375,
      "learning_rate": 2.2321131074622647e-05,
      "loss": 1.0905,
      "num_input_tokens_seen": 45243536,
      "step": 77985
    },
    {
      "epoch": 11.616026213881442,
      "grad_norm": 8.875,
      "learning_rate": 2.2317900384824132e-05,
      "loss": 1.1971,
      "num_input_tokens_seen": 45246224,
      "step": 77990
    },
    {
      "epoch": 11.6167709264224,
      "grad_norm": 5.6875,
      "learning_rate": 2.2314669740336957e-05,
      "loss": 1.3043,
      "num_input_tokens_seen": 45249360,
      "step": 77995
    },
    {
      "epoch": 11.61751563896336,
      "grad_norm": 5.3125,
      "learning_rate": 2.2311439141215715e-05,
      "loss": 1.0453,
      "num_input_tokens_seen": 45252368,
      "step": 78000
    },
    {
      "epoch": 11.61826035150432,
      "grad_norm": 16.75,
      "learning_rate": 2.2308208587514967e-05,
      "loss": 0.922,
      "num_input_tokens_seen": 45255248,
      "step": 78005
    },
    {
      "epoch": 11.619005064045279,
      "grad_norm": 12.125,
      "learning_rate": 2.230497807928931e-05,
      "loss": 1.1571,
      "num_input_tokens_seen": 45258160,
      "step": 78010
    },
    {
      "epoch": 11.619749776586238,
      "grad_norm": 13.1875,
      "learning_rate": 2.2301747616593306e-05,
      "loss": 1.1033,
      "num_input_tokens_seen": 45261008,
      "step": 78015
    },
    {
      "epoch": 11.620494489127196,
      "grad_norm": 14.8125,
      "learning_rate": 2.2298517199481534e-05,
      "loss": 0.9914,
      "num_input_tokens_seen": 45264144,
      "step": 78020
    },
    {
      "epoch": 11.621239201668157,
      "grad_norm": 9.4375,
      "learning_rate": 2.2295286828008572e-05,
      "loss": 0.8754,
      "num_input_tokens_seen": 45266864,
      "step": 78025
    },
    {
      "epoch": 11.621983914209116,
      "grad_norm": 12.375,
      "learning_rate": 2.2292056502228975e-05,
      "loss": 1.3274,
      "num_input_tokens_seen": 45269680,
      "step": 78030
    },
    {
      "epoch": 11.622728626750074,
      "grad_norm": 10.875,
      "learning_rate": 2.2288826222197346e-05,
      "loss": 1.3029,
      "num_input_tokens_seen": 45272944,
      "step": 78035
    },
    {
      "epoch": 11.623473339291033,
      "grad_norm": 13.5,
      "learning_rate": 2.228559598796823e-05,
      "loss": 1.0766,
      "num_input_tokens_seen": 45276208,
      "step": 78040
    },
    {
      "epoch": 11.624218051831992,
      "grad_norm": 8.0625,
      "learning_rate": 2.2282365799596222e-05,
      "loss": 1.3475,
      "num_input_tokens_seen": 45279120,
      "step": 78045
    },
    {
      "epoch": 11.624962764372953,
      "grad_norm": 13.125,
      "learning_rate": 2.2279135657135876e-05,
      "loss": 1.3653,
      "num_input_tokens_seen": 45281968,
      "step": 78050
    },
    {
      "epoch": 11.625707476913911,
      "grad_norm": 12.9375,
      "learning_rate": 2.2275905560641775e-05,
      "loss": 1.0955,
      "num_input_tokens_seen": 45285072,
      "step": 78055
    },
    {
      "epoch": 11.62645218945487,
      "grad_norm": 5.65625,
      "learning_rate": 2.2272675510168482e-05,
      "loss": 1.0789,
      "num_input_tokens_seen": 45288304,
      "step": 78060
    },
    {
      "epoch": 11.627196901995829,
      "grad_norm": 10.4375,
      "learning_rate": 2.226944550577055e-05,
      "loss": 0.9331,
      "num_input_tokens_seen": 45291120,
      "step": 78065
    },
    {
      "epoch": 11.62794161453679,
      "grad_norm": 14.3125,
      "learning_rate": 2.2266215547502573e-05,
      "loss": 0.8759,
      "num_input_tokens_seen": 45293776,
      "step": 78070
    },
    {
      "epoch": 11.628686327077748,
      "grad_norm": 6.3125,
      "learning_rate": 2.22629856354191e-05,
      "loss": 0.8627,
      "num_input_tokens_seen": 45296624,
      "step": 78075
    },
    {
      "epoch": 11.629431039618707,
      "grad_norm": 10.3125,
      "learning_rate": 2.22597557695747e-05,
      "loss": 1.3464,
      "num_input_tokens_seen": 45299376,
      "step": 78080
    },
    {
      "epoch": 11.630175752159666,
      "grad_norm": 15.8125,
      "learning_rate": 2.225652595002395e-05,
      "loss": 1.2627,
      "num_input_tokens_seen": 45302416,
      "step": 78085
    },
    {
      "epoch": 11.630920464700626,
      "grad_norm": 15.4375,
      "learning_rate": 2.2253296176821402e-05,
      "loss": 1.4152,
      "num_input_tokens_seen": 45305392,
      "step": 78090
    },
    {
      "epoch": 11.631665177241585,
      "grad_norm": 5.90625,
      "learning_rate": 2.2250066450021628e-05,
      "loss": 1.0986,
      "num_input_tokens_seen": 45308432,
      "step": 78095
    },
    {
      "epoch": 11.632409889782544,
      "grad_norm": 9.5,
      "learning_rate": 2.2246836769679175e-05,
      "loss": 1.3224,
      "num_input_tokens_seen": 45311408,
      "step": 78100
    },
    {
      "epoch": 11.633154602323502,
      "grad_norm": 13.625,
      "learning_rate": 2.2243607135848625e-05,
      "loss": 1.1968,
      "num_input_tokens_seen": 45314192,
      "step": 78105
    },
    {
      "epoch": 11.633899314864463,
      "grad_norm": 8.625,
      "learning_rate": 2.2240377548584532e-05,
      "loss": 0.928,
      "num_input_tokens_seen": 45317168,
      "step": 78110
    },
    {
      "epoch": 11.634644027405422,
      "grad_norm": 6.59375,
      "learning_rate": 2.2237148007941455e-05,
      "loss": 0.8838,
      "num_input_tokens_seen": 45319856,
      "step": 78115
    },
    {
      "epoch": 11.63538873994638,
      "grad_norm": 10.625,
      "learning_rate": 2.2233918513973944e-05,
      "loss": 1.2475,
      "num_input_tokens_seen": 45322576,
      "step": 78120
    },
    {
      "epoch": 11.63613345248734,
      "grad_norm": 13.75,
      "learning_rate": 2.223068906673658e-05,
      "loss": 1.01,
      "num_input_tokens_seen": 45325264,
      "step": 78125
    },
    {
      "epoch": 11.6368781650283,
      "grad_norm": 15.875,
      "learning_rate": 2.22274596662839e-05,
      "loss": 1.2709,
      "num_input_tokens_seen": 45328048,
      "step": 78130
    },
    {
      "epoch": 11.637622877569259,
      "grad_norm": 7.875,
      "learning_rate": 2.222423031267048e-05,
      "loss": 0.8557,
      "num_input_tokens_seen": 45331344,
      "step": 78135
    },
    {
      "epoch": 11.638367590110217,
      "grad_norm": 10.375,
      "learning_rate": 2.222100100595087e-05,
      "loss": 1.3322,
      "num_input_tokens_seen": 45334256,
      "step": 78140
    },
    {
      "epoch": 11.639112302651176,
      "grad_norm": 10.3125,
      "learning_rate": 2.221777174617962e-05,
      "loss": 1.2009,
      "num_input_tokens_seen": 45337072,
      "step": 78145
    },
    {
      "epoch": 11.639857015192137,
      "grad_norm": 7.875,
      "learning_rate": 2.221454253341129e-05,
      "loss": 0.987,
      "num_input_tokens_seen": 45339696,
      "step": 78150
    },
    {
      "epoch": 11.640601727733095,
      "grad_norm": 15.4375,
      "learning_rate": 2.2211313367700422e-05,
      "loss": 1.2113,
      "num_input_tokens_seen": 45342544,
      "step": 78155
    },
    {
      "epoch": 11.641346440274054,
      "grad_norm": 7.4375,
      "learning_rate": 2.2208084249101593e-05,
      "loss": 1.0687,
      "num_input_tokens_seen": 45345360,
      "step": 78160
    },
    {
      "epoch": 11.642091152815013,
      "grad_norm": 12.25,
      "learning_rate": 2.220485517766933e-05,
      "loss": 1.4665,
      "num_input_tokens_seen": 45348016,
      "step": 78165
    },
    {
      "epoch": 11.642835865355973,
      "grad_norm": 13.9375,
      "learning_rate": 2.220162615345821e-05,
      "loss": 0.9345,
      "num_input_tokens_seen": 45350640,
      "step": 78170
    },
    {
      "epoch": 11.643580577896932,
      "grad_norm": 11.4375,
      "learning_rate": 2.2198397176522773e-05,
      "loss": 0.9538,
      "num_input_tokens_seen": 45353232,
      "step": 78175
    },
    {
      "epoch": 11.64432529043789,
      "grad_norm": 5.9375,
      "learning_rate": 2.2195168246917564e-05,
      "loss": 1.0142,
      "num_input_tokens_seen": 45356304,
      "step": 78180
    },
    {
      "epoch": 11.64507000297885,
      "grad_norm": 8.5,
      "learning_rate": 2.219193936469714e-05,
      "loss": 1.0466,
      "num_input_tokens_seen": 45359120,
      "step": 78185
    },
    {
      "epoch": 11.64581471551981,
      "grad_norm": 13.1875,
      "learning_rate": 2.2188710529916033e-05,
      "loss": 1.1499,
      "num_input_tokens_seen": 45361968,
      "step": 78190
    },
    {
      "epoch": 11.646559428060769,
      "grad_norm": 9.25,
      "learning_rate": 2.218548174262882e-05,
      "loss": 0.9503,
      "num_input_tokens_seen": 45364880,
      "step": 78195
    },
    {
      "epoch": 11.647304140601728,
      "grad_norm": 8.0,
      "learning_rate": 2.218225300289002e-05,
      "loss": 0.8432,
      "num_input_tokens_seen": 45367760,
      "step": 78200
    },
    {
      "epoch": 11.648048853142686,
      "grad_norm": 7.6875,
      "learning_rate": 2.21790243107542e-05,
      "loss": 1.3487,
      "num_input_tokens_seen": 45370672,
      "step": 78205
    },
    {
      "epoch": 11.648793565683647,
      "grad_norm": 8.125,
      "learning_rate": 2.2175795666275894e-05,
      "loss": 1.1059,
      "num_input_tokens_seen": 45373584,
      "step": 78210
    },
    {
      "epoch": 11.649538278224606,
      "grad_norm": 17.25,
      "learning_rate": 2.2172567069509656e-05,
      "loss": 0.9863,
      "num_input_tokens_seen": 45376400,
      "step": 78215
    },
    {
      "epoch": 11.650282990765565,
      "grad_norm": 4.625,
      "learning_rate": 2.2169338520510025e-05,
      "loss": 0.7575,
      "num_input_tokens_seen": 45379504,
      "step": 78220
    },
    {
      "epoch": 11.651027703306523,
      "grad_norm": 15.8125,
      "learning_rate": 2.2166110019331526e-05,
      "loss": 0.9808,
      "num_input_tokens_seen": 45382544,
      "step": 78225
    },
    {
      "epoch": 11.651772415847482,
      "grad_norm": 10.125,
      "learning_rate": 2.2162881566028736e-05,
      "loss": 1.0717,
      "num_input_tokens_seen": 45385392,
      "step": 78230
    },
    {
      "epoch": 11.652517128388443,
      "grad_norm": 21.75,
      "learning_rate": 2.2159653160656162e-05,
      "loss": 0.6823,
      "num_input_tokens_seen": 45388304,
      "step": 78235
    },
    {
      "epoch": 11.653261840929401,
      "grad_norm": 6.875,
      "learning_rate": 2.2156424803268374e-05,
      "loss": 1.08,
      "num_input_tokens_seen": 45391312,
      "step": 78240
    },
    {
      "epoch": 11.65400655347036,
      "grad_norm": 14.625,
      "learning_rate": 2.2153196493919896e-05,
      "loss": 1.2471,
      "num_input_tokens_seen": 45394224,
      "step": 78245
    },
    {
      "epoch": 11.654751266011319,
      "grad_norm": 17.375,
      "learning_rate": 2.214996823266527e-05,
      "loss": 1.1664,
      "num_input_tokens_seen": 45397328,
      "step": 78250
    },
    {
      "epoch": 11.65549597855228,
      "grad_norm": 7.84375,
      "learning_rate": 2.2146740019559036e-05,
      "loss": 1.0486,
      "num_input_tokens_seen": 45400336,
      "step": 78255
    },
    {
      "epoch": 11.656240691093238,
      "grad_norm": 15.875,
      "learning_rate": 2.214351185465572e-05,
      "loss": 1.1493,
      "num_input_tokens_seen": 45403248,
      "step": 78260
    },
    {
      "epoch": 11.656985403634197,
      "grad_norm": 8.1875,
      "learning_rate": 2.214028373800988e-05,
      "loss": 0.9071,
      "num_input_tokens_seen": 45405936,
      "step": 78265
    },
    {
      "epoch": 11.657730116175156,
      "grad_norm": 13.125,
      "learning_rate": 2.2137055669676027e-05,
      "loss": 0.9151,
      "num_input_tokens_seen": 45408688,
      "step": 78270
    },
    {
      "epoch": 11.658474828716116,
      "grad_norm": 13.125,
      "learning_rate": 2.213382764970872e-05,
      "loss": 1.3359,
      "num_input_tokens_seen": 45411472,
      "step": 78275
    },
    {
      "epoch": 11.659219541257075,
      "grad_norm": 10.25,
      "learning_rate": 2.2130599678162474e-05,
      "loss": 0.9305,
      "num_input_tokens_seen": 45414256,
      "step": 78280
    },
    {
      "epoch": 11.659964253798034,
      "grad_norm": 10.6875,
      "learning_rate": 2.212737175509184e-05,
      "loss": 0.805,
      "num_input_tokens_seen": 45417072,
      "step": 78285
    },
    {
      "epoch": 11.660708966338992,
      "grad_norm": 6.71875,
      "learning_rate": 2.2124143880551327e-05,
      "loss": 0.951,
      "num_input_tokens_seen": 45419952,
      "step": 78290
    },
    {
      "epoch": 11.661453678879953,
      "grad_norm": 7.9375,
      "learning_rate": 2.2120916054595492e-05,
      "loss": 1.1848,
      "num_input_tokens_seen": 45422864,
      "step": 78295
    },
    {
      "epoch": 11.662198391420912,
      "grad_norm": 7.6875,
      "learning_rate": 2.211768827727885e-05,
      "loss": 1.3343,
      "num_input_tokens_seen": 45425808,
      "step": 78300
    },
    {
      "epoch": 11.66294310396187,
      "grad_norm": 9.0625,
      "learning_rate": 2.211446054865593e-05,
      "loss": 0.9263,
      "num_input_tokens_seen": 45428720,
      "step": 78305
    },
    {
      "epoch": 11.66368781650283,
      "grad_norm": 5.875,
      "learning_rate": 2.2111232868781277e-05,
      "loss": 0.9611,
      "num_input_tokens_seen": 45431728,
      "step": 78310
    },
    {
      "epoch": 11.66443252904379,
      "grad_norm": 8.5625,
      "learning_rate": 2.21080052377094e-05,
      "loss": 1.4366,
      "num_input_tokens_seen": 45435088,
      "step": 78315
    },
    {
      "epoch": 11.665177241584749,
      "grad_norm": 19.875,
      "learning_rate": 2.210477765549484e-05,
      "loss": 1.2524,
      "num_input_tokens_seen": 45438416,
      "step": 78320
    },
    {
      "epoch": 11.665921954125707,
      "grad_norm": 10.1875,
      "learning_rate": 2.210155012219211e-05,
      "loss": 0.9783,
      "num_input_tokens_seen": 45441648,
      "step": 78325
    },
    {
      "epoch": 11.666666666666666,
      "grad_norm": 12.875,
      "learning_rate": 2.2098322637855757e-05,
      "loss": 0.9909,
      "num_input_tokens_seen": 45444528,
      "step": 78330
    },
    {
      "epoch": 11.667411379207627,
      "grad_norm": 9.0,
      "learning_rate": 2.2095095202540293e-05,
      "loss": 1.1066,
      "num_input_tokens_seen": 45447280,
      "step": 78335
    },
    {
      "epoch": 11.668156091748585,
      "grad_norm": 6.15625,
      "learning_rate": 2.209186781630023e-05,
      "loss": 1.2252,
      "num_input_tokens_seen": 45450288,
      "step": 78340
    },
    {
      "epoch": 11.668900804289544,
      "grad_norm": 8.375,
      "learning_rate": 2.2088640479190116e-05,
      "loss": 1.0875,
      "num_input_tokens_seen": 45453008,
      "step": 78345
    },
    {
      "epoch": 11.669645516830503,
      "grad_norm": 14.8125,
      "learning_rate": 2.208541319126446e-05,
      "loss": 1.0382,
      "num_input_tokens_seen": 45455536,
      "step": 78350
    },
    {
      "epoch": 11.670390229371463,
      "grad_norm": 4.90625,
      "learning_rate": 2.2082185952577788e-05,
      "loss": 1.0489,
      "num_input_tokens_seen": 45458352,
      "step": 78355
    },
    {
      "epoch": 11.671134941912422,
      "grad_norm": 9.5625,
      "learning_rate": 2.207895876318461e-05,
      "loss": 1.129,
      "num_input_tokens_seen": 45461072,
      "step": 78360
    },
    {
      "epoch": 11.671879654453381,
      "grad_norm": 7.5625,
      "learning_rate": 2.2075731623139463e-05,
      "loss": 0.8595,
      "num_input_tokens_seen": 45464272,
      "step": 78365
    },
    {
      "epoch": 11.67262436699434,
      "grad_norm": 13.5,
      "learning_rate": 2.207250453249685e-05,
      "loss": 1.2298,
      "num_input_tokens_seen": 45467248,
      "step": 78370
    },
    {
      "epoch": 11.673369079535298,
      "grad_norm": 9.9375,
      "learning_rate": 2.2069277491311306e-05,
      "loss": 0.9296,
      "num_input_tokens_seen": 45470192,
      "step": 78375
    },
    {
      "epoch": 11.674113792076259,
      "grad_norm": 4.65625,
      "learning_rate": 2.2066050499637344e-05,
      "loss": 1.2516,
      "num_input_tokens_seen": 45472912,
      "step": 78380
    },
    {
      "epoch": 11.674858504617218,
      "grad_norm": 12.0625,
      "learning_rate": 2.2062823557529467e-05,
      "loss": 1.072,
      "num_input_tokens_seen": 45476176,
      "step": 78385
    },
    {
      "epoch": 11.675603217158177,
      "grad_norm": 11.4375,
      "learning_rate": 2.2059596665042213e-05,
      "loss": 0.9588,
      "num_input_tokens_seen": 45479216,
      "step": 78390
    },
    {
      "epoch": 11.676347929699137,
      "grad_norm": 4.53125,
      "learning_rate": 2.2056369822230067e-05,
      "loss": 0.8042,
      "num_input_tokens_seen": 45482256,
      "step": 78395
    },
    {
      "epoch": 11.677092642240096,
      "grad_norm": 5.1875,
      "learning_rate": 2.2053143029147574e-05,
      "loss": 0.7686,
      "num_input_tokens_seen": 45485232,
      "step": 78400
    },
    {
      "epoch": 11.677837354781055,
      "grad_norm": 14.5,
      "learning_rate": 2.2049916285849233e-05,
      "loss": 0.851,
      "num_input_tokens_seen": 45487984,
      "step": 78405
    },
    {
      "epoch": 11.678582067322013,
      "grad_norm": 9.9375,
      "learning_rate": 2.204668959238955e-05,
      "loss": 1.4475,
      "num_input_tokens_seen": 45490672,
      "step": 78410
    },
    {
      "epoch": 11.679326779862972,
      "grad_norm": 8.3125,
      "learning_rate": 2.2043462948823057e-05,
      "loss": 1.1876,
      "num_input_tokens_seen": 45493744,
      "step": 78415
    },
    {
      "epoch": 11.680071492403933,
      "grad_norm": 12.1875,
      "learning_rate": 2.2040236355204244e-05,
      "loss": 1.0112,
      "num_input_tokens_seen": 45496432,
      "step": 78420
    },
    {
      "epoch": 11.680816204944891,
      "grad_norm": 11.25,
      "learning_rate": 2.2037009811587638e-05,
      "loss": 0.9782,
      "num_input_tokens_seen": 45499216,
      "step": 78425
    },
    {
      "epoch": 11.68156091748585,
      "grad_norm": 10.25,
      "learning_rate": 2.2033783318027725e-05,
      "loss": 0.7732,
      "num_input_tokens_seen": 45502160,
      "step": 78430
    },
    {
      "epoch": 11.682305630026809,
      "grad_norm": 7.15625,
      "learning_rate": 2.203055687457904e-05,
      "loss": 1.004,
      "num_input_tokens_seen": 45505296,
      "step": 78435
    },
    {
      "epoch": 11.68305034256777,
      "grad_norm": 8.625,
      "learning_rate": 2.2027330481296074e-05,
      "loss": 1.2481,
      "num_input_tokens_seen": 45508112,
      "step": 78440
    },
    {
      "epoch": 11.683795055108728,
      "grad_norm": 13.8125,
      "learning_rate": 2.2024104138233343e-05,
      "loss": 0.8945,
      "num_input_tokens_seen": 45510672,
      "step": 78445
    },
    {
      "epoch": 11.684539767649687,
      "grad_norm": 7.3125,
      "learning_rate": 2.2020877845445338e-05,
      "loss": 0.8998,
      "num_input_tokens_seen": 45513520,
      "step": 78450
    },
    {
      "epoch": 11.685284480190646,
      "grad_norm": 8.0,
      "learning_rate": 2.2017651602986584e-05,
      "loss": 1.0046,
      "num_input_tokens_seen": 45516272,
      "step": 78455
    },
    {
      "epoch": 11.686029192731606,
      "grad_norm": 7.625,
      "learning_rate": 2.2014425410911575e-05,
      "loss": 1.3583,
      "num_input_tokens_seen": 45519088,
      "step": 78460
    },
    {
      "epoch": 11.686773905272565,
      "grad_norm": 13.5625,
      "learning_rate": 2.2011199269274804e-05,
      "loss": 1.2267,
      "num_input_tokens_seen": 45522000,
      "step": 78465
    },
    {
      "epoch": 11.687518617813524,
      "grad_norm": 10.3125,
      "learning_rate": 2.2007973178130795e-05,
      "loss": 1.067,
      "num_input_tokens_seen": 45525200,
      "step": 78470
    },
    {
      "epoch": 11.688263330354483,
      "grad_norm": 9.75,
      "learning_rate": 2.2004747137534032e-05,
      "loss": 0.9788,
      "num_input_tokens_seen": 45527824,
      "step": 78475
    },
    {
      "epoch": 11.689008042895443,
      "grad_norm": 11.75,
      "learning_rate": 2.2001521147539028e-05,
      "loss": 1.2302,
      "num_input_tokens_seen": 45530544,
      "step": 78480
    },
    {
      "epoch": 11.689752755436402,
      "grad_norm": 19.0,
      "learning_rate": 2.1998295208200263e-05,
      "loss": 1.1812,
      "num_input_tokens_seen": 45533360,
      "step": 78485
    },
    {
      "epoch": 11.69049746797736,
      "grad_norm": 14.5,
      "learning_rate": 2.1995069319572264e-05,
      "loss": 1.3255,
      "num_input_tokens_seen": 45536112,
      "step": 78490
    },
    {
      "epoch": 11.69124218051832,
      "grad_norm": 9.5625,
      "learning_rate": 2.1991843481709513e-05,
      "loss": 0.7881,
      "num_input_tokens_seen": 45538800,
      "step": 78495
    },
    {
      "epoch": 11.69198689305928,
      "grad_norm": 11.1875,
      "learning_rate": 2.19886176946665e-05,
      "loss": 1.0789,
      "num_input_tokens_seen": 45541744,
      "step": 78500
    },
    {
      "epoch": 11.692731605600239,
      "grad_norm": 14.9375,
      "learning_rate": 2.1985391958497743e-05,
      "loss": 1.0487,
      "num_input_tokens_seen": 45544336,
      "step": 78505
    },
    {
      "epoch": 11.693476318141197,
      "grad_norm": 7.15625,
      "learning_rate": 2.1982166273257716e-05,
      "loss": 1.0761,
      "num_input_tokens_seen": 45547056,
      "step": 78510
    },
    {
      "epoch": 11.694221030682156,
      "grad_norm": 6.46875,
      "learning_rate": 2.1978940639000927e-05,
      "loss": 0.7827,
      "num_input_tokens_seen": 45549968,
      "step": 78515
    },
    {
      "epoch": 11.694965743223117,
      "grad_norm": 6.4375,
      "learning_rate": 2.1975715055781858e-05,
      "loss": 0.868,
      "num_input_tokens_seen": 45553232,
      "step": 78520
    },
    {
      "epoch": 11.695710455764075,
      "grad_norm": 9.3125,
      "learning_rate": 2.1972489523655016e-05,
      "loss": 0.9569,
      "num_input_tokens_seen": 45556240,
      "step": 78525
    },
    {
      "epoch": 11.696455168305034,
      "grad_norm": 9.9375,
      "learning_rate": 2.1969264042674877e-05,
      "loss": 0.9783,
      "num_input_tokens_seen": 45559216,
      "step": 78530
    },
    {
      "epoch": 11.697199880845993,
      "grad_norm": 7.09375,
      "learning_rate": 2.1966038612895958e-05,
      "loss": 0.886,
      "num_input_tokens_seen": 45562128,
      "step": 78535
    },
    {
      "epoch": 11.697944593386953,
      "grad_norm": 10.125,
      "learning_rate": 2.1962813234372727e-05,
      "loss": 1.0662,
      "num_input_tokens_seen": 45565200,
      "step": 78540
    },
    {
      "epoch": 11.698689305927912,
      "grad_norm": 10.8125,
      "learning_rate": 2.1959587907159673e-05,
      "loss": 1.1596,
      "num_input_tokens_seen": 45568016,
      "step": 78545
    },
    {
      "epoch": 11.699434018468871,
      "grad_norm": 10.4375,
      "learning_rate": 2.19563626313113e-05,
      "loss": 1.2074,
      "num_input_tokens_seen": 45571120,
      "step": 78550
    },
    {
      "epoch": 11.70017873100983,
      "grad_norm": 9.3125,
      "learning_rate": 2.1953137406882078e-05,
      "loss": 1.0263,
      "num_input_tokens_seen": 45574096,
      "step": 78555
    },
    {
      "epoch": 11.700923443550789,
      "grad_norm": 5.59375,
      "learning_rate": 2.194991223392651e-05,
      "loss": 1.2502,
      "num_input_tokens_seen": 45577136,
      "step": 78560
    },
    {
      "epoch": 11.701668156091749,
      "grad_norm": 6.59375,
      "learning_rate": 2.1946687112499066e-05,
      "loss": 0.8884,
      "num_input_tokens_seen": 45579920,
      "step": 78565
    },
    {
      "epoch": 11.702412868632708,
      "grad_norm": 5.0,
      "learning_rate": 2.194346204265425e-05,
      "loss": 1.0297,
      "num_input_tokens_seen": 45583472,
      "step": 78570
    },
    {
      "epoch": 11.703157581173667,
      "grad_norm": 10.5625,
      "learning_rate": 2.1940237024446535e-05,
      "loss": 1.1182,
      "num_input_tokens_seen": 45586192,
      "step": 78575
    },
    {
      "epoch": 11.703902293714627,
      "grad_norm": 19.0,
      "learning_rate": 2.19370120579304e-05,
      "loss": 1.2327,
      "num_input_tokens_seen": 45589328,
      "step": 78580
    },
    {
      "epoch": 11.704647006255586,
      "grad_norm": 9.9375,
      "learning_rate": 2.1933787143160343e-05,
      "loss": 1.4196,
      "num_input_tokens_seen": 45592464,
      "step": 78585
    },
    {
      "epoch": 11.705391718796545,
      "grad_norm": 12.375,
      "learning_rate": 2.193056228019082e-05,
      "loss": 1.0157,
      "num_input_tokens_seen": 45595568,
      "step": 78590
    },
    {
      "epoch": 11.706136431337503,
      "grad_norm": 12.375,
      "learning_rate": 2.1927337469076343e-05,
      "loss": 1.0033,
      "num_input_tokens_seen": 45598512,
      "step": 78595
    },
    {
      "epoch": 11.706881143878462,
      "grad_norm": 8.9375,
      "learning_rate": 2.1924112709871362e-05,
      "loss": 1.1345,
      "num_input_tokens_seen": 45601680,
      "step": 78600
    },
    {
      "epoch": 11.707625856419423,
      "grad_norm": 8.625,
      "learning_rate": 2.1920888002630382e-05,
      "loss": 0.7777,
      "num_input_tokens_seen": 45604528,
      "step": 78605
    },
    {
      "epoch": 11.708370568960381,
      "grad_norm": 9.6875,
      "learning_rate": 2.1917663347407867e-05,
      "loss": 0.9805,
      "num_input_tokens_seen": 45607408,
      "step": 78610
    },
    {
      "epoch": 11.70911528150134,
      "grad_norm": 5.875,
      "learning_rate": 2.1914438744258298e-05,
      "loss": 1.1918,
      "num_input_tokens_seen": 45610320,
      "step": 78615
    },
    {
      "epoch": 11.709859994042299,
      "grad_norm": 11.0,
      "learning_rate": 2.1911214193236153e-05,
      "loss": 1.0514,
      "num_input_tokens_seen": 45613072,
      "step": 78620
    },
    {
      "epoch": 11.71060470658326,
      "grad_norm": 17.625,
      "learning_rate": 2.1907989694395893e-05,
      "loss": 1.2933,
      "num_input_tokens_seen": 45615600,
      "step": 78625
    },
    {
      "epoch": 11.711349419124218,
      "grad_norm": 4.09375,
      "learning_rate": 2.1904765247792016e-05,
      "loss": 1.0339,
      "num_input_tokens_seen": 45618480,
      "step": 78630
    },
    {
      "epoch": 11.712094131665177,
      "grad_norm": 8.5625,
      "learning_rate": 2.1901540853478976e-05,
      "loss": 1.1239,
      "num_input_tokens_seen": 45621392,
      "step": 78635
    },
    {
      "epoch": 11.712838844206136,
      "grad_norm": 13.1875,
      "learning_rate": 2.1898316511511264e-05,
      "loss": 0.8005,
      "num_input_tokens_seen": 45624144,
      "step": 78640
    },
    {
      "epoch": 11.713583556747096,
      "grad_norm": 8.0,
      "learning_rate": 2.1895092221943335e-05,
      "loss": 1.0646,
      "num_input_tokens_seen": 45627280,
      "step": 78645
    },
    {
      "epoch": 11.714328269288055,
      "grad_norm": 13.375,
      "learning_rate": 2.1891867984829672e-05,
      "loss": 1.2181,
      "num_input_tokens_seen": 45630288,
      "step": 78650
    },
    {
      "epoch": 11.715072981829014,
      "grad_norm": 11.25,
      "learning_rate": 2.1888643800224728e-05,
      "loss": 1.3093,
      "num_input_tokens_seen": 45633104,
      "step": 78655
    },
    {
      "epoch": 11.715817694369973,
      "grad_norm": 6.03125,
      "learning_rate": 2.1885419668183e-05,
      "loss": 0.9119,
      "num_input_tokens_seen": 45636176,
      "step": 78660
    },
    {
      "epoch": 11.716562406910933,
      "grad_norm": 11.75,
      "learning_rate": 2.188219558875894e-05,
      "loss": 1.1132,
      "num_input_tokens_seen": 45639120,
      "step": 78665
    },
    {
      "epoch": 11.717307119451892,
      "grad_norm": 8.3125,
      "learning_rate": 2.1878971562007007e-05,
      "loss": 1.3298,
      "num_input_tokens_seen": 45642000,
      "step": 78670
    },
    {
      "epoch": 11.71805183199285,
      "grad_norm": 9.3125,
      "learning_rate": 2.1875747587981686e-05,
      "loss": 1.174,
      "num_input_tokens_seen": 45644752,
      "step": 78675
    },
    {
      "epoch": 11.71879654453381,
      "grad_norm": 14.125,
      "learning_rate": 2.1872523666737428e-05,
      "loss": 0.9096,
      "num_input_tokens_seen": 45647696,
      "step": 78680
    },
    {
      "epoch": 11.71954125707477,
      "grad_norm": 8.25,
      "learning_rate": 2.186929979832871e-05,
      "loss": 1.0347,
      "num_input_tokens_seen": 45650896,
      "step": 78685
    },
    {
      "epoch": 11.720285969615729,
      "grad_norm": 8.875,
      "learning_rate": 2.186607598280998e-05,
      "loss": 1.0805,
      "num_input_tokens_seen": 45654192,
      "step": 78690
    },
    {
      "epoch": 11.721030682156687,
      "grad_norm": 13.0625,
      "learning_rate": 2.186285222023572e-05,
      "loss": 0.822,
      "num_input_tokens_seen": 45656880,
      "step": 78695
    },
    {
      "epoch": 11.721775394697646,
      "grad_norm": 8.125,
      "learning_rate": 2.185962851066039e-05,
      "loss": 1.1647,
      "num_input_tokens_seen": 45660272,
      "step": 78700
    },
    {
      "epoch": 11.722520107238607,
      "grad_norm": 8.4375,
      "learning_rate": 2.1856404854138426e-05,
      "loss": 1.0587,
      "num_input_tokens_seen": 45663088,
      "step": 78705
    },
    {
      "epoch": 11.723264819779565,
      "grad_norm": 9.4375,
      "learning_rate": 2.1853181250724318e-05,
      "loss": 1.0766,
      "num_input_tokens_seen": 45666064,
      "step": 78710
    },
    {
      "epoch": 11.724009532320524,
      "grad_norm": 3.796875,
      "learning_rate": 2.1849957700472515e-05,
      "loss": 0.9732,
      "num_input_tokens_seen": 45669008,
      "step": 78715
    },
    {
      "epoch": 11.724754244861483,
      "grad_norm": 11.3125,
      "learning_rate": 2.1846734203437478e-05,
      "loss": 1.3001,
      "num_input_tokens_seen": 45671728,
      "step": 78720
    },
    {
      "epoch": 11.725498957402444,
      "grad_norm": 11.0625,
      "learning_rate": 2.1843510759673648e-05,
      "loss": 1.2476,
      "num_input_tokens_seen": 45674864,
      "step": 78725
    },
    {
      "epoch": 11.726243669943402,
      "grad_norm": 7.9375,
      "learning_rate": 2.184028736923551e-05,
      "loss": 1.0894,
      "num_input_tokens_seen": 45677712,
      "step": 78730
    },
    {
      "epoch": 11.726988382484361,
      "grad_norm": 12.875,
      "learning_rate": 2.1837064032177497e-05,
      "loss": 0.8646,
      "num_input_tokens_seen": 45680272,
      "step": 78735
    },
    {
      "epoch": 11.72773309502532,
      "grad_norm": 8.5625,
      "learning_rate": 2.1833840748554075e-05,
      "loss": 1.1914,
      "num_input_tokens_seen": 45683248,
      "step": 78740
    },
    {
      "epoch": 11.728477807566279,
      "grad_norm": 17.125,
      "learning_rate": 2.18306175184197e-05,
      "loss": 1.3898,
      "num_input_tokens_seen": 45686192,
      "step": 78745
    },
    {
      "epoch": 11.729222520107239,
      "grad_norm": 16.0,
      "learning_rate": 2.1827394341828817e-05,
      "loss": 0.9799,
      "num_input_tokens_seen": 45689360,
      "step": 78750
    },
    {
      "epoch": 11.729967232648198,
      "grad_norm": 11.4375,
      "learning_rate": 2.1824171218835886e-05,
      "loss": 1.5215,
      "num_input_tokens_seen": 45692272,
      "step": 78755
    },
    {
      "epoch": 11.730711945189157,
      "grad_norm": 5.125,
      "learning_rate": 2.1820948149495343e-05,
      "loss": 0.9573,
      "num_input_tokens_seen": 45695280,
      "step": 78760
    },
    {
      "epoch": 11.731456657730115,
      "grad_norm": 21.0,
      "learning_rate": 2.181772513386166e-05,
      "loss": 1.7102,
      "num_input_tokens_seen": 45697968,
      "step": 78765
    },
    {
      "epoch": 11.732201370271076,
      "grad_norm": 6.90625,
      "learning_rate": 2.1814502171989276e-05,
      "loss": 1.1485,
      "num_input_tokens_seen": 45700944,
      "step": 78770
    },
    {
      "epoch": 11.732946082812035,
      "grad_norm": 6.3125,
      "learning_rate": 2.1811279263932642e-05,
      "loss": 0.8326,
      "num_input_tokens_seen": 45703888,
      "step": 78775
    },
    {
      "epoch": 11.733690795352993,
      "grad_norm": 12.3125,
      "learning_rate": 2.1808056409746196e-05,
      "loss": 1.2475,
      "num_input_tokens_seen": 45706672,
      "step": 78780
    },
    {
      "epoch": 11.734435507893952,
      "grad_norm": 5.28125,
      "learning_rate": 2.18048336094844e-05,
      "loss": 0.8267,
      "num_input_tokens_seen": 45709744,
      "step": 78785
    },
    {
      "epoch": 11.735180220434913,
      "grad_norm": 11.0625,
      "learning_rate": 2.180161086320169e-05,
      "loss": 1.2149,
      "num_input_tokens_seen": 45712432,
      "step": 78790
    },
    {
      "epoch": 11.735924932975871,
      "grad_norm": 18.75,
      "learning_rate": 2.1798388170952508e-05,
      "loss": 0.8503,
      "num_input_tokens_seen": 45715248,
      "step": 78795
    },
    {
      "epoch": 11.73666964551683,
      "grad_norm": 6.78125,
      "learning_rate": 2.1795165532791315e-05,
      "loss": 0.9811,
      "num_input_tokens_seen": 45718288,
      "step": 78800
    },
    {
      "epoch": 11.737414358057789,
      "grad_norm": 6.25,
      "learning_rate": 2.1791942948772533e-05,
      "loss": 0.7639,
      "num_input_tokens_seen": 45721136,
      "step": 78805
    },
    {
      "epoch": 11.73815907059875,
      "grad_norm": 10.3125,
      "learning_rate": 2.1788720418950626e-05,
      "loss": 1.2343,
      "num_input_tokens_seen": 45724144,
      "step": 78810
    },
    {
      "epoch": 11.738903783139708,
      "grad_norm": 10.75,
      "learning_rate": 2.178549794338001e-05,
      "loss": 1.4707,
      "num_input_tokens_seen": 45726960,
      "step": 78815
    },
    {
      "epoch": 11.739648495680667,
      "grad_norm": 11.0,
      "learning_rate": 2.178227552211515e-05,
      "loss": 0.8833,
      "num_input_tokens_seen": 45729936,
      "step": 78820
    },
    {
      "epoch": 11.740393208221626,
      "grad_norm": 13.625,
      "learning_rate": 2.1779053155210474e-05,
      "loss": 0.8952,
      "num_input_tokens_seen": 45732752,
      "step": 78825
    },
    {
      "epoch": 11.741137920762586,
      "grad_norm": 16.5,
      "learning_rate": 2.177583084272041e-05,
      "loss": 1.149,
      "num_input_tokens_seen": 45735792,
      "step": 78830
    },
    {
      "epoch": 11.741882633303545,
      "grad_norm": 8.625,
      "learning_rate": 2.177260858469942e-05,
      "loss": 0.9259,
      "num_input_tokens_seen": 45738608,
      "step": 78835
    },
    {
      "epoch": 11.742627345844504,
      "grad_norm": 12.5625,
      "learning_rate": 2.176938638120192e-05,
      "loss": 1.069,
      "num_input_tokens_seen": 45741584,
      "step": 78840
    },
    {
      "epoch": 11.743372058385463,
      "grad_norm": 9.875,
      "learning_rate": 2.176616423228236e-05,
      "loss": 1.0826,
      "num_input_tokens_seen": 45744592,
      "step": 78845
    },
    {
      "epoch": 11.744116770926423,
      "grad_norm": 11.75,
      "learning_rate": 2.1762942137995158e-05,
      "loss": 1.0891,
      "num_input_tokens_seen": 45747504,
      "step": 78850
    },
    {
      "epoch": 11.744861483467382,
      "grad_norm": 9.0,
      "learning_rate": 2.175972009839477e-05,
      "loss": 1.1599,
      "num_input_tokens_seen": 45750288,
      "step": 78855
    },
    {
      "epoch": 11.74560619600834,
      "grad_norm": 22.5,
      "learning_rate": 2.1756498113535617e-05,
      "loss": 1.0841,
      "num_input_tokens_seen": 45753264,
      "step": 78860
    },
    {
      "epoch": 11.7463509085493,
      "grad_norm": 12.0,
      "learning_rate": 2.1753276183472122e-05,
      "loss": 1.3461,
      "num_input_tokens_seen": 45756144,
      "step": 78865
    },
    {
      "epoch": 11.74709562109026,
      "grad_norm": 7.1875,
      "learning_rate": 2.1750054308258737e-05,
      "loss": 0.949,
      "num_input_tokens_seen": 45758960,
      "step": 78870
    },
    {
      "epoch": 11.747840333631219,
      "grad_norm": 16.375,
      "learning_rate": 2.1746832487949874e-05,
      "loss": 0.8753,
      "num_input_tokens_seen": 45761680,
      "step": 78875
    },
    {
      "epoch": 11.748585046172177,
      "grad_norm": 22.5,
      "learning_rate": 2.174361072259998e-05,
      "loss": 1.4558,
      "num_input_tokens_seen": 45764944,
      "step": 78880
    },
    {
      "epoch": 11.749329758713136,
      "grad_norm": 13.5625,
      "learning_rate": 2.1740389012263454e-05,
      "loss": 1.0051,
      "num_input_tokens_seen": 45768048,
      "step": 78885
    },
    {
      "epoch": 11.750074471254095,
      "grad_norm": 5.875,
      "learning_rate": 2.173716735699476e-05,
      "loss": 1.1305,
      "num_input_tokens_seen": 45770992,
      "step": 78890
    },
    {
      "epoch": 11.750819183795056,
      "grad_norm": 6.84375,
      "learning_rate": 2.173394575684829e-05,
      "loss": 0.9931,
      "num_input_tokens_seen": 45773936,
      "step": 78895
    },
    {
      "epoch": 11.751563896336014,
      "grad_norm": 5.96875,
      "learning_rate": 2.1730724211878506e-05,
      "loss": 1.1181,
      "num_input_tokens_seen": 45776592,
      "step": 78900
    },
    {
      "epoch": 11.752308608876973,
      "grad_norm": 11.75,
      "learning_rate": 2.172750272213981e-05,
      "loss": 1.0775,
      "num_input_tokens_seen": 45779376,
      "step": 78905
    },
    {
      "epoch": 11.753053321417934,
      "grad_norm": 9.0,
      "learning_rate": 2.1724281287686622e-05,
      "loss": 0.9504,
      "num_input_tokens_seen": 45782288,
      "step": 78910
    },
    {
      "epoch": 11.753798033958892,
      "grad_norm": 17.5,
      "learning_rate": 2.1721059908573383e-05,
      "loss": 1.0765,
      "num_input_tokens_seen": 45785296,
      "step": 78915
    },
    {
      "epoch": 11.754542746499851,
      "grad_norm": 5.75,
      "learning_rate": 2.171783858485449e-05,
      "loss": 0.8161,
      "num_input_tokens_seen": 45787952,
      "step": 78920
    },
    {
      "epoch": 11.75528745904081,
      "grad_norm": 5.4375,
      "learning_rate": 2.171461731658439e-05,
      "loss": 1.1615,
      "num_input_tokens_seen": 45790704,
      "step": 78925
    },
    {
      "epoch": 11.756032171581769,
      "grad_norm": 8.6875,
      "learning_rate": 2.1711396103817477e-05,
      "loss": 1.311,
      "num_input_tokens_seen": 45793552,
      "step": 78930
    },
    {
      "epoch": 11.75677688412273,
      "grad_norm": 11.375,
      "learning_rate": 2.17081749466082e-05,
      "loss": 1.2486,
      "num_input_tokens_seen": 45796528,
      "step": 78935
    },
    {
      "epoch": 11.757521596663688,
      "grad_norm": 8.8125,
      "learning_rate": 2.170495384501096e-05,
      "loss": 0.9226,
      "num_input_tokens_seen": 45799536,
      "step": 78940
    },
    {
      "epoch": 11.758266309204647,
      "grad_norm": 12.5625,
      "learning_rate": 2.1701732799080173e-05,
      "loss": 0.8804,
      "num_input_tokens_seen": 45802544,
      "step": 78945
    },
    {
      "epoch": 11.759011021745605,
      "grad_norm": 10.875,
      "learning_rate": 2.169851180887026e-05,
      "loss": 1.3314,
      "num_input_tokens_seen": 45805520,
      "step": 78950
    },
    {
      "epoch": 11.759755734286566,
      "grad_norm": 10.125,
      "learning_rate": 2.1695290874435623e-05,
      "loss": 1.4192,
      "num_input_tokens_seen": 45808304,
      "step": 78955
    },
    {
      "epoch": 11.760500446827525,
      "grad_norm": 21.375,
      "learning_rate": 2.16920699958307e-05,
      "loss": 1.4115,
      "num_input_tokens_seen": 45811120,
      "step": 78960
    },
    {
      "epoch": 11.761245159368483,
      "grad_norm": 10.0625,
      "learning_rate": 2.168884917310988e-05,
      "loss": 0.8294,
      "num_input_tokens_seen": 45814096,
      "step": 78965
    },
    {
      "epoch": 11.761989871909442,
      "grad_norm": 7.40625,
      "learning_rate": 2.16856284063276e-05,
      "loss": 1.3753,
      "num_input_tokens_seen": 45817008,
      "step": 78970
    },
    {
      "epoch": 11.762734584450403,
      "grad_norm": 12.0625,
      "learning_rate": 2.1682407695538255e-05,
      "loss": 0.9873,
      "num_input_tokens_seen": 45820080,
      "step": 78975
    },
    {
      "epoch": 11.763479296991362,
      "grad_norm": 15.0,
      "learning_rate": 2.1679187040796266e-05,
      "loss": 1.0979,
      "num_input_tokens_seen": 45822896,
      "step": 78980
    },
    {
      "epoch": 11.76422400953232,
      "grad_norm": 11.3125,
      "learning_rate": 2.1675966442156038e-05,
      "loss": 1.1568,
      "num_input_tokens_seen": 45825904,
      "step": 78985
    },
    {
      "epoch": 11.764968722073279,
      "grad_norm": 7.9375,
      "learning_rate": 2.1672745899671965e-05,
      "loss": 1.1169,
      "num_input_tokens_seen": 45829072,
      "step": 78990
    },
    {
      "epoch": 11.76571343461424,
      "grad_norm": 16.75,
      "learning_rate": 2.1669525413398477e-05,
      "loss": 1.1115,
      "num_input_tokens_seen": 45831920,
      "step": 78995
    },
    {
      "epoch": 11.766458147155198,
      "grad_norm": 8.375,
      "learning_rate": 2.166630498338997e-05,
      "loss": 0.8975,
      "num_input_tokens_seen": 45835152,
      "step": 79000
    },
    {
      "epoch": 11.767202859696157,
      "grad_norm": 6.75,
      "learning_rate": 2.1663084609700853e-05,
      "loss": 0.9943,
      "num_input_tokens_seen": 45837904,
      "step": 79005
    },
    {
      "epoch": 11.767947572237116,
      "grad_norm": 8.25,
      "learning_rate": 2.1659864292385528e-05,
      "loss": 0.9663,
      "num_input_tokens_seen": 45840912,
      "step": 79010
    },
    {
      "epoch": 11.768692284778076,
      "grad_norm": 20.375,
      "learning_rate": 2.1656644031498407e-05,
      "loss": 0.9605,
      "num_input_tokens_seen": 45843792,
      "step": 79015
    },
    {
      "epoch": 11.769436997319035,
      "grad_norm": 6.53125,
      "learning_rate": 2.1653423827093888e-05,
      "loss": 1.0331,
      "num_input_tokens_seen": 45846960,
      "step": 79020
    },
    {
      "epoch": 11.770181709859994,
      "grad_norm": 12.25,
      "learning_rate": 2.1650203679226362e-05,
      "loss": 1.1103,
      "num_input_tokens_seen": 45849776,
      "step": 79025
    },
    {
      "epoch": 11.770926422400953,
      "grad_norm": 8.1875,
      "learning_rate": 2.164698358795025e-05,
      "loss": 1.2957,
      "num_input_tokens_seen": 45852848,
      "step": 79030
    },
    {
      "epoch": 11.771671134941913,
      "grad_norm": 17.875,
      "learning_rate": 2.164376355331993e-05,
      "loss": 1.4328,
      "num_input_tokens_seen": 45855920,
      "step": 79035
    },
    {
      "epoch": 11.772415847482872,
      "grad_norm": 10.5,
      "learning_rate": 2.1640543575389828e-05,
      "loss": 1.2498,
      "num_input_tokens_seen": 45858800,
      "step": 79040
    },
    {
      "epoch": 11.77316056002383,
      "grad_norm": 7.375,
      "learning_rate": 2.163732365421432e-05,
      "loss": 0.9825,
      "num_input_tokens_seen": 45861744,
      "step": 79045
    },
    {
      "epoch": 11.77390527256479,
      "grad_norm": 8.5,
      "learning_rate": 2.1634103789847813e-05,
      "loss": 0.9125,
      "num_input_tokens_seen": 45864528,
      "step": 79050
    },
    {
      "epoch": 11.77464998510575,
      "grad_norm": 18.625,
      "learning_rate": 2.1630883982344695e-05,
      "loss": 1.3487,
      "num_input_tokens_seen": 45867280,
      "step": 79055
    },
    {
      "epoch": 11.775394697646709,
      "grad_norm": 8.875,
      "learning_rate": 2.1627664231759383e-05,
      "loss": 0.9983,
      "num_input_tokens_seen": 45870032,
      "step": 79060
    },
    {
      "epoch": 11.776139410187668,
      "grad_norm": 10.125,
      "learning_rate": 2.1624444538146248e-05,
      "loss": 0.8468,
      "num_input_tokens_seen": 45872912,
      "step": 79065
    },
    {
      "epoch": 11.776884122728626,
      "grad_norm": 8.375,
      "learning_rate": 2.1621224901559685e-05,
      "loss": 1.1412,
      "num_input_tokens_seen": 45875504,
      "step": 79070
    },
    {
      "epoch": 11.777628835269585,
      "grad_norm": 10.75,
      "learning_rate": 2.1618005322054103e-05,
      "loss": 1.1778,
      "num_input_tokens_seen": 45878256,
      "step": 79075
    },
    {
      "epoch": 11.778373547810546,
      "grad_norm": 7.25,
      "learning_rate": 2.1614785799683877e-05,
      "loss": 0.9133,
      "num_input_tokens_seen": 45881136,
      "step": 79080
    },
    {
      "epoch": 11.779118260351504,
      "grad_norm": 6.71875,
      "learning_rate": 2.1611566334503413e-05,
      "loss": 1.1265,
      "num_input_tokens_seen": 45884368,
      "step": 79085
    },
    {
      "epoch": 11.779862972892463,
      "grad_norm": 7.125,
      "learning_rate": 2.160834692656708e-05,
      "loss": 1.125,
      "num_input_tokens_seen": 45887280,
      "step": 79090
    },
    {
      "epoch": 11.780607685433424,
      "grad_norm": 10.4375,
      "learning_rate": 2.160512757592929e-05,
      "loss": 1.3845,
      "num_input_tokens_seen": 45890000,
      "step": 79095
    },
    {
      "epoch": 11.781352397974382,
      "grad_norm": 6.28125,
      "learning_rate": 2.1601908282644418e-05,
      "loss": 0.8935,
      "num_input_tokens_seen": 45892720,
      "step": 79100
    },
    {
      "epoch": 11.782097110515341,
      "grad_norm": 12.75,
      "learning_rate": 2.1598689046766848e-05,
      "loss": 1.3516,
      "num_input_tokens_seen": 45895728,
      "step": 79105
    },
    {
      "epoch": 11.7828418230563,
      "grad_norm": 9.625,
      "learning_rate": 2.1595469868350966e-05,
      "loss": 0.9872,
      "num_input_tokens_seen": 45898576,
      "step": 79110
    },
    {
      "epoch": 11.783586535597259,
      "grad_norm": 6.53125,
      "learning_rate": 2.1592250747451166e-05,
      "loss": 1.0808,
      "num_input_tokens_seen": 45901648,
      "step": 79115
    },
    {
      "epoch": 11.78433124813822,
      "grad_norm": 8.8125,
      "learning_rate": 2.1589031684121828e-05,
      "loss": 0.8124,
      "num_input_tokens_seen": 45904496,
      "step": 79120
    },
    {
      "epoch": 11.785075960679178,
      "grad_norm": 12.5,
      "learning_rate": 2.1585812678417323e-05,
      "loss": 1.1041,
      "num_input_tokens_seen": 45907440,
      "step": 79125
    },
    {
      "epoch": 11.785820673220137,
      "grad_norm": 8.125,
      "learning_rate": 2.1582593730392055e-05,
      "loss": 1.3643,
      "num_input_tokens_seen": 45910160,
      "step": 79130
    },
    {
      "epoch": 11.786565385761095,
      "grad_norm": 4.8125,
      "learning_rate": 2.1579374840100383e-05,
      "loss": 0.8123,
      "num_input_tokens_seen": 45913168,
      "step": 79135
    },
    {
      "epoch": 11.787310098302056,
      "grad_norm": 8.125,
      "learning_rate": 2.1576156007596705e-05,
      "loss": 1.0699,
      "num_input_tokens_seen": 45916016,
      "step": 79140
    },
    {
      "epoch": 11.788054810843015,
      "grad_norm": 9.125,
      "learning_rate": 2.1572937232935385e-05,
      "loss": 1.1188,
      "num_input_tokens_seen": 45918576,
      "step": 79145
    },
    {
      "epoch": 11.788799523383974,
      "grad_norm": 9.875,
      "learning_rate": 2.1569718516170806e-05,
      "loss": 0.995,
      "num_input_tokens_seen": 45921616,
      "step": 79150
    },
    {
      "epoch": 11.789544235924932,
      "grad_norm": 15.25,
      "learning_rate": 2.1566499857357352e-05,
      "loss": 1.6703,
      "num_input_tokens_seen": 45924400,
      "step": 79155
    },
    {
      "epoch": 11.790288948465893,
      "grad_norm": 15.5625,
      "learning_rate": 2.1563281256549385e-05,
      "loss": 1.248,
      "num_input_tokens_seen": 45927440,
      "step": 79160
    },
    {
      "epoch": 11.791033661006852,
      "grad_norm": 13.5,
      "learning_rate": 2.15600627138013e-05,
      "loss": 1.0241,
      "num_input_tokens_seen": 45930448,
      "step": 79165
    },
    {
      "epoch": 11.79177837354781,
      "grad_norm": 14.625,
      "learning_rate": 2.155684422916745e-05,
      "loss": 1.0508,
      "num_input_tokens_seen": 45933392,
      "step": 79170
    },
    {
      "epoch": 11.792523086088769,
      "grad_norm": 16.625,
      "learning_rate": 2.1553625802702226e-05,
      "loss": 0.9971,
      "num_input_tokens_seen": 45936144,
      "step": 79175
    },
    {
      "epoch": 11.79326779862973,
      "grad_norm": 7.6875,
      "learning_rate": 2.155040743445999e-05,
      "loss": 0.8004,
      "num_input_tokens_seen": 45938896,
      "step": 79180
    },
    {
      "epoch": 11.794012511170688,
      "grad_norm": 4.84375,
      "learning_rate": 2.1547189124495103e-05,
      "loss": 0.8113,
      "num_input_tokens_seen": 45941872,
      "step": 79185
    },
    {
      "epoch": 11.794757223711647,
      "grad_norm": 12.75,
      "learning_rate": 2.1543970872861957e-05,
      "loss": 1.2204,
      "num_input_tokens_seen": 45945104,
      "step": 79190
    },
    {
      "epoch": 11.795501936252606,
      "grad_norm": 10.625,
      "learning_rate": 2.15407526796149e-05,
      "loss": 1.1651,
      "num_input_tokens_seen": 45948240,
      "step": 79195
    },
    {
      "epoch": 11.796246648793566,
      "grad_norm": 12.4375,
      "learning_rate": 2.153753454480832e-05,
      "loss": 1.028,
      "num_input_tokens_seen": 45951408,
      "step": 79200
    },
    {
      "epoch": 11.796991361334525,
      "grad_norm": 12.8125,
      "learning_rate": 2.1534316468496575e-05,
      "loss": 1.223,
      "num_input_tokens_seen": 45954320,
      "step": 79205
    },
    {
      "epoch": 11.797736073875484,
      "grad_norm": 17.375,
      "learning_rate": 2.153109845073403e-05,
      "loss": 1.7218,
      "num_input_tokens_seen": 45957136,
      "step": 79210
    },
    {
      "epoch": 11.798480786416443,
      "grad_norm": 9.5,
      "learning_rate": 2.1527880491575042e-05,
      "loss": 0.8773,
      "num_input_tokens_seen": 45959792,
      "step": 79215
    },
    {
      "epoch": 11.799225498957403,
      "grad_norm": 7.65625,
      "learning_rate": 2.1524662591073997e-05,
      "loss": 0.93,
      "num_input_tokens_seen": 45962416,
      "step": 79220
    },
    {
      "epoch": 11.799970211498362,
      "grad_norm": 12.4375,
      "learning_rate": 2.1521444749285244e-05,
      "loss": 0.9692,
      "num_input_tokens_seen": 45965616,
      "step": 79225
    },
    {
      "epoch": 11.80071492403932,
      "grad_norm": 8.6875,
      "learning_rate": 2.1518226966263136e-05,
      "loss": 1.167,
      "num_input_tokens_seen": 45968304,
      "step": 79230
    },
    {
      "epoch": 11.80145963658028,
      "grad_norm": 15.3125,
      "learning_rate": 2.1515009242062055e-05,
      "loss": 1.3207,
      "num_input_tokens_seen": 45970928,
      "step": 79235
    },
    {
      "epoch": 11.80220434912124,
      "grad_norm": 5.03125,
      "learning_rate": 2.1511791576736346e-05,
      "loss": 1.1829,
      "num_input_tokens_seen": 45973904,
      "step": 79240
    },
    {
      "epoch": 11.802949061662199,
      "grad_norm": 7.46875,
      "learning_rate": 2.1508573970340377e-05,
      "loss": 0.8679,
      "num_input_tokens_seen": 45976560,
      "step": 79245
    },
    {
      "epoch": 11.803693774203158,
      "grad_norm": 12.375,
      "learning_rate": 2.1505356422928493e-05,
      "loss": 1.1558,
      "num_input_tokens_seen": 45979312,
      "step": 79250
    },
    {
      "epoch": 11.804438486744116,
      "grad_norm": 18.0,
      "learning_rate": 2.1502138934555072e-05,
      "loss": 1.4331,
      "num_input_tokens_seen": 45982224,
      "step": 79255
    },
    {
      "epoch": 11.805183199285075,
      "grad_norm": 15.1875,
      "learning_rate": 2.1498921505274444e-05,
      "loss": 0.8978,
      "num_input_tokens_seen": 45984944,
      "step": 79260
    },
    {
      "epoch": 11.805927911826036,
      "grad_norm": 9.6875,
      "learning_rate": 2.1495704135140992e-05,
      "loss": 0.8925,
      "num_input_tokens_seen": 45988144,
      "step": 79265
    },
    {
      "epoch": 11.806672624366994,
      "grad_norm": 16.0,
      "learning_rate": 2.1492486824209058e-05,
      "loss": 1.2613,
      "num_input_tokens_seen": 45990672,
      "step": 79270
    },
    {
      "epoch": 11.807417336907953,
      "grad_norm": 7.9375,
      "learning_rate": 2.1489269572532987e-05,
      "loss": 0.9854,
      "num_input_tokens_seen": 45993392,
      "step": 79275
    },
    {
      "epoch": 11.808162049448912,
      "grad_norm": 6.125,
      "learning_rate": 2.1486052380167146e-05,
      "loss": 1.2441,
      "num_input_tokens_seen": 45996368,
      "step": 79280
    },
    {
      "epoch": 11.808906761989872,
      "grad_norm": 5.5625,
      "learning_rate": 2.1482835247165867e-05,
      "loss": 1.1751,
      "num_input_tokens_seen": 45999536,
      "step": 79285
    },
    {
      "epoch": 11.809651474530831,
      "grad_norm": 13.25,
      "learning_rate": 2.1479618173583522e-05,
      "loss": 1.3472,
      "num_input_tokens_seen": 46002288,
      "step": 79290
    },
    {
      "epoch": 11.81039618707179,
      "grad_norm": 7.625,
      "learning_rate": 2.147640115947444e-05,
      "loss": 1.0594,
      "num_input_tokens_seen": 46005136,
      "step": 79295
    },
    {
      "epoch": 11.811140899612749,
      "grad_norm": 9.25,
      "learning_rate": 2.147318420489299e-05,
      "loss": 0.9426,
      "num_input_tokens_seen": 46008080,
      "step": 79300
    },
    {
      "epoch": 11.81188561215371,
      "grad_norm": 11.5625,
      "learning_rate": 2.1469967309893508e-05,
      "loss": 1.2192,
      "num_input_tokens_seen": 46011088,
      "step": 79305
    },
    {
      "epoch": 11.812630324694668,
      "grad_norm": 4.46875,
      "learning_rate": 2.1466750474530333e-05,
      "loss": 1.3833,
      "num_input_tokens_seen": 46013616,
      "step": 79310
    },
    {
      "epoch": 11.813375037235627,
      "grad_norm": 7.46875,
      "learning_rate": 2.1463533698857827e-05,
      "loss": 0.9412,
      "num_input_tokens_seen": 46016592,
      "step": 79315
    },
    {
      "epoch": 11.814119749776586,
      "grad_norm": 25.0,
      "learning_rate": 2.1460316982930313e-05,
      "loss": 1.4772,
      "num_input_tokens_seen": 46019344,
      "step": 79320
    },
    {
      "epoch": 11.814864462317546,
      "grad_norm": 9.8125,
      "learning_rate": 2.1457100326802155e-05,
      "loss": 1.5253,
      "num_input_tokens_seen": 46022672,
      "step": 79325
    },
    {
      "epoch": 11.815609174858505,
      "grad_norm": 7.15625,
      "learning_rate": 2.1453883730527677e-05,
      "loss": 0.8077,
      "num_input_tokens_seen": 46025520,
      "step": 79330
    },
    {
      "epoch": 11.816353887399464,
      "grad_norm": 10.5,
      "learning_rate": 2.145066719416124e-05,
      "loss": 1.2144,
      "num_input_tokens_seen": 46028592,
      "step": 79335
    },
    {
      "epoch": 11.817098599940422,
      "grad_norm": 14.6875,
      "learning_rate": 2.1447450717757167e-05,
      "loss": 1.2222,
      "num_input_tokens_seen": 46031824,
      "step": 79340
    },
    {
      "epoch": 11.817843312481383,
      "grad_norm": 13.625,
      "learning_rate": 2.144423430136981e-05,
      "loss": 0.9443,
      "num_input_tokens_seen": 46034480,
      "step": 79345
    },
    {
      "epoch": 11.818588025022342,
      "grad_norm": 11.5625,
      "learning_rate": 2.1441017945053497e-05,
      "loss": 1.1488,
      "num_input_tokens_seen": 46037424,
      "step": 79350
    },
    {
      "epoch": 11.8193327375633,
      "grad_norm": 13.0,
      "learning_rate": 2.143780164886256e-05,
      "loss": 1.0688,
      "num_input_tokens_seen": 46040432,
      "step": 79355
    },
    {
      "epoch": 11.82007745010426,
      "grad_norm": 9.8125,
      "learning_rate": 2.143458541285136e-05,
      "loss": 1.3584,
      "num_input_tokens_seen": 46043888,
      "step": 79360
    },
    {
      "epoch": 11.82082216264522,
      "grad_norm": 8.1875,
      "learning_rate": 2.1431369237074196e-05,
      "loss": 0.978,
      "num_input_tokens_seen": 46046704,
      "step": 79365
    },
    {
      "epoch": 11.821566875186178,
      "grad_norm": 4.96875,
      "learning_rate": 2.1428153121585438e-05,
      "loss": 0.9523,
      "num_input_tokens_seen": 46049808,
      "step": 79370
    },
    {
      "epoch": 11.822311587727137,
      "grad_norm": 11.75,
      "learning_rate": 2.1424937066439398e-05,
      "loss": 1.1618,
      "num_input_tokens_seen": 46052656,
      "step": 79375
    },
    {
      "epoch": 11.823056300268096,
      "grad_norm": 10.8125,
      "learning_rate": 2.1421721071690415e-05,
      "loss": 0.8631,
      "num_input_tokens_seen": 46055568,
      "step": 79380
    },
    {
      "epoch": 11.823801012809056,
      "grad_norm": 5.1875,
      "learning_rate": 2.141850513739282e-05,
      "loss": 0.8041,
      "num_input_tokens_seen": 46058288,
      "step": 79385
    },
    {
      "epoch": 11.824545725350015,
      "grad_norm": 9.5,
      "learning_rate": 2.1415289263600927e-05,
      "loss": 1.3155,
      "num_input_tokens_seen": 46061552,
      "step": 79390
    },
    {
      "epoch": 11.825290437890974,
      "grad_norm": 12.8125,
      "learning_rate": 2.1412073450369092e-05,
      "loss": 1.2143,
      "num_input_tokens_seen": 46064432,
      "step": 79395
    },
    {
      "epoch": 11.826035150431933,
      "grad_norm": 8.8125,
      "learning_rate": 2.1408857697751617e-05,
      "loss": 1.0446,
      "num_input_tokens_seen": 46067184,
      "step": 79400
    },
    {
      "epoch": 11.826779862972892,
      "grad_norm": 10.5625,
      "learning_rate": 2.1405642005802852e-05,
      "loss": 0.822,
      "num_input_tokens_seen": 46069968,
      "step": 79405
    },
    {
      "epoch": 11.827524575513852,
      "grad_norm": 10.0625,
      "learning_rate": 2.1402426374577107e-05,
      "loss": 0.9948,
      "num_input_tokens_seen": 46072880,
      "step": 79410
    },
    {
      "epoch": 11.82826928805481,
      "grad_norm": 13.1875,
      "learning_rate": 2.139921080412872e-05,
      "loss": 1.1929,
      "num_input_tokens_seen": 46075632,
      "step": 79415
    },
    {
      "epoch": 11.82901400059577,
      "grad_norm": 9.9375,
      "learning_rate": 2.1395995294511993e-05,
      "loss": 0.8863,
      "num_input_tokens_seen": 46078544,
      "step": 79420
    },
    {
      "epoch": 11.82975871313673,
      "grad_norm": 10.5625,
      "learning_rate": 2.1392779845781275e-05,
      "loss": 1.0537,
      "num_input_tokens_seen": 46081456,
      "step": 79425
    },
    {
      "epoch": 11.830503425677689,
      "grad_norm": 12.25,
      "learning_rate": 2.1389564457990875e-05,
      "loss": 1.2866,
      "num_input_tokens_seen": 46084528,
      "step": 79430
    },
    {
      "epoch": 11.831248138218648,
      "grad_norm": 8.8125,
      "learning_rate": 2.1386349131195103e-05,
      "loss": 0.8824,
      "num_input_tokens_seen": 46087280,
      "step": 79435
    },
    {
      "epoch": 11.831992850759606,
      "grad_norm": 10.0625,
      "learning_rate": 2.13831338654483e-05,
      "loss": 1.3036,
      "num_input_tokens_seen": 46090256,
      "step": 79440
    },
    {
      "epoch": 11.832737563300565,
      "grad_norm": 6.84375,
      "learning_rate": 2.1379918660804766e-05,
      "loss": 0.7579,
      "num_input_tokens_seen": 46093072,
      "step": 79445
    },
    {
      "epoch": 11.833482275841526,
      "grad_norm": 7.625,
      "learning_rate": 2.1376703517318837e-05,
      "loss": 1.1116,
      "num_input_tokens_seen": 46095984,
      "step": 79450
    },
    {
      "epoch": 11.834226988382484,
      "grad_norm": 5.28125,
      "learning_rate": 2.1373488435044804e-05,
      "loss": 1.2341,
      "num_input_tokens_seen": 46098928,
      "step": 79455
    },
    {
      "epoch": 11.834971700923443,
      "grad_norm": 11.25,
      "learning_rate": 2.1370273414037013e-05,
      "loss": 1.241,
      "num_input_tokens_seen": 46102032,
      "step": 79460
    },
    {
      "epoch": 11.835716413464402,
      "grad_norm": 10.125,
      "learning_rate": 2.1367058454349763e-05,
      "loss": 1.2778,
      "num_input_tokens_seen": 46104688,
      "step": 79465
    },
    {
      "epoch": 11.836461126005362,
      "grad_norm": 8.375,
      "learning_rate": 2.1363843556037365e-05,
      "loss": 0.9827,
      "num_input_tokens_seen": 46107664,
      "step": 79470
    },
    {
      "epoch": 11.837205838546321,
      "grad_norm": 9.625,
      "learning_rate": 2.136062871915413e-05,
      "loss": 1.0381,
      "num_input_tokens_seen": 46110480,
      "step": 79475
    },
    {
      "epoch": 11.83795055108728,
      "grad_norm": 11.375,
      "learning_rate": 2.1357413943754374e-05,
      "loss": 0.7124,
      "num_input_tokens_seen": 46113296,
      "step": 79480
    },
    {
      "epoch": 11.838695263628239,
      "grad_norm": 11.9375,
      "learning_rate": 2.1354199229892416e-05,
      "loss": 1.2179,
      "num_input_tokens_seen": 46116272,
      "step": 79485
    },
    {
      "epoch": 11.8394399761692,
      "grad_norm": 12.8125,
      "learning_rate": 2.1350984577622547e-05,
      "loss": 1.155,
      "num_input_tokens_seen": 46119120,
      "step": 79490
    },
    {
      "epoch": 11.840184688710158,
      "grad_norm": 6.90625,
      "learning_rate": 2.1347769986999088e-05,
      "loss": 1.1213,
      "num_input_tokens_seen": 46121872,
      "step": 79495
    },
    {
      "epoch": 11.840929401251117,
      "grad_norm": 10.125,
      "learning_rate": 2.1344555458076345e-05,
      "loss": 1.1314,
      "num_input_tokens_seen": 46124720,
      "step": 79500
    },
    {
      "epoch": 11.841674113792076,
      "grad_norm": 11.625,
      "learning_rate": 2.1341340990908627e-05,
      "loss": 0.9976,
      "num_input_tokens_seen": 46127856,
      "step": 79505
    },
    {
      "epoch": 11.842418826333036,
      "grad_norm": 13.375,
      "learning_rate": 2.133812658555023e-05,
      "loss": 0.9125,
      "num_input_tokens_seen": 46130800,
      "step": 79510
    },
    {
      "epoch": 11.843163538873995,
      "grad_norm": 16.5,
      "learning_rate": 2.1334912242055454e-05,
      "loss": 1.2372,
      "num_input_tokens_seen": 46133584,
      "step": 79515
    },
    {
      "epoch": 11.843908251414954,
      "grad_norm": 8.0,
      "learning_rate": 2.1331697960478624e-05,
      "loss": 1.1785,
      "num_input_tokens_seen": 46136304,
      "step": 79520
    },
    {
      "epoch": 11.844652963955912,
      "grad_norm": 7.03125,
      "learning_rate": 2.1328483740874014e-05,
      "loss": 0.9912,
      "num_input_tokens_seen": 46139376,
      "step": 79525
    },
    {
      "epoch": 11.845397676496873,
      "grad_norm": 21.375,
      "learning_rate": 2.1325269583295953e-05,
      "loss": 1.1697,
      "num_input_tokens_seen": 46142224,
      "step": 79530
    },
    {
      "epoch": 11.846142389037832,
      "grad_norm": 17.75,
      "learning_rate": 2.132205548779872e-05,
      "loss": 1.165,
      "num_input_tokens_seen": 46145008,
      "step": 79535
    },
    {
      "epoch": 11.84688710157879,
      "grad_norm": 13.3125,
      "learning_rate": 2.131884145443663e-05,
      "loss": 1.2724,
      "num_input_tokens_seen": 46147888,
      "step": 79540
    },
    {
      "epoch": 11.84763181411975,
      "grad_norm": 7.65625,
      "learning_rate": 2.131562748326397e-05,
      "loss": 0.9143,
      "num_input_tokens_seen": 46150832,
      "step": 79545
    },
    {
      "epoch": 11.84837652666071,
      "grad_norm": 7.65625,
      "learning_rate": 2.131241357433503e-05,
      "loss": 1.0516,
      "num_input_tokens_seen": 46153712,
      "step": 79550
    },
    {
      "epoch": 11.849121239201668,
      "grad_norm": 13.9375,
      "learning_rate": 2.1309199727704125e-05,
      "loss": 1.0133,
      "num_input_tokens_seen": 46156400,
      "step": 79555
    },
    {
      "epoch": 11.849865951742627,
      "grad_norm": 9.0625,
      "learning_rate": 2.130598594342553e-05,
      "loss": 1.0481,
      "num_input_tokens_seen": 46159280,
      "step": 79560
    },
    {
      "epoch": 11.850610664283586,
      "grad_norm": 6.8125,
      "learning_rate": 2.130277222155355e-05,
      "loss": 0.9743,
      "num_input_tokens_seen": 46162384,
      "step": 79565
    },
    {
      "epoch": 11.851355376824547,
      "grad_norm": 9.9375,
      "learning_rate": 2.129955856214248e-05,
      "loss": 1.0309,
      "num_input_tokens_seen": 46165392,
      "step": 79570
    },
    {
      "epoch": 11.852100089365505,
      "grad_norm": 6.59375,
      "learning_rate": 2.129634496524661e-05,
      "loss": 1.3618,
      "num_input_tokens_seen": 46168368,
      "step": 79575
    },
    {
      "epoch": 11.852844801906464,
      "grad_norm": 12.5625,
      "learning_rate": 2.1293131430920215e-05,
      "loss": 1.1809,
      "num_input_tokens_seen": 46171248,
      "step": 79580
    },
    {
      "epoch": 11.853589514447423,
      "grad_norm": 16.75,
      "learning_rate": 2.128991795921761e-05,
      "loss": 1.2968,
      "num_input_tokens_seen": 46174128,
      "step": 79585
    },
    {
      "epoch": 11.854334226988382,
      "grad_norm": 13.75,
      "learning_rate": 2.128670455019307e-05,
      "loss": 1.3084,
      "num_input_tokens_seen": 46177008,
      "step": 79590
    },
    {
      "epoch": 11.855078939529342,
      "grad_norm": 10.0,
      "learning_rate": 2.128349120390087e-05,
      "loss": 1.1401,
      "num_input_tokens_seen": 46179856,
      "step": 79595
    },
    {
      "epoch": 11.8558236520703,
      "grad_norm": 11.875,
      "learning_rate": 2.1280277920395322e-05,
      "loss": 1.0321,
      "num_input_tokens_seen": 46183024,
      "step": 79600
    },
    {
      "epoch": 11.85656836461126,
      "grad_norm": 5.6875,
      "learning_rate": 2.1277064699730694e-05,
      "loss": 0.8154,
      "num_input_tokens_seen": 46186320,
      "step": 79605
    },
    {
      "epoch": 11.85731307715222,
      "grad_norm": 13.5625,
      "learning_rate": 2.1273851541961274e-05,
      "loss": 1.2814,
      "num_input_tokens_seen": 46189296,
      "step": 79610
    },
    {
      "epoch": 11.858057789693179,
      "grad_norm": 15.8125,
      "learning_rate": 2.1270638447141337e-05,
      "loss": 0.9707,
      "num_input_tokens_seen": 46192432,
      "step": 79615
    },
    {
      "epoch": 11.858802502234138,
      "grad_norm": 6.75,
      "learning_rate": 2.1267425415325185e-05,
      "loss": 1.0988,
      "num_input_tokens_seen": 46195248,
      "step": 79620
    },
    {
      "epoch": 11.859547214775096,
      "grad_norm": 8.0625,
      "learning_rate": 2.1264212446567084e-05,
      "loss": 1.0596,
      "num_input_tokens_seen": 46198224,
      "step": 79625
    },
    {
      "epoch": 11.860291927316055,
      "grad_norm": 9.25,
      "learning_rate": 2.1260999540921307e-05,
      "loss": 1.0915,
      "num_input_tokens_seen": 46201040,
      "step": 79630
    },
    {
      "epoch": 11.861036639857016,
      "grad_norm": 13.75,
      "learning_rate": 2.1257786698442155e-05,
      "loss": 1.1991,
      "num_input_tokens_seen": 46203952,
      "step": 79635
    },
    {
      "epoch": 11.861781352397974,
      "grad_norm": 8.3125,
      "learning_rate": 2.125457391918389e-05,
      "loss": 1.0307,
      "num_input_tokens_seen": 46207024,
      "step": 79640
    },
    {
      "epoch": 11.862526064938933,
      "grad_norm": 8.4375,
      "learning_rate": 2.1251361203200793e-05,
      "loss": 1.2391,
      "num_input_tokens_seen": 46210096,
      "step": 79645
    },
    {
      "epoch": 11.863270777479892,
      "grad_norm": 5.5,
      "learning_rate": 2.124814855054713e-05,
      "loss": 1.0117,
      "num_input_tokens_seen": 46213008,
      "step": 79650
    },
    {
      "epoch": 11.864015490020853,
      "grad_norm": 4.6875,
      "learning_rate": 2.1244935961277197e-05,
      "loss": 0.8598,
      "num_input_tokens_seen": 46215792,
      "step": 79655
    },
    {
      "epoch": 11.864760202561811,
      "grad_norm": 14.8125,
      "learning_rate": 2.124172343544524e-05,
      "loss": 1.2512,
      "num_input_tokens_seen": 46218800,
      "step": 79660
    },
    {
      "epoch": 11.86550491510277,
      "grad_norm": 21.75,
      "learning_rate": 2.123851097310556e-05,
      "loss": 1.11,
      "num_input_tokens_seen": 46221680,
      "step": 79665
    },
    {
      "epoch": 11.866249627643729,
      "grad_norm": 6.34375,
      "learning_rate": 2.1235298574312405e-05,
      "loss": 1.1442,
      "num_input_tokens_seen": 46224656,
      "step": 79670
    },
    {
      "epoch": 11.86699434018469,
      "grad_norm": 12.5,
      "learning_rate": 2.123208623912006e-05,
      "loss": 1.0255,
      "num_input_tokens_seen": 46227728,
      "step": 79675
    },
    {
      "epoch": 11.867739052725648,
      "grad_norm": 13.5625,
      "learning_rate": 2.1228873967582787e-05,
      "loss": 1.5969,
      "num_input_tokens_seen": 46230704,
      "step": 79680
    },
    {
      "epoch": 11.868483765266607,
      "grad_norm": 12.5,
      "learning_rate": 2.1225661759754848e-05,
      "loss": 1.0248,
      "num_input_tokens_seen": 46233712,
      "step": 79685
    },
    {
      "epoch": 11.869228477807566,
      "grad_norm": 7.0625,
      "learning_rate": 2.1222449615690525e-05,
      "loss": 1.2422,
      "num_input_tokens_seen": 46236816,
      "step": 79690
    },
    {
      "epoch": 11.869973190348526,
      "grad_norm": 13.25,
      "learning_rate": 2.121923753544407e-05,
      "loss": 1.4714,
      "num_input_tokens_seen": 46239824,
      "step": 79695
    },
    {
      "epoch": 11.870717902889485,
      "grad_norm": 13.25,
      "learning_rate": 2.1216025519069766e-05,
      "loss": 1.5192,
      "num_input_tokens_seen": 46242672,
      "step": 79700
    },
    {
      "epoch": 11.871462615430444,
      "grad_norm": 6.34375,
      "learning_rate": 2.121281356662186e-05,
      "loss": 1.2001,
      "num_input_tokens_seen": 46245360,
      "step": 79705
    },
    {
      "epoch": 11.872207327971402,
      "grad_norm": 11.875,
      "learning_rate": 2.1209601678154615e-05,
      "loss": 1.1015,
      "num_input_tokens_seen": 46248208,
      "step": 79710
    },
    {
      "epoch": 11.872952040512363,
      "grad_norm": 9.125,
      "learning_rate": 2.1206389853722306e-05,
      "loss": 1.0915,
      "num_input_tokens_seen": 46250928,
      "step": 79715
    },
    {
      "epoch": 11.873696753053322,
      "grad_norm": 6.09375,
      "learning_rate": 2.1203178093379172e-05,
      "loss": 1.0936,
      "num_input_tokens_seen": 46254000,
      "step": 79720
    },
    {
      "epoch": 11.87444146559428,
      "grad_norm": 13.8125,
      "learning_rate": 2.1199966397179492e-05,
      "loss": 1.3027,
      "num_input_tokens_seen": 46257168,
      "step": 79725
    },
    {
      "epoch": 11.87518617813524,
      "grad_norm": 13.4375,
      "learning_rate": 2.1196754765177514e-05,
      "loss": 1.1039,
      "num_input_tokens_seen": 46260112,
      "step": 79730
    },
    {
      "epoch": 11.8759308906762,
      "grad_norm": 14.75,
      "learning_rate": 2.1193543197427507e-05,
      "loss": 0.8041,
      "num_input_tokens_seen": 46263120,
      "step": 79735
    },
    {
      "epoch": 11.876675603217159,
      "grad_norm": 10.1875,
      "learning_rate": 2.119033169398371e-05,
      "loss": 0.962,
      "num_input_tokens_seen": 46266128,
      "step": 79740
    },
    {
      "epoch": 11.877420315758117,
      "grad_norm": 4.9375,
      "learning_rate": 2.1187120254900397e-05,
      "loss": 1.0497,
      "num_input_tokens_seen": 46268816,
      "step": 79745
    },
    {
      "epoch": 11.878165028299076,
      "grad_norm": 13.5625,
      "learning_rate": 2.118390888023181e-05,
      "loss": 1.139,
      "num_input_tokens_seen": 46271568,
      "step": 79750
    },
    {
      "epoch": 11.878909740840037,
      "grad_norm": 22.375,
      "learning_rate": 2.1180697570032195e-05,
      "loss": 0.9546,
      "num_input_tokens_seen": 46274288,
      "step": 79755
    },
    {
      "epoch": 11.879654453380995,
      "grad_norm": 7.875,
      "learning_rate": 2.117748632435582e-05,
      "loss": 1.1957,
      "num_input_tokens_seen": 46276880,
      "step": 79760
    },
    {
      "epoch": 11.880399165921954,
      "grad_norm": 8.0,
      "learning_rate": 2.1174275143256927e-05,
      "loss": 0.8595,
      "num_input_tokens_seen": 46279632,
      "step": 79765
    },
    {
      "epoch": 11.881143878462913,
      "grad_norm": 9.0,
      "learning_rate": 2.1171064026789768e-05,
      "loss": 1.0636,
      "num_input_tokens_seen": 46282608,
      "step": 79770
    },
    {
      "epoch": 11.881888591003872,
      "grad_norm": 11.0,
      "learning_rate": 2.1167852975008587e-05,
      "loss": 1.2821,
      "num_input_tokens_seen": 46285296,
      "step": 79775
    },
    {
      "epoch": 11.882633303544832,
      "grad_norm": 7.09375,
      "learning_rate": 2.1164641987967638e-05,
      "loss": 1.0184,
      "num_input_tokens_seen": 46288048,
      "step": 79780
    },
    {
      "epoch": 11.883378016085791,
      "grad_norm": 13.875,
      "learning_rate": 2.116143106572117e-05,
      "loss": 1.2786,
      "num_input_tokens_seen": 46290960,
      "step": 79785
    },
    {
      "epoch": 11.88412272862675,
      "grad_norm": 11.0625,
      "learning_rate": 2.115822020832341e-05,
      "loss": 0.7799,
      "num_input_tokens_seen": 46293872,
      "step": 79790
    },
    {
      "epoch": 11.884867441167708,
      "grad_norm": 8.75,
      "learning_rate": 2.1155009415828628e-05,
      "loss": 1.2449,
      "num_input_tokens_seen": 46296624,
      "step": 79795
    },
    {
      "epoch": 11.885612153708669,
      "grad_norm": 8.625,
      "learning_rate": 2.1151798688291046e-05,
      "loss": 1.1937,
      "num_input_tokens_seen": 46299984,
      "step": 79800
    },
    {
      "epoch": 11.886356866249628,
      "grad_norm": 13.6875,
      "learning_rate": 2.1148588025764916e-05,
      "loss": 1.071,
      "num_input_tokens_seen": 46302928,
      "step": 79805
    },
    {
      "epoch": 11.887101578790586,
      "grad_norm": 11.875,
      "learning_rate": 2.1145377428304476e-05,
      "loss": 0.984,
      "num_input_tokens_seen": 46305840,
      "step": 79810
    },
    {
      "epoch": 11.887846291331545,
      "grad_norm": 13.125,
      "learning_rate": 2.1142166895963973e-05,
      "loss": 0.9066,
      "num_input_tokens_seen": 46308432,
      "step": 79815
    },
    {
      "epoch": 11.888591003872506,
      "grad_norm": 8.0625,
      "learning_rate": 2.1138956428797624e-05,
      "loss": 1.1548,
      "num_input_tokens_seen": 46311344,
      "step": 79820
    },
    {
      "epoch": 11.889335716413465,
      "grad_norm": 10.375,
      "learning_rate": 2.1135746026859697e-05,
      "loss": 0.8976,
      "num_input_tokens_seen": 46314032,
      "step": 79825
    },
    {
      "epoch": 11.890080428954423,
      "grad_norm": 14.375,
      "learning_rate": 2.1132535690204415e-05,
      "loss": 1.305,
      "num_input_tokens_seen": 46316976,
      "step": 79830
    },
    {
      "epoch": 11.890825141495382,
      "grad_norm": 15.625,
      "learning_rate": 2.1129325418886e-05,
      "loss": 1.0842,
      "num_input_tokens_seen": 46319792,
      "step": 79835
    },
    {
      "epoch": 11.891569854036343,
      "grad_norm": 11.4375,
      "learning_rate": 2.1126115212958708e-05,
      "loss": 1.2161,
      "num_input_tokens_seen": 46322704,
      "step": 79840
    },
    {
      "epoch": 11.892314566577301,
      "grad_norm": 9.4375,
      "learning_rate": 2.112290507247675e-05,
      "loss": 1.0336,
      "num_input_tokens_seen": 46325520,
      "step": 79845
    },
    {
      "epoch": 11.89305927911826,
      "grad_norm": 5.5,
      "learning_rate": 2.1119694997494382e-05,
      "loss": 1.3727,
      "num_input_tokens_seen": 46328176,
      "step": 79850
    },
    {
      "epoch": 11.893803991659219,
      "grad_norm": 7.5,
      "learning_rate": 2.1116484988065813e-05,
      "loss": 1.4605,
      "num_input_tokens_seen": 46331344,
      "step": 79855
    },
    {
      "epoch": 11.89454870420018,
      "grad_norm": 8.125,
      "learning_rate": 2.1113275044245293e-05,
      "loss": 0.8913,
      "num_input_tokens_seen": 46334096,
      "step": 79860
    },
    {
      "epoch": 11.895293416741138,
      "grad_norm": 10.5,
      "learning_rate": 2.1110065166087037e-05,
      "loss": 0.9625,
      "num_input_tokens_seen": 46336816,
      "step": 79865
    },
    {
      "epoch": 11.896038129282097,
      "grad_norm": 6.75,
      "learning_rate": 2.110685535364528e-05,
      "loss": 1.4065,
      "num_input_tokens_seen": 46339696,
      "step": 79870
    },
    {
      "epoch": 11.896782841823056,
      "grad_norm": 11.9375,
      "learning_rate": 2.1103645606974244e-05,
      "loss": 1.0014,
      "num_input_tokens_seen": 46342480,
      "step": 79875
    },
    {
      "epoch": 11.897527554364016,
      "grad_norm": 5.78125,
      "learning_rate": 2.1100435926128146e-05,
      "loss": 0.7058,
      "num_input_tokens_seen": 46345200,
      "step": 79880
    },
    {
      "epoch": 11.898272266904975,
      "grad_norm": 9.1875,
      "learning_rate": 2.1097226311161232e-05,
      "loss": 1.2087,
      "num_input_tokens_seen": 46347856,
      "step": 79885
    },
    {
      "epoch": 11.899016979445934,
      "grad_norm": 8.5625,
      "learning_rate": 2.1094016762127698e-05,
      "loss": 1.3199,
      "num_input_tokens_seen": 46350704,
      "step": 79890
    },
    {
      "epoch": 11.899761691986892,
      "grad_norm": 12.6875,
      "learning_rate": 2.10908072790818e-05,
      "loss": 1.3787,
      "num_input_tokens_seen": 46353648,
      "step": 79895
    },
    {
      "epoch": 11.900506404527853,
      "grad_norm": 10.5,
      "learning_rate": 2.1087597862077726e-05,
      "loss": 1.0174,
      "num_input_tokens_seen": 46356336,
      "step": 79900
    },
    {
      "epoch": 11.901251117068812,
      "grad_norm": 4.96875,
      "learning_rate": 2.1084388511169718e-05,
      "loss": 0.8783,
      "num_input_tokens_seen": 46360528,
      "step": 79905
    },
    {
      "epoch": 11.90199582960977,
      "grad_norm": 8.0,
      "learning_rate": 2.1081179226411985e-05,
      "loss": 1.1811,
      "num_input_tokens_seen": 46363824,
      "step": 79910
    },
    {
      "epoch": 11.90274054215073,
      "grad_norm": 8.1875,
      "learning_rate": 2.107797000785874e-05,
      "loss": 1.2278,
      "num_input_tokens_seen": 46366704,
      "step": 79915
    },
    {
      "epoch": 11.90348525469169,
      "grad_norm": 11.125,
      "learning_rate": 2.107476085556421e-05,
      "loss": 1.0482,
      "num_input_tokens_seen": 46369712,
      "step": 79920
    },
    {
      "epoch": 11.904229967232649,
      "grad_norm": 6.03125,
      "learning_rate": 2.10715517695826e-05,
      "loss": 0.9965,
      "num_input_tokens_seen": 46372592,
      "step": 79925
    },
    {
      "epoch": 11.904974679773607,
      "grad_norm": 11.6875,
      "learning_rate": 2.106834274996814e-05,
      "loss": 0.8491,
      "num_input_tokens_seen": 46375504,
      "step": 79930
    },
    {
      "epoch": 11.905719392314566,
      "grad_norm": 10.9375,
      "learning_rate": 2.1065133796775026e-05,
      "loss": 1.1361,
      "num_input_tokens_seen": 46378320,
      "step": 79935
    },
    {
      "epoch": 11.906464104855527,
      "grad_norm": 9.875,
      "learning_rate": 2.1061924910057485e-05,
      "loss": 1.0272,
      "num_input_tokens_seen": 46381296,
      "step": 79940
    },
    {
      "epoch": 11.907208817396485,
      "grad_norm": 12.0625,
      "learning_rate": 2.1058716089869707e-05,
      "loss": 1.3212,
      "num_input_tokens_seen": 46384048,
      "step": 79945
    },
    {
      "epoch": 11.907953529937444,
      "grad_norm": 11.375,
      "learning_rate": 2.1055507336265925e-05,
      "loss": 1.1869,
      "num_input_tokens_seen": 46387152,
      "step": 79950
    },
    {
      "epoch": 11.908698242478403,
      "grad_norm": 12.8125,
      "learning_rate": 2.105229864930034e-05,
      "loss": 1.3321,
      "num_input_tokens_seen": 46389936,
      "step": 79955
    },
    {
      "epoch": 11.909442955019362,
      "grad_norm": 14.1875,
      "learning_rate": 2.1049090029027146e-05,
      "loss": 0.9679,
      "num_input_tokens_seen": 46392752,
      "step": 79960
    },
    {
      "epoch": 11.910187667560322,
      "grad_norm": 9.875,
      "learning_rate": 2.104588147550057e-05,
      "loss": 1.0857,
      "num_input_tokens_seen": 46395824,
      "step": 79965
    },
    {
      "epoch": 11.910932380101281,
      "grad_norm": 8.25,
      "learning_rate": 2.1042672988774805e-05,
      "loss": 1.3423,
      "num_input_tokens_seen": 46399024,
      "step": 79970
    },
    {
      "epoch": 11.91167709264224,
      "grad_norm": 20.25,
      "learning_rate": 2.103946456890406e-05,
      "loss": 1.1485,
      "num_input_tokens_seen": 46401904,
      "step": 79975
    },
    {
      "epoch": 11.912421805183198,
      "grad_norm": 23.25,
      "learning_rate": 2.1036256215942526e-05,
      "loss": 1.2202,
      "num_input_tokens_seen": 46404688,
      "step": 79980
    },
    {
      "epoch": 11.913166517724159,
      "grad_norm": 12.0625,
      "learning_rate": 2.1033047929944427e-05,
      "loss": 1.1125,
      "num_input_tokens_seen": 46407312,
      "step": 79985
    },
    {
      "epoch": 11.913911230265118,
      "grad_norm": 4.65625,
      "learning_rate": 2.102983971096395e-05,
      "loss": 0.9053,
      "num_input_tokens_seen": 46410352,
      "step": 79990
    },
    {
      "epoch": 11.914655942806077,
      "grad_norm": 15.0625,
      "learning_rate": 2.1026631559055285e-05,
      "loss": 1.1713,
      "num_input_tokens_seen": 46413296,
      "step": 79995
    },
    {
      "epoch": 11.915400655347035,
      "grad_norm": 5.71875,
      "learning_rate": 2.1023423474272652e-05,
      "loss": 0.9935,
      "num_input_tokens_seen": 46416176,
      "step": 80000
    },
    {
      "epoch": 11.916145367887996,
      "grad_norm": 10.75,
      "learning_rate": 2.1020215456670234e-05,
      "loss": 0.9213,
      "num_input_tokens_seen": 46418992,
      "step": 80005
    },
    {
      "epoch": 11.916890080428955,
      "grad_norm": 7.46875,
      "learning_rate": 2.1017007506302233e-05,
      "loss": 0.9428,
      "num_input_tokens_seen": 46421712,
      "step": 80010
    },
    {
      "epoch": 11.917634792969913,
      "grad_norm": 16.875,
      "learning_rate": 2.1013799623222833e-05,
      "loss": 0.9298,
      "num_input_tokens_seen": 46424368,
      "step": 80015
    },
    {
      "epoch": 11.918379505510872,
      "grad_norm": 7.21875,
      "learning_rate": 2.1010591807486253e-05,
      "loss": 1.2758,
      "num_input_tokens_seen": 46427088,
      "step": 80020
    },
    {
      "epoch": 11.919124218051833,
      "grad_norm": 7.28125,
      "learning_rate": 2.100738405914665e-05,
      "loss": 0.9137,
      "num_input_tokens_seen": 46429904,
      "step": 80025
    },
    {
      "epoch": 11.919868930592791,
      "grad_norm": 9.0,
      "learning_rate": 2.1004176378258252e-05,
      "loss": 1.0613,
      "num_input_tokens_seen": 46432528,
      "step": 80030
    },
    {
      "epoch": 11.92061364313375,
      "grad_norm": 14.25,
      "learning_rate": 2.100096876487523e-05,
      "loss": 1.0988,
      "num_input_tokens_seen": 46435568,
      "step": 80035
    },
    {
      "epoch": 11.921358355674709,
      "grad_norm": 10.5625,
      "learning_rate": 2.0997761219051777e-05,
      "loss": 0.9432,
      "num_input_tokens_seen": 46438512,
      "step": 80040
    },
    {
      "epoch": 11.92210306821567,
      "grad_norm": 7.40625,
      "learning_rate": 2.099455374084208e-05,
      "loss": 0.9747,
      "num_input_tokens_seen": 46441456,
      "step": 80045
    },
    {
      "epoch": 11.922847780756628,
      "grad_norm": 5.9375,
      "learning_rate": 2.0991346330300314e-05,
      "loss": 1.0223,
      "num_input_tokens_seen": 46444432,
      "step": 80050
    },
    {
      "epoch": 11.923592493297587,
      "grad_norm": 15.375,
      "learning_rate": 2.0988138987480694e-05,
      "loss": 1.4241,
      "num_input_tokens_seen": 46447216,
      "step": 80055
    },
    {
      "epoch": 11.924337205838546,
      "grad_norm": 10.25,
      "learning_rate": 2.0984931712437377e-05,
      "loss": 1.1241,
      "num_input_tokens_seen": 46449872,
      "step": 80060
    },
    {
      "epoch": 11.925081918379506,
      "grad_norm": 5.0,
      "learning_rate": 2.0981724505224563e-05,
      "loss": 0.873,
      "num_input_tokens_seen": 46452976,
      "step": 80065
    },
    {
      "epoch": 11.925826630920465,
      "grad_norm": 11.5,
      "learning_rate": 2.0978517365896433e-05,
      "loss": 0.9129,
      "num_input_tokens_seen": 46455856,
      "step": 80070
    },
    {
      "epoch": 11.926571343461424,
      "grad_norm": 6.875,
      "learning_rate": 2.0975310294507162e-05,
      "loss": 0.9553,
      "num_input_tokens_seen": 46458576,
      "step": 80075
    },
    {
      "epoch": 11.927316056002383,
      "grad_norm": 13.1875,
      "learning_rate": 2.0972103291110933e-05,
      "loss": 1.3311,
      "num_input_tokens_seen": 46461488,
      "step": 80080
    },
    {
      "epoch": 11.928060768543343,
      "grad_norm": 11.125,
      "learning_rate": 2.096889635576192e-05,
      "loss": 0.6911,
      "num_input_tokens_seen": 46464208,
      "step": 80085
    },
    {
      "epoch": 11.928805481084302,
      "grad_norm": 6.0625,
      "learning_rate": 2.0965689488514314e-05,
      "loss": 1.1866,
      "num_input_tokens_seen": 46466992,
      "step": 80090
    },
    {
      "epoch": 11.92955019362526,
      "grad_norm": 18.875,
      "learning_rate": 2.0962482689422276e-05,
      "loss": 1.2145,
      "num_input_tokens_seen": 46470096,
      "step": 80095
    },
    {
      "epoch": 11.93029490616622,
      "grad_norm": 9.125,
      "learning_rate": 2.0959275958539996e-05,
      "loss": 1.3334,
      "num_input_tokens_seen": 46472880,
      "step": 80100
    },
    {
      "epoch": 11.931039618707178,
      "grad_norm": 12.8125,
      "learning_rate": 2.095606929592164e-05,
      "loss": 0.9202,
      "num_input_tokens_seen": 46475600,
      "step": 80105
    },
    {
      "epoch": 11.931784331248139,
      "grad_norm": 14.9375,
      "learning_rate": 2.0952862701621385e-05,
      "loss": 1.1685,
      "num_input_tokens_seen": 46478512,
      "step": 80110
    },
    {
      "epoch": 11.932529043789097,
      "grad_norm": 15.3125,
      "learning_rate": 2.09496561756934e-05,
      "loss": 1.5417,
      "num_input_tokens_seen": 46481520,
      "step": 80115
    },
    {
      "epoch": 11.933273756330056,
      "grad_norm": 7.625,
      "learning_rate": 2.094644971819185e-05,
      "loss": 1.2051,
      "num_input_tokens_seen": 46484208,
      "step": 80120
    },
    {
      "epoch": 11.934018468871017,
      "grad_norm": 7.0625,
      "learning_rate": 2.0943243329170922e-05,
      "loss": 0.8115,
      "num_input_tokens_seen": 46487248,
      "step": 80125
    },
    {
      "epoch": 11.934763181411975,
      "grad_norm": 9.1875,
      "learning_rate": 2.0940037008684772e-05,
      "loss": 1.011,
      "num_input_tokens_seen": 46490192,
      "step": 80130
    },
    {
      "epoch": 11.935507893952934,
      "grad_norm": 17.875,
      "learning_rate": 2.0936830756787568e-05,
      "loss": 1.0238,
      "num_input_tokens_seen": 46492816,
      "step": 80135
    },
    {
      "epoch": 11.936252606493893,
      "grad_norm": 11.125,
      "learning_rate": 2.0933624573533477e-05,
      "loss": 0.8189,
      "num_input_tokens_seen": 46495856,
      "step": 80140
    },
    {
      "epoch": 11.936997319034852,
      "grad_norm": 9.125,
      "learning_rate": 2.0930418458976676e-05,
      "loss": 1.2351,
      "num_input_tokens_seen": 46498864,
      "step": 80145
    },
    {
      "epoch": 11.937742031575812,
      "grad_norm": 16.75,
      "learning_rate": 2.0927212413171316e-05,
      "loss": 1.1812,
      "num_input_tokens_seen": 46501520,
      "step": 80150
    },
    {
      "epoch": 11.938486744116771,
      "grad_norm": 7.71875,
      "learning_rate": 2.092400643617155e-05,
      "loss": 0.7038,
      "num_input_tokens_seen": 46504432,
      "step": 80155
    },
    {
      "epoch": 11.93923145665773,
      "grad_norm": 16.375,
      "learning_rate": 2.092080052803157e-05,
      "loss": 1.4016,
      "num_input_tokens_seen": 46507120,
      "step": 80160
    },
    {
      "epoch": 11.939976169198689,
      "grad_norm": 9.0,
      "learning_rate": 2.0917594688805507e-05,
      "loss": 0.8372,
      "num_input_tokens_seen": 46509872,
      "step": 80165
    },
    {
      "epoch": 11.940720881739649,
      "grad_norm": 6.6875,
      "learning_rate": 2.091438891854754e-05,
      "loss": 0.8606,
      "num_input_tokens_seen": 46512624,
      "step": 80170
    },
    {
      "epoch": 11.941465594280608,
      "grad_norm": 3.859375,
      "learning_rate": 2.091118321731181e-05,
      "loss": 0.745,
      "num_input_tokens_seen": 46515952,
      "step": 80175
    },
    {
      "epoch": 11.942210306821567,
      "grad_norm": 11.5625,
      "learning_rate": 2.0907977585152495e-05,
      "loss": 0.911,
      "num_input_tokens_seen": 46519120,
      "step": 80180
    },
    {
      "epoch": 11.942955019362525,
      "grad_norm": 11.0,
      "learning_rate": 2.0904772022123725e-05,
      "loss": 0.9747,
      "num_input_tokens_seen": 46521904,
      "step": 80185
    },
    {
      "epoch": 11.943699731903486,
      "grad_norm": 9.8125,
      "learning_rate": 2.0901566528279687e-05,
      "loss": 1.3622,
      "num_input_tokens_seen": 46524848,
      "step": 80190
    },
    {
      "epoch": 11.944444444444445,
      "grad_norm": 17.0,
      "learning_rate": 2.089836110367451e-05,
      "loss": 1.3955,
      "num_input_tokens_seen": 46527984,
      "step": 80195
    },
    {
      "epoch": 11.945189156985403,
      "grad_norm": 8.6875,
      "learning_rate": 2.0895155748362353e-05,
      "loss": 0.9779,
      "num_input_tokens_seen": 46530704,
      "step": 80200
    },
    {
      "epoch": 11.945933869526362,
      "grad_norm": 8.6875,
      "learning_rate": 2.0891950462397372e-05,
      "loss": 1.141,
      "num_input_tokens_seen": 46533712,
      "step": 80205
    },
    {
      "epoch": 11.946678582067323,
      "grad_norm": 13.3125,
      "learning_rate": 2.0888745245833703e-05,
      "loss": 1.0427,
      "num_input_tokens_seen": 46536528,
      "step": 80210
    },
    {
      "epoch": 11.947423294608281,
      "grad_norm": 8.25,
      "learning_rate": 2.0885540098725513e-05,
      "loss": 1.0848,
      "num_input_tokens_seen": 46539408,
      "step": 80215
    },
    {
      "epoch": 11.94816800714924,
      "grad_norm": 8.0625,
      "learning_rate": 2.088233502112693e-05,
      "loss": 1.0378,
      "num_input_tokens_seen": 46542288,
      "step": 80220
    },
    {
      "epoch": 11.948912719690199,
      "grad_norm": 8.5,
      "learning_rate": 2.0879130013092124e-05,
      "loss": 1.0487,
      "num_input_tokens_seen": 46545104,
      "step": 80225
    },
    {
      "epoch": 11.94965743223116,
      "grad_norm": 10.125,
      "learning_rate": 2.087592507467523e-05,
      "loss": 1.0131,
      "num_input_tokens_seen": 46548400,
      "step": 80230
    },
    {
      "epoch": 11.950402144772118,
      "grad_norm": 12.375,
      "learning_rate": 2.087272020593038e-05,
      "loss": 0.9424,
      "num_input_tokens_seen": 46551216,
      "step": 80235
    },
    {
      "epoch": 11.951146857313077,
      "grad_norm": 12.125,
      "learning_rate": 2.086951540691174e-05,
      "loss": 1.3205,
      "num_input_tokens_seen": 46554480,
      "step": 80240
    },
    {
      "epoch": 11.951891569854036,
      "grad_norm": 10.0,
      "learning_rate": 2.086631067767342e-05,
      "loss": 1.3005,
      "num_input_tokens_seen": 46557200,
      "step": 80245
    },
    {
      "epoch": 11.952636282394996,
      "grad_norm": 10.6875,
      "learning_rate": 2.0863106018269596e-05,
      "loss": 1.368,
      "num_input_tokens_seen": 46560304,
      "step": 80250
    },
    {
      "epoch": 11.953380994935955,
      "grad_norm": 8.75,
      "learning_rate": 2.085990142875438e-05,
      "loss": 0.823,
      "num_input_tokens_seen": 46563152,
      "step": 80255
    },
    {
      "epoch": 11.954125707476914,
      "grad_norm": 8.625,
      "learning_rate": 2.0856696909181932e-05,
      "loss": 0.8265,
      "num_input_tokens_seen": 46565904,
      "step": 80260
    },
    {
      "epoch": 11.954870420017873,
      "grad_norm": 5.34375,
      "learning_rate": 2.0853492459606373e-05,
      "loss": 1.0238,
      "num_input_tokens_seen": 46568784,
      "step": 80265
    },
    {
      "epoch": 11.955615132558833,
      "grad_norm": 13.875,
      "learning_rate": 2.085028808008185e-05,
      "loss": 1.1466,
      "num_input_tokens_seen": 46571568,
      "step": 80270
    },
    {
      "epoch": 11.956359845099792,
      "grad_norm": 10.125,
      "learning_rate": 2.084708377066249e-05,
      "loss": 1.0984,
      "num_input_tokens_seen": 46574480,
      "step": 80275
    },
    {
      "epoch": 11.95710455764075,
      "grad_norm": 13.8125,
      "learning_rate": 2.084387953140242e-05,
      "loss": 1.1713,
      "num_input_tokens_seen": 46577264,
      "step": 80280
    },
    {
      "epoch": 11.95784927018171,
      "grad_norm": 10.9375,
      "learning_rate": 2.0840675362355792e-05,
      "loss": 1.0409,
      "num_input_tokens_seen": 46580080,
      "step": 80285
    },
    {
      "epoch": 11.958593982722668,
      "grad_norm": 7.75,
      "learning_rate": 2.0837471263576716e-05,
      "loss": 1.1814,
      "num_input_tokens_seen": 46582864,
      "step": 80290
    },
    {
      "epoch": 11.959338695263629,
      "grad_norm": 10.6875,
      "learning_rate": 2.0834267235119342e-05,
      "loss": 0.8297,
      "num_input_tokens_seen": 46585616,
      "step": 80295
    },
    {
      "epoch": 11.960083407804587,
      "grad_norm": 17.375,
      "learning_rate": 2.083106327703779e-05,
      "loss": 1.2489,
      "num_input_tokens_seen": 46588624,
      "step": 80300
    },
    {
      "epoch": 11.960828120345546,
      "grad_norm": 10.9375,
      "learning_rate": 2.0827859389386184e-05,
      "loss": 1.1735,
      "num_input_tokens_seen": 46591472,
      "step": 80305
    },
    {
      "epoch": 11.961572832886507,
      "grad_norm": 14.875,
      "learning_rate": 2.0824655572218655e-05,
      "loss": 1.3341,
      "num_input_tokens_seen": 46594352,
      "step": 80310
    },
    {
      "epoch": 11.962317545427466,
      "grad_norm": 16.125,
      "learning_rate": 2.0821451825589315e-05,
      "loss": 1.0475,
      "num_input_tokens_seen": 46597040,
      "step": 80315
    },
    {
      "epoch": 11.963062257968424,
      "grad_norm": 13.9375,
      "learning_rate": 2.0818248149552315e-05,
      "loss": 1.0376,
      "num_input_tokens_seen": 46599728,
      "step": 80320
    },
    {
      "epoch": 11.963806970509383,
      "grad_norm": 10.0625,
      "learning_rate": 2.0815044544161748e-05,
      "loss": 1.2202,
      "num_input_tokens_seen": 46602640,
      "step": 80325
    },
    {
      "epoch": 11.964551683050342,
      "grad_norm": 9.1875,
      "learning_rate": 2.081184100947176e-05,
      "loss": 1.0542,
      "num_input_tokens_seen": 46605520,
      "step": 80330
    },
    {
      "epoch": 11.965296395591302,
      "grad_norm": 15.625,
      "learning_rate": 2.080863754553646e-05,
      "loss": 1.0469,
      "num_input_tokens_seen": 46608464,
      "step": 80335
    },
    {
      "epoch": 11.966041108132261,
      "grad_norm": 13.0625,
      "learning_rate": 2.080543415240997e-05,
      "loss": 1.2098,
      "num_input_tokens_seen": 46611472,
      "step": 80340
    },
    {
      "epoch": 11.96678582067322,
      "grad_norm": 9.6875,
      "learning_rate": 2.0802230830146398e-05,
      "loss": 0.9606,
      "num_input_tokens_seen": 46614608,
      "step": 80345
    },
    {
      "epoch": 11.967530533214179,
      "grad_norm": 8.75,
      "learning_rate": 2.0799027578799882e-05,
      "loss": 1.2143,
      "num_input_tokens_seen": 46617584,
      "step": 80350
    },
    {
      "epoch": 11.96827524575514,
      "grad_norm": 6.21875,
      "learning_rate": 2.0795824398424523e-05,
      "loss": 0.7605,
      "num_input_tokens_seen": 46620496,
      "step": 80355
    },
    {
      "epoch": 11.969019958296098,
      "grad_norm": 7.21875,
      "learning_rate": 2.079262128907443e-05,
      "loss": 0.8249,
      "num_input_tokens_seen": 46623120,
      "step": 80360
    },
    {
      "epoch": 11.969764670837057,
      "grad_norm": 11.3125,
      "learning_rate": 2.0789418250803732e-05,
      "loss": 0.7987,
      "num_input_tokens_seen": 46626256,
      "step": 80365
    },
    {
      "epoch": 11.970509383378015,
      "grad_norm": 5.96875,
      "learning_rate": 2.078621528366653e-05,
      "loss": 1.029,
      "num_input_tokens_seen": 46629200,
      "step": 80370
    },
    {
      "epoch": 11.971254095918976,
      "grad_norm": 10.375,
      "learning_rate": 2.078301238771694e-05,
      "loss": 1.1123,
      "num_input_tokens_seen": 46631952,
      "step": 80375
    },
    {
      "epoch": 11.971998808459935,
      "grad_norm": 4.125,
      "learning_rate": 2.0779809563009063e-05,
      "loss": 0.947,
      "num_input_tokens_seen": 46634704,
      "step": 80380
    },
    {
      "epoch": 11.972743521000893,
      "grad_norm": 12.25,
      "learning_rate": 2.0776606809597022e-05,
      "loss": 0.9972,
      "num_input_tokens_seen": 46637648,
      "step": 80385
    },
    {
      "epoch": 11.973488233541852,
      "grad_norm": 11.8125,
      "learning_rate": 2.077340412753492e-05,
      "loss": 1.337,
      "num_input_tokens_seen": 46640656,
      "step": 80390
    },
    {
      "epoch": 11.974232946082813,
      "grad_norm": 8.5,
      "learning_rate": 2.077020151687684e-05,
      "loss": 1.0614,
      "num_input_tokens_seen": 46643536,
      "step": 80395
    },
    {
      "epoch": 11.974977658623772,
      "grad_norm": 13.9375,
      "learning_rate": 2.0766998977676928e-05,
      "loss": 1.1532,
      "num_input_tokens_seen": 46646288,
      "step": 80400
    },
    {
      "epoch": 11.97572237116473,
      "grad_norm": 10.875,
      "learning_rate": 2.0763796509989252e-05,
      "loss": 0.9666,
      "num_input_tokens_seen": 46649072,
      "step": 80405
    },
    {
      "epoch": 11.976467083705689,
      "grad_norm": 15.3125,
      "learning_rate": 2.076059411386794e-05,
      "loss": 0.9614,
      "num_input_tokens_seen": 46652336,
      "step": 80410
    },
    {
      "epoch": 11.97721179624665,
      "grad_norm": 5.15625,
      "learning_rate": 2.075739178936707e-05,
      "loss": 0.9766,
      "num_input_tokens_seen": 46655184,
      "step": 80415
    },
    {
      "epoch": 11.977956508787608,
      "grad_norm": 11.0,
      "learning_rate": 2.0754189536540764e-05,
      "loss": 0.864,
      "num_input_tokens_seen": 46658128,
      "step": 80420
    },
    {
      "epoch": 11.978701221328567,
      "grad_norm": 6.59375,
      "learning_rate": 2.07509873554431e-05,
      "loss": 1.2299,
      "num_input_tokens_seen": 46660816,
      "step": 80425
    },
    {
      "epoch": 11.979445933869526,
      "grad_norm": 4.875,
      "learning_rate": 2.074778524612819e-05,
      "loss": 0.9997,
      "num_input_tokens_seen": 46663824,
      "step": 80430
    },
    {
      "epoch": 11.980190646410486,
      "grad_norm": 10.5625,
      "learning_rate": 2.0744583208650136e-05,
      "loss": 1.0432,
      "num_input_tokens_seen": 46666704,
      "step": 80435
    },
    {
      "epoch": 11.980935358951445,
      "grad_norm": 8.625,
      "learning_rate": 2.0741381243063015e-05,
      "loss": 1.0995,
      "num_input_tokens_seen": 46669520,
      "step": 80440
    },
    {
      "epoch": 11.981680071492404,
      "grad_norm": 9.9375,
      "learning_rate": 2.0738179349420935e-05,
      "loss": 0.9445,
      "num_input_tokens_seen": 46672240,
      "step": 80445
    },
    {
      "epoch": 11.982424784033363,
      "grad_norm": 15.5,
      "learning_rate": 2.0734977527777974e-05,
      "loss": 1.5162,
      "num_input_tokens_seen": 46675184,
      "step": 80450
    },
    {
      "epoch": 11.983169496574323,
      "grad_norm": 6.96875,
      "learning_rate": 2.0731775778188242e-05,
      "loss": 1.3072,
      "num_input_tokens_seen": 46679088,
      "step": 80455
    },
    {
      "epoch": 11.983914209115282,
      "grad_norm": 8.3125,
      "learning_rate": 2.0728574100705813e-05,
      "loss": 0.7658,
      "num_input_tokens_seen": 46681680,
      "step": 80460
    },
    {
      "epoch": 11.98465892165624,
      "grad_norm": 11.875,
      "learning_rate": 2.0725372495384786e-05,
      "loss": 0.8083,
      "num_input_tokens_seen": 46684336,
      "step": 80465
    },
    {
      "epoch": 11.9854036341972,
      "grad_norm": 16.25,
      "learning_rate": 2.0722170962279248e-05,
      "loss": 1.522,
      "num_input_tokens_seen": 46686800,
      "step": 80470
    },
    {
      "epoch": 11.986148346738158,
      "grad_norm": 4.6875,
      "learning_rate": 2.0718969501443286e-05,
      "loss": 0.9726,
      "num_input_tokens_seen": 46689744,
      "step": 80475
    },
    {
      "epoch": 11.986893059279119,
      "grad_norm": 10.9375,
      "learning_rate": 2.0715768112930984e-05,
      "loss": 1.0218,
      "num_input_tokens_seen": 46692624,
      "step": 80480
    },
    {
      "epoch": 11.987637771820078,
      "grad_norm": 6.8125,
      "learning_rate": 2.071256679679641e-05,
      "loss": 0.818,
      "num_input_tokens_seen": 46695280,
      "step": 80485
    },
    {
      "epoch": 11.988382484361036,
      "grad_norm": 8.8125,
      "learning_rate": 2.070936555309368e-05,
      "loss": 0.9236,
      "num_input_tokens_seen": 46698192,
      "step": 80490
    },
    {
      "epoch": 11.989127196901995,
      "grad_norm": 8.6875,
      "learning_rate": 2.0706164381876852e-05,
      "loss": 1.0514,
      "num_input_tokens_seen": 46701136,
      "step": 80495
    },
    {
      "epoch": 11.989871909442956,
      "grad_norm": 10.75,
      "learning_rate": 2.0702963283200018e-05,
      "loss": 0.9614,
      "num_input_tokens_seen": 46704144,
      "step": 80500
    },
    {
      "epoch": 11.990616621983914,
      "grad_norm": 9.5625,
      "learning_rate": 2.0699762257117235e-05,
      "loss": 1.2067,
      "num_input_tokens_seen": 46707152,
      "step": 80505
    },
    {
      "epoch": 11.991361334524873,
      "grad_norm": 11.875,
      "learning_rate": 2.0696561303682617e-05,
      "loss": 0.783,
      "num_input_tokens_seen": 46709936,
      "step": 80510
    },
    {
      "epoch": 11.992106047065832,
      "grad_norm": 8.375,
      "learning_rate": 2.0693360422950217e-05,
      "loss": 1.2188,
      "num_input_tokens_seen": 46712912,
      "step": 80515
    },
    {
      "epoch": 11.992850759606792,
      "grad_norm": 10.9375,
      "learning_rate": 2.0690159614974107e-05,
      "loss": 0.955,
      "num_input_tokens_seen": 46715856,
      "step": 80520
    },
    {
      "epoch": 11.993595472147751,
      "grad_norm": 4.125,
      "learning_rate": 2.068695887980838e-05,
      "loss": 0.6425,
      "num_input_tokens_seen": 46718480,
      "step": 80525
    },
    {
      "epoch": 11.99434018468871,
      "grad_norm": 12.5,
      "learning_rate": 2.0683758217507092e-05,
      "loss": 1.0401,
      "num_input_tokens_seen": 46721296,
      "step": 80530
    },
    {
      "epoch": 11.995084897229669,
      "grad_norm": 5.71875,
      "learning_rate": 2.068055762812433e-05,
      "loss": 1.0373,
      "num_input_tokens_seen": 46724016,
      "step": 80535
    },
    {
      "epoch": 11.99582960977063,
      "grad_norm": 6.3125,
      "learning_rate": 2.0677357111714147e-05,
      "loss": 1.1013,
      "num_input_tokens_seen": 46726864,
      "step": 80540
    },
    {
      "epoch": 11.996574322311588,
      "grad_norm": 7.75,
      "learning_rate": 2.067415666833063e-05,
      "loss": 1.0933,
      "num_input_tokens_seen": 46729584,
      "step": 80545
    },
    {
      "epoch": 11.997319034852547,
      "grad_norm": 16.375,
      "learning_rate": 2.0670956298027833e-05,
      "loss": 1.1528,
      "num_input_tokens_seen": 46732400,
      "step": 80550
    },
    {
      "epoch": 11.998063747393505,
      "grad_norm": 11.5,
      "learning_rate": 2.0667756000859835e-05,
      "loss": 1.3722,
      "num_input_tokens_seen": 46734896,
      "step": 80555
    },
    {
      "epoch": 11.998808459934466,
      "grad_norm": 9.875,
      "learning_rate": 2.06645557768807e-05,
      "loss": 1.1627,
      "num_input_tokens_seen": 46737904,
      "step": 80560
    },
    {
      "epoch": 11.999553172475425,
      "grad_norm": 10.0,
      "learning_rate": 2.0661355626144483e-05,
      "loss": 0.9584,
      "num_input_tokens_seen": 46740624,
      "step": 80565
    },
    {
      "epoch": 12.0,
      "eval_loss": 1.0869436264038086,
      "eval_runtime": 43.7721,
      "eval_samples_per_second": 68.171,
      "eval_steps_per_second": 17.043,
      "num_input_tokens_seen": 46741816,
      "step": 80568
    },
    {
      "epoch": 12.000297885016384,
      "grad_norm": 12.8125,
      "learning_rate": 2.0658155548705258e-05,
      "loss": 0.9112,
      "num_input_tokens_seen": 46743320,
      "step": 80570
    },
    {
      "epoch": 12.001042597557342,
      "grad_norm": 11.9375,
      "learning_rate": 2.065495554461707e-05,
      "loss": 1.1675,
      "num_input_tokens_seen": 46745944,
      "step": 80575
    },
    {
      "epoch": 12.001787310098303,
      "grad_norm": 6.21875,
      "learning_rate": 2.0651755613934005e-05,
      "loss": 0.9224,
      "num_input_tokens_seen": 46748792,
      "step": 80580
    },
    {
      "epoch": 12.002532022639262,
      "grad_norm": 3.28125,
      "learning_rate": 2.0648555756710098e-05,
      "loss": 0.8887,
      "num_input_tokens_seen": 46752088,
      "step": 80585
    },
    {
      "epoch": 12.00327673518022,
      "grad_norm": 20.375,
      "learning_rate": 2.064535597299943e-05,
      "loss": 1.6662,
      "num_input_tokens_seen": 46755288,
      "step": 80590
    },
    {
      "epoch": 12.004021447721179,
      "grad_norm": 11.9375,
      "learning_rate": 2.0642156262856045e-05,
      "loss": 1.3391,
      "num_input_tokens_seen": 46758296,
      "step": 80595
    },
    {
      "epoch": 12.00476616026214,
      "grad_norm": 11.625,
      "learning_rate": 2.0638956626333993e-05,
      "loss": 0.8531,
      "num_input_tokens_seen": 46761048,
      "step": 80600
    },
    {
      "epoch": 12.005510872803098,
      "grad_norm": 12.0625,
      "learning_rate": 2.0635757063487348e-05,
      "loss": 1.1779,
      "num_input_tokens_seen": 46764024,
      "step": 80605
    },
    {
      "epoch": 12.006255585344057,
      "grad_norm": 9.6875,
      "learning_rate": 2.0632557574370137e-05,
      "loss": 0.9574,
      "num_input_tokens_seen": 46766776,
      "step": 80610
    },
    {
      "epoch": 12.007000297885016,
      "grad_norm": 10.3125,
      "learning_rate": 2.0629358159036437e-05,
      "loss": 1.0388,
      "num_input_tokens_seen": 46769400,
      "step": 80615
    },
    {
      "epoch": 12.007745010425976,
      "grad_norm": 12.375,
      "learning_rate": 2.0626158817540284e-05,
      "loss": 1.1874,
      "num_input_tokens_seen": 46772280,
      "step": 80620
    },
    {
      "epoch": 12.008489722966935,
      "grad_norm": 10.125,
      "learning_rate": 2.0622959549935738e-05,
      "loss": 0.8539,
      "num_input_tokens_seen": 46775384,
      "step": 80625
    },
    {
      "epoch": 12.009234435507894,
      "grad_norm": 6.40625,
      "learning_rate": 2.061976035627684e-05,
      "loss": 0.8249,
      "num_input_tokens_seen": 46778264,
      "step": 80630
    },
    {
      "epoch": 12.009979148048853,
      "grad_norm": 17.125,
      "learning_rate": 2.061656123661764e-05,
      "loss": 1.2922,
      "num_input_tokens_seen": 46781112,
      "step": 80635
    },
    {
      "epoch": 12.010723860589811,
      "grad_norm": 11.1875,
      "learning_rate": 2.0613362191012185e-05,
      "loss": 1.079,
      "num_input_tokens_seen": 46784024,
      "step": 80640
    },
    {
      "epoch": 12.011468573130772,
      "grad_norm": 9.625,
      "learning_rate": 2.0610163219514504e-05,
      "loss": 0.9113,
      "num_input_tokens_seen": 46787032,
      "step": 80645
    },
    {
      "epoch": 12.01221328567173,
      "grad_norm": 6.375,
      "learning_rate": 2.0606964322178667e-05,
      "loss": 1.0581,
      "num_input_tokens_seen": 46790072,
      "step": 80650
    },
    {
      "epoch": 12.01295799821269,
      "grad_norm": 5.21875,
      "learning_rate": 2.0603765499058695e-05,
      "loss": 0.9978,
      "num_input_tokens_seen": 46792856,
      "step": 80655
    },
    {
      "epoch": 12.013702710753648,
      "grad_norm": 13.0625,
      "learning_rate": 2.0600566750208642e-05,
      "loss": 1.0746,
      "num_input_tokens_seen": 46795672,
      "step": 80660
    },
    {
      "epoch": 12.014447423294609,
      "grad_norm": 13.1875,
      "learning_rate": 2.0597368075682542e-05,
      "loss": 1.2016,
      "num_input_tokens_seen": 46798584,
      "step": 80665
    },
    {
      "epoch": 12.015192135835568,
      "grad_norm": 13.0625,
      "learning_rate": 2.0594169475534436e-05,
      "loss": 1.0206,
      "num_input_tokens_seen": 46801304,
      "step": 80670
    },
    {
      "epoch": 12.015936848376526,
      "grad_norm": 7.5,
      "learning_rate": 2.0590970949818357e-05,
      "loss": 1.078,
      "num_input_tokens_seen": 46804056,
      "step": 80675
    },
    {
      "epoch": 12.016681560917485,
      "grad_norm": 7.1875,
      "learning_rate": 2.0587772498588336e-05,
      "loss": 0.9405,
      "num_input_tokens_seen": 46806744,
      "step": 80680
    },
    {
      "epoch": 12.017426273458446,
      "grad_norm": 17.875,
      "learning_rate": 2.0584574121898424e-05,
      "loss": 1.2138,
      "num_input_tokens_seen": 46809720,
      "step": 80685
    },
    {
      "epoch": 12.018170985999404,
      "grad_norm": 7.03125,
      "learning_rate": 2.0581375819802635e-05,
      "loss": 1.3338,
      "num_input_tokens_seen": 46812568,
      "step": 80690
    },
    {
      "epoch": 12.018915698540363,
      "grad_norm": 6.875,
      "learning_rate": 2.057817759235502e-05,
      "loss": 0.9658,
      "num_input_tokens_seen": 46815160,
      "step": 80695
    },
    {
      "epoch": 12.019660411081322,
      "grad_norm": 13.3125,
      "learning_rate": 2.0574979439609593e-05,
      "loss": 0.9011,
      "num_input_tokens_seen": 46817848,
      "step": 80700
    },
    {
      "epoch": 12.020405123622282,
      "grad_norm": 11.9375,
      "learning_rate": 2.0571781361620398e-05,
      "loss": 0.9192,
      "num_input_tokens_seen": 46820632,
      "step": 80705
    },
    {
      "epoch": 12.021149836163241,
      "grad_norm": 8.125,
      "learning_rate": 2.0568583358441445e-05,
      "loss": 1.3538,
      "num_input_tokens_seen": 46823640,
      "step": 80710
    },
    {
      "epoch": 12.0218945487042,
      "grad_norm": 6.46875,
      "learning_rate": 2.0565385430126783e-05,
      "loss": 0.7678,
      "num_input_tokens_seen": 46826360,
      "step": 80715
    },
    {
      "epoch": 12.022639261245159,
      "grad_norm": 14.375,
      "learning_rate": 2.0562187576730428e-05,
      "loss": 0.9131,
      "num_input_tokens_seen": 46829176,
      "step": 80720
    },
    {
      "epoch": 12.02338397378612,
      "grad_norm": 8.4375,
      "learning_rate": 2.0558989798306395e-05,
      "loss": 0.8683,
      "num_input_tokens_seen": 46832536,
      "step": 80725
    },
    {
      "epoch": 12.024128686327078,
      "grad_norm": 9.375,
      "learning_rate": 2.0555792094908722e-05,
      "loss": 0.869,
      "num_input_tokens_seen": 46835480,
      "step": 80730
    },
    {
      "epoch": 12.024873398868037,
      "grad_norm": 8.625,
      "learning_rate": 2.055259446659142e-05,
      "loss": 1.2383,
      "num_input_tokens_seen": 46838712,
      "step": 80735
    },
    {
      "epoch": 12.025618111408996,
      "grad_norm": 12.9375,
      "learning_rate": 2.0549396913408522e-05,
      "loss": 1.2404,
      "num_input_tokens_seen": 46841624,
      "step": 80740
    },
    {
      "epoch": 12.026362823949956,
      "grad_norm": 6.5625,
      "learning_rate": 2.0546199435414028e-05,
      "loss": 0.725,
      "num_input_tokens_seen": 46844600,
      "step": 80745
    },
    {
      "epoch": 12.027107536490915,
      "grad_norm": 5.375,
      "learning_rate": 2.054300203266198e-05,
      "loss": 0.8766,
      "num_input_tokens_seen": 46847256,
      "step": 80750
    },
    {
      "epoch": 12.027852249031874,
      "grad_norm": 10.5,
      "learning_rate": 2.0539804705206378e-05,
      "loss": 1.0227,
      "num_input_tokens_seen": 46849816,
      "step": 80755
    },
    {
      "epoch": 12.028596961572832,
      "grad_norm": 11.8125,
      "learning_rate": 2.0536607453101236e-05,
      "loss": 1.3088,
      "num_input_tokens_seen": 46853112,
      "step": 80760
    },
    {
      "epoch": 12.029341674113793,
      "grad_norm": 13.375,
      "learning_rate": 2.0533410276400582e-05,
      "loss": 1.1224,
      "num_input_tokens_seen": 46855800,
      "step": 80765
    },
    {
      "epoch": 12.030086386654752,
      "grad_norm": 8.5625,
      "learning_rate": 2.053021317515842e-05,
      "loss": 1.14,
      "num_input_tokens_seen": 46858552,
      "step": 80770
    },
    {
      "epoch": 12.03083109919571,
      "grad_norm": 7.65625,
      "learning_rate": 2.0527016149428767e-05,
      "loss": 1.0965,
      "num_input_tokens_seen": 46861688,
      "step": 80775
    },
    {
      "epoch": 12.03157581173667,
      "grad_norm": 5.65625,
      "learning_rate": 2.052381919926562e-05,
      "loss": 1.1376,
      "num_input_tokens_seen": 46864280,
      "step": 80780
    },
    {
      "epoch": 12.03232052427763,
      "grad_norm": 11.9375,
      "learning_rate": 2.052062232472301e-05,
      "loss": 1.3105,
      "num_input_tokens_seen": 46866840,
      "step": 80785
    },
    {
      "epoch": 12.033065236818588,
      "grad_norm": 8.375,
      "learning_rate": 2.0517425525854926e-05,
      "loss": 0.8398,
      "num_input_tokens_seen": 46869464,
      "step": 80790
    },
    {
      "epoch": 12.033809949359547,
      "grad_norm": 11.625,
      "learning_rate": 2.051422880271538e-05,
      "loss": 1.397,
      "num_input_tokens_seen": 46872408,
      "step": 80795
    },
    {
      "epoch": 12.034554661900506,
      "grad_norm": 8.9375,
      "learning_rate": 2.051103215535839e-05,
      "loss": 0.963,
      "num_input_tokens_seen": 46875256,
      "step": 80800
    },
    {
      "epoch": 12.035299374441466,
      "grad_norm": 10.125,
      "learning_rate": 2.0507835583837943e-05,
      "loss": 1.4167,
      "num_input_tokens_seen": 46878296,
      "step": 80805
    },
    {
      "epoch": 12.036044086982425,
      "grad_norm": 14.125,
      "learning_rate": 2.050463908820805e-05,
      "loss": 1.2467,
      "num_input_tokens_seen": 46881112,
      "step": 80810
    },
    {
      "epoch": 12.036788799523384,
      "grad_norm": 12.75,
      "learning_rate": 2.0501442668522703e-05,
      "loss": 1.1497,
      "num_input_tokens_seen": 46883928,
      "step": 80815
    },
    {
      "epoch": 12.037533512064343,
      "grad_norm": 8.6875,
      "learning_rate": 2.0498246324835918e-05,
      "loss": 1.1309,
      "num_input_tokens_seen": 46886744,
      "step": 80820
    },
    {
      "epoch": 12.038278224605302,
      "grad_norm": 15.875,
      "learning_rate": 2.0495050057201683e-05,
      "loss": 0.976,
      "num_input_tokens_seen": 46889848,
      "step": 80825
    },
    {
      "epoch": 12.039022937146262,
      "grad_norm": 10.5,
      "learning_rate": 2.0491853865674002e-05,
      "loss": 0.8431,
      "num_input_tokens_seen": 46892888,
      "step": 80830
    },
    {
      "epoch": 12.03976764968722,
      "grad_norm": 10.25,
      "learning_rate": 2.0488657750306865e-05,
      "loss": 1.2581,
      "num_input_tokens_seen": 46895928,
      "step": 80835
    },
    {
      "epoch": 12.04051236222818,
      "grad_norm": 11.75,
      "learning_rate": 2.0485461711154265e-05,
      "loss": 1.0559,
      "num_input_tokens_seen": 46898840,
      "step": 80840
    },
    {
      "epoch": 12.041257074769138,
      "grad_norm": 5.75,
      "learning_rate": 2.0482265748270213e-05,
      "loss": 1.2532,
      "num_input_tokens_seen": 46901368,
      "step": 80845
    },
    {
      "epoch": 12.042001787310099,
      "grad_norm": 10.75,
      "learning_rate": 2.0479069861708674e-05,
      "loss": 0.9954,
      "num_input_tokens_seen": 46904312,
      "step": 80850
    },
    {
      "epoch": 12.042746499851058,
      "grad_norm": 5.84375,
      "learning_rate": 2.047587405152367e-05,
      "loss": 1.2599,
      "num_input_tokens_seen": 46907064,
      "step": 80855
    },
    {
      "epoch": 12.043491212392016,
      "grad_norm": 14.5625,
      "learning_rate": 2.0472678317769168e-05,
      "loss": 1.0731,
      "num_input_tokens_seen": 46910008,
      "step": 80860
    },
    {
      "epoch": 12.044235924932975,
      "grad_norm": 10.125,
      "learning_rate": 2.0469482660499167e-05,
      "loss": 1.1412,
      "num_input_tokens_seen": 46912824,
      "step": 80865
    },
    {
      "epoch": 12.044980637473936,
      "grad_norm": 8.1875,
      "learning_rate": 2.0466287079767646e-05,
      "loss": 0.9617,
      "num_input_tokens_seen": 46915736,
      "step": 80870
    },
    {
      "epoch": 12.045725350014894,
      "grad_norm": 6.96875,
      "learning_rate": 2.0463091575628608e-05,
      "loss": 0.8409,
      "num_input_tokens_seen": 46918904,
      "step": 80875
    },
    {
      "epoch": 12.046470062555853,
      "grad_norm": 8.125,
      "learning_rate": 2.045989614813602e-05,
      "loss": 0.8987,
      "num_input_tokens_seen": 46921816,
      "step": 80880
    },
    {
      "epoch": 12.047214775096812,
      "grad_norm": 10.0625,
      "learning_rate": 2.0456700797343867e-05,
      "loss": 0.8739,
      "num_input_tokens_seen": 46924504,
      "step": 80885
    },
    {
      "epoch": 12.047959487637772,
      "grad_norm": 15.125,
      "learning_rate": 2.0453505523306147e-05,
      "loss": 1.4323,
      "num_input_tokens_seen": 46927416,
      "step": 80890
    },
    {
      "epoch": 12.048704200178731,
      "grad_norm": 11.4375,
      "learning_rate": 2.0450310326076823e-05,
      "loss": 1.2411,
      "num_input_tokens_seen": 46930360,
      "step": 80895
    },
    {
      "epoch": 12.04944891271969,
      "grad_norm": 8.3125,
      "learning_rate": 2.0447115205709887e-05,
      "loss": 0.98,
      "num_input_tokens_seen": 46933496,
      "step": 80900
    },
    {
      "epoch": 12.050193625260649,
      "grad_norm": 8.75,
      "learning_rate": 2.0443920162259306e-05,
      "loss": 0.979,
      "num_input_tokens_seen": 46936568,
      "step": 80905
    },
    {
      "epoch": 12.05093833780161,
      "grad_norm": 13.0625,
      "learning_rate": 2.044072519577907e-05,
      "loss": 0.9168,
      "num_input_tokens_seen": 46939480,
      "step": 80910
    },
    {
      "epoch": 12.051683050342568,
      "grad_norm": 8.9375,
      "learning_rate": 2.043753030632315e-05,
      "loss": 1.129,
      "num_input_tokens_seen": 46942168,
      "step": 80915
    },
    {
      "epoch": 12.052427762883527,
      "grad_norm": 11.25,
      "learning_rate": 2.0434335493945506e-05,
      "loss": 1.1559,
      "num_input_tokens_seen": 46944952,
      "step": 80920
    },
    {
      "epoch": 12.053172475424486,
      "grad_norm": 15.1875,
      "learning_rate": 2.043114075870013e-05,
      "loss": 1.021,
      "num_input_tokens_seen": 46947640,
      "step": 80925
    },
    {
      "epoch": 12.053917187965446,
      "grad_norm": 4.875,
      "learning_rate": 2.042794610064099e-05,
      "loss": 1.0522,
      "num_input_tokens_seen": 46950264,
      "step": 80930
    },
    {
      "epoch": 12.054661900506405,
      "grad_norm": 10.25,
      "learning_rate": 2.0424751519822054e-05,
      "loss": 1.2725,
      "num_input_tokens_seen": 46952984,
      "step": 80935
    },
    {
      "epoch": 12.055406613047364,
      "grad_norm": 16.125,
      "learning_rate": 2.0421557016297283e-05,
      "loss": 0.8593,
      "num_input_tokens_seen": 46956120,
      "step": 80940
    },
    {
      "epoch": 12.056151325588322,
      "grad_norm": 11.0625,
      "learning_rate": 2.041836259012066e-05,
      "loss": 1.1213,
      "num_input_tokens_seen": 46958840,
      "step": 80945
    },
    {
      "epoch": 12.056896038129283,
      "grad_norm": 16.75,
      "learning_rate": 2.0415168241346138e-05,
      "loss": 1.042,
      "num_input_tokens_seen": 46961496,
      "step": 80950
    },
    {
      "epoch": 12.057640750670242,
      "grad_norm": 6.03125,
      "learning_rate": 2.0411973970027698e-05,
      "loss": 0.9066,
      "num_input_tokens_seen": 46964536,
      "step": 80955
    },
    {
      "epoch": 12.0583854632112,
      "grad_norm": 9.9375,
      "learning_rate": 2.0408779776219295e-05,
      "loss": 1.1184,
      "num_input_tokens_seen": 46967192,
      "step": 80960
    },
    {
      "epoch": 12.05913017575216,
      "grad_norm": 7.4375,
      "learning_rate": 2.0405585659974885e-05,
      "loss": 0.9791,
      "num_input_tokens_seen": 46970136,
      "step": 80965
    },
    {
      "epoch": 12.05987488829312,
      "grad_norm": 6.71875,
      "learning_rate": 2.0402391621348444e-05,
      "loss": 1.1886,
      "num_input_tokens_seen": 46972984,
      "step": 80970
    },
    {
      "epoch": 12.060619600834078,
      "grad_norm": 8.5,
      "learning_rate": 2.039919766039391e-05,
      "loss": 1.3475,
      "num_input_tokens_seen": 46976088,
      "step": 80975
    },
    {
      "epoch": 12.061364313375037,
      "grad_norm": 11.625,
      "learning_rate": 2.0396003777165266e-05,
      "loss": 1.0614,
      "num_input_tokens_seen": 46978680,
      "step": 80980
    },
    {
      "epoch": 12.062109025915996,
      "grad_norm": 8.1875,
      "learning_rate": 2.0392809971716448e-05,
      "loss": 1.1834,
      "num_input_tokens_seen": 46981656,
      "step": 80985
    },
    {
      "epoch": 12.062853738456955,
      "grad_norm": 4.875,
      "learning_rate": 2.0389616244101437e-05,
      "loss": 0.7437,
      "num_input_tokens_seen": 46984440,
      "step": 80990
    },
    {
      "epoch": 12.063598450997915,
      "grad_norm": 3.828125,
      "learning_rate": 2.038642259437417e-05,
      "loss": 1.1038,
      "num_input_tokens_seen": 46987064,
      "step": 80995
    },
    {
      "epoch": 12.064343163538874,
      "grad_norm": 12.8125,
      "learning_rate": 2.03832290225886e-05,
      "loss": 1.2813,
      "num_input_tokens_seen": 46989624,
      "step": 81000
    },
    {
      "epoch": 12.065087876079833,
      "grad_norm": 19.375,
      "learning_rate": 2.0380035528798692e-05,
      "loss": 1.3898,
      "num_input_tokens_seen": 46992504,
      "step": 81005
    },
    {
      "epoch": 12.065832588620792,
      "grad_norm": 5.09375,
      "learning_rate": 2.0376842113058372e-05,
      "loss": 1.1794,
      "num_input_tokens_seen": 46995608,
      "step": 81010
    },
    {
      "epoch": 12.066577301161752,
      "grad_norm": 11.4375,
      "learning_rate": 2.037364877542162e-05,
      "loss": 1.4558,
      "num_input_tokens_seen": 46998648,
      "step": 81015
    },
    {
      "epoch": 12.06732201370271,
      "grad_norm": 6.71875,
      "learning_rate": 2.037045551594236e-05,
      "loss": 0.9146,
      "num_input_tokens_seen": 47001688,
      "step": 81020
    },
    {
      "epoch": 12.06806672624367,
      "grad_norm": 11.75,
      "learning_rate": 2.0367262334674556e-05,
      "loss": 1.4873,
      "num_input_tokens_seen": 47004536,
      "step": 81025
    },
    {
      "epoch": 12.068811438784628,
      "grad_norm": 6.53125,
      "learning_rate": 2.0364069231672143e-05,
      "loss": 1.2135,
      "num_input_tokens_seen": 47007160,
      "step": 81030
    },
    {
      "epoch": 12.069556151325589,
      "grad_norm": 10.9375,
      "learning_rate": 2.0360876206989073e-05,
      "loss": 1.0755,
      "num_input_tokens_seen": 47009720,
      "step": 81035
    },
    {
      "epoch": 12.070300863866548,
      "grad_norm": 10.625,
      "learning_rate": 2.0357683260679285e-05,
      "loss": 0.8788,
      "num_input_tokens_seen": 47012504,
      "step": 81040
    },
    {
      "epoch": 12.071045576407506,
      "grad_norm": 8.875,
      "learning_rate": 2.035449039279671e-05,
      "loss": 0.9119,
      "num_input_tokens_seen": 47015576,
      "step": 81045
    },
    {
      "epoch": 12.071790288948465,
      "grad_norm": 12.5625,
      "learning_rate": 2.035129760339531e-05,
      "loss": 0.9615,
      "num_input_tokens_seen": 47018488,
      "step": 81050
    },
    {
      "epoch": 12.072535001489426,
      "grad_norm": 13.8125,
      "learning_rate": 2.0348104892528998e-05,
      "loss": 1.0493,
      "num_input_tokens_seen": 47021336,
      "step": 81055
    },
    {
      "epoch": 12.073279714030384,
      "grad_norm": 18.0,
      "learning_rate": 2.0344912260251742e-05,
      "loss": 0.9177,
      "num_input_tokens_seen": 47024472,
      "step": 81060
    },
    {
      "epoch": 12.074024426571343,
      "grad_norm": 7.8125,
      "learning_rate": 2.034171970661745e-05,
      "loss": 1.192,
      "num_input_tokens_seen": 47027480,
      "step": 81065
    },
    {
      "epoch": 12.074769139112302,
      "grad_norm": 13.5625,
      "learning_rate": 2.0338527231680078e-05,
      "loss": 1.082,
      "num_input_tokens_seen": 47030392,
      "step": 81070
    },
    {
      "epoch": 12.075513851653263,
      "grad_norm": 20.5,
      "learning_rate": 2.033533483549354e-05,
      "loss": 1.1602,
      "num_input_tokens_seen": 47033176,
      "step": 81075
    },
    {
      "epoch": 12.076258564194221,
      "grad_norm": 6.65625,
      "learning_rate": 2.033214251811179e-05,
      "loss": 0.915,
      "num_input_tokens_seen": 47036088,
      "step": 81080
    },
    {
      "epoch": 12.07700327673518,
      "grad_norm": 7.34375,
      "learning_rate": 2.0328950279588748e-05,
      "loss": 1.0166,
      "num_input_tokens_seen": 47038872,
      "step": 81085
    },
    {
      "epoch": 12.077747989276139,
      "grad_norm": 7.0625,
      "learning_rate": 2.0325758119978334e-05,
      "loss": 1.2844,
      "num_input_tokens_seen": 47041752,
      "step": 81090
    },
    {
      "epoch": 12.0784927018171,
      "grad_norm": 10.1875,
      "learning_rate": 2.0322566039334497e-05,
      "loss": 0.9605,
      "num_input_tokens_seen": 47044440,
      "step": 81095
    },
    {
      "epoch": 12.079237414358058,
      "grad_norm": 8.3125,
      "learning_rate": 2.0319374037711143e-05,
      "loss": 1.1602,
      "num_input_tokens_seen": 47047224,
      "step": 81100
    },
    {
      "epoch": 12.079982126899017,
      "grad_norm": 5.75,
      "learning_rate": 2.0316182115162218e-05,
      "loss": 1.0454,
      "num_input_tokens_seen": 47050136,
      "step": 81105
    },
    {
      "epoch": 12.080726839439976,
      "grad_norm": 13.25,
      "learning_rate": 2.031299027174162e-05,
      "loss": 1.013,
      "num_input_tokens_seen": 47053208,
      "step": 81110
    },
    {
      "epoch": 12.081471551980936,
      "grad_norm": 6.1875,
      "learning_rate": 2.03097985075033e-05,
      "loss": 0.956,
      "num_input_tokens_seen": 47056408,
      "step": 81115
    },
    {
      "epoch": 12.082216264521895,
      "grad_norm": 6.375,
      "learning_rate": 2.030660682250117e-05,
      "loss": 1.1286,
      "num_input_tokens_seen": 47059224,
      "step": 81120
    },
    {
      "epoch": 12.082960977062854,
      "grad_norm": 7.96875,
      "learning_rate": 2.0303415216789135e-05,
      "loss": 0.9497,
      "num_input_tokens_seen": 47062232,
      "step": 81125
    },
    {
      "epoch": 12.083705689603812,
      "grad_norm": 6.78125,
      "learning_rate": 2.0300223690421135e-05,
      "loss": 1.1512,
      "num_input_tokens_seen": 47065048,
      "step": 81130
    },
    {
      "epoch": 12.084450402144773,
      "grad_norm": 8.0625,
      "learning_rate": 2.029703224345108e-05,
      "loss": 1.0574,
      "num_input_tokens_seen": 47067640,
      "step": 81135
    },
    {
      "epoch": 12.085195114685732,
      "grad_norm": 6.25,
      "learning_rate": 2.0293840875932886e-05,
      "loss": 1.011,
      "num_input_tokens_seen": 47070648,
      "step": 81140
    },
    {
      "epoch": 12.08593982722669,
      "grad_norm": 11.125,
      "learning_rate": 2.029064958792046e-05,
      "loss": 1.4417,
      "num_input_tokens_seen": 47073336,
      "step": 81145
    },
    {
      "epoch": 12.08668453976765,
      "grad_norm": 11.1875,
      "learning_rate": 2.0287458379467728e-05,
      "loss": 1.3783,
      "num_input_tokens_seen": 47075960,
      "step": 81150
    },
    {
      "epoch": 12.08742925230861,
      "grad_norm": 9.5625,
      "learning_rate": 2.028426725062859e-05,
      "loss": 1.0778,
      "num_input_tokens_seen": 47078776,
      "step": 81155
    },
    {
      "epoch": 12.088173964849569,
      "grad_norm": 11.125,
      "learning_rate": 2.0281076201456977e-05,
      "loss": 1.397,
      "num_input_tokens_seen": 47081688,
      "step": 81160
    },
    {
      "epoch": 12.088918677390527,
      "grad_norm": 9.375,
      "learning_rate": 2.0277885232006776e-05,
      "loss": 0.8813,
      "num_input_tokens_seen": 47084472,
      "step": 81165
    },
    {
      "epoch": 12.089663389931486,
      "grad_norm": 2.109375,
      "learning_rate": 2.0274694342331907e-05,
      "loss": 0.8306,
      "num_input_tokens_seen": 47087288,
      "step": 81170
    },
    {
      "epoch": 12.090408102472445,
      "grad_norm": 17.625,
      "learning_rate": 2.027150353248628e-05,
      "loss": 0.8781,
      "num_input_tokens_seen": 47090104,
      "step": 81175
    },
    {
      "epoch": 12.091152815013405,
      "grad_norm": 12.625,
      "learning_rate": 2.026831280252378e-05,
      "loss": 1.0301,
      "num_input_tokens_seen": 47092920,
      "step": 81180
    },
    {
      "epoch": 12.091897527554364,
      "grad_norm": 8.5625,
      "learning_rate": 2.026512215249834e-05,
      "loss": 0.9089,
      "num_input_tokens_seen": 47095736,
      "step": 81185
    },
    {
      "epoch": 12.092642240095323,
      "grad_norm": 6.96875,
      "learning_rate": 2.0261931582463844e-05,
      "loss": 1.0526,
      "num_input_tokens_seen": 47098424,
      "step": 81190
    },
    {
      "epoch": 12.093386952636282,
      "grad_norm": 11.125,
      "learning_rate": 2.0258741092474204e-05,
      "loss": 1.1541,
      "num_input_tokens_seen": 47101272,
      "step": 81195
    },
    {
      "epoch": 12.094131665177242,
      "grad_norm": 10.9375,
      "learning_rate": 2.0255550682583313e-05,
      "loss": 1.1,
      "num_input_tokens_seen": 47104088,
      "step": 81200
    },
    {
      "epoch": 12.094876377718201,
      "grad_norm": 7.1875,
      "learning_rate": 2.025236035284506e-05,
      "loss": 0.8818,
      "num_input_tokens_seen": 47106904,
      "step": 81205
    },
    {
      "epoch": 12.09562109025916,
      "grad_norm": 6.59375,
      "learning_rate": 2.0249170103313365e-05,
      "loss": 0.962,
      "num_input_tokens_seen": 47109944,
      "step": 81210
    },
    {
      "epoch": 12.096365802800118,
      "grad_norm": 4.8125,
      "learning_rate": 2.0245979934042104e-05,
      "loss": 0.8479,
      "num_input_tokens_seen": 47113176,
      "step": 81215
    },
    {
      "epoch": 12.097110515341079,
      "grad_norm": 16.0,
      "learning_rate": 2.0242789845085187e-05,
      "loss": 1.0336,
      "num_input_tokens_seen": 47116088,
      "step": 81220
    },
    {
      "epoch": 12.097855227882038,
      "grad_norm": 20.75,
      "learning_rate": 2.0239599836496497e-05,
      "loss": 1.0274,
      "num_input_tokens_seen": 47119000,
      "step": 81225
    },
    {
      "epoch": 12.098599940422996,
      "grad_norm": 8.625,
      "learning_rate": 2.0236409908329933e-05,
      "loss": 1.1207,
      "num_input_tokens_seen": 47121720,
      "step": 81230
    },
    {
      "epoch": 12.099344652963955,
      "grad_norm": 6.3125,
      "learning_rate": 2.0233220060639373e-05,
      "loss": 1.217,
      "num_input_tokens_seen": 47124440,
      "step": 81235
    },
    {
      "epoch": 12.100089365504916,
      "grad_norm": 11.9375,
      "learning_rate": 2.023003029347873e-05,
      "loss": 1.0888,
      "num_input_tokens_seen": 47127480,
      "step": 81240
    },
    {
      "epoch": 12.100834078045875,
      "grad_norm": 8.625,
      "learning_rate": 2.0226840606901872e-05,
      "loss": 1.3732,
      "num_input_tokens_seen": 47130360,
      "step": 81245
    },
    {
      "epoch": 12.101578790586833,
      "grad_norm": 20.25,
      "learning_rate": 2.022365100096268e-05,
      "loss": 1.2294,
      "num_input_tokens_seen": 47133592,
      "step": 81250
    },
    {
      "epoch": 12.102323503127792,
      "grad_norm": 11.125,
      "learning_rate": 2.0220461475715063e-05,
      "loss": 0.9069,
      "num_input_tokens_seen": 47136376,
      "step": 81255
    },
    {
      "epoch": 12.103068215668753,
      "grad_norm": 3.03125,
      "learning_rate": 2.0217272031212887e-05,
      "loss": 0.8262,
      "num_input_tokens_seen": 47139288,
      "step": 81260
    },
    {
      "epoch": 12.103812928209711,
      "grad_norm": 10.4375,
      "learning_rate": 2.021408266751004e-05,
      "loss": 0.7697,
      "num_input_tokens_seen": 47142072,
      "step": 81265
    },
    {
      "epoch": 12.10455764075067,
      "grad_norm": 20.5,
      "learning_rate": 2.0210893384660396e-05,
      "loss": 1.5007,
      "num_input_tokens_seen": 47145304,
      "step": 81270
    },
    {
      "epoch": 12.105302353291629,
      "grad_norm": 8.625,
      "learning_rate": 2.0207704182717852e-05,
      "loss": 1.1118,
      "num_input_tokens_seen": 47148280,
      "step": 81275
    },
    {
      "epoch": 12.10604706583259,
      "grad_norm": 8.3125,
      "learning_rate": 2.0204515061736275e-05,
      "loss": 1.166,
      "num_input_tokens_seen": 47151160,
      "step": 81280
    },
    {
      "epoch": 12.106791778373548,
      "grad_norm": 6.3125,
      "learning_rate": 2.0201326021769526e-05,
      "loss": 1.1056,
      "num_input_tokens_seen": 47153880,
      "step": 81285
    },
    {
      "epoch": 12.107536490914507,
      "grad_norm": 16.75,
      "learning_rate": 2.0198137062871512e-05,
      "loss": 1.1685,
      "num_input_tokens_seen": 47156760,
      "step": 81290
    },
    {
      "epoch": 12.108281203455466,
      "grad_norm": 3.625,
      "learning_rate": 2.0194948185096086e-05,
      "loss": 1.0441,
      "num_input_tokens_seen": 47159640,
      "step": 81295
    },
    {
      "epoch": 12.109025915996426,
      "grad_norm": 10.875,
      "learning_rate": 2.019175938849713e-05,
      "loss": 1.0888,
      "num_input_tokens_seen": 47162616,
      "step": 81300
    },
    {
      "epoch": 12.109770628537385,
      "grad_norm": 9.125,
      "learning_rate": 2.0188570673128504e-05,
      "loss": 1.472,
      "num_input_tokens_seen": 47165368,
      "step": 81305
    },
    {
      "epoch": 12.110515341078344,
      "grad_norm": 7.3125,
      "learning_rate": 2.0185382039044094e-05,
      "loss": 1.1347,
      "num_input_tokens_seen": 47168504,
      "step": 81310
    },
    {
      "epoch": 12.111260053619302,
      "grad_norm": 11.25,
      "learning_rate": 2.0182193486297755e-05,
      "loss": 1.2975,
      "num_input_tokens_seen": 47171288,
      "step": 81315
    },
    {
      "epoch": 12.112004766160263,
      "grad_norm": 13.375,
      "learning_rate": 2.017900501494337e-05,
      "loss": 1.1126,
      "num_input_tokens_seen": 47173848,
      "step": 81320
    },
    {
      "epoch": 12.112749478701222,
      "grad_norm": 21.75,
      "learning_rate": 2.0175816625034795e-05,
      "loss": 1.2617,
      "num_input_tokens_seen": 47176664,
      "step": 81325
    },
    {
      "epoch": 12.11349419124218,
      "grad_norm": 7.71875,
      "learning_rate": 2.0172628316625887e-05,
      "loss": 0.8563,
      "num_input_tokens_seen": 47179640,
      "step": 81330
    },
    {
      "epoch": 12.11423890378314,
      "grad_norm": 8.5,
      "learning_rate": 2.0169440089770523e-05,
      "loss": 1.1196,
      "num_input_tokens_seen": 47182456,
      "step": 81335
    },
    {
      "epoch": 12.114983616324098,
      "grad_norm": 9.0625,
      "learning_rate": 2.0166251944522553e-05,
      "loss": 0.9379,
      "num_input_tokens_seen": 47185400,
      "step": 81340
    },
    {
      "epoch": 12.115728328865059,
      "grad_norm": 10.5625,
      "learning_rate": 2.016306388093585e-05,
      "loss": 1.0791,
      "num_input_tokens_seen": 47188504,
      "step": 81345
    },
    {
      "epoch": 12.116473041406017,
      "grad_norm": 14.0,
      "learning_rate": 2.0159875899064258e-05,
      "loss": 1.2753,
      "num_input_tokens_seen": 47191448,
      "step": 81350
    },
    {
      "epoch": 12.117217753946976,
      "grad_norm": 7.625,
      "learning_rate": 2.0156687998961653e-05,
      "loss": 1.0581,
      "num_input_tokens_seen": 47194264,
      "step": 81355
    },
    {
      "epoch": 12.117962466487935,
      "grad_norm": 9.5,
      "learning_rate": 2.015350018068188e-05,
      "loss": 1.1956,
      "num_input_tokens_seen": 47197016,
      "step": 81360
    },
    {
      "epoch": 12.118707179028895,
      "grad_norm": 8.0,
      "learning_rate": 2.0150312444278795e-05,
      "loss": 0.9875,
      "num_input_tokens_seen": 47199640,
      "step": 81365
    },
    {
      "epoch": 12.119451891569854,
      "grad_norm": 10.6875,
      "learning_rate": 2.0147124789806254e-05,
      "loss": 1.0752,
      "num_input_tokens_seen": 47202488,
      "step": 81370
    },
    {
      "epoch": 12.120196604110813,
      "grad_norm": 13.0,
      "learning_rate": 2.01439372173181e-05,
      "loss": 1.1636,
      "num_input_tokens_seen": 47205432,
      "step": 81375
    },
    {
      "epoch": 12.120941316651772,
      "grad_norm": 7.34375,
      "learning_rate": 2.0140749726868197e-05,
      "loss": 1.0115,
      "num_input_tokens_seen": 47208376,
      "step": 81380
    },
    {
      "epoch": 12.121686029192732,
      "grad_norm": 17.75,
      "learning_rate": 2.013756231851038e-05,
      "loss": 1.1585,
      "num_input_tokens_seen": 47211352,
      "step": 81385
    },
    {
      "epoch": 12.122430741733691,
      "grad_norm": 4.28125,
      "learning_rate": 2.0134374992298515e-05,
      "loss": 0.9077,
      "num_input_tokens_seen": 47214104,
      "step": 81390
    },
    {
      "epoch": 12.12317545427465,
      "grad_norm": 11.25,
      "learning_rate": 2.0131187748286438e-05,
      "loss": 1.3996,
      "num_input_tokens_seen": 47216728,
      "step": 81395
    },
    {
      "epoch": 12.123920166815608,
      "grad_norm": 9.125,
      "learning_rate": 2.0128000586528e-05,
      "loss": 0.9423,
      "num_input_tokens_seen": 47219576,
      "step": 81400
    },
    {
      "epoch": 12.124664879356569,
      "grad_norm": 12.125,
      "learning_rate": 2.012481350707704e-05,
      "loss": 0.8897,
      "num_input_tokens_seen": 47222744,
      "step": 81405
    },
    {
      "epoch": 12.125409591897528,
      "grad_norm": 12.0,
      "learning_rate": 2.012162650998739e-05,
      "loss": 1.1533,
      "num_input_tokens_seen": 47225816,
      "step": 81410
    },
    {
      "epoch": 12.126154304438487,
      "grad_norm": 11.1875,
      "learning_rate": 2.011843959531291e-05,
      "loss": 0.9932,
      "num_input_tokens_seen": 47228600,
      "step": 81415
    },
    {
      "epoch": 12.126899016979445,
      "grad_norm": 10.0625,
      "learning_rate": 2.0115252763107424e-05,
      "loss": 1.2385,
      "num_input_tokens_seen": 47231672,
      "step": 81420
    },
    {
      "epoch": 12.127643729520406,
      "grad_norm": 16.75,
      "learning_rate": 2.0112066013424785e-05,
      "loss": 1.2691,
      "num_input_tokens_seen": 47234392,
      "step": 81425
    },
    {
      "epoch": 12.128388442061365,
      "grad_norm": 5.96875,
      "learning_rate": 2.010887934631882e-05,
      "loss": 1.0475,
      "num_input_tokens_seen": 47237624,
      "step": 81430
    },
    {
      "epoch": 12.129133154602323,
      "grad_norm": 7.96875,
      "learning_rate": 2.0105692761843375e-05,
      "loss": 1.0461,
      "num_input_tokens_seen": 47240600,
      "step": 81435
    },
    {
      "epoch": 12.129877867143282,
      "grad_norm": 16.75,
      "learning_rate": 2.0102506260052273e-05,
      "loss": 1.1193,
      "num_input_tokens_seen": 47243448,
      "step": 81440
    },
    {
      "epoch": 12.130622579684243,
      "grad_norm": 13.5,
      "learning_rate": 2.0099319840999343e-05,
      "loss": 0.9755,
      "num_input_tokens_seen": 47246456,
      "step": 81445
    },
    {
      "epoch": 12.131367292225201,
      "grad_norm": 11.5625,
      "learning_rate": 2.0096133504738428e-05,
      "loss": 0.9457,
      "num_input_tokens_seen": 47249496,
      "step": 81450
    },
    {
      "epoch": 12.13211200476616,
      "grad_norm": 8.875,
      "learning_rate": 2.009294725132335e-05,
      "loss": 0.732,
      "num_input_tokens_seen": 47252312,
      "step": 81455
    },
    {
      "epoch": 12.132856717307119,
      "grad_norm": 9.0625,
      "learning_rate": 2.0089761080807945e-05,
      "loss": 0.6371,
      "num_input_tokens_seen": 47255224,
      "step": 81460
    },
    {
      "epoch": 12.13360142984808,
      "grad_norm": 13.5625,
      "learning_rate": 2.0086574993246034e-05,
      "loss": 1.0653,
      "num_input_tokens_seen": 47257944,
      "step": 81465
    },
    {
      "epoch": 12.134346142389038,
      "grad_norm": 8.0,
      "learning_rate": 2.008338898869145e-05,
      "loss": 0.946,
      "num_input_tokens_seen": 47261080,
      "step": 81470
    },
    {
      "epoch": 12.135090854929997,
      "grad_norm": 7.59375,
      "learning_rate": 2.0080203067198003e-05,
      "loss": 1.1595,
      "num_input_tokens_seen": 47263864,
      "step": 81475
    },
    {
      "epoch": 12.135835567470956,
      "grad_norm": 13.0,
      "learning_rate": 2.0077017228819534e-05,
      "loss": 1.3251,
      "num_input_tokens_seen": 47267064,
      "step": 81480
    },
    {
      "epoch": 12.136580280011916,
      "grad_norm": 7.15625,
      "learning_rate": 2.0073831473609855e-05,
      "loss": 0.7587,
      "num_input_tokens_seen": 47269976,
      "step": 81485
    },
    {
      "epoch": 12.137324992552875,
      "grad_norm": 10.8125,
      "learning_rate": 2.007064580162278e-05,
      "loss": 1.2418,
      "num_input_tokens_seen": 47272952,
      "step": 81490
    },
    {
      "epoch": 12.138069705093834,
      "grad_norm": 7.3125,
      "learning_rate": 2.0067460212912137e-05,
      "loss": 1.4862,
      "num_input_tokens_seen": 47275736,
      "step": 81495
    },
    {
      "epoch": 12.138814417634793,
      "grad_norm": 16.125,
      "learning_rate": 2.006427470753174e-05,
      "loss": 1.0242,
      "num_input_tokens_seen": 47278616,
      "step": 81500
    },
    {
      "epoch": 12.139559130175751,
      "grad_norm": 6.4375,
      "learning_rate": 2.006108928553541e-05,
      "loss": 0.8153,
      "num_input_tokens_seen": 47281368,
      "step": 81505
    },
    {
      "epoch": 12.140303842716712,
      "grad_norm": 8.625,
      "learning_rate": 2.0057903946976944e-05,
      "loss": 0.9481,
      "num_input_tokens_seen": 47284248,
      "step": 81510
    },
    {
      "epoch": 12.14104855525767,
      "grad_norm": 16.375,
      "learning_rate": 2.0054718691910178e-05,
      "loss": 1.0466,
      "num_input_tokens_seen": 47287096,
      "step": 81515
    },
    {
      "epoch": 12.14179326779863,
      "grad_norm": 14.5625,
      "learning_rate": 2.0051533520388918e-05,
      "loss": 0.8062,
      "num_input_tokens_seen": 47290008,
      "step": 81520
    },
    {
      "epoch": 12.142537980339588,
      "grad_norm": 9.5,
      "learning_rate": 2.0048348432466963e-05,
      "loss": 1.1691,
      "num_input_tokens_seen": 47292728,
      "step": 81525
    },
    {
      "epoch": 12.143282692880549,
      "grad_norm": 9.4375,
      "learning_rate": 2.004516342819813e-05,
      "loss": 1.0448,
      "num_input_tokens_seen": 47295416,
      "step": 81530
    },
    {
      "epoch": 12.144027405421507,
      "grad_norm": 10.5,
      "learning_rate": 2.0041978507636222e-05,
      "loss": 0.7569,
      "num_input_tokens_seen": 47298136,
      "step": 81535
    },
    {
      "epoch": 12.144772117962466,
      "grad_norm": 9.375,
      "learning_rate": 2.0038793670835054e-05,
      "loss": 1.0842,
      "num_input_tokens_seen": 47300824,
      "step": 81540
    },
    {
      "epoch": 12.145516830503425,
      "grad_norm": 8.3125,
      "learning_rate": 2.0035608917848415e-05,
      "loss": 0.8469,
      "num_input_tokens_seen": 47303480,
      "step": 81545
    },
    {
      "epoch": 12.146261543044385,
      "grad_norm": 4.25,
      "learning_rate": 2.0032424248730124e-05,
      "loss": 0.8963,
      "num_input_tokens_seen": 47306360,
      "step": 81550
    },
    {
      "epoch": 12.147006255585344,
      "grad_norm": 16.125,
      "learning_rate": 2.0029239663533977e-05,
      "loss": 1.6731,
      "num_input_tokens_seen": 47309304,
      "step": 81555
    },
    {
      "epoch": 12.147750968126303,
      "grad_norm": 11.25,
      "learning_rate": 2.0026055162313778e-05,
      "loss": 0.715,
      "num_input_tokens_seen": 47312088,
      "step": 81560
    },
    {
      "epoch": 12.148495680667262,
      "grad_norm": 6.625,
      "learning_rate": 2.002287074512332e-05,
      "loss": 1.0547,
      "num_input_tokens_seen": 47314840,
      "step": 81565
    },
    {
      "epoch": 12.149240393208222,
      "grad_norm": 8.25,
      "learning_rate": 2.001968641201639e-05,
      "loss": 1.3508,
      "num_input_tokens_seen": 47317528,
      "step": 81570
    },
    {
      "epoch": 12.149985105749181,
      "grad_norm": 17.0,
      "learning_rate": 2.0016502163046815e-05,
      "loss": 1.2143,
      "num_input_tokens_seen": 47320312,
      "step": 81575
    },
    {
      "epoch": 12.15072981829014,
      "grad_norm": 9.5625,
      "learning_rate": 2.0013317998268352e-05,
      "loss": 1.063,
      "num_input_tokens_seen": 47323384,
      "step": 81580
    },
    {
      "epoch": 12.151474530831099,
      "grad_norm": 8.375,
      "learning_rate": 2.0010133917734825e-05,
      "loss": 1.1342,
      "num_input_tokens_seen": 47326648,
      "step": 81585
    },
    {
      "epoch": 12.152219243372059,
      "grad_norm": 17.5,
      "learning_rate": 2.0006949921500012e-05,
      "loss": 1.0923,
      "num_input_tokens_seen": 47329560,
      "step": 81590
    },
    {
      "epoch": 12.152963955913018,
      "grad_norm": 7.09375,
      "learning_rate": 2.0003766009617707e-05,
      "loss": 0.8795,
      "num_input_tokens_seen": 47332248,
      "step": 81595
    },
    {
      "epoch": 12.153708668453977,
      "grad_norm": 8.375,
      "learning_rate": 2.00005821821417e-05,
      "loss": 1.0101,
      "num_input_tokens_seen": 47335800,
      "step": 81600
    },
    {
      "epoch": 12.154453380994935,
      "grad_norm": 8.0625,
      "learning_rate": 1.9997398439125763e-05,
      "loss": 1.0958,
      "num_input_tokens_seen": 47338680,
      "step": 81605
    },
    {
      "epoch": 12.155198093535896,
      "grad_norm": 12.6875,
      "learning_rate": 1.999421478062371e-05,
      "loss": 1.0996,
      "num_input_tokens_seen": 47341688,
      "step": 81610
    },
    {
      "epoch": 12.155942806076855,
      "grad_norm": 14.125,
      "learning_rate": 1.9991031206689294e-05,
      "loss": 1.3373,
      "num_input_tokens_seen": 47344440,
      "step": 81615
    },
    {
      "epoch": 12.156687518617813,
      "grad_norm": 8.625,
      "learning_rate": 1.998784771737633e-05,
      "loss": 1.2409,
      "num_input_tokens_seen": 47347800,
      "step": 81620
    },
    {
      "epoch": 12.157432231158772,
      "grad_norm": 9.125,
      "learning_rate": 1.9984664312738578e-05,
      "loss": 0.8049,
      "num_input_tokens_seen": 47350648,
      "step": 81625
    },
    {
      "epoch": 12.158176943699733,
      "grad_norm": 9.9375,
      "learning_rate": 1.9981480992829832e-05,
      "loss": 0.9927,
      "num_input_tokens_seen": 47353496,
      "step": 81630
    },
    {
      "epoch": 12.158921656240691,
      "grad_norm": 10.5625,
      "learning_rate": 1.997829775770385e-05,
      "loss": 0.972,
      "num_input_tokens_seen": 47356408,
      "step": 81635
    },
    {
      "epoch": 12.15966636878165,
      "grad_norm": 8.625,
      "learning_rate": 1.9975114607414434e-05,
      "loss": 0.939,
      "num_input_tokens_seen": 47359480,
      "step": 81640
    },
    {
      "epoch": 12.160411081322609,
      "grad_norm": 11.9375,
      "learning_rate": 1.9971931542015355e-05,
      "loss": 1.1389,
      "num_input_tokens_seen": 47362392,
      "step": 81645
    },
    {
      "epoch": 12.16115579386357,
      "grad_norm": 6.59375,
      "learning_rate": 1.9968748561560366e-05,
      "loss": 0.8932,
      "num_input_tokens_seen": 47365432,
      "step": 81650
    },
    {
      "epoch": 12.161900506404528,
      "grad_norm": 18.75,
      "learning_rate": 1.996556566610327e-05,
      "loss": 1.3006,
      "num_input_tokens_seen": 47368504,
      "step": 81655
    },
    {
      "epoch": 12.162645218945487,
      "grad_norm": 7.4375,
      "learning_rate": 1.996238285569782e-05,
      "loss": 0.8528,
      "num_input_tokens_seen": 47371352,
      "step": 81660
    },
    {
      "epoch": 12.163389931486446,
      "grad_norm": 9.0625,
      "learning_rate": 1.9959200130397795e-05,
      "loss": 0.9044,
      "num_input_tokens_seen": 47374232,
      "step": 81665
    },
    {
      "epoch": 12.164134644027406,
      "grad_norm": 10.4375,
      "learning_rate": 1.995601749025695e-05,
      "loss": 1.1435,
      "num_input_tokens_seen": 47376888,
      "step": 81670
    },
    {
      "epoch": 12.164879356568365,
      "grad_norm": 8.1875,
      "learning_rate": 1.9952834935329077e-05,
      "loss": 1.2194,
      "num_input_tokens_seen": 47379736,
      "step": 81675
    },
    {
      "epoch": 12.165624069109324,
      "grad_norm": 10.125,
      "learning_rate": 1.9949652465667915e-05,
      "loss": 1.2946,
      "num_input_tokens_seen": 47382744,
      "step": 81680
    },
    {
      "epoch": 12.166368781650283,
      "grad_norm": 9.5,
      "learning_rate": 1.9946470081327253e-05,
      "loss": 1.087,
      "num_input_tokens_seen": 47385624,
      "step": 81685
    },
    {
      "epoch": 12.167113494191241,
      "grad_norm": 19.5,
      "learning_rate": 1.9943287782360844e-05,
      "loss": 1.2342,
      "num_input_tokens_seen": 47388696,
      "step": 81690
    },
    {
      "epoch": 12.167858206732202,
      "grad_norm": 7.8125,
      "learning_rate": 1.9940105568822437e-05,
      "loss": 0.7624,
      "num_input_tokens_seen": 47391256,
      "step": 81695
    },
    {
      "epoch": 12.16860291927316,
      "grad_norm": 28.0,
      "learning_rate": 1.9936923440765813e-05,
      "loss": 1.1365,
      "num_input_tokens_seen": 47394200,
      "step": 81700
    },
    {
      "epoch": 12.16934763181412,
      "grad_norm": 13.9375,
      "learning_rate": 1.9933741398244714e-05,
      "loss": 0.9037,
      "num_input_tokens_seen": 47397048,
      "step": 81705
    },
    {
      "epoch": 12.170092344355078,
      "grad_norm": 13.9375,
      "learning_rate": 1.9930559441312913e-05,
      "loss": 1.2359,
      "num_input_tokens_seen": 47399768,
      "step": 81710
    },
    {
      "epoch": 12.170837056896039,
      "grad_norm": 4.5,
      "learning_rate": 1.9927377570024146e-05,
      "loss": 0.8202,
      "num_input_tokens_seen": 47402392,
      "step": 81715
    },
    {
      "epoch": 12.171581769436997,
      "grad_norm": 8.8125,
      "learning_rate": 1.9924195784432192e-05,
      "loss": 1.0355,
      "num_input_tokens_seen": 47405496,
      "step": 81720
    },
    {
      "epoch": 12.172326481977956,
      "grad_norm": 11.5625,
      "learning_rate": 1.992101408459079e-05,
      "loss": 1.0681,
      "num_input_tokens_seen": 47408376,
      "step": 81725
    },
    {
      "epoch": 12.173071194518915,
      "grad_norm": 7.15625,
      "learning_rate": 1.9917832470553692e-05,
      "loss": 1.2391,
      "num_input_tokens_seen": 47411160,
      "step": 81730
    },
    {
      "epoch": 12.173815907059875,
      "grad_norm": 13.0625,
      "learning_rate": 1.9914650942374648e-05,
      "loss": 1.4113,
      "num_input_tokens_seen": 47414360,
      "step": 81735
    },
    {
      "epoch": 12.174560619600834,
      "grad_norm": 8.9375,
      "learning_rate": 1.9911469500107398e-05,
      "loss": 1.2553,
      "num_input_tokens_seen": 47417112,
      "step": 81740
    },
    {
      "epoch": 12.175305332141793,
      "grad_norm": 7.90625,
      "learning_rate": 1.9908288143805714e-05,
      "loss": 0.9454,
      "num_input_tokens_seen": 47420120,
      "step": 81745
    },
    {
      "epoch": 12.176050044682752,
      "grad_norm": 20.875,
      "learning_rate": 1.9905106873523316e-05,
      "loss": 1.3279,
      "num_input_tokens_seen": 47423000,
      "step": 81750
    },
    {
      "epoch": 12.176794757223712,
      "grad_norm": 11.0625,
      "learning_rate": 1.9901925689313967e-05,
      "loss": 1.1931,
      "num_input_tokens_seen": 47425880,
      "step": 81755
    },
    {
      "epoch": 12.177539469764671,
      "grad_norm": 14.6875,
      "learning_rate": 1.9898744591231396e-05,
      "loss": 1.137,
      "num_input_tokens_seen": 47429080,
      "step": 81760
    },
    {
      "epoch": 12.17828418230563,
      "grad_norm": 14.0625,
      "learning_rate": 1.989556357932936e-05,
      "loss": 1.2012,
      "num_input_tokens_seen": 47432248,
      "step": 81765
    },
    {
      "epoch": 12.179028894846589,
      "grad_norm": 7.4375,
      "learning_rate": 1.9892382653661584e-05,
      "loss": 0.9286,
      "num_input_tokens_seen": 47434872,
      "step": 81770
    },
    {
      "epoch": 12.179773607387549,
      "grad_norm": 7.375,
      "learning_rate": 1.9889201814281804e-05,
      "loss": 1.1591,
      "num_input_tokens_seen": 47437624,
      "step": 81775
    },
    {
      "epoch": 12.180518319928508,
      "grad_norm": 13.625,
      "learning_rate": 1.9886021061243775e-05,
      "loss": 1.1588,
      "num_input_tokens_seen": 47440376,
      "step": 81780
    },
    {
      "epoch": 12.181263032469467,
      "grad_norm": 12.25,
      "learning_rate": 1.9882840394601213e-05,
      "loss": 1.2998,
      "num_input_tokens_seen": 47443672,
      "step": 81785
    },
    {
      "epoch": 12.182007745010425,
      "grad_norm": 4.6875,
      "learning_rate": 1.987965981440787e-05,
      "loss": 0.9684,
      "num_input_tokens_seen": 47446584,
      "step": 81790
    },
    {
      "epoch": 12.182752457551386,
      "grad_norm": 9.0,
      "learning_rate": 1.987647932071747e-05,
      "loss": 1.0361,
      "num_input_tokens_seen": 47449496,
      "step": 81795
    },
    {
      "epoch": 12.183497170092345,
      "grad_norm": 10.8125,
      "learning_rate": 1.9873298913583746e-05,
      "loss": 1.1239,
      "num_input_tokens_seen": 47452344,
      "step": 81800
    },
    {
      "epoch": 12.184241882633303,
      "grad_norm": 11.75,
      "learning_rate": 1.987011859306043e-05,
      "loss": 1.2623,
      "num_input_tokens_seen": 47455192,
      "step": 81805
    },
    {
      "epoch": 12.184986595174262,
      "grad_norm": 14.8125,
      "learning_rate": 1.986693835920123e-05,
      "loss": 1.0777,
      "num_input_tokens_seen": 47458072,
      "step": 81810
    },
    {
      "epoch": 12.185731307715223,
      "grad_norm": 12.5625,
      "learning_rate": 1.9863758212059902e-05,
      "loss": 1.227,
      "num_input_tokens_seen": 47460888,
      "step": 81815
    },
    {
      "epoch": 12.186476020256181,
      "grad_norm": 14.25,
      "learning_rate": 1.9860578151690154e-05,
      "loss": 1.1277,
      "num_input_tokens_seen": 47463480,
      "step": 81820
    },
    {
      "epoch": 12.18722073279714,
      "grad_norm": 19.625,
      "learning_rate": 1.9857398178145718e-05,
      "loss": 1.1346,
      "num_input_tokens_seen": 47466648,
      "step": 81825
    },
    {
      "epoch": 12.187965445338099,
      "grad_norm": 7.34375,
      "learning_rate": 1.985421829148031e-05,
      "loss": 1.1508,
      "num_input_tokens_seen": 47469752,
      "step": 81830
    },
    {
      "epoch": 12.18871015787906,
      "grad_norm": 13.875,
      "learning_rate": 1.985103849174766e-05,
      "loss": 0.9724,
      "num_input_tokens_seen": 47472792,
      "step": 81835
    },
    {
      "epoch": 12.189454870420018,
      "grad_norm": 12.1875,
      "learning_rate": 1.984785877900147e-05,
      "loss": 1.6517,
      "num_input_tokens_seen": 47475384,
      "step": 81840
    },
    {
      "epoch": 12.190199582960977,
      "grad_norm": 6.125,
      "learning_rate": 1.984467915329548e-05,
      "loss": 1.2308,
      "num_input_tokens_seen": 47478296,
      "step": 81845
    },
    {
      "epoch": 12.190944295501936,
      "grad_norm": 9.0,
      "learning_rate": 1.9841499614683394e-05,
      "loss": 0.7771,
      "num_input_tokens_seen": 47481368,
      "step": 81850
    },
    {
      "epoch": 12.191689008042895,
      "grad_norm": 17.875,
      "learning_rate": 1.9838320163218927e-05,
      "loss": 1.0901,
      "num_input_tokens_seen": 47484248,
      "step": 81855
    },
    {
      "epoch": 12.192433720583855,
      "grad_norm": 14.875,
      "learning_rate": 1.98351407989558e-05,
      "loss": 1.1164,
      "num_input_tokens_seen": 47487544,
      "step": 81860
    },
    {
      "epoch": 12.193178433124814,
      "grad_norm": 7.78125,
      "learning_rate": 1.983196152194771e-05,
      "loss": 1.1787,
      "num_input_tokens_seen": 47490424,
      "step": 81865
    },
    {
      "epoch": 12.193923145665773,
      "grad_norm": 4.78125,
      "learning_rate": 1.9828782332248385e-05,
      "loss": 0.9103,
      "num_input_tokens_seen": 47493304,
      "step": 81870
    },
    {
      "epoch": 12.194667858206731,
      "grad_norm": 9.5625,
      "learning_rate": 1.982560322991152e-05,
      "loss": 1.3434,
      "num_input_tokens_seen": 47496184,
      "step": 81875
    },
    {
      "epoch": 12.195412570747692,
      "grad_norm": 6.6875,
      "learning_rate": 1.9822424214990837e-05,
      "loss": 0.968,
      "num_input_tokens_seen": 47498840,
      "step": 81880
    },
    {
      "epoch": 12.19615728328865,
      "grad_norm": 13.1875,
      "learning_rate": 1.9819245287540036e-05,
      "loss": 1.4419,
      "num_input_tokens_seen": 47501880,
      "step": 81885
    },
    {
      "epoch": 12.19690199582961,
      "grad_norm": 4.75,
      "learning_rate": 1.9816066447612815e-05,
      "loss": 1.0334,
      "num_input_tokens_seen": 47504728,
      "step": 81890
    },
    {
      "epoch": 12.197646708370568,
      "grad_norm": 8.5,
      "learning_rate": 1.9812887695262887e-05,
      "loss": 1.3756,
      "num_input_tokens_seen": 47507640,
      "step": 81895
    },
    {
      "epoch": 12.198391420911529,
      "grad_norm": 17.0,
      "learning_rate": 1.980970903054394e-05,
      "loss": 0.9224,
      "num_input_tokens_seen": 47510328,
      "step": 81900
    },
    {
      "epoch": 12.199136133452487,
      "grad_norm": 10.125,
      "learning_rate": 1.9806530453509693e-05,
      "loss": 0.7286,
      "num_input_tokens_seen": 47513016,
      "step": 81905
    },
    {
      "epoch": 12.199880845993446,
      "grad_norm": 6.9375,
      "learning_rate": 1.9803351964213827e-05,
      "loss": 1.3963,
      "num_input_tokens_seen": 47516024,
      "step": 81910
    },
    {
      "epoch": 12.200625558534405,
      "grad_norm": 12.1875,
      "learning_rate": 1.9800173562710055e-05,
      "loss": 1.2017,
      "num_input_tokens_seen": 47518776,
      "step": 81915
    },
    {
      "epoch": 12.201370271075366,
      "grad_norm": 11.375,
      "learning_rate": 1.9796995249052064e-05,
      "loss": 0.8621,
      "num_input_tokens_seen": 47521592,
      "step": 81920
    },
    {
      "epoch": 12.202114983616324,
      "grad_norm": 6.65625,
      "learning_rate": 1.9793817023293555e-05,
      "loss": 0.9921,
      "num_input_tokens_seen": 47524632,
      "step": 81925
    },
    {
      "epoch": 12.202859696157283,
      "grad_norm": 6.90625,
      "learning_rate": 1.9790638885488216e-05,
      "loss": 0.974,
      "num_input_tokens_seen": 47527768,
      "step": 81930
    },
    {
      "epoch": 12.203604408698242,
      "grad_norm": 14.8125,
      "learning_rate": 1.9787460835689726e-05,
      "loss": 1.1661,
      "num_input_tokens_seen": 47530616,
      "step": 81935
    },
    {
      "epoch": 12.204349121239202,
      "grad_norm": 10.5625,
      "learning_rate": 1.9784282873951797e-05,
      "loss": 1.1126,
      "num_input_tokens_seen": 47533976,
      "step": 81940
    },
    {
      "epoch": 12.205093833780161,
      "grad_norm": 9.3125,
      "learning_rate": 1.97811050003281e-05,
      "loss": 0.9585,
      "num_input_tokens_seen": 47536792,
      "step": 81945
    },
    {
      "epoch": 12.20583854632112,
      "grad_norm": 7.65625,
      "learning_rate": 1.977792721487234e-05,
      "loss": 1.0593,
      "num_input_tokens_seen": 47539768,
      "step": 81950
    },
    {
      "epoch": 12.206583258862079,
      "grad_norm": 10.625,
      "learning_rate": 1.9774749517638188e-05,
      "loss": 1.2211,
      "num_input_tokens_seen": 47542392,
      "step": 81955
    },
    {
      "epoch": 12.20732797140304,
      "grad_norm": 14.4375,
      "learning_rate": 1.9771571908679337e-05,
      "loss": 1.1072,
      "num_input_tokens_seen": 47545304,
      "step": 81960
    },
    {
      "epoch": 12.208072683943998,
      "grad_norm": 11.6875,
      "learning_rate": 1.976839438804946e-05,
      "loss": 1.2732,
      "num_input_tokens_seen": 47548088,
      "step": 81965
    },
    {
      "epoch": 12.208817396484957,
      "grad_norm": 22.125,
      "learning_rate": 1.976521695580224e-05,
      "loss": 1.1414,
      "num_input_tokens_seen": 47551032,
      "step": 81970
    },
    {
      "epoch": 12.209562109025915,
      "grad_norm": 9.875,
      "learning_rate": 1.9762039611991365e-05,
      "loss": 1.328,
      "num_input_tokens_seen": 47553848,
      "step": 81975
    },
    {
      "epoch": 12.210306821566876,
      "grad_norm": 10.6875,
      "learning_rate": 1.9758862356670498e-05,
      "loss": 0.864,
      "num_input_tokens_seen": 47556728,
      "step": 81980
    },
    {
      "epoch": 12.211051534107835,
      "grad_norm": 13.375,
      "learning_rate": 1.9755685189893332e-05,
      "loss": 0.8934,
      "num_input_tokens_seen": 47559640,
      "step": 81985
    },
    {
      "epoch": 12.211796246648793,
      "grad_norm": 9.875,
      "learning_rate": 1.9752508111713532e-05,
      "loss": 0.8913,
      "num_input_tokens_seen": 47562712,
      "step": 81990
    },
    {
      "epoch": 12.212540959189752,
      "grad_norm": 8.9375,
      "learning_rate": 1.974933112218478e-05,
      "loss": 0.9108,
      "num_input_tokens_seen": 47565720,
      "step": 81995
    },
    {
      "epoch": 12.213285671730713,
      "grad_norm": 9.1875,
      "learning_rate": 1.9746154221360732e-05,
      "loss": 1.0508,
      "num_input_tokens_seen": 47568600,
      "step": 82000
    },
    {
      "epoch": 12.214030384271672,
      "grad_norm": 13.0,
      "learning_rate": 1.9742977409295076e-05,
      "loss": 1.1224,
      "num_input_tokens_seen": 47571448,
      "step": 82005
    },
    {
      "epoch": 12.21477509681263,
      "grad_norm": 12.375,
      "learning_rate": 1.9739800686041478e-05,
      "loss": 0.5204,
      "num_input_tokens_seen": 47574328,
      "step": 82010
    },
    {
      "epoch": 12.215519809353589,
      "grad_norm": 11.875,
      "learning_rate": 1.973662405165359e-05,
      "loss": 1.0125,
      "num_input_tokens_seen": 47577336,
      "step": 82015
    },
    {
      "epoch": 12.216264521894548,
      "grad_norm": 22.125,
      "learning_rate": 1.9733447506185095e-05,
      "loss": 1.6872,
      "num_input_tokens_seen": 47580152,
      "step": 82020
    },
    {
      "epoch": 12.217009234435508,
      "grad_norm": 16.125,
      "learning_rate": 1.973027104968965e-05,
      "loss": 1.1703,
      "num_input_tokens_seen": 47583096,
      "step": 82025
    },
    {
      "epoch": 12.217753946976467,
      "grad_norm": 5.28125,
      "learning_rate": 1.9727094682220925e-05,
      "loss": 1.0397,
      "num_input_tokens_seen": 47585976,
      "step": 82030
    },
    {
      "epoch": 12.218498659517426,
      "grad_norm": 7.09375,
      "learning_rate": 1.9723918403832565e-05,
      "loss": 1.0576,
      "num_input_tokens_seen": 47588664,
      "step": 82035
    },
    {
      "epoch": 12.219243372058385,
      "grad_norm": 11.1875,
      "learning_rate": 1.9720742214578254e-05,
      "loss": 1.1109,
      "num_input_tokens_seen": 47591608,
      "step": 82040
    },
    {
      "epoch": 12.219988084599345,
      "grad_norm": 12.5625,
      "learning_rate": 1.9717566114511634e-05,
      "loss": 1.2327,
      "num_input_tokens_seen": 47595512,
      "step": 82045
    },
    {
      "epoch": 12.220732797140304,
      "grad_norm": 17.125,
      "learning_rate": 1.9714390103686355e-05,
      "loss": 1.1691,
      "num_input_tokens_seen": 47598328,
      "step": 82050
    },
    {
      "epoch": 12.221477509681263,
      "grad_norm": 10.375,
      "learning_rate": 1.9711214182156096e-05,
      "loss": 1.2965,
      "num_input_tokens_seen": 47601304,
      "step": 82055
    },
    {
      "epoch": 12.222222222222221,
      "grad_norm": 13.6875,
      "learning_rate": 1.9708038349974494e-05,
      "loss": 0.9429,
      "num_input_tokens_seen": 47604120,
      "step": 82060
    },
    {
      "epoch": 12.222966934763182,
      "grad_norm": 17.125,
      "learning_rate": 1.9704862607195207e-05,
      "loss": 1.0961,
      "num_input_tokens_seen": 47607224,
      "step": 82065
    },
    {
      "epoch": 12.22371164730414,
      "grad_norm": 12.375,
      "learning_rate": 1.970168695387188e-05,
      "loss": 0.9862,
      "num_input_tokens_seen": 47610168,
      "step": 82070
    },
    {
      "epoch": 12.2244563598451,
      "grad_norm": 9.125,
      "learning_rate": 1.9698511390058172e-05,
      "loss": 0.9603,
      "num_input_tokens_seen": 47612920,
      "step": 82075
    },
    {
      "epoch": 12.225201072386058,
      "grad_norm": 14.125,
      "learning_rate": 1.9695335915807716e-05,
      "loss": 1.1155,
      "num_input_tokens_seen": 47616056,
      "step": 82080
    },
    {
      "epoch": 12.225945784927019,
      "grad_norm": 19.625,
      "learning_rate": 1.969216053117418e-05,
      "loss": 1.4534,
      "num_input_tokens_seen": 47619000,
      "step": 82085
    },
    {
      "epoch": 12.226690497467978,
      "grad_norm": 13.875,
      "learning_rate": 1.9688985236211197e-05,
      "loss": 1.0641,
      "num_input_tokens_seen": 47621848,
      "step": 82090
    },
    {
      "epoch": 12.227435210008936,
      "grad_norm": 8.125,
      "learning_rate": 1.9685810030972405e-05,
      "loss": 0.8835,
      "num_input_tokens_seen": 47624856,
      "step": 82095
    },
    {
      "epoch": 12.228179922549895,
      "grad_norm": 12.875,
      "learning_rate": 1.9682634915511455e-05,
      "loss": 1.0807,
      "num_input_tokens_seen": 47627928,
      "step": 82100
    },
    {
      "epoch": 12.228924635090856,
      "grad_norm": 14.9375,
      "learning_rate": 1.9679459889881977e-05,
      "loss": 1.424,
      "num_input_tokens_seen": 47630904,
      "step": 82105
    },
    {
      "epoch": 12.229669347631814,
      "grad_norm": 7.46875,
      "learning_rate": 1.9676284954137624e-05,
      "loss": 1.3281,
      "num_input_tokens_seen": 47633624,
      "step": 82110
    },
    {
      "epoch": 12.230414060172773,
      "grad_norm": 16.125,
      "learning_rate": 1.9673110108332014e-05,
      "loss": 1.2647,
      "num_input_tokens_seen": 47636536,
      "step": 82115
    },
    {
      "epoch": 12.231158772713732,
      "grad_norm": 14.1875,
      "learning_rate": 1.966993535251881e-05,
      "loss": 1.0934,
      "num_input_tokens_seen": 47639640,
      "step": 82120
    },
    {
      "epoch": 12.231903485254692,
      "grad_norm": 11.125,
      "learning_rate": 1.966676068675163e-05,
      "loss": 1.1413,
      "num_input_tokens_seen": 47642584,
      "step": 82125
    },
    {
      "epoch": 12.232648197795651,
      "grad_norm": 11.0625,
      "learning_rate": 1.96635861110841e-05,
      "loss": 1.213,
      "num_input_tokens_seen": 47645208,
      "step": 82130
    },
    {
      "epoch": 12.23339291033661,
      "grad_norm": 5.28125,
      "learning_rate": 1.9660411625569867e-05,
      "loss": 0.9031,
      "num_input_tokens_seen": 47647992,
      "step": 82135
    },
    {
      "epoch": 12.234137622877569,
      "grad_norm": 6.78125,
      "learning_rate": 1.965723723026254e-05,
      "loss": 1.0567,
      "num_input_tokens_seen": 47650936,
      "step": 82140
    },
    {
      "epoch": 12.23488233541853,
      "grad_norm": 17.625,
      "learning_rate": 1.965406292521577e-05,
      "loss": 0.9271,
      "num_input_tokens_seen": 47653784,
      "step": 82145
    },
    {
      "epoch": 12.235627047959488,
      "grad_norm": 11.5,
      "learning_rate": 1.965088871048317e-05,
      "loss": 1.2015,
      "num_input_tokens_seen": 47656504,
      "step": 82150
    },
    {
      "epoch": 12.236371760500447,
      "grad_norm": 10.4375,
      "learning_rate": 1.964771458611837e-05,
      "loss": 0.7933,
      "num_input_tokens_seen": 47659224,
      "step": 82155
    },
    {
      "epoch": 12.237116473041405,
      "grad_norm": 7.46875,
      "learning_rate": 1.964454055217499e-05,
      "loss": 1.293,
      "num_input_tokens_seen": 47662712,
      "step": 82160
    },
    {
      "epoch": 12.237861185582366,
      "grad_norm": 20.375,
      "learning_rate": 1.9641366608706656e-05,
      "loss": 1.0681,
      "num_input_tokens_seen": 47665720,
      "step": 82165
    },
    {
      "epoch": 12.238605898123325,
      "grad_norm": 5.59375,
      "learning_rate": 1.9638192755766993e-05,
      "loss": 1.0934,
      "num_input_tokens_seen": 47668792,
      "step": 82170
    },
    {
      "epoch": 12.239350610664284,
      "grad_norm": 9.9375,
      "learning_rate": 1.9635018993409602e-05,
      "loss": 1.0301,
      "num_input_tokens_seen": 47671736,
      "step": 82175
    },
    {
      "epoch": 12.240095323205242,
      "grad_norm": 19.0,
      "learning_rate": 1.963184532168812e-05,
      "loss": 1.2286,
      "num_input_tokens_seen": 47674264,
      "step": 82180
    },
    {
      "epoch": 12.240840035746203,
      "grad_norm": 10.0,
      "learning_rate": 1.9628671740656154e-05,
      "loss": 1.1867,
      "num_input_tokens_seen": 47676920,
      "step": 82185
    },
    {
      "epoch": 12.241584748287162,
      "grad_norm": 11.25,
      "learning_rate": 1.962549825036732e-05,
      "loss": 0.9909,
      "num_input_tokens_seen": 47679832,
      "step": 82190
    },
    {
      "epoch": 12.24232946082812,
      "grad_norm": 8.375,
      "learning_rate": 1.9622324850875227e-05,
      "loss": 1.1868,
      "num_input_tokens_seen": 47682392,
      "step": 82195
    },
    {
      "epoch": 12.243074173369079,
      "grad_norm": 13.9375,
      "learning_rate": 1.9619151542233494e-05,
      "loss": 1.2445,
      "num_input_tokens_seen": 47685240,
      "step": 82200
    },
    {
      "epoch": 12.243818885910038,
      "grad_norm": 12.875,
      "learning_rate": 1.9615978324495733e-05,
      "loss": 1.0112,
      "num_input_tokens_seen": 47688248,
      "step": 82205
    },
    {
      "epoch": 12.244563598450998,
      "grad_norm": 10.0,
      "learning_rate": 1.961280519771553e-05,
      "loss": 1.1687,
      "num_input_tokens_seen": 47691288,
      "step": 82210
    },
    {
      "epoch": 12.245308310991957,
      "grad_norm": 15.0,
      "learning_rate": 1.960963216194652e-05,
      "loss": 1.2526,
      "num_input_tokens_seen": 47694040,
      "step": 82215
    },
    {
      "epoch": 12.246053023532916,
      "grad_norm": 15.5,
      "learning_rate": 1.960645921724229e-05,
      "loss": 1.577,
      "num_input_tokens_seen": 47696856,
      "step": 82220
    },
    {
      "epoch": 12.246797736073875,
      "grad_norm": 7.4375,
      "learning_rate": 1.960328636365646e-05,
      "loss": 1.0418,
      "num_input_tokens_seen": 47699832,
      "step": 82225
    },
    {
      "epoch": 12.247542448614835,
      "grad_norm": 20.125,
      "learning_rate": 1.9600113601242605e-05,
      "loss": 1.0025,
      "num_input_tokens_seen": 47702904,
      "step": 82230
    },
    {
      "epoch": 12.248287161155794,
      "grad_norm": 4.65625,
      "learning_rate": 1.9596940930054358e-05,
      "loss": 0.9061,
      "num_input_tokens_seen": 47705912,
      "step": 82235
    },
    {
      "epoch": 12.249031873696753,
      "grad_norm": 14.9375,
      "learning_rate": 1.9593768350145288e-05,
      "loss": 1.0256,
      "num_input_tokens_seen": 47708536,
      "step": 82240
    },
    {
      "epoch": 12.249776586237711,
      "grad_norm": 9.125,
      "learning_rate": 1.9590595861569023e-05,
      "loss": 1.0801,
      "num_input_tokens_seen": 47711544,
      "step": 82245
    },
    {
      "epoch": 12.250521298778672,
      "grad_norm": 12.625,
      "learning_rate": 1.9587423464379136e-05,
      "loss": 0.9364,
      "num_input_tokens_seen": 47714424,
      "step": 82250
    },
    {
      "epoch": 12.25126601131963,
      "grad_norm": 10.8125,
      "learning_rate": 1.9584251158629228e-05,
      "loss": 1.0573,
      "num_input_tokens_seen": 47717272,
      "step": 82255
    },
    {
      "epoch": 12.25201072386059,
      "grad_norm": 10.4375,
      "learning_rate": 1.9581078944372897e-05,
      "loss": 1.0481,
      "num_input_tokens_seen": 47720056,
      "step": 82260
    },
    {
      "epoch": 12.252755436401548,
      "grad_norm": 10.0,
      "learning_rate": 1.957790682166372e-05,
      "loss": 0.9598,
      "num_input_tokens_seen": 47722936,
      "step": 82265
    },
    {
      "epoch": 12.253500148942509,
      "grad_norm": 9.25,
      "learning_rate": 1.9574734790555305e-05,
      "loss": 1.1521,
      "num_input_tokens_seen": 47725656,
      "step": 82270
    },
    {
      "epoch": 12.254244861483468,
      "grad_norm": 12.5625,
      "learning_rate": 1.9571562851101223e-05,
      "loss": 0.9889,
      "num_input_tokens_seen": 47728536,
      "step": 82275
    },
    {
      "epoch": 12.254989574024426,
      "grad_norm": 7.0625,
      "learning_rate": 1.9568391003355073e-05,
      "loss": 0.9546,
      "num_input_tokens_seen": 47732024,
      "step": 82280
    },
    {
      "epoch": 12.255734286565385,
      "grad_norm": 13.0,
      "learning_rate": 1.956521924737044e-05,
      "loss": 1.3111,
      "num_input_tokens_seen": 47734712,
      "step": 82285
    },
    {
      "epoch": 12.256478999106346,
      "grad_norm": 16.0,
      "learning_rate": 1.9562047583200906e-05,
      "loss": 1.2922,
      "num_input_tokens_seen": 47737432,
      "step": 82290
    },
    {
      "epoch": 12.257223711647304,
      "grad_norm": 7.25,
      "learning_rate": 1.955887601090005e-05,
      "loss": 1.2411,
      "num_input_tokens_seen": 47739896,
      "step": 82295
    },
    {
      "epoch": 12.257968424188263,
      "grad_norm": 14.6875,
      "learning_rate": 1.9555704530521445e-05,
      "loss": 1.1796,
      "num_input_tokens_seen": 47742968,
      "step": 82300
    },
    {
      "epoch": 12.258713136729222,
      "grad_norm": 12.4375,
      "learning_rate": 1.955253314211869e-05,
      "loss": 0.9022,
      "num_input_tokens_seen": 47745528,
      "step": 82305
    },
    {
      "epoch": 12.259457849270182,
      "grad_norm": 11.8125,
      "learning_rate": 1.9549361845745338e-05,
      "loss": 0.9465,
      "num_input_tokens_seen": 47748376,
      "step": 82310
    },
    {
      "epoch": 12.260202561811141,
      "grad_norm": 8.625,
      "learning_rate": 1.9546190641454993e-05,
      "loss": 1.1953,
      "num_input_tokens_seen": 47751288,
      "step": 82315
    },
    {
      "epoch": 12.2609472743521,
      "grad_norm": 7.15625,
      "learning_rate": 1.9543019529301203e-05,
      "loss": 1.1003,
      "num_input_tokens_seen": 47754008,
      "step": 82320
    },
    {
      "epoch": 12.261691986893059,
      "grad_norm": 12.5625,
      "learning_rate": 1.953984850933756e-05,
      "loss": 1.0104,
      "num_input_tokens_seen": 47756888,
      "step": 82325
    },
    {
      "epoch": 12.26243669943402,
      "grad_norm": 10.375,
      "learning_rate": 1.953667758161763e-05,
      "loss": 1.4095,
      "num_input_tokens_seen": 47760088,
      "step": 82330
    },
    {
      "epoch": 12.263181411974978,
      "grad_norm": 14.375,
      "learning_rate": 1.9533506746194964e-05,
      "loss": 0.8269,
      "num_input_tokens_seen": 47763160,
      "step": 82335
    },
    {
      "epoch": 12.263926124515937,
      "grad_norm": 7.3125,
      "learning_rate": 1.9530336003123156e-05,
      "loss": 1.2442,
      "num_input_tokens_seen": 47766104,
      "step": 82340
    },
    {
      "epoch": 12.264670837056896,
      "grad_norm": 8.9375,
      "learning_rate": 1.9527165352455755e-05,
      "loss": 1.2888,
      "num_input_tokens_seen": 47768952,
      "step": 82345
    },
    {
      "epoch": 12.265415549597856,
      "grad_norm": 11.8125,
      "learning_rate": 1.9523994794246344e-05,
      "loss": 1.0545,
      "num_input_tokens_seen": 47771640,
      "step": 82350
    },
    {
      "epoch": 12.266160262138815,
      "grad_norm": 16.5,
      "learning_rate": 1.9520824328548465e-05,
      "loss": 1.0319,
      "num_input_tokens_seen": 47774552,
      "step": 82355
    },
    {
      "epoch": 12.266904974679774,
      "grad_norm": 8.9375,
      "learning_rate": 1.9517653955415698e-05,
      "loss": 0.8301,
      "num_input_tokens_seen": 47777432,
      "step": 82360
    },
    {
      "epoch": 12.267649687220732,
      "grad_norm": 10.625,
      "learning_rate": 1.9514483674901586e-05,
      "loss": 0.9207,
      "num_input_tokens_seen": 47780280,
      "step": 82365
    },
    {
      "epoch": 12.268394399761693,
      "grad_norm": 9.3125,
      "learning_rate": 1.9511313487059706e-05,
      "loss": 1.0498,
      "num_input_tokens_seen": 47783064,
      "step": 82370
    },
    {
      "epoch": 12.269139112302652,
      "grad_norm": 8.5625,
      "learning_rate": 1.950814339194361e-05,
      "loss": 1.1072,
      "num_input_tokens_seen": 47786072,
      "step": 82375
    },
    {
      "epoch": 12.26988382484361,
      "grad_norm": 7.15625,
      "learning_rate": 1.9504973389606834e-05,
      "loss": 0.9289,
      "num_input_tokens_seen": 47788888,
      "step": 82380
    },
    {
      "epoch": 12.27062853738457,
      "grad_norm": 10.5,
      "learning_rate": 1.9501803480102962e-05,
      "loss": 1.0176,
      "num_input_tokens_seen": 47792056,
      "step": 82385
    },
    {
      "epoch": 12.271373249925528,
      "grad_norm": 18.5,
      "learning_rate": 1.9498633663485526e-05,
      "loss": 1.3163,
      "num_input_tokens_seen": 47794712,
      "step": 82390
    },
    {
      "epoch": 12.272117962466488,
      "grad_norm": 10.0,
      "learning_rate": 1.9495463939808085e-05,
      "loss": 1.2308,
      "num_input_tokens_seen": 47797464,
      "step": 82395
    },
    {
      "epoch": 12.272862675007447,
      "grad_norm": 6.625,
      "learning_rate": 1.9492294309124183e-05,
      "loss": 0.8886,
      "num_input_tokens_seen": 47800120,
      "step": 82400
    },
    {
      "epoch": 12.273607387548406,
      "grad_norm": 6.8125,
      "learning_rate": 1.9489124771487375e-05,
      "loss": 0.8304,
      "num_input_tokens_seen": 47803032,
      "step": 82405
    },
    {
      "epoch": 12.274352100089365,
      "grad_norm": 11.125,
      "learning_rate": 1.9485955326951204e-05,
      "loss": 1.1183,
      "num_input_tokens_seen": 47805912,
      "step": 82410
    },
    {
      "epoch": 12.275096812630325,
      "grad_norm": 6.0,
      "learning_rate": 1.9482785975569202e-05,
      "loss": 0.8618,
      "num_input_tokens_seen": 47809144,
      "step": 82415
    },
    {
      "epoch": 12.275841525171284,
      "grad_norm": 12.4375,
      "learning_rate": 1.9479616717394937e-05,
      "loss": 1.1448,
      "num_input_tokens_seen": 47811800,
      "step": 82420
    },
    {
      "epoch": 12.276586237712243,
      "grad_norm": 6.375,
      "learning_rate": 1.947644755248193e-05,
      "loss": 0.6003,
      "num_input_tokens_seen": 47814616,
      "step": 82425
    },
    {
      "epoch": 12.277330950253202,
      "grad_norm": 13.3125,
      "learning_rate": 1.9473278480883733e-05,
      "loss": 0.8707,
      "num_input_tokens_seen": 47817848,
      "step": 82430
    },
    {
      "epoch": 12.278075662794162,
      "grad_norm": 4.34375,
      "learning_rate": 1.947010950265387e-05,
      "loss": 1.1458,
      "num_input_tokens_seen": 47820504,
      "step": 82435
    },
    {
      "epoch": 12.27882037533512,
      "grad_norm": 15.6875,
      "learning_rate": 1.9466940617845897e-05,
      "loss": 1.0334,
      "num_input_tokens_seen": 47823096,
      "step": 82440
    },
    {
      "epoch": 12.27956508787608,
      "grad_norm": 8.125,
      "learning_rate": 1.9463771826513326e-05,
      "loss": 1.4932,
      "num_input_tokens_seen": 47826072,
      "step": 82445
    },
    {
      "epoch": 12.280309800417038,
      "grad_norm": 15.25,
      "learning_rate": 1.9460603128709715e-05,
      "loss": 1.0599,
      "num_input_tokens_seen": 47828824,
      "step": 82450
    },
    {
      "epoch": 12.281054512957999,
      "grad_norm": 5.28125,
      "learning_rate": 1.9457434524488582e-05,
      "loss": 0.7925,
      "num_input_tokens_seen": 47831576,
      "step": 82455
    },
    {
      "epoch": 12.281799225498958,
      "grad_norm": 16.875,
      "learning_rate": 1.9454266013903458e-05,
      "loss": 1.0932,
      "num_input_tokens_seen": 47834328,
      "step": 82460
    },
    {
      "epoch": 12.282543938039916,
      "grad_norm": 10.375,
      "learning_rate": 1.945109759700788e-05,
      "loss": 0.8759,
      "num_input_tokens_seen": 47837304,
      "step": 82465
    },
    {
      "epoch": 12.283288650580875,
      "grad_norm": 17.75,
      "learning_rate": 1.9447929273855354e-05,
      "loss": 1.1251,
      "num_input_tokens_seen": 47840088,
      "step": 82470
    },
    {
      "epoch": 12.284033363121836,
      "grad_norm": 8.1875,
      "learning_rate": 1.944476104449943e-05,
      "loss": 1.1496,
      "num_input_tokens_seen": 47843000,
      "step": 82475
    },
    {
      "epoch": 12.284778075662794,
      "grad_norm": 11.1875,
      "learning_rate": 1.9441592908993616e-05,
      "loss": 1.1883,
      "num_input_tokens_seen": 47845688,
      "step": 82480
    },
    {
      "epoch": 12.285522788203753,
      "grad_norm": 7.3125,
      "learning_rate": 1.9438424867391444e-05,
      "loss": 1.2099,
      "num_input_tokens_seen": 47848568,
      "step": 82485
    },
    {
      "epoch": 12.286267500744712,
      "grad_norm": 7.40625,
      "learning_rate": 1.9435256919746436e-05,
      "loss": 0.9573,
      "num_input_tokens_seen": 47851224,
      "step": 82490
    },
    {
      "epoch": 12.287012213285673,
      "grad_norm": 15.0,
      "learning_rate": 1.94320890661121e-05,
      "loss": 1.1327,
      "num_input_tokens_seen": 47853944,
      "step": 82495
    },
    {
      "epoch": 12.287756925826631,
      "grad_norm": 7.46875,
      "learning_rate": 1.9428921306541963e-05,
      "loss": 0.8944,
      "num_input_tokens_seen": 47856760,
      "step": 82500
    },
    {
      "epoch": 12.28850163836759,
      "grad_norm": 16.75,
      "learning_rate": 1.9425753641089535e-05,
      "loss": 1.3323,
      "num_input_tokens_seen": 47859928,
      "step": 82505
    },
    {
      "epoch": 12.289246350908549,
      "grad_norm": 5.6875,
      "learning_rate": 1.9422586069808337e-05,
      "loss": 1.1779,
      "num_input_tokens_seen": 47863128,
      "step": 82510
    },
    {
      "epoch": 12.28999106344951,
      "grad_norm": 10.4375,
      "learning_rate": 1.941941859275188e-05,
      "loss": 0.9513,
      "num_input_tokens_seen": 47866136,
      "step": 82515
    },
    {
      "epoch": 12.290735775990468,
      "grad_norm": 15.625,
      "learning_rate": 1.9416251209973672e-05,
      "loss": 1.2768,
      "num_input_tokens_seen": 47869112,
      "step": 82520
    },
    {
      "epoch": 12.291480488531427,
      "grad_norm": 7.125,
      "learning_rate": 1.941308392152722e-05,
      "loss": 1.1993,
      "num_input_tokens_seen": 47871960,
      "step": 82525
    },
    {
      "epoch": 12.292225201072386,
      "grad_norm": 7.75,
      "learning_rate": 1.9409916727466047e-05,
      "loss": 0.9272,
      "num_input_tokens_seen": 47874808,
      "step": 82530
    },
    {
      "epoch": 12.292969913613344,
      "grad_norm": 10.6875,
      "learning_rate": 1.9406749627843645e-05,
      "loss": 1.1057,
      "num_input_tokens_seen": 47877496,
      "step": 82535
    },
    {
      "epoch": 12.293714626154305,
      "grad_norm": 11.375,
      "learning_rate": 1.940358262271352e-05,
      "loss": 1.3054,
      "num_input_tokens_seen": 47880248,
      "step": 82540
    },
    {
      "epoch": 12.294459338695264,
      "grad_norm": 7.5625,
      "learning_rate": 1.9400415712129184e-05,
      "loss": 0.9984,
      "num_input_tokens_seen": 47882808,
      "step": 82545
    },
    {
      "epoch": 12.295204051236222,
      "grad_norm": 13.3125,
      "learning_rate": 1.9397248896144127e-05,
      "loss": 1.1474,
      "num_input_tokens_seen": 47885528,
      "step": 82550
    },
    {
      "epoch": 12.295948763777181,
      "grad_norm": 7.25,
      "learning_rate": 1.939408217481186e-05,
      "loss": 1.0833,
      "num_input_tokens_seen": 47888504,
      "step": 82555
    },
    {
      "epoch": 12.296693476318142,
      "grad_norm": 9.25,
      "learning_rate": 1.939091554818587e-05,
      "loss": 1.0056,
      "num_input_tokens_seen": 47891288,
      "step": 82560
    },
    {
      "epoch": 12.2974381888591,
      "grad_norm": 18.25,
      "learning_rate": 1.9387749016319673e-05,
      "loss": 1.445,
      "num_input_tokens_seen": 47894232,
      "step": 82565
    },
    {
      "epoch": 12.29818290140006,
      "grad_norm": 10.625,
      "learning_rate": 1.938458257926675e-05,
      "loss": 1.3498,
      "num_input_tokens_seen": 47897112,
      "step": 82570
    },
    {
      "epoch": 12.298927613941018,
      "grad_norm": 5.78125,
      "learning_rate": 1.9381416237080586e-05,
      "loss": 0.8398,
      "num_input_tokens_seen": 47900152,
      "step": 82575
    },
    {
      "epoch": 12.299672326481979,
      "grad_norm": 8.3125,
      "learning_rate": 1.9378249989814697e-05,
      "loss": 0.7871,
      "num_input_tokens_seen": 47903256,
      "step": 82580
    },
    {
      "epoch": 12.300417039022937,
      "grad_norm": 7.90625,
      "learning_rate": 1.937508383752255e-05,
      "loss": 0.9783,
      "num_input_tokens_seen": 47906584,
      "step": 82585
    },
    {
      "epoch": 12.301161751563896,
      "grad_norm": 11.0,
      "learning_rate": 1.9371917780257657e-05,
      "loss": 1.2749,
      "num_input_tokens_seen": 47909592,
      "step": 82590
    },
    {
      "epoch": 12.301906464104855,
      "grad_norm": 7.1875,
      "learning_rate": 1.9368751818073476e-05,
      "loss": 1.0532,
      "num_input_tokens_seen": 47912600,
      "step": 82595
    },
    {
      "epoch": 12.302651176645815,
      "grad_norm": 9.25,
      "learning_rate": 1.9365585951023523e-05,
      "loss": 1.1883,
      "num_input_tokens_seen": 47915480,
      "step": 82600
    },
    {
      "epoch": 12.303395889186774,
      "grad_norm": 13.125,
      "learning_rate": 1.9362420179161262e-05,
      "loss": 1.0076,
      "num_input_tokens_seen": 47918168,
      "step": 82605
    },
    {
      "epoch": 12.304140601727733,
      "grad_norm": 5.40625,
      "learning_rate": 1.935925450254019e-05,
      "loss": 0.8525,
      "num_input_tokens_seen": 47921272,
      "step": 82610
    },
    {
      "epoch": 12.304885314268692,
      "grad_norm": 8.5,
      "learning_rate": 1.935608892121378e-05,
      "loss": 1.0519,
      "num_input_tokens_seen": 47924184,
      "step": 82615
    },
    {
      "epoch": 12.305630026809652,
      "grad_norm": 11.1875,
      "learning_rate": 1.9352923435235505e-05,
      "loss": 1.0509,
      "num_input_tokens_seen": 47927448,
      "step": 82620
    },
    {
      "epoch": 12.30637473935061,
      "grad_norm": 18.625,
      "learning_rate": 1.9349758044658854e-05,
      "loss": 1.363,
      "num_input_tokens_seen": 47930328,
      "step": 82625
    },
    {
      "epoch": 12.30711945189157,
      "grad_norm": 10.625,
      "learning_rate": 1.934659274953729e-05,
      "loss": 0.9545,
      "num_input_tokens_seen": 47932952,
      "step": 82630
    },
    {
      "epoch": 12.307864164432528,
      "grad_norm": 11.5625,
      "learning_rate": 1.9343427549924302e-05,
      "loss": 1.0623,
      "num_input_tokens_seen": 47935832,
      "step": 82635
    },
    {
      "epoch": 12.308608876973489,
      "grad_norm": 19.875,
      "learning_rate": 1.9340262445873354e-05,
      "loss": 0.9686,
      "num_input_tokens_seen": 47939000,
      "step": 82640
    },
    {
      "epoch": 12.309353589514448,
      "grad_norm": 12.1875,
      "learning_rate": 1.9337097437437924e-05,
      "loss": 1.048,
      "num_input_tokens_seen": 47941976,
      "step": 82645
    },
    {
      "epoch": 12.310098302055406,
      "grad_norm": 7.1875,
      "learning_rate": 1.9333932524671478e-05,
      "loss": 1.1634,
      "num_input_tokens_seen": 47944696,
      "step": 82650
    },
    {
      "epoch": 12.310843014596365,
      "grad_norm": 15.5,
      "learning_rate": 1.933076770762748e-05,
      "loss": 1.0239,
      "num_input_tokens_seen": 47947928,
      "step": 82655
    },
    {
      "epoch": 12.311587727137326,
      "grad_norm": 8.5625,
      "learning_rate": 1.93276029863594e-05,
      "loss": 1.1458,
      "num_input_tokens_seen": 47950840,
      "step": 82660
    },
    {
      "epoch": 12.312332439678285,
      "grad_norm": 8.5625,
      "learning_rate": 1.9324438360920696e-05,
      "loss": 1.2269,
      "num_input_tokens_seen": 47953688,
      "step": 82665
    },
    {
      "epoch": 12.313077152219243,
      "grad_norm": 6.3125,
      "learning_rate": 1.9321273831364847e-05,
      "loss": 0.8971,
      "num_input_tokens_seen": 47956856,
      "step": 82670
    },
    {
      "epoch": 12.313821864760202,
      "grad_norm": 8.125,
      "learning_rate": 1.9318109397745295e-05,
      "loss": 0.9682,
      "num_input_tokens_seen": 47959896,
      "step": 82675
    },
    {
      "epoch": 12.314566577301163,
      "grad_norm": 15.5,
      "learning_rate": 1.9314945060115517e-05,
      "loss": 0.9392,
      "num_input_tokens_seen": 47962584,
      "step": 82680
    },
    {
      "epoch": 12.315311289842121,
      "grad_norm": 14.8125,
      "learning_rate": 1.9311780818528966e-05,
      "loss": 0.9143,
      "num_input_tokens_seen": 47965432,
      "step": 82685
    },
    {
      "epoch": 12.31605600238308,
      "grad_norm": 26.125,
      "learning_rate": 1.93086166730391e-05,
      "loss": 1.0787,
      "num_input_tokens_seen": 47968088,
      "step": 82690
    },
    {
      "epoch": 12.316800714924039,
      "grad_norm": 11.8125,
      "learning_rate": 1.9305452623699367e-05,
      "loss": 1.0916,
      "num_input_tokens_seen": 47971096,
      "step": 82695
    },
    {
      "epoch": 12.317545427465,
      "grad_norm": 16.5,
      "learning_rate": 1.9302288670563215e-05,
      "loss": 1.3658,
      "num_input_tokens_seen": 47974072,
      "step": 82700
    },
    {
      "epoch": 12.318290140005958,
      "grad_norm": 7.75,
      "learning_rate": 1.9299124813684117e-05,
      "loss": 0.8934,
      "num_input_tokens_seen": 47976664,
      "step": 82705
    },
    {
      "epoch": 12.319034852546917,
      "grad_norm": 10.0625,
      "learning_rate": 1.92959610531155e-05,
      "loss": 1.2479,
      "num_input_tokens_seen": 47979736,
      "step": 82710
    },
    {
      "epoch": 12.319779565087876,
      "grad_norm": 11.4375,
      "learning_rate": 1.9292797388910832e-05,
      "loss": 1.3208,
      "num_input_tokens_seen": 47982648,
      "step": 82715
    },
    {
      "epoch": 12.320524277628834,
      "grad_norm": 11.0,
      "learning_rate": 1.928963382112355e-05,
      "loss": 1.3493,
      "num_input_tokens_seen": 47985624,
      "step": 82720
    },
    {
      "epoch": 12.321268990169795,
      "grad_norm": 9.125,
      "learning_rate": 1.9286470349807108e-05,
      "loss": 1.2234,
      "num_input_tokens_seen": 47988376,
      "step": 82725
    },
    {
      "epoch": 12.322013702710754,
      "grad_norm": 6.875,
      "learning_rate": 1.9283306975014935e-05,
      "loss": 1.1147,
      "num_input_tokens_seen": 47991320,
      "step": 82730
    },
    {
      "epoch": 12.322758415251712,
      "grad_norm": 8.9375,
      "learning_rate": 1.9280143696800473e-05,
      "loss": 1.216,
      "num_input_tokens_seen": 47994168,
      "step": 82735
    },
    {
      "epoch": 12.323503127792671,
      "grad_norm": 7.84375,
      "learning_rate": 1.9276980515217183e-05,
      "loss": 0.7846,
      "num_input_tokens_seen": 47997400,
      "step": 82740
    },
    {
      "epoch": 12.324247840333632,
      "grad_norm": 9.0,
      "learning_rate": 1.927381743031848e-05,
      "loss": 0.8422,
      "num_input_tokens_seen": 48000248,
      "step": 82745
    },
    {
      "epoch": 12.32499255287459,
      "grad_norm": 11.25,
      "learning_rate": 1.927065444215782e-05,
      "loss": 1.2059,
      "num_input_tokens_seen": 48003576,
      "step": 82750
    },
    {
      "epoch": 12.32573726541555,
      "grad_norm": 12.25,
      "learning_rate": 1.9267491550788626e-05,
      "loss": 1.0472,
      "num_input_tokens_seen": 48006616,
      "step": 82755
    },
    {
      "epoch": 12.326481977956508,
      "grad_norm": 11.25,
      "learning_rate": 1.926432875626434e-05,
      "loss": 1.1046,
      "num_input_tokens_seen": 48009336,
      "step": 82760
    },
    {
      "epoch": 12.327226690497469,
      "grad_norm": 14.6875,
      "learning_rate": 1.926116605863838e-05,
      "loss": 1.3244,
      "num_input_tokens_seen": 48012024,
      "step": 82765
    },
    {
      "epoch": 12.327971403038427,
      "grad_norm": 14.9375,
      "learning_rate": 1.9258003457964198e-05,
      "loss": 1.2496,
      "num_input_tokens_seen": 48014840,
      "step": 82770
    },
    {
      "epoch": 12.328716115579386,
      "grad_norm": 12.875,
      "learning_rate": 1.925484095429521e-05,
      "loss": 1.3018,
      "num_input_tokens_seen": 48017400,
      "step": 82775
    },
    {
      "epoch": 12.329460828120345,
      "grad_norm": 8.4375,
      "learning_rate": 1.9251678547684836e-05,
      "loss": 1.0514,
      "num_input_tokens_seen": 48020440,
      "step": 82780
    },
    {
      "epoch": 12.330205540661305,
      "grad_norm": 17.5,
      "learning_rate": 1.924851623818652e-05,
      "loss": 1.3174,
      "num_input_tokens_seen": 48023320,
      "step": 82785
    },
    {
      "epoch": 12.330950253202264,
      "grad_norm": 9.625,
      "learning_rate": 1.9245354025853673e-05,
      "loss": 1.0137,
      "num_input_tokens_seen": 48026168,
      "step": 82790
    },
    {
      "epoch": 12.331694965743223,
      "grad_norm": 6.8125,
      "learning_rate": 1.9242191910739727e-05,
      "loss": 1.0919,
      "num_input_tokens_seen": 48029048,
      "step": 82795
    },
    {
      "epoch": 12.332439678284182,
      "grad_norm": 9.25,
      "learning_rate": 1.9239029892898083e-05,
      "loss": 1.0397,
      "num_input_tokens_seen": 48031704,
      "step": 82800
    },
    {
      "epoch": 12.333184390825142,
      "grad_norm": 11.0625,
      "learning_rate": 1.9235867972382188e-05,
      "loss": 0.9162,
      "num_input_tokens_seen": 48034552,
      "step": 82805
    },
    {
      "epoch": 12.333929103366101,
      "grad_norm": 8.125,
      "learning_rate": 1.9232706149245443e-05,
      "loss": 1.2483,
      "num_input_tokens_seen": 48037432,
      "step": 82810
    },
    {
      "epoch": 12.33467381590706,
      "grad_norm": 14.4375,
      "learning_rate": 1.9229544423541254e-05,
      "loss": 1.2493,
      "num_input_tokens_seen": 48040184,
      "step": 82815
    },
    {
      "epoch": 12.335418528448018,
      "grad_norm": 9.875,
      "learning_rate": 1.922638279532306e-05,
      "loss": 1.6523,
      "num_input_tokens_seen": 48042776,
      "step": 82820
    },
    {
      "epoch": 12.336163240988979,
      "grad_norm": 11.8125,
      "learning_rate": 1.9223221264644253e-05,
      "loss": 1.0918,
      "num_input_tokens_seen": 48045720,
      "step": 82825
    },
    {
      "epoch": 12.336907953529938,
      "grad_norm": 9.3125,
      "learning_rate": 1.922005983155826e-05,
      "loss": 0.8757,
      "num_input_tokens_seen": 48048952,
      "step": 82830
    },
    {
      "epoch": 12.337652666070897,
      "grad_norm": 13.375,
      "learning_rate": 1.921689849611847e-05,
      "loss": 1.2244,
      "num_input_tokens_seen": 48052056,
      "step": 82835
    },
    {
      "epoch": 12.338397378611855,
      "grad_norm": 14.5625,
      "learning_rate": 1.921373725837831e-05,
      "loss": 1.3078,
      "num_input_tokens_seen": 48055160,
      "step": 82840
    },
    {
      "epoch": 12.339142091152816,
      "grad_norm": 9.75,
      "learning_rate": 1.9210576118391177e-05,
      "loss": 1.0792,
      "num_input_tokens_seen": 48058136,
      "step": 82845
    },
    {
      "epoch": 12.339886803693775,
      "grad_norm": 12.4375,
      "learning_rate": 1.920741507621048e-05,
      "loss": 1.0027,
      "num_input_tokens_seen": 48060888,
      "step": 82850
    },
    {
      "epoch": 12.340631516234733,
      "grad_norm": 18.5,
      "learning_rate": 1.9204254131889612e-05,
      "loss": 1.4008,
      "num_input_tokens_seen": 48063768,
      "step": 82855
    },
    {
      "epoch": 12.341376228775692,
      "grad_norm": 9.25,
      "learning_rate": 1.920109328548198e-05,
      "loss": 1.3165,
      "num_input_tokens_seen": 48066808,
      "step": 82860
    },
    {
      "epoch": 12.342120941316653,
      "grad_norm": 8.125,
      "learning_rate": 1.919793253704099e-05,
      "loss": 1.0483,
      "num_input_tokens_seen": 48070072,
      "step": 82865
    },
    {
      "epoch": 12.342865653857611,
      "grad_norm": 19.5,
      "learning_rate": 1.9194771886620023e-05,
      "loss": 1.1756,
      "num_input_tokens_seen": 48072952,
      "step": 82870
    },
    {
      "epoch": 12.34361036639857,
      "grad_norm": 7.21875,
      "learning_rate": 1.919161133427249e-05,
      "loss": 0.8911,
      "num_input_tokens_seen": 48075672,
      "step": 82875
    },
    {
      "epoch": 12.344355078939529,
      "grad_norm": 8.25,
      "learning_rate": 1.918845088005178e-05,
      "loss": 1.2128,
      "num_input_tokens_seen": 48078648,
      "step": 82880
    },
    {
      "epoch": 12.34509979148049,
      "grad_norm": 7.875,
      "learning_rate": 1.918529052401129e-05,
      "loss": 1.1433,
      "num_input_tokens_seen": 48081912,
      "step": 82885
    },
    {
      "epoch": 12.345844504021448,
      "grad_norm": 7.21875,
      "learning_rate": 1.9182130266204396e-05,
      "loss": 1.1135,
      "num_input_tokens_seen": 48085176,
      "step": 82890
    },
    {
      "epoch": 12.346589216562407,
      "grad_norm": 21.875,
      "learning_rate": 1.9178970106684506e-05,
      "loss": 1.1746,
      "num_input_tokens_seen": 48088056,
      "step": 82895
    },
    {
      "epoch": 12.347333929103366,
      "grad_norm": 10.1875,
      "learning_rate": 1.9175810045505006e-05,
      "loss": 1.124,
      "num_input_tokens_seen": 48091224,
      "step": 82900
    },
    {
      "epoch": 12.348078641644324,
      "grad_norm": 9.125,
      "learning_rate": 1.917265008271926e-05,
      "loss": 1.1411,
      "num_input_tokens_seen": 48094168,
      "step": 82905
    },
    {
      "epoch": 12.348823354185285,
      "grad_norm": 7.90625,
      "learning_rate": 1.916949021838068e-05,
      "loss": 0.7667,
      "num_input_tokens_seen": 48097208,
      "step": 82910
    },
    {
      "epoch": 12.349568066726244,
      "grad_norm": 14.9375,
      "learning_rate": 1.916633045254263e-05,
      "loss": 0.9924,
      "num_input_tokens_seen": 48100440,
      "step": 82915
    },
    {
      "epoch": 12.350312779267203,
      "grad_norm": 13.3125,
      "learning_rate": 1.9163170785258507e-05,
      "loss": 0.9063,
      "num_input_tokens_seen": 48102968,
      "step": 82920
    },
    {
      "epoch": 12.351057491808161,
      "grad_norm": 13.75,
      "learning_rate": 1.916001121658167e-05,
      "loss": 1.2086,
      "num_input_tokens_seen": 48105656,
      "step": 82925
    },
    {
      "epoch": 12.351802204349122,
      "grad_norm": 10.1875,
      "learning_rate": 1.9156851746565514e-05,
      "loss": 1.0704,
      "num_input_tokens_seen": 48108440,
      "step": 82930
    },
    {
      "epoch": 12.35254691689008,
      "grad_norm": 9.625,
      "learning_rate": 1.9153692375263413e-05,
      "loss": 0.8827,
      "num_input_tokens_seen": 48111544,
      "step": 82935
    },
    {
      "epoch": 12.35329162943104,
      "grad_norm": 8.625,
      "learning_rate": 1.9150533102728728e-05,
      "loss": 0.9603,
      "num_input_tokens_seen": 48114488,
      "step": 82940
    },
    {
      "epoch": 12.354036341971998,
      "grad_norm": 6.4375,
      "learning_rate": 1.914737392901485e-05,
      "loss": 0.9966,
      "num_input_tokens_seen": 48117240,
      "step": 82945
    },
    {
      "epoch": 12.354781054512959,
      "grad_norm": 9.8125,
      "learning_rate": 1.9144214854175136e-05,
      "loss": 1.0865,
      "num_input_tokens_seen": 48119832,
      "step": 82950
    },
    {
      "epoch": 12.355525767053917,
      "grad_norm": 8.8125,
      "learning_rate": 1.9141055878262963e-05,
      "loss": 0.8996,
      "num_input_tokens_seen": 48122840,
      "step": 82955
    },
    {
      "epoch": 12.356270479594876,
      "grad_norm": 11.8125,
      "learning_rate": 1.913789700133169e-05,
      "loss": 1.2613,
      "num_input_tokens_seen": 48125976,
      "step": 82960
    },
    {
      "epoch": 12.357015192135835,
      "grad_norm": 11.25,
      "learning_rate": 1.9134738223434697e-05,
      "loss": 1.3478,
      "num_input_tokens_seen": 48128920,
      "step": 82965
    },
    {
      "epoch": 12.357759904676795,
      "grad_norm": 8.8125,
      "learning_rate": 1.913157954462533e-05,
      "loss": 1.218,
      "num_input_tokens_seen": 48131640,
      "step": 82970
    },
    {
      "epoch": 12.358504617217754,
      "grad_norm": 19.375,
      "learning_rate": 1.9128420964956972e-05,
      "loss": 0.9843,
      "num_input_tokens_seen": 48134584,
      "step": 82975
    },
    {
      "epoch": 12.359249329758713,
      "grad_norm": 16.25,
      "learning_rate": 1.912526248448298e-05,
      "loss": 1.2136,
      "num_input_tokens_seen": 48137432,
      "step": 82980
    },
    {
      "epoch": 12.359994042299672,
      "grad_norm": 11.0625,
      "learning_rate": 1.9122104103256693e-05,
      "loss": 0.8601,
      "num_input_tokens_seen": 48140152,
      "step": 82985
    },
    {
      "epoch": 12.360738754840632,
      "grad_norm": 6.71875,
      "learning_rate": 1.9118945821331495e-05,
      "loss": 0.8894,
      "num_input_tokens_seen": 48142712,
      "step": 82990
    },
    {
      "epoch": 12.361483467381591,
      "grad_norm": 10.4375,
      "learning_rate": 1.9115787638760717e-05,
      "loss": 0.9598,
      "num_input_tokens_seen": 48145944,
      "step": 82995
    },
    {
      "epoch": 12.36222817992255,
      "grad_norm": 10.6875,
      "learning_rate": 1.911262955559774e-05,
      "loss": 1.3055,
      "num_input_tokens_seen": 48148408,
      "step": 83000
    },
    {
      "epoch": 12.362972892463509,
      "grad_norm": 7.75,
      "learning_rate": 1.910947157189589e-05,
      "loss": 1.2335,
      "num_input_tokens_seen": 48151320,
      "step": 83005
    },
    {
      "epoch": 12.363717605004469,
      "grad_norm": 5.125,
      "learning_rate": 1.9106313687708543e-05,
      "loss": 0.8705,
      "num_input_tokens_seen": 48154104,
      "step": 83010
    },
    {
      "epoch": 12.364462317545428,
      "grad_norm": 17.375,
      "learning_rate": 1.9103155903089036e-05,
      "loss": 1.1968,
      "num_input_tokens_seen": 48156888,
      "step": 83015
    },
    {
      "epoch": 12.365207030086387,
      "grad_norm": 11.6875,
      "learning_rate": 1.9099998218090707e-05,
      "loss": 1.2556,
      "num_input_tokens_seen": 48159960,
      "step": 83020
    },
    {
      "epoch": 12.365951742627345,
      "grad_norm": 7.46875,
      "learning_rate": 1.9096840632766923e-05,
      "loss": 0.9732,
      "num_input_tokens_seen": 48162712,
      "step": 83025
    },
    {
      "epoch": 12.366696455168306,
      "grad_norm": 9.125,
      "learning_rate": 1.9093683147171002e-05,
      "loss": 1.3448,
      "num_input_tokens_seen": 48165752,
      "step": 83030
    },
    {
      "epoch": 12.367441167709265,
      "grad_norm": 7.15625,
      "learning_rate": 1.9090525761356315e-05,
      "loss": 1.1676,
      "num_input_tokens_seen": 48168632,
      "step": 83035
    },
    {
      "epoch": 12.368185880250223,
      "grad_norm": 7.9375,
      "learning_rate": 1.9087368475376176e-05,
      "loss": 1.5224,
      "num_input_tokens_seen": 48171768,
      "step": 83040
    },
    {
      "epoch": 12.368930592791182,
      "grad_norm": 9.3125,
      "learning_rate": 1.908421128928395e-05,
      "loss": 1.1561,
      "num_input_tokens_seen": 48174552,
      "step": 83045
    },
    {
      "epoch": 12.36967530533214,
      "grad_norm": 11.125,
      "learning_rate": 1.9081054203132955e-05,
      "loss": 0.9702,
      "num_input_tokens_seen": 48177656,
      "step": 83050
    },
    {
      "epoch": 12.370420017873101,
      "grad_norm": 13.375,
      "learning_rate": 1.9077897216976537e-05,
      "loss": 0.8379,
      "num_input_tokens_seen": 48180472,
      "step": 83055
    },
    {
      "epoch": 12.37116473041406,
      "grad_norm": 12.4375,
      "learning_rate": 1.907474033086803e-05,
      "loss": 1.3379,
      "num_input_tokens_seen": 48183256,
      "step": 83060
    },
    {
      "epoch": 12.371909442955019,
      "grad_norm": 11.4375,
      "learning_rate": 1.907158354486075e-05,
      "loss": 1.1909,
      "num_input_tokens_seen": 48186168,
      "step": 83065
    },
    {
      "epoch": 12.37265415549598,
      "grad_norm": 8.5625,
      "learning_rate": 1.9068426859008055e-05,
      "loss": 1.0604,
      "num_input_tokens_seen": 48188984,
      "step": 83070
    },
    {
      "epoch": 12.373398868036938,
      "grad_norm": 6.875,
      "learning_rate": 1.9065270273363244e-05,
      "loss": 1.0547,
      "num_input_tokens_seen": 48191736,
      "step": 83075
    },
    {
      "epoch": 12.374143580577897,
      "grad_norm": 8.5625,
      "learning_rate": 1.9062113787979674e-05,
      "loss": 0.8389,
      "num_input_tokens_seen": 48194520,
      "step": 83080
    },
    {
      "epoch": 12.374888293118856,
      "grad_norm": 7.09375,
      "learning_rate": 1.905895740291065e-05,
      "loss": 0.8648,
      "num_input_tokens_seen": 48197304,
      "step": 83085
    },
    {
      "epoch": 12.375633005659815,
      "grad_norm": 8.75,
      "learning_rate": 1.9055801118209507e-05,
      "loss": 1.2222,
      "num_input_tokens_seen": 48200024,
      "step": 83090
    },
    {
      "epoch": 12.376377718200775,
      "grad_norm": 7.0625,
      "learning_rate": 1.9052644933929564e-05,
      "loss": 1.2672,
      "num_input_tokens_seen": 48202808,
      "step": 83095
    },
    {
      "epoch": 12.377122430741734,
      "grad_norm": 7.53125,
      "learning_rate": 1.9049488850124128e-05,
      "loss": 1.1677,
      "num_input_tokens_seen": 48205432,
      "step": 83100
    },
    {
      "epoch": 12.377867143282693,
      "grad_norm": 13.4375,
      "learning_rate": 1.9046332866846544e-05,
      "loss": 0.8039,
      "num_input_tokens_seen": 48208600,
      "step": 83105
    },
    {
      "epoch": 12.378611855823651,
      "grad_norm": 12.0625,
      "learning_rate": 1.90431769841501e-05,
      "loss": 1.1874,
      "num_input_tokens_seen": 48211544,
      "step": 83110
    },
    {
      "epoch": 12.379356568364612,
      "grad_norm": 7.625,
      "learning_rate": 1.904002120208814e-05,
      "loss": 1.2496,
      "num_input_tokens_seen": 48214520,
      "step": 83115
    },
    {
      "epoch": 12.38010128090557,
      "grad_norm": 12.75,
      "learning_rate": 1.903686552071396e-05,
      "loss": 1.256,
      "num_input_tokens_seen": 48217368,
      "step": 83120
    },
    {
      "epoch": 12.38084599344653,
      "grad_norm": 9.5625,
      "learning_rate": 1.903370994008088e-05,
      "loss": 0.9348,
      "num_input_tokens_seen": 48220504,
      "step": 83125
    },
    {
      "epoch": 12.381590705987488,
      "grad_norm": 3.8125,
      "learning_rate": 1.9030554460242194e-05,
      "loss": 0.7365,
      "num_input_tokens_seen": 48223288,
      "step": 83130
    },
    {
      "epoch": 12.382335418528449,
      "grad_norm": 5.53125,
      "learning_rate": 1.902739908125124e-05,
      "loss": 0.8987,
      "num_input_tokens_seen": 48225912,
      "step": 83135
    },
    {
      "epoch": 12.383080131069407,
      "grad_norm": 4.84375,
      "learning_rate": 1.9024243803161304e-05,
      "loss": 0.8944,
      "num_input_tokens_seen": 48228792,
      "step": 83140
    },
    {
      "epoch": 12.383824843610366,
      "grad_norm": 5.25,
      "learning_rate": 1.9021088626025684e-05,
      "loss": 1.0485,
      "num_input_tokens_seen": 48231416,
      "step": 83145
    },
    {
      "epoch": 12.384569556151325,
      "grad_norm": 6.21875,
      "learning_rate": 1.9017933549897706e-05,
      "loss": 0.8569,
      "num_input_tokens_seen": 48234296,
      "step": 83150
    },
    {
      "epoch": 12.385314268692285,
      "grad_norm": 9.0625,
      "learning_rate": 1.901477857483066e-05,
      "loss": 0.985,
      "num_input_tokens_seen": 48237048,
      "step": 83155
    },
    {
      "epoch": 12.386058981233244,
      "grad_norm": 8.9375,
      "learning_rate": 1.9011623700877845e-05,
      "loss": 1.2634,
      "num_input_tokens_seen": 48239896,
      "step": 83160
    },
    {
      "epoch": 12.386803693774203,
      "grad_norm": 9.9375,
      "learning_rate": 1.9008468928092555e-05,
      "loss": 0.9527,
      "num_input_tokens_seen": 48243032,
      "step": 83165
    },
    {
      "epoch": 12.387548406315162,
      "grad_norm": 14.75,
      "learning_rate": 1.9005314256528103e-05,
      "loss": 1.0098,
      "num_input_tokens_seen": 48245880,
      "step": 83170
    },
    {
      "epoch": 12.388293118856122,
      "grad_norm": 9.125,
      "learning_rate": 1.9002159686237776e-05,
      "loss": 0.9565,
      "num_input_tokens_seen": 48248760,
      "step": 83175
    },
    {
      "epoch": 12.389037831397081,
      "grad_norm": 10.125,
      "learning_rate": 1.8999005217274857e-05,
      "loss": 0.864,
      "num_input_tokens_seen": 48251672,
      "step": 83180
    },
    {
      "epoch": 12.38978254393804,
      "grad_norm": 10.375,
      "learning_rate": 1.8995850849692646e-05,
      "loss": 0.8552,
      "num_input_tokens_seen": 48254552,
      "step": 83185
    },
    {
      "epoch": 12.390527256478999,
      "grad_norm": 10.25,
      "learning_rate": 1.8992696583544434e-05,
      "loss": 1.1431,
      "num_input_tokens_seen": 48257432,
      "step": 83190
    },
    {
      "epoch": 12.391271969019959,
      "grad_norm": 10.3125,
      "learning_rate": 1.898954241888351e-05,
      "loss": 1.1408,
      "num_input_tokens_seen": 48260120,
      "step": 83195
    },
    {
      "epoch": 12.392016681560918,
      "grad_norm": 21.625,
      "learning_rate": 1.8986388355763147e-05,
      "loss": 1.3058,
      "num_input_tokens_seen": 48263000,
      "step": 83200
    },
    {
      "epoch": 12.392761394101877,
      "grad_norm": 7.0,
      "learning_rate": 1.8983234394236657e-05,
      "loss": 1.0935,
      "num_input_tokens_seen": 48265720,
      "step": 83205
    },
    {
      "epoch": 12.393506106642835,
      "grad_norm": 7.90625,
      "learning_rate": 1.8980080534357298e-05,
      "loss": 0.9029,
      "num_input_tokens_seen": 48268664,
      "step": 83210
    },
    {
      "epoch": 12.394250819183796,
      "grad_norm": 13.375,
      "learning_rate": 1.8976926776178366e-05,
      "loss": 0.8508,
      "num_input_tokens_seen": 48271320,
      "step": 83215
    },
    {
      "epoch": 12.394995531724755,
      "grad_norm": 9.1875,
      "learning_rate": 1.8973773119753132e-05,
      "loss": 1.0675,
      "num_input_tokens_seen": 48274296,
      "step": 83220
    },
    {
      "epoch": 12.395740244265713,
      "grad_norm": 11.1875,
      "learning_rate": 1.8970619565134866e-05,
      "loss": 1.1907,
      "num_input_tokens_seen": 48277240,
      "step": 83225
    },
    {
      "epoch": 12.396484956806672,
      "grad_norm": 6.125,
      "learning_rate": 1.896746611237687e-05,
      "loss": 0.8763,
      "num_input_tokens_seen": 48279928,
      "step": 83230
    },
    {
      "epoch": 12.397229669347631,
      "grad_norm": 5.84375,
      "learning_rate": 1.8964312761532388e-05,
      "loss": 0.8778,
      "num_input_tokens_seen": 48282680,
      "step": 83235
    },
    {
      "epoch": 12.397974381888591,
      "grad_norm": 7.71875,
      "learning_rate": 1.896115951265472e-05,
      "loss": 0.7699,
      "num_input_tokens_seen": 48285784,
      "step": 83240
    },
    {
      "epoch": 12.39871909442955,
      "grad_norm": 19.25,
      "learning_rate": 1.8958006365797118e-05,
      "loss": 1.1028,
      "num_input_tokens_seen": 48288952,
      "step": 83245
    },
    {
      "epoch": 12.399463806970509,
      "grad_norm": 12.0,
      "learning_rate": 1.8954853321012865e-05,
      "loss": 1.0402,
      "num_input_tokens_seen": 48292120,
      "step": 83250
    },
    {
      "epoch": 12.400208519511468,
      "grad_norm": 15.0,
      "learning_rate": 1.8951700378355218e-05,
      "loss": 1.2677,
      "num_input_tokens_seen": 48294872,
      "step": 83255
    },
    {
      "epoch": 12.400953232052428,
      "grad_norm": 15.875,
      "learning_rate": 1.8948547537877436e-05,
      "loss": 1.2998,
      "num_input_tokens_seen": 48297592,
      "step": 83260
    },
    {
      "epoch": 12.401697944593387,
      "grad_norm": 8.6875,
      "learning_rate": 1.8945394799632804e-05,
      "loss": 1.0206,
      "num_input_tokens_seen": 48300472,
      "step": 83265
    },
    {
      "epoch": 12.402442657134346,
      "grad_norm": 13.25,
      "learning_rate": 1.8942242163674563e-05,
      "loss": 1.0654,
      "num_input_tokens_seen": 48303448,
      "step": 83270
    },
    {
      "epoch": 12.403187369675305,
      "grad_norm": 9.9375,
      "learning_rate": 1.8939089630055994e-05,
      "loss": 0.7212,
      "num_input_tokens_seen": 48306424,
      "step": 83275
    },
    {
      "epoch": 12.403932082216265,
      "grad_norm": 6.4375,
      "learning_rate": 1.8935937198830343e-05,
      "loss": 0.8284,
      "num_input_tokens_seen": 48309144,
      "step": 83280
    },
    {
      "epoch": 12.404676794757224,
      "grad_norm": 11.8125,
      "learning_rate": 1.893278487005087e-05,
      "loss": 0.9384,
      "num_input_tokens_seen": 48311864,
      "step": 83285
    },
    {
      "epoch": 12.405421507298183,
      "grad_norm": 15.5625,
      "learning_rate": 1.8929632643770824e-05,
      "loss": 1.2179,
      "num_input_tokens_seen": 48314680,
      "step": 83290
    },
    {
      "epoch": 12.406166219839141,
      "grad_norm": 11.625,
      "learning_rate": 1.8926480520043472e-05,
      "loss": 1.0122,
      "num_input_tokens_seen": 48317816,
      "step": 83295
    },
    {
      "epoch": 12.406910932380102,
      "grad_norm": 10.6875,
      "learning_rate": 1.892332849892206e-05,
      "loss": 1.043,
      "num_input_tokens_seen": 48320600,
      "step": 83300
    },
    {
      "epoch": 12.40765564492106,
      "grad_norm": 15.625,
      "learning_rate": 1.8920176580459827e-05,
      "loss": 1.1688,
      "num_input_tokens_seen": 48323768,
      "step": 83305
    },
    {
      "epoch": 12.40840035746202,
      "grad_norm": 12.875,
      "learning_rate": 1.8917024764710043e-05,
      "loss": 1.0925,
      "num_input_tokens_seen": 48326968,
      "step": 83310
    },
    {
      "epoch": 12.409145070002978,
      "grad_norm": 7.8125,
      "learning_rate": 1.8913873051725935e-05,
      "loss": 1.0836,
      "num_input_tokens_seen": 48329784,
      "step": 83315
    },
    {
      "epoch": 12.409889782543939,
      "grad_norm": 14.375,
      "learning_rate": 1.8910721441560765e-05,
      "loss": 0.9464,
      "num_input_tokens_seen": 48332760,
      "step": 83320
    },
    {
      "epoch": 12.410634495084897,
      "grad_norm": 15.0625,
      "learning_rate": 1.8907569934267756e-05,
      "loss": 1.1283,
      "num_input_tokens_seen": 48335512,
      "step": 83325
    },
    {
      "epoch": 12.411379207625856,
      "grad_norm": 5.65625,
      "learning_rate": 1.890441852990017e-05,
      "loss": 1.0009,
      "num_input_tokens_seen": 48338648,
      "step": 83330
    },
    {
      "epoch": 12.412123920166815,
      "grad_norm": 11.9375,
      "learning_rate": 1.890126722851124e-05,
      "loss": 1.1919,
      "num_input_tokens_seen": 48341624,
      "step": 83335
    },
    {
      "epoch": 12.412868632707776,
      "grad_norm": 22.125,
      "learning_rate": 1.8898116030154185e-05,
      "loss": 1.2069,
      "num_input_tokens_seen": 48344472,
      "step": 83340
    },
    {
      "epoch": 12.413613345248734,
      "grad_norm": 7.8125,
      "learning_rate": 1.8894964934882274e-05,
      "loss": 1.0036,
      "num_input_tokens_seen": 48347384,
      "step": 83345
    },
    {
      "epoch": 12.414358057789693,
      "grad_norm": 16.875,
      "learning_rate": 1.8891813942748717e-05,
      "loss": 1.4536,
      "num_input_tokens_seen": 48350040,
      "step": 83350
    },
    {
      "epoch": 12.415102770330652,
      "grad_norm": 14.125,
      "learning_rate": 1.8888663053806765e-05,
      "loss": 0.8301,
      "num_input_tokens_seen": 48352888,
      "step": 83355
    },
    {
      "epoch": 12.415847482871612,
      "grad_norm": 14.3125,
      "learning_rate": 1.8885512268109625e-05,
      "loss": 1.1107,
      "num_input_tokens_seen": 48355768,
      "step": 83360
    },
    {
      "epoch": 12.416592195412571,
      "grad_norm": 11.0,
      "learning_rate": 1.8882361585710554e-05,
      "loss": 1.0112,
      "num_input_tokens_seen": 48358840,
      "step": 83365
    },
    {
      "epoch": 12.41733690795353,
      "grad_norm": 10.875,
      "learning_rate": 1.887921100666275e-05,
      "loss": 1.0495,
      "num_input_tokens_seen": 48361816,
      "step": 83370
    },
    {
      "epoch": 12.418081620494489,
      "grad_norm": 12.125,
      "learning_rate": 1.8876060531019474e-05,
      "loss": 0.9432,
      "num_input_tokens_seen": 48364600,
      "step": 83375
    },
    {
      "epoch": 12.41882633303545,
      "grad_norm": 4.9375,
      "learning_rate": 1.887291015883393e-05,
      "loss": 1.2236,
      "num_input_tokens_seen": 48367384,
      "step": 83380
    },
    {
      "epoch": 12.419571045576408,
      "grad_norm": 21.25,
      "learning_rate": 1.8869759890159333e-05,
      "loss": 1.43,
      "num_input_tokens_seen": 48370392,
      "step": 83385
    },
    {
      "epoch": 12.420315758117367,
      "grad_norm": 14.0,
      "learning_rate": 1.8866609725048918e-05,
      "loss": 1.0259,
      "num_input_tokens_seen": 48373240,
      "step": 83390
    },
    {
      "epoch": 12.421060470658325,
      "grad_norm": 9.9375,
      "learning_rate": 1.8863459663555885e-05,
      "loss": 0.8659,
      "num_input_tokens_seen": 48376408,
      "step": 83395
    },
    {
      "epoch": 12.421805183199286,
      "grad_norm": 7.625,
      "learning_rate": 1.8860309705733477e-05,
      "loss": 1.0954,
      "num_input_tokens_seen": 48379256,
      "step": 83400
    },
    {
      "epoch": 12.422549895740245,
      "grad_norm": 11.0625,
      "learning_rate": 1.8857159851634888e-05,
      "loss": 0.9585,
      "num_input_tokens_seen": 48382072,
      "step": 83405
    },
    {
      "epoch": 12.423294608281203,
      "grad_norm": 7.34375,
      "learning_rate": 1.885401010131335e-05,
      "loss": 1.0884,
      "num_input_tokens_seen": 48385528,
      "step": 83410
    },
    {
      "epoch": 12.424039320822162,
      "grad_norm": 8.5,
      "learning_rate": 1.8850860454822056e-05,
      "loss": 0.7677,
      "num_input_tokens_seen": 48388440,
      "step": 83415
    },
    {
      "epoch": 12.424784033363121,
      "grad_norm": 14.25,
      "learning_rate": 1.8847710912214233e-05,
      "loss": 1.2274,
      "num_input_tokens_seen": 48391320,
      "step": 83420
    },
    {
      "epoch": 12.425528745904082,
      "grad_norm": 25.0,
      "learning_rate": 1.8844561473543082e-05,
      "loss": 1.1703,
      "num_input_tokens_seen": 48394008,
      "step": 83425
    },
    {
      "epoch": 12.42627345844504,
      "grad_norm": 8.8125,
      "learning_rate": 1.8841412138861797e-05,
      "loss": 1.1412,
      "num_input_tokens_seen": 48397080,
      "step": 83430
    },
    {
      "epoch": 12.427018170985999,
      "grad_norm": 8.75,
      "learning_rate": 1.8838262908223602e-05,
      "loss": 0.957,
      "num_input_tokens_seen": 48400056,
      "step": 83435
    },
    {
      "epoch": 12.427762883526958,
      "grad_norm": 9.625,
      "learning_rate": 1.8835113781681686e-05,
      "loss": 1.1918,
      "num_input_tokens_seen": 48402808,
      "step": 83440
    },
    {
      "epoch": 12.428507596067918,
      "grad_norm": 7.28125,
      "learning_rate": 1.8831964759289265e-05,
      "loss": 1.1727,
      "num_input_tokens_seen": 48405560,
      "step": 83445
    },
    {
      "epoch": 12.429252308608877,
      "grad_norm": 9.4375,
      "learning_rate": 1.8828815841099528e-05,
      "loss": 0.8986,
      "num_input_tokens_seen": 48408568,
      "step": 83450
    },
    {
      "epoch": 12.429997021149836,
      "grad_norm": 10.4375,
      "learning_rate": 1.882566702716568e-05,
      "loss": 0.8014,
      "num_input_tokens_seen": 48411544,
      "step": 83455
    },
    {
      "epoch": 12.430741733690795,
      "grad_norm": 9.25,
      "learning_rate": 1.8822518317540913e-05,
      "loss": 0.9328,
      "num_input_tokens_seen": 48414808,
      "step": 83460
    },
    {
      "epoch": 12.431486446231755,
      "grad_norm": 5.34375,
      "learning_rate": 1.8819369712278408e-05,
      "loss": 0.7753,
      "num_input_tokens_seen": 48417656,
      "step": 83465
    },
    {
      "epoch": 12.432231158772714,
      "grad_norm": 6.0625,
      "learning_rate": 1.8816221211431382e-05,
      "loss": 1.0962,
      "num_input_tokens_seen": 48420536,
      "step": 83470
    },
    {
      "epoch": 12.432975871313673,
      "grad_norm": 7.3125,
      "learning_rate": 1.8813072815053003e-05,
      "loss": 1.1797,
      "num_input_tokens_seen": 48423288,
      "step": 83475
    },
    {
      "epoch": 12.433720583854631,
      "grad_norm": 12.8125,
      "learning_rate": 1.880992452319648e-05,
      "loss": 1.0274,
      "num_input_tokens_seen": 48426488,
      "step": 83480
    },
    {
      "epoch": 12.434465296395592,
      "grad_norm": 12.375,
      "learning_rate": 1.8806776335914986e-05,
      "loss": 1.3867,
      "num_input_tokens_seen": 48429400,
      "step": 83485
    },
    {
      "epoch": 12.43521000893655,
      "grad_norm": 11.8125,
      "learning_rate": 1.8803628253261717e-05,
      "loss": 0.9519,
      "num_input_tokens_seen": 48432408,
      "step": 83490
    },
    {
      "epoch": 12.43595472147751,
      "grad_norm": 9.6875,
      "learning_rate": 1.880048027528984e-05,
      "loss": 1.0238,
      "num_input_tokens_seen": 48435800,
      "step": 83495
    },
    {
      "epoch": 12.436699434018468,
      "grad_norm": 14.4375,
      "learning_rate": 1.879733240205256e-05,
      "loss": 0.9323,
      "num_input_tokens_seen": 48438648,
      "step": 83500
    },
    {
      "epoch": 12.437444146559429,
      "grad_norm": 6.65625,
      "learning_rate": 1.879418463360304e-05,
      "loss": 1.2224,
      "num_input_tokens_seen": 48441368,
      "step": 83505
    },
    {
      "epoch": 12.438188859100388,
      "grad_norm": 9.3125,
      "learning_rate": 1.8791036969994462e-05,
      "loss": 0.7711,
      "num_input_tokens_seen": 48444312,
      "step": 83510
    },
    {
      "epoch": 12.438933571641346,
      "grad_norm": 4.6875,
      "learning_rate": 1.8787889411280005e-05,
      "loss": 0.971,
      "num_input_tokens_seen": 48447288,
      "step": 83515
    },
    {
      "epoch": 12.439678284182305,
      "grad_norm": 18.5,
      "learning_rate": 1.8784741957512842e-05,
      "loss": 1.4155,
      "num_input_tokens_seen": 48450200,
      "step": 83520
    },
    {
      "epoch": 12.440422996723266,
      "grad_norm": 9.375,
      "learning_rate": 1.878159460874615e-05,
      "loss": 1.3023,
      "num_input_tokens_seen": 48453048,
      "step": 83525
    },
    {
      "epoch": 12.441167709264224,
      "grad_norm": 8.3125,
      "learning_rate": 1.8778447365033085e-05,
      "loss": 1.2238,
      "num_input_tokens_seen": 48455896,
      "step": 83530
    },
    {
      "epoch": 12.441912421805183,
      "grad_norm": 8.9375,
      "learning_rate": 1.877530022642684e-05,
      "loss": 1.2882,
      "num_input_tokens_seen": 48458776,
      "step": 83535
    },
    {
      "epoch": 12.442657134346142,
      "grad_norm": 7.15625,
      "learning_rate": 1.8772153192980578e-05,
      "loss": 1.1137,
      "num_input_tokens_seen": 48461688,
      "step": 83540
    },
    {
      "epoch": 12.443401846887102,
      "grad_norm": 6.53125,
      "learning_rate": 1.8769006264747445e-05,
      "loss": 0.7182,
      "num_input_tokens_seen": 48464376,
      "step": 83545
    },
    {
      "epoch": 12.444146559428061,
      "grad_norm": 7.59375,
      "learning_rate": 1.8765859441780625e-05,
      "loss": 1.4587,
      "num_input_tokens_seen": 48467096,
      "step": 83550
    },
    {
      "epoch": 12.44489127196902,
      "grad_norm": 7.84375,
      "learning_rate": 1.8762712724133266e-05,
      "loss": 0.9904,
      "num_input_tokens_seen": 48469880,
      "step": 83555
    },
    {
      "epoch": 12.445635984509979,
      "grad_norm": 6.9375,
      "learning_rate": 1.8759566111858544e-05,
      "loss": 0.865,
      "num_input_tokens_seen": 48472920,
      "step": 83560
    },
    {
      "epoch": 12.44638069705094,
      "grad_norm": 11.375,
      "learning_rate": 1.87564196050096e-05,
      "loss": 1.2945,
      "num_input_tokens_seen": 48476120,
      "step": 83565
    },
    {
      "epoch": 12.447125409591898,
      "grad_norm": 6.1875,
      "learning_rate": 1.8753273203639614e-05,
      "loss": 0.9061,
      "num_input_tokens_seen": 48478936,
      "step": 83570
    },
    {
      "epoch": 12.447870122132857,
      "grad_norm": 10.0,
      "learning_rate": 1.875012690780172e-05,
      "loss": 1.276,
      "num_input_tokens_seen": 48481592,
      "step": 83575
    },
    {
      "epoch": 12.448614834673815,
      "grad_norm": 17.25,
      "learning_rate": 1.8746980717549088e-05,
      "loss": 1.3728,
      "num_input_tokens_seen": 48484280,
      "step": 83580
    },
    {
      "epoch": 12.449359547214776,
      "grad_norm": 3.640625,
      "learning_rate": 1.8743834632934858e-05,
      "loss": 0.9818,
      "num_input_tokens_seen": 48487224,
      "step": 83585
    },
    {
      "epoch": 12.450104259755735,
      "grad_norm": 12.6875,
      "learning_rate": 1.8740688654012172e-05,
      "loss": 1.3744,
      "num_input_tokens_seen": 48490104,
      "step": 83590
    },
    {
      "epoch": 12.450848972296694,
      "grad_norm": 4.65625,
      "learning_rate": 1.8737542780834205e-05,
      "loss": 1.0889,
      "num_input_tokens_seen": 48492888,
      "step": 83595
    },
    {
      "epoch": 12.451593684837652,
      "grad_norm": 15.1875,
      "learning_rate": 1.8734397013454075e-05,
      "loss": 1.2644,
      "num_input_tokens_seen": 48495672,
      "step": 83600
    },
    {
      "epoch": 12.452338397378611,
      "grad_norm": 7.9375,
      "learning_rate": 1.873125135192495e-05,
      "loss": 1.102,
      "num_input_tokens_seen": 48498584,
      "step": 83605
    },
    {
      "epoch": 12.453083109919572,
      "grad_norm": 17.125,
      "learning_rate": 1.8728105796299954e-05,
      "loss": 1.2249,
      "num_input_tokens_seen": 48501496,
      "step": 83610
    },
    {
      "epoch": 12.45382782246053,
      "grad_norm": 8.75,
      "learning_rate": 1.8724960346632247e-05,
      "loss": 1.1206,
      "num_input_tokens_seen": 48504312,
      "step": 83615
    },
    {
      "epoch": 12.454572535001489,
      "grad_norm": 10.625,
      "learning_rate": 1.8721815002974954e-05,
      "loss": 0.9342,
      "num_input_tokens_seen": 48507096,
      "step": 83620
    },
    {
      "epoch": 12.455317247542448,
      "grad_norm": 7.5625,
      "learning_rate": 1.8718669765381207e-05,
      "loss": 1.2724,
      "num_input_tokens_seen": 48510200,
      "step": 83625
    },
    {
      "epoch": 12.456061960083408,
      "grad_norm": 9.75,
      "learning_rate": 1.8715524633904157e-05,
      "loss": 0.9472,
      "num_input_tokens_seen": 48513240,
      "step": 83630
    },
    {
      "epoch": 12.456806672624367,
      "grad_norm": 9.5625,
      "learning_rate": 1.8712379608596926e-05,
      "loss": 1.35,
      "num_input_tokens_seen": 48516120,
      "step": 83635
    },
    {
      "epoch": 12.457551385165326,
      "grad_norm": 17.125,
      "learning_rate": 1.8709234689512656e-05,
      "loss": 1.0037,
      "num_input_tokens_seen": 48519128,
      "step": 83640
    },
    {
      "epoch": 12.458296097706285,
      "grad_norm": 4.5625,
      "learning_rate": 1.8706089876704468e-05,
      "loss": 0.9024,
      "num_input_tokens_seen": 48522008,
      "step": 83645
    },
    {
      "epoch": 12.459040810247245,
      "grad_norm": 12.875,
      "learning_rate": 1.8702945170225504e-05,
      "loss": 1.2338,
      "num_input_tokens_seen": 48524856,
      "step": 83650
    },
    {
      "epoch": 12.459785522788204,
      "grad_norm": 9.375,
      "learning_rate": 1.8699800570128868e-05,
      "loss": 0.8224,
      "num_input_tokens_seen": 48527736,
      "step": 83655
    },
    {
      "epoch": 12.460530235329163,
      "grad_norm": 7.15625,
      "learning_rate": 1.8696656076467705e-05,
      "loss": 1.1419,
      "num_input_tokens_seen": 48530648,
      "step": 83660
    },
    {
      "epoch": 12.461274947870121,
      "grad_norm": 12.25,
      "learning_rate": 1.8693511689295138e-05,
      "loss": 0.9053,
      "num_input_tokens_seen": 48533560,
      "step": 83665
    },
    {
      "epoch": 12.462019660411082,
      "grad_norm": 7.5,
      "learning_rate": 1.8690367408664265e-05,
      "loss": 0.9724,
      "num_input_tokens_seen": 48536376,
      "step": 83670
    },
    {
      "epoch": 12.46276437295204,
      "grad_norm": 12.875,
      "learning_rate": 1.8687223234628237e-05,
      "loss": 0.871,
      "num_input_tokens_seen": 48539160,
      "step": 83675
    },
    {
      "epoch": 12.463509085493,
      "grad_norm": 12.125,
      "learning_rate": 1.868407916724015e-05,
      "loss": 1.4058,
      "num_input_tokens_seen": 48542328,
      "step": 83680
    },
    {
      "epoch": 12.464253798033958,
      "grad_norm": 8.875,
      "learning_rate": 1.868093520655313e-05,
      "loss": 1.2368,
      "num_input_tokens_seen": 48545240,
      "step": 83685
    },
    {
      "epoch": 12.464998510574919,
      "grad_norm": 12.5,
      "learning_rate": 1.8677791352620278e-05,
      "loss": 1.0259,
      "num_input_tokens_seen": 48548376,
      "step": 83690
    },
    {
      "epoch": 12.465743223115878,
      "grad_norm": 5.46875,
      "learning_rate": 1.8674647605494727e-05,
      "loss": 0.9423,
      "num_input_tokens_seen": 48551576,
      "step": 83695
    },
    {
      "epoch": 12.466487935656836,
      "grad_norm": 11.4375,
      "learning_rate": 1.8671503965229572e-05,
      "loss": 1.0735,
      "num_input_tokens_seen": 48554296,
      "step": 83700
    },
    {
      "epoch": 12.467232648197795,
      "grad_norm": 10.0625,
      "learning_rate": 1.8668360431877918e-05,
      "loss": 0.962,
      "num_input_tokens_seen": 48556824,
      "step": 83705
    },
    {
      "epoch": 12.467977360738756,
      "grad_norm": 7.3125,
      "learning_rate": 1.8665217005492892e-05,
      "loss": 1.1556,
      "num_input_tokens_seen": 48560056,
      "step": 83710
    },
    {
      "epoch": 12.468722073279714,
      "grad_norm": 9.0,
      "learning_rate": 1.8662073686127575e-05,
      "loss": 1.4047,
      "num_input_tokens_seen": 48562936,
      "step": 83715
    },
    {
      "epoch": 12.469466785820673,
      "grad_norm": 2.71875,
      "learning_rate": 1.865893047383509e-05,
      "loss": 0.7224,
      "num_input_tokens_seen": 48565912,
      "step": 83720
    },
    {
      "epoch": 12.470211498361632,
      "grad_norm": 18.625,
      "learning_rate": 1.865578736866852e-05,
      "loss": 1.3795,
      "num_input_tokens_seen": 48568728,
      "step": 83725
    },
    {
      "epoch": 12.470956210902592,
      "grad_norm": 8.25,
      "learning_rate": 1.8652644370680986e-05,
      "loss": 0.9352,
      "num_input_tokens_seen": 48571640,
      "step": 83730
    },
    {
      "epoch": 12.471700923443551,
      "grad_norm": 18.375,
      "learning_rate": 1.8649501479925562e-05,
      "loss": 1.1946,
      "num_input_tokens_seen": 48574648,
      "step": 83735
    },
    {
      "epoch": 12.47244563598451,
      "grad_norm": 7.875,
      "learning_rate": 1.8646358696455365e-05,
      "loss": 0.9477,
      "num_input_tokens_seen": 48577400,
      "step": 83740
    },
    {
      "epoch": 12.473190348525469,
      "grad_norm": 7.96875,
      "learning_rate": 1.8643216020323483e-05,
      "loss": 1.1348,
      "num_input_tokens_seen": 48580408,
      "step": 83745
    },
    {
      "epoch": 12.473935061066427,
      "grad_norm": 7.78125,
      "learning_rate": 1.8640073451583003e-05,
      "loss": 0.8737,
      "num_input_tokens_seen": 48583384,
      "step": 83750
    },
    {
      "epoch": 12.474679773607388,
      "grad_norm": 9.5625,
      "learning_rate": 1.8636930990287015e-05,
      "loss": 1.0179,
      "num_input_tokens_seen": 48586456,
      "step": 83755
    },
    {
      "epoch": 12.475424486148347,
      "grad_norm": 10.6875,
      "learning_rate": 1.8633788636488605e-05,
      "loss": 1.2878,
      "num_input_tokens_seen": 48589048,
      "step": 83760
    },
    {
      "epoch": 12.476169198689306,
      "grad_norm": 13.0625,
      "learning_rate": 1.8630646390240876e-05,
      "loss": 1.1409,
      "num_input_tokens_seen": 48592056,
      "step": 83765
    },
    {
      "epoch": 12.476913911230264,
      "grad_norm": 11.125,
      "learning_rate": 1.8627504251596895e-05,
      "loss": 1.0065,
      "num_input_tokens_seen": 48595128,
      "step": 83770
    },
    {
      "epoch": 12.477658623771225,
      "grad_norm": 14.375,
      "learning_rate": 1.862436222060976e-05,
      "loss": 0.9306,
      "num_input_tokens_seen": 48597880,
      "step": 83775
    },
    {
      "epoch": 12.478403336312184,
      "grad_norm": 15.6875,
      "learning_rate": 1.8621220297332544e-05,
      "loss": 0.7764,
      "num_input_tokens_seen": 48600888,
      "step": 83780
    },
    {
      "epoch": 12.479148048853142,
      "grad_norm": 19.5,
      "learning_rate": 1.8618078481818324e-05,
      "loss": 1.4113,
      "num_input_tokens_seen": 48603864,
      "step": 83785
    },
    {
      "epoch": 12.479892761394101,
      "grad_norm": 12.375,
      "learning_rate": 1.861493677412019e-05,
      "loss": 1.1832,
      "num_input_tokens_seen": 48606456,
      "step": 83790
    },
    {
      "epoch": 12.480637473935062,
      "grad_norm": 7.625,
      "learning_rate": 1.8611795174291198e-05,
      "loss": 1.1286,
      "num_input_tokens_seen": 48609240,
      "step": 83795
    },
    {
      "epoch": 12.48138218647602,
      "grad_norm": 9.1875,
      "learning_rate": 1.8608653682384442e-05,
      "loss": 1.0507,
      "num_input_tokens_seen": 48612088,
      "step": 83800
    },
    {
      "epoch": 12.48212689901698,
      "grad_norm": 12.25,
      "learning_rate": 1.8605512298452977e-05,
      "loss": 0.9609,
      "num_input_tokens_seen": 48615128,
      "step": 83805
    },
    {
      "epoch": 12.482871611557938,
      "grad_norm": 5.4375,
      "learning_rate": 1.8602371022549895e-05,
      "loss": 0.996,
      "num_input_tokens_seen": 48618104,
      "step": 83810
    },
    {
      "epoch": 12.483616324098898,
      "grad_norm": 11.4375,
      "learning_rate": 1.8599229854728244e-05,
      "loss": 1.354,
      "num_input_tokens_seen": 48620984,
      "step": 83815
    },
    {
      "epoch": 12.484361036639857,
      "grad_norm": 10.3125,
      "learning_rate": 1.8596088795041106e-05,
      "loss": 0.9354,
      "num_input_tokens_seen": 48623672,
      "step": 83820
    },
    {
      "epoch": 12.485105749180816,
      "grad_norm": 13.0,
      "learning_rate": 1.859294784354154e-05,
      "loss": 1.0838,
      "num_input_tokens_seen": 48626648,
      "step": 83825
    },
    {
      "epoch": 12.485850461721775,
      "grad_norm": 12.75,
      "learning_rate": 1.8589807000282592e-05,
      "loss": 1.0753,
      "num_input_tokens_seen": 48629912,
      "step": 83830
    },
    {
      "epoch": 12.486595174262735,
      "grad_norm": 15.5625,
      "learning_rate": 1.858666626531736e-05,
      "loss": 1.232,
      "num_input_tokens_seen": 48633016,
      "step": 83835
    },
    {
      "epoch": 12.487339886803694,
      "grad_norm": 6.03125,
      "learning_rate": 1.8583525638698873e-05,
      "loss": 0.67,
      "num_input_tokens_seen": 48635928,
      "step": 83840
    },
    {
      "epoch": 12.488084599344653,
      "grad_norm": 10.125,
      "learning_rate": 1.85803851204802e-05,
      "loss": 0.9873,
      "num_input_tokens_seen": 48638776,
      "step": 83845
    },
    {
      "epoch": 12.488829311885612,
      "grad_norm": 7.59375,
      "learning_rate": 1.857724471071439e-05,
      "loss": 0.9063,
      "num_input_tokens_seen": 48642168,
      "step": 83850
    },
    {
      "epoch": 12.489574024426572,
      "grad_norm": 6.59375,
      "learning_rate": 1.8574104409454514e-05,
      "loss": 1.0195,
      "num_input_tokens_seen": 48645208,
      "step": 83855
    },
    {
      "epoch": 12.49031873696753,
      "grad_norm": 11.75,
      "learning_rate": 1.857096421675361e-05,
      "loss": 0.9992,
      "num_input_tokens_seen": 48648216,
      "step": 83860
    },
    {
      "epoch": 12.49106344950849,
      "grad_norm": 7.71875,
      "learning_rate": 1.8567824132664724e-05,
      "loss": 0.9089,
      "num_input_tokens_seen": 48651000,
      "step": 83865
    },
    {
      "epoch": 12.491808162049448,
      "grad_norm": 16.0,
      "learning_rate": 1.856468415724092e-05,
      "loss": 1.7315,
      "num_input_tokens_seen": 48653560,
      "step": 83870
    },
    {
      "epoch": 12.492552874590409,
      "grad_norm": 14.0625,
      "learning_rate": 1.8561544290535234e-05,
      "loss": 1.1132,
      "num_input_tokens_seen": 48656312,
      "step": 83875
    },
    {
      "epoch": 12.493297587131368,
      "grad_norm": 15.5625,
      "learning_rate": 1.8558404532600717e-05,
      "loss": 1.167,
      "num_input_tokens_seen": 48659512,
      "step": 83880
    },
    {
      "epoch": 12.494042299672326,
      "grad_norm": 6.6875,
      "learning_rate": 1.8555264883490397e-05,
      "loss": 1.1538,
      "num_input_tokens_seen": 48662488,
      "step": 83885
    },
    {
      "epoch": 12.494787012213285,
      "grad_norm": 8.4375,
      "learning_rate": 1.8552125343257337e-05,
      "loss": 0.8436,
      "num_input_tokens_seen": 48665368,
      "step": 83890
    },
    {
      "epoch": 12.495531724754246,
      "grad_norm": 11.875,
      "learning_rate": 1.8548985911954557e-05,
      "loss": 1.2813,
      "num_input_tokens_seen": 48668056,
      "step": 83895
    },
    {
      "epoch": 12.496276437295204,
      "grad_norm": 11.625,
      "learning_rate": 1.8545846589635115e-05,
      "loss": 1.3121,
      "num_input_tokens_seen": 48670808,
      "step": 83900
    },
    {
      "epoch": 12.497021149836163,
      "grad_norm": 17.125,
      "learning_rate": 1.8542707376352033e-05,
      "loss": 1.2301,
      "num_input_tokens_seen": 48673816,
      "step": 83905
    },
    {
      "epoch": 12.497765862377122,
      "grad_norm": 14.5,
      "learning_rate": 1.853956827215834e-05,
      "loss": 1.203,
      "num_input_tokens_seen": 48676632,
      "step": 83910
    },
    {
      "epoch": 12.498510574918082,
      "grad_norm": 6.9375,
      "learning_rate": 1.8536429277107086e-05,
      "loss": 1.0807,
      "num_input_tokens_seen": 48679576,
      "step": 83915
    },
    {
      "epoch": 12.499255287459041,
      "grad_norm": 19.0,
      "learning_rate": 1.8533290391251278e-05,
      "loss": 1.2871,
      "num_input_tokens_seen": 48682360,
      "step": 83920
    },
    {
      "epoch": 12.5,
      "grad_norm": 6.40625,
      "learning_rate": 1.8530151614643966e-05,
      "loss": 1.0848,
      "num_input_tokens_seen": 48685272,
      "step": 83925
    },
    {
      "epoch": 12.500744712540959,
      "grad_norm": 9.75,
      "learning_rate": 1.8527012947338155e-05,
      "loss": 1.0212,
      "num_input_tokens_seen": 48687800,
      "step": 83930
    },
    {
      "epoch": 12.501489425081918,
      "grad_norm": 6.1875,
      "learning_rate": 1.852387438938689e-05,
      "loss": 1.0139,
      "num_input_tokens_seen": 48690808,
      "step": 83935
    },
    {
      "epoch": 12.502234137622878,
      "grad_norm": 12.0,
      "learning_rate": 1.8520735940843187e-05,
      "loss": 1.3107,
      "num_input_tokens_seen": 48693720,
      "step": 83940
    },
    {
      "epoch": 12.502978850163837,
      "grad_norm": 10.0625,
      "learning_rate": 1.8517597601760062e-05,
      "loss": 1.2322,
      "num_input_tokens_seen": 48696440,
      "step": 83945
    },
    {
      "epoch": 12.503723562704796,
      "grad_norm": 11.625,
      "learning_rate": 1.851445937219054e-05,
      "loss": 1.058,
      "num_input_tokens_seen": 48699448,
      "step": 83950
    },
    {
      "epoch": 12.504468275245754,
      "grad_norm": 10.4375,
      "learning_rate": 1.8511321252187625e-05,
      "loss": 0.965,
      "num_input_tokens_seen": 48702168,
      "step": 83955
    },
    {
      "epoch": 12.505212987786715,
      "grad_norm": 8.0625,
      "learning_rate": 1.8508183241804356e-05,
      "loss": 1.0618,
      "num_input_tokens_seen": 48705176,
      "step": 83960
    },
    {
      "epoch": 12.505957700327674,
      "grad_norm": 15.4375,
      "learning_rate": 1.850504534109372e-05,
      "loss": 1.164,
      "num_input_tokens_seen": 48708216,
      "step": 83965
    },
    {
      "epoch": 12.506702412868632,
      "grad_norm": 11.1875,
      "learning_rate": 1.8501907550108752e-05,
      "loss": 1.0977,
      "num_input_tokens_seen": 48711160,
      "step": 83970
    },
    {
      "epoch": 12.507447125409591,
      "grad_norm": 6.53125,
      "learning_rate": 1.8498769868902445e-05,
      "loss": 1.119,
      "num_input_tokens_seen": 48713944,
      "step": 83975
    },
    {
      "epoch": 12.508191837950552,
      "grad_norm": 11.5625,
      "learning_rate": 1.849563229752782e-05,
      "loss": 1.3051,
      "num_input_tokens_seen": 48717080,
      "step": 83980
    },
    {
      "epoch": 12.50893655049151,
      "grad_norm": 11.625,
      "learning_rate": 1.849249483603788e-05,
      "loss": 1.0959,
      "num_input_tokens_seen": 48720120,
      "step": 83985
    },
    {
      "epoch": 12.50968126303247,
      "grad_norm": 20.0,
      "learning_rate": 1.8489357484485616e-05,
      "loss": 1.2937,
      "num_input_tokens_seen": 48722936,
      "step": 83990
    },
    {
      "epoch": 12.510425975573428,
      "grad_norm": 12.3125,
      "learning_rate": 1.8486220242924042e-05,
      "loss": 1.2788,
      "num_input_tokens_seen": 48725944,
      "step": 83995
    },
    {
      "epoch": 12.511170688114388,
      "grad_norm": 9.75,
      "learning_rate": 1.8483083111406154e-05,
      "loss": 1.1325,
      "num_input_tokens_seen": 48728824,
      "step": 84000
    },
    {
      "epoch": 12.511915400655347,
      "grad_norm": 8.0625,
      "learning_rate": 1.8479946089984963e-05,
      "loss": 1.2419,
      "num_input_tokens_seen": 48731768,
      "step": 84005
    },
    {
      "epoch": 12.512660113196306,
      "grad_norm": 9.6875,
      "learning_rate": 1.8476809178713446e-05,
      "loss": 1.1245,
      "num_input_tokens_seen": 48734936,
      "step": 84010
    },
    {
      "epoch": 12.513404825737265,
      "grad_norm": 9.6875,
      "learning_rate": 1.8473672377644617e-05,
      "loss": 1.1249,
      "num_input_tokens_seen": 48737624,
      "step": 84015
    },
    {
      "epoch": 12.514149538278225,
      "grad_norm": 16.25,
      "learning_rate": 1.8470535686831446e-05,
      "loss": 0.8861,
      "num_input_tokens_seen": 48740568,
      "step": 84020
    },
    {
      "epoch": 12.514894250819184,
      "grad_norm": 12.375,
      "learning_rate": 1.8467399106326954e-05,
      "loss": 1.093,
      "num_input_tokens_seen": 48743576,
      "step": 84025
    },
    {
      "epoch": 12.515638963360143,
      "grad_norm": 13.9375,
      "learning_rate": 1.8464262636184117e-05,
      "loss": 1.1561,
      "num_input_tokens_seen": 48746264,
      "step": 84030
    },
    {
      "epoch": 12.516383675901102,
      "grad_norm": 11.1875,
      "learning_rate": 1.8461126276455904e-05,
      "loss": 1.3023,
      "num_input_tokens_seen": 48748984,
      "step": 84035
    },
    {
      "epoch": 12.517128388442062,
      "grad_norm": 9.0625,
      "learning_rate": 1.8457990027195325e-05,
      "loss": 1.1577,
      "num_input_tokens_seen": 48751672,
      "step": 84040
    },
    {
      "epoch": 12.51787310098302,
      "grad_norm": 15.25,
      "learning_rate": 1.8454853888455352e-05,
      "loss": 1.0117,
      "num_input_tokens_seen": 48754552,
      "step": 84045
    },
    {
      "epoch": 12.51861781352398,
      "grad_norm": 10.1875,
      "learning_rate": 1.845171786028898e-05,
      "loss": 1.2012,
      "num_input_tokens_seen": 48757240,
      "step": 84050
    },
    {
      "epoch": 12.519362526064938,
      "grad_norm": 10.1875,
      "learning_rate": 1.8448581942749167e-05,
      "loss": 0.8264,
      "num_input_tokens_seen": 48760152,
      "step": 84055
    },
    {
      "epoch": 12.520107238605899,
      "grad_norm": 10.875,
      "learning_rate": 1.844544613588891e-05,
      "loss": 0.8931,
      "num_input_tokens_seen": 48762872,
      "step": 84060
    },
    {
      "epoch": 12.520851951146858,
      "grad_norm": 10.375,
      "learning_rate": 1.8442310439761185e-05,
      "loss": 1.2015,
      "num_input_tokens_seen": 48765752,
      "step": 84065
    },
    {
      "epoch": 12.521596663687816,
      "grad_norm": 9.875,
      "learning_rate": 1.8439174854418946e-05,
      "loss": 1.1864,
      "num_input_tokens_seen": 48768632,
      "step": 84070
    },
    {
      "epoch": 12.522341376228775,
      "grad_norm": 14.4375,
      "learning_rate": 1.843603937991519e-05,
      "loss": 1.2148,
      "num_input_tokens_seen": 48771416,
      "step": 84075
    },
    {
      "epoch": 12.523086088769734,
      "grad_norm": 12.125,
      "learning_rate": 1.8432904016302872e-05,
      "loss": 0.9384,
      "num_input_tokens_seen": 48774232,
      "step": 84080
    },
    {
      "epoch": 12.523830801310694,
      "grad_norm": 7.125,
      "learning_rate": 1.8429768763634974e-05,
      "loss": 0.8243,
      "num_input_tokens_seen": 48777368,
      "step": 84085
    },
    {
      "epoch": 12.524575513851653,
      "grad_norm": 4.625,
      "learning_rate": 1.8426633621964443e-05,
      "loss": 0.7226,
      "num_input_tokens_seen": 48780344,
      "step": 84090
    },
    {
      "epoch": 12.525320226392612,
      "grad_norm": 13.125,
      "learning_rate": 1.8423498591344267e-05,
      "loss": 1.2588,
      "num_input_tokens_seen": 48783032,
      "step": 84095
    },
    {
      "epoch": 12.526064938933573,
      "grad_norm": 14.3125,
      "learning_rate": 1.8420363671827387e-05,
      "loss": 1.0198,
      "num_input_tokens_seen": 48785816,
      "step": 84100
    },
    {
      "epoch": 12.526809651474531,
      "grad_norm": 8.1875,
      "learning_rate": 1.8417228863466786e-05,
      "loss": 1.0146,
      "num_input_tokens_seen": 48788696,
      "step": 84105
    },
    {
      "epoch": 12.52755436401549,
      "grad_norm": 11.75,
      "learning_rate": 1.841409416631541e-05,
      "loss": 0.9703,
      "num_input_tokens_seen": 48791544,
      "step": 84110
    },
    {
      "epoch": 12.528299076556449,
      "grad_norm": 6.53125,
      "learning_rate": 1.8410959580426222e-05,
      "loss": 1.2103,
      "num_input_tokens_seen": 48794456,
      "step": 84115
    },
    {
      "epoch": 12.529043789097408,
      "grad_norm": 11.8125,
      "learning_rate": 1.8407825105852175e-05,
      "loss": 0.8611,
      "num_input_tokens_seen": 48797528,
      "step": 84120
    },
    {
      "epoch": 12.529788501638368,
      "grad_norm": 6.78125,
      "learning_rate": 1.8404690742646212e-05,
      "loss": 1.1263,
      "num_input_tokens_seen": 48800760,
      "step": 84125
    },
    {
      "epoch": 12.530533214179327,
      "grad_norm": 20.375,
      "learning_rate": 1.840155649086131e-05,
      "loss": 1.5034,
      "num_input_tokens_seen": 48803608,
      "step": 84130
    },
    {
      "epoch": 12.531277926720286,
      "grad_norm": 9.125,
      "learning_rate": 1.8398422350550386e-05,
      "loss": 1.207,
      "num_input_tokens_seen": 48806712,
      "step": 84135
    },
    {
      "epoch": 12.532022639261244,
      "grad_norm": 7.40625,
      "learning_rate": 1.8395288321766424e-05,
      "loss": 1.2087,
      "num_input_tokens_seen": 48809624,
      "step": 84140
    },
    {
      "epoch": 12.532767351802205,
      "grad_norm": 11.0,
      "learning_rate": 1.8392154404562354e-05,
      "loss": 0.9903,
      "num_input_tokens_seen": 48812472,
      "step": 84145
    },
    {
      "epoch": 12.533512064343164,
      "grad_norm": 14.1875,
      "learning_rate": 1.8389020598991113e-05,
      "loss": 1.3694,
      "num_input_tokens_seen": 48815480,
      "step": 84150
    },
    {
      "epoch": 12.534256776884122,
      "grad_norm": 14.375,
      "learning_rate": 1.8385886905105653e-05,
      "loss": 1.1116,
      "num_input_tokens_seen": 48818072,
      "step": 84155
    },
    {
      "epoch": 12.535001489425081,
      "grad_norm": 5.6875,
      "learning_rate": 1.8382753322958902e-05,
      "loss": 1.3794,
      "num_input_tokens_seen": 48820920,
      "step": 84160
    },
    {
      "epoch": 12.535746201966042,
      "grad_norm": 13.625,
      "learning_rate": 1.837961985260382e-05,
      "loss": 1.2161,
      "num_input_tokens_seen": 48823608,
      "step": 84165
    },
    {
      "epoch": 12.536490914507,
      "grad_norm": 10.375,
      "learning_rate": 1.8376486494093327e-05,
      "loss": 0.814,
      "num_input_tokens_seen": 48826616,
      "step": 84170
    },
    {
      "epoch": 12.53723562704796,
      "grad_norm": 6.53125,
      "learning_rate": 1.837335324748036e-05,
      "loss": 1.0676,
      "num_input_tokens_seen": 48829592,
      "step": 84175
    },
    {
      "epoch": 12.537980339588918,
      "grad_norm": 14.25,
      "learning_rate": 1.8370220112817854e-05,
      "loss": 0.9398,
      "num_input_tokens_seen": 48832472,
      "step": 84180
    },
    {
      "epoch": 12.538725052129879,
      "grad_norm": 5.5,
      "learning_rate": 1.836708709015875e-05,
      "loss": 1.2818,
      "num_input_tokens_seen": 48835544,
      "step": 84185
    },
    {
      "epoch": 12.539469764670837,
      "grad_norm": 11.375,
      "learning_rate": 1.836395417955597e-05,
      "loss": 1.3496,
      "num_input_tokens_seen": 48838808,
      "step": 84190
    },
    {
      "epoch": 12.540214477211796,
      "grad_norm": 14.3125,
      "learning_rate": 1.836082138106242e-05,
      "loss": 1.3052,
      "num_input_tokens_seen": 48841816,
      "step": 84195
    },
    {
      "epoch": 12.540959189752755,
      "grad_norm": 7.03125,
      "learning_rate": 1.8357688694731063e-05,
      "loss": 1.1572,
      "num_input_tokens_seen": 48845208,
      "step": 84200
    },
    {
      "epoch": 12.541703902293715,
      "grad_norm": 5.65625,
      "learning_rate": 1.8354556120614796e-05,
      "loss": 1.1035,
      "num_input_tokens_seen": 48848440,
      "step": 84205
    },
    {
      "epoch": 12.542448614834674,
      "grad_norm": 15.875,
      "learning_rate": 1.8351423658766557e-05,
      "loss": 1.209,
      "num_input_tokens_seen": 48851224,
      "step": 84210
    },
    {
      "epoch": 12.543193327375633,
      "grad_norm": 10.8125,
      "learning_rate": 1.8348291309239248e-05,
      "loss": 0.899,
      "num_input_tokens_seen": 48854200,
      "step": 84215
    },
    {
      "epoch": 12.543938039916592,
      "grad_norm": 22.375,
      "learning_rate": 1.8345159072085803e-05,
      "loss": 1.19,
      "num_input_tokens_seen": 48857368,
      "step": 84220
    },
    {
      "epoch": 12.544682752457552,
      "grad_norm": 9.6875,
      "learning_rate": 1.8342026947359137e-05,
      "loss": 1.1316,
      "num_input_tokens_seen": 48860472,
      "step": 84225
    },
    {
      "epoch": 12.545427464998511,
      "grad_norm": 10.8125,
      "learning_rate": 1.8338894935112144e-05,
      "loss": 1.0627,
      "num_input_tokens_seen": 48863224,
      "step": 84230
    },
    {
      "epoch": 12.54617217753947,
      "grad_norm": 11.875,
      "learning_rate": 1.8335763035397765e-05,
      "loss": 0.7914,
      "num_input_tokens_seen": 48866168,
      "step": 84235
    },
    {
      "epoch": 12.546916890080428,
      "grad_norm": 5.6875,
      "learning_rate": 1.833263124826889e-05,
      "loss": 0.6612,
      "num_input_tokens_seen": 48869208,
      "step": 84240
    },
    {
      "epoch": 12.547661602621389,
      "grad_norm": 11.5,
      "learning_rate": 1.832949957377844e-05,
      "loss": 1.0597,
      "num_input_tokens_seen": 48871960,
      "step": 84245
    },
    {
      "epoch": 12.548406315162348,
      "grad_norm": 7.5,
      "learning_rate": 1.83263680119793e-05,
      "loss": 1.5813,
      "num_input_tokens_seen": 48875064,
      "step": 84250
    },
    {
      "epoch": 12.549151027703306,
      "grad_norm": 11.875,
      "learning_rate": 1.8323236562924405e-05,
      "loss": 0.9218,
      "num_input_tokens_seen": 48877720,
      "step": 84255
    },
    {
      "epoch": 12.549895740244265,
      "grad_norm": 13.25,
      "learning_rate": 1.8320105226666628e-05,
      "loss": 1.1842,
      "num_input_tokens_seen": 48880600,
      "step": 84260
    },
    {
      "epoch": 12.550640452785224,
      "grad_norm": 8.3125,
      "learning_rate": 1.8316974003258898e-05,
      "loss": 0.9291,
      "num_input_tokens_seen": 48883224,
      "step": 84265
    },
    {
      "epoch": 12.551385165326185,
      "grad_norm": 7.8125,
      "learning_rate": 1.8313842892754097e-05,
      "loss": 0.9353,
      "num_input_tokens_seen": 48885976,
      "step": 84270
    },
    {
      "epoch": 12.552129877867143,
      "grad_norm": 3.375,
      "learning_rate": 1.8310711895205125e-05,
      "loss": 0.8288,
      "num_input_tokens_seen": 48888856,
      "step": 84275
    },
    {
      "epoch": 12.552874590408102,
      "grad_norm": 11.8125,
      "learning_rate": 1.8307581010664875e-05,
      "loss": 1.0871,
      "num_input_tokens_seen": 48891832,
      "step": 84280
    },
    {
      "epoch": 12.553619302949063,
      "grad_norm": 16.25,
      "learning_rate": 1.8304450239186235e-05,
      "loss": 1.0984,
      "num_input_tokens_seen": 48894808,
      "step": 84285
    },
    {
      "epoch": 12.554364015490021,
      "grad_norm": 8.5,
      "learning_rate": 1.8301319580822112e-05,
      "loss": 1.0451,
      "num_input_tokens_seen": 48898072,
      "step": 84290
    },
    {
      "epoch": 12.55510872803098,
      "grad_norm": 8.5,
      "learning_rate": 1.829818903562538e-05,
      "loss": 1.0238,
      "num_input_tokens_seen": 48901080,
      "step": 84295
    },
    {
      "epoch": 12.555853440571939,
      "grad_norm": 9.6875,
      "learning_rate": 1.8295058603648942e-05,
      "loss": 1.2024,
      "num_input_tokens_seen": 48904344,
      "step": 84300
    },
    {
      "epoch": 12.556598153112898,
      "grad_norm": 15.4375,
      "learning_rate": 1.8291928284945668e-05,
      "loss": 1.153,
      "num_input_tokens_seen": 48907192,
      "step": 84305
    },
    {
      "epoch": 12.557342865653858,
      "grad_norm": 7.65625,
      "learning_rate": 1.828879807956845e-05,
      "loss": 1.2399,
      "num_input_tokens_seen": 48910072,
      "step": 84310
    },
    {
      "epoch": 12.558087578194817,
      "grad_norm": 3.625,
      "learning_rate": 1.828566798757017e-05,
      "loss": 0.9554,
      "num_input_tokens_seen": 48912824,
      "step": 84315
    },
    {
      "epoch": 12.558832290735776,
      "grad_norm": 4.03125,
      "learning_rate": 1.8282538009003696e-05,
      "loss": 0.6576,
      "num_input_tokens_seen": 48915416,
      "step": 84320
    },
    {
      "epoch": 12.559577003276734,
      "grad_norm": 9.4375,
      "learning_rate": 1.827940814392192e-05,
      "loss": 0.7106,
      "num_input_tokens_seen": 48918616,
      "step": 84325
    },
    {
      "epoch": 12.560321715817695,
      "grad_norm": 7.09375,
      "learning_rate": 1.827627839237771e-05,
      "loss": 1.3079,
      "num_input_tokens_seen": 48921528,
      "step": 84330
    },
    {
      "epoch": 12.561066428358654,
      "grad_norm": 8.875,
      "learning_rate": 1.8273148754423953e-05,
      "loss": 0.9066,
      "num_input_tokens_seen": 48924792,
      "step": 84335
    },
    {
      "epoch": 12.561811140899612,
      "grad_norm": 9.0625,
      "learning_rate": 1.82700192301135e-05,
      "loss": 1.1124,
      "num_input_tokens_seen": 48927992,
      "step": 84340
    },
    {
      "epoch": 12.562555853440571,
      "grad_norm": 11.125,
      "learning_rate": 1.826688981949924e-05,
      "loss": 0.9592,
      "num_input_tokens_seen": 48931064,
      "step": 84345
    },
    {
      "epoch": 12.563300565981532,
      "grad_norm": 7.0625,
      "learning_rate": 1.8263760522634033e-05,
      "loss": 1.0936,
      "num_input_tokens_seen": 48933752,
      "step": 84350
    },
    {
      "epoch": 12.56404527852249,
      "grad_norm": 11.75,
      "learning_rate": 1.826063133957074e-05,
      "loss": 0.9159,
      "num_input_tokens_seen": 48936472,
      "step": 84355
    },
    {
      "epoch": 12.56478999106345,
      "grad_norm": 8.25,
      "learning_rate": 1.8257502270362235e-05,
      "loss": 0.8858,
      "num_input_tokens_seen": 48939352,
      "step": 84360
    },
    {
      "epoch": 12.565534703604408,
      "grad_norm": 10.75,
      "learning_rate": 1.8254373315061364e-05,
      "loss": 1.1783,
      "num_input_tokens_seen": 48942296,
      "step": 84365
    },
    {
      "epoch": 12.566279416145369,
      "grad_norm": 6.65625,
      "learning_rate": 1.8251244473721017e-05,
      "loss": 1.1463,
      "num_input_tokens_seen": 48945304,
      "step": 84370
    },
    {
      "epoch": 12.567024128686327,
      "grad_norm": 6.09375,
      "learning_rate": 1.8248115746394025e-05,
      "loss": 1.0429,
      "num_input_tokens_seen": 48948056,
      "step": 84375
    },
    {
      "epoch": 12.567768841227286,
      "grad_norm": 9.0,
      "learning_rate": 1.8244987133133264e-05,
      "loss": 1.013,
      "num_input_tokens_seen": 48951288,
      "step": 84380
    },
    {
      "epoch": 12.568513553768245,
      "grad_norm": 14.75,
      "learning_rate": 1.8241858633991578e-05,
      "loss": 1.1589,
      "num_input_tokens_seen": 48954328,
      "step": 84385
    },
    {
      "epoch": 12.569258266309205,
      "grad_norm": 11.1875,
      "learning_rate": 1.8238730249021812e-05,
      "loss": 1.1257,
      "num_input_tokens_seen": 48957368,
      "step": 84390
    },
    {
      "epoch": 12.570002978850164,
      "grad_norm": 23.625,
      "learning_rate": 1.8235601978276838e-05,
      "loss": 1.1008,
      "num_input_tokens_seen": 48960216,
      "step": 84395
    },
    {
      "epoch": 12.570747691391123,
      "grad_norm": 12.625,
      "learning_rate": 1.823247382180948e-05,
      "loss": 1.2372,
      "num_input_tokens_seen": 48962872,
      "step": 84400
    },
    {
      "epoch": 12.571492403932082,
      "grad_norm": 11.0,
      "learning_rate": 1.8229345779672613e-05,
      "loss": 1.2443,
      "num_input_tokens_seen": 48965496,
      "step": 84405
    },
    {
      "epoch": 12.572237116473042,
      "grad_norm": 12.9375,
      "learning_rate": 1.8226217851919062e-05,
      "loss": 1.1579,
      "num_input_tokens_seen": 48968312,
      "step": 84410
    },
    {
      "epoch": 12.572981829014001,
      "grad_norm": 12.75,
      "learning_rate": 1.8223090038601678e-05,
      "loss": 1.3206,
      "num_input_tokens_seen": 48971256,
      "step": 84415
    },
    {
      "epoch": 12.57372654155496,
      "grad_norm": 11.6875,
      "learning_rate": 1.8219962339773292e-05,
      "loss": 1.2601,
      "num_input_tokens_seen": 48973944,
      "step": 84420
    },
    {
      "epoch": 12.574471254095918,
      "grad_norm": 7.65625,
      "learning_rate": 1.8216834755486763e-05,
      "loss": 1.5558,
      "num_input_tokens_seen": 48976856,
      "step": 84425
    },
    {
      "epoch": 12.575215966636879,
      "grad_norm": 10.6875,
      "learning_rate": 1.821370728579491e-05,
      "loss": 0.9998,
      "num_input_tokens_seen": 48979640,
      "step": 84430
    },
    {
      "epoch": 12.575960679177838,
      "grad_norm": 7.84375,
      "learning_rate": 1.821057993075057e-05,
      "loss": 1.1359,
      "num_input_tokens_seen": 48982712,
      "step": 84435
    },
    {
      "epoch": 12.576705391718797,
      "grad_norm": 9.4375,
      "learning_rate": 1.8207452690406594e-05,
      "loss": 1.247,
      "num_input_tokens_seen": 48985432,
      "step": 84440
    },
    {
      "epoch": 12.577450104259755,
      "grad_norm": 20.125,
      "learning_rate": 1.8204325564815796e-05,
      "loss": 1.5018,
      "num_input_tokens_seen": 48988376,
      "step": 84445
    },
    {
      "epoch": 12.578194816800714,
      "grad_norm": 11.125,
      "learning_rate": 1.820119855403101e-05,
      "loss": 0.7353,
      "num_input_tokens_seen": 48991000,
      "step": 84450
    },
    {
      "epoch": 12.578939529341675,
      "grad_norm": 12.6875,
      "learning_rate": 1.819807165810506e-05,
      "loss": 1.0552,
      "num_input_tokens_seen": 48993944,
      "step": 84455
    },
    {
      "epoch": 12.579684241882633,
      "grad_norm": 13.875,
      "learning_rate": 1.819494487709078e-05,
      "loss": 1.3799,
      "num_input_tokens_seen": 48996632,
      "step": 84460
    },
    {
      "epoch": 12.580428954423592,
      "grad_norm": 10.8125,
      "learning_rate": 1.8191818211040997e-05,
      "loss": 1.1302,
      "num_input_tokens_seen": 48999576,
      "step": 84465
    },
    {
      "epoch": 12.58117366696455,
      "grad_norm": 9.0,
      "learning_rate": 1.8188691660008513e-05,
      "loss": 1.0583,
      "num_input_tokens_seen": 49002264,
      "step": 84470
    },
    {
      "epoch": 12.581918379505511,
      "grad_norm": 8.0625,
      "learning_rate": 1.818556522404617e-05,
      "loss": 1.5916,
      "num_input_tokens_seen": 49005048,
      "step": 84475
    },
    {
      "epoch": 12.58266309204647,
      "grad_norm": 10.0625,
      "learning_rate": 1.818243890320677e-05,
      "loss": 0.8861,
      "num_input_tokens_seen": 49007832,
      "step": 84480
    },
    {
      "epoch": 12.583407804587429,
      "grad_norm": 7.71875,
      "learning_rate": 1.8179312697543145e-05,
      "loss": 1.2564,
      "num_input_tokens_seen": 49010872,
      "step": 84485
    },
    {
      "epoch": 12.584152517128388,
      "grad_norm": 7.4375,
      "learning_rate": 1.8176186607108086e-05,
      "loss": 0.9507,
      "num_input_tokens_seen": 49013752,
      "step": 84490
    },
    {
      "epoch": 12.584897229669348,
      "grad_norm": 9.0,
      "learning_rate": 1.817306063195443e-05,
      "loss": 1.005,
      "num_input_tokens_seen": 49016600,
      "step": 84495
    },
    {
      "epoch": 12.585641942210307,
      "grad_norm": 18.125,
      "learning_rate": 1.8169934772134974e-05,
      "loss": 1.0773,
      "num_input_tokens_seen": 49019352,
      "step": 84500
    },
    {
      "epoch": 12.586386654751266,
      "grad_norm": 6.90625,
      "learning_rate": 1.8166809027702522e-05,
      "loss": 1.1834,
      "num_input_tokens_seen": 49022616,
      "step": 84505
    },
    {
      "epoch": 12.587131367292224,
      "grad_norm": 4.15625,
      "learning_rate": 1.8163683398709898e-05,
      "loss": 1.1801,
      "num_input_tokens_seen": 49025464,
      "step": 84510
    },
    {
      "epoch": 12.587876079833185,
      "grad_norm": 10.6875,
      "learning_rate": 1.8160557885209884e-05,
      "loss": 0.9402,
      "num_input_tokens_seen": 49028632,
      "step": 84515
    },
    {
      "epoch": 12.588620792374144,
      "grad_norm": 6.09375,
      "learning_rate": 1.81574324872553e-05,
      "loss": 1.1078,
      "num_input_tokens_seen": 49031384,
      "step": 84520
    },
    {
      "epoch": 12.589365504915103,
      "grad_norm": 7.78125,
      "learning_rate": 1.8154307204898933e-05,
      "loss": 1.2005,
      "num_input_tokens_seen": 49034360,
      "step": 84525
    },
    {
      "epoch": 12.590110217456061,
      "grad_norm": 10.6875,
      "learning_rate": 1.8151182038193594e-05,
      "loss": 1.172,
      "num_input_tokens_seen": 49037176,
      "step": 84530
    },
    {
      "epoch": 12.590854929997022,
      "grad_norm": 10.375,
      "learning_rate": 1.814805698719207e-05,
      "loss": 0.8874,
      "num_input_tokens_seen": 49040152,
      "step": 84535
    },
    {
      "epoch": 12.59159964253798,
      "grad_norm": 8.625,
      "learning_rate": 1.8144932051947166e-05,
      "loss": 0.9856,
      "num_input_tokens_seen": 49043000,
      "step": 84540
    },
    {
      "epoch": 12.59234435507894,
      "grad_norm": 6.8125,
      "learning_rate": 1.814180723251166e-05,
      "loss": 1.0492,
      "num_input_tokens_seen": 49045880,
      "step": 84545
    },
    {
      "epoch": 12.593089067619898,
      "grad_norm": 16.5,
      "learning_rate": 1.8138682528938354e-05,
      "loss": 1.1646,
      "num_input_tokens_seen": 49048920,
      "step": 84550
    },
    {
      "epoch": 12.593833780160859,
      "grad_norm": 9.9375,
      "learning_rate": 1.8135557941280035e-05,
      "loss": 1.2281,
      "num_input_tokens_seen": 49051768,
      "step": 84555
    },
    {
      "epoch": 12.594578492701817,
      "grad_norm": 12.125,
      "learning_rate": 1.813243346958948e-05,
      "loss": 1.2285,
      "num_input_tokens_seen": 49054520,
      "step": 84560
    },
    {
      "epoch": 12.595323205242776,
      "grad_norm": 21.75,
      "learning_rate": 1.812930911391949e-05,
      "loss": 1.0705,
      "num_input_tokens_seen": 49057560,
      "step": 84565
    },
    {
      "epoch": 12.596067917783735,
      "grad_norm": 16.625,
      "learning_rate": 1.8126184874322837e-05,
      "loss": 1.2993,
      "num_input_tokens_seen": 49060696,
      "step": 84570
    },
    {
      "epoch": 12.596812630324695,
      "grad_norm": 9.25,
      "learning_rate": 1.8123060750852305e-05,
      "loss": 0.988,
      "num_input_tokens_seen": 49063896,
      "step": 84575
    },
    {
      "epoch": 12.597557342865654,
      "grad_norm": 10.75,
      "learning_rate": 1.8119936743560667e-05,
      "loss": 1.1766,
      "num_input_tokens_seen": 49066584,
      "step": 84580
    },
    {
      "epoch": 12.598302055406613,
      "grad_norm": 10.3125,
      "learning_rate": 1.8116812852500713e-05,
      "loss": 1.2324,
      "num_input_tokens_seen": 49069304,
      "step": 84585
    },
    {
      "epoch": 12.599046767947572,
      "grad_norm": 12.3125,
      "learning_rate": 1.811368907772521e-05,
      "loss": 1.1303,
      "num_input_tokens_seen": 49072408,
      "step": 84590
    },
    {
      "epoch": 12.599791480488532,
      "grad_norm": 11.125,
      "learning_rate": 1.8110565419286916e-05,
      "loss": 0.8934,
      "num_input_tokens_seen": 49075224,
      "step": 84595
    },
    {
      "epoch": 12.600536193029491,
      "grad_norm": 7.09375,
      "learning_rate": 1.8107441877238634e-05,
      "loss": 1.122,
      "num_input_tokens_seen": 49078264,
      "step": 84600
    },
    {
      "epoch": 12.60128090557045,
      "grad_norm": 8.0625,
      "learning_rate": 1.8104318451633114e-05,
      "loss": 0.8115,
      "num_input_tokens_seen": 49081272,
      "step": 84605
    },
    {
      "epoch": 12.602025618111409,
      "grad_norm": 5.1875,
      "learning_rate": 1.810119514252312e-05,
      "loss": 0.8411,
      "num_input_tokens_seen": 49083992,
      "step": 84610
    },
    {
      "epoch": 12.602770330652369,
      "grad_norm": 6.625,
      "learning_rate": 1.809807194996142e-05,
      "loss": 1.1551,
      "num_input_tokens_seen": 49086616,
      "step": 84615
    },
    {
      "epoch": 12.603515043193328,
      "grad_norm": 10.125,
      "learning_rate": 1.809494887400079e-05,
      "loss": 1.3577,
      "num_input_tokens_seen": 49089624,
      "step": 84620
    },
    {
      "epoch": 12.604259755734287,
      "grad_norm": 9.5625,
      "learning_rate": 1.8091825914693966e-05,
      "loss": 0.9085,
      "num_input_tokens_seen": 49092632,
      "step": 84625
    },
    {
      "epoch": 12.605004468275245,
      "grad_norm": 7.84375,
      "learning_rate": 1.8088703072093735e-05,
      "loss": 0.9013,
      "num_input_tokens_seen": 49095544,
      "step": 84630
    },
    {
      "epoch": 12.605749180816204,
      "grad_norm": 8.625,
      "learning_rate": 1.808558034625284e-05,
      "loss": 1.2222,
      "num_input_tokens_seen": 49098456,
      "step": 84635
    },
    {
      "epoch": 12.606493893357165,
      "grad_norm": 17.375,
      "learning_rate": 1.8082457737224034e-05,
      "loss": 1.1666,
      "num_input_tokens_seen": 49101176,
      "step": 84640
    },
    {
      "epoch": 12.607238605898123,
      "grad_norm": 10.75,
      "learning_rate": 1.8079335245060076e-05,
      "loss": 1.138,
      "num_input_tokens_seen": 49103992,
      "step": 84645
    },
    {
      "epoch": 12.607983318439082,
      "grad_norm": 9.875,
      "learning_rate": 1.8076212869813706e-05,
      "loss": 0.8588,
      "num_input_tokens_seen": 49107128,
      "step": 84650
    },
    {
      "epoch": 12.608728030980041,
      "grad_norm": 6.4375,
      "learning_rate": 1.8073090611537697e-05,
      "loss": 1.0562,
      "num_input_tokens_seen": 49109944,
      "step": 84655
    },
    {
      "epoch": 12.609472743521001,
      "grad_norm": 7.15625,
      "learning_rate": 1.8069968470284768e-05,
      "loss": 0.9628,
      "num_input_tokens_seen": 49112792,
      "step": 84660
    },
    {
      "epoch": 12.61021745606196,
      "grad_norm": 6.875,
      "learning_rate": 1.806684644610769e-05,
      "loss": 1.1868,
      "num_input_tokens_seen": 49115576,
      "step": 84665
    },
    {
      "epoch": 12.610962168602919,
      "grad_norm": 11.5625,
      "learning_rate": 1.8063724539059195e-05,
      "loss": 1.0302,
      "num_input_tokens_seen": 49118520,
      "step": 84670
    },
    {
      "epoch": 12.611706881143878,
      "grad_norm": 5.5,
      "learning_rate": 1.806060274919202e-05,
      "loss": 1.218,
      "num_input_tokens_seen": 49121656,
      "step": 84675
    },
    {
      "epoch": 12.612451593684838,
      "grad_norm": 9.9375,
      "learning_rate": 1.8057481076558906e-05,
      "loss": 1.2533,
      "num_input_tokens_seen": 49124632,
      "step": 84680
    },
    {
      "epoch": 12.613196306225797,
      "grad_norm": 27.375,
      "learning_rate": 1.8054359521212592e-05,
      "loss": 1.3688,
      "num_input_tokens_seen": 49127704,
      "step": 84685
    },
    {
      "epoch": 12.613941018766756,
      "grad_norm": 13.375,
      "learning_rate": 1.805123808320582e-05,
      "loss": 1.1191,
      "num_input_tokens_seen": 49130744,
      "step": 84690
    },
    {
      "epoch": 12.614685731307715,
      "grad_norm": 5.5625,
      "learning_rate": 1.804811676259131e-05,
      "loss": 1.0714,
      "num_input_tokens_seen": 49133688,
      "step": 84695
    },
    {
      "epoch": 12.615430443848675,
      "grad_norm": 9.25,
      "learning_rate": 1.8044995559421813e-05,
      "loss": 1.2113,
      "num_input_tokens_seen": 49136472,
      "step": 84700
    },
    {
      "epoch": 12.616175156389634,
      "grad_norm": 16.375,
      "learning_rate": 1.804187447375004e-05,
      "loss": 1.3358,
      "num_input_tokens_seen": 49139672,
      "step": 84705
    },
    {
      "epoch": 12.616919868930593,
      "grad_norm": 9.3125,
      "learning_rate": 1.803875350562873e-05,
      "loss": 1.2818,
      "num_input_tokens_seen": 49142520,
      "step": 84710
    },
    {
      "epoch": 12.617664581471551,
      "grad_norm": 9.125,
      "learning_rate": 1.8035632655110607e-05,
      "loss": 1.2112,
      "num_input_tokens_seen": 49145592,
      "step": 84715
    },
    {
      "epoch": 12.618409294012512,
      "grad_norm": 9.0625,
      "learning_rate": 1.803251192224838e-05,
      "loss": 1.1056,
      "num_input_tokens_seen": 49148376,
      "step": 84720
    },
    {
      "epoch": 12.61915400655347,
      "grad_norm": 17.125,
      "learning_rate": 1.8029391307094796e-05,
      "loss": 1.4148,
      "num_input_tokens_seen": 49151160,
      "step": 84725
    },
    {
      "epoch": 12.61989871909443,
      "grad_norm": 14.4375,
      "learning_rate": 1.8026270809702547e-05,
      "loss": 1.1515,
      "num_input_tokens_seen": 49154072,
      "step": 84730
    },
    {
      "epoch": 12.620643431635388,
      "grad_norm": 11.0625,
      "learning_rate": 1.8023150430124375e-05,
      "loss": 1.3131,
      "num_input_tokens_seen": 49156792,
      "step": 84735
    },
    {
      "epoch": 12.621388144176349,
      "grad_norm": 9.0625,
      "learning_rate": 1.802003016841298e-05,
      "loss": 1.2561,
      "num_input_tokens_seen": 49159960,
      "step": 84740
    },
    {
      "epoch": 12.622132856717307,
      "grad_norm": 11.0625,
      "learning_rate": 1.801691002462109e-05,
      "loss": 1.1718,
      "num_input_tokens_seen": 49162680,
      "step": 84745
    },
    {
      "epoch": 12.622877569258266,
      "grad_norm": 6.53125,
      "learning_rate": 1.8013789998801407e-05,
      "loss": 0.8061,
      "num_input_tokens_seen": 49165432,
      "step": 84750
    },
    {
      "epoch": 12.623622281799225,
      "grad_norm": 6.03125,
      "learning_rate": 1.801067009100663e-05,
      "loss": 0.8217,
      "num_input_tokens_seen": 49168504,
      "step": 84755
    },
    {
      "epoch": 12.624366994340185,
      "grad_norm": 8.1875,
      "learning_rate": 1.800755030128949e-05,
      "loss": 0.8489,
      "num_input_tokens_seen": 49171000,
      "step": 84760
    },
    {
      "epoch": 12.625111706881144,
      "grad_norm": 15.75,
      "learning_rate": 1.800443062970267e-05,
      "loss": 1.0421,
      "num_input_tokens_seen": 49173720,
      "step": 84765
    },
    {
      "epoch": 12.625856419422103,
      "grad_norm": 15.5625,
      "learning_rate": 1.8001311076298895e-05,
      "loss": 1.1588,
      "num_input_tokens_seen": 49176376,
      "step": 84770
    },
    {
      "epoch": 12.626601131963062,
      "grad_norm": 9.0,
      "learning_rate": 1.799819164113085e-05,
      "loss": 1.2351,
      "num_input_tokens_seen": 49179160,
      "step": 84775
    },
    {
      "epoch": 12.62734584450402,
      "grad_norm": 3.9375,
      "learning_rate": 1.799507232425125e-05,
      "loss": 1.3118,
      "num_input_tokens_seen": 49181944,
      "step": 84780
    },
    {
      "epoch": 12.628090557044981,
      "grad_norm": 14.75,
      "learning_rate": 1.799195312571277e-05,
      "loss": 0.9152,
      "num_input_tokens_seen": 49184856,
      "step": 84785
    },
    {
      "epoch": 12.62883526958594,
      "grad_norm": 12.9375,
      "learning_rate": 1.7988834045568126e-05,
      "loss": 0.9432,
      "num_input_tokens_seen": 49188216,
      "step": 84790
    },
    {
      "epoch": 12.629579982126899,
      "grad_norm": 9.125,
      "learning_rate": 1.7985715083870008e-05,
      "loss": 1.149,
      "num_input_tokens_seen": 49191224,
      "step": 84795
    },
    {
      "epoch": 12.63032469466786,
      "grad_norm": 11.125,
      "learning_rate": 1.7982596240671095e-05,
      "loss": 1.258,
      "num_input_tokens_seen": 49193912,
      "step": 84800
    },
    {
      "epoch": 12.631069407208818,
      "grad_norm": 9.9375,
      "learning_rate": 1.7979477516024096e-05,
      "loss": 0.9822,
      "num_input_tokens_seen": 49196536,
      "step": 84805
    },
    {
      "epoch": 12.631814119749777,
      "grad_norm": 14.625,
      "learning_rate": 1.7976358909981686e-05,
      "loss": 1.2392,
      "num_input_tokens_seen": 49200024,
      "step": 84810
    },
    {
      "epoch": 12.632558832290735,
      "grad_norm": 11.9375,
      "learning_rate": 1.7973240422596557e-05,
      "loss": 0.9666,
      "num_input_tokens_seen": 49203096,
      "step": 84815
    },
    {
      "epoch": 12.633303544831694,
      "grad_norm": 4.28125,
      "learning_rate": 1.7970122053921378e-05,
      "loss": 0.9208,
      "num_input_tokens_seen": 49205816,
      "step": 84820
    },
    {
      "epoch": 12.634048257372655,
      "grad_norm": 15.0,
      "learning_rate": 1.7967003804008855e-05,
      "loss": 0.9824,
      "num_input_tokens_seen": 49208696,
      "step": 84825
    },
    {
      "epoch": 12.634792969913613,
      "grad_norm": 17.125,
      "learning_rate": 1.7963885672911655e-05,
      "loss": 1.1337,
      "num_input_tokens_seen": 49211928,
      "step": 84830
    },
    {
      "epoch": 12.635537682454572,
      "grad_norm": 9.875,
      "learning_rate": 1.7960767660682442e-05,
      "loss": 0.919,
      "num_input_tokens_seen": 49215000,
      "step": 84835
    },
    {
      "epoch": 12.636282394995531,
      "grad_norm": 13.625,
      "learning_rate": 1.7957649767373916e-05,
      "loss": 1.1729,
      "num_input_tokens_seen": 49217848,
      "step": 84840
    },
    {
      "epoch": 12.637027107536491,
      "grad_norm": 13.375,
      "learning_rate": 1.7954531993038737e-05,
      "loss": 1.0392,
      "num_input_tokens_seen": 49221080,
      "step": 84845
    },
    {
      "epoch": 12.63777182007745,
      "grad_norm": 9.125,
      "learning_rate": 1.7951414337729584e-05,
      "loss": 1.07,
      "num_input_tokens_seen": 49223992,
      "step": 84850
    },
    {
      "epoch": 12.638516532618409,
      "grad_norm": 12.4375,
      "learning_rate": 1.794829680149911e-05,
      "loss": 0.9598,
      "num_input_tokens_seen": 49226744,
      "step": 84855
    },
    {
      "epoch": 12.639261245159368,
      "grad_norm": 6.65625,
      "learning_rate": 1.7945179384400002e-05,
      "loss": 0.7049,
      "num_input_tokens_seen": 49229720,
      "step": 84860
    },
    {
      "epoch": 12.640005957700328,
      "grad_norm": 9.125,
      "learning_rate": 1.794206208648492e-05,
      "loss": 0.9693,
      "num_input_tokens_seen": 49232856,
      "step": 84865
    },
    {
      "epoch": 12.640750670241287,
      "grad_norm": 7.84375,
      "learning_rate": 1.7938944907806523e-05,
      "loss": 0.8304,
      "num_input_tokens_seen": 49235704,
      "step": 84870
    },
    {
      "epoch": 12.641495382782246,
      "grad_norm": 14.4375,
      "learning_rate": 1.7935827848417476e-05,
      "loss": 1.3055,
      "num_input_tokens_seen": 49238776,
      "step": 84875
    },
    {
      "epoch": 12.642240095323205,
      "grad_norm": 16.875,
      "learning_rate": 1.7932710908370434e-05,
      "loss": 1.0499,
      "num_input_tokens_seen": 49241432,
      "step": 84880
    },
    {
      "epoch": 12.642984807864165,
      "grad_norm": 12.6875,
      "learning_rate": 1.7929594087718067e-05,
      "loss": 1.085,
      "num_input_tokens_seen": 49244472,
      "step": 84885
    },
    {
      "epoch": 12.643729520405124,
      "grad_norm": 12.125,
      "learning_rate": 1.7926477386513008e-05,
      "loss": 0.7337,
      "num_input_tokens_seen": 49247320,
      "step": 84890
    },
    {
      "epoch": 12.644474232946083,
      "grad_norm": 12.5,
      "learning_rate": 1.7923360804807937e-05,
      "loss": 1.1225,
      "num_input_tokens_seen": 49250264,
      "step": 84895
    },
    {
      "epoch": 12.645218945487041,
      "grad_norm": 8.375,
      "learning_rate": 1.7920244342655485e-05,
      "loss": 0.9192,
      "num_input_tokens_seen": 49253144,
      "step": 84900
    },
    {
      "epoch": 12.645963658028002,
      "grad_norm": 7.71875,
      "learning_rate": 1.791712800010832e-05,
      "loss": 0.8841,
      "num_input_tokens_seen": 49256216,
      "step": 84905
    },
    {
      "epoch": 12.64670837056896,
      "grad_norm": 13.5625,
      "learning_rate": 1.7914011777219074e-05,
      "loss": 1.3782,
      "num_input_tokens_seen": 49258968,
      "step": 84910
    },
    {
      "epoch": 12.64745308310992,
      "grad_norm": 8.5,
      "learning_rate": 1.7910895674040387e-05,
      "loss": 1.1371,
      "num_input_tokens_seen": 49261912,
      "step": 84915
    },
    {
      "epoch": 12.648197795650878,
      "grad_norm": 6.9375,
      "learning_rate": 1.7907779690624923e-05,
      "loss": 0.8601,
      "num_input_tokens_seen": 49264952,
      "step": 84920
    },
    {
      "epoch": 12.648942508191839,
      "grad_norm": 4.09375,
      "learning_rate": 1.7904663827025304e-05,
      "loss": 0.7451,
      "num_input_tokens_seen": 49268248,
      "step": 84925
    },
    {
      "epoch": 12.649687220732797,
      "grad_norm": 14.1875,
      "learning_rate": 1.790154808329419e-05,
      "loss": 0.9432,
      "num_input_tokens_seen": 49271032,
      "step": 84930
    },
    {
      "epoch": 12.650431933273756,
      "grad_norm": 12.125,
      "learning_rate": 1.78984324594842e-05,
      "loss": 1.1623,
      "num_input_tokens_seen": 49273816,
      "step": 84935
    },
    {
      "epoch": 12.651176645814715,
      "grad_norm": 19.375,
      "learning_rate": 1.7895316955647977e-05,
      "loss": 0.918,
      "num_input_tokens_seen": 49276568,
      "step": 84940
    },
    {
      "epoch": 12.651921358355676,
      "grad_norm": 16.75,
      "learning_rate": 1.7892201571838147e-05,
      "loss": 1.1044,
      "num_input_tokens_seen": 49279352,
      "step": 84945
    },
    {
      "epoch": 12.652666070896634,
      "grad_norm": 6.5625,
      "learning_rate": 1.788908630810736e-05,
      "loss": 0.8944,
      "num_input_tokens_seen": 49281944,
      "step": 84950
    },
    {
      "epoch": 12.653410783437593,
      "grad_norm": 11.875,
      "learning_rate": 1.7885971164508227e-05,
      "loss": 1.0609,
      "num_input_tokens_seen": 49284888,
      "step": 84955
    },
    {
      "epoch": 12.654155495978552,
      "grad_norm": 9.625,
      "learning_rate": 1.7882856141093372e-05,
      "loss": 1.1118,
      "num_input_tokens_seen": 49287896,
      "step": 84960
    },
    {
      "epoch": 12.65490020851951,
      "grad_norm": 8.375,
      "learning_rate": 1.7879741237915444e-05,
      "loss": 0.9135,
      "num_input_tokens_seen": 49290616,
      "step": 84965
    },
    {
      "epoch": 12.655644921060471,
      "grad_norm": 20.75,
      "learning_rate": 1.787662645502704e-05,
      "loss": 1.0087,
      "num_input_tokens_seen": 49293688,
      "step": 84970
    },
    {
      "epoch": 12.65638963360143,
      "grad_norm": 9.5625,
      "learning_rate": 1.78735117924808e-05,
      "loss": 1.2059,
      "num_input_tokens_seen": 49296760,
      "step": 84975
    },
    {
      "epoch": 12.657134346142389,
      "grad_norm": 11.5625,
      "learning_rate": 1.7870397250329325e-05,
      "loss": 0.9346,
      "num_input_tokens_seen": 49299736,
      "step": 84980
    },
    {
      "epoch": 12.657879058683347,
      "grad_norm": 7.46875,
      "learning_rate": 1.7867282828625253e-05,
      "loss": 0.9739,
      "num_input_tokens_seen": 49302488,
      "step": 84985
    },
    {
      "epoch": 12.658623771224308,
      "grad_norm": 14.6875,
      "learning_rate": 1.786416852742119e-05,
      "loss": 1.1767,
      "num_input_tokens_seen": 49305720,
      "step": 84990
    },
    {
      "epoch": 12.659368483765267,
      "grad_norm": 7.375,
      "learning_rate": 1.786105434676973e-05,
      "loss": 1.0741,
      "num_input_tokens_seen": 49308504,
      "step": 84995
    },
    {
      "epoch": 12.660113196306225,
      "grad_norm": 10.3125,
      "learning_rate": 1.785794028672352e-05,
      "loss": 1.4466,
      "num_input_tokens_seen": 49311768,
      "step": 85000
    },
    {
      "epoch": 12.660857908847184,
      "grad_norm": 8.125,
      "learning_rate": 1.785482634733514e-05,
      "loss": 1.1127,
      "num_input_tokens_seen": 49314584,
      "step": 85005
    },
    {
      "epoch": 12.661602621388145,
      "grad_norm": 5.15625,
      "learning_rate": 1.785171252865721e-05,
      "loss": 1.3439,
      "num_input_tokens_seen": 49317336,
      "step": 85010
    },
    {
      "epoch": 12.662347333929103,
      "grad_norm": 5.28125,
      "learning_rate": 1.7848598830742323e-05,
      "loss": 1.2395,
      "num_input_tokens_seen": 49320184,
      "step": 85015
    },
    {
      "epoch": 12.663092046470062,
      "grad_norm": 9.9375,
      "learning_rate": 1.78454852536431e-05,
      "loss": 1.273,
      "num_input_tokens_seen": 49323096,
      "step": 85020
    },
    {
      "epoch": 12.663836759011021,
      "grad_norm": 15.125,
      "learning_rate": 1.784237179741213e-05,
      "loss": 0.9386,
      "num_input_tokens_seen": 49325880,
      "step": 85025
    },
    {
      "epoch": 12.664581471551982,
      "grad_norm": 5.75,
      "learning_rate": 1.7839258462102015e-05,
      "loss": 0.9577,
      "num_input_tokens_seen": 49328888,
      "step": 85030
    },
    {
      "epoch": 12.66532618409294,
      "grad_norm": 6.59375,
      "learning_rate": 1.783614524776535e-05,
      "loss": 0.8315,
      "num_input_tokens_seen": 49331960,
      "step": 85035
    },
    {
      "epoch": 12.666070896633899,
      "grad_norm": 10.125,
      "learning_rate": 1.783303215445473e-05,
      "loss": 0.986,
      "num_input_tokens_seen": 49334744,
      "step": 85040
    },
    {
      "epoch": 12.666815609174858,
      "grad_norm": 20.25,
      "learning_rate": 1.7829919182222752e-05,
      "loss": 1.5283,
      "num_input_tokens_seen": 49337528,
      "step": 85045
    },
    {
      "epoch": 12.667560321715818,
      "grad_norm": 12.1875,
      "learning_rate": 1.7826806331121987e-05,
      "loss": 1.0634,
      "num_input_tokens_seen": 49340472,
      "step": 85050
    },
    {
      "epoch": 12.668305034256777,
      "grad_norm": 8.375,
      "learning_rate": 1.7823693601205054e-05,
      "loss": 1.0805,
      "num_input_tokens_seen": 49343512,
      "step": 85055
    },
    {
      "epoch": 12.669049746797736,
      "grad_norm": 13.875,
      "learning_rate": 1.782058099252451e-05,
      "loss": 0.9062,
      "num_input_tokens_seen": 49346456,
      "step": 85060
    },
    {
      "epoch": 12.669794459338695,
      "grad_norm": 4.5,
      "learning_rate": 1.7817468505132966e-05,
      "loss": 0.7536,
      "num_input_tokens_seen": 49349272,
      "step": 85065
    },
    {
      "epoch": 12.670539171879655,
      "grad_norm": 6.8125,
      "learning_rate": 1.7814356139082993e-05,
      "loss": 1.0647,
      "num_input_tokens_seen": 49352024,
      "step": 85070
    },
    {
      "epoch": 12.671283884420614,
      "grad_norm": 12.5625,
      "learning_rate": 1.781124389442716e-05,
      "loss": 1.1606,
      "num_input_tokens_seen": 49354648,
      "step": 85075
    },
    {
      "epoch": 12.672028596961573,
      "grad_norm": 12.1875,
      "learning_rate": 1.7808131771218065e-05,
      "loss": 1.3968,
      "num_input_tokens_seen": 49357592,
      "step": 85080
    },
    {
      "epoch": 12.672773309502531,
      "grad_norm": 16.5,
      "learning_rate": 1.7805019769508262e-05,
      "loss": 1.1057,
      "num_input_tokens_seen": 49360408,
      "step": 85085
    },
    {
      "epoch": 12.673518022043492,
      "grad_norm": 13.0625,
      "learning_rate": 1.7801907889350346e-05,
      "loss": 1.2709,
      "num_input_tokens_seen": 49363096,
      "step": 85090
    },
    {
      "epoch": 12.67426273458445,
      "grad_norm": 10.125,
      "learning_rate": 1.7798796130796875e-05,
      "loss": 0.9939,
      "num_input_tokens_seen": 49366552,
      "step": 85095
    },
    {
      "epoch": 12.67500744712541,
      "grad_norm": 11.8125,
      "learning_rate": 1.779568449390043e-05,
      "loss": 1.0207,
      "num_input_tokens_seen": 49369240,
      "step": 85100
    },
    {
      "epoch": 12.675752159666368,
      "grad_norm": 11.25,
      "learning_rate": 1.7792572978713567e-05,
      "loss": 1.453,
      "num_input_tokens_seen": 49372152,
      "step": 85105
    },
    {
      "epoch": 12.676496872207329,
      "grad_norm": 12.25,
      "learning_rate": 1.778946158528887e-05,
      "loss": 0.8873,
      "num_input_tokens_seen": 49375096,
      "step": 85110
    },
    {
      "epoch": 12.677241584748288,
      "grad_norm": 12.4375,
      "learning_rate": 1.7786350313678885e-05,
      "loss": 1.161,
      "num_input_tokens_seen": 49377752,
      "step": 85115
    },
    {
      "epoch": 12.677986297289246,
      "grad_norm": 14.5,
      "learning_rate": 1.7783239163936172e-05,
      "loss": 1.1235,
      "num_input_tokens_seen": 49380600,
      "step": 85120
    },
    {
      "epoch": 12.678731009830205,
      "grad_norm": 8.9375,
      "learning_rate": 1.7780128136113305e-05,
      "loss": 1.195,
      "num_input_tokens_seen": 49383512,
      "step": 85125
    },
    {
      "epoch": 12.679475722371166,
      "grad_norm": 15.8125,
      "learning_rate": 1.777701723026283e-05,
      "loss": 1.115,
      "num_input_tokens_seen": 49386360,
      "step": 85130
    },
    {
      "epoch": 12.680220434912124,
      "grad_norm": 8.4375,
      "learning_rate": 1.7773906446437316e-05,
      "loss": 1.0009,
      "num_input_tokens_seen": 49389304,
      "step": 85135
    },
    {
      "epoch": 12.680965147453083,
      "grad_norm": 11.5,
      "learning_rate": 1.7770795784689302e-05,
      "loss": 1.3491,
      "num_input_tokens_seen": 49392024,
      "step": 85140
    },
    {
      "epoch": 12.681709859994042,
      "grad_norm": 13.375,
      "learning_rate": 1.7767685245071353e-05,
      "loss": 0.9166,
      "num_input_tokens_seen": 49394936,
      "step": 85145
    },
    {
      "epoch": 12.682454572535,
      "grad_norm": 4.5625,
      "learning_rate": 1.776457482763601e-05,
      "loss": 0.8605,
      "num_input_tokens_seen": 49397784,
      "step": 85150
    },
    {
      "epoch": 12.683199285075961,
      "grad_norm": 6.6875,
      "learning_rate": 1.776146453243581e-05,
      "loss": 0.9373,
      "num_input_tokens_seen": 49400504,
      "step": 85155
    },
    {
      "epoch": 12.68394399761692,
      "grad_norm": 14.75,
      "learning_rate": 1.7758354359523328e-05,
      "loss": 0.8692,
      "num_input_tokens_seen": 49403064,
      "step": 85160
    },
    {
      "epoch": 12.684688710157879,
      "grad_norm": 8.375,
      "learning_rate": 1.775524430895107e-05,
      "loss": 1.2384,
      "num_input_tokens_seen": 49406104,
      "step": 85165
    },
    {
      "epoch": 12.685433422698837,
      "grad_norm": 5.34375,
      "learning_rate": 1.775213438077161e-05,
      "loss": 1.2207,
      "num_input_tokens_seen": 49408984,
      "step": 85170
    },
    {
      "epoch": 12.686178135239798,
      "grad_norm": 10.9375,
      "learning_rate": 1.7749024575037466e-05,
      "loss": 0.9379,
      "num_input_tokens_seen": 49411736,
      "step": 85175
    },
    {
      "epoch": 12.686922847780757,
      "grad_norm": 11.625,
      "learning_rate": 1.774591489180119e-05,
      "loss": 0.9008,
      "num_input_tokens_seen": 49414456,
      "step": 85180
    },
    {
      "epoch": 12.687667560321715,
      "grad_norm": 12.375,
      "learning_rate": 1.77428053311153e-05,
      "loss": 1.0877,
      "num_input_tokens_seen": 49417272,
      "step": 85185
    },
    {
      "epoch": 12.688412272862674,
      "grad_norm": 8.5,
      "learning_rate": 1.7739695893032344e-05,
      "loss": 1.2044,
      "num_input_tokens_seen": 49420312,
      "step": 85190
    },
    {
      "epoch": 12.689156985403635,
      "grad_norm": 7.1875,
      "learning_rate": 1.7736586577604846e-05,
      "loss": 0.8925,
      "num_input_tokens_seen": 49423256,
      "step": 85195
    },
    {
      "epoch": 12.689901697944594,
      "grad_norm": 6.34375,
      "learning_rate": 1.7733477384885333e-05,
      "loss": 0.8114,
      "num_input_tokens_seen": 49426040,
      "step": 85200
    },
    {
      "epoch": 12.690646410485552,
      "grad_norm": 21.25,
      "learning_rate": 1.7730368314926336e-05,
      "loss": 1.705,
      "num_input_tokens_seen": 49428920,
      "step": 85205
    },
    {
      "epoch": 12.691391123026511,
      "grad_norm": 6.3125,
      "learning_rate": 1.7727259367780374e-05,
      "loss": 0.8888,
      "num_input_tokens_seen": 49431864,
      "step": 85210
    },
    {
      "epoch": 12.692135835567472,
      "grad_norm": 9.0625,
      "learning_rate": 1.772415054349998e-05,
      "loss": 1.254,
      "num_input_tokens_seen": 49434776,
      "step": 85215
    },
    {
      "epoch": 12.69288054810843,
      "grad_norm": 8.6875,
      "learning_rate": 1.772104184213766e-05,
      "loss": 1.237,
      "num_input_tokens_seen": 49437592,
      "step": 85220
    },
    {
      "epoch": 12.69362526064939,
      "grad_norm": 13.125,
      "learning_rate": 1.7717933263745947e-05,
      "loss": 1.1163,
      "num_input_tokens_seen": 49440248,
      "step": 85225
    },
    {
      "epoch": 12.694369973190348,
      "grad_norm": 7.34375,
      "learning_rate": 1.7714824808377346e-05,
      "loss": 1.0332,
      "num_input_tokens_seen": 49443288,
      "step": 85230
    },
    {
      "epoch": 12.695114685731308,
      "grad_norm": 11.75,
      "learning_rate": 1.7711716476084384e-05,
      "loss": 1.3238,
      "num_input_tokens_seen": 49446360,
      "step": 85235
    },
    {
      "epoch": 12.695859398272267,
      "grad_norm": 8.0625,
      "learning_rate": 1.770860826691956e-05,
      "loss": 0.9551,
      "num_input_tokens_seen": 49449208,
      "step": 85240
    },
    {
      "epoch": 12.696604110813226,
      "grad_norm": 13.875,
      "learning_rate": 1.770550018093538e-05,
      "loss": 1.0614,
      "num_input_tokens_seen": 49451992,
      "step": 85245
    },
    {
      "epoch": 12.697348823354185,
      "grad_norm": 15.125,
      "learning_rate": 1.770239221818437e-05,
      "loss": 1.0683,
      "num_input_tokens_seen": 49454808,
      "step": 85250
    },
    {
      "epoch": 12.698093535895145,
      "grad_norm": 9.6875,
      "learning_rate": 1.7699284378719017e-05,
      "loss": 1.0826,
      "num_input_tokens_seen": 49457720,
      "step": 85255
    },
    {
      "epoch": 12.698838248436104,
      "grad_norm": 4.75,
      "learning_rate": 1.7696176662591844e-05,
      "loss": 0.8337,
      "num_input_tokens_seen": 49460664,
      "step": 85260
    },
    {
      "epoch": 12.699582960977063,
      "grad_norm": 8.875,
      "learning_rate": 1.7693069069855343e-05,
      "loss": 0.9457,
      "num_input_tokens_seen": 49463192,
      "step": 85265
    },
    {
      "epoch": 12.700327673518021,
      "grad_norm": 12.125,
      "learning_rate": 1.7689961600562014e-05,
      "loss": 1.3022,
      "num_input_tokens_seen": 49466008,
      "step": 85270
    },
    {
      "epoch": 12.701072386058982,
      "grad_norm": 6.84375,
      "learning_rate": 1.7686854254764355e-05,
      "loss": 0.9063,
      "num_input_tokens_seen": 49468664,
      "step": 85275
    },
    {
      "epoch": 12.70181709859994,
      "grad_norm": 8.5625,
      "learning_rate": 1.768374703251485e-05,
      "loss": 0.9021,
      "num_input_tokens_seen": 49471512,
      "step": 85280
    },
    {
      "epoch": 12.7025618111409,
      "grad_norm": 8.75,
      "learning_rate": 1.768063993386601e-05,
      "loss": 0.9912,
      "num_input_tokens_seen": 49474264,
      "step": 85285
    },
    {
      "epoch": 12.703306523681858,
      "grad_norm": 3.578125,
      "learning_rate": 1.7677532958870315e-05,
      "loss": 0.8281,
      "num_input_tokens_seen": 49477144,
      "step": 85290
    },
    {
      "epoch": 12.704051236222817,
      "grad_norm": 10.625,
      "learning_rate": 1.7674426107580268e-05,
      "loss": 1.0266,
      "num_input_tokens_seen": 49479896,
      "step": 85295
    },
    {
      "epoch": 12.704795948763778,
      "grad_norm": 7.75,
      "learning_rate": 1.767131938004834e-05,
      "loss": 0.873,
      "num_input_tokens_seen": 49482616,
      "step": 85300
    },
    {
      "epoch": 12.705540661304736,
      "grad_norm": 7.40625,
      "learning_rate": 1.7668212776327025e-05,
      "loss": 0.9043,
      "num_input_tokens_seen": 49485176,
      "step": 85305
    },
    {
      "epoch": 12.706285373845695,
      "grad_norm": 12.125,
      "learning_rate": 1.7665106296468793e-05,
      "loss": 1.1242,
      "num_input_tokens_seen": 49488056,
      "step": 85310
    },
    {
      "epoch": 12.707030086386656,
      "grad_norm": 8.5625,
      "learning_rate": 1.766199994052615e-05,
      "loss": 1.1011,
      "num_input_tokens_seen": 49491000,
      "step": 85315
    },
    {
      "epoch": 12.707774798927614,
      "grad_norm": 9.625,
      "learning_rate": 1.7658893708551557e-05,
      "loss": 0.9006,
      "num_input_tokens_seen": 49494072,
      "step": 85320
    },
    {
      "epoch": 12.708519511468573,
      "grad_norm": 6.1875,
      "learning_rate": 1.7655787600597486e-05,
      "loss": 0.9216,
      "num_input_tokens_seen": 49497016,
      "step": 85325
    },
    {
      "epoch": 12.709264224009532,
      "grad_norm": 8.0625,
      "learning_rate": 1.7652681616716428e-05,
      "loss": 1.0571,
      "num_input_tokens_seen": 49499896,
      "step": 85330
    },
    {
      "epoch": 12.71000893655049,
      "grad_norm": 17.625,
      "learning_rate": 1.764957575696084e-05,
      "loss": 0.957,
      "num_input_tokens_seen": 49502648,
      "step": 85335
    },
    {
      "epoch": 12.710753649091451,
      "grad_norm": 5.4375,
      "learning_rate": 1.7646470021383204e-05,
      "loss": 0.8263,
      "num_input_tokens_seen": 49506616,
      "step": 85340
    },
    {
      "epoch": 12.71149836163241,
      "grad_norm": 8.9375,
      "learning_rate": 1.7643364410035974e-05,
      "loss": 0.8089,
      "num_input_tokens_seen": 49509528,
      "step": 85345
    },
    {
      "epoch": 12.712243074173369,
      "grad_norm": 10.125,
      "learning_rate": 1.7640258922971636e-05,
      "loss": 0.9673,
      "num_input_tokens_seen": 49512376,
      "step": 85350
    },
    {
      "epoch": 12.712987786714327,
      "grad_norm": 10.625,
      "learning_rate": 1.763715356024264e-05,
      "loss": 1.1987,
      "num_input_tokens_seen": 49515032,
      "step": 85355
    },
    {
      "epoch": 12.713732499255288,
      "grad_norm": 7.375,
      "learning_rate": 1.7634048321901447e-05,
      "loss": 0.9317,
      "num_input_tokens_seen": 49517912,
      "step": 85360
    },
    {
      "epoch": 12.714477211796247,
      "grad_norm": 12.375,
      "learning_rate": 1.7630943208000526e-05,
      "loss": 0.8489,
      "num_input_tokens_seen": 49520728,
      "step": 85365
    },
    {
      "epoch": 12.715221924337206,
      "grad_norm": 10.625,
      "learning_rate": 1.762783821859233e-05,
      "loss": 1.2144,
      "num_input_tokens_seen": 49523800,
      "step": 85370
    },
    {
      "epoch": 12.715966636878164,
      "grad_norm": 7.8125,
      "learning_rate": 1.762473335372932e-05,
      "loss": 1.2025,
      "num_input_tokens_seen": 49526712,
      "step": 85375
    },
    {
      "epoch": 12.716711349419125,
      "grad_norm": 8.125,
      "learning_rate": 1.7621628613463928e-05,
      "loss": 0.9496,
      "num_input_tokens_seen": 49529592,
      "step": 85380
    },
    {
      "epoch": 12.717456061960084,
      "grad_norm": 10.375,
      "learning_rate": 1.7618523997848634e-05,
      "loss": 0.8873,
      "num_input_tokens_seen": 49532312,
      "step": 85385
    },
    {
      "epoch": 12.718200774501042,
      "grad_norm": 4.9375,
      "learning_rate": 1.7615419506935866e-05,
      "loss": 1.1621,
      "num_input_tokens_seen": 49534936,
      "step": 85390
    },
    {
      "epoch": 12.718945487042001,
      "grad_norm": 10.5,
      "learning_rate": 1.761231514077809e-05,
      "loss": 0.9059,
      "num_input_tokens_seen": 49537784,
      "step": 85395
    },
    {
      "epoch": 12.719690199582962,
      "grad_norm": 7.8125,
      "learning_rate": 1.760921089942774e-05,
      "loss": 1.0473,
      "num_input_tokens_seen": 49540600,
      "step": 85400
    },
    {
      "epoch": 12.72043491212392,
      "grad_norm": 10.0625,
      "learning_rate": 1.7606106782937256e-05,
      "loss": 1.0261,
      "num_input_tokens_seen": 49543640,
      "step": 85405
    },
    {
      "epoch": 12.72117962466488,
      "grad_norm": 10.4375,
      "learning_rate": 1.760300279135909e-05,
      "loss": 1.1241,
      "num_input_tokens_seen": 49546392,
      "step": 85410
    },
    {
      "epoch": 12.721924337205838,
      "grad_norm": 10.1875,
      "learning_rate": 1.759989892474566e-05,
      "loss": 0.8905,
      "num_input_tokens_seen": 49549144,
      "step": 85415
    },
    {
      "epoch": 12.722669049746798,
      "grad_norm": 16.25,
      "learning_rate": 1.7596795183149428e-05,
      "loss": 0.9779,
      "num_input_tokens_seen": 49551992,
      "step": 85420
    },
    {
      "epoch": 12.723413762287757,
      "grad_norm": 4.34375,
      "learning_rate": 1.759369156662281e-05,
      "loss": 1.1143,
      "num_input_tokens_seen": 49554584,
      "step": 85425
    },
    {
      "epoch": 12.724158474828716,
      "grad_norm": 14.5,
      "learning_rate": 1.759058807521825e-05,
      "loss": 1.1784,
      "num_input_tokens_seen": 49557400,
      "step": 85430
    },
    {
      "epoch": 12.724903187369675,
      "grad_norm": 6.75,
      "learning_rate": 1.7587484708988176e-05,
      "loss": 1.1639,
      "num_input_tokens_seen": 49560024,
      "step": 85435
    },
    {
      "epoch": 12.725647899910635,
      "grad_norm": 7.09375,
      "learning_rate": 1.758438146798501e-05,
      "loss": 0.9284,
      "num_input_tokens_seen": 49562776,
      "step": 85440
    },
    {
      "epoch": 12.726392612451594,
      "grad_norm": 14.1875,
      "learning_rate": 1.758127835226119e-05,
      "loss": 1.0271,
      "num_input_tokens_seen": 49565528,
      "step": 85445
    },
    {
      "epoch": 12.727137324992553,
      "grad_norm": 8.25,
      "learning_rate": 1.757817536186912e-05,
      "loss": 1.1179,
      "num_input_tokens_seen": 49568696,
      "step": 85450
    },
    {
      "epoch": 12.727882037533512,
      "grad_norm": 10.0,
      "learning_rate": 1.7575072496861243e-05,
      "loss": 1.2732,
      "num_input_tokens_seen": 49571960,
      "step": 85455
    },
    {
      "epoch": 12.728626750074472,
      "grad_norm": 18.875,
      "learning_rate": 1.757196975728996e-05,
      "loss": 1.3055,
      "num_input_tokens_seen": 49574840,
      "step": 85460
    },
    {
      "epoch": 12.72937146261543,
      "grad_norm": 13.8125,
      "learning_rate": 1.7568867143207708e-05,
      "loss": 1.4199,
      "num_input_tokens_seen": 49577624,
      "step": 85465
    },
    {
      "epoch": 12.73011617515639,
      "grad_norm": 6.75,
      "learning_rate": 1.7565764654666888e-05,
      "loss": 0.9681,
      "num_input_tokens_seen": 49580568,
      "step": 85470
    },
    {
      "epoch": 12.730860887697348,
      "grad_norm": 15.6875,
      "learning_rate": 1.756266229171993e-05,
      "loss": 1.1442,
      "num_input_tokens_seen": 49583416,
      "step": 85475
    },
    {
      "epoch": 12.731605600238307,
      "grad_norm": 9.375,
      "learning_rate": 1.7559560054419225e-05,
      "loss": 0.821,
      "num_input_tokens_seen": 49586168,
      "step": 85480
    },
    {
      "epoch": 12.732350312779268,
      "grad_norm": 11.4375,
      "learning_rate": 1.7556457942817184e-05,
      "loss": 1.2727,
      "num_input_tokens_seen": 49589304,
      "step": 85485
    },
    {
      "epoch": 12.733095025320226,
      "grad_norm": 10.0,
      "learning_rate": 1.7553355956966227e-05,
      "loss": 1.1085,
      "num_input_tokens_seen": 49592376,
      "step": 85490
    },
    {
      "epoch": 12.733839737861185,
      "grad_norm": 14.625,
      "learning_rate": 1.7550254096918748e-05,
      "loss": 1.2711,
      "num_input_tokens_seen": 49595288,
      "step": 85495
    },
    {
      "epoch": 12.734584450402146,
      "grad_norm": 8.4375,
      "learning_rate": 1.754715236272716e-05,
      "loss": 1.14,
      "num_input_tokens_seen": 49598296,
      "step": 85500
    },
    {
      "epoch": 12.735329162943104,
      "grad_norm": 7.4375,
      "learning_rate": 1.7544050754443857e-05,
      "loss": 1.2231,
      "num_input_tokens_seen": 49601176,
      "step": 85505
    },
    {
      "epoch": 12.736073875484063,
      "grad_norm": 11.9375,
      "learning_rate": 1.7540949272121244e-05,
      "loss": 1.12,
      "num_input_tokens_seen": 49604280,
      "step": 85510
    },
    {
      "epoch": 12.736818588025022,
      "grad_norm": 8.3125,
      "learning_rate": 1.753784791581171e-05,
      "loss": 0.82,
      "num_input_tokens_seen": 49607416,
      "step": 85515
    },
    {
      "epoch": 12.73756330056598,
      "grad_norm": 6.03125,
      "learning_rate": 1.753474668556764e-05,
      "loss": 1.1064,
      "num_input_tokens_seen": 49610520,
      "step": 85520
    },
    {
      "epoch": 12.738308013106941,
      "grad_norm": 12.75,
      "learning_rate": 1.7531645581441447e-05,
      "loss": 1.2675,
      "num_input_tokens_seen": 49613272,
      "step": 85525
    },
    {
      "epoch": 12.7390527256479,
      "grad_norm": 8.5625,
      "learning_rate": 1.7528544603485507e-05,
      "loss": 0.8659,
      "num_input_tokens_seen": 49616312,
      "step": 85530
    },
    {
      "epoch": 12.739797438188859,
      "grad_norm": 12.5625,
      "learning_rate": 1.7525443751752218e-05,
      "loss": 0.8758,
      "num_input_tokens_seen": 49619224,
      "step": 85535
    },
    {
      "epoch": 12.740542150729818,
      "grad_norm": 9.125,
      "learning_rate": 1.7522343026293953e-05,
      "loss": 1.0778,
      "num_input_tokens_seen": 49622360,
      "step": 85540
    },
    {
      "epoch": 12.741286863270778,
      "grad_norm": 7.96875,
      "learning_rate": 1.751924242716311e-05,
      "loss": 0.9794,
      "num_input_tokens_seen": 49625304,
      "step": 85545
    },
    {
      "epoch": 12.742031575811737,
      "grad_norm": 11.75,
      "learning_rate": 1.751614195441205e-05,
      "loss": 1.1994,
      "num_input_tokens_seen": 49628120,
      "step": 85550
    },
    {
      "epoch": 12.742776288352696,
      "grad_norm": 6.90625,
      "learning_rate": 1.7513041608093185e-05,
      "loss": 1.2139,
      "num_input_tokens_seen": 49630936,
      "step": 85555
    },
    {
      "epoch": 12.743521000893654,
      "grad_norm": 9.5,
      "learning_rate": 1.7509941388258865e-05,
      "loss": 0.8801,
      "num_input_tokens_seen": 49633880,
      "step": 85560
    },
    {
      "epoch": 12.744265713434615,
      "grad_norm": 8.375,
      "learning_rate": 1.750684129496147e-05,
      "loss": 0.9046,
      "num_input_tokens_seen": 49636568,
      "step": 85565
    },
    {
      "epoch": 12.745010425975574,
      "grad_norm": 5.125,
      "learning_rate": 1.7503741328253377e-05,
      "loss": 0.7593,
      "num_input_tokens_seen": 49639096,
      "step": 85570
    },
    {
      "epoch": 12.745755138516532,
      "grad_norm": 8.75,
      "learning_rate": 1.7500641488186946e-05,
      "loss": 0.8293,
      "num_input_tokens_seen": 49641912,
      "step": 85575
    },
    {
      "epoch": 12.746499851057491,
      "grad_norm": 8.125,
      "learning_rate": 1.7497541774814568e-05,
      "loss": 1.0838,
      "num_input_tokens_seen": 49644760,
      "step": 85580
    },
    {
      "epoch": 12.747244563598452,
      "grad_norm": 10.9375,
      "learning_rate": 1.7494442188188588e-05,
      "loss": 1.1582,
      "num_input_tokens_seen": 49647800,
      "step": 85585
    },
    {
      "epoch": 12.74798927613941,
      "grad_norm": 6.1875,
      "learning_rate": 1.7491342728361383e-05,
      "loss": 0.9302,
      "num_input_tokens_seen": 49650648,
      "step": 85590
    },
    {
      "epoch": 12.74873398868037,
      "grad_norm": 10.5,
      "learning_rate": 1.7488243395385317e-05,
      "loss": 0.8789,
      "num_input_tokens_seen": 49653688,
      "step": 85595
    },
    {
      "epoch": 12.749478701221328,
      "grad_norm": 7.46875,
      "learning_rate": 1.7485144189312735e-05,
      "loss": 1.0168,
      "num_input_tokens_seen": 49656728,
      "step": 85600
    },
    {
      "epoch": 12.750223413762289,
      "grad_norm": 7.8125,
      "learning_rate": 1.7482045110196014e-05,
      "loss": 1.0099,
      "num_input_tokens_seen": 49659480,
      "step": 85605
    },
    {
      "epoch": 12.750968126303247,
      "grad_norm": 13.4375,
      "learning_rate": 1.7478946158087484e-05,
      "loss": 1.3482,
      "num_input_tokens_seen": 49662552,
      "step": 85610
    },
    {
      "epoch": 12.751712838844206,
      "grad_norm": 10.125,
      "learning_rate": 1.747584733303953e-05,
      "loss": 1.2105,
      "num_input_tokens_seen": 49665336,
      "step": 85615
    },
    {
      "epoch": 12.752457551385165,
      "grad_norm": 14.3125,
      "learning_rate": 1.7472748635104475e-05,
      "loss": 1.0993,
      "num_input_tokens_seen": 49667928,
      "step": 85620
    },
    {
      "epoch": 12.753202263926125,
      "grad_norm": 10.25,
      "learning_rate": 1.7469650064334693e-05,
      "loss": 1.2414,
      "num_input_tokens_seen": 49670712,
      "step": 85625
    },
    {
      "epoch": 12.753946976467084,
      "grad_norm": 11.1875,
      "learning_rate": 1.746655162078251e-05,
      "loss": 1.19,
      "num_input_tokens_seen": 49673656,
      "step": 85630
    },
    {
      "epoch": 12.754691689008043,
      "grad_norm": 13.25,
      "learning_rate": 1.7463453304500292e-05,
      "loss": 1.1514,
      "num_input_tokens_seen": 49676504,
      "step": 85635
    },
    {
      "epoch": 12.755436401549002,
      "grad_norm": 8.75,
      "learning_rate": 1.7460355115540366e-05,
      "loss": 0.9489,
      "num_input_tokens_seen": 49679768,
      "step": 85640
    },
    {
      "epoch": 12.756181114089962,
      "grad_norm": 8.5,
      "learning_rate": 1.7457257053955073e-05,
      "loss": 1.0126,
      "num_input_tokens_seen": 49682680,
      "step": 85645
    },
    {
      "epoch": 12.756925826630921,
      "grad_norm": 7.5,
      "learning_rate": 1.7454159119796758e-05,
      "loss": 1.0471,
      "num_input_tokens_seen": 49685624,
      "step": 85650
    },
    {
      "epoch": 12.75767053917188,
      "grad_norm": 10.5625,
      "learning_rate": 1.745106131311775e-05,
      "loss": 1.2333,
      "num_input_tokens_seen": 49688888,
      "step": 85655
    },
    {
      "epoch": 12.758415251712838,
      "grad_norm": 7.90625,
      "learning_rate": 1.74479636339704e-05,
      "loss": 1.1775,
      "num_input_tokens_seen": 49691960,
      "step": 85660
    },
    {
      "epoch": 12.759159964253797,
      "grad_norm": 8.4375,
      "learning_rate": 1.7444866082407026e-05,
      "loss": 0.942,
      "num_input_tokens_seen": 49694776,
      "step": 85665
    },
    {
      "epoch": 12.759904676794758,
      "grad_norm": 13.1875,
      "learning_rate": 1.744176865847996e-05,
      "loss": 1.2765,
      "num_input_tokens_seen": 49697816,
      "step": 85670
    },
    {
      "epoch": 12.760649389335716,
      "grad_norm": 14.875,
      "learning_rate": 1.743867136224153e-05,
      "loss": 1.2924,
      "num_input_tokens_seen": 49700536,
      "step": 85675
    },
    {
      "epoch": 12.761394101876675,
      "grad_norm": 8.4375,
      "learning_rate": 1.743557419374406e-05,
      "loss": 0.8005,
      "num_input_tokens_seen": 49703352,
      "step": 85680
    },
    {
      "epoch": 12.762138814417634,
      "grad_norm": 9.0,
      "learning_rate": 1.7432477153039877e-05,
      "loss": 1.3228,
      "num_input_tokens_seen": 49706072,
      "step": 85685
    },
    {
      "epoch": 12.762883526958595,
      "grad_norm": 8.125,
      "learning_rate": 1.7429380240181296e-05,
      "loss": 1.0284,
      "num_input_tokens_seen": 49708824,
      "step": 85690
    },
    {
      "epoch": 12.763628239499553,
      "grad_norm": 11.5625,
      "learning_rate": 1.7426283455220652e-05,
      "loss": 0.9253,
      "num_input_tokens_seen": 49711544,
      "step": 85695
    },
    {
      "epoch": 12.764372952040512,
      "grad_norm": 16.875,
      "learning_rate": 1.7423186798210244e-05,
      "loss": 1.0504,
      "num_input_tokens_seen": 49714872,
      "step": 85700
    },
    {
      "epoch": 12.76511766458147,
      "grad_norm": 11.9375,
      "learning_rate": 1.74200902692024e-05,
      "loss": 0.7213,
      "num_input_tokens_seen": 49717688,
      "step": 85705
    },
    {
      "epoch": 12.765862377122431,
      "grad_norm": 4.875,
      "learning_rate": 1.7416993868249422e-05,
      "loss": 0.9479,
      "num_input_tokens_seen": 49720728,
      "step": 85710
    },
    {
      "epoch": 12.76660708966339,
      "grad_norm": 18.25,
      "learning_rate": 1.7413897595403627e-05,
      "loss": 1.1563,
      "num_input_tokens_seen": 49723544,
      "step": 85715
    },
    {
      "epoch": 12.767351802204349,
      "grad_norm": 9.875,
      "learning_rate": 1.741080145071733e-05,
      "loss": 1.082,
      "num_input_tokens_seen": 49726392,
      "step": 85720
    },
    {
      "epoch": 12.768096514745308,
      "grad_norm": 7.40625,
      "learning_rate": 1.740770543424281e-05,
      "loss": 0.9962,
      "num_input_tokens_seen": 49729272,
      "step": 85725
    },
    {
      "epoch": 12.768841227286268,
      "grad_norm": 9.375,
      "learning_rate": 1.7404609546032407e-05,
      "loss": 1.3138,
      "num_input_tokens_seen": 49732312,
      "step": 85730
    },
    {
      "epoch": 12.769585939827227,
      "grad_norm": 12.4375,
      "learning_rate": 1.7401513786138402e-05,
      "loss": 1.4484,
      "num_input_tokens_seen": 49734840,
      "step": 85735
    },
    {
      "epoch": 12.770330652368186,
      "grad_norm": 10.3125,
      "learning_rate": 1.73984181546131e-05,
      "loss": 1.1581,
      "num_input_tokens_seen": 49737976,
      "step": 85740
    },
    {
      "epoch": 12.771075364909144,
      "grad_norm": 8.4375,
      "learning_rate": 1.7395322651508788e-05,
      "loss": 1.3047,
      "num_input_tokens_seen": 49740888,
      "step": 85745
    },
    {
      "epoch": 12.771820077450105,
      "grad_norm": 12.5625,
      "learning_rate": 1.739222727687778e-05,
      "loss": 0.8626,
      "num_input_tokens_seen": 49743672,
      "step": 85750
    },
    {
      "epoch": 12.772564789991064,
      "grad_norm": 11.75,
      "learning_rate": 1.7389132030772365e-05,
      "loss": 1.2814,
      "num_input_tokens_seen": 49746456,
      "step": 85755
    },
    {
      "epoch": 12.773309502532022,
      "grad_norm": 7.21875,
      "learning_rate": 1.7386036913244812e-05,
      "loss": 0.9411,
      "num_input_tokens_seen": 49749304,
      "step": 85760
    },
    {
      "epoch": 12.774054215072981,
      "grad_norm": 6.09375,
      "learning_rate": 1.7382941924347443e-05,
      "loss": 1.1414,
      "num_input_tokens_seen": 49752216,
      "step": 85765
    },
    {
      "epoch": 12.774798927613942,
      "grad_norm": 12.1875,
      "learning_rate": 1.737984706413252e-05,
      "loss": 0.8393,
      "num_input_tokens_seen": 49755096,
      "step": 85770
    },
    {
      "epoch": 12.7755436401549,
      "grad_norm": 10.0,
      "learning_rate": 1.737675233265234e-05,
      "loss": 1.2867,
      "num_input_tokens_seen": 49758072,
      "step": 85775
    },
    {
      "epoch": 12.77628835269586,
      "grad_norm": 9.625,
      "learning_rate": 1.7373657729959176e-05,
      "loss": 1.5143,
      "num_input_tokens_seen": 49761272,
      "step": 85780
    },
    {
      "epoch": 12.777033065236818,
      "grad_norm": 9.8125,
      "learning_rate": 1.7370563256105322e-05,
      "loss": 1.1657,
      "num_input_tokens_seen": 49763992,
      "step": 85785
    },
    {
      "epoch": 12.777777777777779,
      "grad_norm": 13.625,
      "learning_rate": 1.7367468911143036e-05,
      "loss": 0.9852,
      "num_input_tokens_seen": 49766648,
      "step": 85790
    },
    {
      "epoch": 12.778522490318737,
      "grad_norm": 10.875,
      "learning_rate": 1.7364374695124613e-05,
      "loss": 1.0671,
      "num_input_tokens_seen": 49769496,
      "step": 85795
    },
    {
      "epoch": 12.779267202859696,
      "grad_norm": 6.34375,
      "learning_rate": 1.7361280608102325e-05,
      "loss": 0.8966,
      "num_input_tokens_seen": 49772408,
      "step": 85800
    },
    {
      "epoch": 12.780011915400655,
      "grad_norm": 10.8125,
      "learning_rate": 1.7358186650128427e-05,
      "loss": 1.2487,
      "num_input_tokens_seen": 49775096,
      "step": 85805
    },
    {
      "epoch": 12.780756627941614,
      "grad_norm": 20.125,
      "learning_rate": 1.7355092821255208e-05,
      "loss": 1.3419,
      "num_input_tokens_seen": 49777848,
      "step": 85810
    },
    {
      "epoch": 12.781501340482574,
      "grad_norm": 18.375,
      "learning_rate": 1.7351999121534913e-05,
      "loss": 0.9778,
      "num_input_tokens_seen": 49780760,
      "step": 85815
    },
    {
      "epoch": 12.782246053023533,
      "grad_norm": 6.03125,
      "learning_rate": 1.734890555101983e-05,
      "loss": 1.107,
      "num_input_tokens_seen": 49783480,
      "step": 85820
    },
    {
      "epoch": 12.782990765564492,
      "grad_norm": 11.9375,
      "learning_rate": 1.7345812109762204e-05,
      "loss": 0.9893,
      "num_input_tokens_seen": 49786520,
      "step": 85825
    },
    {
      "epoch": 12.783735478105452,
      "grad_norm": 10.125,
      "learning_rate": 1.734271879781431e-05,
      "loss": 1.0785,
      "num_input_tokens_seen": 49789400,
      "step": 85830
    },
    {
      "epoch": 12.784480190646411,
      "grad_norm": 11.3125,
      "learning_rate": 1.7339625615228396e-05,
      "loss": 1.0866,
      "num_input_tokens_seen": 49792760,
      "step": 85835
    },
    {
      "epoch": 12.78522490318737,
      "grad_norm": 12.75,
      "learning_rate": 1.733653256205673e-05,
      "loss": 1.1458,
      "num_input_tokens_seen": 49795672,
      "step": 85840
    },
    {
      "epoch": 12.785969615728328,
      "grad_norm": 8.4375,
      "learning_rate": 1.733343963835155e-05,
      "loss": 0.9485,
      "num_input_tokens_seen": 49798968,
      "step": 85845
    },
    {
      "epoch": 12.786714328269287,
      "grad_norm": 13.375,
      "learning_rate": 1.733034684416511e-05,
      "loss": 1.3273,
      "num_input_tokens_seen": 49801848,
      "step": 85850
    },
    {
      "epoch": 12.787459040810248,
      "grad_norm": 14.875,
      "learning_rate": 1.7327254179549674e-05,
      "loss": 0.938,
      "num_input_tokens_seen": 49804888,
      "step": 85855
    },
    {
      "epoch": 12.788203753351207,
      "grad_norm": 7.4375,
      "learning_rate": 1.7324161644557472e-05,
      "loss": 1.0303,
      "num_input_tokens_seen": 49808024,
      "step": 85860
    },
    {
      "epoch": 12.788948465892165,
      "grad_norm": 21.875,
      "learning_rate": 1.7321069239240767e-05,
      "loss": 1.5719,
      "num_input_tokens_seen": 49810840,
      "step": 85865
    },
    {
      "epoch": 12.789693178433124,
      "grad_norm": 12.9375,
      "learning_rate": 1.7317976963651783e-05,
      "loss": 1.0205,
      "num_input_tokens_seen": 49814040,
      "step": 85870
    },
    {
      "epoch": 12.790437890974085,
      "grad_norm": 10.25,
      "learning_rate": 1.731488481784278e-05,
      "loss": 1.3055,
      "num_input_tokens_seen": 49817112,
      "step": 85875
    },
    {
      "epoch": 12.791182603515043,
      "grad_norm": 9.5,
      "learning_rate": 1.731179280186599e-05,
      "loss": 1.0778,
      "num_input_tokens_seen": 49820408,
      "step": 85880
    },
    {
      "epoch": 12.791927316056002,
      "grad_norm": 10.125,
      "learning_rate": 1.730870091577363e-05,
      "loss": 1.1785,
      "num_input_tokens_seen": 49823640,
      "step": 85885
    },
    {
      "epoch": 12.79267202859696,
      "grad_norm": 9.75,
      "learning_rate": 1.7305609159617965e-05,
      "loss": 0.8653,
      "num_input_tokens_seen": 49826616,
      "step": 85890
    },
    {
      "epoch": 12.793416741137921,
      "grad_norm": 12.4375,
      "learning_rate": 1.730251753345121e-05,
      "loss": 1.0205,
      "num_input_tokens_seen": 49829464,
      "step": 85895
    },
    {
      "epoch": 12.79416145367888,
      "grad_norm": 6.21875,
      "learning_rate": 1.72994260373256e-05,
      "loss": 1.1213,
      "num_input_tokens_seen": 49832248,
      "step": 85900
    },
    {
      "epoch": 12.794906166219839,
      "grad_norm": 8.125,
      "learning_rate": 1.729633467129335e-05,
      "loss": 1.3773,
      "num_input_tokens_seen": 49835256,
      "step": 85905
    },
    {
      "epoch": 12.795650878760798,
      "grad_norm": 9.8125,
      "learning_rate": 1.7293243435406705e-05,
      "loss": 1.1586,
      "num_input_tokens_seen": 49838264,
      "step": 85910
    },
    {
      "epoch": 12.796395591301758,
      "grad_norm": 11.0625,
      "learning_rate": 1.7290152329717873e-05,
      "loss": 1.0757,
      "num_input_tokens_seen": 49841176,
      "step": 85915
    },
    {
      "epoch": 12.797140303842717,
      "grad_norm": 9.4375,
      "learning_rate": 1.7287061354279088e-05,
      "loss": 0.7822,
      "num_input_tokens_seen": 49844184,
      "step": 85920
    },
    {
      "epoch": 12.797885016383676,
      "grad_norm": 8.4375,
      "learning_rate": 1.7283970509142567e-05,
      "loss": 1.0436,
      "num_input_tokens_seen": 49847000,
      "step": 85925
    },
    {
      "epoch": 12.798629728924634,
      "grad_norm": 10.0625,
      "learning_rate": 1.728087979436051e-05,
      "loss": 1.4989,
      "num_input_tokens_seen": 49850168,
      "step": 85930
    },
    {
      "epoch": 12.799374441465595,
      "grad_norm": 9.6875,
      "learning_rate": 1.7277789209985155e-05,
      "loss": 1.032,
      "num_input_tokens_seen": 49853368,
      "step": 85935
    },
    {
      "epoch": 12.800119154006554,
      "grad_norm": 9.375,
      "learning_rate": 1.727469875606869e-05,
      "loss": 0.9027,
      "num_input_tokens_seen": 49856536,
      "step": 85940
    },
    {
      "epoch": 12.800863866547513,
      "grad_norm": 17.125,
      "learning_rate": 1.727160843266335e-05,
      "loss": 1.1653,
      "num_input_tokens_seen": 49859896,
      "step": 85945
    },
    {
      "epoch": 12.801608579088471,
      "grad_norm": 6.3125,
      "learning_rate": 1.7268518239821318e-05,
      "loss": 1.1064,
      "num_input_tokens_seen": 49862776,
      "step": 85950
    },
    {
      "epoch": 12.802353291629432,
      "grad_norm": 14.25,
      "learning_rate": 1.7265428177594822e-05,
      "loss": 1.3035,
      "num_input_tokens_seen": 49865592,
      "step": 85955
    },
    {
      "epoch": 12.80309800417039,
      "grad_norm": 10.125,
      "learning_rate": 1.7262338246036057e-05,
      "loss": 1.1725,
      "num_input_tokens_seen": 49868536,
      "step": 85960
    },
    {
      "epoch": 12.80384271671135,
      "grad_norm": 11.375,
      "learning_rate": 1.7259248445197217e-05,
      "loss": 1.0281,
      "num_input_tokens_seen": 49871384,
      "step": 85965
    },
    {
      "epoch": 12.804587429252308,
      "grad_norm": 14.8125,
      "learning_rate": 1.7256158775130517e-05,
      "loss": 0.7498,
      "num_input_tokens_seen": 49874232,
      "step": 85970
    },
    {
      "epoch": 12.805332141793269,
      "grad_norm": 11.1875,
      "learning_rate": 1.725306923588813e-05,
      "loss": 0.9859,
      "num_input_tokens_seen": 49877144,
      "step": 85975
    },
    {
      "epoch": 12.806076854334227,
      "grad_norm": 14.0625,
      "learning_rate": 1.7249979827522274e-05,
      "loss": 1.3562,
      "num_input_tokens_seen": 49880344,
      "step": 85980
    },
    {
      "epoch": 12.806821566875186,
      "grad_norm": 14.75,
      "learning_rate": 1.7246890550085122e-05,
      "loss": 0.8601,
      "num_input_tokens_seen": 49883256,
      "step": 85985
    },
    {
      "epoch": 12.807566279416145,
      "grad_norm": 4.34375,
      "learning_rate": 1.7243801403628887e-05,
      "loss": 0.7931,
      "num_input_tokens_seen": 49886296,
      "step": 85990
    },
    {
      "epoch": 12.808310991957104,
      "grad_norm": 6.625,
      "learning_rate": 1.724071238820574e-05,
      "loss": 1.1644,
      "num_input_tokens_seen": 49889080,
      "step": 85995
    },
    {
      "epoch": 12.809055704498064,
      "grad_norm": 9.25,
      "learning_rate": 1.723762350386787e-05,
      "loss": 1.1556,
      "num_input_tokens_seen": 49892056,
      "step": 86000
    },
    {
      "epoch": 12.809800417039023,
      "grad_norm": 6.03125,
      "learning_rate": 1.7234534750667468e-05,
      "loss": 1.0532,
      "num_input_tokens_seen": 49894968,
      "step": 86005
    },
    {
      "epoch": 12.810545129579982,
      "grad_norm": 11.75,
      "learning_rate": 1.7231446128656693e-05,
      "loss": 1.0512,
      "num_input_tokens_seen": 49897784,
      "step": 86010
    },
    {
      "epoch": 12.811289842120942,
      "grad_norm": 26.375,
      "learning_rate": 1.7228357637887755e-05,
      "loss": 1.2664,
      "num_input_tokens_seen": 49900600,
      "step": 86015
    },
    {
      "epoch": 12.812034554661901,
      "grad_norm": 16.125,
      "learning_rate": 1.7225269278412802e-05,
      "loss": 1.1637,
      "num_input_tokens_seen": 49903288,
      "step": 86020
    },
    {
      "epoch": 12.81277926720286,
      "grad_norm": 11.1875,
      "learning_rate": 1.7222181050284037e-05,
      "loss": 1.3853,
      "num_input_tokens_seen": 49906168,
      "step": 86025
    },
    {
      "epoch": 12.813523979743819,
      "grad_norm": 11.0625,
      "learning_rate": 1.7219092953553612e-05,
      "loss": 0.8448,
      "num_input_tokens_seen": 49909272,
      "step": 86030
    },
    {
      "epoch": 12.814268692284777,
      "grad_norm": 8.6875,
      "learning_rate": 1.7216004988273706e-05,
      "loss": 1.0304,
      "num_input_tokens_seen": 49912088,
      "step": 86035
    },
    {
      "epoch": 12.815013404825738,
      "grad_norm": 14.0625,
      "learning_rate": 1.7212917154496488e-05,
      "loss": 0.8904,
      "num_input_tokens_seen": 49914936,
      "step": 86040
    },
    {
      "epoch": 12.815758117366697,
      "grad_norm": 12.0625,
      "learning_rate": 1.7209829452274108e-05,
      "loss": 1.253,
      "num_input_tokens_seen": 49917624,
      "step": 86045
    },
    {
      "epoch": 12.816502829907655,
      "grad_norm": 15.4375,
      "learning_rate": 1.720674188165875e-05,
      "loss": 1.6541,
      "num_input_tokens_seen": 49920312,
      "step": 86050
    },
    {
      "epoch": 12.817247542448614,
      "grad_norm": 8.25,
      "learning_rate": 1.720365444270256e-05,
      "loss": 0.8649,
      "num_input_tokens_seen": 49922968,
      "step": 86055
    },
    {
      "epoch": 12.817992254989575,
      "grad_norm": 8.875,
      "learning_rate": 1.720056713545771e-05,
      "loss": 1.0061,
      "num_input_tokens_seen": 49925912,
      "step": 86060
    },
    {
      "epoch": 12.818736967530533,
      "grad_norm": 10.6875,
      "learning_rate": 1.7197479959976353e-05,
      "loss": 1.1346,
      "num_input_tokens_seen": 49928760,
      "step": 86065
    },
    {
      "epoch": 12.819481680071492,
      "grad_norm": 14.1875,
      "learning_rate": 1.719439291631064e-05,
      "loss": 1.291,
      "num_input_tokens_seen": 49931672,
      "step": 86070
    },
    {
      "epoch": 12.820226392612451,
      "grad_norm": 10.125,
      "learning_rate": 1.7191306004512723e-05,
      "loss": 1.0909,
      "num_input_tokens_seen": 49934552,
      "step": 86075
    },
    {
      "epoch": 12.820971105153411,
      "grad_norm": 9.1875,
      "learning_rate": 1.7188219224634762e-05,
      "loss": 1.2739,
      "num_input_tokens_seen": 49937848,
      "step": 86080
    },
    {
      "epoch": 12.82171581769437,
      "grad_norm": 8.3125,
      "learning_rate": 1.7185132576728898e-05,
      "loss": 1.1092,
      "num_input_tokens_seen": 49940696,
      "step": 86085
    },
    {
      "epoch": 12.822460530235329,
      "grad_norm": 15.0,
      "learning_rate": 1.718204606084726e-05,
      "loss": 1.2085,
      "num_input_tokens_seen": 49943480,
      "step": 86090
    },
    {
      "epoch": 12.823205242776288,
      "grad_norm": 14.6875,
      "learning_rate": 1.717895967704202e-05,
      "loss": 1.2124,
      "num_input_tokens_seen": 49946136,
      "step": 86095
    },
    {
      "epoch": 12.823949955317248,
      "grad_norm": 6.125,
      "learning_rate": 1.7175873425365308e-05,
      "loss": 1.3267,
      "num_input_tokens_seen": 49949016,
      "step": 86100
    },
    {
      "epoch": 12.824694667858207,
      "grad_norm": 9.875,
      "learning_rate": 1.7172787305869266e-05,
      "loss": 1.04,
      "num_input_tokens_seen": 49951800,
      "step": 86105
    },
    {
      "epoch": 12.825439380399166,
      "grad_norm": 16.5,
      "learning_rate": 1.7169701318606014e-05,
      "loss": 0.8826,
      "num_input_tokens_seen": 49954520,
      "step": 86110
    },
    {
      "epoch": 12.826184092940125,
      "grad_norm": 11.375,
      "learning_rate": 1.7166615463627712e-05,
      "loss": 1.2817,
      "num_input_tokens_seen": 49957400,
      "step": 86115
    },
    {
      "epoch": 12.826928805481085,
      "grad_norm": 9.5625,
      "learning_rate": 1.716352974098648e-05,
      "loss": 1.2502,
      "num_input_tokens_seen": 49960568,
      "step": 86120
    },
    {
      "epoch": 12.827673518022044,
      "grad_norm": 7.78125,
      "learning_rate": 1.716044415073444e-05,
      "loss": 1.0051,
      "num_input_tokens_seen": 49963256,
      "step": 86125
    },
    {
      "epoch": 12.828418230563003,
      "grad_norm": 8.5625,
      "learning_rate": 1.7157358692923737e-05,
      "loss": 0.8989,
      "num_input_tokens_seen": 49966040,
      "step": 86130
    },
    {
      "epoch": 12.829162943103961,
      "grad_norm": 12.0625,
      "learning_rate": 1.7154273367606484e-05,
      "loss": 1.1599,
      "num_input_tokens_seen": 49968760,
      "step": 86135
    },
    {
      "epoch": 12.829907655644922,
      "grad_norm": 6.375,
      "learning_rate": 1.715118817483481e-05,
      "loss": 1.0306,
      "num_input_tokens_seen": 49971544,
      "step": 86140
    },
    {
      "epoch": 12.83065236818588,
      "grad_norm": 13.375,
      "learning_rate": 1.7148103114660825e-05,
      "loss": 0.9846,
      "num_input_tokens_seen": 49974232,
      "step": 86145
    },
    {
      "epoch": 12.83139708072684,
      "grad_norm": 9.5625,
      "learning_rate": 1.7145018187136668e-05,
      "loss": 0.9317,
      "num_input_tokens_seen": 49977016,
      "step": 86150
    },
    {
      "epoch": 12.832141793267798,
      "grad_norm": 6.28125,
      "learning_rate": 1.7141933392314436e-05,
      "loss": 0.9682,
      "num_input_tokens_seen": 49979928,
      "step": 86155
    },
    {
      "epoch": 12.832886505808759,
      "grad_norm": 10.1875,
      "learning_rate": 1.7138848730246264e-05,
      "loss": 0.9503,
      "num_input_tokens_seen": 49982904,
      "step": 86160
    },
    {
      "epoch": 12.833631218349717,
      "grad_norm": 10.5625,
      "learning_rate": 1.7135764200984253e-05,
      "loss": 1.2858,
      "num_input_tokens_seen": 49986008,
      "step": 86165
    },
    {
      "epoch": 12.834375930890676,
      "grad_norm": 9.125,
      "learning_rate": 1.7132679804580505e-05,
      "loss": 0.9948,
      "num_input_tokens_seen": 49988856,
      "step": 86170
    },
    {
      "epoch": 12.835120643431635,
      "grad_norm": 6.15625,
      "learning_rate": 1.7129595541087146e-05,
      "loss": 1.1436,
      "num_input_tokens_seen": 49991608,
      "step": 86175
    },
    {
      "epoch": 12.835865355972594,
      "grad_norm": 12.6875,
      "learning_rate": 1.7126511410556256e-05,
      "loss": 1.4335,
      "num_input_tokens_seen": 49994616,
      "step": 86180
    },
    {
      "epoch": 12.836610068513554,
      "grad_norm": 10.9375,
      "learning_rate": 1.7123427413039967e-05,
      "loss": 1.2406,
      "num_input_tokens_seen": 49997720,
      "step": 86185
    },
    {
      "epoch": 12.837354781054513,
      "grad_norm": 11.6875,
      "learning_rate": 1.712034354859036e-05,
      "loss": 1.0886,
      "num_input_tokens_seen": 50000440,
      "step": 86190
    },
    {
      "epoch": 12.838099493595472,
      "grad_norm": 12.375,
      "learning_rate": 1.711725981725954e-05,
      "loss": 1.0838,
      "num_input_tokens_seen": 50003480,
      "step": 86195
    },
    {
      "epoch": 12.83884420613643,
      "grad_norm": 9.5625,
      "learning_rate": 1.7114176219099607e-05,
      "loss": 1.1668,
      "num_input_tokens_seen": 50006200,
      "step": 86200
    },
    {
      "epoch": 12.839588918677391,
      "grad_norm": 7.71875,
      "learning_rate": 1.711109275416265e-05,
      "loss": 0.9044,
      "num_input_tokens_seen": 50009112,
      "step": 86205
    },
    {
      "epoch": 12.84033363121835,
      "grad_norm": 11.9375,
      "learning_rate": 1.7108009422500767e-05,
      "loss": 1.1603,
      "num_input_tokens_seen": 50012056,
      "step": 86210
    },
    {
      "epoch": 12.841078343759309,
      "grad_norm": 11.4375,
      "learning_rate": 1.7104926224166033e-05,
      "loss": 0.8616,
      "num_input_tokens_seen": 50014936,
      "step": 86215
    },
    {
      "epoch": 12.841823056300267,
      "grad_norm": 6.25,
      "learning_rate": 1.7101843159210556e-05,
      "loss": 0.7378,
      "num_input_tokens_seen": 50017912,
      "step": 86220
    },
    {
      "epoch": 12.842567768841228,
      "grad_norm": 19.875,
      "learning_rate": 1.709876022768641e-05,
      "loss": 1.134,
      "num_input_tokens_seen": 50020824,
      "step": 86225
    },
    {
      "epoch": 12.843312481382187,
      "grad_norm": 9.6875,
      "learning_rate": 1.7095677429645682e-05,
      "loss": 0.8811,
      "num_input_tokens_seen": 50024056,
      "step": 86230
    },
    {
      "epoch": 12.844057193923145,
      "grad_norm": 15.4375,
      "learning_rate": 1.709259476514044e-05,
      "loss": 1.3832,
      "num_input_tokens_seen": 50026968,
      "step": 86235
    },
    {
      "epoch": 12.844801906464104,
      "grad_norm": 4.0,
      "learning_rate": 1.7089512234222783e-05,
      "loss": 0.8208,
      "num_input_tokens_seen": 50030168,
      "step": 86240
    },
    {
      "epoch": 12.845546619005065,
      "grad_norm": 8.0,
      "learning_rate": 1.7086429836944777e-05,
      "loss": 0.9505,
      "num_input_tokens_seen": 50033144,
      "step": 86245
    },
    {
      "epoch": 12.846291331546023,
      "grad_norm": 14.3125,
      "learning_rate": 1.7083347573358484e-05,
      "loss": 0.9997,
      "num_input_tokens_seen": 50036056,
      "step": 86250
    },
    {
      "epoch": 12.847036044086982,
      "grad_norm": 16.375,
      "learning_rate": 1.7080265443516e-05,
      "loss": 1.2102,
      "num_input_tokens_seen": 50038968,
      "step": 86255
    },
    {
      "epoch": 12.847780756627941,
      "grad_norm": 7.75,
      "learning_rate": 1.7077183447469376e-05,
      "loss": 0.9845,
      "num_input_tokens_seen": 50041912,
      "step": 86260
    },
    {
      "epoch": 12.848525469168901,
      "grad_norm": 7.03125,
      "learning_rate": 1.7074101585270692e-05,
      "loss": 1.0265,
      "num_input_tokens_seen": 50044760,
      "step": 86265
    },
    {
      "epoch": 12.84927018170986,
      "grad_norm": 6.0625,
      "learning_rate": 1.7071019856971993e-05,
      "loss": 0.7345,
      "num_input_tokens_seen": 50047256,
      "step": 86270
    },
    {
      "epoch": 12.850014894250819,
      "grad_norm": 6.5,
      "learning_rate": 1.7067938262625364e-05,
      "loss": 0.9696,
      "num_input_tokens_seen": 50050104,
      "step": 86275
    },
    {
      "epoch": 12.850759606791778,
      "grad_norm": 7.28125,
      "learning_rate": 1.7064856802282865e-05,
      "loss": 1.0148,
      "num_input_tokens_seen": 50053176,
      "step": 86280
    },
    {
      "epoch": 12.851504319332738,
      "grad_norm": 11.4375,
      "learning_rate": 1.706177547599653e-05,
      "loss": 1.349,
      "num_input_tokens_seen": 50056088,
      "step": 86285
    },
    {
      "epoch": 12.852249031873697,
      "grad_norm": 13.625,
      "learning_rate": 1.7058694283818437e-05,
      "loss": 1.0175,
      "num_input_tokens_seen": 50059224,
      "step": 86290
    },
    {
      "epoch": 12.852993744414656,
      "grad_norm": 9.6875,
      "learning_rate": 1.705561322580063e-05,
      "loss": 1.1114,
      "num_input_tokens_seen": 50061944,
      "step": 86295
    },
    {
      "epoch": 12.853738456955615,
      "grad_norm": 15.3125,
      "learning_rate": 1.7052532301995168e-05,
      "loss": 1.2231,
      "num_input_tokens_seen": 50064824,
      "step": 86300
    },
    {
      "epoch": 12.854483169496575,
      "grad_norm": 7.96875,
      "learning_rate": 1.7049451512454085e-05,
      "loss": 0.9597,
      "num_input_tokens_seen": 50067768,
      "step": 86305
    },
    {
      "epoch": 12.855227882037534,
      "grad_norm": 19.375,
      "learning_rate": 1.704637085722945e-05,
      "loss": 1.456,
      "num_input_tokens_seen": 50070488,
      "step": 86310
    },
    {
      "epoch": 12.855972594578493,
      "grad_norm": 13.5625,
      "learning_rate": 1.7043290336373286e-05,
      "loss": 1.3042,
      "num_input_tokens_seen": 50073176,
      "step": 86315
    },
    {
      "epoch": 12.856717307119451,
      "grad_norm": 6.0,
      "learning_rate": 1.7040209949937653e-05,
      "loss": 1.3307,
      "num_input_tokens_seen": 50075928,
      "step": 86320
    },
    {
      "epoch": 12.857462019660412,
      "grad_norm": 10.3125,
      "learning_rate": 1.7037129697974585e-05,
      "loss": 0.9106,
      "num_input_tokens_seen": 50078840,
      "step": 86325
    },
    {
      "epoch": 12.85820673220137,
      "grad_norm": 14.1875,
      "learning_rate": 1.703404958053611e-05,
      "loss": 1.2071,
      "num_input_tokens_seen": 50082360,
      "step": 86330
    },
    {
      "epoch": 12.85895144474233,
      "grad_norm": 12.0,
      "learning_rate": 1.703096959767428e-05,
      "loss": 1.202,
      "num_input_tokens_seen": 50085176,
      "step": 86335
    },
    {
      "epoch": 12.859696157283288,
      "grad_norm": 24.125,
      "learning_rate": 1.7027889749441108e-05,
      "loss": 1.23,
      "num_input_tokens_seen": 50088184,
      "step": 86340
    },
    {
      "epoch": 12.860440869824249,
      "grad_norm": 21.5,
      "learning_rate": 1.702481003588864e-05,
      "loss": 1.3364,
      "num_input_tokens_seen": 50091128,
      "step": 86345
    },
    {
      "epoch": 12.861185582365207,
      "grad_norm": 23.5,
      "learning_rate": 1.7021730457068898e-05,
      "loss": 0.7653,
      "num_input_tokens_seen": 50094360,
      "step": 86350
    },
    {
      "epoch": 12.861930294906166,
      "grad_norm": 8.9375,
      "learning_rate": 1.701865101303392e-05,
      "loss": 1.2069,
      "num_input_tokens_seen": 50097400,
      "step": 86355
    },
    {
      "epoch": 12.862675007447125,
      "grad_norm": 9.5625,
      "learning_rate": 1.701557170383572e-05,
      "loss": 1.0814,
      "num_input_tokens_seen": 50100248,
      "step": 86360
    },
    {
      "epoch": 12.863419719988084,
      "grad_norm": 10.9375,
      "learning_rate": 1.7012492529526315e-05,
      "loss": 0.9657,
      "num_input_tokens_seen": 50103032,
      "step": 86365
    },
    {
      "epoch": 12.864164432529044,
      "grad_norm": 11.375,
      "learning_rate": 1.700941349015774e-05,
      "loss": 0.9222,
      "num_input_tokens_seen": 50106040,
      "step": 86370
    },
    {
      "epoch": 12.864909145070003,
      "grad_norm": 9.625,
      "learning_rate": 1.7006334585781986e-05,
      "loss": 0.9636,
      "num_input_tokens_seen": 50108920,
      "step": 86375
    },
    {
      "epoch": 12.865653857610962,
      "grad_norm": 10.3125,
      "learning_rate": 1.7003255816451098e-05,
      "loss": 1.1482,
      "num_input_tokens_seen": 50111960,
      "step": 86380
    },
    {
      "epoch": 12.86639857015192,
      "grad_norm": 4.71875,
      "learning_rate": 1.7000177182217066e-05,
      "loss": 1.0527,
      "num_input_tokens_seen": 50114648,
      "step": 86385
    },
    {
      "epoch": 12.867143282692881,
      "grad_norm": 9.75,
      "learning_rate": 1.6997098683131918e-05,
      "loss": 1.2175,
      "num_input_tokens_seen": 50117592,
      "step": 86390
    },
    {
      "epoch": 12.86788799523384,
      "grad_norm": 9.5625,
      "learning_rate": 1.699402031924765e-05,
      "loss": 0.8216,
      "num_input_tokens_seen": 50120472,
      "step": 86395
    },
    {
      "epoch": 12.868632707774799,
      "grad_norm": 6.34375,
      "learning_rate": 1.699094209061628e-05,
      "loss": 1.5152,
      "num_input_tokens_seen": 50123096,
      "step": 86400
    },
    {
      "epoch": 12.869377420315757,
      "grad_norm": 10.5,
      "learning_rate": 1.69878639972898e-05,
      "loss": 1.4643,
      "num_input_tokens_seen": 50125976,
      "step": 86405
    },
    {
      "epoch": 12.870122132856718,
      "grad_norm": 12.4375,
      "learning_rate": 1.6984786039320207e-05,
      "loss": 0.8605,
      "num_input_tokens_seen": 50128856,
      "step": 86410
    },
    {
      "epoch": 12.870866845397677,
      "grad_norm": 9.375,
      "learning_rate": 1.6981708216759515e-05,
      "loss": 1.0846,
      "num_input_tokens_seen": 50131704,
      "step": 86415
    },
    {
      "epoch": 12.871611557938635,
      "grad_norm": 12.25,
      "learning_rate": 1.6978630529659706e-05,
      "loss": 1.2151,
      "num_input_tokens_seen": 50134680,
      "step": 86420
    },
    {
      "epoch": 12.872356270479594,
      "grad_norm": 5.78125,
      "learning_rate": 1.697555297807279e-05,
      "loss": 0.97,
      "num_input_tokens_seen": 50137656,
      "step": 86425
    },
    {
      "epoch": 12.873100983020555,
      "grad_norm": 11.5625,
      "learning_rate": 1.6972475562050744e-05,
      "loss": 1.2152,
      "num_input_tokens_seen": 50140952,
      "step": 86430
    },
    {
      "epoch": 12.873845695561513,
      "grad_norm": 16.25,
      "learning_rate": 1.6969398281645572e-05,
      "loss": 1.1004,
      "num_input_tokens_seen": 50143640,
      "step": 86435
    },
    {
      "epoch": 12.874590408102472,
      "grad_norm": 5.625,
      "learning_rate": 1.696632113690924e-05,
      "loss": 0.9713,
      "num_input_tokens_seen": 50146456,
      "step": 86440
    },
    {
      "epoch": 12.875335120643431,
      "grad_norm": 8.9375,
      "learning_rate": 1.6963244127893763e-05,
      "loss": 1.1557,
      "num_input_tokens_seen": 50149336,
      "step": 86445
    },
    {
      "epoch": 12.876079833184392,
      "grad_norm": 9.5,
      "learning_rate": 1.6960167254651105e-05,
      "loss": 1.1441,
      "num_input_tokens_seen": 50151992,
      "step": 86450
    },
    {
      "epoch": 12.87682454572535,
      "grad_norm": 9.375,
      "learning_rate": 1.6957090517233242e-05,
      "loss": 0.8909,
      "num_input_tokens_seen": 50155128,
      "step": 86455
    },
    {
      "epoch": 12.877569258266309,
      "grad_norm": 6.75,
      "learning_rate": 1.6954013915692167e-05,
      "loss": 1.4249,
      "num_input_tokens_seen": 50158232,
      "step": 86460
    },
    {
      "epoch": 12.878313970807268,
      "grad_norm": 14.625,
      "learning_rate": 1.695093745007985e-05,
      "loss": 1.2695,
      "num_input_tokens_seen": 50161080,
      "step": 86465
    },
    {
      "epoch": 12.879058683348228,
      "grad_norm": 11.375,
      "learning_rate": 1.6947861120448262e-05,
      "loss": 1.2585,
      "num_input_tokens_seen": 50164376,
      "step": 86470
    },
    {
      "epoch": 12.879803395889187,
      "grad_norm": 10.4375,
      "learning_rate": 1.694478492684937e-05,
      "loss": 0.9811,
      "num_input_tokens_seen": 50167608,
      "step": 86475
    },
    {
      "epoch": 12.880548108430146,
      "grad_norm": 17.625,
      "learning_rate": 1.694170886933516e-05,
      "loss": 1.2348,
      "num_input_tokens_seen": 50170488,
      "step": 86480
    },
    {
      "epoch": 12.881292820971105,
      "grad_norm": 5.625,
      "learning_rate": 1.693863294795759e-05,
      "loss": 0.8274,
      "num_input_tokens_seen": 50173624,
      "step": 86485
    },
    {
      "epoch": 12.882037533512065,
      "grad_norm": 6.75,
      "learning_rate": 1.6935557162768612e-05,
      "loss": 1.0309,
      "num_input_tokens_seen": 50176792,
      "step": 86490
    },
    {
      "epoch": 12.882782246053024,
      "grad_norm": 14.375,
      "learning_rate": 1.693248151382021e-05,
      "loss": 1.341,
      "num_input_tokens_seen": 50179704,
      "step": 86495
    },
    {
      "epoch": 12.883526958593983,
      "grad_norm": 12.0,
      "learning_rate": 1.6929406001164325e-05,
      "loss": 0.9623,
      "num_input_tokens_seen": 50182744,
      "step": 86500
    },
    {
      "epoch": 12.884271671134941,
      "grad_norm": 4.96875,
      "learning_rate": 1.6926330624852932e-05,
      "loss": 0.7996,
      "num_input_tokens_seen": 50185528,
      "step": 86505
    },
    {
      "epoch": 12.8850163836759,
      "grad_norm": 9.6875,
      "learning_rate": 1.6923255384937963e-05,
      "loss": 1.064,
      "num_input_tokens_seen": 50188536,
      "step": 86510
    },
    {
      "epoch": 12.88576109621686,
      "grad_norm": 6.15625,
      "learning_rate": 1.69201802814714e-05,
      "loss": 1.0255,
      "num_input_tokens_seen": 50191224,
      "step": 86515
    },
    {
      "epoch": 12.88650580875782,
      "grad_norm": 12.5,
      "learning_rate": 1.691710531450517e-05,
      "loss": 1.1242,
      "num_input_tokens_seen": 50194072,
      "step": 86520
    },
    {
      "epoch": 12.887250521298778,
      "grad_norm": 6.65625,
      "learning_rate": 1.6914030484091235e-05,
      "loss": 1.1406,
      "num_input_tokens_seen": 50197144,
      "step": 86525
    },
    {
      "epoch": 12.887995233839739,
      "grad_norm": 4.65625,
      "learning_rate": 1.6910955790281538e-05,
      "loss": 1.056,
      "num_input_tokens_seen": 50199896,
      "step": 86530
    },
    {
      "epoch": 12.888739946380698,
      "grad_norm": 24.5,
      "learning_rate": 1.690788123312802e-05,
      "loss": 1.238,
      "num_input_tokens_seen": 50202616,
      "step": 86535
    },
    {
      "epoch": 12.889484658921656,
      "grad_norm": 5.28125,
      "learning_rate": 1.6904806812682628e-05,
      "loss": 0.9059,
      "num_input_tokens_seen": 50205432,
      "step": 86540
    },
    {
      "epoch": 12.890229371462615,
      "grad_norm": 11.125,
      "learning_rate": 1.6901732528997282e-05,
      "loss": 0.9291,
      "num_input_tokens_seen": 50208184,
      "step": 86545
    },
    {
      "epoch": 12.890974084003574,
      "grad_norm": 9.25,
      "learning_rate": 1.689865838212395e-05,
      "loss": 1.0455,
      "num_input_tokens_seen": 50211224,
      "step": 86550
    },
    {
      "epoch": 12.891718796544534,
      "grad_norm": 22.5,
      "learning_rate": 1.6895584372114544e-05,
      "loss": 1.409,
      "num_input_tokens_seen": 50214168,
      "step": 86555
    },
    {
      "epoch": 12.892463509085493,
      "grad_norm": 12.3125,
      "learning_rate": 1.689251049902101e-05,
      "loss": 0.9304,
      "num_input_tokens_seen": 50217080,
      "step": 86560
    },
    {
      "epoch": 12.893208221626452,
      "grad_norm": 8.9375,
      "learning_rate": 1.6889436762895267e-05,
      "loss": 1.1567,
      "num_input_tokens_seen": 50220344,
      "step": 86565
    },
    {
      "epoch": 12.89395293416741,
      "grad_norm": 10.8125,
      "learning_rate": 1.6886363163789243e-05,
      "loss": 0.9475,
      "num_input_tokens_seen": 50223352,
      "step": 86570
    },
    {
      "epoch": 12.894697646708371,
      "grad_norm": 13.125,
      "learning_rate": 1.6883289701754872e-05,
      "loss": 1.149,
      "num_input_tokens_seen": 50226136,
      "step": 86575
    },
    {
      "epoch": 12.89544235924933,
      "grad_norm": 5.65625,
      "learning_rate": 1.6880216376844066e-05,
      "loss": 1.011,
      "num_input_tokens_seen": 50229240,
      "step": 86580
    },
    {
      "epoch": 12.896187071790289,
      "grad_norm": 20.0,
      "learning_rate": 1.6877143189108758e-05,
      "loss": 0.8693,
      "num_input_tokens_seen": 50232152,
      "step": 86585
    },
    {
      "epoch": 12.896931784331247,
      "grad_norm": 10.5625,
      "learning_rate": 1.6874070138600855e-05,
      "loss": 1.0319,
      "num_input_tokens_seen": 50234968,
      "step": 86590
    },
    {
      "epoch": 12.897676496872208,
      "grad_norm": 12.1875,
      "learning_rate": 1.6870997225372286e-05,
      "loss": 1.2222,
      "num_input_tokens_seen": 50237880,
      "step": 86595
    },
    {
      "epoch": 12.898421209413167,
      "grad_norm": 9.0,
      "learning_rate": 1.686792444947494e-05,
      "loss": 1.6025,
      "num_input_tokens_seen": 50240696,
      "step": 86600
    },
    {
      "epoch": 12.899165921954125,
      "grad_norm": 10.1875,
      "learning_rate": 1.6864851810960763e-05,
      "loss": 1.3757,
      "num_input_tokens_seen": 50243704,
      "step": 86605
    },
    {
      "epoch": 12.899910634495084,
      "grad_norm": 20.75,
      "learning_rate": 1.6861779309881648e-05,
      "loss": 1.2348,
      "num_input_tokens_seen": 50247032,
      "step": 86610
    },
    {
      "epoch": 12.900655347036045,
      "grad_norm": 13.9375,
      "learning_rate": 1.6858706946289486e-05,
      "loss": 1.3507,
      "num_input_tokens_seen": 50249912,
      "step": 86615
    },
    {
      "epoch": 12.901400059577004,
      "grad_norm": 18.875,
      "learning_rate": 1.6855634720236206e-05,
      "loss": 1.3975,
      "num_input_tokens_seen": 50253176,
      "step": 86620
    },
    {
      "epoch": 12.902144772117962,
      "grad_norm": 9.3125,
      "learning_rate": 1.6852562631773694e-05,
      "loss": 1.0751,
      "num_input_tokens_seen": 50255896,
      "step": 86625
    },
    {
      "epoch": 12.902889484658921,
      "grad_norm": 6.0,
      "learning_rate": 1.684949068095386e-05,
      "loss": 0.9097,
      "num_input_tokens_seen": 50259160,
      "step": 86630
    },
    {
      "epoch": 12.903634197199882,
      "grad_norm": 7.15625,
      "learning_rate": 1.684641886782859e-05,
      "loss": 0.8958,
      "num_input_tokens_seen": 50262008,
      "step": 86635
    },
    {
      "epoch": 12.90437890974084,
      "grad_norm": 12.9375,
      "learning_rate": 1.6843347192449793e-05,
      "loss": 1.2346,
      "num_input_tokens_seen": 50264760,
      "step": 86640
    },
    {
      "epoch": 12.905123622281799,
      "grad_norm": 10.0,
      "learning_rate": 1.6840275654869358e-05,
      "loss": 1.2755,
      "num_input_tokens_seen": 50267768,
      "step": 86645
    },
    {
      "epoch": 12.905868334822758,
      "grad_norm": 14.8125,
      "learning_rate": 1.6837204255139164e-05,
      "loss": 1.3161,
      "num_input_tokens_seen": 50270744,
      "step": 86650
    },
    {
      "epoch": 12.906613047363718,
      "grad_norm": 7.375,
      "learning_rate": 1.6834132993311115e-05,
      "loss": 1.0739,
      "num_input_tokens_seen": 50273880,
      "step": 86655
    },
    {
      "epoch": 12.907357759904677,
      "grad_norm": 5.46875,
      "learning_rate": 1.6831061869437086e-05,
      "loss": 1.2302,
      "num_input_tokens_seen": 50276760,
      "step": 86660
    },
    {
      "epoch": 12.908102472445636,
      "grad_norm": 8.1875,
      "learning_rate": 1.6827990883568966e-05,
      "loss": 1.0137,
      "num_input_tokens_seen": 50279736,
      "step": 86665
    },
    {
      "epoch": 12.908847184986595,
      "grad_norm": 10.75,
      "learning_rate": 1.6824920035758628e-05,
      "loss": 0.7472,
      "num_input_tokens_seen": 50282392,
      "step": 86670
    },
    {
      "epoch": 12.909591897527555,
      "grad_norm": 11.9375,
      "learning_rate": 1.6821849326057963e-05,
      "loss": 1.3085,
      "num_input_tokens_seen": 50285016,
      "step": 86675
    },
    {
      "epoch": 12.910336610068514,
      "grad_norm": 10.25,
      "learning_rate": 1.681877875451884e-05,
      "loss": 1.0308,
      "num_input_tokens_seen": 50288088,
      "step": 86680
    },
    {
      "epoch": 12.911081322609473,
      "grad_norm": 10.9375,
      "learning_rate": 1.681570832119314e-05,
      "loss": 1.3057,
      "num_input_tokens_seen": 50291288,
      "step": 86685
    },
    {
      "epoch": 12.911826035150431,
      "grad_norm": 9.6875,
      "learning_rate": 1.6812638026132728e-05,
      "loss": 1.2831,
      "num_input_tokens_seen": 50294136,
      "step": 86690
    },
    {
      "epoch": 12.91257074769139,
      "grad_norm": 10.375,
      "learning_rate": 1.680956786938947e-05,
      "loss": 1.1252,
      "num_input_tokens_seen": 50297016,
      "step": 86695
    },
    {
      "epoch": 12.91331546023235,
      "grad_norm": 8.4375,
      "learning_rate": 1.6806497851015246e-05,
      "loss": 0.8582,
      "num_input_tokens_seen": 50299832,
      "step": 86700
    },
    {
      "epoch": 12.91406017277331,
      "grad_norm": 19.5,
      "learning_rate": 1.68034279710619e-05,
      "loss": 1.268,
      "num_input_tokens_seen": 50302712,
      "step": 86705
    },
    {
      "epoch": 12.914804885314268,
      "grad_norm": 15.0625,
      "learning_rate": 1.6800358229581326e-05,
      "loss": 1.1123,
      "num_input_tokens_seen": 50305624,
      "step": 86710
    },
    {
      "epoch": 12.915549597855229,
      "grad_norm": 8.3125,
      "learning_rate": 1.6797288626625345e-05,
      "loss": 0.8721,
      "num_input_tokens_seen": 50308600,
      "step": 86715
    },
    {
      "epoch": 12.916294310396188,
      "grad_norm": 7.15625,
      "learning_rate": 1.6794219162245855e-05,
      "loss": 0.9016,
      "num_input_tokens_seen": 50311512,
      "step": 86720
    },
    {
      "epoch": 12.917039022937146,
      "grad_norm": 18.125,
      "learning_rate": 1.679114983649469e-05,
      "loss": 1.1084,
      "num_input_tokens_seen": 50314424,
      "step": 86725
    },
    {
      "epoch": 12.917783735478105,
      "grad_norm": 21.875,
      "learning_rate": 1.6788080649423696e-05,
      "loss": 1.2398,
      "num_input_tokens_seen": 50317208,
      "step": 86730
    },
    {
      "epoch": 12.918528448019064,
      "grad_norm": 11.0625,
      "learning_rate": 1.678501160108474e-05,
      "loss": 1.2451,
      "num_input_tokens_seen": 50319896,
      "step": 86735
    },
    {
      "epoch": 12.919273160560024,
      "grad_norm": 13.1875,
      "learning_rate": 1.6781942691529656e-05,
      "loss": 1.0813,
      "num_input_tokens_seen": 50323288,
      "step": 86740
    },
    {
      "epoch": 12.920017873100983,
      "grad_norm": 9.625,
      "learning_rate": 1.6778873920810305e-05,
      "loss": 0.929,
      "num_input_tokens_seen": 50326104,
      "step": 86745
    },
    {
      "epoch": 12.920762585641942,
      "grad_norm": 8.9375,
      "learning_rate": 1.6775805288978517e-05,
      "loss": 1.5525,
      "num_input_tokens_seen": 50328824,
      "step": 86750
    },
    {
      "epoch": 12.9215072981829,
      "grad_norm": 12.625,
      "learning_rate": 1.6772736796086146e-05,
      "loss": 1.1244,
      "num_input_tokens_seen": 50331960,
      "step": 86755
    },
    {
      "epoch": 12.922252010723861,
      "grad_norm": 15.875,
      "learning_rate": 1.676966844218502e-05,
      "loss": 1.3446,
      "num_input_tokens_seen": 50335448,
      "step": 86760
    },
    {
      "epoch": 12.92299672326482,
      "grad_norm": 11.0625,
      "learning_rate": 1.676660022732699e-05,
      "loss": 0.7958,
      "num_input_tokens_seen": 50338648,
      "step": 86765
    },
    {
      "epoch": 12.923741435805779,
      "grad_norm": 9.4375,
      "learning_rate": 1.6763532151563878e-05,
      "loss": 1.3594,
      "num_input_tokens_seen": 50341400,
      "step": 86770
    },
    {
      "epoch": 12.924486148346737,
      "grad_norm": 9.0625,
      "learning_rate": 1.676046421494751e-05,
      "loss": 1.2504,
      "num_input_tokens_seen": 50344120,
      "step": 86775
    },
    {
      "epoch": 12.925230860887698,
      "grad_norm": 9.3125,
      "learning_rate": 1.6757396417529735e-05,
      "loss": 1.0368,
      "num_input_tokens_seen": 50347096,
      "step": 86780
    },
    {
      "epoch": 12.925975573428657,
      "grad_norm": 10.1875,
      "learning_rate": 1.675432875936236e-05,
      "loss": 0.9405,
      "num_input_tokens_seen": 50350104,
      "step": 86785
    },
    {
      "epoch": 12.926720285969616,
      "grad_norm": 5.96875,
      "learning_rate": 1.6751261240497228e-05,
      "loss": 0.8604,
      "num_input_tokens_seen": 50353080,
      "step": 86790
    },
    {
      "epoch": 12.927464998510574,
      "grad_norm": 13.375,
      "learning_rate": 1.6748193860986152e-05,
      "loss": 0.9671,
      "num_input_tokens_seen": 50355896,
      "step": 86795
    },
    {
      "epoch": 12.928209711051535,
      "grad_norm": 19.5,
      "learning_rate": 1.674512662088096e-05,
      "loss": 0.9581,
      "num_input_tokens_seen": 50358680,
      "step": 86800
    },
    {
      "epoch": 12.928954423592494,
      "grad_norm": 14.125,
      "learning_rate": 1.674205952023346e-05,
      "loss": 1.201,
      "num_input_tokens_seen": 50361944,
      "step": 86805
    },
    {
      "epoch": 12.929699136133452,
      "grad_norm": 10.6875,
      "learning_rate": 1.6738992559095462e-05,
      "loss": 0.7635,
      "num_input_tokens_seen": 50365080,
      "step": 86810
    },
    {
      "epoch": 12.930443848674411,
      "grad_norm": 12.375,
      "learning_rate": 1.67359257375188e-05,
      "loss": 1.1017,
      "num_input_tokens_seen": 50367864,
      "step": 86815
    },
    {
      "epoch": 12.931188561215372,
      "grad_norm": 6.28125,
      "learning_rate": 1.673285905555526e-05,
      "loss": 1.1945,
      "num_input_tokens_seen": 50370744,
      "step": 86820
    },
    {
      "epoch": 12.93193327375633,
      "grad_norm": 9.1875,
      "learning_rate": 1.6729792513256682e-05,
      "loss": 1.0163,
      "num_input_tokens_seen": 50373464,
      "step": 86825
    },
    {
      "epoch": 12.93267798629729,
      "grad_norm": 10.6875,
      "learning_rate": 1.672672611067484e-05,
      "loss": 0.8181,
      "num_input_tokens_seen": 50376376,
      "step": 86830
    },
    {
      "epoch": 12.933422698838248,
      "grad_norm": 15.8125,
      "learning_rate": 1.672365984786156e-05,
      "loss": 1.3723,
      "num_input_tokens_seen": 50379448,
      "step": 86835
    },
    {
      "epoch": 12.934167411379208,
      "grad_norm": 11.875,
      "learning_rate": 1.6720593724868626e-05,
      "loss": 1.0921,
      "num_input_tokens_seen": 50382392,
      "step": 86840
    },
    {
      "epoch": 12.934912123920167,
      "grad_norm": 10.875,
      "learning_rate": 1.6717527741747857e-05,
      "loss": 1.1564,
      "num_input_tokens_seen": 50385400,
      "step": 86845
    },
    {
      "epoch": 12.935656836461126,
      "grad_norm": 8.375,
      "learning_rate": 1.6714461898551037e-05,
      "loss": 1.1907,
      "num_input_tokens_seen": 50388472,
      "step": 86850
    },
    {
      "epoch": 12.936401549002085,
      "grad_norm": 6.6875,
      "learning_rate": 1.6711396195329955e-05,
      "loss": 1.0625,
      "num_input_tokens_seen": 50391000,
      "step": 86855
    },
    {
      "epoch": 12.937146261543045,
      "grad_norm": 10.75,
      "learning_rate": 1.670833063213642e-05,
      "loss": 0.9149,
      "num_input_tokens_seen": 50394200,
      "step": 86860
    },
    {
      "epoch": 12.937890974084004,
      "grad_norm": 6.9375,
      "learning_rate": 1.6705265209022204e-05,
      "loss": 1.1517,
      "num_input_tokens_seen": 50396888,
      "step": 86865
    },
    {
      "epoch": 12.938635686624963,
      "grad_norm": 17.75,
      "learning_rate": 1.6702199926039107e-05,
      "loss": 1.1034,
      "num_input_tokens_seen": 50399736,
      "step": 86870
    },
    {
      "epoch": 12.939380399165922,
      "grad_norm": 8.375,
      "learning_rate": 1.66991347832389e-05,
      "loss": 1.4227,
      "num_input_tokens_seen": 50403192,
      "step": 86875
    },
    {
      "epoch": 12.94012511170688,
      "grad_norm": 10.1875,
      "learning_rate": 1.669606978067338e-05,
      "loss": 1.4787,
      "num_input_tokens_seen": 50406232,
      "step": 86880
    },
    {
      "epoch": 12.94086982424784,
      "grad_norm": 5.75,
      "learning_rate": 1.669300491839433e-05,
      "loss": 1.1416,
      "num_input_tokens_seen": 50409400,
      "step": 86885
    },
    {
      "epoch": 12.9416145367888,
      "grad_norm": 16.875,
      "learning_rate": 1.6689940196453507e-05,
      "loss": 1.0033,
      "num_input_tokens_seen": 50412632,
      "step": 86890
    },
    {
      "epoch": 12.942359249329758,
      "grad_norm": 9.25,
      "learning_rate": 1.66868756149027e-05,
      "loss": 0.9622,
      "num_input_tokens_seen": 50415448,
      "step": 86895
    },
    {
      "epoch": 12.943103961870717,
      "grad_norm": 8.0625,
      "learning_rate": 1.668381117379368e-05,
      "loss": 1.1135,
      "num_input_tokens_seen": 50418488,
      "step": 86900
    },
    {
      "epoch": 12.943848674411678,
      "grad_norm": 9.25,
      "learning_rate": 1.6680746873178225e-05,
      "loss": 1.2426,
      "num_input_tokens_seen": 50421656,
      "step": 86905
    },
    {
      "epoch": 12.944593386952636,
      "grad_norm": 17.25,
      "learning_rate": 1.6677682713108082e-05,
      "loss": 1.0955,
      "num_input_tokens_seen": 50424760,
      "step": 86910
    },
    {
      "epoch": 12.945338099493595,
      "grad_norm": 9.9375,
      "learning_rate": 1.6674618693635047e-05,
      "loss": 1.148,
      "num_input_tokens_seen": 50427640,
      "step": 86915
    },
    {
      "epoch": 12.946082812034554,
      "grad_norm": 7.4375,
      "learning_rate": 1.6671554814810857e-05,
      "loss": 1.0793,
      "num_input_tokens_seen": 50430616,
      "step": 86920
    },
    {
      "epoch": 12.946827524575514,
      "grad_norm": 14.5,
      "learning_rate": 1.6668491076687294e-05,
      "loss": 0.917,
      "num_input_tokens_seen": 50433368,
      "step": 86925
    },
    {
      "epoch": 12.947572237116473,
      "grad_norm": 6.0,
      "learning_rate": 1.66654274793161e-05,
      "loss": 0.9768,
      "num_input_tokens_seen": 50436024,
      "step": 86930
    },
    {
      "epoch": 12.948316949657432,
      "grad_norm": 11.25,
      "learning_rate": 1.6662364022749035e-05,
      "loss": 1.1975,
      "num_input_tokens_seen": 50438904,
      "step": 86935
    },
    {
      "epoch": 12.94906166219839,
      "grad_norm": 21.75,
      "learning_rate": 1.6659300707037864e-05,
      "loss": 1.0977,
      "num_input_tokens_seen": 50441880,
      "step": 86940
    },
    {
      "epoch": 12.949806374739351,
      "grad_norm": 16.25,
      "learning_rate": 1.665623753223432e-05,
      "loss": 0.7893,
      "num_input_tokens_seen": 50445208,
      "step": 86945
    },
    {
      "epoch": 12.95055108728031,
      "grad_norm": 12.8125,
      "learning_rate": 1.6653174498390172e-05,
      "loss": 1.2661,
      "num_input_tokens_seen": 50448248,
      "step": 86950
    },
    {
      "epoch": 12.951295799821269,
      "grad_norm": 15.625,
      "learning_rate": 1.665011160555715e-05,
      "loss": 0.8248,
      "num_input_tokens_seen": 50451160,
      "step": 86955
    },
    {
      "epoch": 12.952040512362228,
      "grad_norm": 10.375,
      "learning_rate": 1.6647048853787018e-05,
      "loss": 0.7456,
      "num_input_tokens_seen": 50453912,
      "step": 86960
    },
    {
      "epoch": 12.952785224903188,
      "grad_norm": 5.84375,
      "learning_rate": 1.6643986243131497e-05,
      "loss": 0.8256,
      "num_input_tokens_seen": 50456696,
      "step": 86965
    },
    {
      "epoch": 12.953529937444147,
      "grad_norm": 7.4375,
      "learning_rate": 1.664092377364233e-05,
      "loss": 0.7467,
      "num_input_tokens_seen": 50459256,
      "step": 86970
    },
    {
      "epoch": 12.954274649985106,
      "grad_norm": 8.6875,
      "learning_rate": 1.663786144537127e-05,
      "loss": 1.069,
      "num_input_tokens_seen": 50461976,
      "step": 86975
    },
    {
      "epoch": 12.955019362526064,
      "grad_norm": 8.125,
      "learning_rate": 1.6634799258370036e-05,
      "loss": 0.8608,
      "num_input_tokens_seen": 50464856,
      "step": 86980
    },
    {
      "epoch": 12.955764075067025,
      "grad_norm": 10.0625,
      "learning_rate": 1.6631737212690373e-05,
      "loss": 1.0337,
      "num_input_tokens_seen": 50467704,
      "step": 86985
    },
    {
      "epoch": 12.956508787607984,
      "grad_norm": 11.375,
      "learning_rate": 1.6628675308384e-05,
      "loss": 1.1287,
      "num_input_tokens_seen": 50470232,
      "step": 86990
    },
    {
      "epoch": 12.957253500148942,
      "grad_norm": 18.125,
      "learning_rate": 1.6625613545502653e-05,
      "loss": 1.2902,
      "num_input_tokens_seen": 50473080,
      "step": 86995
    },
    {
      "epoch": 12.957998212689901,
      "grad_norm": 10.0,
      "learning_rate": 1.6622551924098046e-05,
      "loss": 1.1173,
      "num_input_tokens_seen": 50475896,
      "step": 87000
    },
    {
      "epoch": 12.958742925230862,
      "grad_norm": 12.125,
      "learning_rate": 1.6619490444221918e-05,
      "loss": 0.8067,
      "num_input_tokens_seen": 50478968,
      "step": 87005
    },
    {
      "epoch": 12.95948763777182,
      "grad_norm": 9.625,
      "learning_rate": 1.6616429105925978e-05,
      "loss": 1.0691,
      "num_input_tokens_seen": 50481816,
      "step": 87010
    },
    {
      "epoch": 12.96023235031278,
      "grad_norm": 4.875,
      "learning_rate": 1.6613367909261946e-05,
      "loss": 1.0325,
      "num_input_tokens_seen": 50484568,
      "step": 87015
    },
    {
      "epoch": 12.960977062853738,
      "grad_norm": 27.5,
      "learning_rate": 1.6610306854281542e-05,
      "loss": 1.334,
      "num_input_tokens_seen": 50487416,
      "step": 87020
    },
    {
      "epoch": 12.961721775394697,
      "grad_norm": 11.0,
      "learning_rate": 1.6607245941036476e-05,
      "loss": 1.1026,
      "num_input_tokens_seen": 50490456,
      "step": 87025
    },
    {
      "epoch": 12.962466487935657,
      "grad_norm": 12.9375,
      "learning_rate": 1.660418516957846e-05,
      "loss": 1.3668,
      "num_input_tokens_seen": 50493336,
      "step": 87030
    },
    {
      "epoch": 12.963211200476616,
      "grad_norm": 8.6875,
      "learning_rate": 1.66011245399592e-05,
      "loss": 1.3242,
      "num_input_tokens_seen": 50496408,
      "step": 87035
    },
    {
      "epoch": 12.963955913017575,
      "grad_norm": 8.1875,
      "learning_rate": 1.6598064052230407e-05,
      "loss": 1.0414,
      "num_input_tokens_seen": 50499128,
      "step": 87040
    },
    {
      "epoch": 12.964700625558535,
      "grad_norm": 8.6875,
      "learning_rate": 1.659500370644378e-05,
      "loss": 1.072,
      "num_input_tokens_seen": 50501816,
      "step": 87045
    },
    {
      "epoch": 12.965445338099494,
      "grad_norm": 11.9375,
      "learning_rate": 1.6591943502651025e-05,
      "loss": 1.2436,
      "num_input_tokens_seen": 50504984,
      "step": 87050
    },
    {
      "epoch": 12.966190050640453,
      "grad_norm": 5.15625,
      "learning_rate": 1.6588883440903847e-05,
      "loss": 0.8576,
      "num_input_tokens_seen": 50507672,
      "step": 87055
    },
    {
      "epoch": 12.966934763181412,
      "grad_norm": 7.65625,
      "learning_rate": 1.6585823521253924e-05,
      "loss": 1.073,
      "num_input_tokens_seen": 50510744,
      "step": 87060
    },
    {
      "epoch": 12.96767947572237,
      "grad_norm": 13.0,
      "learning_rate": 1.6582763743752965e-05,
      "loss": 1.114,
      "num_input_tokens_seen": 50513368,
      "step": 87065
    },
    {
      "epoch": 12.96842418826333,
      "grad_norm": 4.53125,
      "learning_rate": 1.6579704108452653e-05,
      "loss": 1.3133,
      "num_input_tokens_seen": 50516504,
      "step": 87070
    },
    {
      "epoch": 12.96916890080429,
      "grad_norm": 13.6875,
      "learning_rate": 1.6576644615404686e-05,
      "loss": 0.9733,
      "num_input_tokens_seen": 50519352,
      "step": 87075
    },
    {
      "epoch": 12.969913613345248,
      "grad_norm": 10.0625,
      "learning_rate": 1.657358526466074e-05,
      "loss": 0.9614,
      "num_input_tokens_seen": 50522104,
      "step": 87080
    },
    {
      "epoch": 12.970658325886207,
      "grad_norm": 11.5625,
      "learning_rate": 1.6570526056272516e-05,
      "loss": 1.4488,
      "num_input_tokens_seen": 50525080,
      "step": 87085
    },
    {
      "epoch": 12.971403038427168,
      "grad_norm": 8.25,
      "learning_rate": 1.656746699029169e-05,
      "loss": 1.3207,
      "num_input_tokens_seen": 50527736,
      "step": 87090
    },
    {
      "epoch": 12.972147750968126,
      "grad_norm": 8.25,
      "learning_rate": 1.6564408066769932e-05,
      "loss": 1.3161,
      "num_input_tokens_seen": 50530776,
      "step": 87095
    },
    {
      "epoch": 12.972892463509085,
      "grad_norm": 11.8125,
      "learning_rate": 1.6561349285758932e-05,
      "loss": 1.1993,
      "num_input_tokens_seen": 50533944,
      "step": 87100
    },
    {
      "epoch": 12.973637176050044,
      "grad_norm": 16.625,
      "learning_rate": 1.6558290647310347e-05,
      "loss": 1.045,
      "num_input_tokens_seen": 50536728,
      "step": 87105
    },
    {
      "epoch": 12.974381888591004,
      "grad_norm": 6.875,
      "learning_rate": 1.6555232151475872e-05,
      "loss": 0.9915,
      "num_input_tokens_seen": 50539512,
      "step": 87110
    },
    {
      "epoch": 12.975126601131963,
      "grad_norm": 10.0625,
      "learning_rate": 1.6552173798307157e-05,
      "loss": 1.0527,
      "num_input_tokens_seen": 50542392,
      "step": 87115
    },
    {
      "epoch": 12.975871313672922,
      "grad_norm": 11.1875,
      "learning_rate": 1.654911558785589e-05,
      "loss": 1.3156,
      "num_input_tokens_seen": 50545208,
      "step": 87120
    },
    {
      "epoch": 12.97661602621388,
      "grad_norm": 11.3125,
      "learning_rate": 1.654605752017372e-05,
      "loss": 0.9453,
      "num_input_tokens_seen": 50547992,
      "step": 87125
    },
    {
      "epoch": 12.977360738754841,
      "grad_norm": 15.9375,
      "learning_rate": 1.6542999595312324e-05,
      "loss": 1.1399,
      "num_input_tokens_seen": 50550808,
      "step": 87130
    },
    {
      "epoch": 12.9781054512958,
      "grad_norm": 8.75,
      "learning_rate": 1.6539941813323353e-05,
      "loss": 1.0004,
      "num_input_tokens_seen": 50553848,
      "step": 87135
    },
    {
      "epoch": 12.978850163836759,
      "grad_norm": 4.03125,
      "learning_rate": 1.653688417425846e-05,
      "loss": 0.7489,
      "num_input_tokens_seen": 50556664,
      "step": 87140
    },
    {
      "epoch": 12.979594876377718,
      "grad_norm": 10.6875,
      "learning_rate": 1.653382667816931e-05,
      "loss": 1.3738,
      "num_input_tokens_seen": 50559352,
      "step": 87145
    },
    {
      "epoch": 12.980339588918678,
      "grad_norm": 8.8125,
      "learning_rate": 1.653076932510755e-05,
      "loss": 1.1364,
      "num_input_tokens_seen": 50562072,
      "step": 87150
    },
    {
      "epoch": 12.981084301459637,
      "grad_norm": 10.4375,
      "learning_rate": 1.652771211512484e-05,
      "loss": 0.9674,
      "num_input_tokens_seen": 50564984,
      "step": 87155
    },
    {
      "epoch": 12.981829014000596,
      "grad_norm": 13.6875,
      "learning_rate": 1.652465504827282e-05,
      "loss": 1.2381,
      "num_input_tokens_seen": 50567864,
      "step": 87160
    },
    {
      "epoch": 12.982573726541554,
      "grad_norm": 8.1875,
      "learning_rate": 1.6521598124603143e-05,
      "loss": 1.0984,
      "num_input_tokens_seen": 50571032,
      "step": 87165
    },
    {
      "epoch": 12.983318439082515,
      "grad_norm": 12.3125,
      "learning_rate": 1.651854134416745e-05,
      "loss": 1.1311,
      "num_input_tokens_seen": 50573880,
      "step": 87170
    },
    {
      "epoch": 12.984063151623474,
      "grad_norm": 5.875,
      "learning_rate": 1.651548470701737e-05,
      "loss": 0.9733,
      "num_input_tokens_seen": 50576792,
      "step": 87175
    },
    {
      "epoch": 12.984807864164432,
      "grad_norm": 13.0,
      "learning_rate": 1.6512428213204564e-05,
      "loss": 0.8191,
      "num_input_tokens_seen": 50579768,
      "step": 87180
    },
    {
      "epoch": 12.985552576705391,
      "grad_norm": 11.6875,
      "learning_rate": 1.6509371862780644e-05,
      "loss": 0.9729,
      "num_input_tokens_seen": 50582712,
      "step": 87185
    },
    {
      "epoch": 12.986297289246352,
      "grad_norm": 12.75,
      "learning_rate": 1.650631565579727e-05,
      "loss": 1.0855,
      "num_input_tokens_seen": 50585784,
      "step": 87190
    },
    {
      "epoch": 12.98704200178731,
      "grad_norm": 18.0,
      "learning_rate": 1.6503259592306053e-05,
      "loss": 1.4783,
      "num_input_tokens_seen": 50588472,
      "step": 87195
    },
    {
      "epoch": 12.98778671432827,
      "grad_norm": 10.25,
      "learning_rate": 1.650020367235864e-05,
      "loss": 1.2633,
      "num_input_tokens_seen": 50591352,
      "step": 87200
    },
    {
      "epoch": 12.988531426869228,
      "grad_norm": 11.3125,
      "learning_rate": 1.649714789600663e-05,
      "loss": 1.5059,
      "num_input_tokens_seen": 50594200,
      "step": 87205
    },
    {
      "epoch": 12.989276139410187,
      "grad_norm": 9.9375,
      "learning_rate": 1.649409226330168e-05,
      "loss": 0.8816,
      "num_input_tokens_seen": 50596984,
      "step": 87210
    },
    {
      "epoch": 12.990020851951147,
      "grad_norm": 6.84375,
      "learning_rate": 1.6491036774295393e-05,
      "loss": 0.9068,
      "num_input_tokens_seen": 50600184,
      "step": 87215
    },
    {
      "epoch": 12.990765564492106,
      "grad_norm": 5.46875,
      "learning_rate": 1.6487981429039383e-05,
      "loss": 1.1608,
      "num_input_tokens_seen": 50603192,
      "step": 87220
    },
    {
      "epoch": 12.991510277033065,
      "grad_norm": 8.125,
      "learning_rate": 1.648492622758528e-05,
      "loss": 0.9838,
      "num_input_tokens_seen": 50606232,
      "step": 87225
    },
    {
      "epoch": 12.992254989574025,
      "grad_norm": 12.6875,
      "learning_rate": 1.6481871169984696e-05,
      "loss": 1.0671,
      "num_input_tokens_seen": 50608984,
      "step": 87230
    },
    {
      "epoch": 12.992999702114984,
      "grad_norm": 6.28125,
      "learning_rate": 1.647881625628924e-05,
      "loss": 0.9471,
      "num_input_tokens_seen": 50611928,
      "step": 87235
    },
    {
      "epoch": 12.993744414655943,
      "grad_norm": 7.40625,
      "learning_rate": 1.6475761486550516e-05,
      "loss": 1.1537,
      "num_input_tokens_seen": 50614968,
      "step": 87240
    },
    {
      "epoch": 12.994489127196902,
      "grad_norm": 8.625,
      "learning_rate": 1.6472706860820152e-05,
      "loss": 1.1575,
      "num_input_tokens_seen": 50617624,
      "step": 87245
    },
    {
      "epoch": 12.99523383973786,
      "grad_norm": 6.1875,
      "learning_rate": 1.6469652379149736e-05,
      "loss": 1.0563,
      "num_input_tokens_seen": 50620344,
      "step": 87250
    },
    {
      "epoch": 12.995978552278821,
      "grad_norm": 8.6875,
      "learning_rate": 1.6466598041590866e-05,
      "loss": 0.8622,
      "num_input_tokens_seen": 50623064,
      "step": 87255
    },
    {
      "epoch": 12.99672326481978,
      "grad_norm": 12.4375,
      "learning_rate": 1.646354384819515e-05,
      "loss": 1.1558,
      "num_input_tokens_seen": 50625912,
      "step": 87260
    },
    {
      "epoch": 12.997467977360738,
      "grad_norm": 13.5,
      "learning_rate": 1.6460489799014188e-05,
      "loss": 1.2897,
      "num_input_tokens_seen": 50628664,
      "step": 87265
    },
    {
      "epoch": 12.998212689901697,
      "grad_norm": 7.53125,
      "learning_rate": 1.6457435894099575e-05,
      "loss": 0.9963,
      "num_input_tokens_seen": 50631896,
      "step": 87270
    },
    {
      "epoch": 12.998957402442658,
      "grad_norm": 5.34375,
      "learning_rate": 1.645438213350289e-05,
      "loss": 0.8621,
      "num_input_tokens_seen": 50634872,
      "step": 87275
    },
    {
      "epoch": 12.999702114983616,
      "grad_norm": 15.5,
      "learning_rate": 1.645132851727574e-05,
      "loss": 0.9073,
      "num_input_tokens_seen": 50637720,
      "step": 87280
    },
    {
      "epoch": 13.0,
      "eval_loss": 1.0850152969360352,
      "eval_runtime": 43.7432,
      "eval_samples_per_second": 68.216,
      "eval_steps_per_second": 17.054,
      "num_input_tokens_seen": 50638456,
      "step": 87282
    },
    {
      "epoch": 13.000446827524575,
      "grad_norm": 12.25,
      "learning_rate": 1.6448275045469702e-05,
      "loss": 1.0228,
      "num_input_tokens_seen": 50640152,
      "step": 87285
    },
    {
      "epoch": 13.001191540065534,
      "grad_norm": 3.28125,
      "learning_rate": 1.6445221718136376e-05,
      "loss": 0.9394,
      "num_input_tokens_seen": 50643064,
      "step": 87290
    },
    {
      "epoch": 13.001936252606495,
      "grad_norm": 13.25,
      "learning_rate": 1.644216853532733e-05,
      "loss": 0.9199,
      "num_input_tokens_seen": 50645688,
      "step": 87295
    },
    {
      "epoch": 13.002680965147453,
      "grad_norm": 12.875,
      "learning_rate": 1.6439115497094137e-05,
      "loss": 0.9453,
      "num_input_tokens_seen": 50648728,
      "step": 87300
    },
    {
      "epoch": 13.003425677688412,
      "grad_norm": 14.25,
      "learning_rate": 1.64360626034884e-05,
      "loss": 1.3213,
      "num_input_tokens_seen": 50651576,
      "step": 87305
    },
    {
      "epoch": 13.00417039022937,
      "grad_norm": 12.625,
      "learning_rate": 1.6433009854561672e-05,
      "loss": 1.0752,
      "num_input_tokens_seen": 50654360,
      "step": 87310
    },
    {
      "epoch": 13.004915102770331,
      "grad_norm": 11.4375,
      "learning_rate": 1.6429957250365547e-05,
      "loss": 0.9375,
      "num_input_tokens_seen": 50657176,
      "step": 87315
    },
    {
      "epoch": 13.00565981531129,
      "grad_norm": 12.375,
      "learning_rate": 1.6426904790951575e-05,
      "loss": 1.1008,
      "num_input_tokens_seen": 50659992,
      "step": 87320
    },
    {
      "epoch": 13.006404527852249,
      "grad_norm": 7.0,
      "learning_rate": 1.642385247637134e-05,
      "loss": 0.9251,
      "num_input_tokens_seen": 50663064,
      "step": 87325
    },
    {
      "epoch": 13.007149240393208,
      "grad_norm": 15.6875,
      "learning_rate": 1.6420800306676397e-05,
      "loss": 1.1169,
      "num_input_tokens_seen": 50665816,
      "step": 87330
    },
    {
      "epoch": 13.007893952934168,
      "grad_norm": 15.25,
      "learning_rate": 1.641774828191831e-05,
      "loss": 1.2213,
      "num_input_tokens_seen": 50668792,
      "step": 87335
    },
    {
      "epoch": 13.008638665475127,
      "grad_norm": 17.375,
      "learning_rate": 1.641469640214865e-05,
      "loss": 1.1008,
      "num_input_tokens_seen": 50671576,
      "step": 87340
    },
    {
      "epoch": 13.009383378016086,
      "grad_norm": 7.59375,
      "learning_rate": 1.6411644667418958e-05,
      "loss": 0.9697,
      "num_input_tokens_seen": 50674616,
      "step": 87345
    },
    {
      "epoch": 13.010128090557044,
      "grad_norm": 15.0,
      "learning_rate": 1.6408593077780808e-05,
      "loss": 1.4337,
      "num_input_tokens_seen": 50677752,
      "step": 87350
    },
    {
      "epoch": 13.010872803098005,
      "grad_norm": 10.5625,
      "learning_rate": 1.6405541633285748e-05,
      "loss": 1.246,
      "num_input_tokens_seen": 50680664,
      "step": 87355
    },
    {
      "epoch": 13.011617515638964,
      "grad_norm": 7.75,
      "learning_rate": 1.6402490333985325e-05,
      "loss": 1.1183,
      "num_input_tokens_seen": 50683224,
      "step": 87360
    },
    {
      "epoch": 13.012362228179922,
      "grad_norm": 13.4375,
      "learning_rate": 1.6399439179931087e-05,
      "loss": 1.2607,
      "num_input_tokens_seen": 50686104,
      "step": 87365
    },
    {
      "epoch": 13.013106940720881,
      "grad_norm": 11.5,
      "learning_rate": 1.6396388171174586e-05,
      "loss": 0.851,
      "num_input_tokens_seen": 50689144,
      "step": 87370
    },
    {
      "epoch": 13.013851653261842,
      "grad_norm": 9.5,
      "learning_rate": 1.6393337307767364e-05,
      "loss": 1.1203,
      "num_input_tokens_seen": 50692184,
      "step": 87375
    },
    {
      "epoch": 13.0145963658028,
      "grad_norm": 6.0625,
      "learning_rate": 1.6390286589760957e-05,
      "loss": 0.8067,
      "num_input_tokens_seen": 50695000,
      "step": 87380
    },
    {
      "epoch": 13.01534107834376,
      "grad_norm": 13.125,
      "learning_rate": 1.6387236017206908e-05,
      "loss": 1.2213,
      "num_input_tokens_seen": 50698072,
      "step": 87385
    },
    {
      "epoch": 13.016085790884718,
      "grad_norm": 9.9375,
      "learning_rate": 1.6384185590156752e-05,
      "loss": 1.0341,
      "num_input_tokens_seen": 50700888,
      "step": 87390
    },
    {
      "epoch": 13.016830503425677,
      "grad_norm": 13.5625,
      "learning_rate": 1.6381135308662032e-05,
      "loss": 1.0074,
      "num_input_tokens_seen": 50703672,
      "step": 87395
    },
    {
      "epoch": 13.017575215966637,
      "grad_norm": 13.375,
      "learning_rate": 1.6378085172774258e-05,
      "loss": 1.0193,
      "num_input_tokens_seen": 50706776,
      "step": 87400
    },
    {
      "epoch": 13.018319928507596,
      "grad_norm": 7.25,
      "learning_rate": 1.6375035182544983e-05,
      "loss": 0.9112,
      "num_input_tokens_seen": 50709400,
      "step": 87405
    },
    {
      "epoch": 13.019064641048555,
      "grad_norm": 4.09375,
      "learning_rate": 1.637198533802572e-05,
      "loss": 0.9201,
      "num_input_tokens_seen": 50712248,
      "step": 87410
    },
    {
      "epoch": 13.019809353589514,
      "grad_norm": 9.375,
      "learning_rate": 1.636893563926799e-05,
      "loss": 1.1695,
      "num_input_tokens_seen": 50715000,
      "step": 87415
    },
    {
      "epoch": 13.020554066130474,
      "grad_norm": 9.5,
      "learning_rate": 1.6365886086323327e-05,
      "loss": 1.0945,
      "num_input_tokens_seen": 50717624,
      "step": 87420
    },
    {
      "epoch": 13.021298778671433,
      "grad_norm": 9.0625,
      "learning_rate": 1.636283667924324e-05,
      "loss": 1.0965,
      "num_input_tokens_seen": 50720472,
      "step": 87425
    },
    {
      "epoch": 13.022043491212392,
      "grad_norm": 14.9375,
      "learning_rate": 1.6359787418079254e-05,
      "loss": 0.9848,
      "num_input_tokens_seen": 50723160,
      "step": 87430
    },
    {
      "epoch": 13.02278820375335,
      "grad_norm": 7.4375,
      "learning_rate": 1.6356738302882864e-05,
      "loss": 1.0532,
      "num_input_tokens_seen": 50725912,
      "step": 87435
    },
    {
      "epoch": 13.023532916294311,
      "grad_norm": 12.75,
      "learning_rate": 1.6353689333705606e-05,
      "loss": 1.2136,
      "num_input_tokens_seen": 50728952,
      "step": 87440
    },
    {
      "epoch": 13.02427762883527,
      "grad_norm": 13.8125,
      "learning_rate": 1.6350640510598974e-05,
      "loss": 1.3987,
      "num_input_tokens_seen": 50731800,
      "step": 87445
    },
    {
      "epoch": 13.025022341376228,
      "grad_norm": 10.5625,
      "learning_rate": 1.634759183361449e-05,
      "loss": 1.2359,
      "num_input_tokens_seen": 50734776,
      "step": 87450
    },
    {
      "epoch": 13.025767053917187,
      "grad_norm": 14.25,
      "learning_rate": 1.6344543302803643e-05,
      "loss": 0.941,
      "num_input_tokens_seen": 50737432,
      "step": 87455
    },
    {
      "epoch": 13.026511766458148,
      "grad_norm": 11.3125,
      "learning_rate": 1.6341494918217938e-05,
      "loss": 1.2865,
      "num_input_tokens_seen": 50740920,
      "step": 87460
    },
    {
      "epoch": 13.027256478999107,
      "grad_norm": 9.375,
      "learning_rate": 1.633844667990888e-05,
      "loss": 1.0555,
      "num_input_tokens_seen": 50744152,
      "step": 87465
    },
    {
      "epoch": 13.028001191540065,
      "grad_norm": 11.5625,
      "learning_rate": 1.633539858792795e-05,
      "loss": 1.0178,
      "num_input_tokens_seen": 50747000,
      "step": 87470
    },
    {
      "epoch": 13.028745904081024,
      "grad_norm": 9.5,
      "learning_rate": 1.6332350642326673e-05,
      "loss": 0.8996,
      "num_input_tokens_seen": 50750104,
      "step": 87475
    },
    {
      "epoch": 13.029490616621985,
      "grad_norm": 7.34375,
      "learning_rate": 1.6329302843156503e-05,
      "loss": 0.9425,
      "num_input_tokens_seen": 50753400,
      "step": 87480
    },
    {
      "epoch": 13.030235329162943,
      "grad_norm": 14.25,
      "learning_rate": 1.6326255190468965e-05,
      "loss": 1.0664,
      "num_input_tokens_seen": 50756376,
      "step": 87485
    },
    {
      "epoch": 13.030980041703902,
      "grad_norm": 9.1875,
      "learning_rate": 1.632320768431553e-05,
      "loss": 1.5221,
      "num_input_tokens_seen": 50759864,
      "step": 87490
    },
    {
      "epoch": 13.03172475424486,
      "grad_norm": 6.5,
      "learning_rate": 1.6320160324747672e-05,
      "loss": 0.8857,
      "num_input_tokens_seen": 50762552,
      "step": 87495
    },
    {
      "epoch": 13.032469466785821,
      "grad_norm": 14.75,
      "learning_rate": 1.631711311181689e-05,
      "loss": 1.1759,
      "num_input_tokens_seen": 50765304,
      "step": 87500
    },
    {
      "epoch": 13.03321417932678,
      "grad_norm": 16.625,
      "learning_rate": 1.631406604557465e-05,
      "loss": 0.9617,
      "num_input_tokens_seen": 50768344,
      "step": 87505
    },
    {
      "epoch": 13.033958891867739,
      "grad_norm": 15.0,
      "learning_rate": 1.6311019126072447e-05,
      "loss": 1.0748,
      "num_input_tokens_seen": 50771256,
      "step": 87510
    },
    {
      "epoch": 13.034703604408698,
      "grad_norm": 10.875,
      "learning_rate": 1.630797235336173e-05,
      "loss": 1.0638,
      "num_input_tokens_seen": 50774200,
      "step": 87515
    },
    {
      "epoch": 13.035448316949658,
      "grad_norm": 16.875,
      "learning_rate": 1.6304925727493998e-05,
      "loss": 1.2097,
      "num_input_tokens_seen": 50777176,
      "step": 87520
    },
    {
      "epoch": 13.036193029490617,
      "grad_norm": 5.375,
      "learning_rate": 1.6301879248520707e-05,
      "loss": 0.9494,
      "num_input_tokens_seen": 50780088,
      "step": 87525
    },
    {
      "epoch": 13.036937742031576,
      "grad_norm": 20.0,
      "learning_rate": 1.629883291649333e-05,
      "loss": 1.3283,
      "num_input_tokens_seen": 50782936,
      "step": 87530
    },
    {
      "epoch": 13.037682454572534,
      "grad_norm": 10.0,
      "learning_rate": 1.6295786731463324e-05,
      "loss": 1.3237,
      "num_input_tokens_seen": 50785752,
      "step": 87535
    },
    {
      "epoch": 13.038427167113495,
      "grad_norm": 16.375,
      "learning_rate": 1.6292740693482144e-05,
      "loss": 1.0895,
      "num_input_tokens_seen": 50788472,
      "step": 87540
    },
    {
      "epoch": 13.039171879654454,
      "grad_norm": 12.875,
      "learning_rate": 1.6289694802601273e-05,
      "loss": 1.4141,
      "num_input_tokens_seen": 50791064,
      "step": 87545
    },
    {
      "epoch": 13.039916592195413,
      "grad_norm": 14.3125,
      "learning_rate": 1.628664905887215e-05,
      "loss": 1.247,
      "num_input_tokens_seen": 50793816,
      "step": 87550
    },
    {
      "epoch": 13.040661304736371,
      "grad_norm": 8.75,
      "learning_rate": 1.6283603462346235e-05,
      "loss": 1.1609,
      "num_input_tokens_seen": 50796856,
      "step": 87555
    },
    {
      "epoch": 13.041406017277332,
      "grad_norm": 8.875,
      "learning_rate": 1.628055801307498e-05,
      "loss": 0.9907,
      "num_input_tokens_seen": 50799512,
      "step": 87560
    },
    {
      "epoch": 13.04215072981829,
      "grad_norm": 7.25,
      "learning_rate": 1.6277512711109842e-05,
      "loss": 1.1279,
      "num_input_tokens_seen": 50802360,
      "step": 87565
    },
    {
      "epoch": 13.04289544235925,
      "grad_norm": 11.3125,
      "learning_rate": 1.627446755650226e-05,
      "loss": 0.9554,
      "num_input_tokens_seen": 50805048,
      "step": 87570
    },
    {
      "epoch": 13.043640154900208,
      "grad_norm": 10.75,
      "learning_rate": 1.627142254930367e-05,
      "loss": 0.9727,
      "num_input_tokens_seen": 50807704,
      "step": 87575
    },
    {
      "epoch": 13.044384867441167,
      "grad_norm": 17.5,
      "learning_rate": 1.6268377689565533e-05,
      "loss": 1.0105,
      "num_input_tokens_seen": 50810328,
      "step": 87580
    },
    {
      "epoch": 13.045129579982127,
      "grad_norm": 15.5625,
      "learning_rate": 1.6265332977339282e-05,
      "loss": 1.1975,
      "num_input_tokens_seen": 50813336,
      "step": 87585
    },
    {
      "epoch": 13.045874292523086,
      "grad_norm": 6.0625,
      "learning_rate": 1.6262288412676345e-05,
      "loss": 1.0315,
      "num_input_tokens_seen": 50815992,
      "step": 87590
    },
    {
      "epoch": 13.046619005064045,
      "grad_norm": 8.625,
      "learning_rate": 1.625924399562817e-05,
      "loss": 0.6577,
      "num_input_tokens_seen": 50818936,
      "step": 87595
    },
    {
      "epoch": 13.047363717605004,
      "grad_norm": 9.25,
      "learning_rate": 1.625619972624619e-05,
      "loss": 0.9552,
      "num_input_tokens_seen": 50821496,
      "step": 87600
    },
    {
      "epoch": 13.048108430145964,
      "grad_norm": 8.1875,
      "learning_rate": 1.6253155604581817e-05,
      "loss": 1.221,
      "num_input_tokens_seen": 50824408,
      "step": 87605
    },
    {
      "epoch": 13.048853142686923,
      "grad_norm": 15.6875,
      "learning_rate": 1.6250111630686498e-05,
      "loss": 1.021,
      "num_input_tokens_seen": 50827256,
      "step": 87610
    },
    {
      "epoch": 13.049597855227882,
      "grad_norm": 9.4375,
      "learning_rate": 1.6247067804611652e-05,
      "loss": 1.0931,
      "num_input_tokens_seen": 50830328,
      "step": 87615
    },
    {
      "epoch": 13.05034256776884,
      "grad_norm": 18.125,
      "learning_rate": 1.6244024126408695e-05,
      "loss": 1.3422,
      "num_input_tokens_seen": 50833080,
      "step": 87620
    },
    {
      "epoch": 13.051087280309801,
      "grad_norm": 8.5,
      "learning_rate": 1.6240980596129053e-05,
      "loss": 0.8822,
      "num_input_tokens_seen": 50835960,
      "step": 87625
    },
    {
      "epoch": 13.05183199285076,
      "grad_norm": 16.875,
      "learning_rate": 1.6237937213824134e-05,
      "loss": 1.179,
      "num_input_tokens_seen": 50838808,
      "step": 87630
    },
    {
      "epoch": 13.052576705391719,
      "grad_norm": 12.3125,
      "learning_rate": 1.623489397954537e-05,
      "loss": 1.3157,
      "num_input_tokens_seen": 50841656,
      "step": 87635
    },
    {
      "epoch": 13.053321417932677,
      "grad_norm": 7.84375,
      "learning_rate": 1.623185089334415e-05,
      "loss": 1.0061,
      "num_input_tokens_seen": 50844472,
      "step": 87640
    },
    {
      "epoch": 13.054066130473638,
      "grad_norm": 14.25,
      "learning_rate": 1.6228807955271915e-05,
      "loss": 1.0934,
      "num_input_tokens_seen": 50847416,
      "step": 87645
    },
    {
      "epoch": 13.054810843014597,
      "grad_norm": 8.125,
      "learning_rate": 1.6225765165380046e-05,
      "loss": 1.1307,
      "num_input_tokens_seen": 50850328,
      "step": 87650
    },
    {
      "epoch": 13.055555555555555,
      "grad_norm": 13.9375,
      "learning_rate": 1.6222722523719963e-05,
      "loss": 0.9527,
      "num_input_tokens_seen": 50853176,
      "step": 87655
    },
    {
      "epoch": 13.056300268096514,
      "grad_norm": 16.125,
      "learning_rate": 1.6219680030343063e-05,
      "loss": 1.2217,
      "num_input_tokens_seen": 50856184,
      "step": 87660
    },
    {
      "epoch": 13.057044980637475,
      "grad_norm": 12.375,
      "learning_rate": 1.6216637685300735e-05,
      "loss": 1.1199,
      "num_input_tokens_seen": 50859224,
      "step": 87665
    },
    {
      "epoch": 13.057789693178433,
      "grad_norm": 14.875,
      "learning_rate": 1.6213595488644393e-05,
      "loss": 0.8322,
      "num_input_tokens_seen": 50861848,
      "step": 87670
    },
    {
      "epoch": 13.058534405719392,
      "grad_norm": 4.59375,
      "learning_rate": 1.6210553440425415e-05,
      "loss": 1.0238,
      "num_input_tokens_seen": 50864856,
      "step": 87675
    },
    {
      "epoch": 13.059279118260351,
      "grad_norm": 4.625,
      "learning_rate": 1.6207511540695215e-05,
      "loss": 1.0557,
      "num_input_tokens_seen": 50868088,
      "step": 87680
    },
    {
      "epoch": 13.060023830801311,
      "grad_norm": 7.1875,
      "learning_rate": 1.6204469789505165e-05,
      "loss": 1.0325,
      "num_input_tokens_seen": 50870872,
      "step": 87685
    },
    {
      "epoch": 13.06076854334227,
      "grad_norm": 6.53125,
      "learning_rate": 1.620142818690667e-05,
      "loss": 0.7892,
      "num_input_tokens_seen": 50873976,
      "step": 87690
    },
    {
      "epoch": 13.061513255883229,
      "grad_norm": 10.25,
      "learning_rate": 1.61983867329511e-05,
      "loss": 0.9036,
      "num_input_tokens_seen": 50876888,
      "step": 87695
    },
    {
      "epoch": 13.062257968424188,
      "grad_norm": 8.0,
      "learning_rate": 1.6195345427689826e-05,
      "loss": 1.0862,
      "num_input_tokens_seen": 50879992,
      "step": 87700
    },
    {
      "epoch": 13.063002680965148,
      "grad_norm": 9.25,
      "learning_rate": 1.6192304271174256e-05,
      "loss": 1.0465,
      "num_input_tokens_seen": 50882776,
      "step": 87705
    },
    {
      "epoch": 13.063747393506107,
      "grad_norm": 7.1875,
      "learning_rate": 1.618926326345574e-05,
      "loss": 1.2541,
      "num_input_tokens_seen": 50885400,
      "step": 87710
    },
    {
      "epoch": 13.064492106047066,
      "grad_norm": 6.90625,
      "learning_rate": 1.618622240458568e-05,
      "loss": 1.1632,
      "num_input_tokens_seen": 50888120,
      "step": 87715
    },
    {
      "epoch": 13.065236818588025,
      "grad_norm": 8.0,
      "learning_rate": 1.618318169461543e-05,
      "loss": 1.185,
      "num_input_tokens_seen": 50891128,
      "step": 87720
    },
    {
      "epoch": 13.065981531128985,
      "grad_norm": 15.375,
      "learning_rate": 1.6180141133596367e-05,
      "loss": 1.2015,
      "num_input_tokens_seen": 50894040,
      "step": 87725
    },
    {
      "epoch": 13.066726243669944,
      "grad_norm": 16.0,
      "learning_rate": 1.6177100721579847e-05,
      "loss": 1.0589,
      "num_input_tokens_seen": 50896664,
      "step": 87730
    },
    {
      "epoch": 13.067470956210903,
      "grad_norm": 11.5,
      "learning_rate": 1.617406045861725e-05,
      "loss": 1.1899,
      "num_input_tokens_seen": 50899800,
      "step": 87735
    },
    {
      "epoch": 13.068215668751861,
      "grad_norm": 12.375,
      "learning_rate": 1.6171020344759936e-05,
      "loss": 1.4384,
      "num_input_tokens_seen": 50902776,
      "step": 87740
    },
    {
      "epoch": 13.06896038129282,
      "grad_norm": 11.3125,
      "learning_rate": 1.616798038005925e-05,
      "loss": 0.9478,
      "num_input_tokens_seen": 50905528,
      "step": 87745
    },
    {
      "epoch": 13.06970509383378,
      "grad_norm": 15.75,
      "learning_rate": 1.6164940564566566e-05,
      "loss": 0.7379,
      "num_input_tokens_seen": 50908408,
      "step": 87750
    },
    {
      "epoch": 13.07044980637474,
      "grad_norm": 11.0625,
      "learning_rate": 1.6161900898333225e-05,
      "loss": 1.3058,
      "num_input_tokens_seen": 50911192,
      "step": 87755
    },
    {
      "epoch": 13.071194518915698,
      "grad_norm": 30.625,
      "learning_rate": 1.6158861381410593e-05,
      "loss": 1.1658,
      "num_input_tokens_seen": 50914232,
      "step": 87760
    },
    {
      "epoch": 13.071939231456657,
      "grad_norm": 11.8125,
      "learning_rate": 1.6155822013850004e-05,
      "loss": 1.0368,
      "num_input_tokens_seen": 50916952,
      "step": 87765
    },
    {
      "epoch": 13.072683943997617,
      "grad_norm": 12.6875,
      "learning_rate": 1.615278279570282e-05,
      "loss": 0.887,
      "num_input_tokens_seen": 50920056,
      "step": 87770
    },
    {
      "epoch": 13.073428656538576,
      "grad_norm": 7.71875,
      "learning_rate": 1.614974372702038e-05,
      "loss": 1.1101,
      "num_input_tokens_seen": 50923128,
      "step": 87775
    },
    {
      "epoch": 13.074173369079535,
      "grad_norm": 9.0,
      "learning_rate": 1.6146704807854014e-05,
      "loss": 1.0697,
      "num_input_tokens_seen": 50926136,
      "step": 87780
    },
    {
      "epoch": 13.074918081620494,
      "grad_norm": 9.0625,
      "learning_rate": 1.6143666038255084e-05,
      "loss": 0.8696,
      "num_input_tokens_seen": 50929080,
      "step": 87785
    },
    {
      "epoch": 13.075662794161454,
      "grad_norm": 16.25,
      "learning_rate": 1.614062741827491e-05,
      "loss": 1.2419,
      "num_input_tokens_seen": 50932088,
      "step": 87790
    },
    {
      "epoch": 13.076407506702413,
      "grad_norm": 8.5,
      "learning_rate": 1.6137588947964838e-05,
      "loss": 0.84,
      "num_input_tokens_seen": 50935064,
      "step": 87795
    },
    {
      "epoch": 13.077152219243372,
      "grad_norm": 11.25,
      "learning_rate": 1.613455062737618e-05,
      "loss": 0.7427,
      "num_input_tokens_seen": 50938392,
      "step": 87800
    },
    {
      "epoch": 13.07789693178433,
      "grad_norm": 8.5625,
      "learning_rate": 1.613151245656029e-05,
      "loss": 1.0685,
      "num_input_tokens_seen": 50941208,
      "step": 87805
    },
    {
      "epoch": 13.078641644325291,
      "grad_norm": 4.5625,
      "learning_rate": 1.612847443556847e-05,
      "loss": 1.0868,
      "num_input_tokens_seen": 50943928,
      "step": 87810
    },
    {
      "epoch": 13.07938635686625,
      "grad_norm": 8.625,
      "learning_rate": 1.6125436564452075e-05,
      "loss": 1.12,
      "num_input_tokens_seen": 50947032,
      "step": 87815
    },
    {
      "epoch": 13.080131069407209,
      "grad_norm": 11.375,
      "learning_rate": 1.6122398843262405e-05,
      "loss": 0.8204,
      "num_input_tokens_seen": 50949816,
      "step": 87820
    },
    {
      "epoch": 13.080875781948167,
      "grad_norm": 3.46875,
      "learning_rate": 1.6119361272050777e-05,
      "loss": 1.0555,
      "num_input_tokens_seen": 50952792,
      "step": 87825
    },
    {
      "epoch": 13.081620494489128,
      "grad_norm": 5.1875,
      "learning_rate": 1.6116323850868526e-05,
      "loss": 1.0673,
      "num_input_tokens_seen": 50955992,
      "step": 87830
    },
    {
      "epoch": 13.082365207030087,
      "grad_norm": 12.1875,
      "learning_rate": 1.611328657976694e-05,
      "loss": 1.1571,
      "num_input_tokens_seen": 50958840,
      "step": 87835
    },
    {
      "epoch": 13.083109919571045,
      "grad_norm": 6.9375,
      "learning_rate": 1.6110249458797355e-05,
      "loss": 0.8337,
      "num_input_tokens_seen": 50961688,
      "step": 87840
    },
    {
      "epoch": 13.083854632112004,
      "grad_norm": 9.4375,
      "learning_rate": 1.610721248801106e-05,
      "loss": 1.44,
      "num_input_tokens_seen": 50964696,
      "step": 87845
    },
    {
      "epoch": 13.084599344652965,
      "grad_norm": 9.9375,
      "learning_rate": 1.610417566745938e-05,
      "loss": 1.197,
      "num_input_tokens_seen": 50967576,
      "step": 87850
    },
    {
      "epoch": 13.085344057193923,
      "grad_norm": 15.0625,
      "learning_rate": 1.6101138997193615e-05,
      "loss": 1.065,
      "num_input_tokens_seen": 50970488,
      "step": 87855
    },
    {
      "epoch": 13.086088769734882,
      "grad_norm": 5.96875,
      "learning_rate": 1.6098102477265057e-05,
      "loss": 1.1907,
      "num_input_tokens_seen": 50974040,
      "step": 87860
    },
    {
      "epoch": 13.086833482275841,
      "grad_norm": 19.0,
      "learning_rate": 1.6095066107725015e-05,
      "loss": 1.0821,
      "num_input_tokens_seen": 50976952,
      "step": 87865
    },
    {
      "epoch": 13.087578194816802,
      "grad_norm": 23.75,
      "learning_rate": 1.609202988862477e-05,
      "loss": 0.9198,
      "num_input_tokens_seen": 50980216,
      "step": 87870
    },
    {
      "epoch": 13.08832290735776,
      "grad_norm": 6.34375,
      "learning_rate": 1.6088993820015634e-05,
      "loss": 0.9932,
      "num_input_tokens_seen": 50983096,
      "step": 87875
    },
    {
      "epoch": 13.089067619898719,
      "grad_norm": 12.5625,
      "learning_rate": 1.608595790194889e-05,
      "loss": 1.3934,
      "num_input_tokens_seen": 50986040,
      "step": 87880
    },
    {
      "epoch": 13.089812332439678,
      "grad_norm": 7.46875,
      "learning_rate": 1.6082922134475823e-05,
      "loss": 0.8569,
      "num_input_tokens_seen": 50988920,
      "step": 87885
    },
    {
      "epoch": 13.090557044980638,
      "grad_norm": 11.375,
      "learning_rate": 1.6079886517647723e-05,
      "loss": 0.8277,
      "num_input_tokens_seen": 50991832,
      "step": 87890
    },
    {
      "epoch": 13.091301757521597,
      "grad_norm": 7.5625,
      "learning_rate": 1.6076851051515884e-05,
      "loss": 1.3718,
      "num_input_tokens_seen": 50995000,
      "step": 87895
    },
    {
      "epoch": 13.092046470062556,
      "grad_norm": 7.625,
      "learning_rate": 1.6073815736131577e-05,
      "loss": 1.2479,
      "num_input_tokens_seen": 50997784,
      "step": 87900
    },
    {
      "epoch": 13.092791182603515,
      "grad_norm": 16.75,
      "learning_rate": 1.6070780571546066e-05,
      "loss": 1.1789,
      "num_input_tokens_seen": 51000632,
      "step": 87905
    },
    {
      "epoch": 13.093535895144473,
      "grad_norm": 10.5625,
      "learning_rate": 1.6067745557810656e-05,
      "loss": 1.4833,
      "num_input_tokens_seen": 51003320,
      "step": 87910
    },
    {
      "epoch": 13.094280607685434,
      "grad_norm": 14.5625,
      "learning_rate": 1.60647106949766e-05,
      "loss": 1.0727,
      "num_input_tokens_seen": 51006232,
      "step": 87915
    },
    {
      "epoch": 13.095025320226393,
      "grad_norm": 9.25,
      "learning_rate": 1.6061675983095177e-05,
      "loss": 0.8463,
      "num_input_tokens_seen": 51009080,
      "step": 87920
    },
    {
      "epoch": 13.095770032767351,
      "grad_norm": 6.5,
      "learning_rate": 1.605864142221765e-05,
      "loss": 1.1732,
      "num_input_tokens_seen": 51011864,
      "step": 87925
    },
    {
      "epoch": 13.09651474530831,
      "grad_norm": 17.0,
      "learning_rate": 1.60556070123953e-05,
      "loss": 1.069,
      "num_input_tokens_seen": 51014808,
      "step": 87930
    },
    {
      "epoch": 13.09725945784927,
      "grad_norm": 10.625,
      "learning_rate": 1.6052572753679372e-05,
      "loss": 1.4115,
      "num_input_tokens_seen": 51017880,
      "step": 87935
    },
    {
      "epoch": 13.09800417039023,
      "grad_norm": 16.875,
      "learning_rate": 1.604953864612113e-05,
      "loss": 1.0921,
      "num_input_tokens_seen": 51020696,
      "step": 87940
    },
    {
      "epoch": 13.098748882931188,
      "grad_norm": 6.84375,
      "learning_rate": 1.6046504689771842e-05,
      "loss": 0.9902,
      "num_input_tokens_seen": 51023608,
      "step": 87945
    },
    {
      "epoch": 13.099493595472147,
      "grad_norm": 10.9375,
      "learning_rate": 1.6043470884682753e-05,
      "loss": 0.9097,
      "num_input_tokens_seen": 51026232,
      "step": 87950
    },
    {
      "epoch": 13.100238308013108,
      "grad_norm": 12.75,
      "learning_rate": 1.6040437230905126e-05,
      "loss": 1.1988,
      "num_input_tokens_seen": 51029496,
      "step": 87955
    },
    {
      "epoch": 13.100983020554066,
      "grad_norm": 15.875,
      "learning_rate": 1.6037403728490193e-05,
      "loss": 1.1881,
      "num_input_tokens_seen": 51032408,
      "step": 87960
    },
    {
      "epoch": 13.101727733095025,
      "grad_norm": 6.5,
      "learning_rate": 1.6034370377489227e-05,
      "loss": 1.1718,
      "num_input_tokens_seen": 51035224,
      "step": 87965
    },
    {
      "epoch": 13.102472445635984,
      "grad_norm": 9.125,
      "learning_rate": 1.6031337177953455e-05,
      "loss": 1.0359,
      "num_input_tokens_seen": 51038264,
      "step": 87970
    },
    {
      "epoch": 13.103217158176944,
      "grad_norm": 9.875,
      "learning_rate": 1.602830412993413e-05,
      "loss": 1.2632,
      "num_input_tokens_seen": 51041176,
      "step": 87975
    },
    {
      "epoch": 13.103961870717903,
      "grad_norm": 7.65625,
      "learning_rate": 1.6025271233482492e-05,
      "loss": 1.1123,
      "num_input_tokens_seen": 51044088,
      "step": 87980
    },
    {
      "epoch": 13.104706583258862,
      "grad_norm": 7.46875,
      "learning_rate": 1.6022238488649764e-05,
      "loss": 1.1561,
      "num_input_tokens_seen": 51047224,
      "step": 87985
    },
    {
      "epoch": 13.10545129579982,
      "grad_norm": 4.3125,
      "learning_rate": 1.6019205895487204e-05,
      "loss": 0.7527,
      "num_input_tokens_seen": 51050040,
      "step": 87990
    },
    {
      "epoch": 13.106196008340781,
      "grad_norm": 16.375,
      "learning_rate": 1.6016173454046018e-05,
      "loss": 1.2732,
      "num_input_tokens_seen": 51053048,
      "step": 87995
    },
    {
      "epoch": 13.10694072088174,
      "grad_norm": 11.0,
      "learning_rate": 1.6013141164377467e-05,
      "loss": 1.3223,
      "num_input_tokens_seen": 51055864,
      "step": 88000
    },
    {
      "epoch": 13.107685433422699,
      "grad_norm": 14.125,
      "learning_rate": 1.6010109026532747e-05,
      "loss": 1.0804,
      "num_input_tokens_seen": 51058776,
      "step": 88005
    },
    {
      "epoch": 13.108430145963657,
      "grad_norm": 7.78125,
      "learning_rate": 1.600707704056311e-05,
      "loss": 1.3536,
      "num_input_tokens_seen": 51061752,
      "step": 88010
    },
    {
      "epoch": 13.109174858504618,
      "grad_norm": 11.875,
      "learning_rate": 1.6004045206519763e-05,
      "loss": 0.9312,
      "num_input_tokens_seen": 51064664,
      "step": 88015
    },
    {
      "epoch": 13.109919571045577,
      "grad_norm": 13.0625,
      "learning_rate": 1.6001013524453928e-05,
      "loss": 1.1234,
      "num_input_tokens_seen": 51067608,
      "step": 88020
    },
    {
      "epoch": 13.110664283586535,
      "grad_norm": 18.25,
      "learning_rate": 1.599798199441683e-05,
      "loss": 1.5146,
      "num_input_tokens_seen": 51070200,
      "step": 88025
    },
    {
      "epoch": 13.111408996127494,
      "grad_norm": 5.8125,
      "learning_rate": 1.5994950616459664e-05,
      "loss": 1.0663,
      "num_input_tokens_seen": 51073080,
      "step": 88030
    },
    {
      "epoch": 13.112153708668455,
      "grad_norm": 8.5625,
      "learning_rate": 1.5991919390633662e-05,
      "loss": 0.9944,
      "num_input_tokens_seen": 51076152,
      "step": 88035
    },
    {
      "epoch": 13.112898421209414,
      "grad_norm": 5.75,
      "learning_rate": 1.5988888316990018e-05,
      "loss": 1.1307,
      "num_input_tokens_seen": 51078872,
      "step": 88040
    },
    {
      "epoch": 13.113643133750372,
      "grad_norm": 17.0,
      "learning_rate": 1.5985857395579963e-05,
      "loss": 1.3201,
      "num_input_tokens_seen": 51081624,
      "step": 88045
    },
    {
      "epoch": 13.114387846291331,
      "grad_norm": 7.15625,
      "learning_rate": 1.5982826626454678e-05,
      "loss": 1.2797,
      "num_input_tokens_seen": 51084440,
      "step": 88050
    },
    {
      "epoch": 13.115132558832292,
      "grad_norm": 16.0,
      "learning_rate": 1.5979796009665376e-05,
      "loss": 1.043,
      "num_input_tokens_seen": 51087224,
      "step": 88055
    },
    {
      "epoch": 13.11587727137325,
      "grad_norm": 10.4375,
      "learning_rate": 1.5976765545263254e-05,
      "loss": 1.2082,
      "num_input_tokens_seen": 51090104,
      "step": 88060
    },
    {
      "epoch": 13.116621983914209,
      "grad_norm": 14.1875,
      "learning_rate": 1.5973735233299496e-05,
      "loss": 1.1484,
      "num_input_tokens_seen": 51092984,
      "step": 88065
    },
    {
      "epoch": 13.117366696455168,
      "grad_norm": 10.5625,
      "learning_rate": 1.5970705073825315e-05,
      "loss": 1.1341,
      "num_input_tokens_seen": 51095704,
      "step": 88070
    },
    {
      "epoch": 13.118111408996128,
      "grad_norm": 13.9375,
      "learning_rate": 1.596767506689189e-05,
      "loss": 1.1281,
      "num_input_tokens_seen": 51098520,
      "step": 88075
    },
    {
      "epoch": 13.118856121537087,
      "grad_norm": 11.6875,
      "learning_rate": 1.5964645212550422e-05,
      "loss": 1.0256,
      "num_input_tokens_seen": 51101400,
      "step": 88080
    },
    {
      "epoch": 13.119600834078046,
      "grad_norm": 7.625,
      "learning_rate": 1.5961615510852083e-05,
      "loss": 1.0796,
      "num_input_tokens_seen": 51104728,
      "step": 88085
    },
    {
      "epoch": 13.120345546619005,
      "grad_norm": 9.0625,
      "learning_rate": 1.5958585961848072e-05,
      "loss": 0.96,
      "num_input_tokens_seen": 51107576,
      "step": 88090
    },
    {
      "epoch": 13.121090259159963,
      "grad_norm": 4.71875,
      "learning_rate": 1.5955556565589564e-05,
      "loss": 0.8723,
      "num_input_tokens_seen": 51110584,
      "step": 88095
    },
    {
      "epoch": 13.121834971700924,
      "grad_norm": 7.0625,
      "learning_rate": 1.5952527322127718e-05,
      "loss": 1.0724,
      "num_input_tokens_seen": 51113432,
      "step": 88100
    },
    {
      "epoch": 13.122579684241883,
      "grad_norm": 9.625,
      "learning_rate": 1.5949498231513744e-05,
      "loss": 1.0711,
      "num_input_tokens_seen": 51116312,
      "step": 88105
    },
    {
      "epoch": 13.123324396782841,
      "grad_norm": 6.03125,
      "learning_rate": 1.5946469293798788e-05,
      "loss": 1.1697,
      "num_input_tokens_seen": 51119288,
      "step": 88110
    },
    {
      "epoch": 13.1240691093238,
      "grad_norm": 5.53125,
      "learning_rate": 1.5943440509034038e-05,
      "loss": 1.5735,
      "num_input_tokens_seen": 51122616,
      "step": 88115
    },
    {
      "epoch": 13.12481382186476,
      "grad_norm": 14.8125,
      "learning_rate": 1.5940411877270655e-05,
      "loss": 1.1711,
      "num_input_tokens_seen": 51125272,
      "step": 88120
    },
    {
      "epoch": 13.12555853440572,
      "grad_norm": 6.78125,
      "learning_rate": 1.5937383398559808e-05,
      "loss": 1.0788,
      "num_input_tokens_seen": 51128248,
      "step": 88125
    },
    {
      "epoch": 13.126303246946678,
      "grad_norm": 12.0,
      "learning_rate": 1.593435507295265e-05,
      "loss": 1.0443,
      "num_input_tokens_seen": 51131384,
      "step": 88130
    },
    {
      "epoch": 13.127047959487637,
      "grad_norm": 13.1875,
      "learning_rate": 1.5931326900500353e-05,
      "loss": 1.2244,
      "num_input_tokens_seen": 51134232,
      "step": 88135
    },
    {
      "epoch": 13.127792672028598,
      "grad_norm": 7.90625,
      "learning_rate": 1.5928298881254077e-05,
      "loss": 0.9667,
      "num_input_tokens_seen": 51137080,
      "step": 88140
    },
    {
      "epoch": 13.128537384569556,
      "grad_norm": 12.3125,
      "learning_rate": 1.5925271015264962e-05,
      "loss": 1.1694,
      "num_input_tokens_seen": 51140120,
      "step": 88145
    },
    {
      "epoch": 13.129282097110515,
      "grad_norm": 5.78125,
      "learning_rate": 1.5922243302584176e-05,
      "loss": 1.0085,
      "num_input_tokens_seen": 51142904,
      "step": 88150
    },
    {
      "epoch": 13.130026809651474,
      "grad_norm": 8.0,
      "learning_rate": 1.5919215743262862e-05,
      "loss": 0.8936,
      "num_input_tokens_seen": 51145528,
      "step": 88155
    },
    {
      "epoch": 13.130771522192434,
      "grad_norm": 16.625,
      "learning_rate": 1.591618833735217e-05,
      "loss": 1.3038,
      "num_input_tokens_seen": 51148440,
      "step": 88160
    },
    {
      "epoch": 13.131516234733393,
      "grad_norm": 8.3125,
      "learning_rate": 1.5913161084903237e-05,
      "loss": 0.879,
      "num_input_tokens_seen": 51151512,
      "step": 88165
    },
    {
      "epoch": 13.132260947274352,
      "grad_norm": 9.25,
      "learning_rate": 1.591013398596722e-05,
      "loss": 0.8659,
      "num_input_tokens_seen": 51154776,
      "step": 88170
    },
    {
      "epoch": 13.13300565981531,
      "grad_norm": 14.0625,
      "learning_rate": 1.5907107040595255e-05,
      "loss": 0.9647,
      "num_input_tokens_seen": 51157656,
      "step": 88175
    },
    {
      "epoch": 13.133750372356271,
      "grad_norm": 5.875,
      "learning_rate": 1.590408024883846e-05,
      "loss": 0.9821,
      "num_input_tokens_seen": 51160664,
      "step": 88180
    },
    {
      "epoch": 13.13449508489723,
      "grad_norm": 5.4375,
      "learning_rate": 1.5901053610747995e-05,
      "loss": 0.8728,
      "num_input_tokens_seen": 51163384,
      "step": 88185
    },
    {
      "epoch": 13.135239797438189,
      "grad_norm": 5.75,
      "learning_rate": 1.5898027126374974e-05,
      "loss": 1.0261,
      "num_input_tokens_seen": 51166200,
      "step": 88190
    },
    {
      "epoch": 13.135984509979147,
      "grad_norm": 10.6875,
      "learning_rate": 1.5895000795770547e-05,
      "loss": 1.1959,
      "num_input_tokens_seen": 51169304,
      "step": 88195
    },
    {
      "epoch": 13.136729222520108,
      "grad_norm": 9.0625,
      "learning_rate": 1.589197461898581e-05,
      "loss": 1.1761,
      "num_input_tokens_seen": 51172408,
      "step": 88200
    },
    {
      "epoch": 13.137473935061067,
      "grad_norm": 4.25,
      "learning_rate": 1.588894859607192e-05,
      "loss": 0.8498,
      "num_input_tokens_seen": 51175224,
      "step": 88205
    },
    {
      "epoch": 13.138218647602026,
      "grad_norm": 10.75,
      "learning_rate": 1.5885922727079977e-05,
      "loss": 1.0847,
      "num_input_tokens_seen": 51178296,
      "step": 88210
    },
    {
      "epoch": 13.138963360142984,
      "grad_norm": 9.3125,
      "learning_rate": 1.5882897012061104e-05,
      "loss": 0.9921,
      "num_input_tokens_seen": 51181208,
      "step": 88215
    },
    {
      "epoch": 13.139708072683945,
      "grad_norm": 7.1875,
      "learning_rate": 1.5879871451066424e-05,
      "loss": 0.7722,
      "num_input_tokens_seen": 51184088,
      "step": 88220
    },
    {
      "epoch": 13.140452785224904,
      "grad_norm": 9.1875,
      "learning_rate": 1.5876846044147048e-05,
      "loss": 0.7177,
      "num_input_tokens_seen": 51186776,
      "step": 88225
    },
    {
      "epoch": 13.141197497765862,
      "grad_norm": 14.4375,
      "learning_rate": 1.5873820791354085e-05,
      "loss": 0.9742,
      "num_input_tokens_seen": 51189656,
      "step": 88230
    },
    {
      "epoch": 13.141942210306821,
      "grad_norm": 7.1875,
      "learning_rate": 1.5870795692738635e-05,
      "loss": 0.945,
      "num_input_tokens_seen": 51192472,
      "step": 88235
    },
    {
      "epoch": 13.142686922847782,
      "grad_norm": 8.0625,
      "learning_rate": 1.5867770748351822e-05,
      "loss": 1.0481,
      "num_input_tokens_seen": 51195576,
      "step": 88240
    },
    {
      "epoch": 13.14343163538874,
      "grad_norm": 9.5,
      "learning_rate": 1.586474595824474e-05,
      "loss": 0.953,
      "num_input_tokens_seen": 51198296,
      "step": 88245
    },
    {
      "epoch": 13.1441763479297,
      "grad_norm": 10.3125,
      "learning_rate": 1.5861721322468487e-05,
      "loss": 1.2116,
      "num_input_tokens_seen": 51201304,
      "step": 88250
    },
    {
      "epoch": 13.144921060470658,
      "grad_norm": 7.71875,
      "learning_rate": 1.5858696841074166e-05,
      "loss": 0.8396,
      "num_input_tokens_seen": 51204280,
      "step": 88255
    },
    {
      "epoch": 13.145665773011617,
      "grad_norm": 5.25,
      "learning_rate": 1.5855672514112876e-05,
      "loss": 1.1141,
      "num_input_tokens_seen": 51207000,
      "step": 88260
    },
    {
      "epoch": 13.146410485552577,
      "grad_norm": 12.0625,
      "learning_rate": 1.5852648341635705e-05,
      "loss": 1.1147,
      "num_input_tokens_seen": 51209848,
      "step": 88265
    },
    {
      "epoch": 13.147155198093536,
      "grad_norm": 6.0,
      "learning_rate": 1.5849624323693736e-05,
      "loss": 1.0038,
      "num_input_tokens_seen": 51212792,
      "step": 88270
    },
    {
      "epoch": 13.147899910634495,
      "grad_norm": 10.625,
      "learning_rate": 1.5846600460338068e-05,
      "loss": 1.14,
      "num_input_tokens_seen": 51215544,
      "step": 88275
    },
    {
      "epoch": 13.148644623175453,
      "grad_norm": 6.125,
      "learning_rate": 1.584357675161978e-05,
      "loss": 1.3164,
      "num_input_tokens_seen": 51218360,
      "step": 88280
    },
    {
      "epoch": 13.149389335716414,
      "grad_norm": 7.375,
      "learning_rate": 1.5840553197589964e-05,
      "loss": 0.9945,
      "num_input_tokens_seen": 51221464,
      "step": 88285
    },
    {
      "epoch": 13.150134048257373,
      "grad_norm": 9.25,
      "learning_rate": 1.5837529798299677e-05,
      "loss": 0.8009,
      "num_input_tokens_seen": 51224152,
      "step": 88290
    },
    {
      "epoch": 13.150878760798332,
      "grad_norm": 10.3125,
      "learning_rate": 1.5834506553800026e-05,
      "loss": 0.9612,
      "num_input_tokens_seen": 51226904,
      "step": 88295
    },
    {
      "epoch": 13.15162347333929,
      "grad_norm": 17.75,
      "learning_rate": 1.583148346414207e-05,
      "loss": 1.2168,
      "num_input_tokens_seen": 51229976,
      "step": 88300
    },
    {
      "epoch": 13.15236818588025,
      "grad_norm": 20.0,
      "learning_rate": 1.5828460529376876e-05,
      "loss": 1.2169,
      "num_input_tokens_seen": 51233016,
      "step": 88305
    },
    {
      "epoch": 13.15311289842121,
      "grad_norm": 12.125,
      "learning_rate": 1.5825437749555525e-05,
      "loss": 0.9196,
      "num_input_tokens_seen": 51236024,
      "step": 88310
    },
    {
      "epoch": 13.153857610962168,
      "grad_norm": 8.3125,
      "learning_rate": 1.582241512472907e-05,
      "loss": 0.736,
      "num_input_tokens_seen": 51238680,
      "step": 88315
    },
    {
      "epoch": 13.154602323503127,
      "grad_norm": 11.0625,
      "learning_rate": 1.58193926549486e-05,
      "loss": 1.0157,
      "num_input_tokens_seen": 51241624,
      "step": 88320
    },
    {
      "epoch": 13.155347036044088,
      "grad_norm": 17.0,
      "learning_rate": 1.5816370340265144e-05,
      "loss": 1.4809,
      "num_input_tokens_seen": 51244760,
      "step": 88325
    },
    {
      "epoch": 13.156091748585046,
      "grad_norm": 13.6875,
      "learning_rate": 1.5813348180729788e-05,
      "loss": 0.9017,
      "num_input_tokens_seen": 51247448,
      "step": 88330
    },
    {
      "epoch": 13.156836461126005,
      "grad_norm": 19.125,
      "learning_rate": 1.5810326176393566e-05,
      "loss": 1.3419,
      "num_input_tokens_seen": 51250520,
      "step": 88335
    },
    {
      "epoch": 13.157581173666964,
      "grad_norm": 4.75,
      "learning_rate": 1.5807304327307556e-05,
      "loss": 0.9901,
      "num_input_tokens_seen": 51253560,
      "step": 88340
    },
    {
      "epoch": 13.158325886207924,
      "grad_norm": 12.25,
      "learning_rate": 1.5804282633522793e-05,
      "loss": 1.0275,
      "num_input_tokens_seen": 51256312,
      "step": 88345
    },
    {
      "epoch": 13.159070598748883,
      "grad_norm": 11.375,
      "learning_rate": 1.5801261095090325e-05,
      "loss": 1.0038,
      "num_input_tokens_seen": 51259160,
      "step": 88350
    },
    {
      "epoch": 13.159815311289842,
      "grad_norm": 5.84375,
      "learning_rate": 1.579823971206121e-05,
      "loss": 0.8133,
      "num_input_tokens_seen": 51262008,
      "step": 88355
    },
    {
      "epoch": 13.1605600238308,
      "grad_norm": 9.875,
      "learning_rate": 1.5795218484486468e-05,
      "loss": 0.7633,
      "num_input_tokens_seen": 51264728,
      "step": 88360
    },
    {
      "epoch": 13.161304736371761,
      "grad_norm": 9.5,
      "learning_rate": 1.5792197412417167e-05,
      "loss": 1.1466,
      "num_input_tokens_seen": 51267928,
      "step": 88365
    },
    {
      "epoch": 13.16204944891272,
      "grad_norm": 8.5,
      "learning_rate": 1.578917649590432e-05,
      "loss": 0.9303,
      "num_input_tokens_seen": 51270680,
      "step": 88370
    },
    {
      "epoch": 13.162794161453679,
      "grad_norm": 11.75,
      "learning_rate": 1.5786155734998988e-05,
      "loss": 1.3183,
      "num_input_tokens_seen": 51273688,
      "step": 88375
    },
    {
      "epoch": 13.163538873994638,
      "grad_norm": 7.0625,
      "learning_rate": 1.578313512975219e-05,
      "loss": 1.438,
      "num_input_tokens_seen": 51276984,
      "step": 88380
    },
    {
      "epoch": 13.164283586535598,
      "grad_norm": 9.5625,
      "learning_rate": 1.5780114680214948e-05,
      "loss": 1.2152,
      "num_input_tokens_seen": 51279928,
      "step": 88385
    },
    {
      "epoch": 13.165028299076557,
      "grad_norm": 17.875,
      "learning_rate": 1.5777094386438306e-05,
      "loss": 0.8241,
      "num_input_tokens_seen": 51283000,
      "step": 88390
    },
    {
      "epoch": 13.165773011617516,
      "grad_norm": 16.625,
      "learning_rate": 1.577407424847327e-05,
      "loss": 1.1513,
      "num_input_tokens_seen": 51285752,
      "step": 88395
    },
    {
      "epoch": 13.166517724158474,
      "grad_norm": 9.1875,
      "learning_rate": 1.5771054266370882e-05,
      "loss": 1.0102,
      "num_input_tokens_seen": 51288856,
      "step": 88400
    },
    {
      "epoch": 13.167262436699435,
      "grad_norm": 13.0,
      "learning_rate": 1.5768034440182143e-05,
      "loss": 1.186,
      "num_input_tokens_seen": 51291672,
      "step": 88405
    },
    {
      "epoch": 13.168007149240394,
      "grad_norm": 11.25,
      "learning_rate": 1.576501476995809e-05,
      "loss": 1.2027,
      "num_input_tokens_seen": 51294360,
      "step": 88410
    },
    {
      "epoch": 13.168751861781352,
      "grad_norm": 16.0,
      "learning_rate": 1.576199525574972e-05,
      "loss": 1.3212,
      "num_input_tokens_seen": 51297176,
      "step": 88415
    },
    {
      "epoch": 13.169496574322311,
      "grad_norm": 10.125,
      "learning_rate": 1.5758975897608056e-05,
      "loss": 1.3519,
      "num_input_tokens_seen": 51299736,
      "step": 88420
    },
    {
      "epoch": 13.17024128686327,
      "grad_norm": 5.5625,
      "learning_rate": 1.57559566955841e-05,
      "loss": 1.1125,
      "num_input_tokens_seen": 51302616,
      "step": 88425
    },
    {
      "epoch": 13.17098599940423,
      "grad_norm": 12.0,
      "learning_rate": 1.5752937649728854e-05,
      "loss": 0.963,
      "num_input_tokens_seen": 51305304,
      "step": 88430
    },
    {
      "epoch": 13.17173071194519,
      "grad_norm": 19.25,
      "learning_rate": 1.574991876009334e-05,
      "loss": 1.1607,
      "num_input_tokens_seen": 51308216,
      "step": 88435
    },
    {
      "epoch": 13.172475424486148,
      "grad_norm": 9.125,
      "learning_rate": 1.574690002672853e-05,
      "loss": 0.7,
      "num_input_tokens_seen": 51311000,
      "step": 88440
    },
    {
      "epoch": 13.173220137027107,
      "grad_norm": 7.84375,
      "learning_rate": 1.5743881449685456e-05,
      "loss": 1.0059,
      "num_input_tokens_seen": 51314328,
      "step": 88445
    },
    {
      "epoch": 13.173964849568067,
      "grad_norm": 10.0625,
      "learning_rate": 1.574086302901509e-05,
      "loss": 1.1991,
      "num_input_tokens_seen": 51317304,
      "step": 88450
    },
    {
      "epoch": 13.174709562109026,
      "grad_norm": 8.9375,
      "learning_rate": 1.5737844764768437e-05,
      "loss": 1.1594,
      "num_input_tokens_seen": 51320536,
      "step": 88455
    },
    {
      "epoch": 13.175454274649985,
      "grad_norm": 9.1875,
      "learning_rate": 1.5734826656996482e-05,
      "loss": 1.1203,
      "num_input_tokens_seen": 51323352,
      "step": 88460
    },
    {
      "epoch": 13.176198987190944,
      "grad_norm": 11.25,
      "learning_rate": 1.5731808705750206e-05,
      "loss": 0.9981,
      "num_input_tokens_seen": 51326072,
      "step": 88465
    },
    {
      "epoch": 13.176943699731904,
      "grad_norm": 6.09375,
      "learning_rate": 1.5728790911080612e-05,
      "loss": 0.8042,
      "num_input_tokens_seen": 51328952,
      "step": 88470
    },
    {
      "epoch": 13.177688412272863,
      "grad_norm": 9.9375,
      "learning_rate": 1.5725773273038662e-05,
      "loss": 0.9465,
      "num_input_tokens_seen": 51331832,
      "step": 88475
    },
    {
      "epoch": 13.178433124813822,
      "grad_norm": 11.625,
      "learning_rate": 1.5722755791675358e-05,
      "loss": 1.1291,
      "num_input_tokens_seen": 51334776,
      "step": 88480
    },
    {
      "epoch": 13.17917783735478,
      "grad_norm": 13.5,
      "learning_rate": 1.571973846704166e-05,
      "loss": 1.0599,
      "num_input_tokens_seen": 51337432,
      "step": 88485
    },
    {
      "epoch": 13.17992254989574,
      "grad_norm": 14.6875,
      "learning_rate": 1.5716721299188553e-05,
      "loss": 1.2277,
      "num_input_tokens_seen": 51340472,
      "step": 88490
    },
    {
      "epoch": 13.1806672624367,
      "grad_norm": 14.6875,
      "learning_rate": 1.5713704288166998e-05,
      "loss": 1.0744,
      "num_input_tokens_seen": 51343160,
      "step": 88495
    },
    {
      "epoch": 13.181411974977658,
      "grad_norm": 10.9375,
      "learning_rate": 1.5710687434027976e-05,
      "loss": 1.2488,
      "num_input_tokens_seen": 51345976,
      "step": 88500
    },
    {
      "epoch": 13.182156687518617,
      "grad_norm": 15.875,
      "learning_rate": 1.5707670736822448e-05,
      "loss": 1.0632,
      "num_input_tokens_seen": 51348600,
      "step": 88505
    },
    {
      "epoch": 13.182901400059578,
      "grad_norm": 10.125,
      "learning_rate": 1.5704654196601375e-05,
      "loss": 1.0873,
      "num_input_tokens_seen": 51351512,
      "step": 88510
    },
    {
      "epoch": 13.183646112600536,
      "grad_norm": 8.875,
      "learning_rate": 1.5701637813415728e-05,
      "loss": 1.0754,
      "num_input_tokens_seen": 51354456,
      "step": 88515
    },
    {
      "epoch": 13.184390825141495,
      "grad_norm": 6.84375,
      "learning_rate": 1.5698621587316454e-05,
      "loss": 0.8451,
      "num_input_tokens_seen": 51357496,
      "step": 88520
    },
    {
      "epoch": 13.185135537682454,
      "grad_norm": 8.5625,
      "learning_rate": 1.5695605518354524e-05,
      "loss": 0.8657,
      "num_input_tokens_seen": 51360312,
      "step": 88525
    },
    {
      "epoch": 13.185880250223414,
      "grad_norm": 8.75,
      "learning_rate": 1.5692589606580866e-05,
      "loss": 1.1354,
      "num_input_tokens_seen": 51363064,
      "step": 88530
    },
    {
      "epoch": 13.186624962764373,
      "grad_norm": 14.3125,
      "learning_rate": 1.568957385204646e-05,
      "loss": 1.0947,
      "num_input_tokens_seen": 51366072,
      "step": 88535
    },
    {
      "epoch": 13.187369675305332,
      "grad_norm": 11.1875,
      "learning_rate": 1.568655825480224e-05,
      "loss": 0.7274,
      "num_input_tokens_seen": 51368856,
      "step": 88540
    },
    {
      "epoch": 13.18811438784629,
      "grad_norm": 9.5625,
      "learning_rate": 1.5683542814899144e-05,
      "loss": 0.8469,
      "num_input_tokens_seen": 51371544,
      "step": 88545
    },
    {
      "epoch": 13.188859100387251,
      "grad_norm": 7.03125,
      "learning_rate": 1.5680527532388133e-05,
      "loss": 1.2427,
      "num_input_tokens_seen": 51374360,
      "step": 88550
    },
    {
      "epoch": 13.18960381292821,
      "grad_norm": 14.875,
      "learning_rate": 1.5677512407320133e-05,
      "loss": 1.2448,
      "num_input_tokens_seen": 51377240,
      "step": 88555
    },
    {
      "epoch": 13.190348525469169,
      "grad_norm": 9.5625,
      "learning_rate": 1.5674497439746088e-05,
      "loss": 0.9147,
      "num_input_tokens_seen": 51379896,
      "step": 88560
    },
    {
      "epoch": 13.191093238010128,
      "grad_norm": 13.625,
      "learning_rate": 1.5671482629716926e-05,
      "loss": 1.0612,
      "num_input_tokens_seen": 51382968,
      "step": 88565
    },
    {
      "epoch": 13.191837950551088,
      "grad_norm": 11.25,
      "learning_rate": 1.566846797728359e-05,
      "loss": 1.3546,
      "num_input_tokens_seen": 51385976,
      "step": 88570
    },
    {
      "epoch": 13.192582663092047,
      "grad_norm": 12.5,
      "learning_rate": 1.5665453482497e-05,
      "loss": 1.1082,
      "num_input_tokens_seen": 51388920,
      "step": 88575
    },
    {
      "epoch": 13.193327375633006,
      "grad_norm": 11.375,
      "learning_rate": 1.5662439145408084e-05,
      "loss": 1.1479,
      "num_input_tokens_seen": 51391640,
      "step": 88580
    },
    {
      "epoch": 13.194072088173964,
      "grad_norm": 16.125,
      "learning_rate": 1.5659424966067775e-05,
      "loss": 0.7986,
      "num_input_tokens_seen": 51394488,
      "step": 88585
    },
    {
      "epoch": 13.194816800714925,
      "grad_norm": 13.8125,
      "learning_rate": 1.5656410944526984e-05,
      "loss": 1.3052,
      "num_input_tokens_seen": 51397304,
      "step": 88590
    },
    {
      "epoch": 13.195561513255884,
      "grad_norm": 5.0,
      "learning_rate": 1.5653397080836633e-05,
      "loss": 1.0243,
      "num_input_tokens_seen": 51400344,
      "step": 88595
    },
    {
      "epoch": 13.196306225796842,
      "grad_norm": 9.625,
      "learning_rate": 1.5650383375047634e-05,
      "loss": 1.2832,
      "num_input_tokens_seen": 51403096,
      "step": 88600
    },
    {
      "epoch": 13.197050938337801,
      "grad_norm": 4.5625,
      "learning_rate": 1.5647369827210917e-05,
      "loss": 1.0417,
      "num_input_tokens_seen": 51405976,
      "step": 88605
    },
    {
      "epoch": 13.19779565087876,
      "grad_norm": 15.9375,
      "learning_rate": 1.5644356437377373e-05,
      "loss": 1.2864,
      "num_input_tokens_seen": 51408792,
      "step": 88610
    },
    {
      "epoch": 13.19854036341972,
      "grad_norm": 6.8125,
      "learning_rate": 1.5641343205597925e-05,
      "loss": 0.8838,
      "num_input_tokens_seen": 51411800,
      "step": 88615
    },
    {
      "epoch": 13.19928507596068,
      "grad_norm": 16.5,
      "learning_rate": 1.5638330131923465e-05,
      "loss": 1.2944,
      "num_input_tokens_seen": 51414712,
      "step": 88620
    },
    {
      "epoch": 13.200029788501638,
      "grad_norm": 8.1875,
      "learning_rate": 1.5635317216404906e-05,
      "loss": 0.9949,
      "num_input_tokens_seen": 51417592,
      "step": 88625
    },
    {
      "epoch": 13.200774501042597,
      "grad_norm": 9.75,
      "learning_rate": 1.5632304459093145e-05,
      "loss": 1.3226,
      "num_input_tokens_seen": 51420568,
      "step": 88630
    },
    {
      "epoch": 13.201519213583557,
      "grad_norm": 11.9375,
      "learning_rate": 1.562929186003907e-05,
      "loss": 0.9798,
      "num_input_tokens_seen": 51423096,
      "step": 88635
    },
    {
      "epoch": 13.202263926124516,
      "grad_norm": 11.0,
      "learning_rate": 1.56262794192936e-05,
      "loss": 1.1048,
      "num_input_tokens_seen": 51425816,
      "step": 88640
    },
    {
      "epoch": 13.203008638665475,
      "grad_norm": 4.875,
      "learning_rate": 1.5623267136907602e-05,
      "loss": 0.9052,
      "num_input_tokens_seen": 51428760,
      "step": 88645
    },
    {
      "epoch": 13.203753351206434,
      "grad_norm": 9.4375,
      "learning_rate": 1.5620255012931984e-05,
      "loss": 1.194,
      "num_input_tokens_seen": 51431832,
      "step": 88650
    },
    {
      "epoch": 13.204498063747394,
      "grad_norm": 9.8125,
      "learning_rate": 1.5617243047417614e-05,
      "loss": 0.9594,
      "num_input_tokens_seen": 51434872,
      "step": 88655
    },
    {
      "epoch": 13.205242776288353,
      "grad_norm": 6.25,
      "learning_rate": 1.5614231240415393e-05,
      "loss": 1.145,
      "num_input_tokens_seen": 51437752,
      "step": 88660
    },
    {
      "epoch": 13.205987488829312,
      "grad_norm": 7.75,
      "learning_rate": 1.5611219591976198e-05,
      "loss": 0.7738,
      "num_input_tokens_seen": 51440536,
      "step": 88665
    },
    {
      "epoch": 13.20673220137027,
      "grad_norm": 12.625,
      "learning_rate": 1.5608208102150895e-05,
      "loss": 1.0973,
      "num_input_tokens_seen": 51443736,
      "step": 88670
    },
    {
      "epoch": 13.207476913911231,
      "grad_norm": 6.5625,
      "learning_rate": 1.560519677099038e-05,
      "loss": 0.9092,
      "num_input_tokens_seen": 51446808,
      "step": 88675
    },
    {
      "epoch": 13.20822162645219,
      "grad_norm": 13.5625,
      "learning_rate": 1.5602185598545515e-05,
      "loss": 1.1601,
      "num_input_tokens_seen": 51449880,
      "step": 88680
    },
    {
      "epoch": 13.208966338993148,
      "grad_norm": 12.125,
      "learning_rate": 1.5599174584867177e-05,
      "loss": 1.2994,
      "num_input_tokens_seen": 51452696,
      "step": 88685
    },
    {
      "epoch": 13.209711051534107,
      "grad_norm": 8.9375,
      "learning_rate": 1.5596163730006218e-05,
      "loss": 0.9066,
      "num_input_tokens_seen": 51455672,
      "step": 88690
    },
    {
      "epoch": 13.210455764075068,
      "grad_norm": 6.34375,
      "learning_rate": 1.5593153034013526e-05,
      "loss": 0.9958,
      "num_input_tokens_seen": 51458424,
      "step": 88695
    },
    {
      "epoch": 13.211200476616026,
      "grad_norm": 10.375,
      "learning_rate": 1.5590142496939956e-05,
      "loss": 0.9986,
      "num_input_tokens_seen": 51461528,
      "step": 88700
    },
    {
      "epoch": 13.211945189156985,
      "grad_norm": 16.25,
      "learning_rate": 1.558713211883635e-05,
      "loss": 1.0906,
      "num_input_tokens_seen": 51464248,
      "step": 88705
    },
    {
      "epoch": 13.212689901697944,
      "grad_norm": 5.875,
      "learning_rate": 1.5584121899753595e-05,
      "loss": 1.218,
      "num_input_tokens_seen": 51467128,
      "step": 88710
    },
    {
      "epoch": 13.213434614238905,
      "grad_norm": 8.8125,
      "learning_rate": 1.558111183974252e-05,
      "loss": 1.0064,
      "num_input_tokens_seen": 51470200,
      "step": 88715
    },
    {
      "epoch": 13.214179326779863,
      "grad_norm": 10.6875,
      "learning_rate": 1.5578101938853994e-05,
      "loss": 0.9729,
      "num_input_tokens_seen": 51473208,
      "step": 88720
    },
    {
      "epoch": 13.214924039320822,
      "grad_norm": 6.96875,
      "learning_rate": 1.5575092197138852e-05,
      "loss": 1.0475,
      "num_input_tokens_seen": 51475864,
      "step": 88725
    },
    {
      "epoch": 13.21566875186178,
      "grad_norm": 20.5,
      "learning_rate": 1.557208261464796e-05,
      "loss": 1.2475,
      "num_input_tokens_seen": 51478552,
      "step": 88730
    },
    {
      "epoch": 13.216413464402741,
      "grad_norm": 7.0625,
      "learning_rate": 1.556907319143214e-05,
      "loss": 0.9931,
      "num_input_tokens_seen": 51481400,
      "step": 88735
    },
    {
      "epoch": 13.2171581769437,
      "grad_norm": 14.9375,
      "learning_rate": 1.5566063927542245e-05,
      "loss": 1.0236,
      "num_input_tokens_seen": 51483960,
      "step": 88740
    },
    {
      "epoch": 13.217902889484659,
      "grad_norm": 4.90625,
      "learning_rate": 1.5563054823029122e-05,
      "loss": 0.9135,
      "num_input_tokens_seen": 51486904,
      "step": 88745
    },
    {
      "epoch": 13.218647602025618,
      "grad_norm": 10.375,
      "learning_rate": 1.5560045877943585e-05,
      "loss": 1.0356,
      "num_input_tokens_seen": 51489592,
      "step": 88750
    },
    {
      "epoch": 13.219392314566578,
      "grad_norm": 10.625,
      "learning_rate": 1.5557037092336486e-05,
      "loss": 1.2471,
      "num_input_tokens_seen": 51492632,
      "step": 88755
    },
    {
      "epoch": 13.220137027107537,
      "grad_norm": 17.0,
      "learning_rate": 1.5554028466258634e-05,
      "loss": 1.1949,
      "num_input_tokens_seen": 51495576,
      "step": 88760
    },
    {
      "epoch": 13.220881739648496,
      "grad_norm": 6.78125,
      "learning_rate": 1.5551019999760885e-05,
      "loss": 0.9019,
      "num_input_tokens_seen": 51498424,
      "step": 88765
    },
    {
      "epoch": 13.221626452189454,
      "grad_norm": 9.6875,
      "learning_rate": 1.554801169289404e-05,
      "loss": 1.323,
      "num_input_tokens_seen": 51501656,
      "step": 88770
    },
    {
      "epoch": 13.222371164730415,
      "grad_norm": 10.3125,
      "learning_rate": 1.554500354570894e-05,
      "loss": 0.8953,
      "num_input_tokens_seen": 51504728,
      "step": 88775
    },
    {
      "epoch": 13.223115877271374,
      "grad_norm": 8.25,
      "learning_rate": 1.5541995558256394e-05,
      "loss": 1.0879,
      "num_input_tokens_seen": 51507992,
      "step": 88780
    },
    {
      "epoch": 13.223860589812332,
      "grad_norm": 14.125,
      "learning_rate": 1.5538987730587217e-05,
      "loss": 1.1329,
      "num_input_tokens_seen": 51510808,
      "step": 88785
    },
    {
      "epoch": 13.224605302353291,
      "grad_norm": 11.0625,
      "learning_rate": 1.553598006275223e-05,
      "loss": 1.0063,
      "num_input_tokens_seen": 51513848,
      "step": 88790
    },
    {
      "epoch": 13.22535001489425,
      "grad_norm": 8.5,
      "learning_rate": 1.5532972554802232e-05,
      "loss": 0.8355,
      "num_input_tokens_seen": 51516568,
      "step": 88795
    },
    {
      "epoch": 13.22609472743521,
      "grad_norm": 11.75,
      "learning_rate": 1.552996520678805e-05,
      "loss": 1.1975,
      "num_input_tokens_seen": 51519448,
      "step": 88800
    },
    {
      "epoch": 13.22683943997617,
      "grad_norm": 7.875,
      "learning_rate": 1.5526958018760473e-05,
      "loss": 1.1076,
      "num_input_tokens_seen": 51522328,
      "step": 88805
    },
    {
      "epoch": 13.227584152517128,
      "grad_norm": 5.53125,
      "learning_rate": 1.552395099077032e-05,
      "loss": 0.953,
      "num_input_tokens_seen": 51525464,
      "step": 88810
    },
    {
      "epoch": 13.228328865058087,
      "grad_norm": 4.15625,
      "learning_rate": 1.552094412286838e-05,
      "loss": 0.9321,
      "num_input_tokens_seen": 51528344,
      "step": 88815
    },
    {
      "epoch": 13.229073577599047,
      "grad_norm": 11.5625,
      "learning_rate": 1.551793741510546e-05,
      "loss": 1.498,
      "num_input_tokens_seen": 51531384,
      "step": 88820
    },
    {
      "epoch": 13.229818290140006,
      "grad_norm": 10.9375,
      "learning_rate": 1.5514930867532352e-05,
      "loss": 1.1917,
      "num_input_tokens_seen": 51534264,
      "step": 88825
    },
    {
      "epoch": 13.230563002680965,
      "grad_norm": 5.9375,
      "learning_rate": 1.5511924480199836e-05,
      "loss": 1.425,
      "num_input_tokens_seen": 51537048,
      "step": 88830
    },
    {
      "epoch": 13.231307715221924,
      "grad_norm": 16.0,
      "learning_rate": 1.550891825315872e-05,
      "loss": 1.1899,
      "num_input_tokens_seen": 51539992,
      "step": 88835
    },
    {
      "epoch": 13.232052427762884,
      "grad_norm": 13.0,
      "learning_rate": 1.5505912186459775e-05,
      "loss": 1.1323,
      "num_input_tokens_seen": 51543032,
      "step": 88840
    },
    {
      "epoch": 13.232797140303843,
      "grad_norm": 11.75,
      "learning_rate": 1.5502906280153806e-05,
      "loss": 1.1603,
      "num_input_tokens_seen": 51546104,
      "step": 88845
    },
    {
      "epoch": 13.233541852844802,
      "grad_norm": 11.5625,
      "learning_rate": 1.5499900534291575e-05,
      "loss": 1.0637,
      "num_input_tokens_seen": 51549144,
      "step": 88850
    },
    {
      "epoch": 13.23428656538576,
      "grad_norm": 9.1875,
      "learning_rate": 1.5496894948923873e-05,
      "loss": 1.1383,
      "num_input_tokens_seen": 51552056,
      "step": 88855
    },
    {
      "epoch": 13.235031277926721,
      "grad_norm": 7.125,
      "learning_rate": 1.5493889524101467e-05,
      "loss": 1.3143,
      "num_input_tokens_seen": 51554872,
      "step": 88860
    },
    {
      "epoch": 13.23577599046768,
      "grad_norm": 10.25,
      "learning_rate": 1.5490884259875143e-05,
      "loss": 0.8452,
      "num_input_tokens_seen": 51557912,
      "step": 88865
    },
    {
      "epoch": 13.236520703008638,
      "grad_norm": 7.53125,
      "learning_rate": 1.5487879156295665e-05,
      "loss": 1.0035,
      "num_input_tokens_seen": 51561048,
      "step": 88870
    },
    {
      "epoch": 13.237265415549597,
      "grad_norm": 6.8125,
      "learning_rate": 1.548487421341379e-05,
      "loss": 1.083,
      "num_input_tokens_seen": 51564152,
      "step": 88875
    },
    {
      "epoch": 13.238010128090558,
      "grad_norm": 16.75,
      "learning_rate": 1.548186943128031e-05,
      "loss": 1.1093,
      "num_input_tokens_seen": 51566872,
      "step": 88880
    },
    {
      "epoch": 13.238754840631517,
      "grad_norm": 10.5,
      "learning_rate": 1.5478864809945965e-05,
      "loss": 0.9225,
      "num_input_tokens_seen": 51569624,
      "step": 88885
    },
    {
      "epoch": 13.239499553172475,
      "grad_norm": 12.375,
      "learning_rate": 1.5475860349461524e-05,
      "loss": 1.0264,
      "num_input_tokens_seen": 51572568,
      "step": 88890
    },
    {
      "epoch": 13.240244265713434,
      "grad_norm": 17.0,
      "learning_rate": 1.5472856049877733e-05,
      "loss": 1.0835,
      "num_input_tokens_seen": 51575416,
      "step": 88895
    },
    {
      "epoch": 13.240988978254395,
      "grad_norm": 19.75,
      "learning_rate": 1.5469851911245368e-05,
      "loss": 1.1385,
      "num_input_tokens_seen": 51578328,
      "step": 88900
    },
    {
      "epoch": 13.241733690795353,
      "grad_norm": 14.4375,
      "learning_rate": 1.5466847933615165e-05,
      "loss": 1.3773,
      "num_input_tokens_seen": 51581048,
      "step": 88905
    },
    {
      "epoch": 13.242478403336312,
      "grad_norm": 5.5,
      "learning_rate": 1.5463844117037872e-05,
      "loss": 0.8207,
      "num_input_tokens_seen": 51583672,
      "step": 88910
    },
    {
      "epoch": 13.24322311587727,
      "grad_norm": 20.5,
      "learning_rate": 1.5460840461564247e-05,
      "loss": 1.272,
      "num_input_tokens_seen": 51586744,
      "step": 88915
    },
    {
      "epoch": 13.243967828418231,
      "grad_norm": 8.25,
      "learning_rate": 1.5457836967245027e-05,
      "loss": 0.8519,
      "num_input_tokens_seen": 51589368,
      "step": 88920
    },
    {
      "epoch": 13.24471254095919,
      "grad_norm": 6.25,
      "learning_rate": 1.5454833634130955e-05,
      "loss": 1.0083,
      "num_input_tokens_seen": 51592472,
      "step": 88925
    },
    {
      "epoch": 13.245457253500149,
      "grad_norm": 4.625,
      "learning_rate": 1.5451830462272753e-05,
      "loss": 1.3156,
      "num_input_tokens_seen": 51595192,
      "step": 88930
    },
    {
      "epoch": 13.246201966041108,
      "grad_norm": 7.34375,
      "learning_rate": 1.5448827451721188e-05,
      "loss": 1.1594,
      "num_input_tokens_seen": 51598616,
      "step": 88935
    },
    {
      "epoch": 13.246946678582066,
      "grad_norm": 11.25,
      "learning_rate": 1.5445824602526966e-05,
      "loss": 0.8627,
      "num_input_tokens_seen": 51601528,
      "step": 88940
    },
    {
      "epoch": 13.247691391123027,
      "grad_norm": 10.375,
      "learning_rate": 1.5442821914740836e-05,
      "loss": 1.1552,
      "num_input_tokens_seen": 51604728,
      "step": 88945
    },
    {
      "epoch": 13.248436103663986,
      "grad_norm": 15.0625,
      "learning_rate": 1.543981938841351e-05,
      "loss": 1.0932,
      "num_input_tokens_seen": 51607640,
      "step": 88950
    },
    {
      "epoch": 13.249180816204944,
      "grad_norm": 12.875,
      "learning_rate": 1.5436817023595716e-05,
      "loss": 1.0251,
      "num_input_tokens_seen": 51610776,
      "step": 88955
    },
    {
      "epoch": 13.249925528745903,
      "grad_norm": 18.125,
      "learning_rate": 1.543381482033819e-05,
      "loss": 0.9532,
      "num_input_tokens_seen": 51613912,
      "step": 88960
    },
    {
      "epoch": 13.250670241286864,
      "grad_norm": 14.25,
      "learning_rate": 1.5430812778691626e-05,
      "loss": 1.0795,
      "num_input_tokens_seen": 51616568,
      "step": 88965
    },
    {
      "epoch": 13.251414953827823,
      "grad_norm": 9.0625,
      "learning_rate": 1.5427810898706764e-05,
      "loss": 1.3343,
      "num_input_tokens_seen": 51619480,
      "step": 88970
    },
    {
      "epoch": 13.252159666368781,
      "grad_norm": 13.125,
      "learning_rate": 1.54248091804343e-05,
      "loss": 1.1481,
      "num_input_tokens_seen": 51622520,
      "step": 88975
    },
    {
      "epoch": 13.25290437890974,
      "grad_norm": 12.1875,
      "learning_rate": 1.5421807623924968e-05,
      "loss": 1.371,
      "num_input_tokens_seen": 51625656,
      "step": 88980
    },
    {
      "epoch": 13.2536490914507,
      "grad_norm": 6.65625,
      "learning_rate": 1.5418806229229452e-05,
      "loss": 0.7249,
      "num_input_tokens_seen": 51628344,
      "step": 88985
    },
    {
      "epoch": 13.25439380399166,
      "grad_norm": 6.625,
      "learning_rate": 1.541580499639846e-05,
      "loss": 0.8302,
      "num_input_tokens_seen": 51631480,
      "step": 88990
    },
    {
      "epoch": 13.255138516532618,
      "grad_norm": 9.8125,
      "learning_rate": 1.541280392548271e-05,
      "loss": 0.8678,
      "num_input_tokens_seen": 51634424,
      "step": 88995
    },
    {
      "epoch": 13.255883229073577,
      "grad_norm": 11.25,
      "learning_rate": 1.5409803016532888e-05,
      "loss": 1.192,
      "num_input_tokens_seen": 51637176,
      "step": 89000
    },
    {
      "epoch": 13.256627941614537,
      "grad_norm": 5.625,
      "learning_rate": 1.5406802269599703e-05,
      "loss": 1.1211,
      "num_input_tokens_seen": 51640088,
      "step": 89005
    },
    {
      "epoch": 13.257372654155496,
      "grad_norm": 11.5,
      "learning_rate": 1.540380168473384e-05,
      "loss": 0.8638,
      "num_input_tokens_seen": 51642872,
      "step": 89010
    },
    {
      "epoch": 13.258117366696455,
      "grad_norm": 18.125,
      "learning_rate": 1.5400801261986e-05,
      "loss": 1.1988,
      "num_input_tokens_seen": 51645560,
      "step": 89015
    },
    {
      "epoch": 13.258862079237414,
      "grad_norm": 8.8125,
      "learning_rate": 1.5397801001406857e-05,
      "loss": 1.3256,
      "num_input_tokens_seen": 51648216,
      "step": 89020
    },
    {
      "epoch": 13.259606791778374,
      "grad_norm": 12.0,
      "learning_rate": 1.5394800903047114e-05,
      "loss": 1.4078,
      "num_input_tokens_seen": 51651320,
      "step": 89025
    },
    {
      "epoch": 13.260351504319333,
      "grad_norm": 11.0,
      "learning_rate": 1.5391800966957448e-05,
      "loss": 1.3342,
      "num_input_tokens_seen": 51654072,
      "step": 89030
    },
    {
      "epoch": 13.261096216860292,
      "grad_norm": 9.6875,
      "learning_rate": 1.538880119318853e-05,
      "loss": 1.1055,
      "num_input_tokens_seen": 51657336,
      "step": 89035
    },
    {
      "epoch": 13.26184092940125,
      "grad_norm": 10.0,
      "learning_rate": 1.538580158179106e-05,
      "loss": 1.127,
      "num_input_tokens_seen": 51660248,
      "step": 89040
    },
    {
      "epoch": 13.262585641942211,
      "grad_norm": 17.0,
      "learning_rate": 1.5382802132815694e-05,
      "loss": 0.9423,
      "num_input_tokens_seen": 51663160,
      "step": 89045
    },
    {
      "epoch": 13.26333035448317,
      "grad_norm": 11.375,
      "learning_rate": 1.5379802846313115e-05,
      "loss": 0.8444,
      "num_input_tokens_seen": 51665944,
      "step": 89050
    },
    {
      "epoch": 13.264075067024129,
      "grad_norm": 19.25,
      "learning_rate": 1.5376803722333983e-05,
      "loss": 1.4704,
      "num_input_tokens_seen": 51669016,
      "step": 89055
    },
    {
      "epoch": 13.264819779565087,
      "grad_norm": 11.375,
      "learning_rate": 1.5373804760928978e-05,
      "loss": 1.0193,
      "num_input_tokens_seen": 51671960,
      "step": 89060
    },
    {
      "epoch": 13.265564492106048,
      "grad_norm": 10.625,
      "learning_rate": 1.537080596214876e-05,
      "loss": 1.1695,
      "num_input_tokens_seen": 51674776,
      "step": 89065
    },
    {
      "epoch": 13.266309204647007,
      "grad_norm": 12.8125,
      "learning_rate": 1.5367807326043976e-05,
      "loss": 1.2902,
      "num_input_tokens_seen": 51677624,
      "step": 89070
    },
    {
      "epoch": 13.267053917187965,
      "grad_norm": 23.0,
      "learning_rate": 1.5364808852665307e-05,
      "loss": 1.11,
      "num_input_tokens_seen": 51680376,
      "step": 89075
    },
    {
      "epoch": 13.267798629728924,
      "grad_norm": 5.59375,
      "learning_rate": 1.53618105420634e-05,
      "loss": 0.7758,
      "num_input_tokens_seen": 51683128,
      "step": 89080
    },
    {
      "epoch": 13.268543342269885,
      "grad_norm": 7.875,
      "learning_rate": 1.5358812394288906e-05,
      "loss": 1.1172,
      "num_input_tokens_seen": 51686040,
      "step": 89085
    },
    {
      "epoch": 13.269288054810843,
      "grad_norm": 15.1875,
      "learning_rate": 1.5355814409392475e-05,
      "loss": 1.0783,
      "num_input_tokens_seen": 51688920,
      "step": 89090
    },
    {
      "epoch": 13.270032767351802,
      "grad_norm": 11.75,
      "learning_rate": 1.5352816587424762e-05,
      "loss": 1.1826,
      "num_input_tokens_seen": 51691896,
      "step": 89095
    },
    {
      "epoch": 13.270777479892761,
      "grad_norm": 11.125,
      "learning_rate": 1.53498189284364e-05,
      "loss": 1.2857,
      "num_input_tokens_seen": 51694776,
      "step": 89100
    },
    {
      "epoch": 13.271522192433721,
      "grad_norm": 7.71875,
      "learning_rate": 1.534682143247805e-05,
      "loss": 0.8476,
      "num_input_tokens_seen": 51697560,
      "step": 89105
    },
    {
      "epoch": 13.27226690497468,
      "grad_norm": 12.6875,
      "learning_rate": 1.534382409960034e-05,
      "loss": 1.195,
      "num_input_tokens_seen": 51700344,
      "step": 89110
    },
    {
      "epoch": 13.273011617515639,
      "grad_norm": 11.4375,
      "learning_rate": 1.5340826929853903e-05,
      "loss": 0.9557,
      "num_input_tokens_seen": 51702904,
      "step": 89115
    },
    {
      "epoch": 13.273756330056598,
      "grad_norm": 11.0,
      "learning_rate": 1.5337829923289382e-05,
      "loss": 0.9886,
      "num_input_tokens_seen": 51705496,
      "step": 89120
    },
    {
      "epoch": 13.274501042597556,
      "grad_norm": 13.5625,
      "learning_rate": 1.5334833079957394e-05,
      "loss": 1.2087,
      "num_input_tokens_seen": 51708344,
      "step": 89125
    },
    {
      "epoch": 13.275245755138517,
      "grad_norm": 9.0,
      "learning_rate": 1.5331836399908588e-05,
      "loss": 0.7839,
      "num_input_tokens_seen": 51711032,
      "step": 89130
    },
    {
      "epoch": 13.275990467679476,
      "grad_norm": 10.0,
      "learning_rate": 1.5328839883193575e-05,
      "loss": 1.0379,
      "num_input_tokens_seen": 51713976,
      "step": 89135
    },
    {
      "epoch": 13.276735180220435,
      "grad_norm": 7.21875,
      "learning_rate": 1.5325843529862987e-05,
      "loss": 0.9233,
      "num_input_tokens_seen": 51716824,
      "step": 89140
    },
    {
      "epoch": 13.277479892761393,
      "grad_norm": 8.6875,
      "learning_rate": 1.532284733996744e-05,
      "loss": 1.01,
      "num_input_tokens_seen": 51719736,
      "step": 89145
    },
    {
      "epoch": 13.278224605302354,
      "grad_norm": 7.9375,
      "learning_rate": 1.5319851313557548e-05,
      "loss": 0.9566,
      "num_input_tokens_seen": 51722712,
      "step": 89150
    },
    {
      "epoch": 13.278969317843313,
      "grad_norm": 4.53125,
      "learning_rate": 1.5316855450683937e-05,
      "loss": 1.1225,
      "num_input_tokens_seen": 51725784,
      "step": 89155
    },
    {
      "epoch": 13.279714030384271,
      "grad_norm": 6.9375,
      "learning_rate": 1.53138597513972e-05,
      "loss": 1.1119,
      "num_input_tokens_seen": 51728728,
      "step": 89160
    },
    {
      "epoch": 13.28045874292523,
      "grad_norm": 9.125,
      "learning_rate": 1.5310864215747966e-05,
      "loss": 1.0866,
      "num_input_tokens_seen": 51731576,
      "step": 89165
    },
    {
      "epoch": 13.28120345546619,
      "grad_norm": 15.1875,
      "learning_rate": 1.5307868843786828e-05,
      "loss": 1.0331,
      "num_input_tokens_seen": 51734616,
      "step": 89170
    },
    {
      "epoch": 13.28194816800715,
      "grad_norm": 11.3125,
      "learning_rate": 1.53048736355644e-05,
      "loss": 1.11,
      "num_input_tokens_seen": 51737624,
      "step": 89175
    },
    {
      "epoch": 13.282692880548108,
      "grad_norm": 12.0,
      "learning_rate": 1.5301878591131273e-05,
      "loss": 1.2003,
      "num_input_tokens_seen": 51740568,
      "step": 89180
    },
    {
      "epoch": 13.283437593089067,
      "grad_norm": 12.1875,
      "learning_rate": 1.529888371053806e-05,
      "loss": 1.3043,
      "num_input_tokens_seen": 51743384,
      "step": 89185
    },
    {
      "epoch": 13.284182305630027,
      "grad_norm": 7.5,
      "learning_rate": 1.5295888993835345e-05,
      "loss": 1.0345,
      "num_input_tokens_seen": 51746328,
      "step": 89190
    },
    {
      "epoch": 13.284927018170986,
      "grad_norm": 6.75,
      "learning_rate": 1.5292894441073712e-05,
      "loss": 1.0046,
      "num_input_tokens_seen": 51749720,
      "step": 89195
    },
    {
      "epoch": 13.285671730711945,
      "grad_norm": 15.3125,
      "learning_rate": 1.5289900052303774e-05,
      "loss": 1.2207,
      "num_input_tokens_seen": 51752344,
      "step": 89200
    },
    {
      "epoch": 13.286416443252904,
      "grad_norm": 26.5,
      "learning_rate": 1.5286905827576094e-05,
      "loss": 1.3524,
      "num_input_tokens_seen": 51755160,
      "step": 89205
    },
    {
      "epoch": 13.287161155793864,
      "grad_norm": 8.75,
      "learning_rate": 1.5283911766941277e-05,
      "loss": 1.2068,
      "num_input_tokens_seen": 51757912,
      "step": 89210
    },
    {
      "epoch": 13.287905868334823,
      "grad_norm": 11.8125,
      "learning_rate": 1.528091787044989e-05,
      "loss": 1.0037,
      "num_input_tokens_seen": 51760920,
      "step": 89215
    },
    {
      "epoch": 13.288650580875782,
      "grad_norm": 10.5625,
      "learning_rate": 1.5277924138152528e-05,
      "loss": 1.061,
      "num_input_tokens_seen": 51763928,
      "step": 89220
    },
    {
      "epoch": 13.28939529341674,
      "grad_norm": 13.1875,
      "learning_rate": 1.527493057009975e-05,
      "loss": 0.8973,
      "num_input_tokens_seen": 51767032,
      "step": 89225
    },
    {
      "epoch": 13.290140005957701,
      "grad_norm": 8.1875,
      "learning_rate": 1.5271937166342132e-05,
      "loss": 1.2016,
      "num_input_tokens_seen": 51769432,
      "step": 89230
    },
    {
      "epoch": 13.29088471849866,
      "grad_norm": 12.125,
      "learning_rate": 1.526894392693025e-05,
      "loss": 1.253,
      "num_input_tokens_seen": 51772664,
      "step": 89235
    },
    {
      "epoch": 13.291629431039619,
      "grad_norm": 11.6875,
      "learning_rate": 1.5265950851914668e-05,
      "loss": 1.1483,
      "num_input_tokens_seen": 51775576,
      "step": 89240
    },
    {
      "epoch": 13.292374143580577,
      "grad_norm": 12.0625,
      "learning_rate": 1.526295794134596e-05,
      "loss": 1.5674,
      "num_input_tokens_seen": 51778552,
      "step": 89245
    },
    {
      "epoch": 13.293118856121538,
      "grad_norm": 8.25,
      "learning_rate": 1.5259965195274678e-05,
      "loss": 1.0458,
      "num_input_tokens_seen": 51781336,
      "step": 89250
    },
    {
      "epoch": 13.293863568662497,
      "grad_norm": 11.0625,
      "learning_rate": 1.5256972613751386e-05,
      "loss": 0.8943,
      "num_input_tokens_seen": 51784056,
      "step": 89255
    },
    {
      "epoch": 13.294608281203455,
      "grad_norm": 11.0,
      "learning_rate": 1.5253980196826634e-05,
      "loss": 1.257,
      "num_input_tokens_seen": 51787000,
      "step": 89260
    },
    {
      "epoch": 13.295352993744414,
      "grad_norm": 12.9375,
      "learning_rate": 1.5250987944550988e-05,
      "loss": 0.9755,
      "num_input_tokens_seen": 51789880,
      "step": 89265
    },
    {
      "epoch": 13.296097706285375,
      "grad_norm": 19.25,
      "learning_rate": 1.5247995856974995e-05,
      "loss": 1.08,
      "num_input_tokens_seen": 51792984,
      "step": 89270
    },
    {
      "epoch": 13.296842418826333,
      "grad_norm": 14.0,
      "learning_rate": 1.5245003934149194e-05,
      "loss": 1.1586,
      "num_input_tokens_seen": 51795960,
      "step": 89275
    },
    {
      "epoch": 13.297587131367292,
      "grad_norm": 9.4375,
      "learning_rate": 1.5242012176124135e-05,
      "loss": 1.1317,
      "num_input_tokens_seen": 51798872,
      "step": 89280
    },
    {
      "epoch": 13.298331843908251,
      "grad_norm": 11.125,
      "learning_rate": 1.5239020582950364e-05,
      "loss": 0.8205,
      "num_input_tokens_seen": 51801752,
      "step": 89285
    },
    {
      "epoch": 13.299076556449211,
      "grad_norm": 11.875,
      "learning_rate": 1.5236029154678425e-05,
      "loss": 0.8034,
      "num_input_tokens_seen": 51804568,
      "step": 89290
    },
    {
      "epoch": 13.29982126899017,
      "grad_norm": 5.21875,
      "learning_rate": 1.523303789135884e-05,
      "loss": 0.9574,
      "num_input_tokens_seen": 51807224,
      "step": 89295
    },
    {
      "epoch": 13.300565981531129,
      "grad_norm": 8.8125,
      "learning_rate": 1.5230046793042163e-05,
      "loss": 1.0358,
      "num_input_tokens_seen": 51810328,
      "step": 89300
    },
    {
      "epoch": 13.301310694072088,
      "grad_norm": 6.53125,
      "learning_rate": 1.5227055859778917e-05,
      "loss": 1.1557,
      "num_input_tokens_seen": 51813016,
      "step": 89305
    },
    {
      "epoch": 13.302055406613047,
      "grad_norm": 14.875,
      "learning_rate": 1.5224065091619622e-05,
      "loss": 1.1081,
      "num_input_tokens_seen": 51816056,
      "step": 89310
    },
    {
      "epoch": 13.302800119154007,
      "grad_norm": 9.75,
      "learning_rate": 1.5221074488614818e-05,
      "loss": 1.3845,
      "num_input_tokens_seen": 51819192,
      "step": 89315
    },
    {
      "epoch": 13.303544831694966,
      "grad_norm": 5.78125,
      "learning_rate": 1.521808405081501e-05,
      "loss": 1.2182,
      "num_input_tokens_seen": 51821848,
      "step": 89320
    },
    {
      "epoch": 13.304289544235925,
      "grad_norm": 14.25,
      "learning_rate": 1.521509377827074e-05,
      "loss": 1.208,
      "num_input_tokens_seen": 51824504,
      "step": 89325
    },
    {
      "epoch": 13.305034256776883,
      "grad_norm": 13.4375,
      "learning_rate": 1.5212103671032507e-05,
      "loss": 0.8346,
      "num_input_tokens_seen": 51827384,
      "step": 89330
    },
    {
      "epoch": 13.305778969317844,
      "grad_norm": 11.3125,
      "learning_rate": 1.5209113729150845e-05,
      "loss": 1.0403,
      "num_input_tokens_seen": 51830008,
      "step": 89335
    },
    {
      "epoch": 13.306523681858803,
      "grad_norm": 9.375,
      "learning_rate": 1.520612395267625e-05,
      "loss": 1.0353,
      "num_input_tokens_seen": 51832888,
      "step": 89340
    },
    {
      "epoch": 13.307268394399761,
      "grad_norm": 15.75,
      "learning_rate": 1.5203134341659242e-05,
      "loss": 1.0958,
      "num_input_tokens_seen": 51836056,
      "step": 89345
    },
    {
      "epoch": 13.30801310694072,
      "grad_norm": 3.46875,
      "learning_rate": 1.520014489615032e-05,
      "loss": 1.0972,
      "num_input_tokens_seen": 51838744,
      "step": 89350
    },
    {
      "epoch": 13.30875781948168,
      "grad_norm": 12.25,
      "learning_rate": 1.5197155616199982e-05,
      "loss": 1.226,
      "num_input_tokens_seen": 51841528,
      "step": 89355
    },
    {
      "epoch": 13.30950253202264,
      "grad_norm": 9.75,
      "learning_rate": 1.5194166501858747e-05,
      "loss": 0.9132,
      "num_input_tokens_seen": 51844568,
      "step": 89360
    },
    {
      "epoch": 13.310247244563598,
      "grad_norm": 10.5625,
      "learning_rate": 1.5191177553177094e-05,
      "loss": 1.2826,
      "num_input_tokens_seen": 51847832,
      "step": 89365
    },
    {
      "epoch": 13.310991957104557,
      "grad_norm": 9.9375,
      "learning_rate": 1.5188188770205533e-05,
      "loss": 1.1208,
      "num_input_tokens_seen": 51851992,
      "step": 89370
    },
    {
      "epoch": 13.311736669645517,
      "grad_norm": 9.0625,
      "learning_rate": 1.518520015299455e-05,
      "loss": 1.0322,
      "num_input_tokens_seen": 51855192,
      "step": 89375
    },
    {
      "epoch": 13.312481382186476,
      "grad_norm": 7.8125,
      "learning_rate": 1.5182211701594634e-05,
      "loss": 1.5017,
      "num_input_tokens_seen": 51857880,
      "step": 89380
    },
    {
      "epoch": 13.313226094727435,
      "grad_norm": 7.09375,
      "learning_rate": 1.5179223416056268e-05,
      "loss": 1.2623,
      "num_input_tokens_seen": 51861080,
      "step": 89385
    },
    {
      "epoch": 13.313970807268394,
      "grad_norm": 11.25,
      "learning_rate": 1.517623529642995e-05,
      "loss": 0.9973,
      "num_input_tokens_seen": 51863928,
      "step": 89390
    },
    {
      "epoch": 13.314715519809354,
      "grad_norm": 15.625,
      "learning_rate": 1.517324734276615e-05,
      "loss": 1.2308,
      "num_input_tokens_seen": 51866904,
      "step": 89395
    },
    {
      "epoch": 13.315460232350313,
      "grad_norm": 11.1875,
      "learning_rate": 1.5170259555115343e-05,
      "loss": 1.1765,
      "num_input_tokens_seen": 51869784,
      "step": 89400
    },
    {
      "epoch": 13.316204944891272,
      "grad_norm": 9.3125,
      "learning_rate": 1.5167271933528015e-05,
      "loss": 1.4194,
      "num_input_tokens_seen": 51872664,
      "step": 89405
    },
    {
      "epoch": 13.31694965743223,
      "grad_norm": 10.75,
      "learning_rate": 1.5164284478054636e-05,
      "loss": 1.0101,
      "num_input_tokens_seen": 51875640,
      "step": 89410
    },
    {
      "epoch": 13.317694369973191,
      "grad_norm": 11.3125,
      "learning_rate": 1.5161297188745673e-05,
      "loss": 0.8447,
      "num_input_tokens_seen": 51878392,
      "step": 89415
    },
    {
      "epoch": 13.31843908251415,
      "grad_norm": 10.4375,
      "learning_rate": 1.5158310065651588e-05,
      "loss": 1.4126,
      "num_input_tokens_seen": 51881496,
      "step": 89420
    },
    {
      "epoch": 13.319183795055109,
      "grad_norm": 3.765625,
      "learning_rate": 1.5155323108822861e-05,
      "loss": 0.8887,
      "num_input_tokens_seen": 51884440,
      "step": 89425
    },
    {
      "epoch": 13.319928507596067,
      "grad_norm": 19.125,
      "learning_rate": 1.5152336318309942e-05,
      "loss": 1.0793,
      "num_input_tokens_seen": 51887256,
      "step": 89430
    },
    {
      "epoch": 13.320673220137028,
      "grad_norm": 13.0625,
      "learning_rate": 1.5149349694163283e-05,
      "loss": 0.9584,
      "num_input_tokens_seen": 51890136,
      "step": 89435
    },
    {
      "epoch": 13.321417932677987,
      "grad_norm": 12.375,
      "learning_rate": 1.5146363236433362e-05,
      "loss": 0.9937,
      "num_input_tokens_seen": 51893208,
      "step": 89440
    },
    {
      "epoch": 13.322162645218945,
      "grad_norm": 9.25,
      "learning_rate": 1.5143376945170612e-05,
      "loss": 1.3855,
      "num_input_tokens_seen": 51896056,
      "step": 89445
    },
    {
      "epoch": 13.322907357759904,
      "grad_norm": 5.0,
      "learning_rate": 1.5140390820425495e-05,
      "loss": 0.9865,
      "num_input_tokens_seen": 51899096,
      "step": 89450
    },
    {
      "epoch": 13.323652070300863,
      "grad_norm": 13.25,
      "learning_rate": 1.5137404862248447e-05,
      "loss": 1.1125,
      "num_input_tokens_seen": 51901784,
      "step": 89455
    },
    {
      "epoch": 13.324396782841823,
      "grad_norm": 12.5,
      "learning_rate": 1.5134419070689926e-05,
      "loss": 0.9452,
      "num_input_tokens_seen": 51904440,
      "step": 89460
    },
    {
      "epoch": 13.325141495382782,
      "grad_norm": 11.375,
      "learning_rate": 1.5131433445800363e-05,
      "loss": 1.2421,
      "num_input_tokens_seen": 51907192,
      "step": 89465
    },
    {
      "epoch": 13.325886207923741,
      "grad_norm": 14.875,
      "learning_rate": 1.5128447987630207e-05,
      "loss": 1.5143,
      "num_input_tokens_seen": 51910040,
      "step": 89470
    },
    {
      "epoch": 13.3266309204647,
      "grad_norm": 9.625,
      "learning_rate": 1.5125462696229892e-05,
      "loss": 1.1447,
      "num_input_tokens_seen": 51912824,
      "step": 89475
    },
    {
      "epoch": 13.32737563300566,
      "grad_norm": 14.1875,
      "learning_rate": 1.5122477571649846e-05,
      "loss": 0.9966,
      "num_input_tokens_seen": 51915672,
      "step": 89480
    },
    {
      "epoch": 13.328120345546619,
      "grad_norm": 12.5625,
      "learning_rate": 1.5119492613940503e-05,
      "loss": 1.1147,
      "num_input_tokens_seen": 51918744,
      "step": 89485
    },
    {
      "epoch": 13.328865058087578,
      "grad_norm": 14.125,
      "learning_rate": 1.5116507823152282e-05,
      "loss": 1.141,
      "num_input_tokens_seen": 51921432,
      "step": 89490
    },
    {
      "epoch": 13.329609770628537,
      "grad_norm": 15.125,
      "learning_rate": 1.5113523199335624e-05,
      "loss": 1.484,
      "num_input_tokens_seen": 51924248,
      "step": 89495
    },
    {
      "epoch": 13.330354483169497,
      "grad_norm": 9.4375,
      "learning_rate": 1.5110538742540936e-05,
      "loss": 1.2635,
      "num_input_tokens_seen": 51927032,
      "step": 89500
    },
    {
      "epoch": 13.331099195710456,
      "grad_norm": 12.625,
      "learning_rate": 1.5107554452818653e-05,
      "loss": 1.0623,
      "num_input_tokens_seen": 51929656,
      "step": 89505
    },
    {
      "epoch": 13.331843908251415,
      "grad_norm": 18.875,
      "learning_rate": 1.5104570330219187e-05,
      "loss": 1.2195,
      "num_input_tokens_seen": 51932408,
      "step": 89510
    },
    {
      "epoch": 13.332588620792373,
      "grad_norm": 8.5625,
      "learning_rate": 1.510158637479294e-05,
      "loss": 1.3231,
      "num_input_tokens_seen": 51935224,
      "step": 89515
    },
    {
      "epoch": 13.333333333333334,
      "grad_norm": 16.0,
      "learning_rate": 1.5098602586590335e-05,
      "loss": 0.7958,
      "num_input_tokens_seen": 51938168,
      "step": 89520
    },
    {
      "epoch": 13.334078045874293,
      "grad_norm": 10.75,
      "learning_rate": 1.5095618965661767e-05,
      "loss": 1.0823,
      "num_input_tokens_seen": 51941112,
      "step": 89525
    },
    {
      "epoch": 13.334822758415251,
      "grad_norm": 6.25,
      "learning_rate": 1.5092635512057662e-05,
      "loss": 0.9557,
      "num_input_tokens_seen": 51944024,
      "step": 89530
    },
    {
      "epoch": 13.33556747095621,
      "grad_norm": 7.90625,
      "learning_rate": 1.5089652225828399e-05,
      "loss": 1.1443,
      "num_input_tokens_seen": 51946904,
      "step": 89535
    },
    {
      "epoch": 13.33631218349717,
      "grad_norm": 5.46875,
      "learning_rate": 1.5086669107024398e-05,
      "loss": 1.1169,
      "num_input_tokens_seen": 51949592,
      "step": 89540
    },
    {
      "epoch": 13.33705689603813,
      "grad_norm": 5.4375,
      "learning_rate": 1.5083686155696043e-05,
      "loss": 0.6458,
      "num_input_tokens_seen": 51952344,
      "step": 89545
    },
    {
      "epoch": 13.337801608579088,
      "grad_norm": 8.6875,
      "learning_rate": 1.5080703371893737e-05,
      "loss": 0.9174,
      "num_input_tokens_seen": 51955256,
      "step": 89550
    },
    {
      "epoch": 13.338546321120047,
      "grad_norm": 8.9375,
      "learning_rate": 1.5077720755667868e-05,
      "loss": 1.147,
      "num_input_tokens_seen": 51958264,
      "step": 89555
    },
    {
      "epoch": 13.339291033661008,
      "grad_norm": 9.375,
      "learning_rate": 1.5074738307068809e-05,
      "loss": 0.7428,
      "num_input_tokens_seen": 51961112,
      "step": 89560
    },
    {
      "epoch": 13.340035746201966,
      "grad_norm": 4.71875,
      "learning_rate": 1.5071756026146972e-05,
      "loss": 0.887,
      "num_input_tokens_seen": 51964152,
      "step": 89565
    },
    {
      "epoch": 13.340780458742925,
      "grad_norm": 7.1875,
      "learning_rate": 1.506877391295271e-05,
      "loss": 0.828,
      "num_input_tokens_seen": 51967128,
      "step": 89570
    },
    {
      "epoch": 13.341525171283884,
      "grad_norm": 10.625,
      "learning_rate": 1.5065791967536436e-05,
      "loss": 1.2035,
      "num_input_tokens_seen": 51970104,
      "step": 89575
    },
    {
      "epoch": 13.342269883824844,
      "grad_norm": 6.9375,
      "learning_rate": 1.50628101899485e-05,
      "loss": 0.9409,
      "num_input_tokens_seen": 51973304,
      "step": 89580
    },
    {
      "epoch": 13.343014596365803,
      "grad_norm": 7.625,
      "learning_rate": 1.5059828580239296e-05,
      "loss": 0.9757,
      "num_input_tokens_seen": 51976024,
      "step": 89585
    },
    {
      "epoch": 13.343759308906762,
      "grad_norm": 12.0625,
      "learning_rate": 1.5056847138459185e-05,
      "loss": 1.1446,
      "num_input_tokens_seen": 51979128,
      "step": 89590
    },
    {
      "epoch": 13.34450402144772,
      "grad_norm": 13.0625,
      "learning_rate": 1.5053865864658523e-05,
      "loss": 0.9469,
      "num_input_tokens_seen": 51981816,
      "step": 89595
    },
    {
      "epoch": 13.345248733988681,
      "grad_norm": 17.875,
      "learning_rate": 1.5050884758887698e-05,
      "loss": 1.4689,
      "num_input_tokens_seen": 51984984,
      "step": 89600
    },
    {
      "epoch": 13.34599344652964,
      "grad_norm": 9.0,
      "learning_rate": 1.504790382119706e-05,
      "loss": 1.2702,
      "num_input_tokens_seen": 51987768,
      "step": 89605
    },
    {
      "epoch": 13.346738159070599,
      "grad_norm": 9.9375,
      "learning_rate": 1.5044923051636972e-05,
      "loss": 0.96,
      "num_input_tokens_seen": 51990712,
      "step": 89610
    },
    {
      "epoch": 13.347482871611557,
      "grad_norm": 8.375,
      "learning_rate": 1.504194245025779e-05,
      "loss": 0.9049,
      "num_input_tokens_seen": 51993784,
      "step": 89615
    },
    {
      "epoch": 13.348227584152518,
      "grad_norm": 10.5625,
      "learning_rate": 1.5038962017109875e-05,
      "loss": 0.9319,
      "num_input_tokens_seen": 51996536,
      "step": 89620
    },
    {
      "epoch": 13.348972296693477,
      "grad_norm": 11.5625,
      "learning_rate": 1.5035981752243561e-05,
      "loss": 1.0012,
      "num_input_tokens_seen": 51999512,
      "step": 89625
    },
    {
      "epoch": 13.349717009234435,
      "grad_norm": 14.625,
      "learning_rate": 1.5033001655709222e-05,
      "loss": 1.1401,
      "num_input_tokens_seen": 52002136,
      "step": 89630
    },
    {
      "epoch": 13.350461721775394,
      "grad_norm": 16.0,
      "learning_rate": 1.5030021727557189e-05,
      "loss": 1.1632,
      "num_input_tokens_seen": 52004824,
      "step": 89635
    },
    {
      "epoch": 13.351206434316353,
      "grad_norm": 13.0,
      "learning_rate": 1.5027041967837802e-05,
      "loss": 1.2265,
      "num_input_tokens_seen": 52007768,
      "step": 89640
    },
    {
      "epoch": 13.351951146857314,
      "grad_norm": 12.6875,
      "learning_rate": 1.5024062376601406e-05,
      "loss": 0.9771,
      "num_input_tokens_seen": 52010520,
      "step": 89645
    },
    {
      "epoch": 13.352695859398272,
      "grad_norm": 9.8125,
      "learning_rate": 1.502108295389833e-05,
      "loss": 1.1533,
      "num_input_tokens_seen": 52013560,
      "step": 89650
    },
    {
      "epoch": 13.353440571939231,
      "grad_norm": 11.625,
      "learning_rate": 1.5018103699778923e-05,
      "loss": 1.1146,
      "num_input_tokens_seen": 52016728,
      "step": 89655
    },
    {
      "epoch": 13.35418528448019,
      "grad_norm": 10.6875,
      "learning_rate": 1.5015124614293501e-05,
      "loss": 0.9136,
      "num_input_tokens_seen": 52019768,
      "step": 89660
    },
    {
      "epoch": 13.35492999702115,
      "grad_norm": 9.625,
      "learning_rate": 1.5012145697492407e-05,
      "loss": 0.8638,
      "num_input_tokens_seen": 52022392,
      "step": 89665
    },
    {
      "epoch": 13.35567470956211,
      "grad_norm": 7.03125,
      "learning_rate": 1.5009166949425965e-05,
      "loss": 0.9895,
      "num_input_tokens_seen": 52025560,
      "step": 89670
    },
    {
      "epoch": 13.356419422103068,
      "grad_norm": 9.6875,
      "learning_rate": 1.5006188370144486e-05,
      "loss": 1.333,
      "num_input_tokens_seen": 52028344,
      "step": 89675
    },
    {
      "epoch": 13.357164134644027,
      "grad_norm": 17.0,
      "learning_rate": 1.5003209959698302e-05,
      "loss": 0.8307,
      "num_input_tokens_seen": 52031000,
      "step": 89680
    },
    {
      "epoch": 13.357908847184987,
      "grad_norm": 24.125,
      "learning_rate": 1.5000231718137717e-05,
      "loss": 0.9545,
      "num_input_tokens_seen": 52033944,
      "step": 89685
    },
    {
      "epoch": 13.358653559725946,
      "grad_norm": 15.6875,
      "learning_rate": 1.4997253645513063e-05,
      "loss": 1.2597,
      "num_input_tokens_seen": 52036920,
      "step": 89690
    },
    {
      "epoch": 13.359398272266905,
      "grad_norm": 16.375,
      "learning_rate": 1.499427574187463e-05,
      "loss": 1.006,
      "num_input_tokens_seen": 52039864,
      "step": 89695
    },
    {
      "epoch": 13.360142984807863,
      "grad_norm": 6.25,
      "learning_rate": 1.4991298007272753e-05,
      "loss": 1.2177,
      "num_input_tokens_seen": 52042712,
      "step": 89700
    },
    {
      "epoch": 13.360887697348824,
      "grad_norm": 10.6875,
      "learning_rate": 1.4988320441757714e-05,
      "loss": 1.0793,
      "num_input_tokens_seen": 52045688,
      "step": 89705
    },
    {
      "epoch": 13.361632409889783,
      "grad_norm": 9.6875,
      "learning_rate": 1.4985343045379836e-05,
      "loss": 1.1153,
      "num_input_tokens_seen": 52048696,
      "step": 89710
    },
    {
      "epoch": 13.362377122430741,
      "grad_norm": 9.3125,
      "learning_rate": 1.4982365818189407e-05,
      "loss": 1.2575,
      "num_input_tokens_seen": 52051896,
      "step": 89715
    },
    {
      "epoch": 13.3631218349717,
      "grad_norm": 5.71875,
      "learning_rate": 1.4979388760236712e-05,
      "loss": 0.9361,
      "num_input_tokens_seen": 52055128,
      "step": 89720
    },
    {
      "epoch": 13.36386654751266,
      "grad_norm": 7.53125,
      "learning_rate": 1.4976411871572074e-05,
      "loss": 0.8216,
      "num_input_tokens_seen": 52057912,
      "step": 89725
    },
    {
      "epoch": 13.36461126005362,
      "grad_norm": 9.5625,
      "learning_rate": 1.4973435152245757e-05,
      "loss": 1.254,
      "num_input_tokens_seen": 52060696,
      "step": 89730
    },
    {
      "epoch": 13.365355972594578,
      "grad_norm": 6.71875,
      "learning_rate": 1.4970458602308077e-05,
      "loss": 0.8829,
      "num_input_tokens_seen": 52063672,
      "step": 89735
    },
    {
      "epoch": 13.366100685135537,
      "grad_norm": 10.0625,
      "learning_rate": 1.4967482221809299e-05,
      "loss": 1.1554,
      "num_input_tokens_seen": 52066616,
      "step": 89740
    },
    {
      "epoch": 13.366845397676498,
      "grad_norm": 16.875,
      "learning_rate": 1.4964506010799711e-05,
      "loss": 1.3486,
      "num_input_tokens_seen": 52069368,
      "step": 89745
    },
    {
      "epoch": 13.367590110217456,
      "grad_norm": 8.125,
      "learning_rate": 1.4961529969329602e-05,
      "loss": 0.8547,
      "num_input_tokens_seen": 52072088,
      "step": 89750
    },
    {
      "epoch": 13.368334822758415,
      "grad_norm": 15.375,
      "learning_rate": 1.4958554097449228e-05,
      "loss": 0.9913,
      "num_input_tokens_seen": 52075192,
      "step": 89755
    },
    {
      "epoch": 13.369079535299374,
      "grad_norm": 12.6875,
      "learning_rate": 1.4955578395208886e-05,
      "loss": 0.983,
      "num_input_tokens_seen": 52077976,
      "step": 89760
    },
    {
      "epoch": 13.369824247840334,
      "grad_norm": 15.1875,
      "learning_rate": 1.4952602862658832e-05,
      "loss": 1.2209,
      "num_input_tokens_seen": 52080824,
      "step": 89765
    },
    {
      "epoch": 13.370568960381293,
      "grad_norm": 11.8125,
      "learning_rate": 1.494962749984935e-05,
      "loss": 1.1509,
      "num_input_tokens_seen": 52083928,
      "step": 89770
    },
    {
      "epoch": 13.371313672922252,
      "grad_norm": 10.3125,
      "learning_rate": 1.4946652306830688e-05,
      "loss": 1.0881,
      "num_input_tokens_seen": 52086808,
      "step": 89775
    },
    {
      "epoch": 13.37205838546321,
      "grad_norm": 10.875,
      "learning_rate": 1.4943677283653124e-05,
      "loss": 0.986,
      "num_input_tokens_seen": 52089624,
      "step": 89780
    },
    {
      "epoch": 13.372803098004171,
      "grad_norm": 12.625,
      "learning_rate": 1.4940702430366905e-05,
      "loss": 1.0559,
      "num_input_tokens_seen": 52092344,
      "step": 89785
    },
    {
      "epoch": 13.37354781054513,
      "grad_norm": 18.125,
      "learning_rate": 1.4937727747022302e-05,
      "loss": 1.5429,
      "num_input_tokens_seen": 52095448,
      "step": 89790
    },
    {
      "epoch": 13.374292523086089,
      "grad_norm": 6.125,
      "learning_rate": 1.493475323366956e-05,
      "loss": 0.9627,
      "num_input_tokens_seen": 52098520,
      "step": 89795
    },
    {
      "epoch": 13.375037235627047,
      "grad_norm": 18.125,
      "learning_rate": 1.4931778890358924e-05,
      "loss": 1.1644,
      "num_input_tokens_seen": 52101304,
      "step": 89800
    },
    {
      "epoch": 13.375781948168008,
      "grad_norm": 14.6875,
      "learning_rate": 1.492880471714066e-05,
      "loss": 1.2646,
      "num_input_tokens_seen": 52104216,
      "step": 89805
    },
    {
      "epoch": 13.376526660708967,
      "grad_norm": 13.625,
      "learning_rate": 1.4925830714065003e-05,
      "loss": 1.2498,
      "num_input_tokens_seen": 52107192,
      "step": 89810
    },
    {
      "epoch": 13.377271373249926,
      "grad_norm": 13.6875,
      "learning_rate": 1.4922856881182199e-05,
      "loss": 0.846,
      "num_input_tokens_seen": 52109944,
      "step": 89815
    },
    {
      "epoch": 13.378016085790884,
      "grad_norm": 16.125,
      "learning_rate": 1.4919883218542474e-05,
      "loss": 1.3985,
      "num_input_tokens_seen": 52112824,
      "step": 89820
    },
    {
      "epoch": 13.378760798331843,
      "grad_norm": 10.8125,
      "learning_rate": 1.4916909726196093e-05,
      "loss": 0.8291,
      "num_input_tokens_seen": 52115736,
      "step": 89825
    },
    {
      "epoch": 13.379505510872804,
      "grad_norm": 10.5625,
      "learning_rate": 1.4913936404193268e-05,
      "loss": 1.034,
      "num_input_tokens_seen": 52118808,
      "step": 89830
    },
    {
      "epoch": 13.380250223413762,
      "grad_norm": 10.875,
      "learning_rate": 1.4910963252584231e-05,
      "loss": 1.1295,
      "num_input_tokens_seen": 52121560,
      "step": 89835
    },
    {
      "epoch": 13.380994935954721,
      "grad_norm": 9.9375,
      "learning_rate": 1.4907990271419222e-05,
      "loss": 1.0227,
      "num_input_tokens_seen": 52124536,
      "step": 89840
    },
    {
      "epoch": 13.38173964849568,
      "grad_norm": 9.25,
      "learning_rate": 1.4905017460748458e-05,
      "loss": 1.2767,
      "num_input_tokens_seen": 52127128,
      "step": 89845
    },
    {
      "epoch": 13.38248436103664,
      "grad_norm": 7.625,
      "learning_rate": 1.4902044820622168e-05,
      "loss": 0.7587,
      "num_input_tokens_seen": 52130296,
      "step": 89850
    },
    {
      "epoch": 13.3832290735776,
      "grad_norm": 6.71875,
      "learning_rate": 1.489907235109056e-05,
      "loss": 0.871,
      "num_input_tokens_seen": 52133080,
      "step": 89855
    },
    {
      "epoch": 13.383973786118558,
      "grad_norm": 9.0,
      "learning_rate": 1.4896100052203865e-05,
      "loss": 1.1259,
      "num_input_tokens_seen": 52135896,
      "step": 89860
    },
    {
      "epoch": 13.384718498659517,
      "grad_norm": 5.96875,
      "learning_rate": 1.4893127924012281e-05,
      "loss": 1.1245,
      "num_input_tokens_seen": 52138808,
      "step": 89865
    },
    {
      "epoch": 13.385463211200477,
      "grad_norm": 18.125,
      "learning_rate": 1.4890155966566039e-05,
      "loss": 1.5464,
      "num_input_tokens_seen": 52141592,
      "step": 89870
    },
    {
      "epoch": 13.386207923741436,
      "grad_norm": 21.75,
      "learning_rate": 1.4887184179915336e-05,
      "loss": 1.5788,
      "num_input_tokens_seen": 52144248,
      "step": 89875
    },
    {
      "epoch": 13.386952636282395,
      "grad_norm": 6.125,
      "learning_rate": 1.4884212564110379e-05,
      "loss": 0.8432,
      "num_input_tokens_seen": 52146904,
      "step": 89880
    },
    {
      "epoch": 13.387697348823353,
      "grad_norm": 7.875,
      "learning_rate": 1.4881241119201367e-05,
      "loss": 0.9452,
      "num_input_tokens_seen": 52149688,
      "step": 89885
    },
    {
      "epoch": 13.388442061364314,
      "grad_norm": 8.1875,
      "learning_rate": 1.4878269845238496e-05,
      "loss": 1.3155,
      "num_input_tokens_seen": 52152600,
      "step": 89890
    },
    {
      "epoch": 13.389186773905273,
      "grad_norm": 12.9375,
      "learning_rate": 1.4875298742271976e-05,
      "loss": 1.1263,
      "num_input_tokens_seen": 52155480,
      "step": 89895
    },
    {
      "epoch": 13.389931486446232,
      "grad_norm": 18.375,
      "learning_rate": 1.4872327810351986e-05,
      "loss": 1.3784,
      "num_input_tokens_seen": 52158328,
      "step": 89900
    },
    {
      "epoch": 13.39067619898719,
      "grad_norm": 8.875,
      "learning_rate": 1.4869357049528731e-05,
      "loss": 0.9006,
      "num_input_tokens_seen": 52161176,
      "step": 89905
    },
    {
      "epoch": 13.39142091152815,
      "grad_norm": 14.0,
      "learning_rate": 1.4866386459852394e-05,
      "loss": 0.9683,
      "num_input_tokens_seen": 52164216,
      "step": 89910
    },
    {
      "epoch": 13.39216562406911,
      "grad_norm": 12.9375,
      "learning_rate": 1.4863416041373158e-05,
      "loss": 1.3421,
      "num_input_tokens_seen": 52167160,
      "step": 89915
    },
    {
      "epoch": 13.392910336610068,
      "grad_norm": 7.84375,
      "learning_rate": 1.4860445794141204e-05,
      "loss": 1.213,
      "num_input_tokens_seen": 52169848,
      "step": 89920
    },
    {
      "epoch": 13.393655049151027,
      "grad_norm": 11.875,
      "learning_rate": 1.4857475718206706e-05,
      "loss": 1.1248,
      "num_input_tokens_seen": 52172760,
      "step": 89925
    },
    {
      "epoch": 13.394399761691988,
      "grad_norm": 8.375,
      "learning_rate": 1.4854505813619857e-05,
      "loss": 1.1185,
      "num_input_tokens_seen": 52175672,
      "step": 89930
    },
    {
      "epoch": 13.395144474232946,
      "grad_norm": 13.75,
      "learning_rate": 1.4851536080430817e-05,
      "loss": 0.9478,
      "num_input_tokens_seen": 52178968,
      "step": 89935
    },
    {
      "epoch": 13.395889186773905,
      "grad_norm": 6.21875,
      "learning_rate": 1.4848566518689757e-05,
      "loss": 0.8261,
      "num_input_tokens_seen": 52181592,
      "step": 89940
    },
    {
      "epoch": 13.396633899314864,
      "grad_norm": 8.25,
      "learning_rate": 1.4845597128446853e-05,
      "loss": 1.2455,
      "num_input_tokens_seen": 52184408,
      "step": 89945
    },
    {
      "epoch": 13.397378611855824,
      "grad_norm": 10.4375,
      "learning_rate": 1.4842627909752266e-05,
      "loss": 1.2576,
      "num_input_tokens_seen": 52187352,
      "step": 89950
    },
    {
      "epoch": 13.398123324396783,
      "grad_norm": 11.4375,
      "learning_rate": 1.4839658862656158e-05,
      "loss": 1.0891,
      "num_input_tokens_seen": 52190264,
      "step": 89955
    },
    {
      "epoch": 13.398868036937742,
      "grad_norm": 11.875,
      "learning_rate": 1.4836689987208677e-05,
      "loss": 0.9386,
      "num_input_tokens_seen": 52192952,
      "step": 89960
    },
    {
      "epoch": 13.3996127494787,
      "grad_norm": 7.84375,
      "learning_rate": 1.483372128346e-05,
      "loss": 0.7199,
      "num_input_tokens_seen": 52196024,
      "step": 89965
    },
    {
      "epoch": 13.400357462019661,
      "grad_norm": 8.8125,
      "learning_rate": 1.4830752751460264e-05,
      "loss": 1.0225,
      "num_input_tokens_seen": 52198616,
      "step": 89970
    },
    {
      "epoch": 13.40110217456062,
      "grad_norm": 12.875,
      "learning_rate": 1.482778439125963e-05,
      "loss": 1.4273,
      "num_input_tokens_seen": 52201560,
      "step": 89975
    },
    {
      "epoch": 13.401846887101579,
      "grad_norm": 7.15625,
      "learning_rate": 1.482481620290823e-05,
      "loss": 0.9151,
      "num_input_tokens_seen": 52204248,
      "step": 89980
    },
    {
      "epoch": 13.402591599642538,
      "grad_norm": 11.875,
      "learning_rate": 1.4821848186456228e-05,
      "loss": 1.0772,
      "num_input_tokens_seen": 52207320,
      "step": 89985
    },
    {
      "epoch": 13.403336312183498,
      "grad_norm": 8.5625,
      "learning_rate": 1.4818880341953745e-05,
      "loss": 1.1391,
      "num_input_tokens_seen": 52210232,
      "step": 89990
    },
    {
      "epoch": 13.404081024724457,
      "grad_norm": 10.6875,
      "learning_rate": 1.4815912669450943e-05,
      "loss": 1.1361,
      "num_input_tokens_seen": 52213848,
      "step": 89995
    },
    {
      "epoch": 13.404825737265416,
      "grad_norm": 8.75,
      "learning_rate": 1.4812945168997947e-05,
      "loss": 1.3636,
      "num_input_tokens_seen": 52217144,
      "step": 90000
    },
    {
      "epoch": 13.405570449806374,
      "grad_norm": 8.1875,
      "learning_rate": 1.4809977840644881e-05,
      "loss": 0.9553,
      "num_input_tokens_seen": 52219992,
      "step": 90005
    },
    {
      "epoch": 13.406315162347333,
      "grad_norm": 7.40625,
      "learning_rate": 1.4807010684441891e-05,
      "loss": 0.8805,
      "num_input_tokens_seen": 52222872,
      "step": 90010
    },
    {
      "epoch": 13.407059874888294,
      "grad_norm": 17.125,
      "learning_rate": 1.4804043700439083e-05,
      "loss": 1.2286,
      "num_input_tokens_seen": 52225816,
      "step": 90015
    },
    {
      "epoch": 13.407804587429252,
      "grad_norm": 7.96875,
      "learning_rate": 1.4801076888686605e-05,
      "loss": 1.3142,
      "num_input_tokens_seen": 52228952,
      "step": 90020
    },
    {
      "epoch": 13.408549299970211,
      "grad_norm": 7.25,
      "learning_rate": 1.4798110249234556e-05,
      "loss": 0.9883,
      "num_input_tokens_seen": 52231864,
      "step": 90025
    },
    {
      "epoch": 13.40929401251117,
      "grad_norm": 7.25,
      "learning_rate": 1.4795143782133075e-05,
      "loss": 1.0643,
      "num_input_tokens_seen": 52234808,
      "step": 90030
    },
    {
      "epoch": 13.41003872505213,
      "grad_norm": 6.71875,
      "learning_rate": 1.4792177487432271e-05,
      "loss": 1.1807,
      "num_input_tokens_seen": 52237784,
      "step": 90035
    },
    {
      "epoch": 13.41078343759309,
      "grad_norm": 7.71875,
      "learning_rate": 1.4789211365182249e-05,
      "loss": 0.8567,
      "num_input_tokens_seen": 52240856,
      "step": 90040
    },
    {
      "epoch": 13.411528150134048,
      "grad_norm": 10.4375,
      "learning_rate": 1.4786245415433125e-05,
      "loss": 1.2707,
      "num_input_tokens_seen": 52243736,
      "step": 90045
    },
    {
      "epoch": 13.412272862675007,
      "grad_norm": 12.25,
      "learning_rate": 1.4783279638234994e-05,
      "loss": 1.1111,
      "num_input_tokens_seen": 52246488,
      "step": 90050
    },
    {
      "epoch": 13.413017575215967,
      "grad_norm": 6.125,
      "learning_rate": 1.4780314033637982e-05,
      "loss": 1.4635,
      "num_input_tokens_seen": 52249400,
      "step": 90055
    },
    {
      "epoch": 13.413762287756926,
      "grad_norm": 12.25,
      "learning_rate": 1.4777348601692168e-05,
      "loss": 1.0699,
      "num_input_tokens_seen": 52252312,
      "step": 90060
    },
    {
      "epoch": 13.414507000297885,
      "grad_norm": 8.5625,
      "learning_rate": 1.4774383342447667e-05,
      "loss": 1.2091,
      "num_input_tokens_seen": 52255224,
      "step": 90065
    },
    {
      "epoch": 13.415251712838844,
      "grad_norm": 4.125,
      "learning_rate": 1.477141825595456e-05,
      "loss": 0.918,
      "num_input_tokens_seen": 52258008,
      "step": 90070
    },
    {
      "epoch": 13.415996425379804,
      "grad_norm": 16.75,
      "learning_rate": 1.4768453342262955e-05,
      "loss": 1.1487,
      "num_input_tokens_seen": 52261080,
      "step": 90075
    },
    {
      "epoch": 13.416741137920763,
      "grad_norm": 8.375,
      "learning_rate": 1.4765488601422934e-05,
      "loss": 1.2751,
      "num_input_tokens_seen": 52264312,
      "step": 90080
    },
    {
      "epoch": 13.417485850461722,
      "grad_norm": 6.28125,
      "learning_rate": 1.4762524033484565e-05,
      "loss": 0.8981,
      "num_input_tokens_seen": 52266840,
      "step": 90085
    },
    {
      "epoch": 13.41823056300268,
      "grad_norm": 12.1875,
      "learning_rate": 1.475955963849796e-05,
      "loss": 1.1947,
      "num_input_tokens_seen": 52269848,
      "step": 90090
    },
    {
      "epoch": 13.418975275543641,
      "grad_norm": 11.125,
      "learning_rate": 1.4756595416513175e-05,
      "loss": 1.0571,
      "num_input_tokens_seen": 52272760,
      "step": 90095
    },
    {
      "epoch": 13.4197199880846,
      "grad_norm": 10.6875,
      "learning_rate": 1.4753631367580312e-05,
      "loss": 1.2556,
      "num_input_tokens_seen": 52275640,
      "step": 90100
    },
    {
      "epoch": 13.420464700625558,
      "grad_norm": 11.4375,
      "learning_rate": 1.4750667491749428e-05,
      "loss": 1.3103,
      "num_input_tokens_seen": 52278680,
      "step": 90105
    },
    {
      "epoch": 13.421209413166517,
      "grad_norm": 12.0625,
      "learning_rate": 1.4747703789070606e-05,
      "loss": 1.1612,
      "num_input_tokens_seen": 52281336,
      "step": 90110
    },
    {
      "epoch": 13.421954125707478,
      "grad_norm": 25.75,
      "learning_rate": 1.4744740259593907e-05,
      "loss": 1.4108,
      "num_input_tokens_seen": 52284184,
      "step": 90115
    },
    {
      "epoch": 13.422698838248436,
      "grad_norm": 5.03125,
      "learning_rate": 1.4741776903369386e-05,
      "loss": 1.0247,
      "num_input_tokens_seen": 52287448,
      "step": 90120
    },
    {
      "epoch": 13.423443550789395,
      "grad_norm": 9.375,
      "learning_rate": 1.4738813720447132e-05,
      "loss": 1.0294,
      "num_input_tokens_seen": 52290264,
      "step": 90125
    },
    {
      "epoch": 13.424188263330354,
      "grad_norm": 6.78125,
      "learning_rate": 1.473585071087718e-05,
      "loss": 0.959,
      "num_input_tokens_seen": 52292984,
      "step": 90130
    },
    {
      "epoch": 13.424932975871315,
      "grad_norm": 17.5,
      "learning_rate": 1.4732887874709605e-05,
      "loss": 0.7889,
      "num_input_tokens_seen": 52295800,
      "step": 90135
    },
    {
      "epoch": 13.425677688412273,
      "grad_norm": 14.375,
      "learning_rate": 1.4729925211994455e-05,
      "loss": 1.0014,
      "num_input_tokens_seen": 52298776,
      "step": 90140
    },
    {
      "epoch": 13.426422400953232,
      "grad_norm": 10.5,
      "learning_rate": 1.4726962722781783e-05,
      "loss": 0.9589,
      "num_input_tokens_seen": 52301944,
      "step": 90145
    },
    {
      "epoch": 13.42716711349419,
      "grad_norm": 21.75,
      "learning_rate": 1.4724000407121624e-05,
      "loss": 1.2631,
      "num_input_tokens_seen": 52305176,
      "step": 90150
    },
    {
      "epoch": 13.42791182603515,
      "grad_norm": 7.34375,
      "learning_rate": 1.4721038265064044e-05,
      "loss": 0.755,
      "num_input_tokens_seen": 52308056,
      "step": 90155
    },
    {
      "epoch": 13.42865653857611,
      "grad_norm": 10.125,
      "learning_rate": 1.4718076296659078e-05,
      "loss": 1.0652,
      "num_input_tokens_seen": 52310776,
      "step": 90160
    },
    {
      "epoch": 13.429401251117069,
      "grad_norm": 18.875,
      "learning_rate": 1.471511450195675e-05,
      "loss": 1.04,
      "num_input_tokens_seen": 52313624,
      "step": 90165
    },
    {
      "epoch": 13.430145963658028,
      "grad_norm": 11.25,
      "learning_rate": 1.4712152881007118e-05,
      "loss": 1.3267,
      "num_input_tokens_seen": 52316472,
      "step": 90170
    },
    {
      "epoch": 13.430890676198986,
      "grad_norm": 17.875,
      "learning_rate": 1.4709191433860206e-05,
      "loss": 1.0802,
      "num_input_tokens_seen": 52319288,
      "step": 90175
    },
    {
      "epoch": 13.431635388739947,
      "grad_norm": 11.75,
      "learning_rate": 1.4706230160566048e-05,
      "loss": 1.3941,
      "num_input_tokens_seen": 52322360,
      "step": 90180
    },
    {
      "epoch": 13.432380101280906,
      "grad_norm": 6.75,
      "learning_rate": 1.4703269061174663e-05,
      "loss": 1.0218,
      "num_input_tokens_seen": 52325240,
      "step": 90185
    },
    {
      "epoch": 13.433124813821864,
      "grad_norm": 9.8125,
      "learning_rate": 1.4700308135736085e-05,
      "loss": 1.105,
      "num_input_tokens_seen": 52328312,
      "step": 90190
    },
    {
      "epoch": 13.433869526362823,
      "grad_norm": 6.96875,
      "learning_rate": 1.4697347384300338e-05,
      "loss": 0.9154,
      "num_input_tokens_seen": 52331032,
      "step": 90195
    },
    {
      "epoch": 13.434614238903784,
      "grad_norm": 15.5,
      "learning_rate": 1.4694386806917426e-05,
      "loss": 1.2819,
      "num_input_tokens_seen": 52333944,
      "step": 90200
    },
    {
      "epoch": 13.435358951444742,
      "grad_norm": 6.09375,
      "learning_rate": 1.4691426403637382e-05,
      "loss": 1.0173,
      "num_input_tokens_seen": 52337048,
      "step": 90205
    },
    {
      "epoch": 13.436103663985701,
      "grad_norm": 17.125,
      "learning_rate": 1.4688466174510209e-05,
      "loss": 1.1767,
      "num_input_tokens_seen": 52340088,
      "step": 90210
    },
    {
      "epoch": 13.43684837652666,
      "grad_norm": 6.0,
      "learning_rate": 1.4685506119585924e-05,
      "loss": 1.0422,
      "num_input_tokens_seen": 52342968,
      "step": 90215
    },
    {
      "epoch": 13.43759308906762,
      "grad_norm": 7.375,
      "learning_rate": 1.4682546238914521e-05,
      "loss": 0.8888,
      "num_input_tokens_seen": 52345656,
      "step": 90220
    },
    {
      "epoch": 13.43833780160858,
      "grad_norm": 6.375,
      "learning_rate": 1.4679586532546025e-05,
      "loss": 1.0093,
      "num_input_tokens_seen": 52348760,
      "step": 90225
    },
    {
      "epoch": 13.439082514149538,
      "grad_norm": 14.875,
      "learning_rate": 1.467662700053041e-05,
      "loss": 1.1907,
      "num_input_tokens_seen": 52351544,
      "step": 90230
    },
    {
      "epoch": 13.439827226690497,
      "grad_norm": 10.125,
      "learning_rate": 1.4673667642917705e-05,
      "loss": 1.2844,
      "num_input_tokens_seen": 52354424,
      "step": 90235
    },
    {
      "epoch": 13.440571939231457,
      "grad_norm": 10.25,
      "learning_rate": 1.4670708459757885e-05,
      "loss": 1.2513,
      "num_input_tokens_seen": 52357176,
      "step": 90240
    },
    {
      "epoch": 13.441316651772416,
      "grad_norm": 8.5625,
      "learning_rate": 1.4667749451100943e-05,
      "loss": 1.2347,
      "num_input_tokens_seen": 52359960,
      "step": 90245
    },
    {
      "epoch": 13.442061364313375,
      "grad_norm": 7.4375,
      "learning_rate": 1.4664790616996881e-05,
      "loss": 1.0438,
      "num_input_tokens_seen": 52363096,
      "step": 90250
    },
    {
      "epoch": 13.442806076854334,
      "grad_norm": 8.6875,
      "learning_rate": 1.4661831957495665e-05,
      "loss": 1.0117,
      "num_input_tokens_seen": 52365784,
      "step": 90255
    },
    {
      "epoch": 13.443550789395294,
      "grad_norm": 11.9375,
      "learning_rate": 1.4658873472647299e-05,
      "loss": 0.9134,
      "num_input_tokens_seen": 52368760,
      "step": 90260
    },
    {
      "epoch": 13.444295501936253,
      "grad_norm": 6.09375,
      "learning_rate": 1.4655915162501754e-05,
      "loss": 0.9748,
      "num_input_tokens_seen": 52371672,
      "step": 90265
    },
    {
      "epoch": 13.445040214477212,
      "grad_norm": 5.3125,
      "learning_rate": 1.4652957027109009e-05,
      "loss": 0.8188,
      "num_input_tokens_seen": 52374520,
      "step": 90270
    },
    {
      "epoch": 13.44578492701817,
      "grad_norm": 6.15625,
      "learning_rate": 1.4649999066519043e-05,
      "loss": 1.2317,
      "num_input_tokens_seen": 52377528,
      "step": 90275
    },
    {
      "epoch": 13.446529639559131,
      "grad_norm": 9.0625,
      "learning_rate": 1.4647041280781821e-05,
      "loss": 0.9945,
      "num_input_tokens_seen": 52380312,
      "step": 90280
    },
    {
      "epoch": 13.44727435210009,
      "grad_norm": 6.71875,
      "learning_rate": 1.4644083669947314e-05,
      "loss": 0.9875,
      "num_input_tokens_seen": 52383352,
      "step": 90285
    },
    {
      "epoch": 13.448019064641048,
      "grad_norm": 9.4375,
      "learning_rate": 1.464112623406548e-05,
      "loss": 0.9311,
      "num_input_tokens_seen": 52386232,
      "step": 90290
    },
    {
      "epoch": 13.448763777182007,
      "grad_norm": 16.5,
      "learning_rate": 1.46381689731863e-05,
      "loss": 1.4358,
      "num_input_tokens_seen": 52388984,
      "step": 90295
    },
    {
      "epoch": 13.449508489722968,
      "grad_norm": 6.3125,
      "learning_rate": 1.463521188735972e-05,
      "loss": 0.8247,
      "num_input_tokens_seen": 52391640,
      "step": 90300
    },
    {
      "epoch": 13.450253202263927,
      "grad_norm": 14.6875,
      "learning_rate": 1.4632254976635706e-05,
      "loss": 0.9314,
      "num_input_tokens_seen": 52394392,
      "step": 90305
    },
    {
      "epoch": 13.450997914804885,
      "grad_norm": 16.5,
      "learning_rate": 1.4629298241064196e-05,
      "loss": 1.1462,
      "num_input_tokens_seen": 52397080,
      "step": 90310
    },
    {
      "epoch": 13.451742627345844,
      "grad_norm": 9.875,
      "learning_rate": 1.462634168069516e-05,
      "loss": 1.3736,
      "num_input_tokens_seen": 52400248,
      "step": 90315
    },
    {
      "epoch": 13.452487339886805,
      "grad_norm": 8.5625,
      "learning_rate": 1.4623385295578542e-05,
      "loss": 1.1767,
      "num_input_tokens_seen": 52403128,
      "step": 90320
    },
    {
      "epoch": 13.453232052427763,
      "grad_norm": 15.0,
      "learning_rate": 1.462042908576427e-05,
      "loss": 1.2001,
      "num_input_tokens_seen": 52405880,
      "step": 90325
    },
    {
      "epoch": 13.453976764968722,
      "grad_norm": 13.0625,
      "learning_rate": 1.461747305130231e-05,
      "loss": 1.2076,
      "num_input_tokens_seen": 52408600,
      "step": 90330
    },
    {
      "epoch": 13.45472147750968,
      "grad_norm": 9.25,
      "learning_rate": 1.4614517192242588e-05,
      "loss": 1.1357,
      "num_input_tokens_seen": 52411320,
      "step": 90335
    },
    {
      "epoch": 13.45546619005064,
      "grad_norm": 6.90625,
      "learning_rate": 1.4611561508635047e-05,
      "loss": 0.743,
      "num_input_tokens_seen": 52413816,
      "step": 90340
    },
    {
      "epoch": 13.4562109025916,
      "grad_norm": 10.8125,
      "learning_rate": 1.4608606000529601e-05,
      "loss": 0.9499,
      "num_input_tokens_seen": 52416824,
      "step": 90345
    },
    {
      "epoch": 13.456955615132559,
      "grad_norm": 17.0,
      "learning_rate": 1.4605650667976211e-05,
      "loss": 1.1476,
      "num_input_tokens_seen": 52419992,
      "step": 90350
    },
    {
      "epoch": 13.457700327673518,
      "grad_norm": 8.1875,
      "learning_rate": 1.4602695511024785e-05,
      "loss": 1.122,
      "num_input_tokens_seen": 52422776,
      "step": 90355
    },
    {
      "epoch": 13.458445040214476,
      "grad_norm": 15.8125,
      "learning_rate": 1.4599740529725242e-05,
      "loss": 1.016,
      "num_input_tokens_seen": 52426072,
      "step": 90360
    },
    {
      "epoch": 13.459189752755437,
      "grad_norm": 10.6875,
      "learning_rate": 1.4596785724127526e-05,
      "loss": 1.0321,
      "num_input_tokens_seen": 52428632,
      "step": 90365
    },
    {
      "epoch": 13.459934465296396,
      "grad_norm": 15.875,
      "learning_rate": 1.4593831094281529e-05,
      "loss": 1.0321,
      "num_input_tokens_seen": 52431480,
      "step": 90370
    },
    {
      "epoch": 13.460679177837354,
      "grad_norm": 7.15625,
      "learning_rate": 1.4590876640237189e-05,
      "loss": 1.4857,
      "num_input_tokens_seen": 52434424,
      "step": 90375
    },
    {
      "epoch": 13.461423890378313,
      "grad_norm": 9.875,
      "learning_rate": 1.4587922362044399e-05,
      "loss": 1.1212,
      "num_input_tokens_seen": 52437304,
      "step": 90380
    },
    {
      "epoch": 13.462168602919274,
      "grad_norm": 17.5,
      "learning_rate": 1.4584968259753088e-05,
      "loss": 1.5451,
      "num_input_tokens_seen": 52440184,
      "step": 90385
    },
    {
      "epoch": 13.462913315460233,
      "grad_norm": 12.25,
      "learning_rate": 1.4582014333413153e-05,
      "loss": 1.3008,
      "num_input_tokens_seen": 52442968,
      "step": 90390
    },
    {
      "epoch": 13.463658028001191,
      "grad_norm": 5.1875,
      "learning_rate": 1.4579060583074497e-05,
      "loss": 1.0886,
      "num_input_tokens_seen": 52445784,
      "step": 90395
    },
    {
      "epoch": 13.46440274054215,
      "grad_norm": 14.5,
      "learning_rate": 1.4576107008787022e-05,
      "loss": 1.2713,
      "num_input_tokens_seen": 52448856,
      "step": 90400
    },
    {
      "epoch": 13.46514745308311,
      "grad_norm": 11.625,
      "learning_rate": 1.4573153610600615e-05,
      "loss": 1.1234,
      "num_input_tokens_seen": 52451864,
      "step": 90405
    },
    {
      "epoch": 13.46589216562407,
      "grad_norm": 10.1875,
      "learning_rate": 1.4570200388565189e-05,
      "loss": 0.9561,
      "num_input_tokens_seen": 52454840,
      "step": 90410
    },
    {
      "epoch": 13.466636878165028,
      "grad_norm": 5.28125,
      "learning_rate": 1.4567247342730617e-05,
      "loss": 1.0916,
      "num_input_tokens_seen": 52457848,
      "step": 90415
    },
    {
      "epoch": 13.467381590705987,
      "grad_norm": 5.71875,
      "learning_rate": 1.4564294473146808e-05,
      "loss": 0.7423,
      "num_input_tokens_seen": 52460440,
      "step": 90420
    },
    {
      "epoch": 13.468126303246947,
      "grad_norm": 11.4375,
      "learning_rate": 1.456134177986363e-05,
      "loss": 1.4005,
      "num_input_tokens_seen": 52463288,
      "step": 90425
    },
    {
      "epoch": 13.468871015787906,
      "grad_norm": 14.1875,
      "learning_rate": 1.455838926293098e-05,
      "loss": 0.9385,
      "num_input_tokens_seen": 52466104,
      "step": 90430
    },
    {
      "epoch": 13.469615728328865,
      "grad_norm": 22.875,
      "learning_rate": 1.4555436922398732e-05,
      "loss": 0.9775,
      "num_input_tokens_seen": 52469400,
      "step": 90435
    },
    {
      "epoch": 13.470360440869824,
      "grad_norm": 12.5,
      "learning_rate": 1.4552484758316754e-05,
      "loss": 0.8019,
      "num_input_tokens_seen": 52472312,
      "step": 90440
    },
    {
      "epoch": 13.471105153410784,
      "grad_norm": 19.125,
      "learning_rate": 1.4549532770734936e-05,
      "loss": 1.2776,
      "num_input_tokens_seen": 52475064,
      "step": 90445
    },
    {
      "epoch": 13.471849865951743,
      "grad_norm": 13.5,
      "learning_rate": 1.4546580959703138e-05,
      "loss": 0.9834,
      "num_input_tokens_seen": 52477880,
      "step": 90450
    },
    {
      "epoch": 13.472594578492702,
      "grad_norm": 12.375,
      "learning_rate": 1.4543629325271225e-05,
      "loss": 1.14,
      "num_input_tokens_seen": 52480792,
      "step": 90455
    },
    {
      "epoch": 13.47333929103366,
      "grad_norm": 6.8125,
      "learning_rate": 1.4540677867489072e-05,
      "loss": 0.7886,
      "num_input_tokens_seen": 52483704,
      "step": 90460
    },
    {
      "epoch": 13.474084003574621,
      "grad_norm": 8.0625,
      "learning_rate": 1.4537726586406538e-05,
      "loss": 1.1215,
      "num_input_tokens_seen": 52486680,
      "step": 90465
    },
    {
      "epoch": 13.47482871611558,
      "grad_norm": 24.5,
      "learning_rate": 1.453477548207347e-05,
      "loss": 1.2151,
      "num_input_tokens_seen": 52489848,
      "step": 90470
    },
    {
      "epoch": 13.475573428656539,
      "grad_norm": 9.25,
      "learning_rate": 1.4531824554539747e-05,
      "loss": 0.9946,
      "num_input_tokens_seen": 52492600,
      "step": 90475
    },
    {
      "epoch": 13.476318141197497,
      "grad_norm": 8.4375,
      "learning_rate": 1.4528873803855206e-05,
      "loss": 1.0908,
      "num_input_tokens_seen": 52495416,
      "step": 90480
    },
    {
      "epoch": 13.477062853738458,
      "grad_norm": 5.78125,
      "learning_rate": 1.4525923230069689e-05,
      "loss": 0.9118,
      "num_input_tokens_seen": 52498264,
      "step": 90485
    },
    {
      "epoch": 13.477807566279417,
      "grad_norm": 16.0,
      "learning_rate": 1.4522972833233068e-05,
      "loss": 1.215,
      "num_input_tokens_seen": 52501496,
      "step": 90490
    },
    {
      "epoch": 13.478552278820375,
      "grad_norm": 9.375,
      "learning_rate": 1.4520022613395157e-05,
      "loss": 1.2207,
      "num_input_tokens_seen": 52504568,
      "step": 90495
    },
    {
      "epoch": 13.479296991361334,
      "grad_norm": 7.59375,
      "learning_rate": 1.4517072570605824e-05,
      "loss": 1.111,
      "num_input_tokens_seen": 52507608,
      "step": 90500
    },
    {
      "epoch": 13.480041703902295,
      "grad_norm": 9.125,
      "learning_rate": 1.4514122704914887e-05,
      "loss": 1.0299,
      "num_input_tokens_seen": 52510392,
      "step": 90505
    },
    {
      "epoch": 13.480786416443253,
      "grad_norm": 14.9375,
      "learning_rate": 1.4511173016372199e-05,
      "loss": 1.155,
      "num_input_tokens_seen": 52513464,
      "step": 90510
    },
    {
      "epoch": 13.481531128984212,
      "grad_norm": 18.25,
      "learning_rate": 1.4508223505027581e-05,
      "loss": 1.2443,
      "num_input_tokens_seen": 52516536,
      "step": 90515
    },
    {
      "epoch": 13.482275841525171,
      "grad_norm": 13.875,
      "learning_rate": 1.4505274170930866e-05,
      "loss": 1.0366,
      "num_input_tokens_seen": 52519512,
      "step": 90520
    },
    {
      "epoch": 13.48302055406613,
      "grad_norm": 4.78125,
      "learning_rate": 1.450232501413188e-05,
      "loss": 1.0648,
      "num_input_tokens_seen": 52522456,
      "step": 90525
    },
    {
      "epoch": 13.48376526660709,
      "grad_norm": 11.3125,
      "learning_rate": 1.4499376034680429e-05,
      "loss": 1.3943,
      "num_input_tokens_seen": 52525656,
      "step": 90530
    },
    {
      "epoch": 13.484509979148049,
      "grad_norm": 8.0,
      "learning_rate": 1.4496427232626358e-05,
      "loss": 1.1636,
      "num_input_tokens_seen": 52528568,
      "step": 90535
    },
    {
      "epoch": 13.485254691689008,
      "grad_norm": 10.0625,
      "learning_rate": 1.4493478608019461e-05,
      "loss": 1.2259,
      "num_input_tokens_seen": 52531448,
      "step": 90540
    },
    {
      "epoch": 13.485999404229966,
      "grad_norm": 13.875,
      "learning_rate": 1.449053016090958e-05,
      "loss": 0.9977,
      "num_input_tokens_seen": 52534488,
      "step": 90545
    },
    {
      "epoch": 13.486744116770927,
      "grad_norm": 4.125,
      "learning_rate": 1.4487581891346497e-05,
      "loss": 0.9887,
      "num_input_tokens_seen": 52537304,
      "step": 90550
    },
    {
      "epoch": 13.487488829311886,
      "grad_norm": 6.28125,
      "learning_rate": 1.4484633799380046e-05,
      "loss": 1.3457,
      "num_input_tokens_seen": 52539928,
      "step": 90555
    },
    {
      "epoch": 13.488233541852845,
      "grad_norm": 10.5625,
      "learning_rate": 1.4481685885060017e-05,
      "loss": 1.0387,
      "num_input_tokens_seen": 52542712,
      "step": 90560
    },
    {
      "epoch": 13.488978254393803,
      "grad_norm": 4.25,
      "learning_rate": 1.4478738148436205e-05,
      "loss": 0.9857,
      "num_input_tokens_seen": 52545688,
      "step": 90565
    },
    {
      "epoch": 13.489722966934764,
      "grad_norm": 15.0625,
      "learning_rate": 1.447579058955843e-05,
      "loss": 1.4213,
      "num_input_tokens_seen": 52548856,
      "step": 90570
    },
    {
      "epoch": 13.490467679475723,
      "grad_norm": 7.125,
      "learning_rate": 1.4472843208476466e-05,
      "loss": 0.8921,
      "num_input_tokens_seen": 52552216,
      "step": 90575
    },
    {
      "epoch": 13.491212392016681,
      "grad_norm": 9.1875,
      "learning_rate": 1.446989600524013e-05,
      "loss": 1.0101,
      "num_input_tokens_seen": 52555288,
      "step": 90580
    },
    {
      "epoch": 13.49195710455764,
      "grad_norm": 13.125,
      "learning_rate": 1.4466948979899194e-05,
      "loss": 1.1127,
      "num_input_tokens_seen": 52558264,
      "step": 90585
    },
    {
      "epoch": 13.4927018170986,
      "grad_norm": 14.1875,
      "learning_rate": 1.446400213250345e-05,
      "loss": 0.933,
      "num_input_tokens_seen": 52561208,
      "step": 90590
    },
    {
      "epoch": 13.49344652963956,
      "grad_norm": 6.875,
      "learning_rate": 1.4461055463102673e-05,
      "loss": 1.0669,
      "num_input_tokens_seen": 52563896,
      "step": 90595
    },
    {
      "epoch": 13.494191242180518,
      "grad_norm": 9.6875,
      "learning_rate": 1.4458108971746665e-05,
      "loss": 1.1355,
      "num_input_tokens_seen": 52566808,
      "step": 90600
    },
    {
      "epoch": 13.494935954721477,
      "grad_norm": 4.125,
      "learning_rate": 1.4455162658485188e-05,
      "loss": 0.8403,
      "num_input_tokens_seen": 52569624,
      "step": 90605
    },
    {
      "epoch": 13.495680667262437,
      "grad_norm": 16.25,
      "learning_rate": 1.4452216523368011e-05,
      "loss": 1.2295,
      "num_input_tokens_seen": 52573016,
      "step": 90610
    },
    {
      "epoch": 13.496425379803396,
      "grad_norm": 11.4375,
      "learning_rate": 1.4449270566444929e-05,
      "loss": 1.1404,
      "num_input_tokens_seen": 52575864,
      "step": 90615
    },
    {
      "epoch": 13.497170092344355,
      "grad_norm": 12.625,
      "learning_rate": 1.4446324787765686e-05,
      "loss": 1.0086,
      "num_input_tokens_seen": 52578936,
      "step": 90620
    },
    {
      "epoch": 13.497914804885314,
      "grad_norm": 13.1875,
      "learning_rate": 1.4443379187380068e-05,
      "loss": 0.9149,
      "num_input_tokens_seen": 52581944,
      "step": 90625
    },
    {
      "epoch": 13.498659517426274,
      "grad_norm": 9.5625,
      "learning_rate": 1.4440433765337819e-05,
      "loss": 1.1791,
      "num_input_tokens_seen": 52584632,
      "step": 90630
    },
    {
      "epoch": 13.499404229967233,
      "grad_norm": 10.5625,
      "learning_rate": 1.443748852168872e-05,
      "loss": 1.1862,
      "num_input_tokens_seen": 52587576,
      "step": 90635
    },
    {
      "epoch": 13.500148942508192,
      "grad_norm": 9.25,
      "learning_rate": 1.443454345648252e-05,
      "loss": 1.1355,
      "num_input_tokens_seen": 52590648,
      "step": 90640
    },
    {
      "epoch": 13.50089365504915,
      "grad_norm": 8.6875,
      "learning_rate": 1.4431598569768955e-05,
      "loss": 0.9241,
      "num_input_tokens_seen": 52593496,
      "step": 90645
    },
    {
      "epoch": 13.501638367590111,
      "grad_norm": 13.6875,
      "learning_rate": 1.4428653861597802e-05,
      "loss": 1.2562,
      "num_input_tokens_seen": 52596440,
      "step": 90650
    },
    {
      "epoch": 13.50238308013107,
      "grad_norm": 7.71875,
      "learning_rate": 1.44257093320188e-05,
      "loss": 0.842,
      "num_input_tokens_seen": 52599448,
      "step": 90655
    },
    {
      "epoch": 13.503127792672029,
      "grad_norm": 13.125,
      "learning_rate": 1.4422764981081691e-05,
      "loss": 1.4559,
      "num_input_tokens_seen": 52602168,
      "step": 90660
    },
    {
      "epoch": 13.503872505212987,
      "grad_norm": 13.25,
      "learning_rate": 1.4419820808836207e-05,
      "loss": 1.0953,
      "num_input_tokens_seen": 52605016,
      "step": 90665
    },
    {
      "epoch": 13.504617217753946,
      "grad_norm": 10.6875,
      "learning_rate": 1.441687681533211e-05,
      "loss": 1.0767,
      "num_input_tokens_seen": 52607960,
      "step": 90670
    },
    {
      "epoch": 13.505361930294907,
      "grad_norm": 15.0625,
      "learning_rate": 1.4413933000619106e-05,
      "loss": 0.8498,
      "num_input_tokens_seen": 52610648,
      "step": 90675
    },
    {
      "epoch": 13.506106642835865,
      "grad_norm": 7.25,
      "learning_rate": 1.441098936474696e-05,
      "loss": 0.7356,
      "num_input_tokens_seen": 52613560,
      "step": 90680
    },
    {
      "epoch": 13.506851355376824,
      "grad_norm": 6.34375,
      "learning_rate": 1.4408045907765385e-05,
      "loss": 1.1073,
      "num_input_tokens_seen": 52616312,
      "step": 90685
    },
    {
      "epoch": 13.507596067917785,
      "grad_norm": 26.75,
      "learning_rate": 1.44051026297241e-05,
      "loss": 1.4138,
      "num_input_tokens_seen": 52619000,
      "step": 90690
    },
    {
      "epoch": 13.508340780458743,
      "grad_norm": 6.03125,
      "learning_rate": 1.4402159530672849e-05,
      "loss": 1.2618,
      "num_input_tokens_seen": 52621848,
      "step": 90695
    },
    {
      "epoch": 13.509085492999702,
      "grad_norm": 10.9375,
      "learning_rate": 1.4399216610661329e-05,
      "loss": 0.9816,
      "num_input_tokens_seen": 52624696,
      "step": 90700
    },
    {
      "epoch": 13.509830205540661,
      "grad_norm": 10.125,
      "learning_rate": 1.4396273869739279e-05,
      "loss": 1.1873,
      "num_input_tokens_seen": 52627352,
      "step": 90705
    },
    {
      "epoch": 13.51057491808162,
      "grad_norm": 7.4375,
      "learning_rate": 1.4393331307956399e-05,
      "loss": 0.9497,
      "num_input_tokens_seen": 52630360,
      "step": 90710
    },
    {
      "epoch": 13.51131963062258,
      "grad_norm": 7.375,
      "learning_rate": 1.4390388925362413e-05,
      "loss": 0.7213,
      "num_input_tokens_seen": 52633176,
      "step": 90715
    },
    {
      "epoch": 13.512064343163539,
      "grad_norm": 9.25,
      "learning_rate": 1.4387446722007025e-05,
      "loss": 1.0818,
      "num_input_tokens_seen": 52636152,
      "step": 90720
    },
    {
      "epoch": 13.512809055704498,
      "grad_norm": 10.8125,
      "learning_rate": 1.4384504697939938e-05,
      "loss": 0.8748,
      "num_input_tokens_seen": 52639224,
      "step": 90725
    },
    {
      "epoch": 13.513553768245457,
      "grad_norm": 9.75,
      "learning_rate": 1.4381562853210856e-05,
      "loss": 1.1459,
      "num_input_tokens_seen": 52642072,
      "step": 90730
    },
    {
      "epoch": 13.514298480786417,
      "grad_norm": 9.5,
      "learning_rate": 1.4378621187869467e-05,
      "loss": 1.2376,
      "num_input_tokens_seen": 52645016,
      "step": 90735
    },
    {
      "epoch": 13.515043193327376,
      "grad_norm": 16.875,
      "learning_rate": 1.4375679701965488e-05,
      "loss": 1.0192,
      "num_input_tokens_seen": 52647640,
      "step": 90740
    },
    {
      "epoch": 13.515787905868335,
      "grad_norm": 6.78125,
      "learning_rate": 1.437273839554859e-05,
      "loss": 1.2398,
      "num_input_tokens_seen": 52650872,
      "step": 90745
    },
    {
      "epoch": 13.516532618409293,
      "grad_norm": 16.375,
      "learning_rate": 1.436979726866849e-05,
      "loss": 1.2578,
      "num_input_tokens_seen": 52653752,
      "step": 90750
    },
    {
      "epoch": 13.517277330950254,
      "grad_norm": 10.0,
      "learning_rate": 1.436685632137485e-05,
      "loss": 1.0088,
      "num_input_tokens_seen": 52656600,
      "step": 90755
    },
    {
      "epoch": 13.518022043491213,
      "grad_norm": 11.625,
      "learning_rate": 1.4363915553717371e-05,
      "loss": 1.2025,
      "num_input_tokens_seen": 52659320,
      "step": 90760
    },
    {
      "epoch": 13.518766756032171,
      "grad_norm": 8.3125,
      "learning_rate": 1.4360974965745732e-05,
      "loss": 1.2077,
      "num_input_tokens_seen": 52662360,
      "step": 90765
    },
    {
      "epoch": 13.51951146857313,
      "grad_norm": 9.625,
      "learning_rate": 1.4358034557509598e-05,
      "loss": 1.3758,
      "num_input_tokens_seen": 52665240,
      "step": 90770
    },
    {
      "epoch": 13.52025618111409,
      "grad_norm": 11.6875,
      "learning_rate": 1.4355094329058666e-05,
      "loss": 1.3122,
      "num_input_tokens_seen": 52668024,
      "step": 90775
    },
    {
      "epoch": 13.52100089365505,
      "grad_norm": 10.1875,
      "learning_rate": 1.4352154280442592e-05,
      "loss": 1.0427,
      "num_input_tokens_seen": 52670776,
      "step": 90780
    },
    {
      "epoch": 13.521745606196008,
      "grad_norm": 12.3125,
      "learning_rate": 1.4349214411711043e-05,
      "loss": 1.3004,
      "num_input_tokens_seen": 52673688,
      "step": 90785
    },
    {
      "epoch": 13.522490318736967,
      "grad_norm": 13.0625,
      "learning_rate": 1.43462747229137e-05,
      "loss": 1.3508,
      "num_input_tokens_seen": 52676312,
      "step": 90790
    },
    {
      "epoch": 13.523235031277927,
      "grad_norm": 13.8125,
      "learning_rate": 1.4343335214100218e-05,
      "loss": 1.0931,
      "num_input_tokens_seen": 52679320,
      "step": 90795
    },
    {
      "epoch": 13.523979743818886,
      "grad_norm": 4.84375,
      "learning_rate": 1.4340395885320257e-05,
      "loss": 0.7532,
      "num_input_tokens_seen": 52682232,
      "step": 90800
    },
    {
      "epoch": 13.524724456359845,
      "grad_norm": 9.25,
      "learning_rate": 1.4337456736623462e-05,
      "loss": 1.1107,
      "num_input_tokens_seen": 52685272,
      "step": 90805
    },
    {
      "epoch": 13.525469168900804,
      "grad_norm": 3.953125,
      "learning_rate": 1.433451776805951e-05,
      "loss": 0.8974,
      "num_input_tokens_seen": 52688440,
      "step": 90810
    },
    {
      "epoch": 13.526213881441764,
      "grad_norm": 8.1875,
      "learning_rate": 1.4331578979678029e-05,
      "loss": 1.028,
      "num_input_tokens_seen": 52691192,
      "step": 90815
    },
    {
      "epoch": 13.526958593982723,
      "grad_norm": 11.0625,
      "learning_rate": 1.432864037152869e-05,
      "loss": 1.0217,
      "num_input_tokens_seen": 52694008,
      "step": 90820
    },
    {
      "epoch": 13.527703306523682,
      "grad_norm": 9.8125,
      "learning_rate": 1.4325701943661116e-05,
      "loss": 1.277,
      "num_input_tokens_seen": 52696664,
      "step": 90825
    },
    {
      "epoch": 13.52844801906464,
      "grad_norm": 12.3125,
      "learning_rate": 1.432276369612497e-05,
      "loss": 1.0541,
      "num_input_tokens_seen": 52699576,
      "step": 90830
    },
    {
      "epoch": 13.529192731605601,
      "grad_norm": 5.9375,
      "learning_rate": 1.4319825628969863e-05,
      "loss": 0.9012,
      "num_input_tokens_seen": 52702104,
      "step": 90835
    },
    {
      "epoch": 13.52993744414656,
      "grad_norm": 15.25,
      "learning_rate": 1.4316887742245464e-05,
      "loss": 0.9546,
      "num_input_tokens_seen": 52704856,
      "step": 90840
    },
    {
      "epoch": 13.530682156687519,
      "grad_norm": 7.21875,
      "learning_rate": 1.4313950036001384e-05,
      "loss": 0.9012,
      "num_input_tokens_seen": 52708152,
      "step": 90845
    },
    {
      "epoch": 13.531426869228477,
      "grad_norm": 11.875,
      "learning_rate": 1.431101251028726e-05,
      "loss": 1.3625,
      "num_input_tokens_seen": 52711128,
      "step": 90850
    },
    {
      "epoch": 13.532171581769436,
      "grad_norm": 7.84375,
      "learning_rate": 1.4308075165152718e-05,
      "loss": 1.1308,
      "num_input_tokens_seen": 52713720,
      "step": 90855
    },
    {
      "epoch": 13.532916294310397,
      "grad_norm": 13.9375,
      "learning_rate": 1.4305138000647367e-05,
      "loss": 1.0643,
      "num_input_tokens_seen": 52716408,
      "step": 90860
    },
    {
      "epoch": 13.533661006851355,
      "grad_norm": 11.5,
      "learning_rate": 1.4302201016820849e-05,
      "loss": 0.917,
      "num_input_tokens_seen": 52719032,
      "step": 90865
    },
    {
      "epoch": 13.534405719392314,
      "grad_norm": 12.125,
      "learning_rate": 1.4299264213722762e-05,
      "loss": 1.1967,
      "num_input_tokens_seen": 52722104,
      "step": 90870
    },
    {
      "epoch": 13.535150431933273,
      "grad_norm": 22.5,
      "learning_rate": 1.4296327591402742e-05,
      "loss": 0.9603,
      "num_input_tokens_seen": 52724792,
      "step": 90875
    },
    {
      "epoch": 13.535895144474233,
      "grad_norm": 11.375,
      "learning_rate": 1.4293391149910384e-05,
      "loss": 1.1506,
      "num_input_tokens_seen": 52727608,
      "step": 90880
    },
    {
      "epoch": 13.536639857015192,
      "grad_norm": 14.75,
      "learning_rate": 1.4290454889295296e-05,
      "loss": 1.3431,
      "num_input_tokens_seen": 52730488,
      "step": 90885
    },
    {
      "epoch": 13.537384569556151,
      "grad_norm": 6.9375,
      "learning_rate": 1.4287518809607097e-05,
      "loss": 1.0848,
      "num_input_tokens_seen": 52733048,
      "step": 90890
    },
    {
      "epoch": 13.53812928209711,
      "grad_norm": 6.90625,
      "learning_rate": 1.428458291089537e-05,
      "loss": 1.0263,
      "num_input_tokens_seen": 52735928,
      "step": 90895
    },
    {
      "epoch": 13.53887399463807,
      "grad_norm": 9.5,
      "learning_rate": 1.4281647193209732e-05,
      "loss": 0.8801,
      "num_input_tokens_seen": 52738552,
      "step": 90900
    },
    {
      "epoch": 13.539618707179029,
      "grad_norm": 9.0,
      "learning_rate": 1.4278711656599764e-05,
      "loss": 0.9139,
      "num_input_tokens_seen": 52741528,
      "step": 90905
    },
    {
      "epoch": 13.540363419719988,
      "grad_norm": 21.875,
      "learning_rate": 1.4275776301115074e-05,
      "loss": 1.2268,
      "num_input_tokens_seen": 52744280,
      "step": 90910
    },
    {
      "epoch": 13.541108132260947,
      "grad_norm": 17.5,
      "learning_rate": 1.4272841126805242e-05,
      "loss": 1.3008,
      "num_input_tokens_seen": 52747160,
      "step": 90915
    },
    {
      "epoch": 13.541852844801907,
      "grad_norm": 11.75,
      "learning_rate": 1.4269906133719863e-05,
      "loss": 1.0819,
      "num_input_tokens_seen": 52749912,
      "step": 90920
    },
    {
      "epoch": 13.542597557342866,
      "grad_norm": 7.84375,
      "learning_rate": 1.4266971321908507e-05,
      "loss": 1.2956,
      "num_input_tokens_seen": 52753048,
      "step": 90925
    },
    {
      "epoch": 13.543342269883825,
      "grad_norm": 13.5625,
      "learning_rate": 1.4264036691420756e-05,
      "loss": 1.2699,
      "num_input_tokens_seen": 52756056,
      "step": 90930
    },
    {
      "epoch": 13.544086982424783,
      "grad_norm": 6.84375,
      "learning_rate": 1.42611022423062e-05,
      "loss": 1.0758,
      "num_input_tokens_seen": 52759224,
      "step": 90935
    },
    {
      "epoch": 13.544831694965744,
      "grad_norm": 6.46875,
      "learning_rate": 1.42581679746144e-05,
      "loss": 1.1245,
      "num_input_tokens_seen": 52762040,
      "step": 90940
    },
    {
      "epoch": 13.545576407506703,
      "grad_norm": 9.375,
      "learning_rate": 1.4255233888394947e-05,
      "loss": 1.337,
      "num_input_tokens_seen": 52764760,
      "step": 90945
    },
    {
      "epoch": 13.546321120047661,
      "grad_norm": 13.3125,
      "learning_rate": 1.4252299983697381e-05,
      "loss": 1.2272,
      "num_input_tokens_seen": 52767576,
      "step": 90950
    },
    {
      "epoch": 13.54706583258862,
      "grad_norm": 5.53125,
      "learning_rate": 1.4249366260571299e-05,
      "loss": 0.7237,
      "num_input_tokens_seen": 52770360,
      "step": 90955
    },
    {
      "epoch": 13.54781054512958,
      "grad_norm": 7.40625,
      "learning_rate": 1.4246432719066244e-05,
      "loss": 1.3174,
      "num_input_tokens_seen": 52773240,
      "step": 90960
    },
    {
      "epoch": 13.54855525767054,
      "grad_norm": 17.125,
      "learning_rate": 1.4243499359231771e-05,
      "loss": 1.3314,
      "num_input_tokens_seen": 52776248,
      "step": 90965
    },
    {
      "epoch": 13.549299970211498,
      "grad_norm": 7.84375,
      "learning_rate": 1.4240566181117451e-05,
      "loss": 1.2393,
      "num_input_tokens_seen": 52779160,
      "step": 90970
    },
    {
      "epoch": 13.550044682752457,
      "grad_norm": 11.25,
      "learning_rate": 1.4237633184772822e-05,
      "loss": 0.8958,
      "num_input_tokens_seen": 52781912,
      "step": 90975
    },
    {
      "epoch": 13.550789395293418,
      "grad_norm": 8.9375,
      "learning_rate": 1.423470037024745e-05,
      "loss": 0.9836,
      "num_input_tokens_seen": 52784952,
      "step": 90980
    },
    {
      "epoch": 13.551534107834376,
      "grad_norm": 10.8125,
      "learning_rate": 1.423176773759088e-05,
      "loss": 1.3067,
      "num_input_tokens_seen": 52787832,
      "step": 90985
    },
    {
      "epoch": 13.552278820375335,
      "grad_norm": 12.375,
      "learning_rate": 1.4228835286852643e-05,
      "loss": 1.3588,
      "num_input_tokens_seen": 52791000,
      "step": 90990
    },
    {
      "epoch": 13.553023532916294,
      "grad_norm": 18.875,
      "learning_rate": 1.4225903018082278e-05,
      "loss": 1.166,
      "num_input_tokens_seen": 52794040,
      "step": 90995
    },
    {
      "epoch": 13.553768245457253,
      "grad_norm": 8.75,
      "learning_rate": 1.4222970931329343e-05,
      "loss": 0.6414,
      "num_input_tokens_seen": 52796952,
      "step": 91000
    },
    {
      "epoch": 13.554512957998213,
      "grad_norm": 14.25,
      "learning_rate": 1.4220039026643361e-05,
      "loss": 1.0801,
      "num_input_tokens_seen": 52800024,
      "step": 91005
    },
    {
      "epoch": 13.555257670539172,
      "grad_norm": 6.875,
      "learning_rate": 1.4217107304073851e-05,
      "loss": 0.7629,
      "num_input_tokens_seen": 52803096,
      "step": 91010
    },
    {
      "epoch": 13.55600238308013,
      "grad_norm": 10.6875,
      "learning_rate": 1.4214175763670365e-05,
      "loss": 0.9893,
      "num_input_tokens_seen": 52806040,
      "step": 91015
    },
    {
      "epoch": 13.556747095621091,
      "grad_norm": 17.0,
      "learning_rate": 1.4211244405482408e-05,
      "loss": 1.244,
      "num_input_tokens_seen": 52808888,
      "step": 91020
    },
    {
      "epoch": 13.55749180816205,
      "grad_norm": 11.375,
      "learning_rate": 1.420831322955952e-05,
      "loss": 1.1541,
      "num_input_tokens_seen": 52811640,
      "step": 91025
    },
    {
      "epoch": 13.558236520703009,
      "grad_norm": 6.1875,
      "learning_rate": 1.4205382235951204e-05,
      "loss": 1.3494,
      "num_input_tokens_seen": 52814840,
      "step": 91030
    },
    {
      "epoch": 13.558981233243967,
      "grad_norm": 7.15625,
      "learning_rate": 1.4202451424706991e-05,
      "loss": 1.0029,
      "num_input_tokens_seen": 52817944,
      "step": 91035
    },
    {
      "epoch": 13.559725945784926,
      "grad_norm": 9.625,
      "learning_rate": 1.4199520795876387e-05,
      "loss": 1.2005,
      "num_input_tokens_seen": 52820696,
      "step": 91040
    },
    {
      "epoch": 13.560470658325887,
      "grad_norm": 8.125,
      "learning_rate": 1.4196590349508896e-05,
      "loss": 0.8162,
      "num_input_tokens_seen": 52823544,
      "step": 91045
    },
    {
      "epoch": 13.561215370866845,
      "grad_norm": 7.28125,
      "learning_rate": 1.4193660085654037e-05,
      "loss": 1.3665,
      "num_input_tokens_seen": 52826392,
      "step": 91050
    },
    {
      "epoch": 13.561960083407804,
      "grad_norm": 15.125,
      "learning_rate": 1.419073000436131e-05,
      "loss": 0.868,
      "num_input_tokens_seen": 52829112,
      "step": 91055
    },
    {
      "epoch": 13.562704795948763,
      "grad_norm": 6.3125,
      "learning_rate": 1.4187800105680213e-05,
      "loss": 0.9101,
      "num_input_tokens_seen": 52831928,
      "step": 91060
    },
    {
      "epoch": 13.563449508489724,
      "grad_norm": 12.5625,
      "learning_rate": 1.4184870389660235e-05,
      "loss": 0.7525,
      "num_input_tokens_seen": 52834680,
      "step": 91065
    },
    {
      "epoch": 13.564194221030682,
      "grad_norm": 11.9375,
      "learning_rate": 1.4181940856350889e-05,
      "loss": 0.8038,
      "num_input_tokens_seen": 52837624,
      "step": 91070
    },
    {
      "epoch": 13.564938933571641,
      "grad_norm": 4.96875,
      "learning_rate": 1.4179011505801648e-05,
      "loss": 0.9309,
      "num_input_tokens_seen": 52840312,
      "step": 91075
    },
    {
      "epoch": 13.5656836461126,
      "grad_norm": 9.4375,
      "learning_rate": 1.4176082338062019e-05,
      "loss": 1.0283,
      "num_input_tokens_seen": 52843128,
      "step": 91080
    },
    {
      "epoch": 13.56642835865356,
      "grad_norm": 8.5625,
      "learning_rate": 1.4173153353181477e-05,
      "loss": 0.9067,
      "num_input_tokens_seen": 52845592,
      "step": 91085
    },
    {
      "epoch": 13.567173071194519,
      "grad_norm": 9.5,
      "learning_rate": 1.4170224551209493e-05,
      "loss": 0.9893,
      "num_input_tokens_seen": 52848440,
      "step": 91090
    },
    {
      "epoch": 13.567917783735478,
      "grad_norm": 18.125,
      "learning_rate": 1.4167295932195573e-05,
      "loss": 0.9895,
      "num_input_tokens_seen": 52851448,
      "step": 91095
    },
    {
      "epoch": 13.568662496276437,
      "grad_norm": 16.125,
      "learning_rate": 1.4164367496189169e-05,
      "loss": 1.2404,
      "num_input_tokens_seen": 52854328,
      "step": 91100
    },
    {
      "epoch": 13.569407208817397,
      "grad_norm": 10.5,
      "learning_rate": 1.4161439243239768e-05,
      "loss": 1.3902,
      "num_input_tokens_seen": 52857240,
      "step": 91105
    },
    {
      "epoch": 13.570151921358356,
      "grad_norm": 5.46875,
      "learning_rate": 1.4158511173396838e-05,
      "loss": 0.8888,
      "num_input_tokens_seen": 52860120,
      "step": 91110
    },
    {
      "epoch": 13.570896633899315,
      "grad_norm": 11.25,
      "learning_rate": 1.4155583286709833e-05,
      "loss": 1.1103,
      "num_input_tokens_seen": 52863192,
      "step": 91115
    },
    {
      "epoch": 13.571641346440273,
      "grad_norm": 8.3125,
      "learning_rate": 1.4152655583228235e-05,
      "loss": 0.8724,
      "num_input_tokens_seen": 52866296,
      "step": 91120
    },
    {
      "epoch": 13.572386058981234,
      "grad_norm": 15.25,
      "learning_rate": 1.41497280630015e-05,
      "loss": 1.0692,
      "num_input_tokens_seen": 52869144,
      "step": 91125
    },
    {
      "epoch": 13.573130771522193,
      "grad_norm": 8.375,
      "learning_rate": 1.414680072607908e-05,
      "loss": 1.1249,
      "num_input_tokens_seen": 52872216,
      "step": 91130
    },
    {
      "epoch": 13.573875484063151,
      "grad_norm": 8.5625,
      "learning_rate": 1.414387357251042e-05,
      "loss": 1.3301,
      "num_input_tokens_seen": 52874872,
      "step": 91135
    },
    {
      "epoch": 13.57462019660411,
      "grad_norm": 8.375,
      "learning_rate": 1.4140946602344993e-05,
      "loss": 1.1487,
      "num_input_tokens_seen": 52877976,
      "step": 91140
    },
    {
      "epoch": 13.57536490914507,
      "grad_norm": 11.0,
      "learning_rate": 1.4138019815632226e-05,
      "loss": 0.7361,
      "num_input_tokens_seen": 52880632,
      "step": 91145
    },
    {
      "epoch": 13.57610962168603,
      "grad_norm": 17.375,
      "learning_rate": 1.4135093212421584e-05,
      "loss": 1.0587,
      "num_input_tokens_seen": 52883608,
      "step": 91150
    },
    {
      "epoch": 13.576854334226988,
      "grad_norm": 9.3125,
      "learning_rate": 1.4132166792762491e-05,
      "loss": 0.9601,
      "num_input_tokens_seen": 52886616,
      "step": 91155
    },
    {
      "epoch": 13.577599046767947,
      "grad_norm": 19.0,
      "learning_rate": 1.4129240556704403e-05,
      "loss": 1.4694,
      "num_input_tokens_seen": 52889592,
      "step": 91160
    },
    {
      "epoch": 13.578343759308908,
      "grad_norm": 13.875,
      "learning_rate": 1.4126314504296751e-05,
      "loss": 1.0671,
      "num_input_tokens_seen": 52892536,
      "step": 91165
    },
    {
      "epoch": 13.579088471849866,
      "grad_norm": 16.125,
      "learning_rate": 1.412338863558895e-05,
      "loss": 1.17,
      "num_input_tokens_seen": 52895256,
      "step": 91170
    },
    {
      "epoch": 13.579833184390825,
      "grad_norm": 7.84375,
      "learning_rate": 1.4120462950630453e-05,
      "loss": 1.0699,
      "num_input_tokens_seen": 52898136,
      "step": 91175
    },
    {
      "epoch": 13.580577896931784,
      "grad_norm": 7.78125,
      "learning_rate": 1.411753744947068e-05,
      "loss": 0.7727,
      "num_input_tokens_seen": 52901080,
      "step": 91180
    },
    {
      "epoch": 13.581322609472743,
      "grad_norm": 6.96875,
      "learning_rate": 1.4114612132159049e-05,
      "loss": 0.9027,
      "num_input_tokens_seen": 52903928,
      "step": 91185
    },
    {
      "epoch": 13.582067322013703,
      "grad_norm": 17.25,
      "learning_rate": 1.4111686998744975e-05,
      "loss": 0.8606,
      "num_input_tokens_seen": 52906712,
      "step": 91190
    },
    {
      "epoch": 13.582812034554662,
      "grad_norm": 10.8125,
      "learning_rate": 1.410876204927789e-05,
      "loss": 0.7114,
      "num_input_tokens_seen": 52909496,
      "step": 91195
    },
    {
      "epoch": 13.58355674709562,
      "grad_norm": 6.84375,
      "learning_rate": 1.4105837283807194e-05,
      "loss": 0.9714,
      "num_input_tokens_seen": 52912568,
      "step": 91200
    },
    {
      "epoch": 13.584301459636581,
      "grad_norm": 12.375,
      "learning_rate": 1.410291270238231e-05,
      "loss": 1.1927,
      "num_input_tokens_seen": 52915192,
      "step": 91205
    },
    {
      "epoch": 13.58504617217754,
      "grad_norm": 8.875,
      "learning_rate": 1.4099988305052644e-05,
      "loss": 1.1144,
      "num_input_tokens_seen": 52917976,
      "step": 91210
    },
    {
      "epoch": 13.585790884718499,
      "grad_norm": 11.4375,
      "learning_rate": 1.4097064091867587e-05,
      "loss": 0.7929,
      "num_input_tokens_seen": 52921016,
      "step": 91215
    },
    {
      "epoch": 13.586535597259457,
      "grad_norm": 10.75,
      "learning_rate": 1.4094140062876559e-05,
      "loss": 1.2022,
      "num_input_tokens_seen": 52923992,
      "step": 91220
    },
    {
      "epoch": 13.587280309800416,
      "grad_norm": 4.9375,
      "learning_rate": 1.4091216218128945e-05,
      "loss": 1.0518,
      "num_input_tokens_seen": 52926488,
      "step": 91225
    },
    {
      "epoch": 13.588025022341377,
      "grad_norm": 6.625,
      "learning_rate": 1.4088292557674155e-05,
      "loss": 0.679,
      "num_input_tokens_seen": 52929176,
      "step": 91230
    },
    {
      "epoch": 13.588769734882336,
      "grad_norm": 8.8125,
      "learning_rate": 1.4085369081561556e-05,
      "loss": 1.3421,
      "num_input_tokens_seen": 52932152,
      "step": 91235
    },
    {
      "epoch": 13.589514447423294,
      "grad_norm": 13.0625,
      "learning_rate": 1.408244578984057e-05,
      "loss": 1.0006,
      "num_input_tokens_seen": 52935096,
      "step": 91240
    },
    {
      "epoch": 13.590259159964253,
      "grad_norm": 10.9375,
      "learning_rate": 1.4079522682560563e-05,
      "loss": 0.7278,
      "num_input_tokens_seen": 52937752,
      "step": 91245
    },
    {
      "epoch": 13.591003872505214,
      "grad_norm": 7.15625,
      "learning_rate": 1.4076599759770919e-05,
      "loss": 1.1528,
      "num_input_tokens_seen": 52940440,
      "step": 91250
    },
    {
      "epoch": 13.591748585046172,
      "grad_norm": 10.125,
      "learning_rate": 1.4073677021521026e-05,
      "loss": 0.8786,
      "num_input_tokens_seen": 52943480,
      "step": 91255
    },
    {
      "epoch": 13.592493297587131,
      "grad_norm": 8.0,
      "learning_rate": 1.407075446786024e-05,
      "loss": 1.2842,
      "num_input_tokens_seen": 52946424,
      "step": 91260
    },
    {
      "epoch": 13.59323801012809,
      "grad_norm": 7.8125,
      "learning_rate": 1.406783209883796e-05,
      "loss": 1.0979,
      "num_input_tokens_seen": 52949656,
      "step": 91265
    },
    {
      "epoch": 13.59398272266905,
      "grad_norm": 12.4375,
      "learning_rate": 1.4064909914503537e-05,
      "loss": 1.2423,
      "num_input_tokens_seen": 52952600,
      "step": 91270
    },
    {
      "epoch": 13.59472743521001,
      "grad_norm": 13.75,
      "learning_rate": 1.4061987914906354e-05,
      "loss": 0.9972,
      "num_input_tokens_seen": 52955480,
      "step": 91275
    },
    {
      "epoch": 13.595472147750968,
      "grad_norm": 4.90625,
      "learning_rate": 1.4059066100095763e-05,
      "loss": 1.0026,
      "num_input_tokens_seen": 52958136,
      "step": 91280
    },
    {
      "epoch": 13.596216860291927,
      "grad_norm": 10.4375,
      "learning_rate": 1.4056144470121137e-05,
      "loss": 0.6032,
      "num_input_tokens_seen": 52960952,
      "step": 91285
    },
    {
      "epoch": 13.596961572832887,
      "grad_norm": 11.5,
      "learning_rate": 1.405322302503183e-05,
      "loss": 1.07,
      "num_input_tokens_seen": 52964088,
      "step": 91290
    },
    {
      "epoch": 13.597706285373846,
      "grad_norm": 10.25,
      "learning_rate": 1.4050301764877183e-05,
      "loss": 1.3826,
      "num_input_tokens_seen": 52967096,
      "step": 91295
    },
    {
      "epoch": 13.598450997914805,
      "grad_norm": 8.6875,
      "learning_rate": 1.4047380689706568e-05,
      "loss": 0.9178,
      "num_input_tokens_seen": 52970104,
      "step": 91300
    },
    {
      "epoch": 13.599195710455763,
      "grad_norm": 5.59375,
      "learning_rate": 1.4044459799569316e-05,
      "loss": 0.9773,
      "num_input_tokens_seen": 52972824,
      "step": 91305
    },
    {
      "epoch": 13.599940422996724,
      "grad_norm": 11.9375,
      "learning_rate": 1.4041539094514788e-05,
      "loss": 1.1462,
      "num_input_tokens_seen": 52975704,
      "step": 91310
    },
    {
      "epoch": 13.600685135537683,
      "grad_norm": 11.8125,
      "learning_rate": 1.4038618574592322e-05,
      "loss": 0.9265,
      "num_input_tokens_seen": 52978424,
      "step": 91315
    },
    {
      "epoch": 13.601429848078642,
      "grad_norm": 7.5,
      "learning_rate": 1.4035698239851253e-05,
      "loss": 0.9886,
      "num_input_tokens_seen": 52981208,
      "step": 91320
    },
    {
      "epoch": 13.6021745606196,
      "grad_norm": 5.21875,
      "learning_rate": 1.403277809034092e-05,
      "loss": 0.9118,
      "num_input_tokens_seen": 52984248,
      "step": 91325
    },
    {
      "epoch": 13.60291927316056,
      "grad_norm": 6.0625,
      "learning_rate": 1.4029858126110645e-05,
      "loss": 0.741,
      "num_input_tokens_seen": 52986808,
      "step": 91330
    },
    {
      "epoch": 13.60366398570152,
      "grad_norm": 6.21875,
      "learning_rate": 1.4026938347209778e-05,
      "loss": 0.9239,
      "num_input_tokens_seen": 52990040,
      "step": 91335
    },
    {
      "epoch": 13.604408698242478,
      "grad_norm": 9.1875,
      "learning_rate": 1.4024018753687624e-05,
      "loss": 1.199,
      "num_input_tokens_seen": 52992664,
      "step": 91340
    },
    {
      "epoch": 13.605153410783437,
      "grad_norm": 9.25,
      "learning_rate": 1.4021099345593524e-05,
      "loss": 0.6992,
      "num_input_tokens_seen": 52996536,
      "step": 91345
    },
    {
      "epoch": 13.605898123324398,
      "grad_norm": 17.875,
      "learning_rate": 1.4018180122976788e-05,
      "loss": 1.3998,
      "num_input_tokens_seen": 52999320,
      "step": 91350
    },
    {
      "epoch": 13.606642835865356,
      "grad_norm": 7.46875,
      "learning_rate": 1.4015261085886743e-05,
      "loss": 1.1463,
      "num_input_tokens_seen": 53002200,
      "step": 91355
    },
    {
      "epoch": 13.607387548406315,
      "grad_norm": 10.5,
      "learning_rate": 1.4012342234372688e-05,
      "loss": 1.2373,
      "num_input_tokens_seen": 53005240,
      "step": 91360
    },
    {
      "epoch": 13.608132260947274,
      "grad_norm": 9.3125,
      "learning_rate": 1.4009423568483957e-05,
      "loss": 1.0892,
      "num_input_tokens_seen": 53008056,
      "step": 91365
    },
    {
      "epoch": 13.608876973488233,
      "grad_norm": 7.1875,
      "learning_rate": 1.4006505088269841e-05,
      "loss": 1.1184,
      "num_input_tokens_seen": 53011064,
      "step": 91370
    },
    {
      "epoch": 13.609621686029193,
      "grad_norm": 10.9375,
      "learning_rate": 1.4003586793779641e-05,
      "loss": 1.2668,
      "num_input_tokens_seen": 53013848,
      "step": 91375
    },
    {
      "epoch": 13.610366398570152,
      "grad_norm": 11.4375,
      "learning_rate": 1.4000668685062674e-05,
      "loss": 0.8103,
      "num_input_tokens_seen": 53016888,
      "step": 91380
    },
    {
      "epoch": 13.61111111111111,
      "grad_norm": 12.5,
      "learning_rate": 1.3997750762168232e-05,
      "loss": 0.9022,
      "num_input_tokens_seen": 53019928,
      "step": 91385
    },
    {
      "epoch": 13.61185582365207,
      "grad_norm": 12.0625,
      "learning_rate": 1.3994833025145607e-05,
      "loss": 0.9691,
      "num_input_tokens_seen": 53023000,
      "step": 91390
    },
    {
      "epoch": 13.61260053619303,
      "grad_norm": 13.0625,
      "learning_rate": 1.3991915474044081e-05,
      "loss": 0.9782,
      "num_input_tokens_seen": 53025848,
      "step": 91395
    },
    {
      "epoch": 13.613345248733989,
      "grad_norm": 5.96875,
      "learning_rate": 1.398899810891297e-05,
      "loss": 1.0811,
      "num_input_tokens_seen": 53028504,
      "step": 91400
    },
    {
      "epoch": 13.614089961274948,
      "grad_norm": 12.5625,
      "learning_rate": 1.3986080929801543e-05,
      "loss": 1.0574,
      "num_input_tokens_seen": 53031192,
      "step": 91405
    },
    {
      "epoch": 13.614834673815906,
      "grad_norm": 10.3125,
      "learning_rate": 1.3983163936759072e-05,
      "loss": 0.979,
      "num_input_tokens_seen": 53034008,
      "step": 91410
    },
    {
      "epoch": 13.615579386356867,
      "grad_norm": 8.25,
      "learning_rate": 1.398024712983486e-05,
      "loss": 0.7738,
      "num_input_tokens_seen": 53036984,
      "step": 91415
    },
    {
      "epoch": 13.616324098897826,
      "grad_norm": 10.8125,
      "learning_rate": 1.3977330509078165e-05,
      "loss": 1.2699,
      "num_input_tokens_seen": 53039800,
      "step": 91420
    },
    {
      "epoch": 13.617068811438784,
      "grad_norm": 10.75,
      "learning_rate": 1.3974414074538277e-05,
      "loss": 1.0693,
      "num_input_tokens_seen": 53042488,
      "step": 91425
    },
    {
      "epoch": 13.617813523979743,
      "grad_norm": 6.5625,
      "learning_rate": 1.3971497826264448e-05,
      "loss": 0.9485,
      "num_input_tokens_seen": 53045144,
      "step": 91430
    },
    {
      "epoch": 13.618558236520704,
      "grad_norm": 5.28125,
      "learning_rate": 1.3968581764305965e-05,
      "loss": 0.7934,
      "num_input_tokens_seen": 53048088,
      "step": 91435
    },
    {
      "epoch": 13.619302949061662,
      "grad_norm": 7.0625,
      "learning_rate": 1.396566588871208e-05,
      "loss": 1.0304,
      "num_input_tokens_seen": 53050936,
      "step": 91440
    },
    {
      "epoch": 13.620047661602621,
      "grad_norm": 8.5625,
      "learning_rate": 1.3962750199532042e-05,
      "loss": 0.8504,
      "num_input_tokens_seen": 53053688,
      "step": 91445
    },
    {
      "epoch": 13.62079237414358,
      "grad_norm": 13.375,
      "learning_rate": 1.3959834696815138e-05,
      "loss": 1.147,
      "num_input_tokens_seen": 53056632,
      "step": 91450
    },
    {
      "epoch": 13.62153708668454,
      "grad_norm": 10.8125,
      "learning_rate": 1.39569193806106e-05,
      "loss": 1.3926,
      "num_input_tokens_seen": 53059416,
      "step": 91455
    },
    {
      "epoch": 13.6222817992255,
      "grad_norm": 5.875,
      "learning_rate": 1.395400425096769e-05,
      "loss": 0.9837,
      "num_input_tokens_seen": 53062200,
      "step": 91460
    },
    {
      "epoch": 13.623026511766458,
      "grad_norm": 7.46875,
      "learning_rate": 1.3951089307935639e-05,
      "loss": 1.0588,
      "num_input_tokens_seen": 53065048,
      "step": 91465
    },
    {
      "epoch": 13.623771224307417,
      "grad_norm": 7.53125,
      "learning_rate": 1.3948174551563713e-05,
      "loss": 0.9982,
      "num_input_tokens_seen": 53067896,
      "step": 91470
    },
    {
      "epoch": 13.624515936848377,
      "grad_norm": 4.90625,
      "learning_rate": 1.3945259981901138e-05,
      "loss": 1.0839,
      "num_input_tokens_seen": 53070776,
      "step": 91475
    },
    {
      "epoch": 13.625260649389336,
      "grad_norm": 8.6875,
      "learning_rate": 1.394234559899717e-05,
      "loss": 1.404,
      "num_input_tokens_seen": 53073592,
      "step": 91480
    },
    {
      "epoch": 13.626005361930295,
      "grad_norm": 13.0,
      "learning_rate": 1.3939431402901034e-05,
      "loss": 1.005,
      "num_input_tokens_seen": 53076856,
      "step": 91485
    },
    {
      "epoch": 13.626750074471254,
      "grad_norm": 15.1875,
      "learning_rate": 1.3936517393661955e-05,
      "loss": 1.1582,
      "num_input_tokens_seen": 53080184,
      "step": 91490
    },
    {
      "epoch": 13.627494787012214,
      "grad_norm": 12.125,
      "learning_rate": 1.393360357132918e-05,
      "loss": 0.8464,
      "num_input_tokens_seen": 53082872,
      "step": 91495
    },
    {
      "epoch": 13.628239499553173,
      "grad_norm": 8.6875,
      "learning_rate": 1.3930689935951913e-05,
      "loss": 0.8891,
      "num_input_tokens_seen": 53085688,
      "step": 91500
    },
    {
      "epoch": 13.628984212094132,
      "grad_norm": 13.875,
      "learning_rate": 1.3927776487579397e-05,
      "loss": 0.8688,
      "num_input_tokens_seen": 53088600,
      "step": 91505
    },
    {
      "epoch": 13.62972892463509,
      "grad_norm": 12.0625,
      "learning_rate": 1.3924863226260849e-05,
      "loss": 1.0801,
      "num_input_tokens_seen": 53091384,
      "step": 91510
    },
    {
      "epoch": 13.63047363717605,
      "grad_norm": 17.375,
      "learning_rate": 1.3921950152045477e-05,
      "loss": 1.1968,
      "num_input_tokens_seen": 53094360,
      "step": 91515
    },
    {
      "epoch": 13.63121834971701,
      "grad_norm": 8.5625,
      "learning_rate": 1.391903726498249e-05,
      "loss": 1.1582,
      "num_input_tokens_seen": 53097208,
      "step": 91520
    },
    {
      "epoch": 13.631963062257968,
      "grad_norm": 7.96875,
      "learning_rate": 1.3916124565121114e-05,
      "loss": 1.1564,
      "num_input_tokens_seen": 53100280,
      "step": 91525
    },
    {
      "epoch": 13.632707774798927,
      "grad_norm": 7.78125,
      "learning_rate": 1.391321205251055e-05,
      "loss": 1.2711,
      "num_input_tokens_seen": 53103224,
      "step": 91530
    },
    {
      "epoch": 13.633452487339888,
      "grad_norm": 13.8125,
      "learning_rate": 1.3910299727199991e-05,
      "loss": 1.1616,
      "num_input_tokens_seen": 53105944,
      "step": 91535
    },
    {
      "epoch": 13.634197199880846,
      "grad_norm": 11.6875,
      "learning_rate": 1.3907387589238657e-05,
      "loss": 1.2441,
      "num_input_tokens_seen": 53108920,
      "step": 91540
    },
    {
      "epoch": 13.634941912421805,
      "grad_norm": 5.6875,
      "learning_rate": 1.3904475638675724e-05,
      "loss": 0.9725,
      "num_input_tokens_seen": 53111704,
      "step": 91545
    },
    {
      "epoch": 13.635686624962764,
      "grad_norm": 17.0,
      "learning_rate": 1.3901563875560408e-05,
      "loss": 1.1043,
      "num_input_tokens_seen": 53114552,
      "step": 91550
    },
    {
      "epoch": 13.636431337503723,
      "grad_norm": 16.875,
      "learning_rate": 1.3898652299941883e-05,
      "loss": 1.2686,
      "num_input_tokens_seen": 53117496,
      "step": 91555
    },
    {
      "epoch": 13.637176050044683,
      "grad_norm": 12.25,
      "learning_rate": 1.3895740911869351e-05,
      "loss": 0.9158,
      "num_input_tokens_seen": 53120344,
      "step": 91560
    },
    {
      "epoch": 13.637920762585642,
      "grad_norm": 9.0625,
      "learning_rate": 1.389282971139199e-05,
      "loss": 1.2039,
      "num_input_tokens_seen": 53123096,
      "step": 91565
    },
    {
      "epoch": 13.6386654751266,
      "grad_norm": 12.3125,
      "learning_rate": 1.3889918698558976e-05,
      "loss": 0.8361,
      "num_input_tokens_seen": 53126040,
      "step": 91570
    },
    {
      "epoch": 13.63941018766756,
      "grad_norm": 8.8125,
      "learning_rate": 1.3887007873419503e-05,
      "loss": 1.0166,
      "num_input_tokens_seen": 53128920,
      "step": 91575
    },
    {
      "epoch": 13.64015490020852,
      "grad_norm": 16.0,
      "learning_rate": 1.3884097236022736e-05,
      "loss": 1.3094,
      "num_input_tokens_seen": 53131736,
      "step": 91580
    },
    {
      "epoch": 13.640899612749479,
      "grad_norm": 12.125,
      "learning_rate": 1.3881186786417848e-05,
      "loss": 1.1939,
      "num_input_tokens_seen": 53134552,
      "step": 91585
    },
    {
      "epoch": 13.641644325290438,
      "grad_norm": 9.9375,
      "learning_rate": 1.3878276524654e-05,
      "loss": 1.0956,
      "num_input_tokens_seen": 53137592,
      "step": 91590
    },
    {
      "epoch": 13.642389037831396,
      "grad_norm": 9.0,
      "learning_rate": 1.3875366450780375e-05,
      "loss": 0.861,
      "num_input_tokens_seen": 53140408,
      "step": 91595
    },
    {
      "epoch": 13.643133750372357,
      "grad_norm": 16.625,
      "learning_rate": 1.387245656484612e-05,
      "loss": 1.2864,
      "num_input_tokens_seen": 53143288,
      "step": 91600
    },
    {
      "epoch": 13.643878462913316,
      "grad_norm": 10.8125,
      "learning_rate": 1.3869546866900409e-05,
      "loss": 1.0289,
      "num_input_tokens_seen": 53146296,
      "step": 91605
    },
    {
      "epoch": 13.644623175454274,
      "grad_norm": 11.75,
      "learning_rate": 1.3866637356992393e-05,
      "loss": 1.4002,
      "num_input_tokens_seen": 53149208,
      "step": 91610
    },
    {
      "epoch": 13.645367887995233,
      "grad_norm": 9.4375,
      "learning_rate": 1.3863728035171214e-05,
      "loss": 1.0563,
      "num_input_tokens_seen": 53152024,
      "step": 91615
    },
    {
      "epoch": 13.646112600536194,
      "grad_norm": 10.0,
      "learning_rate": 1.386081890148604e-05,
      "loss": 1.0015,
      "num_input_tokens_seen": 53154712,
      "step": 91620
    },
    {
      "epoch": 13.646857313077152,
      "grad_norm": 11.5,
      "learning_rate": 1.3857909955985999e-05,
      "loss": 1.1958,
      "num_input_tokens_seen": 53157688,
      "step": 91625
    },
    {
      "epoch": 13.647602025618111,
      "grad_norm": 10.3125,
      "learning_rate": 1.3855001198720255e-05,
      "loss": 0.9753,
      "num_input_tokens_seen": 53160472,
      "step": 91630
    },
    {
      "epoch": 13.64834673815907,
      "grad_norm": 6.71875,
      "learning_rate": 1.3852092629737928e-05,
      "loss": 0.9669,
      "num_input_tokens_seen": 53162968,
      "step": 91635
    },
    {
      "epoch": 13.64909145070003,
      "grad_norm": 16.875,
      "learning_rate": 1.3849184249088176e-05,
      "loss": 1.0411,
      "num_input_tokens_seen": 53165656,
      "step": 91640
    },
    {
      "epoch": 13.64983616324099,
      "grad_norm": 15.3125,
      "learning_rate": 1.3846276056820123e-05,
      "loss": 1.0324,
      "num_input_tokens_seen": 53168344,
      "step": 91645
    },
    {
      "epoch": 13.650580875781948,
      "grad_norm": 8.125,
      "learning_rate": 1.3843368052982903e-05,
      "loss": 0.9644,
      "num_input_tokens_seen": 53171320,
      "step": 91650
    },
    {
      "epoch": 13.651325588322907,
      "grad_norm": 4.4375,
      "learning_rate": 1.3840460237625635e-05,
      "loss": 1.0502,
      "num_input_tokens_seen": 53174392,
      "step": 91655
    },
    {
      "epoch": 13.652070300863867,
      "grad_norm": 20.375,
      "learning_rate": 1.3837552610797444e-05,
      "loss": 0.9932,
      "num_input_tokens_seen": 53177336,
      "step": 91660
    },
    {
      "epoch": 13.652815013404826,
      "grad_norm": 7.1875,
      "learning_rate": 1.3834645172547467e-05,
      "loss": 1.4148,
      "num_input_tokens_seen": 53180152,
      "step": 91665
    },
    {
      "epoch": 13.653559725945785,
      "grad_norm": 11.5,
      "learning_rate": 1.3831737922924798e-05,
      "loss": 1.1294,
      "num_input_tokens_seen": 53183480,
      "step": 91670
    },
    {
      "epoch": 13.654304438486744,
      "grad_norm": 13.5625,
      "learning_rate": 1.3828830861978579e-05,
      "loss": 1.3815,
      "num_input_tokens_seen": 53186296,
      "step": 91675
    },
    {
      "epoch": 13.655049151027704,
      "grad_norm": 9.0625,
      "learning_rate": 1.3825923989757896e-05,
      "loss": 1.1139,
      "num_input_tokens_seen": 53189080,
      "step": 91680
    },
    {
      "epoch": 13.655793863568663,
      "grad_norm": 5.15625,
      "learning_rate": 1.382301730631188e-05,
      "loss": 1.0403,
      "num_input_tokens_seen": 53192312,
      "step": 91685
    },
    {
      "epoch": 13.656538576109622,
      "grad_norm": 7.375,
      "learning_rate": 1.382011081168963e-05,
      "loss": 1.1839,
      "num_input_tokens_seen": 53195384,
      "step": 91690
    },
    {
      "epoch": 13.65728328865058,
      "grad_norm": 7.4375,
      "learning_rate": 1.3817204505940235e-05,
      "loss": 0.9873,
      "num_input_tokens_seen": 53198552,
      "step": 91695
    },
    {
      "epoch": 13.65802800119154,
      "grad_norm": 12.375,
      "learning_rate": 1.3814298389112811e-05,
      "loss": 0.9751,
      "num_input_tokens_seen": 53201304,
      "step": 91700
    },
    {
      "epoch": 13.6587727137325,
      "grad_norm": 5.6875,
      "learning_rate": 1.381139246125644e-05,
      "loss": 0.9821,
      "num_input_tokens_seen": 53204184,
      "step": 91705
    },
    {
      "epoch": 13.659517426273458,
      "grad_norm": 8.0,
      "learning_rate": 1.3808486722420233e-05,
      "loss": 1.1939,
      "num_input_tokens_seen": 53207064,
      "step": 91710
    },
    {
      "epoch": 13.660262138814417,
      "grad_norm": 6.125,
      "learning_rate": 1.3805581172653265e-05,
      "loss": 0.9538,
      "num_input_tokens_seen": 53210136,
      "step": 91715
    },
    {
      "epoch": 13.661006851355378,
      "grad_norm": 7.28125,
      "learning_rate": 1.3802675812004626e-05,
      "loss": 1.4045,
      "num_input_tokens_seen": 53213144,
      "step": 91720
    },
    {
      "epoch": 13.661751563896336,
      "grad_norm": 17.875,
      "learning_rate": 1.3799770640523398e-05,
      "loss": 1.1683,
      "num_input_tokens_seen": 53216024,
      "step": 91725
    },
    {
      "epoch": 13.662496276437295,
      "grad_norm": 11.5625,
      "learning_rate": 1.3796865658258654e-05,
      "loss": 1.2074,
      "num_input_tokens_seen": 53219096,
      "step": 91730
    },
    {
      "epoch": 13.663240988978254,
      "grad_norm": 7.3125,
      "learning_rate": 1.3793960865259486e-05,
      "loss": 0.9674,
      "num_input_tokens_seen": 53221976,
      "step": 91735
    },
    {
      "epoch": 13.663985701519213,
      "grad_norm": 7.53125,
      "learning_rate": 1.3791056261574952e-05,
      "loss": 0.8175,
      "num_input_tokens_seen": 53224760,
      "step": 91740
    },
    {
      "epoch": 13.664730414060173,
      "grad_norm": 8.9375,
      "learning_rate": 1.3788151847254139e-05,
      "loss": 1.2876,
      "num_input_tokens_seen": 53227704,
      "step": 91745
    },
    {
      "epoch": 13.665475126601132,
      "grad_norm": 16.25,
      "learning_rate": 1.3785247622346098e-05,
      "loss": 1.2622,
      "num_input_tokens_seen": 53230520,
      "step": 91750
    },
    {
      "epoch": 13.66621983914209,
      "grad_norm": 6.84375,
      "learning_rate": 1.3782343586899906e-05,
      "loss": 1.1289,
      "num_input_tokens_seen": 53233688,
      "step": 91755
    },
    {
      "epoch": 13.66696455168305,
      "grad_norm": 11.875,
      "learning_rate": 1.377943974096461e-05,
      "loss": 1.2102,
      "num_input_tokens_seen": 53236440,
      "step": 91760
    },
    {
      "epoch": 13.66770926422401,
      "grad_norm": 7.5,
      "learning_rate": 1.3776536084589287e-05,
      "loss": 0.6939,
      "num_input_tokens_seen": 53239352,
      "step": 91765
    },
    {
      "epoch": 13.668453976764969,
      "grad_norm": 4.96875,
      "learning_rate": 1.377363261782298e-05,
      "loss": 1.1867,
      "num_input_tokens_seen": 53242200,
      "step": 91770
    },
    {
      "epoch": 13.669198689305928,
      "grad_norm": 11.9375,
      "learning_rate": 1.3770729340714728e-05,
      "loss": 1.0038,
      "num_input_tokens_seen": 53245080,
      "step": 91775
    },
    {
      "epoch": 13.669943401846886,
      "grad_norm": 9.1875,
      "learning_rate": 1.3767826253313599e-05,
      "loss": 0.978,
      "num_input_tokens_seen": 53248248,
      "step": 91780
    },
    {
      "epoch": 13.670688114387847,
      "grad_norm": 7.75,
      "learning_rate": 1.376492335566863e-05,
      "loss": 0.936,
      "num_input_tokens_seen": 53251256,
      "step": 91785
    },
    {
      "epoch": 13.671432826928806,
      "grad_norm": 8.625,
      "learning_rate": 1.3762020647828866e-05,
      "loss": 0.9758,
      "num_input_tokens_seen": 53254136,
      "step": 91790
    },
    {
      "epoch": 13.672177539469764,
      "grad_norm": 5.15625,
      "learning_rate": 1.375911812984333e-05,
      "loss": 1.3155,
      "num_input_tokens_seen": 53256888,
      "step": 91795
    },
    {
      "epoch": 13.672922252010723,
      "grad_norm": 10.3125,
      "learning_rate": 1.3756215801761074e-05,
      "loss": 1.4273,
      "num_input_tokens_seen": 53259928,
      "step": 91800
    },
    {
      "epoch": 13.673666964551684,
      "grad_norm": 13.0,
      "learning_rate": 1.3753313663631119e-05,
      "loss": 0.9203,
      "num_input_tokens_seen": 53262648,
      "step": 91805
    },
    {
      "epoch": 13.674411677092642,
      "grad_norm": 10.0625,
      "learning_rate": 1.375041171550251e-05,
      "loss": 1.027,
      "num_input_tokens_seen": 53265496,
      "step": 91810
    },
    {
      "epoch": 13.675156389633601,
      "grad_norm": 14.3125,
      "learning_rate": 1.3747509957424259e-05,
      "loss": 1.0864,
      "num_input_tokens_seen": 53268408,
      "step": 91815
    },
    {
      "epoch": 13.67590110217456,
      "grad_norm": 10.125,
      "learning_rate": 1.3744608389445379e-05,
      "loss": 1.5366,
      "num_input_tokens_seen": 53271224,
      "step": 91820
    },
    {
      "epoch": 13.67664581471552,
      "grad_norm": 5.8125,
      "learning_rate": 1.3741707011614912e-05,
      "loss": 1.0834,
      "num_input_tokens_seen": 53273944,
      "step": 91825
    },
    {
      "epoch": 13.67739052725648,
      "grad_norm": 11.4375,
      "learning_rate": 1.3738805823981857e-05,
      "loss": 1.4833,
      "num_input_tokens_seen": 53276920,
      "step": 91830
    },
    {
      "epoch": 13.678135239797438,
      "grad_norm": 15.5625,
      "learning_rate": 1.3735904826595236e-05,
      "loss": 0.7311,
      "num_input_tokens_seen": 53279736,
      "step": 91835
    },
    {
      "epoch": 13.678879952338397,
      "grad_norm": 7.78125,
      "learning_rate": 1.3733004019504058e-05,
      "loss": 0.9088,
      "num_input_tokens_seen": 53282744,
      "step": 91840
    },
    {
      "epoch": 13.679624664879357,
      "grad_norm": 10.8125,
      "learning_rate": 1.3730103402757327e-05,
      "loss": 0.9421,
      "num_input_tokens_seen": 53285560,
      "step": 91845
    },
    {
      "epoch": 13.680369377420316,
      "grad_norm": 10.625,
      "learning_rate": 1.3727202976404033e-05,
      "loss": 1.2253,
      "num_input_tokens_seen": 53288312,
      "step": 91850
    },
    {
      "epoch": 13.681114089961275,
      "grad_norm": 10.0625,
      "learning_rate": 1.3724302740493198e-05,
      "loss": 1.0337,
      "num_input_tokens_seen": 53291288,
      "step": 91855
    },
    {
      "epoch": 13.681858802502234,
      "grad_norm": 14.9375,
      "learning_rate": 1.372140269507381e-05,
      "loss": 1.3438,
      "num_input_tokens_seen": 53294200,
      "step": 91860
    },
    {
      "epoch": 13.682603515043194,
      "grad_norm": 10.875,
      "learning_rate": 1.3718502840194847e-05,
      "loss": 1.0925,
      "num_input_tokens_seen": 53296888,
      "step": 91865
    },
    {
      "epoch": 13.683348227584153,
      "grad_norm": 12.375,
      "learning_rate": 1.3715603175905322e-05,
      "loss": 0.9087,
      "num_input_tokens_seen": 53299768,
      "step": 91870
    },
    {
      "epoch": 13.684092940125112,
      "grad_norm": 12.5,
      "learning_rate": 1.3712703702254203e-05,
      "loss": 1.1072,
      "num_input_tokens_seen": 53302424,
      "step": 91875
    },
    {
      "epoch": 13.68483765266607,
      "grad_norm": 9.9375,
      "learning_rate": 1.3709804419290496e-05,
      "loss": 0.9294,
      "num_input_tokens_seen": 53305336,
      "step": 91880
    },
    {
      "epoch": 13.68558236520703,
      "grad_norm": 19.375,
      "learning_rate": 1.3706905327063158e-05,
      "loss": 1.0236,
      "num_input_tokens_seen": 53308152,
      "step": 91885
    },
    {
      "epoch": 13.68632707774799,
      "grad_norm": 5.09375,
      "learning_rate": 1.3704006425621185e-05,
      "loss": 0.9287,
      "num_input_tokens_seen": 53310968,
      "step": 91890
    },
    {
      "epoch": 13.687071790288948,
      "grad_norm": 11.5625,
      "learning_rate": 1.3701107715013542e-05,
      "loss": 1.24,
      "num_input_tokens_seen": 53314008,
      "step": 91895
    },
    {
      "epoch": 13.687816502829907,
      "grad_norm": 7.28125,
      "learning_rate": 1.369820919528919e-05,
      "loss": 0.9913,
      "num_input_tokens_seen": 53317016,
      "step": 91900
    },
    {
      "epoch": 13.688561215370868,
      "grad_norm": 8.1875,
      "learning_rate": 1.3695310866497119e-05,
      "loss": 0.9025,
      "num_input_tokens_seen": 53320344,
      "step": 91905
    },
    {
      "epoch": 13.689305927911827,
      "grad_norm": 13.0,
      "learning_rate": 1.3692412728686282e-05,
      "loss": 1.3373,
      "num_input_tokens_seen": 53323160,
      "step": 91910
    },
    {
      "epoch": 13.690050640452785,
      "grad_norm": 16.75,
      "learning_rate": 1.3689514781905638e-05,
      "loss": 0.9419,
      "num_input_tokens_seen": 53325688,
      "step": 91915
    },
    {
      "epoch": 13.690795352993744,
      "grad_norm": 4.8125,
      "learning_rate": 1.3686617026204138e-05,
      "loss": 1.1097,
      "num_input_tokens_seen": 53328152,
      "step": 91920
    },
    {
      "epoch": 13.691540065534703,
      "grad_norm": 13.5,
      "learning_rate": 1.368371946163075e-05,
      "loss": 1.1431,
      "num_input_tokens_seen": 53331160,
      "step": 91925
    },
    {
      "epoch": 13.692284778075663,
      "grad_norm": 15.9375,
      "learning_rate": 1.3680822088234427e-05,
      "loss": 1.4456,
      "num_input_tokens_seen": 53334200,
      "step": 91930
    },
    {
      "epoch": 13.693029490616622,
      "grad_norm": 12.5625,
      "learning_rate": 1.3677924906064097e-05,
      "loss": 1.0006,
      "num_input_tokens_seen": 53337080,
      "step": 91935
    },
    {
      "epoch": 13.69377420315758,
      "grad_norm": 9.0625,
      "learning_rate": 1.3675027915168729e-05,
      "loss": 1.0394,
      "num_input_tokens_seen": 53339736,
      "step": 91940
    },
    {
      "epoch": 13.69451891569854,
      "grad_norm": 10.25,
      "learning_rate": 1.3672131115597241e-05,
      "loss": 0.7004,
      "num_input_tokens_seen": 53342424,
      "step": 91945
    },
    {
      "epoch": 13.6952636282395,
      "grad_norm": 6.28125,
      "learning_rate": 1.3669234507398601e-05,
      "loss": 1.2047,
      "num_input_tokens_seen": 53345368,
      "step": 91950
    },
    {
      "epoch": 13.696008340780459,
      "grad_norm": 11.75,
      "learning_rate": 1.3666338090621716e-05,
      "loss": 1.1137,
      "num_input_tokens_seen": 53348344,
      "step": 91955
    },
    {
      "epoch": 13.696753053321418,
      "grad_norm": 21.875,
      "learning_rate": 1.3663441865315538e-05,
      "loss": 1.2968,
      "num_input_tokens_seen": 53351256,
      "step": 91960
    },
    {
      "epoch": 13.697497765862376,
      "grad_norm": 8.75,
      "learning_rate": 1.3660545831528975e-05,
      "loss": 1.2668,
      "num_input_tokens_seen": 53354360,
      "step": 91965
    },
    {
      "epoch": 13.698242478403337,
      "grad_norm": 4.59375,
      "learning_rate": 1.365764998931098e-05,
      "loss": 0.9053,
      "num_input_tokens_seen": 53357656,
      "step": 91970
    },
    {
      "epoch": 13.698987190944296,
      "grad_norm": 17.5,
      "learning_rate": 1.365475433871046e-05,
      "loss": 1.3703,
      "num_input_tokens_seen": 53360536,
      "step": 91975
    },
    {
      "epoch": 13.699731903485254,
      "grad_norm": 12.8125,
      "learning_rate": 1.3651858879776336e-05,
      "loss": 1.034,
      "num_input_tokens_seen": 53363256,
      "step": 91980
    },
    {
      "epoch": 13.700476616026213,
      "grad_norm": 6.65625,
      "learning_rate": 1.3648963612557519e-05,
      "loss": 0.9397,
      "num_input_tokens_seen": 53366264,
      "step": 91985
    },
    {
      "epoch": 13.701221328567174,
      "grad_norm": 8.1875,
      "learning_rate": 1.3646068537102916e-05,
      "loss": 1.133,
      "num_input_tokens_seen": 53369048,
      "step": 91990
    },
    {
      "epoch": 13.701966041108133,
      "grad_norm": 9.1875,
      "learning_rate": 1.3643173653461454e-05,
      "loss": 1.241,
      "num_input_tokens_seen": 53371960,
      "step": 91995
    },
    {
      "epoch": 13.702710753649091,
      "grad_norm": 8.1875,
      "learning_rate": 1.3640278961682023e-05,
      "loss": 0.818,
      "num_input_tokens_seen": 53374776,
      "step": 92000
    },
    {
      "epoch": 13.70345546619005,
      "grad_norm": 8.5625,
      "learning_rate": 1.3637384461813546e-05,
      "loss": 0.857,
      "num_input_tokens_seen": 53377688,
      "step": 92005
    },
    {
      "epoch": 13.70420017873101,
      "grad_norm": 14.1875,
      "learning_rate": 1.3634490153904905e-05,
      "loss": 1.2134,
      "num_input_tokens_seen": 53380504,
      "step": 92010
    },
    {
      "epoch": 13.70494489127197,
      "grad_norm": 10.875,
      "learning_rate": 1.3631596038004994e-05,
      "loss": 0.924,
      "num_input_tokens_seen": 53383544,
      "step": 92015
    },
    {
      "epoch": 13.705689603812928,
      "grad_norm": 14.125,
      "learning_rate": 1.3628702114162722e-05,
      "loss": 0.9423,
      "num_input_tokens_seen": 53386136,
      "step": 92020
    },
    {
      "epoch": 13.706434316353887,
      "grad_norm": 11.75,
      "learning_rate": 1.3625808382426964e-05,
      "loss": 1.0796,
      "num_input_tokens_seen": 53389144,
      "step": 92025
    },
    {
      "epoch": 13.707179028894847,
      "grad_norm": 18.5,
      "learning_rate": 1.3622914842846619e-05,
      "loss": 1.1861,
      "num_input_tokens_seen": 53391960,
      "step": 92030
    },
    {
      "epoch": 13.707923741435806,
      "grad_norm": 10.3125,
      "learning_rate": 1.3620021495470556e-05,
      "loss": 0.9247,
      "num_input_tokens_seen": 53394584,
      "step": 92035
    },
    {
      "epoch": 13.708668453976765,
      "grad_norm": 6.84375,
      "learning_rate": 1.3617128340347673e-05,
      "loss": 1.1559,
      "num_input_tokens_seen": 53397848,
      "step": 92040
    },
    {
      "epoch": 13.709413166517724,
      "grad_norm": 17.75,
      "learning_rate": 1.361423537752684e-05,
      "loss": 1.2413,
      "num_input_tokens_seen": 53400888,
      "step": 92045
    },
    {
      "epoch": 13.710157879058684,
      "grad_norm": 10.0,
      "learning_rate": 1.3611342607056925e-05,
      "loss": 1.109,
      "num_input_tokens_seen": 53403736,
      "step": 92050
    },
    {
      "epoch": 13.710902591599643,
      "grad_norm": 4.5,
      "learning_rate": 1.3608450028986804e-05,
      "loss": 1.2906,
      "num_input_tokens_seen": 53406712,
      "step": 92055
    },
    {
      "epoch": 13.711647304140602,
      "grad_norm": 9.375,
      "learning_rate": 1.3605557643365333e-05,
      "loss": 1.0434,
      "num_input_tokens_seen": 53409464,
      "step": 92060
    },
    {
      "epoch": 13.71239201668156,
      "grad_norm": 13.375,
      "learning_rate": 1.3602665450241392e-05,
      "loss": 1.0749,
      "num_input_tokens_seen": 53412312,
      "step": 92065
    },
    {
      "epoch": 13.71313672922252,
      "grad_norm": 15.75,
      "learning_rate": 1.3599773449663828e-05,
      "loss": 1.069,
      "num_input_tokens_seen": 53415128,
      "step": 92070
    },
    {
      "epoch": 13.71388144176348,
      "grad_norm": 9.125,
      "learning_rate": 1.3596881641681513e-05,
      "loss": 0.8958,
      "num_input_tokens_seen": 53417976,
      "step": 92075
    },
    {
      "epoch": 13.714626154304439,
      "grad_norm": 10.0625,
      "learning_rate": 1.3593990026343284e-05,
      "loss": 1.3324,
      "num_input_tokens_seen": 53421016,
      "step": 92080
    },
    {
      "epoch": 13.715370866845397,
      "grad_norm": 11.1875,
      "learning_rate": 1.3591098603698007e-05,
      "loss": 1.0072,
      "num_input_tokens_seen": 53423992,
      "step": 92085
    },
    {
      "epoch": 13.716115579386356,
      "grad_norm": 4.90625,
      "learning_rate": 1.3588207373794526e-05,
      "loss": 0.9713,
      "num_input_tokens_seen": 53427352,
      "step": 92090
    },
    {
      "epoch": 13.716860291927317,
      "grad_norm": 8.3125,
      "learning_rate": 1.3585316336681675e-05,
      "loss": 0.9469,
      "num_input_tokens_seen": 53430008,
      "step": 92095
    },
    {
      "epoch": 13.717605004468275,
      "grad_norm": 10.5,
      "learning_rate": 1.3582425492408313e-05,
      "loss": 1.3802,
      "num_input_tokens_seen": 53432888,
      "step": 92100
    },
    {
      "epoch": 13.718349717009234,
      "grad_norm": 8.3125,
      "learning_rate": 1.3579534841023256e-05,
      "loss": 0.8923,
      "num_input_tokens_seen": 53435864,
      "step": 92105
    },
    {
      "epoch": 13.719094429550193,
      "grad_norm": 14.875,
      "learning_rate": 1.357664438257536e-05,
      "loss": 1.4046,
      "num_input_tokens_seen": 53438808,
      "step": 92110
    },
    {
      "epoch": 13.719839142091153,
      "grad_norm": 10.5,
      "learning_rate": 1.3573754117113446e-05,
      "loss": 1.1582,
      "num_input_tokens_seen": 53441784,
      "step": 92115
    },
    {
      "epoch": 13.720583854632112,
      "grad_norm": 11.8125,
      "learning_rate": 1.3570864044686349e-05,
      "loss": 1.1361,
      "num_input_tokens_seen": 53444600,
      "step": 92120
    },
    {
      "epoch": 13.721328567173071,
      "grad_norm": 6.78125,
      "learning_rate": 1.3567974165342873e-05,
      "loss": 0.9766,
      "num_input_tokens_seen": 53447608,
      "step": 92125
    },
    {
      "epoch": 13.72207327971403,
      "grad_norm": 8.8125,
      "learning_rate": 1.3565084479131865e-05,
      "loss": 1.117,
      "num_input_tokens_seen": 53450360,
      "step": 92130
    },
    {
      "epoch": 13.72281799225499,
      "grad_norm": 14.5,
      "learning_rate": 1.3562194986102134e-05,
      "loss": 0.9572,
      "num_input_tokens_seen": 53453656,
      "step": 92135
    },
    {
      "epoch": 13.723562704795949,
      "grad_norm": 14.8125,
      "learning_rate": 1.3559305686302482e-05,
      "loss": 1.0354,
      "num_input_tokens_seen": 53456280,
      "step": 92140
    },
    {
      "epoch": 13.724307417336908,
      "grad_norm": 8.1875,
      "learning_rate": 1.3556416579781745e-05,
      "loss": 1.0562,
      "num_input_tokens_seen": 53459064,
      "step": 92145
    },
    {
      "epoch": 13.725052129877866,
      "grad_norm": 14.3125,
      "learning_rate": 1.355352766658871e-05,
      "loss": 1.1111,
      "num_input_tokens_seen": 53461880,
      "step": 92150
    },
    {
      "epoch": 13.725796842418827,
      "grad_norm": 11.5625,
      "learning_rate": 1.3550638946772198e-05,
      "loss": 1.0298,
      "num_input_tokens_seen": 53464632,
      "step": 92155
    },
    {
      "epoch": 13.726541554959786,
      "grad_norm": 11.9375,
      "learning_rate": 1.3547750420380994e-05,
      "loss": 1.0136,
      "num_input_tokens_seen": 53467544,
      "step": 92160
    },
    {
      "epoch": 13.727286267500745,
      "grad_norm": 8.875,
      "learning_rate": 1.3544862087463922e-05,
      "loss": 1.0033,
      "num_input_tokens_seen": 53470360,
      "step": 92165
    },
    {
      "epoch": 13.728030980041703,
      "grad_norm": 5.09375,
      "learning_rate": 1.3541973948069757e-05,
      "loss": 1.0039,
      "num_input_tokens_seen": 53473144,
      "step": 92170
    },
    {
      "epoch": 13.728775692582664,
      "grad_norm": 10.375,
      "learning_rate": 1.3539086002247301e-05,
      "loss": 1.2132,
      "num_input_tokens_seen": 53475928,
      "step": 92175
    },
    {
      "epoch": 13.729520405123623,
      "grad_norm": 8.3125,
      "learning_rate": 1.3536198250045326e-05,
      "loss": 1.0902,
      "num_input_tokens_seen": 53478584,
      "step": 92180
    },
    {
      "epoch": 13.730265117664581,
      "grad_norm": 21.25,
      "learning_rate": 1.353331069151264e-05,
      "loss": 1.1725,
      "num_input_tokens_seen": 53481624,
      "step": 92185
    },
    {
      "epoch": 13.73100983020554,
      "grad_norm": 9.625,
      "learning_rate": 1.3530423326698015e-05,
      "loss": 1.2803,
      "num_input_tokens_seen": 53484568,
      "step": 92190
    },
    {
      "epoch": 13.7317545427465,
      "grad_norm": 14.125,
      "learning_rate": 1.3527536155650224e-05,
      "loss": 1.4781,
      "num_input_tokens_seen": 53487608,
      "step": 92195
    },
    {
      "epoch": 13.73249925528746,
      "grad_norm": 14.875,
      "learning_rate": 1.3524649178418058e-05,
      "loss": 1.3381,
      "num_input_tokens_seen": 53490744,
      "step": 92200
    },
    {
      "epoch": 13.733243967828418,
      "grad_norm": 19.375,
      "learning_rate": 1.3521762395050272e-05,
      "loss": 0.9909,
      "num_input_tokens_seen": 53493624,
      "step": 92205
    },
    {
      "epoch": 13.733988680369377,
      "grad_norm": 10.375,
      "learning_rate": 1.3518875805595654e-05,
      "loss": 1.0794,
      "num_input_tokens_seen": 53496312,
      "step": 92210
    },
    {
      "epoch": 13.734733392910336,
      "grad_norm": 6.65625,
      "learning_rate": 1.3515989410102959e-05,
      "loss": 1.0959,
      "num_input_tokens_seen": 53499256,
      "step": 92215
    },
    {
      "epoch": 13.735478105451296,
      "grad_norm": 9.4375,
      "learning_rate": 1.3513103208620941e-05,
      "loss": 0.9528,
      "num_input_tokens_seen": 53502456,
      "step": 92220
    },
    {
      "epoch": 13.736222817992255,
      "grad_norm": 4.84375,
      "learning_rate": 1.3510217201198383e-05,
      "loss": 0.8274,
      "num_input_tokens_seen": 53505624,
      "step": 92225
    },
    {
      "epoch": 13.736967530533214,
      "grad_norm": 5.03125,
      "learning_rate": 1.3507331387884015e-05,
      "loss": 0.9998,
      "num_input_tokens_seen": 53508600,
      "step": 92230
    },
    {
      "epoch": 13.737712243074174,
      "grad_norm": 15.875,
      "learning_rate": 1.3504445768726612e-05,
      "loss": 1.1424,
      "num_input_tokens_seen": 53511448,
      "step": 92235
    },
    {
      "epoch": 13.738456955615133,
      "grad_norm": 10.4375,
      "learning_rate": 1.3501560343774917e-05,
      "loss": 0.9952,
      "num_input_tokens_seen": 53514168,
      "step": 92240
    },
    {
      "epoch": 13.739201668156092,
      "grad_norm": 12.0625,
      "learning_rate": 1.3498675113077669e-05,
      "loss": 1.5393,
      "num_input_tokens_seen": 53517368,
      "step": 92245
    },
    {
      "epoch": 13.73994638069705,
      "grad_norm": 8.875,
      "learning_rate": 1.3495790076683617e-05,
      "loss": 1.0018,
      "num_input_tokens_seen": 53519992,
      "step": 92250
    },
    {
      "epoch": 13.74069109323801,
      "grad_norm": 9.5,
      "learning_rate": 1.3492905234641492e-05,
      "loss": 1.0684,
      "num_input_tokens_seen": 53522808,
      "step": 92255
    },
    {
      "epoch": 13.74143580577897,
      "grad_norm": 14.125,
      "learning_rate": 1.3490020587000046e-05,
      "loss": 1.0899,
      "num_input_tokens_seen": 53525560,
      "step": 92260
    },
    {
      "epoch": 13.742180518319929,
      "grad_norm": 7.78125,
      "learning_rate": 1.3487136133807992e-05,
      "loss": 0.8925,
      "num_input_tokens_seen": 53528600,
      "step": 92265
    },
    {
      "epoch": 13.742925230860887,
      "grad_norm": 17.25,
      "learning_rate": 1.3484251875114085e-05,
      "loss": 1.2928,
      "num_input_tokens_seen": 53531608,
      "step": 92270
    },
    {
      "epoch": 13.743669943401846,
      "grad_norm": 7.9375,
      "learning_rate": 1.3481367810967027e-05,
      "loss": 1.1497,
      "num_input_tokens_seen": 53534552,
      "step": 92275
    },
    {
      "epoch": 13.744414655942807,
      "grad_norm": 20.125,
      "learning_rate": 1.3478483941415565e-05,
      "loss": 1.5336,
      "num_input_tokens_seen": 53537144,
      "step": 92280
    },
    {
      "epoch": 13.745159368483765,
      "grad_norm": 7.0,
      "learning_rate": 1.3475600266508395e-05,
      "loss": 0.9171,
      "num_input_tokens_seen": 53539896,
      "step": 92285
    },
    {
      "epoch": 13.745904081024724,
      "grad_norm": 6.46875,
      "learning_rate": 1.3472716786294254e-05,
      "loss": 1.1377,
      "num_input_tokens_seen": 53542616,
      "step": 92290
    },
    {
      "epoch": 13.746648793565683,
      "grad_norm": 12.4375,
      "learning_rate": 1.3469833500821848e-05,
      "loss": 1.0684,
      "num_input_tokens_seen": 53545784,
      "step": 92295
    },
    {
      "epoch": 13.747393506106643,
      "grad_norm": 12.0625,
      "learning_rate": 1.3466950410139878e-05,
      "loss": 1.3362,
      "num_input_tokens_seen": 53548824,
      "step": 92300
    },
    {
      "epoch": 13.748138218647602,
      "grad_norm": 21.5,
      "learning_rate": 1.3464067514297069e-05,
      "loss": 1.028,
      "num_input_tokens_seen": 53552120,
      "step": 92305
    },
    {
      "epoch": 13.748882931188561,
      "grad_norm": 11.375,
      "learning_rate": 1.3461184813342116e-05,
      "loss": 1.0709,
      "num_input_tokens_seen": 53555544,
      "step": 92310
    },
    {
      "epoch": 13.74962764372952,
      "grad_norm": 12.4375,
      "learning_rate": 1.3458302307323714e-05,
      "loss": 1.0127,
      "num_input_tokens_seen": 53558520,
      "step": 92315
    },
    {
      "epoch": 13.75037235627048,
      "grad_norm": 12.0625,
      "learning_rate": 1.3455419996290558e-05,
      "loss": 1.2224,
      "num_input_tokens_seen": 53561432,
      "step": 92320
    },
    {
      "epoch": 13.751117068811439,
      "grad_norm": 5.1875,
      "learning_rate": 1.3452537880291355e-05,
      "loss": 0.8769,
      "num_input_tokens_seen": 53564312,
      "step": 92325
    },
    {
      "epoch": 13.751861781352398,
      "grad_norm": 7.78125,
      "learning_rate": 1.3449655959374791e-05,
      "loss": 0.9052,
      "num_input_tokens_seen": 53567288,
      "step": 92330
    },
    {
      "epoch": 13.752606493893357,
      "grad_norm": 14.5625,
      "learning_rate": 1.3446774233589537e-05,
      "loss": 1.2012,
      "num_input_tokens_seen": 53570072,
      "step": 92335
    },
    {
      "epoch": 13.753351206434317,
      "grad_norm": 8.0625,
      "learning_rate": 1.3443892702984302e-05,
      "loss": 1.0628,
      "num_input_tokens_seen": 53572856,
      "step": 92340
    },
    {
      "epoch": 13.754095918975276,
      "grad_norm": 9.8125,
      "learning_rate": 1.3441011367607743e-05,
      "loss": 1.2141,
      "num_input_tokens_seen": 53575864,
      "step": 92345
    },
    {
      "epoch": 13.754840631516235,
      "grad_norm": 7.6875,
      "learning_rate": 1.343813022750856e-05,
      "loss": 1.1824,
      "num_input_tokens_seen": 53578936,
      "step": 92350
    },
    {
      "epoch": 13.755585344057193,
      "grad_norm": 6.75,
      "learning_rate": 1.3435249282735407e-05,
      "loss": 1.223,
      "num_input_tokens_seen": 53581912,
      "step": 92355
    },
    {
      "epoch": 13.756330056598154,
      "grad_norm": 8.8125,
      "learning_rate": 1.343236853333697e-05,
      "loss": 1.0105,
      "num_input_tokens_seen": 53584632,
      "step": 92360
    },
    {
      "epoch": 13.757074769139113,
      "grad_norm": 8.8125,
      "learning_rate": 1.3429487979361905e-05,
      "loss": 1.111,
      "num_input_tokens_seen": 53587672,
      "step": 92365
    },
    {
      "epoch": 13.757819481680071,
      "grad_norm": 15.3125,
      "learning_rate": 1.342660762085889e-05,
      "loss": 1.1385,
      "num_input_tokens_seen": 53590712,
      "step": 92370
    },
    {
      "epoch": 13.75856419422103,
      "grad_norm": 13.375,
      "learning_rate": 1.3423727457876572e-05,
      "loss": 1.1684,
      "num_input_tokens_seen": 53593784,
      "step": 92375
    },
    {
      "epoch": 13.75930890676199,
      "grad_norm": 5.34375,
      "learning_rate": 1.3420847490463614e-05,
      "loss": 1.2741,
      "num_input_tokens_seen": 53596600,
      "step": 92380
    },
    {
      "epoch": 13.76005361930295,
      "grad_norm": 8.875,
      "learning_rate": 1.3417967718668672e-05,
      "loss": 1.0926,
      "num_input_tokens_seen": 53599352,
      "step": 92385
    },
    {
      "epoch": 13.760798331843908,
      "grad_norm": 8.5,
      "learning_rate": 1.3415088142540383e-05,
      "loss": 0.774,
      "num_input_tokens_seen": 53601848,
      "step": 92390
    },
    {
      "epoch": 13.761543044384867,
      "grad_norm": 13.25,
      "learning_rate": 1.3412208762127415e-05,
      "loss": 1.2612,
      "num_input_tokens_seen": 53604760,
      "step": 92395
    },
    {
      "epoch": 13.762287756925826,
      "grad_norm": 8.25,
      "learning_rate": 1.3409329577478391e-05,
      "loss": 0.953,
      "num_input_tokens_seen": 53607608,
      "step": 92400
    },
    {
      "epoch": 13.763032469466786,
      "grad_norm": 8.125,
      "learning_rate": 1.3406450588641978e-05,
      "loss": 0.8468,
      "num_input_tokens_seen": 53610520,
      "step": 92405
    },
    {
      "epoch": 13.763777182007745,
      "grad_norm": 15.125,
      "learning_rate": 1.3403571795666786e-05,
      "loss": 1.3522,
      "num_input_tokens_seen": 53613432,
      "step": 92410
    },
    {
      "epoch": 13.764521894548704,
      "grad_norm": 12.1875,
      "learning_rate": 1.3400693198601472e-05,
      "loss": 1.1457,
      "num_input_tokens_seen": 53616248,
      "step": 92415
    },
    {
      "epoch": 13.765266607089664,
      "grad_norm": 11.125,
      "learning_rate": 1.339781479749466e-05,
      "loss": 1.4744,
      "num_input_tokens_seen": 53619064,
      "step": 92420
    },
    {
      "epoch": 13.766011319630623,
      "grad_norm": 10.375,
      "learning_rate": 1.3394936592394963e-05,
      "loss": 1.1974,
      "num_input_tokens_seen": 53622136,
      "step": 92425
    },
    {
      "epoch": 13.766756032171582,
      "grad_norm": 13.25,
      "learning_rate": 1.3392058583351027e-05,
      "loss": 1.0899,
      "num_input_tokens_seen": 53624984,
      "step": 92430
    },
    {
      "epoch": 13.76750074471254,
      "grad_norm": 10.0625,
      "learning_rate": 1.3389180770411456e-05,
      "loss": 1.2595,
      "num_input_tokens_seen": 53627832,
      "step": 92435
    },
    {
      "epoch": 13.7682454572535,
      "grad_norm": 7.78125,
      "learning_rate": 1.3386303153624882e-05,
      "loss": 0.7642,
      "num_input_tokens_seen": 53630680,
      "step": 92440
    },
    {
      "epoch": 13.76899016979446,
      "grad_norm": 12.625,
      "learning_rate": 1.3383425733039914e-05,
      "loss": 1.2249,
      "num_input_tokens_seen": 53633336,
      "step": 92445
    },
    {
      "epoch": 13.769734882335419,
      "grad_norm": 16.875,
      "learning_rate": 1.3380548508705162e-05,
      "loss": 1.2246,
      "num_input_tokens_seen": 53636248,
      "step": 92450
    },
    {
      "epoch": 13.770479594876377,
      "grad_norm": 9.0,
      "learning_rate": 1.3377671480669235e-05,
      "loss": 1.093,
      "num_input_tokens_seen": 53639064,
      "step": 92455
    },
    {
      "epoch": 13.771224307417336,
      "grad_norm": 13.9375,
      "learning_rate": 1.3374794648980721e-05,
      "loss": 1.2325,
      "num_input_tokens_seen": 53642136,
      "step": 92460
    },
    {
      "epoch": 13.771969019958297,
      "grad_norm": 12.875,
      "learning_rate": 1.337191801368825e-05,
      "loss": 1.1318,
      "num_input_tokens_seen": 53645208,
      "step": 92465
    },
    {
      "epoch": 13.772713732499255,
      "grad_norm": 11.1875,
      "learning_rate": 1.3369041574840396e-05,
      "loss": 0.9459,
      "num_input_tokens_seen": 53648280,
      "step": 92470
    },
    {
      "epoch": 13.773458445040214,
      "grad_norm": 15.25,
      "learning_rate": 1.3366165332485772e-05,
      "loss": 1.5194,
      "num_input_tokens_seen": 53651352,
      "step": 92475
    },
    {
      "epoch": 13.774203157581173,
      "grad_norm": 7.03125,
      "learning_rate": 1.3363289286672952e-05,
      "loss": 1.0751,
      "num_input_tokens_seen": 53654136,
      "step": 92480
    },
    {
      "epoch": 13.774947870122134,
      "grad_norm": 4.78125,
      "learning_rate": 1.3360413437450542e-05,
      "loss": 0.9007,
      "num_input_tokens_seen": 53656824,
      "step": 92485
    },
    {
      "epoch": 13.775692582663092,
      "grad_norm": 7.28125,
      "learning_rate": 1.3357537784867105e-05,
      "loss": 1.0299,
      "num_input_tokens_seen": 53659800,
      "step": 92490
    },
    {
      "epoch": 13.776437295204051,
      "grad_norm": 4.21875,
      "learning_rate": 1.3354662328971246e-05,
      "loss": 0.9384,
      "num_input_tokens_seen": 53662616,
      "step": 92495
    },
    {
      "epoch": 13.77718200774501,
      "grad_norm": 9.25,
      "learning_rate": 1.3351787069811533e-05,
      "loss": 0.8204,
      "num_input_tokens_seen": 53665688,
      "step": 92500
    },
    {
      "epoch": 13.77792672028597,
      "grad_norm": 17.25,
      "learning_rate": 1.3348912007436537e-05,
      "loss": 1.1708,
      "num_input_tokens_seen": 53668696,
      "step": 92505
    },
    {
      "epoch": 13.778671432826929,
      "grad_norm": 8.5625,
      "learning_rate": 1.3346037141894829e-05,
      "loss": 1.0261,
      "num_input_tokens_seen": 53671768,
      "step": 92510
    },
    {
      "epoch": 13.779416145367888,
      "grad_norm": 11.0625,
      "learning_rate": 1.3343162473234972e-05,
      "loss": 0.939,
      "num_input_tokens_seen": 53674616,
      "step": 92515
    },
    {
      "epoch": 13.780160857908847,
      "grad_norm": 13.875,
      "learning_rate": 1.3340288001505546e-05,
      "loss": 0.8266,
      "num_input_tokens_seen": 53677368,
      "step": 92520
    },
    {
      "epoch": 13.780905570449807,
      "grad_norm": 14.4375,
      "learning_rate": 1.3337413726755093e-05,
      "loss": 1.1952,
      "num_input_tokens_seen": 53680376,
      "step": 92525
    },
    {
      "epoch": 13.781650282990766,
      "grad_norm": 10.9375,
      "learning_rate": 1.3334539649032193e-05,
      "loss": 1.069,
      "num_input_tokens_seen": 53683448,
      "step": 92530
    },
    {
      "epoch": 13.782394995531725,
      "grad_norm": 7.8125,
      "learning_rate": 1.3331665768385387e-05,
      "loss": 1.0644,
      "num_input_tokens_seen": 53686104,
      "step": 92535
    },
    {
      "epoch": 13.783139708072683,
      "grad_norm": 6.78125,
      "learning_rate": 1.3328792084863223e-05,
      "loss": 0.8036,
      "num_input_tokens_seen": 53689048,
      "step": 92540
    },
    {
      "epoch": 13.783884420613644,
      "grad_norm": 7.46875,
      "learning_rate": 1.3325918598514265e-05,
      "loss": 0.7696,
      "num_input_tokens_seen": 53691800,
      "step": 92545
    },
    {
      "epoch": 13.784629133154603,
      "grad_norm": 11.3125,
      "learning_rate": 1.3323045309387033e-05,
      "loss": 0.7518,
      "num_input_tokens_seen": 53694776,
      "step": 92550
    },
    {
      "epoch": 13.785373845695561,
      "grad_norm": 14.3125,
      "learning_rate": 1.3320172217530094e-05,
      "loss": 1.1781,
      "num_input_tokens_seen": 53697592,
      "step": 92555
    },
    {
      "epoch": 13.78611855823652,
      "grad_norm": 17.5,
      "learning_rate": 1.3317299322991966e-05,
      "loss": 1.2719,
      "num_input_tokens_seen": 53700632,
      "step": 92560
    },
    {
      "epoch": 13.78686327077748,
      "grad_norm": 5.59375,
      "learning_rate": 1.33144266258212e-05,
      "loss": 0.9869,
      "num_input_tokens_seen": 53703704,
      "step": 92565
    },
    {
      "epoch": 13.78760798331844,
      "grad_norm": 4.4375,
      "learning_rate": 1.3311554126066323e-05,
      "loss": 1.3165,
      "num_input_tokens_seen": 53706488,
      "step": 92570
    },
    {
      "epoch": 13.788352695859398,
      "grad_norm": 6.125,
      "learning_rate": 1.3308681823775853e-05,
      "loss": 1.0939,
      "num_input_tokens_seen": 53709464,
      "step": 92575
    },
    {
      "epoch": 13.789097408400357,
      "grad_norm": 10.3125,
      "learning_rate": 1.3305809718998324e-05,
      "loss": 1.0202,
      "num_input_tokens_seen": 53712376,
      "step": 92580
    },
    {
      "epoch": 13.789842120941316,
      "grad_norm": 4.78125,
      "learning_rate": 1.3302937811782249e-05,
      "loss": 0.9658,
      "num_input_tokens_seen": 53715256,
      "step": 92585
    },
    {
      "epoch": 13.790586833482276,
      "grad_norm": 8.75,
      "learning_rate": 1.3300066102176157e-05,
      "loss": 0.8818,
      "num_input_tokens_seen": 53718136,
      "step": 92590
    },
    {
      "epoch": 13.791331546023235,
      "grad_norm": 5.0,
      "learning_rate": 1.3297194590228545e-05,
      "loss": 1.0718,
      "num_input_tokens_seen": 53720984,
      "step": 92595
    },
    {
      "epoch": 13.792076258564194,
      "grad_norm": 9.5625,
      "learning_rate": 1.3294323275987953e-05,
      "loss": 1.2815,
      "num_input_tokens_seen": 53724152,
      "step": 92600
    },
    {
      "epoch": 13.792820971105153,
      "grad_norm": 7.3125,
      "learning_rate": 1.3291452159502853e-05,
      "loss": 0.9775,
      "num_input_tokens_seen": 53726904,
      "step": 92605
    },
    {
      "epoch": 13.793565683646113,
      "grad_norm": 11.6875,
      "learning_rate": 1.3288581240821785e-05,
      "loss": 0.9373,
      "num_input_tokens_seen": 53729944,
      "step": 92610
    },
    {
      "epoch": 13.794310396187072,
      "grad_norm": 7.6875,
      "learning_rate": 1.3285710519993233e-05,
      "loss": 0.6729,
      "num_input_tokens_seen": 53732696,
      "step": 92615
    },
    {
      "epoch": 13.79505510872803,
      "grad_norm": 8.9375,
      "learning_rate": 1.3282839997065689e-05,
      "loss": 0.9378,
      "num_input_tokens_seen": 53735448,
      "step": 92620
    },
    {
      "epoch": 13.79579982126899,
      "grad_norm": 9.0,
      "learning_rate": 1.327996967208766e-05,
      "loss": 0.8036,
      "num_input_tokens_seen": 53738168,
      "step": 92625
    },
    {
      "epoch": 13.79654453380995,
      "grad_norm": 7.375,
      "learning_rate": 1.3277099545107622e-05,
      "loss": 0.8069,
      "num_input_tokens_seen": 53740984,
      "step": 92630
    },
    {
      "epoch": 13.797289246350909,
      "grad_norm": 21.375,
      "learning_rate": 1.3274229616174084e-05,
      "loss": 1.2686,
      "num_input_tokens_seen": 53744216,
      "step": 92635
    },
    {
      "epoch": 13.798033958891867,
      "grad_norm": 12.25,
      "learning_rate": 1.3271359885335515e-05,
      "loss": 1.0106,
      "num_input_tokens_seen": 53747032,
      "step": 92640
    },
    {
      "epoch": 13.798778671432826,
      "grad_norm": 12.125,
      "learning_rate": 1.3268490352640405e-05,
      "loss": 1.044,
      "num_input_tokens_seen": 53749848,
      "step": 92645
    },
    {
      "epoch": 13.799523383973787,
      "grad_norm": 12.125,
      "learning_rate": 1.3265621018137216e-05,
      "loss": 1.192,
      "num_input_tokens_seen": 53752952,
      "step": 92650
    },
    {
      "epoch": 13.800268096514746,
      "grad_norm": 15.9375,
      "learning_rate": 1.3262751881874443e-05,
      "loss": 1.3172,
      "num_input_tokens_seen": 53756344,
      "step": 92655
    },
    {
      "epoch": 13.801012809055704,
      "grad_norm": 6.21875,
      "learning_rate": 1.3259882943900547e-05,
      "loss": 1.0989,
      "num_input_tokens_seen": 53759064,
      "step": 92660
    },
    {
      "epoch": 13.801757521596663,
      "grad_norm": 9.9375,
      "learning_rate": 1.325701420426399e-05,
      "loss": 0.801,
      "num_input_tokens_seen": 53761816,
      "step": 92665
    },
    {
      "epoch": 13.802502234137624,
      "grad_norm": 6.03125,
      "learning_rate": 1.3254145663013251e-05,
      "loss": 0.8127,
      "num_input_tokens_seen": 53764824,
      "step": 92670
    },
    {
      "epoch": 13.803246946678582,
      "grad_norm": 16.875,
      "learning_rate": 1.3251277320196772e-05,
      "loss": 1.3412,
      "num_input_tokens_seen": 53768024,
      "step": 92675
    },
    {
      "epoch": 13.803991659219541,
      "grad_norm": 8.5,
      "learning_rate": 1.3248409175863033e-05,
      "loss": 1.022,
      "num_input_tokens_seen": 53770808,
      "step": 92680
    },
    {
      "epoch": 13.8047363717605,
      "grad_norm": 10.5,
      "learning_rate": 1.3245541230060465e-05,
      "loss": 1.3232,
      "num_input_tokens_seen": 53773624,
      "step": 92685
    },
    {
      "epoch": 13.80548108430146,
      "grad_norm": 9.9375,
      "learning_rate": 1.3242673482837544e-05,
      "loss": 1.0183,
      "num_input_tokens_seen": 53776856,
      "step": 92690
    },
    {
      "epoch": 13.80622579684242,
      "grad_norm": 12.1875,
      "learning_rate": 1.3239805934242704e-05,
      "loss": 1.1891,
      "num_input_tokens_seen": 53779640,
      "step": 92695
    },
    {
      "epoch": 13.806970509383378,
      "grad_norm": 7.90625,
      "learning_rate": 1.3236938584324382e-05,
      "loss": 0.587,
      "num_input_tokens_seen": 53782520,
      "step": 92700
    },
    {
      "epoch": 13.807715221924337,
      "grad_norm": 10.1875,
      "learning_rate": 1.3234071433131034e-05,
      "loss": 1.1499,
      "num_input_tokens_seen": 53785528,
      "step": 92705
    },
    {
      "epoch": 13.808459934465297,
      "grad_norm": 13.375,
      "learning_rate": 1.323120448071109e-05,
      "loss": 1.0346,
      "num_input_tokens_seen": 53788120,
      "step": 92710
    },
    {
      "epoch": 13.809204647006256,
      "grad_norm": 7.5625,
      "learning_rate": 1.3228337727112988e-05,
      "loss": 0.9097,
      "num_input_tokens_seen": 53791064,
      "step": 92715
    },
    {
      "epoch": 13.809949359547215,
      "grad_norm": 5.25,
      "learning_rate": 1.3225471172385145e-05,
      "loss": 1.401,
      "num_input_tokens_seen": 53794136,
      "step": 92720
    },
    {
      "epoch": 13.810694072088173,
      "grad_norm": 8.375,
      "learning_rate": 1.3222604816576011e-05,
      "loss": 1.2844,
      "num_input_tokens_seen": 53797144,
      "step": 92725
    },
    {
      "epoch": 13.811438784629132,
      "grad_norm": 11.75,
      "learning_rate": 1.3219738659733988e-05,
      "loss": 1.322,
      "num_input_tokens_seen": 53800120,
      "step": 92730
    },
    {
      "epoch": 13.812183497170093,
      "grad_norm": 10.625,
      "learning_rate": 1.3216872701907515e-05,
      "loss": 1.0648,
      "num_input_tokens_seen": 53803320,
      "step": 92735
    },
    {
      "epoch": 13.812928209711052,
      "grad_norm": 11.5625,
      "learning_rate": 1.3214006943145002e-05,
      "loss": 1.0897,
      "num_input_tokens_seen": 53806168,
      "step": 92740
    },
    {
      "epoch": 13.81367292225201,
      "grad_norm": 11.0,
      "learning_rate": 1.3211141383494856e-05,
      "loss": 1.0267,
      "num_input_tokens_seen": 53808824,
      "step": 92745
    },
    {
      "epoch": 13.81441763479297,
      "grad_norm": 7.375,
      "learning_rate": 1.32082760230055e-05,
      "loss": 1.094,
      "num_input_tokens_seen": 53811608,
      "step": 92750
    },
    {
      "epoch": 13.81516234733393,
      "grad_norm": 12.6875,
      "learning_rate": 1.3205410861725331e-05,
      "loss": 0.9799,
      "num_input_tokens_seen": 53814552,
      "step": 92755
    },
    {
      "epoch": 13.815907059874888,
      "grad_norm": 12.125,
      "learning_rate": 1.3202545899702768e-05,
      "loss": 1.0438,
      "num_input_tokens_seen": 53817752,
      "step": 92760
    },
    {
      "epoch": 13.816651772415847,
      "grad_norm": 9.0,
      "learning_rate": 1.3199681136986186e-05,
      "loss": 0.9129,
      "num_input_tokens_seen": 53820408,
      "step": 92765
    },
    {
      "epoch": 13.817396484956806,
      "grad_norm": 11.5,
      "learning_rate": 1.3196816573624013e-05,
      "loss": 1.1415,
      "num_input_tokens_seen": 53823416,
      "step": 92770
    },
    {
      "epoch": 13.818141197497766,
      "grad_norm": 10.875,
      "learning_rate": 1.3193952209664625e-05,
      "loss": 1.0506,
      "num_input_tokens_seen": 53826680,
      "step": 92775
    },
    {
      "epoch": 13.818885910038725,
      "grad_norm": 16.125,
      "learning_rate": 1.319108804515642e-05,
      "loss": 1.2812,
      "num_input_tokens_seen": 53829496,
      "step": 92780
    },
    {
      "epoch": 13.819630622579684,
      "grad_norm": 6.46875,
      "learning_rate": 1.3188224080147776e-05,
      "loss": 0.8361,
      "num_input_tokens_seen": 53832696,
      "step": 92785
    },
    {
      "epoch": 13.820375335120643,
      "grad_norm": 21.125,
      "learning_rate": 1.318536031468707e-05,
      "loss": 1.0686,
      "num_input_tokens_seen": 53835512,
      "step": 92790
    },
    {
      "epoch": 13.821120047661603,
      "grad_norm": 14.1875,
      "learning_rate": 1.3182496748822706e-05,
      "loss": 1.428,
      "num_input_tokens_seen": 53838328,
      "step": 92795
    },
    {
      "epoch": 13.821864760202562,
      "grad_norm": 18.625,
      "learning_rate": 1.3179633382603041e-05,
      "loss": 1.1632,
      "num_input_tokens_seen": 53840792,
      "step": 92800
    },
    {
      "epoch": 13.82260947274352,
      "grad_norm": 7.75,
      "learning_rate": 1.3176770216076462e-05,
      "loss": 0.8959,
      "num_input_tokens_seen": 53844600,
      "step": 92805
    },
    {
      "epoch": 13.82335418528448,
      "grad_norm": 6.125,
      "learning_rate": 1.3173907249291326e-05,
      "loss": 1.0201,
      "num_input_tokens_seen": 53847480,
      "step": 92810
    },
    {
      "epoch": 13.82409889782544,
      "grad_norm": 9.0625,
      "learning_rate": 1.3171044482296017e-05,
      "loss": 1.0393,
      "num_input_tokens_seen": 53850328,
      "step": 92815
    },
    {
      "epoch": 13.824843610366399,
      "grad_norm": 18.25,
      "learning_rate": 1.3168181915138889e-05,
      "loss": 1.0584,
      "num_input_tokens_seen": 53853144,
      "step": 92820
    },
    {
      "epoch": 13.825588322907358,
      "grad_norm": 8.6875,
      "learning_rate": 1.316531954786829e-05,
      "loss": 1.1289,
      "num_input_tokens_seen": 53856376,
      "step": 92825
    },
    {
      "epoch": 13.826333035448316,
      "grad_norm": 7.1875,
      "learning_rate": 1.31624573805326e-05,
      "loss": 0.7789,
      "num_input_tokens_seen": 53859288,
      "step": 92830
    },
    {
      "epoch": 13.827077747989277,
      "grad_norm": 9.6875,
      "learning_rate": 1.3159595413180164e-05,
      "loss": 1.0809,
      "num_input_tokens_seen": 53862104,
      "step": 92835
    },
    {
      "epoch": 13.827822460530236,
      "grad_norm": 9.8125,
      "learning_rate": 1.3156733645859328e-05,
      "loss": 1.3547,
      "num_input_tokens_seen": 53864728,
      "step": 92840
    },
    {
      "epoch": 13.828567173071194,
      "grad_norm": 7.59375,
      "learning_rate": 1.3153872078618428e-05,
      "loss": 0.7771,
      "num_input_tokens_seen": 53867736,
      "step": 92845
    },
    {
      "epoch": 13.829311885612153,
      "grad_norm": 8.625,
      "learning_rate": 1.3151010711505835e-05,
      "loss": 0.8325,
      "num_input_tokens_seen": 53870680,
      "step": 92850
    },
    {
      "epoch": 13.830056598153114,
      "grad_norm": 7.625,
      "learning_rate": 1.3148149544569868e-05,
      "loss": 1.3898,
      "num_input_tokens_seen": 53873816,
      "step": 92855
    },
    {
      "epoch": 13.830801310694072,
      "grad_norm": 13.125,
      "learning_rate": 1.3145288577858861e-05,
      "loss": 1.1038,
      "num_input_tokens_seen": 53876536,
      "step": 92860
    },
    {
      "epoch": 13.831546023235031,
      "grad_norm": 16.25,
      "learning_rate": 1.3142427811421165e-05,
      "loss": 0.9269,
      "num_input_tokens_seen": 53879352,
      "step": 92865
    },
    {
      "epoch": 13.83229073577599,
      "grad_norm": 9.3125,
      "learning_rate": 1.313956724530509e-05,
      "loss": 1.4949,
      "num_input_tokens_seen": 53882360,
      "step": 92870
    },
    {
      "epoch": 13.83303544831695,
      "grad_norm": 17.0,
      "learning_rate": 1.3136706879558979e-05,
      "loss": 1.2236,
      "num_input_tokens_seen": 53885336,
      "step": 92875
    },
    {
      "epoch": 13.83378016085791,
      "grad_norm": 8.75,
      "learning_rate": 1.3133846714231141e-05,
      "loss": 0.8814,
      "num_input_tokens_seen": 53888024,
      "step": 92880
    },
    {
      "epoch": 13.834524873398868,
      "grad_norm": 6.15625,
      "learning_rate": 1.3130986749369911e-05,
      "loss": 1.0455,
      "num_input_tokens_seen": 53891160,
      "step": 92885
    },
    {
      "epoch": 13.835269585939827,
      "grad_norm": 14.375,
      "learning_rate": 1.3128126985023586e-05,
      "loss": 0.9804,
      "num_input_tokens_seen": 53893944,
      "step": 92890
    },
    {
      "epoch": 13.836014298480787,
      "grad_norm": 11.125,
      "learning_rate": 1.3125267421240504e-05,
      "loss": 0.8503,
      "num_input_tokens_seen": 53896728,
      "step": 92895
    },
    {
      "epoch": 13.836759011021746,
      "grad_norm": 12.625,
      "learning_rate": 1.3122408058068955e-05,
      "loss": 1.0954,
      "num_input_tokens_seen": 53899416,
      "step": 92900
    },
    {
      "epoch": 13.837503723562705,
      "grad_norm": 17.375,
      "learning_rate": 1.3119548895557252e-05,
      "loss": 1.4502,
      "num_input_tokens_seen": 53902424,
      "step": 92905
    },
    {
      "epoch": 13.838248436103664,
      "grad_norm": 10.8125,
      "learning_rate": 1.3116689933753696e-05,
      "loss": 1.0808,
      "num_input_tokens_seen": 53905432,
      "step": 92910
    },
    {
      "epoch": 13.838993148644622,
      "grad_norm": 7.46875,
      "learning_rate": 1.3113831172706575e-05,
      "loss": 1.367,
      "num_input_tokens_seen": 53908120,
      "step": 92915
    },
    {
      "epoch": 13.839737861185583,
      "grad_norm": 14.4375,
      "learning_rate": 1.3110972612464207e-05,
      "loss": 1.0864,
      "num_input_tokens_seen": 53910840,
      "step": 92920
    },
    {
      "epoch": 13.840482573726542,
      "grad_norm": 12.6875,
      "learning_rate": 1.310811425307486e-05,
      "loss": 1.3492,
      "num_input_tokens_seen": 53913336,
      "step": 92925
    },
    {
      "epoch": 13.8412272862675,
      "grad_norm": 17.0,
      "learning_rate": 1.3105256094586849e-05,
      "loss": 1.1858,
      "num_input_tokens_seen": 53915992,
      "step": 92930
    },
    {
      "epoch": 13.84197199880846,
      "grad_norm": 11.25,
      "learning_rate": 1.310239813704845e-05,
      "loss": 1.1134,
      "num_input_tokens_seen": 53918776,
      "step": 92935
    },
    {
      "epoch": 13.84271671134942,
      "grad_norm": 14.25,
      "learning_rate": 1.3099540380507927e-05,
      "loss": 1.0428,
      "num_input_tokens_seen": 53921560,
      "step": 92940
    },
    {
      "epoch": 13.843461423890378,
      "grad_norm": 12.5,
      "learning_rate": 1.3096682825013584e-05,
      "loss": 1.2513,
      "num_input_tokens_seen": 53924312,
      "step": 92945
    },
    {
      "epoch": 13.844206136431337,
      "grad_norm": 5.75,
      "learning_rate": 1.309382547061368e-05,
      "loss": 1.4531,
      "num_input_tokens_seen": 53927064,
      "step": 92950
    },
    {
      "epoch": 13.844950848972296,
      "grad_norm": 5.59375,
      "learning_rate": 1.3090968317356502e-05,
      "loss": 0.8805,
      "num_input_tokens_seen": 53930168,
      "step": 92955
    },
    {
      "epoch": 13.845695561513256,
      "grad_norm": 11.8125,
      "learning_rate": 1.3088111365290302e-05,
      "loss": 1.2169,
      "num_input_tokens_seen": 53933048,
      "step": 92960
    },
    {
      "epoch": 13.846440274054215,
      "grad_norm": 5.65625,
      "learning_rate": 1.3085254614463362e-05,
      "loss": 1.0148,
      "num_input_tokens_seen": 53936120,
      "step": 92965
    },
    {
      "epoch": 13.847184986595174,
      "grad_norm": 12.4375,
      "learning_rate": 1.308239806492394e-05,
      "loss": 1.7166,
      "num_input_tokens_seen": 53938712,
      "step": 92970
    },
    {
      "epoch": 13.847929699136133,
      "grad_norm": 8.0625,
      "learning_rate": 1.3079541716720284e-05,
      "loss": 1.0928,
      "num_input_tokens_seen": 53941304,
      "step": 92975
    },
    {
      "epoch": 13.848674411677093,
      "grad_norm": 5.53125,
      "learning_rate": 1.307668556990066e-05,
      "loss": 0.8974,
      "num_input_tokens_seen": 53944472,
      "step": 92980
    },
    {
      "epoch": 13.849419124218052,
      "grad_norm": 8.8125,
      "learning_rate": 1.30738296245133e-05,
      "loss": 0.9246,
      "num_input_tokens_seen": 53947544,
      "step": 92985
    },
    {
      "epoch": 13.85016383675901,
      "grad_norm": 9.5625,
      "learning_rate": 1.3070973880606482e-05,
      "loss": 1.0675,
      "num_input_tokens_seen": 53950840,
      "step": 92990
    },
    {
      "epoch": 13.85090854929997,
      "grad_norm": 6.15625,
      "learning_rate": 1.3068118338228425e-05,
      "loss": 0.786,
      "num_input_tokens_seen": 53953912,
      "step": 92995
    },
    {
      "epoch": 13.85165326184093,
      "grad_norm": 5.5,
      "learning_rate": 1.306526299742739e-05,
      "loss": 0.9943,
      "num_input_tokens_seen": 53956856,
      "step": 93000
    },
    {
      "epoch": 13.852397974381889,
      "grad_norm": 6.96875,
      "learning_rate": 1.3062407858251598e-05,
      "loss": 0.9089,
      "num_input_tokens_seen": 53959416,
      "step": 93005
    },
    {
      "epoch": 13.853142686922848,
      "grad_norm": 11.3125,
      "learning_rate": 1.3059552920749301e-05,
      "loss": 1.3381,
      "num_input_tokens_seen": 53962744,
      "step": 93010
    },
    {
      "epoch": 13.853887399463806,
      "grad_norm": 11.5625,
      "learning_rate": 1.3056698184968714e-05,
      "loss": 1.2196,
      "num_input_tokens_seen": 53965784,
      "step": 93015
    },
    {
      "epoch": 13.854632112004767,
      "grad_norm": 14.75,
      "learning_rate": 1.305384365095808e-05,
      "loss": 1.5723,
      "num_input_tokens_seen": 53968600,
      "step": 93020
    },
    {
      "epoch": 13.855376824545726,
      "grad_norm": 7.9375,
      "learning_rate": 1.305098931876562e-05,
      "loss": 1.2463,
      "num_input_tokens_seen": 53971192,
      "step": 93025
    },
    {
      "epoch": 13.856121537086684,
      "grad_norm": 12.625,
      "learning_rate": 1.3048135188439537e-05,
      "loss": 1.3613,
      "num_input_tokens_seen": 53974040,
      "step": 93030
    },
    {
      "epoch": 13.856866249627643,
      "grad_norm": 9.125,
      "learning_rate": 1.3045281260028075e-05,
      "loss": 0.8332,
      "num_input_tokens_seen": 53977080,
      "step": 93035
    },
    {
      "epoch": 13.857610962168604,
      "grad_norm": 8.9375,
      "learning_rate": 1.3042427533579435e-05,
      "loss": 0.921,
      "num_input_tokens_seen": 53979992,
      "step": 93040
    },
    {
      "epoch": 13.858355674709562,
      "grad_norm": 9.4375,
      "learning_rate": 1.303957400914183e-05,
      "loss": 0.7953,
      "num_input_tokens_seen": 53983000,
      "step": 93045
    },
    {
      "epoch": 13.859100387250521,
      "grad_norm": 8.0,
      "learning_rate": 1.3036720686763454e-05,
      "loss": 1.1518,
      "num_input_tokens_seen": 53985944,
      "step": 93050
    },
    {
      "epoch": 13.85984509979148,
      "grad_norm": 12.25,
      "learning_rate": 1.3033867566492534e-05,
      "loss": 1.4188,
      "num_input_tokens_seen": 53989048,
      "step": 93055
    },
    {
      "epoch": 13.86058981233244,
      "grad_norm": 9.25,
      "learning_rate": 1.303101464837726e-05,
      "loss": 1.214,
      "num_input_tokens_seen": 53991832,
      "step": 93060
    },
    {
      "epoch": 13.8613345248734,
      "grad_norm": 9.5625,
      "learning_rate": 1.3028161932465815e-05,
      "loss": 1.0023,
      "num_input_tokens_seen": 53994584,
      "step": 93065
    },
    {
      "epoch": 13.862079237414358,
      "grad_norm": 10.375,
      "learning_rate": 1.3025309418806422e-05,
      "loss": 1.2633,
      "num_input_tokens_seen": 53997848,
      "step": 93070
    },
    {
      "epoch": 13.862823949955317,
      "grad_norm": 10.5,
      "learning_rate": 1.3022457107447244e-05,
      "loss": 1.4122,
      "num_input_tokens_seen": 54000856,
      "step": 93075
    },
    {
      "epoch": 13.863568662496277,
      "grad_norm": 12.75,
      "learning_rate": 1.3019604998436491e-05,
      "loss": 0.9034,
      "num_input_tokens_seen": 54004024,
      "step": 93080
    },
    {
      "epoch": 13.864313375037236,
      "grad_norm": 19.375,
      "learning_rate": 1.301675309182232e-05,
      "loss": 1.226,
      "num_input_tokens_seen": 54007064,
      "step": 93085
    },
    {
      "epoch": 13.865058087578195,
      "grad_norm": 10.875,
      "learning_rate": 1.3013901387652941e-05,
      "loss": 0.8821,
      "num_input_tokens_seen": 54010328,
      "step": 93090
    },
    {
      "epoch": 13.865802800119154,
      "grad_norm": 10.25,
      "learning_rate": 1.3011049885976505e-05,
      "loss": 0.8001,
      "num_input_tokens_seen": 54013272,
      "step": 93095
    },
    {
      "epoch": 13.866547512660112,
      "grad_norm": 6.46875,
      "learning_rate": 1.3008198586841209e-05,
      "loss": 1.2298,
      "num_input_tokens_seen": 54016088,
      "step": 93100
    },
    {
      "epoch": 13.867292225201073,
      "grad_norm": 11.9375,
      "learning_rate": 1.3005347490295205e-05,
      "loss": 1.257,
      "num_input_tokens_seen": 54019032,
      "step": 93105
    },
    {
      "epoch": 13.868036937742032,
      "grad_norm": 7.15625,
      "learning_rate": 1.3002496596386666e-05,
      "loss": 1.3327,
      "num_input_tokens_seen": 54022072,
      "step": 93110
    },
    {
      "epoch": 13.86878165028299,
      "grad_norm": 13.9375,
      "learning_rate": 1.2999645905163754e-05,
      "loss": 1.1003,
      "num_input_tokens_seen": 54025144,
      "step": 93115
    },
    {
      "epoch": 13.86952636282395,
      "grad_norm": 10.1875,
      "learning_rate": 1.2996795416674618e-05,
      "loss": 0.9063,
      "num_input_tokens_seen": 54028056,
      "step": 93120
    },
    {
      "epoch": 13.87027107536491,
      "grad_norm": 6.75,
      "learning_rate": 1.2993945130967434e-05,
      "loss": 1.0194,
      "num_input_tokens_seen": 54031000,
      "step": 93125
    },
    {
      "epoch": 13.871015787905868,
      "grad_norm": 11.875,
      "learning_rate": 1.2991095048090333e-05,
      "loss": 1.0249,
      "num_input_tokens_seen": 54033784,
      "step": 93130
    },
    {
      "epoch": 13.871760500446827,
      "grad_norm": 14.8125,
      "learning_rate": 1.2988245168091485e-05,
      "loss": 1.2811,
      "num_input_tokens_seen": 54036664,
      "step": 93135
    },
    {
      "epoch": 13.872505212987786,
      "grad_norm": 6.8125,
      "learning_rate": 1.2985395491019029e-05,
      "loss": 1.4849,
      "num_input_tokens_seen": 54039512,
      "step": 93140
    },
    {
      "epoch": 13.873249925528746,
      "grad_norm": 5.25,
      "learning_rate": 1.2982546016921093e-05,
      "loss": 0.9157,
      "num_input_tokens_seen": 54042584,
      "step": 93145
    },
    {
      "epoch": 13.873994638069705,
      "grad_norm": 10.5625,
      "learning_rate": 1.297969674584584e-05,
      "loss": 0.8999,
      "num_input_tokens_seen": 54045240,
      "step": 93150
    },
    {
      "epoch": 13.874739350610664,
      "grad_norm": 15.875,
      "learning_rate": 1.2976847677841383e-05,
      "loss": 1.1553,
      "num_input_tokens_seen": 54048344,
      "step": 93155
    },
    {
      "epoch": 13.875484063151623,
      "grad_norm": 17.25,
      "learning_rate": 1.2973998812955876e-05,
      "loss": 1.1383,
      "num_input_tokens_seen": 54051128,
      "step": 93160
    },
    {
      "epoch": 13.876228775692583,
      "grad_norm": 6.3125,
      "learning_rate": 1.2971150151237435e-05,
      "loss": 1.1175,
      "num_input_tokens_seen": 54053848,
      "step": 93165
    },
    {
      "epoch": 13.876973488233542,
      "grad_norm": 14.0625,
      "learning_rate": 1.2968301692734187e-05,
      "loss": 1.2297,
      "num_input_tokens_seen": 54056856,
      "step": 93170
    },
    {
      "epoch": 13.8777182007745,
      "grad_norm": 9.1875,
      "learning_rate": 1.2965453437494243e-05,
      "loss": 1.008,
      "num_input_tokens_seen": 54059832,
      "step": 93175
    },
    {
      "epoch": 13.87846291331546,
      "grad_norm": 7.96875,
      "learning_rate": 1.296260538556574e-05,
      "loss": 1.1252,
      "num_input_tokens_seen": 54062872,
      "step": 93180
    },
    {
      "epoch": 13.87920762585642,
      "grad_norm": 7.0,
      "learning_rate": 1.295975753699679e-05,
      "loss": 1.0214,
      "num_input_tokens_seen": 54065784,
      "step": 93185
    },
    {
      "epoch": 13.879952338397379,
      "grad_norm": 17.875,
      "learning_rate": 1.2956909891835484e-05,
      "loss": 0.8639,
      "num_input_tokens_seen": 54068568,
      "step": 93190
    },
    {
      "epoch": 13.880697050938338,
      "grad_norm": 12.0625,
      "learning_rate": 1.2954062450129959e-05,
      "loss": 1.1302,
      "num_input_tokens_seen": 54071544,
      "step": 93195
    },
    {
      "epoch": 13.881441763479296,
      "grad_norm": 7.5625,
      "learning_rate": 1.2951215211928292e-05,
      "loss": 1.0111,
      "num_input_tokens_seen": 54074584,
      "step": 93200
    },
    {
      "epoch": 13.882186476020257,
      "grad_norm": 12.5625,
      "learning_rate": 1.2948368177278614e-05,
      "loss": 1.4055,
      "num_input_tokens_seen": 54077336,
      "step": 93205
    },
    {
      "epoch": 13.882931188561216,
      "grad_norm": 14.0,
      "learning_rate": 1.2945521346228989e-05,
      "loss": 1.0104,
      "num_input_tokens_seen": 54080088,
      "step": 93210
    },
    {
      "epoch": 13.883675901102174,
      "grad_norm": 12.3125,
      "learning_rate": 1.2942674718827546e-05,
      "loss": 0.9351,
      "num_input_tokens_seen": 54083224,
      "step": 93215
    },
    {
      "epoch": 13.884420613643133,
      "grad_norm": 12.4375,
      "learning_rate": 1.2939828295122358e-05,
      "loss": 1.5939,
      "num_input_tokens_seen": 54086360,
      "step": 93220
    },
    {
      "epoch": 13.885165326184094,
      "grad_norm": 14.875,
      "learning_rate": 1.2936982075161502e-05,
      "loss": 1.1753,
      "num_input_tokens_seen": 54089400,
      "step": 93225
    },
    {
      "epoch": 13.885910038725052,
      "grad_norm": 4.125,
      "learning_rate": 1.2934136058993082e-05,
      "loss": 1.1923,
      "num_input_tokens_seen": 54092024,
      "step": 93230
    },
    {
      "epoch": 13.886654751266011,
      "grad_norm": 9.625,
      "learning_rate": 1.2931290246665173e-05,
      "loss": 1.0138,
      "num_input_tokens_seen": 54094840,
      "step": 93235
    },
    {
      "epoch": 13.88739946380697,
      "grad_norm": 9.4375,
      "learning_rate": 1.2928444638225848e-05,
      "loss": 0.9402,
      "num_input_tokens_seen": 54097592,
      "step": 93240
    },
    {
      "epoch": 13.88814417634793,
      "grad_norm": 13.375,
      "learning_rate": 1.2925599233723174e-05,
      "loss": 1.3167,
      "num_input_tokens_seen": 54100312,
      "step": 93245
    },
    {
      "epoch": 13.88888888888889,
      "grad_norm": 9.9375,
      "learning_rate": 1.2922754033205237e-05,
      "loss": 0.9027,
      "num_input_tokens_seen": 54103160,
      "step": 93250
    },
    {
      "epoch": 13.889633601429848,
      "grad_norm": 11.3125,
      "learning_rate": 1.2919909036720085e-05,
      "loss": 0.8129,
      "num_input_tokens_seen": 54105976,
      "step": 93255
    },
    {
      "epoch": 13.890378313970807,
      "grad_norm": 6.5,
      "learning_rate": 1.2917064244315802e-05,
      "loss": 1.1831,
      "num_input_tokens_seen": 54108664,
      "step": 93260
    },
    {
      "epoch": 13.891123026511767,
      "grad_norm": 10.5,
      "learning_rate": 1.2914219656040437e-05,
      "loss": 1.1854,
      "num_input_tokens_seen": 54111448,
      "step": 93265
    },
    {
      "epoch": 13.891867739052726,
      "grad_norm": 13.25,
      "learning_rate": 1.2911375271942042e-05,
      "loss": 0.79,
      "num_input_tokens_seen": 54114392,
      "step": 93270
    },
    {
      "epoch": 13.892612451593685,
      "grad_norm": 4.875,
      "learning_rate": 1.2908531092068682e-05,
      "loss": 0.9301,
      "num_input_tokens_seen": 54117432,
      "step": 93275
    },
    {
      "epoch": 13.893357164134644,
      "grad_norm": 6.40625,
      "learning_rate": 1.290568711646839e-05,
      "loss": 1.0512,
      "num_input_tokens_seen": 54120568,
      "step": 93280
    },
    {
      "epoch": 13.894101876675602,
      "grad_norm": 6.375,
      "learning_rate": 1.2902843345189237e-05,
      "loss": 1.0055,
      "num_input_tokens_seen": 54123288,
      "step": 93285
    },
    {
      "epoch": 13.894846589216563,
      "grad_norm": 14.0625,
      "learning_rate": 1.2899999778279235e-05,
      "loss": 1.2812,
      "num_input_tokens_seen": 54126616,
      "step": 93290
    },
    {
      "epoch": 13.895591301757522,
      "grad_norm": 8.5625,
      "learning_rate": 1.289715641578645e-05,
      "loss": 1.23,
      "num_input_tokens_seen": 54129592,
      "step": 93295
    },
    {
      "epoch": 13.89633601429848,
      "grad_norm": 12.375,
      "learning_rate": 1.2894313257758906e-05,
      "loss": 1.4976,
      "num_input_tokens_seen": 54132280,
      "step": 93300
    },
    {
      "epoch": 13.89708072683944,
      "grad_norm": 11.6875,
      "learning_rate": 1.2891470304244638e-05,
      "loss": 0.9528,
      "num_input_tokens_seen": 54135192,
      "step": 93305
    },
    {
      "epoch": 13.8978254393804,
      "grad_norm": 6.59375,
      "learning_rate": 1.288862755529167e-05,
      "loss": 0.9588,
      "num_input_tokens_seen": 54138040,
      "step": 93310
    },
    {
      "epoch": 13.898570151921358,
      "grad_norm": 13.3125,
      "learning_rate": 1.2885785010948023e-05,
      "loss": 0.9685,
      "num_input_tokens_seen": 54140984,
      "step": 93315
    },
    {
      "epoch": 13.899314864462317,
      "grad_norm": 10.1875,
      "learning_rate": 1.2882942671261733e-05,
      "loss": 1.2032,
      "num_input_tokens_seen": 54143768,
      "step": 93320
    },
    {
      "epoch": 13.900059577003276,
      "grad_norm": 12.3125,
      "learning_rate": 1.2880100536280803e-05,
      "loss": 1.5192,
      "num_input_tokens_seen": 54147032,
      "step": 93325
    },
    {
      "epoch": 13.900804289544237,
      "grad_norm": 9.0,
      "learning_rate": 1.2877258606053266e-05,
      "loss": 0.8484,
      "num_input_tokens_seen": 54150264,
      "step": 93330
    },
    {
      "epoch": 13.901549002085195,
      "grad_norm": 5.46875,
      "learning_rate": 1.2874416880627116e-05,
      "loss": 1.1357,
      "num_input_tokens_seen": 54153208,
      "step": 93335
    },
    {
      "epoch": 13.902293714626154,
      "grad_norm": 16.125,
      "learning_rate": 1.2871575360050376e-05,
      "loss": 0.8348,
      "num_input_tokens_seen": 54156280,
      "step": 93340
    },
    {
      "epoch": 13.903038427167113,
      "grad_norm": 15.5625,
      "learning_rate": 1.2868734044371044e-05,
      "loss": 1.1147,
      "num_input_tokens_seen": 54159224,
      "step": 93345
    },
    {
      "epoch": 13.903783139708073,
      "grad_norm": 17.5,
      "learning_rate": 1.2865892933637114e-05,
      "loss": 1.4166,
      "num_input_tokens_seen": 54162552,
      "step": 93350
    },
    {
      "epoch": 13.904527852249032,
      "grad_norm": 9.8125,
      "learning_rate": 1.2863052027896597e-05,
      "loss": 1.184,
      "num_input_tokens_seen": 54165336,
      "step": 93355
    },
    {
      "epoch": 13.90527256478999,
      "grad_norm": 5.34375,
      "learning_rate": 1.2860211327197468e-05,
      "loss": 1.0225,
      "num_input_tokens_seen": 54168344,
      "step": 93360
    },
    {
      "epoch": 13.90601727733095,
      "grad_norm": 17.625,
      "learning_rate": 1.2857370831587745e-05,
      "loss": 1.2175,
      "num_input_tokens_seen": 54171224,
      "step": 93365
    },
    {
      "epoch": 13.90676198987191,
      "grad_norm": 11.6875,
      "learning_rate": 1.28545305411154e-05,
      "loss": 1.1178,
      "num_input_tokens_seen": 54174072,
      "step": 93370
    },
    {
      "epoch": 13.907506702412869,
      "grad_norm": 14.5,
      "learning_rate": 1.2851690455828414e-05,
      "loss": 1.0304,
      "num_input_tokens_seen": 54177144,
      "step": 93375
    },
    {
      "epoch": 13.908251414953828,
      "grad_norm": 6.03125,
      "learning_rate": 1.2848850575774774e-05,
      "loss": 1.1706,
      "num_input_tokens_seen": 54179960,
      "step": 93380
    },
    {
      "epoch": 13.908996127494786,
      "grad_norm": 10.0625,
      "learning_rate": 1.2846010901002442e-05,
      "loss": 0.9543,
      "num_input_tokens_seen": 54182872,
      "step": 93385
    },
    {
      "epoch": 13.909740840035747,
      "grad_norm": 12.875,
      "learning_rate": 1.2843171431559414e-05,
      "loss": 1.0102,
      "num_input_tokens_seen": 54185624,
      "step": 93390
    },
    {
      "epoch": 13.910485552576706,
      "grad_norm": 11.25,
      "learning_rate": 1.284033216749364e-05,
      "loss": 1.2637,
      "num_input_tokens_seen": 54188344,
      "step": 93395
    },
    {
      "epoch": 13.911230265117664,
      "grad_norm": 5.875,
      "learning_rate": 1.2837493108853105e-05,
      "loss": 0.8773,
      "num_input_tokens_seen": 54191256,
      "step": 93400
    },
    {
      "epoch": 13.911974977658623,
      "grad_norm": 5.21875,
      "learning_rate": 1.2834654255685752e-05,
      "loss": 1.1241,
      "num_input_tokens_seen": 54194552,
      "step": 93405
    },
    {
      "epoch": 13.912719690199584,
      "grad_norm": 4.71875,
      "learning_rate": 1.283181560803956e-05,
      "loss": 1.0488,
      "num_input_tokens_seen": 54197144,
      "step": 93410
    },
    {
      "epoch": 13.913464402740543,
      "grad_norm": 9.75,
      "learning_rate": 1.282897716596247e-05,
      "loss": 0.9308,
      "num_input_tokens_seen": 54199928,
      "step": 93415
    },
    {
      "epoch": 13.914209115281501,
      "grad_norm": 11.4375,
      "learning_rate": 1.2826138929502446e-05,
      "loss": 0.7642,
      "num_input_tokens_seen": 54202872,
      "step": 93420
    },
    {
      "epoch": 13.91495382782246,
      "grad_norm": 12.8125,
      "learning_rate": 1.2823300898707432e-05,
      "loss": 1.2512,
      "num_input_tokens_seen": 54205464,
      "step": 93425
    },
    {
      "epoch": 13.915698540363419,
      "grad_norm": 12.4375,
      "learning_rate": 1.2820463073625367e-05,
      "loss": 1.2519,
      "num_input_tokens_seen": 54208152,
      "step": 93430
    },
    {
      "epoch": 13.91644325290438,
      "grad_norm": 22.875,
      "learning_rate": 1.2817625454304204e-05,
      "loss": 1.1573,
      "num_input_tokens_seen": 54211000,
      "step": 93435
    },
    {
      "epoch": 13.917187965445338,
      "grad_norm": 8.5,
      "learning_rate": 1.281478804079188e-05,
      "loss": 1.1784,
      "num_input_tokens_seen": 54213528,
      "step": 93440
    },
    {
      "epoch": 13.917932677986297,
      "grad_norm": 10.25,
      "learning_rate": 1.2811950833136332e-05,
      "loss": 1.0272,
      "num_input_tokens_seen": 54216120,
      "step": 93445
    },
    {
      "epoch": 13.918677390527257,
      "grad_norm": 11.25,
      "learning_rate": 1.2809113831385472e-05,
      "loss": 1.1779,
      "num_input_tokens_seen": 54218968,
      "step": 93450
    },
    {
      "epoch": 13.919422103068216,
      "grad_norm": 11.1875,
      "learning_rate": 1.2806277035587256e-05,
      "loss": 0.8529,
      "num_input_tokens_seen": 54221720,
      "step": 93455
    },
    {
      "epoch": 13.920166815609175,
      "grad_norm": 8.4375,
      "learning_rate": 1.2803440445789594e-05,
      "loss": 1.5401,
      "num_input_tokens_seen": 54224248,
      "step": 93460
    },
    {
      "epoch": 13.920911528150134,
      "grad_norm": 10.1875,
      "learning_rate": 1.2800604062040403e-05,
      "loss": 0.854,
      "num_input_tokens_seen": 54227000,
      "step": 93465
    },
    {
      "epoch": 13.921656240691092,
      "grad_norm": 10.4375,
      "learning_rate": 1.2797767884387615e-05,
      "loss": 1.0801,
      "num_input_tokens_seen": 54229816,
      "step": 93470
    },
    {
      "epoch": 13.922400953232053,
      "grad_norm": 9.1875,
      "learning_rate": 1.2794931912879127e-05,
      "loss": 1.299,
      "num_input_tokens_seen": 54232824,
      "step": 93475
    },
    {
      "epoch": 13.923145665773012,
      "grad_norm": 10.3125,
      "learning_rate": 1.2792096147562872e-05,
      "loss": 1.0697,
      "num_input_tokens_seen": 54235640,
      "step": 93480
    },
    {
      "epoch": 13.92389037831397,
      "grad_norm": 7.09375,
      "learning_rate": 1.2789260588486735e-05,
      "loss": 1.2657,
      "num_input_tokens_seen": 54238392,
      "step": 93485
    },
    {
      "epoch": 13.92463509085493,
      "grad_norm": 7.8125,
      "learning_rate": 1.2786425235698634e-05,
      "loss": 0.82,
      "num_input_tokens_seen": 54241336,
      "step": 93490
    },
    {
      "epoch": 13.92537980339589,
      "grad_norm": 16.5,
      "learning_rate": 1.2783590089246473e-05,
      "loss": 1.0661,
      "num_input_tokens_seen": 54244504,
      "step": 93495
    },
    {
      "epoch": 13.926124515936849,
      "grad_norm": 16.75,
      "learning_rate": 1.2780755149178136e-05,
      "loss": 1.1844,
      "num_input_tokens_seen": 54247320,
      "step": 93500
    },
    {
      "epoch": 13.926869228477807,
      "grad_norm": 8.1875,
      "learning_rate": 1.2777920415541514e-05,
      "loss": 1.0941,
      "num_input_tokens_seen": 54250456,
      "step": 93505
    },
    {
      "epoch": 13.927613941018766,
      "grad_norm": 8.375,
      "learning_rate": 1.2775085888384514e-05,
      "loss": 0.8994,
      "num_input_tokens_seen": 54253432,
      "step": 93510
    },
    {
      "epoch": 13.928358653559727,
      "grad_norm": 5.15625,
      "learning_rate": 1.2772251567755011e-05,
      "loss": 1.08,
      "num_input_tokens_seen": 54256152,
      "step": 93515
    },
    {
      "epoch": 13.929103366100685,
      "grad_norm": 11.5,
      "learning_rate": 1.2769417453700882e-05,
      "loss": 1.1134,
      "num_input_tokens_seen": 54259096,
      "step": 93520
    },
    {
      "epoch": 13.929848078641644,
      "grad_norm": 12.75,
      "learning_rate": 1.2766583546270027e-05,
      "loss": 1.1644,
      "num_input_tokens_seen": 54262232,
      "step": 93525
    },
    {
      "epoch": 13.930592791182603,
      "grad_norm": 20.875,
      "learning_rate": 1.2763749845510297e-05,
      "loss": 1.2363,
      "num_input_tokens_seen": 54265368,
      "step": 93530
    },
    {
      "epoch": 13.931337503723563,
      "grad_norm": 9.875,
      "learning_rate": 1.2760916351469588e-05,
      "loss": 0.9087,
      "num_input_tokens_seen": 54268120,
      "step": 93535
    },
    {
      "epoch": 13.932082216264522,
      "grad_norm": 7.3125,
      "learning_rate": 1.2758083064195756e-05,
      "loss": 0.9717,
      "num_input_tokens_seen": 54270776,
      "step": 93540
    },
    {
      "epoch": 13.932826928805481,
      "grad_norm": 8.5,
      "learning_rate": 1.2755249983736662e-05,
      "loss": 1.0415,
      "num_input_tokens_seen": 54273752,
      "step": 93545
    },
    {
      "epoch": 13.93357164134644,
      "grad_norm": 16.375,
      "learning_rate": 1.275241711014018e-05,
      "loss": 1.1515,
      "num_input_tokens_seen": 54276632,
      "step": 93550
    },
    {
      "epoch": 13.9343163538874,
      "grad_norm": 15.75,
      "learning_rate": 1.2749584443454154e-05,
      "loss": 0.9485,
      "num_input_tokens_seen": 54279640,
      "step": 93555
    },
    {
      "epoch": 13.935061066428359,
      "grad_norm": 6.34375,
      "learning_rate": 1.2746751983726459e-05,
      "loss": 1.2799,
      "num_input_tokens_seen": 54282488,
      "step": 93560
    },
    {
      "epoch": 13.935805778969318,
      "grad_norm": 11.375,
      "learning_rate": 1.2743919731004938e-05,
      "loss": 0.9657,
      "num_input_tokens_seen": 54285272,
      "step": 93565
    },
    {
      "epoch": 13.936550491510276,
      "grad_norm": 10.9375,
      "learning_rate": 1.2741087685337432e-05,
      "loss": 1.2188,
      "num_input_tokens_seen": 54288600,
      "step": 93570
    },
    {
      "epoch": 13.937295204051237,
      "grad_norm": 7.21875,
      "learning_rate": 1.2738255846771785e-05,
      "loss": 1.2021,
      "num_input_tokens_seen": 54291576,
      "step": 93575
    },
    {
      "epoch": 13.938039916592196,
      "grad_norm": 10.25,
      "learning_rate": 1.273542421535585e-05,
      "loss": 1.2897,
      "num_input_tokens_seen": 54294488,
      "step": 93580
    },
    {
      "epoch": 13.938784629133155,
      "grad_norm": 10.25,
      "learning_rate": 1.273259279113746e-05,
      "loss": 1.1859,
      "num_input_tokens_seen": 54297208,
      "step": 93585
    },
    {
      "epoch": 13.939529341674113,
      "grad_norm": 17.625,
      "learning_rate": 1.2729761574164434e-05,
      "loss": 0.9827,
      "num_input_tokens_seen": 54300184,
      "step": 93590
    },
    {
      "epoch": 13.940274054215074,
      "grad_norm": 15.0625,
      "learning_rate": 1.2726930564484627e-05,
      "loss": 1.5448,
      "num_input_tokens_seen": 54303096,
      "step": 93595
    },
    {
      "epoch": 13.941018766756033,
      "grad_norm": 11.4375,
      "learning_rate": 1.2724099762145841e-05,
      "loss": 1.0354,
      "num_input_tokens_seen": 54305848,
      "step": 93600
    },
    {
      "epoch": 13.941763479296991,
      "grad_norm": 17.75,
      "learning_rate": 1.2721269167195926e-05,
      "loss": 1.0086,
      "num_input_tokens_seen": 54308760,
      "step": 93605
    },
    {
      "epoch": 13.94250819183795,
      "grad_norm": 16.0,
      "learning_rate": 1.2718438779682678e-05,
      "loss": 1.2213,
      "num_input_tokens_seen": 54311928,
      "step": 93610
    },
    {
      "epoch": 13.943252904378909,
      "grad_norm": 10.6875,
      "learning_rate": 1.2715608599653938e-05,
      "loss": 1.298,
      "num_input_tokens_seen": 54315000,
      "step": 93615
    },
    {
      "epoch": 13.94399761691987,
      "grad_norm": 14.75,
      "learning_rate": 1.271277862715749e-05,
      "loss": 1.2042,
      "num_input_tokens_seen": 54317688,
      "step": 93620
    },
    {
      "epoch": 13.944742329460828,
      "grad_norm": 8.5,
      "learning_rate": 1.2709948862241173e-05,
      "loss": 1.2041,
      "num_input_tokens_seen": 54320568,
      "step": 93625
    },
    {
      "epoch": 13.945487042001787,
      "grad_norm": 23.25,
      "learning_rate": 1.2707119304952777e-05,
      "loss": 0.9836,
      "num_input_tokens_seen": 54323512,
      "step": 93630
    },
    {
      "epoch": 13.946231754542747,
      "grad_norm": 9.3125,
      "learning_rate": 1.2704289955340107e-05,
      "loss": 0.8592,
      "num_input_tokens_seen": 54326296,
      "step": 93635
    },
    {
      "epoch": 13.946976467083706,
      "grad_norm": 11.875,
      "learning_rate": 1.270146081345096e-05,
      "loss": 1.4762,
      "num_input_tokens_seen": 54328952,
      "step": 93640
    },
    {
      "epoch": 13.947721179624665,
      "grad_norm": 12.6875,
      "learning_rate": 1.2698631879333126e-05,
      "loss": 0.8126,
      "num_input_tokens_seen": 54331736,
      "step": 93645
    },
    {
      "epoch": 13.948465892165624,
      "grad_norm": 16.75,
      "learning_rate": 1.2695803153034411e-05,
      "loss": 1.2257,
      "num_input_tokens_seen": 54334712,
      "step": 93650
    },
    {
      "epoch": 13.949210604706582,
      "grad_norm": 18.625,
      "learning_rate": 1.2692974634602586e-05,
      "loss": 1.5369,
      "num_input_tokens_seen": 54337560,
      "step": 93655
    },
    {
      "epoch": 13.949955317247543,
      "grad_norm": 6.9375,
      "learning_rate": 1.2690146324085458e-05,
      "loss": 1.0811,
      "num_input_tokens_seen": 54340344,
      "step": 93660
    },
    {
      "epoch": 13.950700029788502,
      "grad_norm": 15.5,
      "learning_rate": 1.2687318221530797e-05,
      "loss": 1.4666,
      "num_input_tokens_seen": 54343224,
      "step": 93665
    },
    {
      "epoch": 13.95144474232946,
      "grad_norm": 6.6875,
      "learning_rate": 1.268449032698637e-05,
      "loss": 1.1662,
      "num_input_tokens_seen": 54346296,
      "step": 93670
    },
    {
      "epoch": 13.95218945487042,
      "grad_norm": 13.6875,
      "learning_rate": 1.2681662640499969e-05,
      "loss": 1.2313,
      "num_input_tokens_seen": 54349304,
      "step": 93675
    },
    {
      "epoch": 13.95293416741138,
      "grad_norm": 6.71875,
      "learning_rate": 1.2678835162119352e-05,
      "loss": 0.8626,
      "num_input_tokens_seen": 54352472,
      "step": 93680
    },
    {
      "epoch": 13.953678879952339,
      "grad_norm": 11.5,
      "learning_rate": 1.26760078918923e-05,
      "loss": 1.1782,
      "num_input_tokens_seen": 54355352,
      "step": 93685
    },
    {
      "epoch": 13.954423592493297,
      "grad_norm": 8.875,
      "learning_rate": 1.267318082986656e-05,
      "loss": 1.0338,
      "num_input_tokens_seen": 54358232,
      "step": 93690
    },
    {
      "epoch": 13.955168305034256,
      "grad_norm": 16.25,
      "learning_rate": 1.267035397608991e-05,
      "loss": 1.2587,
      "num_input_tokens_seen": 54361176,
      "step": 93695
    },
    {
      "epoch": 13.955913017575217,
      "grad_norm": 8.3125,
      "learning_rate": 1.2667527330610101e-05,
      "loss": 1.0916,
      "num_input_tokens_seen": 54364184,
      "step": 93700
    },
    {
      "epoch": 13.956657730116175,
      "grad_norm": 8.8125,
      "learning_rate": 1.2664700893474884e-05,
      "loss": 0.8084,
      "num_input_tokens_seen": 54366904,
      "step": 93705
    },
    {
      "epoch": 13.957402442657134,
      "grad_norm": 11.0625,
      "learning_rate": 1.2661874664732004e-05,
      "loss": 0.9425,
      "num_input_tokens_seen": 54369848,
      "step": 93710
    },
    {
      "epoch": 13.958147155198093,
      "grad_norm": 17.0,
      "learning_rate": 1.2659048644429205e-05,
      "loss": 1.0045,
      "num_input_tokens_seen": 54372760,
      "step": 93715
    },
    {
      "epoch": 13.958891867739053,
      "grad_norm": 18.125,
      "learning_rate": 1.2656222832614245e-05,
      "loss": 1.2122,
      "num_input_tokens_seen": 54375544,
      "step": 93720
    },
    {
      "epoch": 13.959636580280012,
      "grad_norm": 7.6875,
      "learning_rate": 1.2653397229334846e-05,
      "loss": 1.0963,
      "num_input_tokens_seen": 54378680,
      "step": 93725
    },
    {
      "epoch": 13.960381292820971,
      "grad_norm": 19.25,
      "learning_rate": 1.2650571834638764e-05,
      "loss": 0.9352,
      "num_input_tokens_seen": 54381592,
      "step": 93730
    },
    {
      "epoch": 13.96112600536193,
      "grad_norm": 9.8125,
      "learning_rate": 1.2647746648573705e-05,
      "loss": 1.2852,
      "num_input_tokens_seen": 54384664,
      "step": 93735
    },
    {
      "epoch": 13.96187071790289,
      "grad_norm": 7.65625,
      "learning_rate": 1.2644921671187424e-05,
      "loss": 0.8848,
      "num_input_tokens_seen": 54387480,
      "step": 93740
    },
    {
      "epoch": 13.962615430443849,
      "grad_norm": 5.40625,
      "learning_rate": 1.2642096902527633e-05,
      "loss": 0.8848,
      "num_input_tokens_seen": 54390328,
      "step": 93745
    },
    {
      "epoch": 13.963360142984808,
      "grad_norm": 11.375,
      "learning_rate": 1.2639272342642047e-05,
      "loss": 1.1938,
      "num_input_tokens_seen": 54393304,
      "step": 93750
    },
    {
      "epoch": 13.964104855525767,
      "grad_norm": 13.1875,
      "learning_rate": 1.2636447991578401e-05,
      "loss": 1.3985,
      "num_input_tokens_seen": 54396312,
      "step": 93755
    },
    {
      "epoch": 13.964849568066727,
      "grad_norm": 15.25,
      "learning_rate": 1.263362384938439e-05,
      "loss": 1.1415,
      "num_input_tokens_seen": 54399288,
      "step": 93760
    },
    {
      "epoch": 13.965594280607686,
      "grad_norm": 17.375,
      "learning_rate": 1.2630799916107747e-05,
      "loss": 1.2276,
      "num_input_tokens_seen": 54402296,
      "step": 93765
    },
    {
      "epoch": 13.966338993148645,
      "grad_norm": 8.375,
      "learning_rate": 1.2627976191796165e-05,
      "loss": 1.0971,
      "num_input_tokens_seen": 54405016,
      "step": 93770
    },
    {
      "epoch": 13.967083705689603,
      "grad_norm": 14.0,
      "learning_rate": 1.2625152676497354e-05,
      "loss": 1.0524,
      "num_input_tokens_seen": 54407640,
      "step": 93775
    },
    {
      "epoch": 13.967828418230564,
      "grad_norm": 15.75,
      "learning_rate": 1.2622329370259001e-05,
      "loss": 1.1849,
      "num_input_tokens_seen": 54410456,
      "step": 93780
    },
    {
      "epoch": 13.968573130771523,
      "grad_norm": 6.09375,
      "learning_rate": 1.261950627312882e-05,
      "loss": 1.1219,
      "num_input_tokens_seen": 54413240,
      "step": 93785
    },
    {
      "epoch": 13.969317843312481,
      "grad_norm": 7.6875,
      "learning_rate": 1.2616683385154498e-05,
      "loss": 1.171,
      "num_input_tokens_seen": 54415992,
      "step": 93790
    },
    {
      "epoch": 13.97006255585344,
      "grad_norm": 12.1875,
      "learning_rate": 1.2613860706383718e-05,
      "loss": 0.9177,
      "num_input_tokens_seen": 54418744,
      "step": 93795
    },
    {
      "epoch": 13.970807268394399,
      "grad_norm": 14.0,
      "learning_rate": 1.261103823686418e-05,
      "loss": 0.9964,
      "num_input_tokens_seen": 54421496,
      "step": 93800
    },
    {
      "epoch": 13.97155198093536,
      "grad_norm": 12.75,
      "learning_rate": 1.260821597664355e-05,
      "loss": 1.2406,
      "num_input_tokens_seen": 54424376,
      "step": 93805
    },
    {
      "epoch": 13.972296693476318,
      "grad_norm": 7.40625,
      "learning_rate": 1.2605393925769526e-05,
      "loss": 0.8213,
      "num_input_tokens_seen": 54427320,
      "step": 93810
    },
    {
      "epoch": 13.973041406017277,
      "grad_norm": 14.6875,
      "learning_rate": 1.2602572084289765e-05,
      "loss": 1.2282,
      "num_input_tokens_seen": 54429880,
      "step": 93815
    },
    {
      "epoch": 13.973786118558236,
      "grad_norm": 7.9375,
      "learning_rate": 1.259975045225196e-05,
      "loss": 1.139,
      "num_input_tokens_seen": 54432472,
      "step": 93820
    },
    {
      "epoch": 13.974530831099196,
      "grad_norm": 7.75,
      "learning_rate": 1.2596929029703766e-05,
      "loss": 0.8706,
      "num_input_tokens_seen": 54435128,
      "step": 93825
    },
    {
      "epoch": 13.975275543640155,
      "grad_norm": 11.8125,
      "learning_rate": 1.2594107816692852e-05,
      "loss": 1.1929,
      "num_input_tokens_seen": 54438104,
      "step": 93830
    },
    {
      "epoch": 13.976020256181114,
      "grad_norm": 16.25,
      "learning_rate": 1.2591286813266867e-05,
      "loss": 0.975,
      "num_input_tokens_seen": 54440984,
      "step": 93835
    },
    {
      "epoch": 13.976764968722073,
      "grad_norm": 7.15625,
      "learning_rate": 1.2588466019473488e-05,
      "loss": 1.3659,
      "num_input_tokens_seen": 54443864,
      "step": 93840
    },
    {
      "epoch": 13.977509681263033,
      "grad_norm": 13.25,
      "learning_rate": 1.2585645435360361e-05,
      "loss": 1.2286,
      "num_input_tokens_seen": 54446680,
      "step": 93845
    },
    {
      "epoch": 13.978254393803992,
      "grad_norm": 8.8125,
      "learning_rate": 1.2582825060975128e-05,
      "loss": 1.0623,
      "num_input_tokens_seen": 54449592,
      "step": 93850
    },
    {
      "epoch": 13.97899910634495,
      "grad_norm": 9.5625,
      "learning_rate": 1.2580004896365455e-05,
      "loss": 0.9994,
      "num_input_tokens_seen": 54452600,
      "step": 93855
    },
    {
      "epoch": 13.97974381888591,
      "grad_norm": 15.8125,
      "learning_rate": 1.2577184941578968e-05,
      "loss": 1.1109,
      "num_input_tokens_seen": 54455384,
      "step": 93860
    },
    {
      "epoch": 13.98048853142687,
      "grad_norm": 12.625,
      "learning_rate": 1.2574365196663324e-05,
      "loss": 1.3104,
      "num_input_tokens_seen": 54458712,
      "step": 93865
    },
    {
      "epoch": 13.981233243967829,
      "grad_norm": 15.125,
      "learning_rate": 1.2571545661666151e-05,
      "loss": 1.1109,
      "num_input_tokens_seen": 54461624,
      "step": 93870
    },
    {
      "epoch": 13.981977956508787,
      "grad_norm": 14.6875,
      "learning_rate": 1.2568726336635073e-05,
      "loss": 0.8849,
      "num_input_tokens_seen": 54464792,
      "step": 93875
    },
    {
      "epoch": 13.982722669049746,
      "grad_norm": 9.6875,
      "learning_rate": 1.2565907221617738e-05,
      "loss": 0.8874,
      "num_input_tokens_seen": 54467960,
      "step": 93880
    },
    {
      "epoch": 13.983467381590707,
      "grad_norm": 15.75,
      "learning_rate": 1.2563088316661753e-05,
      "loss": 1.017,
      "num_input_tokens_seen": 54470680,
      "step": 93885
    },
    {
      "epoch": 13.984212094131665,
      "grad_norm": 10.375,
      "learning_rate": 1.256026962181476e-05,
      "loss": 1.099,
      "num_input_tokens_seen": 54473528,
      "step": 93890
    },
    {
      "epoch": 13.984956806672624,
      "grad_norm": 13.5625,
      "learning_rate": 1.255745113712437e-05,
      "loss": 1.4047,
      "num_input_tokens_seen": 54476184,
      "step": 93895
    },
    {
      "epoch": 13.985701519213583,
      "grad_norm": 10.75,
      "learning_rate": 1.2554632862638197e-05,
      "loss": 1.1771,
      "num_input_tokens_seen": 54479128,
      "step": 93900
    },
    {
      "epoch": 13.986446231754543,
      "grad_norm": 8.5625,
      "learning_rate": 1.2551814798403851e-05,
      "loss": 1.2053,
      "num_input_tokens_seen": 54481816,
      "step": 93905
    },
    {
      "epoch": 13.987190944295502,
      "grad_norm": 15.25,
      "learning_rate": 1.2548996944468935e-05,
      "loss": 1.1195,
      "num_input_tokens_seen": 54484920,
      "step": 93910
    },
    {
      "epoch": 13.987935656836461,
      "grad_norm": 14.1875,
      "learning_rate": 1.254617930088107e-05,
      "loss": 1.0665,
      "num_input_tokens_seen": 54487512,
      "step": 93915
    },
    {
      "epoch": 13.98868036937742,
      "grad_norm": 6.46875,
      "learning_rate": 1.2543361867687836e-05,
      "loss": 1.1371,
      "num_input_tokens_seen": 54489976,
      "step": 93920
    },
    {
      "epoch": 13.98942508191838,
      "grad_norm": 6.09375,
      "learning_rate": 1.2540544644936858e-05,
      "loss": 0.9649,
      "num_input_tokens_seen": 54492632,
      "step": 93925
    },
    {
      "epoch": 13.990169794459339,
      "grad_norm": 6.46875,
      "learning_rate": 1.2537727632675699e-05,
      "loss": 1.1555,
      "num_input_tokens_seen": 54495416,
      "step": 93930
    },
    {
      "epoch": 13.990914507000298,
      "grad_norm": 10.0625,
      "learning_rate": 1.253491083095198e-05,
      "loss": 0.549,
      "num_input_tokens_seen": 54498136,
      "step": 93935
    },
    {
      "epoch": 13.991659219541257,
      "grad_norm": 5.0,
      "learning_rate": 1.253209423981326e-05,
      "loss": 1.0132,
      "num_input_tokens_seen": 54501208,
      "step": 93940
    },
    {
      "epoch": 13.992403932082215,
      "grad_norm": 9.625,
      "learning_rate": 1.2529277859307148e-05,
      "loss": 1.1669,
      "num_input_tokens_seen": 54504120,
      "step": 93945
    },
    {
      "epoch": 13.993148644623176,
      "grad_norm": 5.28125,
      "learning_rate": 1.2526461689481212e-05,
      "loss": 1.2744,
      "num_input_tokens_seen": 54507128,
      "step": 93950
    },
    {
      "epoch": 13.993893357164135,
      "grad_norm": 10.3125,
      "learning_rate": 1.2523645730383018e-05,
      "loss": 1.076,
      "num_input_tokens_seen": 54510008,
      "step": 93955
    },
    {
      "epoch": 13.994638069705093,
      "grad_norm": 10.0,
      "learning_rate": 1.2520829982060162e-05,
      "loss": 1.0107,
      "num_input_tokens_seen": 54513080,
      "step": 93960
    },
    {
      "epoch": 13.995382782246054,
      "grad_norm": 6.4375,
      "learning_rate": 1.2518014444560195e-05,
      "loss": 1.2168,
      "num_input_tokens_seen": 54515768,
      "step": 93965
    },
    {
      "epoch": 13.996127494787013,
      "grad_norm": 18.875,
      "learning_rate": 1.251519911793069e-05,
      "loss": 1.2347,
      "num_input_tokens_seen": 54518552,
      "step": 93970
    },
    {
      "epoch": 13.996872207327971,
      "grad_norm": 9.9375,
      "learning_rate": 1.2512384002219196e-05,
      "loss": 1.0494,
      "num_input_tokens_seen": 54521368,
      "step": 93975
    },
    {
      "epoch": 13.99761691986893,
      "grad_norm": 17.125,
      "learning_rate": 1.2509569097473295e-05,
      "loss": 1.209,
      "num_input_tokens_seen": 54524472,
      "step": 93980
    },
    {
      "epoch": 13.998361632409889,
      "grad_norm": 19.125,
      "learning_rate": 1.2506754403740529e-05,
      "loss": 1.084,
      "num_input_tokens_seen": 54527512,
      "step": 93985
    },
    {
      "epoch": 13.99910634495085,
      "grad_norm": 5.4375,
      "learning_rate": 1.2503939921068435e-05,
      "loss": 1.1375,
      "num_input_tokens_seen": 54530296,
      "step": 93990
    },
    {
      "epoch": 13.999851057491808,
      "grad_norm": 10.5,
      "learning_rate": 1.2501125649504591e-05,
      "loss": 1.1854,
      "num_input_tokens_seen": 54533016,
      "step": 93995
    },
    {
      "epoch": 14.0,
      "eval_loss": 1.0829328298568726,
      "eval_runtime": 43.7872,
      "eval_samples_per_second": 68.148,
      "eval_steps_per_second": 17.037,
      "num_input_tokens_seen": 54533112,
      "step": 93996
    },
    {
      "epoch": 14.000595770032767,
      "grad_norm": 11.0625,
      "learning_rate": 1.2498311589096514e-05,
      "loss": 1.0409,
      "num_input_tokens_seen": 54535352,
      "step": 94000
    },
    {
      "epoch": 14.001340482573726,
      "grad_norm": 15.0625,
      "learning_rate": 1.2495497739891764e-05,
      "loss": 1.1682,
      "num_input_tokens_seen": 54538328,
      "step": 94005
    },
    {
      "epoch": 14.002085195114686,
      "grad_norm": 10.375,
      "learning_rate": 1.2492684101937865e-05,
      "loss": 1.2577,
      "num_input_tokens_seen": 54541272,
      "step": 94010
    },
    {
      "epoch": 14.002829907655645,
      "grad_norm": 9.5,
      "learning_rate": 1.2489870675282364e-05,
      "loss": 0.9889,
      "num_input_tokens_seen": 54543960,
      "step": 94015
    },
    {
      "epoch": 14.003574620196604,
      "grad_norm": 7.25,
      "learning_rate": 1.2487057459972775e-05,
      "loss": 0.923,
      "num_input_tokens_seen": 54546840,
      "step": 94020
    },
    {
      "epoch": 14.004319332737563,
      "grad_norm": 10.8125,
      "learning_rate": 1.248424445605664e-05,
      "loss": 1.1154,
      "num_input_tokens_seen": 54549592,
      "step": 94025
    },
    {
      "epoch": 14.005064045278523,
      "grad_norm": 8.375,
      "learning_rate": 1.2481431663581474e-05,
      "loss": 1.1762,
      "num_input_tokens_seen": 54552568,
      "step": 94030
    },
    {
      "epoch": 14.005808757819482,
      "grad_norm": 11.1875,
      "learning_rate": 1.24786190825948e-05,
      "loss": 1.1942,
      "num_input_tokens_seen": 54555608,
      "step": 94035
    },
    {
      "epoch": 14.00655347036044,
      "grad_norm": 7.09375,
      "learning_rate": 1.2475806713144128e-05,
      "loss": 0.9693,
      "num_input_tokens_seen": 54558904,
      "step": 94040
    },
    {
      "epoch": 14.0072981829014,
      "grad_norm": 11.625,
      "learning_rate": 1.2472994555276964e-05,
      "loss": 1.0482,
      "num_input_tokens_seen": 54561912,
      "step": 94045
    },
    {
      "epoch": 14.00804289544236,
      "grad_norm": 18.125,
      "learning_rate": 1.2470182609040833e-05,
      "loss": 1.1567,
      "num_input_tokens_seen": 54564920,
      "step": 94050
    },
    {
      "epoch": 14.008787607983319,
      "grad_norm": 12.75,
      "learning_rate": 1.2467370874483225e-05,
      "loss": 1.026,
      "num_input_tokens_seen": 54567896,
      "step": 94055
    },
    {
      "epoch": 14.009532320524277,
      "grad_norm": 7.375,
      "learning_rate": 1.2464559351651658e-05,
      "loss": 0.9075,
      "num_input_tokens_seen": 54570360,
      "step": 94060
    },
    {
      "epoch": 14.010277033065236,
      "grad_norm": 7.90625,
      "learning_rate": 1.2461748040593621e-05,
      "loss": 0.8404,
      "num_input_tokens_seen": 54573144,
      "step": 94065
    },
    {
      "epoch": 14.011021745606197,
      "grad_norm": 9.0,
      "learning_rate": 1.2458936941356594e-05,
      "loss": 1.3358,
      "num_input_tokens_seen": 54576024,
      "step": 94070
    },
    {
      "epoch": 14.011766458147155,
      "grad_norm": 12.1875,
      "learning_rate": 1.2456126053988093e-05,
      "loss": 1.188,
      "num_input_tokens_seen": 54578712,
      "step": 94075
    },
    {
      "epoch": 14.012511170688114,
      "grad_norm": 9.25,
      "learning_rate": 1.2453315378535584e-05,
      "loss": 1.1099,
      "num_input_tokens_seen": 54581784,
      "step": 94080
    },
    {
      "epoch": 14.013255883229073,
      "grad_norm": 6.46875,
      "learning_rate": 1.245050491504657e-05,
      "loss": 1.0923,
      "num_input_tokens_seen": 54584760,
      "step": 94085
    },
    {
      "epoch": 14.014000595770034,
      "grad_norm": 6.96875,
      "learning_rate": 1.2447694663568509e-05,
      "loss": 0.9228,
      "num_input_tokens_seen": 54587832,
      "step": 94090
    },
    {
      "epoch": 14.014745308310992,
      "grad_norm": 13.5625,
      "learning_rate": 1.24448846241489e-05,
      "loss": 1.1369,
      "num_input_tokens_seen": 54590520,
      "step": 94095
    },
    {
      "epoch": 14.015490020851951,
      "grad_norm": 17.375,
      "learning_rate": 1.2442074796835206e-05,
      "loss": 1.2623,
      "num_input_tokens_seen": 54593304,
      "step": 94100
    },
    {
      "epoch": 14.01623473339291,
      "grad_norm": 23.0,
      "learning_rate": 1.2439265181674895e-05,
      "loss": 1.2191,
      "num_input_tokens_seen": 54595960,
      "step": 94105
    },
    {
      "epoch": 14.01697944593387,
      "grad_norm": 8.625,
      "learning_rate": 1.2436455778715431e-05,
      "loss": 1.1153,
      "num_input_tokens_seen": 54599064,
      "step": 94110
    },
    {
      "epoch": 14.017724158474829,
      "grad_norm": 13.8125,
      "learning_rate": 1.2433646588004266e-05,
      "loss": 0.7881,
      "num_input_tokens_seen": 54601912,
      "step": 94115
    },
    {
      "epoch": 14.018468871015788,
      "grad_norm": 6.21875,
      "learning_rate": 1.2430837609588883e-05,
      "loss": 0.9681,
      "num_input_tokens_seen": 54604856,
      "step": 94120
    },
    {
      "epoch": 14.019213583556747,
      "grad_norm": 8.9375,
      "learning_rate": 1.2428028843516715e-05,
      "loss": 1.1721,
      "num_input_tokens_seen": 54607832,
      "step": 94125
    },
    {
      "epoch": 14.019958296097707,
      "grad_norm": 9.9375,
      "learning_rate": 1.2425220289835229e-05,
      "loss": 0.8572,
      "num_input_tokens_seen": 54610584,
      "step": 94130
    },
    {
      "epoch": 14.020703008638666,
      "grad_norm": 7.21875,
      "learning_rate": 1.2422411948591855e-05,
      "loss": 1.0454,
      "num_input_tokens_seen": 54613560,
      "step": 94135
    },
    {
      "epoch": 14.021447721179625,
      "grad_norm": 8.5,
      "learning_rate": 1.241960381983406e-05,
      "loss": 1.0016,
      "num_input_tokens_seen": 54616952,
      "step": 94140
    },
    {
      "epoch": 14.022192433720583,
      "grad_norm": 5.5,
      "learning_rate": 1.2416795903609274e-05,
      "loss": 0.8601,
      "num_input_tokens_seen": 54620024,
      "step": 94145
    },
    {
      "epoch": 14.022937146261542,
      "grad_norm": 13.375,
      "learning_rate": 1.2413988199964918e-05,
      "loss": 1.0021,
      "num_input_tokens_seen": 54622936,
      "step": 94150
    },
    {
      "epoch": 14.023681858802503,
      "grad_norm": 9.6875,
      "learning_rate": 1.2411180708948453e-05,
      "loss": 0.9623,
      "num_input_tokens_seen": 54626040,
      "step": 94155
    },
    {
      "epoch": 14.024426571343461,
      "grad_norm": 8.25,
      "learning_rate": 1.2408373430607296e-05,
      "loss": 1.4716,
      "num_input_tokens_seen": 54629016,
      "step": 94160
    },
    {
      "epoch": 14.02517128388442,
      "grad_norm": 8.5,
      "learning_rate": 1.2405566364988857e-05,
      "loss": 1.2165,
      "num_input_tokens_seen": 54631800,
      "step": 94165
    },
    {
      "epoch": 14.025915996425379,
      "grad_norm": 11.125,
      "learning_rate": 1.2402759512140588e-05,
      "loss": 1.0799,
      "num_input_tokens_seen": 54634520,
      "step": 94170
    },
    {
      "epoch": 14.02666070896634,
      "grad_norm": 17.25,
      "learning_rate": 1.2399952872109893e-05,
      "loss": 1.139,
      "num_input_tokens_seen": 54637400,
      "step": 94175
    },
    {
      "epoch": 14.027405421507298,
      "grad_norm": 6.625,
      "learning_rate": 1.239714644494418e-05,
      "loss": 1.1601,
      "num_input_tokens_seen": 54640184,
      "step": 94180
    },
    {
      "epoch": 14.028150134048257,
      "grad_norm": 13.5,
      "learning_rate": 1.2394340230690877e-05,
      "loss": 1.1266,
      "num_input_tokens_seen": 54643032,
      "step": 94185
    },
    {
      "epoch": 14.028894846589216,
      "grad_norm": 16.375,
      "learning_rate": 1.2391534229397384e-05,
      "loss": 1.3209,
      "num_input_tokens_seen": 54645848,
      "step": 94190
    },
    {
      "epoch": 14.029639559130176,
      "grad_norm": 5.40625,
      "learning_rate": 1.2388728441111095e-05,
      "loss": 1.0019,
      "num_input_tokens_seen": 54648920,
      "step": 94195
    },
    {
      "epoch": 14.030384271671135,
      "grad_norm": 5.3125,
      "learning_rate": 1.2385922865879432e-05,
      "loss": 0.8414,
      "num_input_tokens_seen": 54651768,
      "step": 94200
    },
    {
      "epoch": 14.031128984212094,
      "grad_norm": 14.6875,
      "learning_rate": 1.2383117503749769e-05,
      "loss": 1.3551,
      "num_input_tokens_seen": 54654808,
      "step": 94205
    },
    {
      "epoch": 14.031873696753053,
      "grad_norm": 8.5,
      "learning_rate": 1.2380312354769526e-05,
      "loss": 1.0752,
      "num_input_tokens_seen": 54657848,
      "step": 94210
    },
    {
      "epoch": 14.032618409294013,
      "grad_norm": 10.0,
      "learning_rate": 1.2377507418986071e-05,
      "loss": 1.0186,
      "num_input_tokens_seen": 54660760,
      "step": 94215
    },
    {
      "epoch": 14.033363121834972,
      "grad_norm": 12.4375,
      "learning_rate": 1.2374702696446806e-05,
      "loss": 1.0838,
      "num_input_tokens_seen": 54663544,
      "step": 94220
    },
    {
      "epoch": 14.03410783437593,
      "grad_norm": 11.5625,
      "learning_rate": 1.2371898187199108e-05,
      "loss": 0.9203,
      "num_input_tokens_seen": 54666296,
      "step": 94225
    },
    {
      "epoch": 14.03485254691689,
      "grad_norm": 6.53125,
      "learning_rate": 1.2369093891290357e-05,
      "loss": 1.1066,
      "num_input_tokens_seen": 54669240,
      "step": 94230
    },
    {
      "epoch": 14.03559725945785,
      "grad_norm": 11.6875,
      "learning_rate": 1.2366289808767926e-05,
      "loss": 1.162,
      "num_input_tokens_seen": 54672088,
      "step": 94235
    },
    {
      "epoch": 14.036341971998809,
      "grad_norm": 11.5625,
      "learning_rate": 1.2363485939679175e-05,
      "loss": 0.9195,
      "num_input_tokens_seen": 54674872,
      "step": 94240
    },
    {
      "epoch": 14.037086684539767,
      "grad_norm": 9.6875,
      "learning_rate": 1.23606822840715e-05,
      "loss": 1.0743,
      "num_input_tokens_seen": 54677624,
      "step": 94245
    },
    {
      "epoch": 14.037831397080726,
      "grad_norm": 11.9375,
      "learning_rate": 1.2357878841992243e-05,
      "loss": 1.1844,
      "num_input_tokens_seen": 54680504,
      "step": 94250
    },
    {
      "epoch": 14.038576109621687,
      "grad_norm": 5.15625,
      "learning_rate": 1.2355075613488782e-05,
      "loss": 1.0316,
      "num_input_tokens_seen": 54683896,
      "step": 94255
    },
    {
      "epoch": 14.039320822162646,
      "grad_norm": 16.875,
      "learning_rate": 1.2352272598608455e-05,
      "loss": 1.1511,
      "num_input_tokens_seen": 54687064,
      "step": 94260
    },
    {
      "epoch": 14.040065534703604,
      "grad_norm": 7.78125,
      "learning_rate": 1.234946979739864e-05,
      "loss": 0.814,
      "num_input_tokens_seen": 54690072,
      "step": 94265
    },
    {
      "epoch": 14.040810247244563,
      "grad_norm": 9.9375,
      "learning_rate": 1.2346667209906677e-05,
      "loss": 1.0223,
      "num_input_tokens_seen": 54693016,
      "step": 94270
    },
    {
      "epoch": 14.041554959785524,
      "grad_norm": 12.75,
      "learning_rate": 1.23438648361799e-05,
      "loss": 0.9907,
      "num_input_tokens_seen": 54695960,
      "step": 94275
    },
    {
      "epoch": 14.042299672326482,
      "grad_norm": 13.875,
      "learning_rate": 1.2341062676265671e-05,
      "loss": 1.1374,
      "num_input_tokens_seen": 54698904,
      "step": 94280
    },
    {
      "epoch": 14.043044384867441,
      "grad_norm": 10.125,
      "learning_rate": 1.2338260730211316e-05,
      "loss": 1.1234,
      "num_input_tokens_seen": 54701944,
      "step": 94285
    },
    {
      "epoch": 14.0437890974084,
      "grad_norm": 11.875,
      "learning_rate": 1.2335458998064184e-05,
      "loss": 1.4048,
      "num_input_tokens_seen": 54704760,
      "step": 94290
    },
    {
      "epoch": 14.04453380994936,
      "grad_norm": 8.6875,
      "learning_rate": 1.23326574798716e-05,
      "loss": 1.1504,
      "num_input_tokens_seen": 54707576,
      "step": 94295
    },
    {
      "epoch": 14.04527852249032,
      "grad_norm": 7.65625,
      "learning_rate": 1.2329856175680896e-05,
      "loss": 1.0885,
      "num_input_tokens_seen": 54710584,
      "step": 94300
    },
    {
      "epoch": 14.046023235031278,
      "grad_norm": 7.9375,
      "learning_rate": 1.2327055085539382e-05,
      "loss": 1.0122,
      "num_input_tokens_seen": 54713240,
      "step": 94305
    },
    {
      "epoch": 14.046767947572237,
      "grad_norm": 10.75,
      "learning_rate": 1.2324254209494405e-05,
      "loss": 1.0107,
      "num_input_tokens_seen": 54715960,
      "step": 94310
    },
    {
      "epoch": 14.047512660113195,
      "grad_norm": 5.3125,
      "learning_rate": 1.2321453547593267e-05,
      "loss": 0.8645,
      "num_input_tokens_seen": 54719032,
      "step": 94315
    },
    {
      "epoch": 14.048257372654156,
      "grad_norm": 13.1875,
      "learning_rate": 1.2318653099883278e-05,
      "loss": 1.2116,
      "num_input_tokens_seen": 54722168,
      "step": 94320
    },
    {
      "epoch": 14.049002085195115,
      "grad_norm": 14.625,
      "learning_rate": 1.2315852866411767e-05,
      "loss": 1.4938,
      "num_input_tokens_seen": 54724984,
      "step": 94325
    },
    {
      "epoch": 14.049746797736073,
      "grad_norm": 13.125,
      "learning_rate": 1.2313052847226018e-05,
      "loss": 0.9648,
      "num_input_tokens_seen": 54727832,
      "step": 94330
    },
    {
      "epoch": 14.050491510277032,
      "grad_norm": 12.0625,
      "learning_rate": 1.2310253042373356e-05,
      "loss": 0.8962,
      "num_input_tokens_seen": 54730744,
      "step": 94335
    },
    {
      "epoch": 14.051236222817993,
      "grad_norm": 10.9375,
      "learning_rate": 1.2307453451901063e-05,
      "loss": 1.1403,
      "num_input_tokens_seen": 54733496,
      "step": 94340
    },
    {
      "epoch": 14.051980935358952,
      "grad_norm": 14.1875,
      "learning_rate": 1.2304654075856452e-05,
      "loss": 1.1335,
      "num_input_tokens_seen": 54736344,
      "step": 94345
    },
    {
      "epoch": 14.05272564789991,
      "grad_norm": 13.4375,
      "learning_rate": 1.2301854914286812e-05,
      "loss": 1.2883,
      "num_input_tokens_seen": 54739480,
      "step": 94350
    },
    {
      "epoch": 14.053470360440869,
      "grad_norm": 5.15625,
      "learning_rate": 1.2299055967239415e-05,
      "loss": 1.115,
      "num_input_tokens_seen": 54742264,
      "step": 94355
    },
    {
      "epoch": 14.05421507298183,
      "grad_norm": 11.6875,
      "learning_rate": 1.2296257234761566e-05,
      "loss": 1.1247,
      "num_input_tokens_seen": 54745080,
      "step": 94360
    },
    {
      "epoch": 14.054959785522788,
      "grad_norm": 8.0625,
      "learning_rate": 1.2293458716900543e-05,
      "loss": 0.6933,
      "num_input_tokens_seen": 54748088,
      "step": 94365
    },
    {
      "epoch": 14.055704498063747,
      "grad_norm": 8.1875,
      "learning_rate": 1.229066041370362e-05,
      "loss": 1.2152,
      "num_input_tokens_seen": 54751064,
      "step": 94370
    },
    {
      "epoch": 14.056449210604706,
      "grad_norm": 13.75,
      "learning_rate": 1.228786232521806e-05,
      "loss": 1.1022,
      "num_input_tokens_seen": 54754232,
      "step": 94375
    },
    {
      "epoch": 14.057193923145666,
      "grad_norm": 11.1875,
      "learning_rate": 1.2285064451491157e-05,
      "loss": 1.5279,
      "num_input_tokens_seen": 54757240,
      "step": 94380
    },
    {
      "epoch": 14.057938635686625,
      "grad_norm": 17.5,
      "learning_rate": 1.2282266792570158e-05,
      "loss": 1.0499,
      "num_input_tokens_seen": 54760120,
      "step": 94385
    },
    {
      "epoch": 14.058683348227584,
      "grad_norm": 12.1875,
      "learning_rate": 1.2279469348502345e-05,
      "loss": 1.2761,
      "num_input_tokens_seen": 54763032,
      "step": 94390
    },
    {
      "epoch": 14.059428060768543,
      "grad_norm": 10.875,
      "learning_rate": 1.227667211933497e-05,
      "loss": 0.9057,
      "num_input_tokens_seen": 54765944,
      "step": 94395
    },
    {
      "epoch": 14.060172773309503,
      "grad_norm": 20.0,
      "learning_rate": 1.2273875105115275e-05,
      "loss": 1.1022,
      "num_input_tokens_seen": 54768760,
      "step": 94400
    },
    {
      "epoch": 14.060917485850462,
      "grad_norm": 7.59375,
      "learning_rate": 1.227107830589054e-05,
      "loss": 1.0611,
      "num_input_tokens_seen": 54771608,
      "step": 94405
    },
    {
      "epoch": 14.06166219839142,
      "grad_norm": 14.0,
      "learning_rate": 1.2268281721707989e-05,
      "loss": 0.955,
      "num_input_tokens_seen": 54774392,
      "step": 94410
    },
    {
      "epoch": 14.06240691093238,
      "grad_norm": 16.875,
      "learning_rate": 1.2265485352614887e-05,
      "loss": 1.0239,
      "num_input_tokens_seen": 54777208,
      "step": 94415
    },
    {
      "epoch": 14.06315162347334,
      "grad_norm": 15.6875,
      "learning_rate": 1.226268919865846e-05,
      "loss": 1.1568,
      "num_input_tokens_seen": 54779992,
      "step": 94420
    },
    {
      "epoch": 14.063896336014299,
      "grad_norm": 6.53125,
      "learning_rate": 1.225989325988596e-05,
      "loss": 1.0057,
      "num_input_tokens_seen": 54782872,
      "step": 94425
    },
    {
      "epoch": 14.064641048555258,
      "grad_norm": 9.875,
      "learning_rate": 1.2257097536344613e-05,
      "loss": 0.7853,
      "num_input_tokens_seen": 54785656,
      "step": 94430
    },
    {
      "epoch": 14.065385761096216,
      "grad_norm": 15.125,
      "learning_rate": 1.2254302028081657e-05,
      "loss": 1.0545,
      "num_input_tokens_seen": 54788472,
      "step": 94435
    },
    {
      "epoch": 14.066130473637177,
      "grad_norm": 10.125,
      "learning_rate": 1.225150673514431e-05,
      "loss": 0.9454,
      "num_input_tokens_seen": 54791480,
      "step": 94440
    },
    {
      "epoch": 14.066875186178136,
      "grad_norm": 7.0,
      "learning_rate": 1.2248711657579792e-05,
      "loss": 0.8322,
      "num_input_tokens_seen": 54794392,
      "step": 94445
    },
    {
      "epoch": 14.067619898719094,
      "grad_norm": 7.5,
      "learning_rate": 1.2245916795435342e-05,
      "loss": 0.9767,
      "num_input_tokens_seen": 54797304,
      "step": 94450
    },
    {
      "epoch": 14.068364611260053,
      "grad_norm": 7.25,
      "learning_rate": 1.2243122148758152e-05,
      "loss": 0.8784,
      "num_input_tokens_seen": 54800152,
      "step": 94455
    },
    {
      "epoch": 14.069109323801014,
      "grad_norm": 5.25,
      "learning_rate": 1.224032771759546e-05,
      "loss": 0.8698,
      "num_input_tokens_seen": 54802904,
      "step": 94460
    },
    {
      "epoch": 14.069854036341972,
      "grad_norm": 13.4375,
      "learning_rate": 1.2237533501994452e-05,
      "loss": 1.3086,
      "num_input_tokens_seen": 54805784,
      "step": 94465
    },
    {
      "epoch": 14.070598748882931,
      "grad_norm": 6.03125,
      "learning_rate": 1.2234739502002353e-05,
      "loss": 0.7812,
      "num_input_tokens_seen": 54808920,
      "step": 94470
    },
    {
      "epoch": 14.07134346142389,
      "grad_norm": 7.3125,
      "learning_rate": 1.2231945717666358e-05,
      "loss": 1.1092,
      "num_input_tokens_seen": 54812152,
      "step": 94475
    },
    {
      "epoch": 14.07208817396485,
      "grad_norm": 14.0,
      "learning_rate": 1.2229152149033655e-05,
      "loss": 1.0889,
      "num_input_tokens_seen": 54814968,
      "step": 94480
    },
    {
      "epoch": 14.07283288650581,
      "grad_norm": 13.4375,
      "learning_rate": 1.2226358796151452e-05,
      "loss": 1.5692,
      "num_input_tokens_seen": 54817784,
      "step": 94485
    },
    {
      "epoch": 14.073577599046768,
      "grad_norm": 12.25,
      "learning_rate": 1.2223565659066938e-05,
      "loss": 0.8986,
      "num_input_tokens_seen": 54820664,
      "step": 94490
    },
    {
      "epoch": 14.074322311587727,
      "grad_norm": 10.1875,
      "learning_rate": 1.2220772737827285e-05,
      "loss": 1.5515,
      "num_input_tokens_seen": 54823448,
      "step": 94495
    },
    {
      "epoch": 14.075067024128685,
      "grad_norm": 8.6875,
      "learning_rate": 1.2217980032479701e-05,
      "loss": 1.3643,
      "num_input_tokens_seen": 54826104,
      "step": 94500
    },
    {
      "epoch": 14.075811736669646,
      "grad_norm": 3.3125,
      "learning_rate": 1.221518754307135e-05,
      "loss": 1.1053,
      "num_input_tokens_seen": 54829016,
      "step": 94505
    },
    {
      "epoch": 14.076556449210605,
      "grad_norm": 9.4375,
      "learning_rate": 1.2212395269649413e-05,
      "loss": 1.3983,
      "num_input_tokens_seen": 54831832,
      "step": 94510
    },
    {
      "epoch": 14.077301161751564,
      "grad_norm": 8.1875,
      "learning_rate": 1.220960321226105e-05,
      "loss": 1.1334,
      "num_input_tokens_seen": 54834968,
      "step": 94515
    },
    {
      "epoch": 14.078045874292522,
      "grad_norm": 7.125,
      "learning_rate": 1.2206811370953453e-05,
      "loss": 1.0691,
      "num_input_tokens_seen": 54838520,
      "step": 94520
    },
    {
      "epoch": 14.078790586833483,
      "grad_norm": 6.0625,
      "learning_rate": 1.2204019745773764e-05,
      "loss": 1.017,
      "num_input_tokens_seen": 54841368,
      "step": 94525
    },
    {
      "epoch": 14.079535299374442,
      "grad_norm": 12.625,
      "learning_rate": 1.2201228336769169e-05,
      "loss": 1.2181,
      "num_input_tokens_seen": 54844120,
      "step": 94530
    },
    {
      "epoch": 14.0802800119154,
      "grad_norm": 13.4375,
      "learning_rate": 1.2198437143986798e-05,
      "loss": 1.2554,
      "num_input_tokens_seen": 54847000,
      "step": 94535
    },
    {
      "epoch": 14.081024724456359,
      "grad_norm": 13.1875,
      "learning_rate": 1.2195646167473835e-05,
      "loss": 0.7618,
      "num_input_tokens_seen": 54849592,
      "step": 94540
    },
    {
      "epoch": 14.08176943699732,
      "grad_norm": 11.0,
      "learning_rate": 1.2192855407277407e-05,
      "loss": 1.346,
      "num_input_tokens_seen": 54852248,
      "step": 94545
    },
    {
      "epoch": 14.082514149538278,
      "grad_norm": 12.3125,
      "learning_rate": 1.2190064863444675e-05,
      "loss": 1.057,
      "num_input_tokens_seen": 54854936,
      "step": 94550
    },
    {
      "epoch": 14.083258862079237,
      "grad_norm": 10.875,
      "learning_rate": 1.2187274536022783e-05,
      "loss": 1.1521,
      "num_input_tokens_seen": 54857848,
      "step": 94555
    },
    {
      "epoch": 14.084003574620196,
      "grad_norm": 8.9375,
      "learning_rate": 1.2184484425058863e-05,
      "loss": 0.9,
      "num_input_tokens_seen": 54860920,
      "step": 94560
    },
    {
      "epoch": 14.084748287161156,
      "grad_norm": 9.375,
      "learning_rate": 1.2181694530600052e-05,
      "loss": 1.0109,
      "num_input_tokens_seen": 54863864,
      "step": 94565
    },
    {
      "epoch": 14.085492999702115,
      "grad_norm": 9.0625,
      "learning_rate": 1.2178904852693476e-05,
      "loss": 1.0493,
      "num_input_tokens_seen": 54866712,
      "step": 94570
    },
    {
      "epoch": 14.086237712243074,
      "grad_norm": 3.28125,
      "learning_rate": 1.217611539138628e-05,
      "loss": 0.982,
      "num_input_tokens_seen": 54869560,
      "step": 94575
    },
    {
      "epoch": 14.086982424784033,
      "grad_norm": 13.875,
      "learning_rate": 1.2173326146725575e-05,
      "loss": 0.8586,
      "num_input_tokens_seen": 54872440,
      "step": 94580
    },
    {
      "epoch": 14.087727137324993,
      "grad_norm": 13.4375,
      "learning_rate": 1.2170537118758496e-05,
      "loss": 0.9885,
      "num_input_tokens_seen": 54875352,
      "step": 94585
    },
    {
      "epoch": 14.088471849865952,
      "grad_norm": 9.4375,
      "learning_rate": 1.216774830753215e-05,
      "loss": 1.2343,
      "num_input_tokens_seen": 54878456,
      "step": 94590
    },
    {
      "epoch": 14.08921656240691,
      "grad_norm": 11.3125,
      "learning_rate": 1.2164959713093649e-05,
      "loss": 0.8861,
      "num_input_tokens_seen": 54881464,
      "step": 94595
    },
    {
      "epoch": 14.08996127494787,
      "grad_norm": 13.0,
      "learning_rate": 1.2162171335490115e-05,
      "loss": 1.2369,
      "num_input_tokens_seen": 54884248,
      "step": 94600
    },
    {
      "epoch": 14.09070598748883,
      "grad_norm": 10.375,
      "learning_rate": 1.2159383174768641e-05,
      "loss": 1.2551,
      "num_input_tokens_seen": 54887352,
      "step": 94605
    },
    {
      "epoch": 14.091450700029789,
      "grad_norm": 10.3125,
      "learning_rate": 1.2156595230976348e-05,
      "loss": 1.2233,
      "num_input_tokens_seen": 54890488,
      "step": 94610
    },
    {
      "epoch": 14.092195412570748,
      "grad_norm": 7.84375,
      "learning_rate": 1.2153807504160313e-05,
      "loss": 1.1566,
      "num_input_tokens_seen": 54893208,
      "step": 94615
    },
    {
      "epoch": 14.092940125111706,
      "grad_norm": 6.9375,
      "learning_rate": 1.2151019994367655e-05,
      "loss": 1.0072,
      "num_input_tokens_seen": 54896376,
      "step": 94620
    },
    {
      "epoch": 14.093684837652667,
      "grad_norm": 10.125,
      "learning_rate": 1.2148232701645453e-05,
      "loss": 0.8854,
      "num_input_tokens_seen": 54899416,
      "step": 94625
    },
    {
      "epoch": 14.094429550193626,
      "grad_norm": 14.0,
      "learning_rate": 1.2145445626040801e-05,
      "loss": 1.0086,
      "num_input_tokens_seen": 54902488,
      "step": 94630
    },
    {
      "epoch": 14.095174262734584,
      "grad_norm": 18.5,
      "learning_rate": 1.2142658767600779e-05,
      "loss": 1.0368,
      "num_input_tokens_seen": 54905464,
      "step": 94635
    },
    {
      "epoch": 14.095918975275543,
      "grad_norm": 16.75,
      "learning_rate": 1.213987212637246e-05,
      "loss": 1.3702,
      "num_input_tokens_seen": 54908088,
      "step": 94640
    },
    {
      "epoch": 14.096663687816504,
      "grad_norm": 5.09375,
      "learning_rate": 1.2137085702402939e-05,
      "loss": 0.8119,
      "num_input_tokens_seen": 54911512,
      "step": 94645
    },
    {
      "epoch": 14.097408400357462,
      "grad_norm": 17.25,
      "learning_rate": 1.2134299495739274e-05,
      "loss": 1.1655,
      "num_input_tokens_seen": 54914328,
      "step": 94650
    },
    {
      "epoch": 14.098153112898421,
      "grad_norm": 7.1875,
      "learning_rate": 1.2131513506428552e-05,
      "loss": 1.081,
      "num_input_tokens_seen": 54917112,
      "step": 94655
    },
    {
      "epoch": 14.09889782543938,
      "grad_norm": 12.25,
      "learning_rate": 1.2128727734517819e-05,
      "loss": 1.1785,
      "num_input_tokens_seen": 54919992,
      "step": 94660
    },
    {
      "epoch": 14.099642537980339,
      "grad_norm": 7.53125,
      "learning_rate": 1.2125942180054161e-05,
      "loss": 0.7836,
      "num_input_tokens_seen": 54922840,
      "step": 94665
    },
    {
      "epoch": 14.1003872505213,
      "grad_norm": 5.6875,
      "learning_rate": 1.2123156843084624e-05,
      "loss": 0.9784,
      "num_input_tokens_seen": 54925880,
      "step": 94670
    },
    {
      "epoch": 14.101131963062258,
      "grad_norm": 14.6875,
      "learning_rate": 1.2120371723656257e-05,
      "loss": 1.3205,
      "num_input_tokens_seen": 54928728,
      "step": 94675
    },
    {
      "epoch": 14.101876675603217,
      "grad_norm": 8.625,
      "learning_rate": 1.2117586821816127e-05,
      "loss": 1.2356,
      "num_input_tokens_seen": 54931704,
      "step": 94680
    },
    {
      "epoch": 14.102621388144176,
      "grad_norm": 14.3125,
      "learning_rate": 1.2114802137611266e-05,
      "loss": 1.3276,
      "num_input_tokens_seen": 54934584,
      "step": 94685
    },
    {
      "epoch": 14.103366100685136,
      "grad_norm": 6.46875,
      "learning_rate": 1.2112017671088737e-05,
      "loss": 0.9871,
      "num_input_tokens_seen": 54937368,
      "step": 94690
    },
    {
      "epoch": 14.104110813226095,
      "grad_norm": 12.125,
      "learning_rate": 1.2109233422295568e-05,
      "loss": 1.1165,
      "num_input_tokens_seen": 54940056,
      "step": 94695
    },
    {
      "epoch": 14.104855525767054,
      "grad_norm": 16.375,
      "learning_rate": 1.2106449391278802e-05,
      "loss": 1.0461,
      "num_input_tokens_seen": 54943032,
      "step": 94700
    },
    {
      "epoch": 14.105600238308012,
      "grad_norm": 6.8125,
      "learning_rate": 1.2103665578085458e-05,
      "loss": 1.2843,
      "num_input_tokens_seen": 54946040,
      "step": 94705
    },
    {
      "epoch": 14.106344950848973,
      "grad_norm": 11.3125,
      "learning_rate": 1.2100881982762589e-05,
      "loss": 1.1523,
      "num_input_tokens_seen": 54948728,
      "step": 94710
    },
    {
      "epoch": 14.107089663389932,
      "grad_norm": 10.375,
      "learning_rate": 1.2098098605357205e-05,
      "loss": 1.1474,
      "num_input_tokens_seen": 54951672,
      "step": 94715
    },
    {
      "epoch": 14.10783437593089,
      "grad_norm": 14.1875,
      "learning_rate": 1.2095315445916323e-05,
      "loss": 1.4301,
      "num_input_tokens_seen": 54954520,
      "step": 94720
    },
    {
      "epoch": 14.10857908847185,
      "grad_norm": 11.25,
      "learning_rate": 1.2092532504486981e-05,
      "loss": 1.2223,
      "num_input_tokens_seen": 54957592,
      "step": 94725
    },
    {
      "epoch": 14.10932380101281,
      "grad_norm": 8.375,
      "learning_rate": 1.2089749781116175e-05,
      "loss": 1.2527,
      "num_input_tokens_seen": 54960696,
      "step": 94730
    },
    {
      "epoch": 14.110068513553768,
      "grad_norm": 9.4375,
      "learning_rate": 1.2086967275850936e-05,
      "loss": 1.0046,
      "num_input_tokens_seen": 54963288,
      "step": 94735
    },
    {
      "epoch": 14.110813226094727,
      "grad_norm": 13.8125,
      "learning_rate": 1.2084184988738247e-05,
      "loss": 1.4301,
      "num_input_tokens_seen": 54966232,
      "step": 94740
    },
    {
      "epoch": 14.111557938635686,
      "grad_norm": 17.625,
      "learning_rate": 1.2081402919825139e-05,
      "loss": 1.187,
      "num_input_tokens_seen": 54969400,
      "step": 94745
    },
    {
      "epoch": 14.112302651176647,
      "grad_norm": 9.0,
      "learning_rate": 1.2078621069158596e-05,
      "loss": 0.9405,
      "num_input_tokens_seen": 54972536,
      "step": 94750
    },
    {
      "epoch": 14.113047363717605,
      "grad_norm": 14.1875,
      "learning_rate": 1.2075839436785611e-05,
      "loss": 1.0944,
      "num_input_tokens_seen": 54975448,
      "step": 94755
    },
    {
      "epoch": 14.113792076258564,
      "grad_norm": 10.8125,
      "learning_rate": 1.2073058022753189e-05,
      "loss": 1.0973,
      "num_input_tokens_seen": 54978424,
      "step": 94760
    },
    {
      "epoch": 14.114536788799523,
      "grad_norm": 9.1875,
      "learning_rate": 1.2070276827108315e-05,
      "loss": 1.0109,
      "num_input_tokens_seen": 54981272,
      "step": 94765
    },
    {
      "epoch": 14.115281501340483,
      "grad_norm": 19.0,
      "learning_rate": 1.2067495849897972e-05,
      "loss": 0.7868,
      "num_input_tokens_seen": 54984216,
      "step": 94770
    },
    {
      "epoch": 14.116026213881442,
      "grad_norm": 8.625,
      "learning_rate": 1.2064715091169135e-05,
      "loss": 0.9256,
      "num_input_tokens_seen": 54987320,
      "step": 94775
    },
    {
      "epoch": 14.1167709264224,
      "grad_norm": 9.3125,
      "learning_rate": 1.2061934550968798e-05,
      "loss": 1.1127,
      "num_input_tokens_seen": 54990104,
      "step": 94780
    },
    {
      "epoch": 14.11751563896336,
      "grad_norm": 9.4375,
      "learning_rate": 1.2059154229343919e-05,
      "loss": 1.2169,
      "num_input_tokens_seen": 54992792,
      "step": 94785
    },
    {
      "epoch": 14.11826035150432,
      "grad_norm": 5.71875,
      "learning_rate": 1.2056374126341485e-05,
      "loss": 0.9519,
      "num_input_tokens_seen": 54995800,
      "step": 94790
    },
    {
      "epoch": 14.119005064045279,
      "grad_norm": 9.75,
      "learning_rate": 1.2053594242008453e-05,
      "loss": 0.9599,
      "num_input_tokens_seen": 54998872,
      "step": 94795
    },
    {
      "epoch": 14.119749776586238,
      "grad_norm": 12.25,
      "learning_rate": 1.205081457639178e-05,
      "loss": 0.956,
      "num_input_tokens_seen": 55001848,
      "step": 94800
    },
    {
      "epoch": 14.120494489127196,
      "grad_norm": 7.9375,
      "learning_rate": 1.2048035129538446e-05,
      "loss": 0.7545,
      "num_input_tokens_seen": 55004504,
      "step": 94805
    },
    {
      "epoch": 14.121239201668157,
      "grad_norm": 12.625,
      "learning_rate": 1.2045255901495384e-05,
      "loss": 1.042,
      "num_input_tokens_seen": 55007640,
      "step": 94810
    },
    {
      "epoch": 14.121983914209116,
      "grad_norm": 20.625,
      "learning_rate": 1.2042476892309565e-05,
      "loss": 1.1124,
      "num_input_tokens_seen": 55010264,
      "step": 94815
    },
    {
      "epoch": 14.122728626750074,
      "grad_norm": 7.1875,
      "learning_rate": 1.203969810202793e-05,
      "loss": 0.9055,
      "num_input_tokens_seen": 55013144,
      "step": 94820
    },
    {
      "epoch": 14.123473339291033,
      "grad_norm": 8.375,
      "learning_rate": 1.2036919530697412e-05,
      "loss": 0.8247,
      "num_input_tokens_seen": 55016152,
      "step": 94825
    },
    {
      "epoch": 14.124218051831992,
      "grad_norm": 6.09375,
      "learning_rate": 1.2034141178364974e-05,
      "loss": 1.0054,
      "num_input_tokens_seen": 55019064,
      "step": 94830
    },
    {
      "epoch": 14.124962764372953,
      "grad_norm": 5.125,
      "learning_rate": 1.2031363045077545e-05,
      "loss": 0.9156,
      "num_input_tokens_seen": 55021848,
      "step": 94835
    },
    {
      "epoch": 14.125707476913911,
      "grad_norm": 8.6875,
      "learning_rate": 1.2028585130882056e-05,
      "loss": 0.7623,
      "num_input_tokens_seen": 55024728,
      "step": 94840
    },
    {
      "epoch": 14.12645218945487,
      "grad_norm": 6.78125,
      "learning_rate": 1.2025807435825426e-05,
      "loss": 1.031,
      "num_input_tokens_seen": 55027672,
      "step": 94845
    },
    {
      "epoch": 14.127196901995829,
      "grad_norm": 9.4375,
      "learning_rate": 1.2023029959954603e-05,
      "loss": 0.94,
      "num_input_tokens_seen": 55030456,
      "step": 94850
    },
    {
      "epoch": 14.12794161453679,
      "grad_norm": 21.25,
      "learning_rate": 1.2020252703316492e-05,
      "loss": 0.9223,
      "num_input_tokens_seen": 55033528,
      "step": 94855
    },
    {
      "epoch": 14.128686327077748,
      "grad_norm": 11.9375,
      "learning_rate": 1.2017475665958028e-05,
      "loss": 1.3527,
      "num_input_tokens_seen": 55036280,
      "step": 94860
    },
    {
      "epoch": 14.129431039618707,
      "grad_norm": 11.875,
      "learning_rate": 1.201469884792611e-05,
      "loss": 1.0112,
      "num_input_tokens_seen": 55039384,
      "step": 94865
    },
    {
      "epoch": 14.130175752159666,
      "grad_norm": 12.125,
      "learning_rate": 1.2011922249267662e-05,
      "loss": 0.817,
      "num_input_tokens_seen": 55042296,
      "step": 94870
    },
    {
      "epoch": 14.130920464700626,
      "grad_norm": 12.625,
      "learning_rate": 1.2009145870029592e-05,
      "loss": 1.2172,
      "num_input_tokens_seen": 55045144,
      "step": 94875
    },
    {
      "epoch": 14.131665177241585,
      "grad_norm": 18.875,
      "learning_rate": 1.200636971025879e-05,
      "loss": 1.4776,
      "num_input_tokens_seen": 55047832,
      "step": 94880
    },
    {
      "epoch": 14.132409889782544,
      "grad_norm": 11.9375,
      "learning_rate": 1.2003593770002169e-05,
      "loss": 1.0684,
      "num_input_tokens_seen": 55050680,
      "step": 94885
    },
    {
      "epoch": 14.133154602323502,
      "grad_norm": 8.3125,
      "learning_rate": 1.2000818049306628e-05,
      "loss": 1.0427,
      "num_input_tokens_seen": 55053656,
      "step": 94890
    },
    {
      "epoch": 14.133899314864463,
      "grad_norm": 17.625,
      "learning_rate": 1.1998042548219052e-05,
      "loss": 1.4583,
      "num_input_tokens_seen": 55056536,
      "step": 94895
    },
    {
      "epoch": 14.134644027405422,
      "grad_norm": 11.0625,
      "learning_rate": 1.1995267266786325e-05,
      "loss": 0.7951,
      "num_input_tokens_seen": 55059256,
      "step": 94900
    },
    {
      "epoch": 14.13538873994638,
      "grad_norm": 8.4375,
      "learning_rate": 1.1992492205055347e-05,
      "loss": 1.0111,
      "num_input_tokens_seen": 55062488,
      "step": 94905
    },
    {
      "epoch": 14.13613345248734,
      "grad_norm": 8.5625,
      "learning_rate": 1.1989717363072986e-05,
      "loss": 1.3582,
      "num_input_tokens_seen": 55065304,
      "step": 94910
    },
    {
      "epoch": 14.1368781650283,
      "grad_norm": 8.25,
      "learning_rate": 1.1986942740886135e-05,
      "loss": 0.9346,
      "num_input_tokens_seen": 55068408,
      "step": 94915
    },
    {
      "epoch": 14.137622877569259,
      "grad_norm": 8.8125,
      "learning_rate": 1.198416833854166e-05,
      "loss": 1.1147,
      "num_input_tokens_seen": 55071416,
      "step": 94920
    },
    {
      "epoch": 14.138367590110217,
      "grad_norm": 10.9375,
      "learning_rate": 1.1981394156086423e-05,
      "loss": 1.2467,
      "num_input_tokens_seen": 55074264,
      "step": 94925
    },
    {
      "epoch": 14.139112302651176,
      "grad_norm": 8.875,
      "learning_rate": 1.197862019356731e-05,
      "loss": 1.112,
      "num_input_tokens_seen": 55077048,
      "step": 94930
    },
    {
      "epoch": 14.139857015192137,
      "grad_norm": 12.75,
      "learning_rate": 1.1975846451031167e-05,
      "loss": 1.0302,
      "num_input_tokens_seen": 55080088,
      "step": 94935
    },
    {
      "epoch": 14.140601727733095,
      "grad_norm": 6.9375,
      "learning_rate": 1.1973072928524868e-05,
      "loss": 0.8683,
      "num_input_tokens_seen": 55083032,
      "step": 94940
    },
    {
      "epoch": 14.141346440274054,
      "grad_norm": 12.625,
      "learning_rate": 1.1970299626095252e-05,
      "loss": 1.1798,
      "num_input_tokens_seen": 55085944,
      "step": 94945
    },
    {
      "epoch": 14.142091152815013,
      "grad_norm": 12.8125,
      "learning_rate": 1.1967526543789192e-05,
      "loss": 0.9103,
      "num_input_tokens_seen": 55089016,
      "step": 94950
    },
    {
      "epoch": 14.142835865355973,
      "grad_norm": 12.0,
      "learning_rate": 1.1964753681653526e-05,
      "loss": 1.0094,
      "num_input_tokens_seen": 55091672,
      "step": 94955
    },
    {
      "epoch": 14.143580577896932,
      "grad_norm": 8.625,
      "learning_rate": 1.1961981039735096e-05,
      "loss": 0.9086,
      "num_input_tokens_seen": 55094488,
      "step": 94960
    },
    {
      "epoch": 14.14432529043789,
      "grad_norm": 12.125,
      "learning_rate": 1.1959208618080747e-05,
      "loss": 1.2654,
      "num_input_tokens_seen": 55097368,
      "step": 94965
    },
    {
      "epoch": 14.14507000297885,
      "grad_norm": 9.3125,
      "learning_rate": 1.1956436416737304e-05,
      "loss": 1.3167,
      "num_input_tokens_seen": 55100184,
      "step": 94970
    },
    {
      "epoch": 14.14581471551981,
      "grad_norm": 8.5,
      "learning_rate": 1.1953664435751621e-05,
      "loss": 0.9388,
      "num_input_tokens_seen": 55103096,
      "step": 94975
    },
    {
      "epoch": 14.146559428060769,
      "grad_norm": 8.0625,
      "learning_rate": 1.1950892675170509e-05,
      "loss": 0.8205,
      "num_input_tokens_seen": 55105784,
      "step": 94980
    },
    {
      "epoch": 14.147304140601728,
      "grad_norm": 8.3125,
      "learning_rate": 1.194812113504081e-05,
      "loss": 0.9952,
      "num_input_tokens_seen": 55108440,
      "step": 94985
    },
    {
      "epoch": 14.148048853142686,
      "grad_norm": 9.8125,
      "learning_rate": 1.194534981540933e-05,
      "loss": 1.1957,
      "num_input_tokens_seen": 55111096,
      "step": 94990
    },
    {
      "epoch": 14.148793565683647,
      "grad_norm": 11.0,
      "learning_rate": 1.1942578716322905e-05,
      "loss": 0.8491,
      "num_input_tokens_seen": 55114200,
      "step": 94995
    },
    {
      "epoch": 14.149538278224606,
      "grad_norm": 6.4375,
      "learning_rate": 1.1939807837828345e-05,
      "loss": 1.1907,
      "num_input_tokens_seen": 55117080,
      "step": 95000
    },
    {
      "epoch": 14.150282990765565,
      "grad_norm": 10.0,
      "learning_rate": 1.1937037179972447e-05,
      "loss": 1.1474,
      "num_input_tokens_seen": 55119544,
      "step": 95005
    },
    {
      "epoch": 14.151027703306523,
      "grad_norm": 8.0625,
      "learning_rate": 1.1934266742802039e-05,
      "loss": 1.3239,
      "num_input_tokens_seen": 55122552,
      "step": 95010
    },
    {
      "epoch": 14.151772415847482,
      "grad_norm": 6.9375,
      "learning_rate": 1.1931496526363903e-05,
      "loss": 0.9352,
      "num_input_tokens_seen": 55125560,
      "step": 95015
    },
    {
      "epoch": 14.152517128388443,
      "grad_norm": 7.1875,
      "learning_rate": 1.1928726530704862e-05,
      "loss": 1.1172,
      "num_input_tokens_seen": 55129080,
      "step": 95020
    },
    {
      "epoch": 14.153261840929401,
      "grad_norm": 5.90625,
      "learning_rate": 1.1925956755871703e-05,
      "loss": 0.9032,
      "num_input_tokens_seen": 55131832,
      "step": 95025
    },
    {
      "epoch": 14.15400655347036,
      "grad_norm": 12.4375,
      "learning_rate": 1.1923187201911215e-05,
      "loss": 1.1274,
      "num_input_tokens_seen": 55134552,
      "step": 95030
    },
    {
      "epoch": 14.154751266011319,
      "grad_norm": 11.4375,
      "learning_rate": 1.1920417868870187e-05,
      "loss": 1.3019,
      "num_input_tokens_seen": 55137656,
      "step": 95035
    },
    {
      "epoch": 14.15549597855228,
      "grad_norm": 20.875,
      "learning_rate": 1.1917648756795399e-05,
      "loss": 1.2669,
      "num_input_tokens_seen": 55140344,
      "step": 95040
    },
    {
      "epoch": 14.156240691093238,
      "grad_norm": 6.9375,
      "learning_rate": 1.1914879865733647e-05,
      "loss": 0.8222,
      "num_input_tokens_seen": 55143448,
      "step": 95045
    },
    {
      "epoch": 14.156985403634197,
      "grad_norm": 6.46875,
      "learning_rate": 1.1912111195731693e-05,
      "loss": 1.0364,
      "num_input_tokens_seen": 55146072,
      "step": 95050
    },
    {
      "epoch": 14.157730116175156,
      "grad_norm": 13.5625,
      "learning_rate": 1.1909342746836325e-05,
      "loss": 1.2124,
      "num_input_tokens_seen": 55148760,
      "step": 95055
    },
    {
      "epoch": 14.158474828716116,
      "grad_norm": 10.25,
      "learning_rate": 1.1906574519094299e-05,
      "loss": 1.0485,
      "num_input_tokens_seen": 55151896,
      "step": 95060
    },
    {
      "epoch": 14.159219541257075,
      "grad_norm": 11.0625,
      "learning_rate": 1.1903806512552395e-05,
      "loss": 0.9132,
      "num_input_tokens_seen": 55154680,
      "step": 95065
    },
    {
      "epoch": 14.159964253798034,
      "grad_norm": 10.75,
      "learning_rate": 1.1901038727257366e-05,
      "loss": 1.1513,
      "num_input_tokens_seen": 55157848,
      "step": 95070
    },
    {
      "epoch": 14.160708966338992,
      "grad_norm": 14.0,
      "learning_rate": 1.189827116325598e-05,
      "loss": 1.3636,
      "num_input_tokens_seen": 55160952,
      "step": 95075
    },
    {
      "epoch": 14.161453678879953,
      "grad_norm": 8.25,
      "learning_rate": 1.1895503820594985e-05,
      "loss": 0.982,
      "num_input_tokens_seen": 55163864,
      "step": 95080
    },
    {
      "epoch": 14.162198391420912,
      "grad_norm": 6.53125,
      "learning_rate": 1.189273669932113e-05,
      "loss": 0.9923,
      "num_input_tokens_seen": 55166872,
      "step": 95085
    },
    {
      "epoch": 14.16294310396187,
      "grad_norm": 10.5625,
      "learning_rate": 1.1889969799481173e-05,
      "loss": 0.7439,
      "num_input_tokens_seen": 55169912,
      "step": 95090
    },
    {
      "epoch": 14.16368781650283,
      "grad_norm": 10.3125,
      "learning_rate": 1.1887203121121851e-05,
      "loss": 0.8819,
      "num_input_tokens_seen": 55173176,
      "step": 95095
    },
    {
      "epoch": 14.16443252904379,
      "grad_norm": 12.3125,
      "learning_rate": 1.1884436664289908e-05,
      "loss": 1.3562,
      "num_input_tokens_seen": 55175896,
      "step": 95100
    },
    {
      "epoch": 14.165177241584749,
      "grad_norm": 16.0,
      "learning_rate": 1.1881670429032066e-05,
      "loss": 0.9965,
      "num_input_tokens_seen": 55178968,
      "step": 95105
    },
    {
      "epoch": 14.165921954125707,
      "grad_norm": 8.25,
      "learning_rate": 1.1878904415395078e-05,
      "loss": 1.1887,
      "num_input_tokens_seen": 55181912,
      "step": 95110
    },
    {
      "epoch": 14.166666666666666,
      "grad_norm": 20.0,
      "learning_rate": 1.1876138623425667e-05,
      "loss": 1.0173,
      "num_input_tokens_seen": 55184696,
      "step": 95115
    },
    {
      "epoch": 14.167411379207627,
      "grad_norm": 15.9375,
      "learning_rate": 1.1873373053170545e-05,
      "loss": 0.9656,
      "num_input_tokens_seen": 55187512,
      "step": 95120
    },
    {
      "epoch": 14.168156091748585,
      "grad_norm": 7.75,
      "learning_rate": 1.187060770467645e-05,
      "loss": 0.8995,
      "num_input_tokens_seen": 55190360,
      "step": 95125
    },
    {
      "epoch": 14.168900804289544,
      "grad_norm": 10.1875,
      "learning_rate": 1.1867842577990087e-05,
      "loss": 1.0669,
      "num_input_tokens_seen": 55193272,
      "step": 95130
    },
    {
      "epoch": 14.169645516830503,
      "grad_norm": 8.625,
      "learning_rate": 1.1865077673158188e-05,
      "loss": 0.8988,
      "num_input_tokens_seen": 55196024,
      "step": 95135
    },
    {
      "epoch": 14.170390229371463,
      "grad_norm": 10.5,
      "learning_rate": 1.186231299022744e-05,
      "loss": 1.1431,
      "num_input_tokens_seen": 55199160,
      "step": 95140
    },
    {
      "epoch": 14.171134941912422,
      "grad_norm": 3.953125,
      "learning_rate": 1.1859548529244571e-05,
      "loss": 1.1193,
      "num_input_tokens_seen": 55203032,
      "step": 95145
    },
    {
      "epoch": 14.171879654453381,
      "grad_norm": 9.75,
      "learning_rate": 1.1856784290256276e-05,
      "loss": 1.0244,
      "num_input_tokens_seen": 55206008,
      "step": 95150
    },
    {
      "epoch": 14.17262436699434,
      "grad_norm": 7.59375,
      "learning_rate": 1.1854020273309241e-05,
      "loss": 0.9181,
      "num_input_tokens_seen": 55208920,
      "step": 95155
    },
    {
      "epoch": 14.1733690795353,
      "grad_norm": 8.3125,
      "learning_rate": 1.1851256478450181e-05,
      "loss": 1.0947,
      "num_input_tokens_seen": 55212056,
      "step": 95160
    },
    {
      "epoch": 14.174113792076259,
      "grad_norm": 10.625,
      "learning_rate": 1.1848492905725781e-05,
      "loss": 0.8181,
      "num_input_tokens_seen": 55214936,
      "step": 95165
    },
    {
      "epoch": 14.174858504617218,
      "grad_norm": 13.0625,
      "learning_rate": 1.1845729555182728e-05,
      "loss": 1.0328,
      "num_input_tokens_seen": 55217912,
      "step": 95170
    },
    {
      "epoch": 14.175603217158177,
      "grad_norm": 8.25,
      "learning_rate": 1.1842966426867694e-05,
      "loss": 1.071,
      "num_input_tokens_seen": 55220920,
      "step": 95175
    },
    {
      "epoch": 14.176347929699135,
      "grad_norm": 14.9375,
      "learning_rate": 1.1840203520827378e-05,
      "loss": 1.3158,
      "num_input_tokens_seen": 55223608,
      "step": 95180
    },
    {
      "epoch": 14.177092642240096,
      "grad_norm": 5.0625,
      "learning_rate": 1.183744083710844e-05,
      "loss": 0.95,
      "num_input_tokens_seen": 55226456,
      "step": 95185
    },
    {
      "epoch": 14.177837354781055,
      "grad_norm": 9.0,
      "learning_rate": 1.1834678375757571e-05,
      "loss": 0.8438,
      "num_input_tokens_seen": 55229528,
      "step": 95190
    },
    {
      "epoch": 14.178582067322013,
      "grad_norm": 5.4375,
      "learning_rate": 1.183191613682143e-05,
      "loss": 0.8798,
      "num_input_tokens_seen": 55232600,
      "step": 95195
    },
    {
      "epoch": 14.179326779862972,
      "grad_norm": 12.125,
      "learning_rate": 1.1829154120346673e-05,
      "loss": 0.8977,
      "num_input_tokens_seen": 55235672,
      "step": 95200
    },
    {
      "epoch": 14.180071492403933,
      "grad_norm": 6.78125,
      "learning_rate": 1.1826392326379981e-05,
      "loss": 1.1447,
      "num_input_tokens_seen": 55238552,
      "step": 95205
    },
    {
      "epoch": 14.180816204944891,
      "grad_norm": 5.40625,
      "learning_rate": 1.1823630754967991e-05,
      "loss": 1.153,
      "num_input_tokens_seen": 55241496,
      "step": 95210
    },
    {
      "epoch": 14.18156091748585,
      "grad_norm": 8.5,
      "learning_rate": 1.1820869406157378e-05,
      "loss": 0.7681,
      "num_input_tokens_seen": 55244344,
      "step": 95215
    },
    {
      "epoch": 14.182305630026809,
      "grad_norm": 11.1875,
      "learning_rate": 1.181810827999478e-05,
      "loss": 1.2501,
      "num_input_tokens_seen": 55247448,
      "step": 95220
    },
    {
      "epoch": 14.18305034256777,
      "grad_norm": 8.875,
      "learning_rate": 1.1815347376526847e-05,
      "loss": 0.9977,
      "num_input_tokens_seen": 55250200,
      "step": 95225
    },
    {
      "epoch": 14.183795055108728,
      "grad_norm": 8.375,
      "learning_rate": 1.181258669580021e-05,
      "loss": 1.1514,
      "num_input_tokens_seen": 55253048,
      "step": 95230
    },
    {
      "epoch": 14.184539767649687,
      "grad_norm": 10.625,
      "learning_rate": 1.1809826237861527e-05,
      "loss": 1.2111,
      "num_input_tokens_seen": 55256216,
      "step": 95235
    },
    {
      "epoch": 14.185284480190646,
      "grad_norm": 9.6875,
      "learning_rate": 1.1807066002757422e-05,
      "loss": 0.9783,
      "num_input_tokens_seen": 55259000,
      "step": 95240
    },
    {
      "epoch": 14.186029192731606,
      "grad_norm": 10.9375,
      "learning_rate": 1.180430599053452e-05,
      "loss": 1.1459,
      "num_input_tokens_seen": 55261592,
      "step": 95245
    },
    {
      "epoch": 14.186773905272565,
      "grad_norm": 8.1875,
      "learning_rate": 1.1801546201239466e-05,
      "loss": 0.9488,
      "num_input_tokens_seen": 55264312,
      "step": 95250
    },
    {
      "epoch": 14.187518617813524,
      "grad_norm": 6.03125,
      "learning_rate": 1.1798786634918868e-05,
      "loss": 1.4439,
      "num_input_tokens_seen": 55267128,
      "step": 95255
    },
    {
      "epoch": 14.188263330354483,
      "grad_norm": 11.5625,
      "learning_rate": 1.1796027291619358e-05,
      "loss": 1.1563,
      "num_input_tokens_seen": 55269880,
      "step": 95260
    },
    {
      "epoch": 14.189008042895443,
      "grad_norm": 13.3125,
      "learning_rate": 1.1793268171387539e-05,
      "loss": 1.2473,
      "num_input_tokens_seen": 55272696,
      "step": 95265
    },
    {
      "epoch": 14.189752755436402,
      "grad_norm": 7.25,
      "learning_rate": 1.1790509274270042e-05,
      "loss": 0.9309,
      "num_input_tokens_seen": 55275544,
      "step": 95270
    },
    {
      "epoch": 14.19049746797736,
      "grad_norm": 2.28125,
      "learning_rate": 1.1787750600313465e-05,
      "loss": 1.212,
      "num_input_tokens_seen": 55278552,
      "step": 95275
    },
    {
      "epoch": 14.19124218051832,
      "grad_norm": 14.75,
      "learning_rate": 1.1784992149564403e-05,
      "loss": 1.0234,
      "num_input_tokens_seen": 55281464,
      "step": 95280
    },
    {
      "epoch": 14.19198689305928,
      "grad_norm": 16.375,
      "learning_rate": 1.1782233922069478e-05,
      "loss": 1.3979,
      "num_input_tokens_seen": 55284408,
      "step": 95285
    },
    {
      "epoch": 14.192731605600239,
      "grad_norm": 22.625,
      "learning_rate": 1.1779475917875278e-05,
      "loss": 1.159,
      "num_input_tokens_seen": 55287064,
      "step": 95290
    },
    {
      "epoch": 14.193476318141197,
      "grad_norm": 8.4375,
      "learning_rate": 1.1776718137028392e-05,
      "loss": 1.1814,
      "num_input_tokens_seen": 55289752,
      "step": 95295
    },
    {
      "epoch": 14.194221030682156,
      "grad_norm": 8.75,
      "learning_rate": 1.1773960579575408e-05,
      "loss": 0.8767,
      "num_input_tokens_seen": 55292504,
      "step": 95300
    },
    {
      "epoch": 14.194965743223117,
      "grad_norm": 8.3125,
      "learning_rate": 1.1771203245562924e-05,
      "loss": 1.1739,
      "num_input_tokens_seen": 55295448,
      "step": 95305
    },
    {
      "epoch": 14.195710455764075,
      "grad_norm": 12.8125,
      "learning_rate": 1.176844613503751e-05,
      "loss": 0.8551,
      "num_input_tokens_seen": 55298520,
      "step": 95310
    },
    {
      "epoch": 14.196455168305034,
      "grad_norm": 10.4375,
      "learning_rate": 1.1765689248045755e-05,
      "loss": 1.0576,
      "num_input_tokens_seen": 55301304,
      "step": 95315
    },
    {
      "epoch": 14.197199880845993,
      "grad_norm": 11.125,
      "learning_rate": 1.1762932584634234e-05,
      "loss": 0.9422,
      "num_input_tokens_seen": 55304056,
      "step": 95320
    },
    {
      "epoch": 14.197944593386953,
      "grad_norm": 10.625,
      "learning_rate": 1.1760176144849502e-05,
      "loss": 1.3125,
      "num_input_tokens_seen": 55306808,
      "step": 95325
    },
    {
      "epoch": 14.198689305927912,
      "grad_norm": 13.25,
      "learning_rate": 1.1757419928738147e-05,
      "loss": 1.0549,
      "num_input_tokens_seen": 55309624,
      "step": 95330
    },
    {
      "epoch": 14.199434018468871,
      "grad_norm": 16.25,
      "learning_rate": 1.1754663936346713e-05,
      "loss": 1.5524,
      "num_input_tokens_seen": 55312504,
      "step": 95335
    },
    {
      "epoch": 14.20017873100983,
      "grad_norm": 15.9375,
      "learning_rate": 1.1751908167721782e-05,
      "loss": 0.8573,
      "num_input_tokens_seen": 55315384,
      "step": 95340
    },
    {
      "epoch": 14.200923443550789,
      "grad_norm": 12.0625,
      "learning_rate": 1.1749152622909884e-05,
      "loss": 1.1134,
      "num_input_tokens_seen": 55318296,
      "step": 95345
    },
    {
      "epoch": 14.201668156091749,
      "grad_norm": 6.4375,
      "learning_rate": 1.1746397301957598e-05,
      "loss": 1.1577,
      "num_input_tokens_seen": 55320920,
      "step": 95350
    },
    {
      "epoch": 14.202412868632708,
      "grad_norm": 14.125,
      "learning_rate": 1.174364220491146e-05,
      "loss": 0.9811,
      "num_input_tokens_seen": 55324152,
      "step": 95355
    },
    {
      "epoch": 14.203157581173667,
      "grad_norm": 9.4375,
      "learning_rate": 1.1740887331818009e-05,
      "loss": 1.2585,
      "num_input_tokens_seen": 55327096,
      "step": 95360
    },
    {
      "epoch": 14.203902293714625,
      "grad_norm": 7.375,
      "learning_rate": 1.1738132682723797e-05,
      "loss": 0.9457,
      "num_input_tokens_seen": 55330424,
      "step": 95365
    },
    {
      "epoch": 14.204647006255586,
      "grad_norm": 20.25,
      "learning_rate": 1.1735378257675338e-05,
      "loss": 1.3087,
      "num_input_tokens_seen": 55333464,
      "step": 95370
    },
    {
      "epoch": 14.205391718796545,
      "grad_norm": 15.5625,
      "learning_rate": 1.1732624056719197e-05,
      "loss": 1.1556,
      "num_input_tokens_seen": 55336184,
      "step": 95375
    },
    {
      "epoch": 14.206136431337503,
      "grad_norm": 12.6875,
      "learning_rate": 1.1729870079901875e-05,
      "loss": 0.8947,
      "num_input_tokens_seen": 55339160,
      "step": 95380
    },
    {
      "epoch": 14.206881143878462,
      "grad_norm": 23.625,
      "learning_rate": 1.1727116327269924e-05,
      "loss": 1.3021,
      "num_input_tokens_seen": 55342008,
      "step": 95385
    },
    {
      "epoch": 14.207625856419423,
      "grad_norm": 14.0625,
      "learning_rate": 1.172436279886984e-05,
      "loss": 0.9641,
      "num_input_tokens_seen": 55344792,
      "step": 95390
    },
    {
      "epoch": 14.208370568960381,
      "grad_norm": 10.75,
      "learning_rate": 1.1721609494748164e-05,
      "loss": 1.3847,
      "num_input_tokens_seen": 55347800,
      "step": 95395
    },
    {
      "epoch": 14.20911528150134,
      "grad_norm": 6.25,
      "learning_rate": 1.1718856414951402e-05,
      "loss": 0.8982,
      "num_input_tokens_seen": 55350712,
      "step": 95400
    },
    {
      "epoch": 14.209859994042299,
      "grad_norm": 10.8125,
      "learning_rate": 1.1716103559526051e-05,
      "loss": 1.1077,
      "num_input_tokens_seen": 55353368,
      "step": 95405
    },
    {
      "epoch": 14.21060470658326,
      "grad_norm": 19.25,
      "learning_rate": 1.1713350928518639e-05,
      "loss": 1.3097,
      "num_input_tokens_seen": 55355992,
      "step": 95410
    },
    {
      "epoch": 14.211349419124218,
      "grad_norm": 13.875,
      "learning_rate": 1.171059852197565e-05,
      "loss": 1.148,
      "num_input_tokens_seen": 55358744,
      "step": 95415
    },
    {
      "epoch": 14.212094131665177,
      "grad_norm": 12.1875,
      "learning_rate": 1.1707846339943601e-05,
      "loss": 1.2428,
      "num_input_tokens_seen": 55361944,
      "step": 95420
    },
    {
      "epoch": 14.212838844206136,
      "grad_norm": 7.75,
      "learning_rate": 1.1705094382468979e-05,
      "loss": 1.2512,
      "num_input_tokens_seen": 55364536,
      "step": 95425
    },
    {
      "epoch": 14.213583556747096,
      "grad_norm": 13.625,
      "learning_rate": 1.1702342649598274e-05,
      "loss": 0.9935,
      "num_input_tokens_seen": 55367320,
      "step": 95430
    },
    {
      "epoch": 14.214328269288055,
      "grad_norm": 10.75,
      "learning_rate": 1.1699591141377967e-05,
      "loss": 1.2245,
      "num_input_tokens_seen": 55370456,
      "step": 95435
    },
    {
      "epoch": 14.215072981829014,
      "grad_norm": 13.0,
      "learning_rate": 1.1696839857854558e-05,
      "loss": 1.2976,
      "num_input_tokens_seen": 55373240,
      "step": 95440
    },
    {
      "epoch": 14.215817694369973,
      "grad_norm": 7.84375,
      "learning_rate": 1.169408879907452e-05,
      "loss": 1.1275,
      "num_input_tokens_seen": 55376120,
      "step": 95445
    },
    {
      "epoch": 14.216562406910933,
      "grad_norm": 10.0,
      "learning_rate": 1.1691337965084321e-05,
      "loss": 1.135,
      "num_input_tokens_seen": 55379352,
      "step": 95450
    },
    {
      "epoch": 14.217307119451892,
      "grad_norm": 8.75,
      "learning_rate": 1.1688587355930444e-05,
      "loss": 0.9098,
      "num_input_tokens_seen": 55382264,
      "step": 95455
    },
    {
      "epoch": 14.21805183199285,
      "grad_norm": 9.5625,
      "learning_rate": 1.168583697165935e-05,
      "loss": 0.9031,
      "num_input_tokens_seen": 55385176,
      "step": 95460
    },
    {
      "epoch": 14.21879654453381,
      "grad_norm": 15.375,
      "learning_rate": 1.1683086812317517e-05,
      "loss": 1.1397,
      "num_input_tokens_seen": 55388152,
      "step": 95465
    },
    {
      "epoch": 14.21954125707477,
      "grad_norm": 10.3125,
      "learning_rate": 1.1680336877951387e-05,
      "loss": 1.534,
      "num_input_tokens_seen": 55391064,
      "step": 95470
    },
    {
      "epoch": 14.220285969615729,
      "grad_norm": 10.625,
      "learning_rate": 1.1677587168607437e-05,
      "loss": 1.0543,
      "num_input_tokens_seen": 55393592,
      "step": 95475
    },
    {
      "epoch": 14.221030682156687,
      "grad_norm": 10.5,
      "learning_rate": 1.1674837684332113e-05,
      "loss": 1.1266,
      "num_input_tokens_seen": 55396472,
      "step": 95480
    },
    {
      "epoch": 14.221775394697646,
      "grad_norm": 10.4375,
      "learning_rate": 1.1672088425171854e-05,
      "loss": 1.0603,
      "num_input_tokens_seen": 55399288,
      "step": 95485
    },
    {
      "epoch": 14.222520107238607,
      "grad_norm": 7.4375,
      "learning_rate": 1.1669339391173122e-05,
      "loss": 1.2113,
      "num_input_tokens_seen": 55402360,
      "step": 95490
    },
    {
      "epoch": 14.223264819779565,
      "grad_norm": 21.375,
      "learning_rate": 1.1666590582382355e-05,
      "loss": 1.2905,
      "num_input_tokens_seen": 55405560,
      "step": 95495
    },
    {
      "epoch": 14.224009532320524,
      "grad_norm": 7.21875,
      "learning_rate": 1.166384199884599e-05,
      "loss": 1.0811,
      "num_input_tokens_seen": 55408664,
      "step": 95500
    },
    {
      "epoch": 14.224754244861483,
      "grad_norm": 16.5,
      "learning_rate": 1.1661093640610445e-05,
      "loss": 1.2289,
      "num_input_tokens_seen": 55411384,
      "step": 95505
    },
    {
      "epoch": 14.225498957402444,
      "grad_norm": 10.5,
      "learning_rate": 1.1658345507722182e-05,
      "loss": 1.3003,
      "num_input_tokens_seen": 55414424,
      "step": 95510
    },
    {
      "epoch": 14.226243669943402,
      "grad_norm": 13.625,
      "learning_rate": 1.1655597600227597e-05,
      "loss": 1.1564,
      "num_input_tokens_seen": 55417240,
      "step": 95515
    },
    {
      "epoch": 14.226988382484361,
      "grad_norm": 12.0625,
      "learning_rate": 1.1652849918173139e-05,
      "loss": 0.9332,
      "num_input_tokens_seen": 55420120,
      "step": 95520
    },
    {
      "epoch": 14.22773309502532,
      "grad_norm": 11.5625,
      "learning_rate": 1.165010246160522e-05,
      "loss": 1.254,
      "num_input_tokens_seen": 55423288,
      "step": 95525
    },
    {
      "epoch": 14.228477807566279,
      "grad_norm": 5.96875,
      "learning_rate": 1.1647355230570237e-05,
      "loss": 0.7635,
      "num_input_tokens_seen": 55426072,
      "step": 95530
    },
    {
      "epoch": 14.229222520107239,
      "grad_norm": 17.125,
      "learning_rate": 1.1644608225114629e-05,
      "loss": 1.2206,
      "num_input_tokens_seen": 55428920,
      "step": 95535
    },
    {
      "epoch": 14.229967232648198,
      "grad_norm": 11.3125,
      "learning_rate": 1.164186144528478e-05,
      "loss": 0.9109,
      "num_input_tokens_seen": 55431960,
      "step": 95540
    },
    {
      "epoch": 14.230711945189157,
      "grad_norm": 11.6875,
      "learning_rate": 1.1639114891127114e-05,
      "loss": 0.9387,
      "num_input_tokens_seen": 55435000,
      "step": 95545
    },
    {
      "epoch": 14.231456657730115,
      "grad_norm": 11.75,
      "learning_rate": 1.1636368562688024e-05,
      "loss": 0.9961,
      "num_input_tokens_seen": 55437848,
      "step": 95550
    },
    {
      "epoch": 14.232201370271076,
      "grad_norm": 7.875,
      "learning_rate": 1.1633622460013904e-05,
      "loss": 0.7264,
      "num_input_tokens_seen": 55441080,
      "step": 95555
    },
    {
      "epoch": 14.232946082812035,
      "grad_norm": 7.71875,
      "learning_rate": 1.163087658315114e-05,
      "loss": 1.5367,
      "num_input_tokens_seen": 55444184,
      "step": 95560
    },
    {
      "epoch": 14.233690795352993,
      "grad_norm": 6.46875,
      "learning_rate": 1.1628130932146137e-05,
      "loss": 1.2022,
      "num_input_tokens_seen": 55447192,
      "step": 95565
    },
    {
      "epoch": 14.234435507893952,
      "grad_norm": 6.96875,
      "learning_rate": 1.1625385507045272e-05,
      "loss": 0.731,
      "num_input_tokens_seen": 55450168,
      "step": 95570
    },
    {
      "epoch": 14.235180220434913,
      "grad_norm": 7.8125,
      "learning_rate": 1.1622640307894913e-05,
      "loss": 1.1117,
      "num_input_tokens_seen": 55453144,
      "step": 95575
    },
    {
      "epoch": 14.235924932975871,
      "grad_norm": 16.75,
      "learning_rate": 1.1619895334741463e-05,
      "loss": 1.0844,
      "num_input_tokens_seen": 55456184,
      "step": 95580
    },
    {
      "epoch": 14.23666964551683,
      "grad_norm": 8.75,
      "learning_rate": 1.161715058763127e-05,
      "loss": 1.1986,
      "num_input_tokens_seen": 55458904,
      "step": 95585
    },
    {
      "epoch": 14.237414358057789,
      "grad_norm": 7.875,
      "learning_rate": 1.1614406066610728e-05,
      "loss": 1.2141,
      "num_input_tokens_seen": 55461976,
      "step": 95590
    },
    {
      "epoch": 14.23815907059875,
      "grad_norm": 9.8125,
      "learning_rate": 1.1611661771726181e-05,
      "loss": 1.0888,
      "num_input_tokens_seen": 55464984,
      "step": 95595
    },
    {
      "epoch": 14.238903783139708,
      "grad_norm": 24.5,
      "learning_rate": 1.1608917703024009e-05,
      "loss": 1.0369,
      "num_input_tokens_seen": 55467896,
      "step": 95600
    },
    {
      "epoch": 14.239648495680667,
      "grad_norm": 5.6875,
      "learning_rate": 1.1606173860550562e-05,
      "loss": 1.0423,
      "num_input_tokens_seen": 55470680,
      "step": 95605
    },
    {
      "epoch": 14.240393208221626,
      "grad_norm": 8.75,
      "learning_rate": 1.1603430244352187e-05,
      "loss": 1.2371,
      "num_input_tokens_seen": 55473592,
      "step": 95610
    },
    {
      "epoch": 14.241137920762586,
      "grad_norm": 7.90625,
      "learning_rate": 1.160068685447525e-05,
      "loss": 1.0372,
      "num_input_tokens_seen": 55476408,
      "step": 95615
    },
    {
      "epoch": 14.241882633303545,
      "grad_norm": 9.875,
      "learning_rate": 1.1597943690966092e-05,
      "loss": 1.2154,
      "num_input_tokens_seen": 55479128,
      "step": 95620
    },
    {
      "epoch": 14.242627345844504,
      "grad_norm": 8.375,
      "learning_rate": 1.1595200753871055e-05,
      "loss": 1.1441,
      "num_input_tokens_seen": 55482168,
      "step": 95625
    },
    {
      "epoch": 14.243372058385463,
      "grad_norm": 7.90625,
      "learning_rate": 1.1592458043236468e-05,
      "loss": 1.1162,
      "num_input_tokens_seen": 55485016,
      "step": 95630
    },
    {
      "epoch": 14.244116770926423,
      "grad_norm": 5.65625,
      "learning_rate": 1.1589715559108682e-05,
      "loss": 1.2378,
      "num_input_tokens_seen": 55487896,
      "step": 95635
    },
    {
      "epoch": 14.244861483467382,
      "grad_norm": 7.03125,
      "learning_rate": 1.1586973301534024e-05,
      "loss": 0.9203,
      "num_input_tokens_seen": 55490584,
      "step": 95640
    },
    {
      "epoch": 14.24560619600834,
      "grad_norm": 9.375,
      "learning_rate": 1.158423127055881e-05,
      "loss": 1.3509,
      "num_input_tokens_seen": 55493592,
      "step": 95645
    },
    {
      "epoch": 14.2463509085493,
      "grad_norm": 11.75,
      "learning_rate": 1.1581489466229381e-05,
      "loss": 1.0783,
      "num_input_tokens_seen": 55496376,
      "step": 95650
    },
    {
      "epoch": 14.24709562109026,
      "grad_norm": 5.71875,
      "learning_rate": 1.1578747888592043e-05,
      "loss": 0.8624,
      "num_input_tokens_seen": 55499576,
      "step": 95655
    },
    {
      "epoch": 14.247840333631219,
      "grad_norm": 10.75,
      "learning_rate": 1.1576006537693127e-05,
      "loss": 1.1536,
      "num_input_tokens_seen": 55502264,
      "step": 95660
    },
    {
      "epoch": 14.248585046172177,
      "grad_norm": 23.0,
      "learning_rate": 1.1573265413578926e-05,
      "loss": 1.0909,
      "num_input_tokens_seen": 55505016,
      "step": 95665
    },
    {
      "epoch": 14.249329758713136,
      "grad_norm": 15.0625,
      "learning_rate": 1.1570524516295773e-05,
      "loss": 1.0442,
      "num_input_tokens_seen": 55507736,
      "step": 95670
    },
    {
      "epoch": 14.250074471254097,
      "grad_norm": 12.6875,
      "learning_rate": 1.1567783845889946e-05,
      "loss": 0.9612,
      "num_input_tokens_seen": 55510488,
      "step": 95675
    },
    {
      "epoch": 14.250819183795056,
      "grad_norm": 17.125,
      "learning_rate": 1.1565043402407768e-05,
      "loss": 1.7123,
      "num_input_tokens_seen": 55513400,
      "step": 95680
    },
    {
      "epoch": 14.251563896336014,
      "grad_norm": 9.8125,
      "learning_rate": 1.1562303185895528e-05,
      "loss": 1.0873,
      "num_input_tokens_seen": 55516568,
      "step": 95685
    },
    {
      "epoch": 14.252308608876973,
      "grad_norm": 15.5625,
      "learning_rate": 1.155956319639952e-05,
      "loss": 1.0225,
      "num_input_tokens_seen": 55519512,
      "step": 95690
    },
    {
      "epoch": 14.253053321417934,
      "grad_norm": 11.1875,
      "learning_rate": 1.155682343396603e-05,
      "loss": 0.8886,
      "num_input_tokens_seen": 55522232,
      "step": 95695
    },
    {
      "epoch": 14.253798033958892,
      "grad_norm": 6.59375,
      "learning_rate": 1.1554083898641335e-05,
      "loss": 0.8346,
      "num_input_tokens_seen": 55525368,
      "step": 95700
    },
    {
      "epoch": 14.254542746499851,
      "grad_norm": 9.25,
      "learning_rate": 1.1551344590471739e-05,
      "loss": 0.9304,
      "num_input_tokens_seen": 55528184,
      "step": 95705
    },
    {
      "epoch": 14.25528745904081,
      "grad_norm": 8.3125,
      "learning_rate": 1.1548605509503496e-05,
      "loss": 0.9039,
      "num_input_tokens_seen": 55531128,
      "step": 95710
    },
    {
      "epoch": 14.256032171581769,
      "grad_norm": 15.9375,
      "learning_rate": 1.15458666557829e-05,
      "loss": 1.173,
      "num_input_tokens_seen": 55533816,
      "step": 95715
    },
    {
      "epoch": 14.25677688412273,
      "grad_norm": 9.1875,
      "learning_rate": 1.1543128029356215e-05,
      "loss": 1.1648,
      "num_input_tokens_seen": 55536888,
      "step": 95720
    },
    {
      "epoch": 14.257521596663688,
      "grad_norm": 6.46875,
      "learning_rate": 1.1540389630269693e-05,
      "loss": 1.3274,
      "num_input_tokens_seen": 55539832,
      "step": 95725
    },
    {
      "epoch": 14.258266309204647,
      "grad_norm": 9.4375,
      "learning_rate": 1.153765145856962e-05,
      "loss": 0.9852,
      "num_input_tokens_seen": 55542872,
      "step": 95730
    },
    {
      "epoch": 14.259011021745605,
      "grad_norm": 15.4375,
      "learning_rate": 1.1534913514302232e-05,
      "loss": 1.5184,
      "num_input_tokens_seen": 55545720,
      "step": 95735
    },
    {
      "epoch": 14.259755734286566,
      "grad_norm": 8.75,
      "learning_rate": 1.1532175797513806e-05,
      "loss": 1.1523,
      "num_input_tokens_seen": 55548664,
      "step": 95740
    },
    {
      "epoch": 14.260500446827525,
      "grad_norm": 12.25,
      "learning_rate": 1.152943830825057e-05,
      "loss": 1.1172,
      "num_input_tokens_seen": 55551608,
      "step": 95745
    },
    {
      "epoch": 14.261245159368483,
      "grad_norm": 13.5625,
      "learning_rate": 1.1526701046558794e-05,
      "loss": 0.9716,
      "num_input_tokens_seen": 55554168,
      "step": 95750
    },
    {
      "epoch": 14.261989871909442,
      "grad_norm": 7.4375,
      "learning_rate": 1.1523964012484712e-05,
      "loss": 0.6483,
      "num_input_tokens_seen": 55557272,
      "step": 95755
    },
    {
      "epoch": 14.262734584450403,
      "grad_norm": 6.9375,
      "learning_rate": 1.1521227206074559e-05,
      "loss": 1.1607,
      "num_input_tokens_seen": 55560344,
      "step": 95760
    },
    {
      "epoch": 14.263479296991362,
      "grad_norm": 10.375,
      "learning_rate": 1.1518490627374572e-05,
      "loss": 0.8244,
      "num_input_tokens_seen": 55563224,
      "step": 95765
    },
    {
      "epoch": 14.26422400953232,
      "grad_norm": 10.25,
      "learning_rate": 1.151575427643098e-05,
      "loss": 1.2034,
      "num_input_tokens_seen": 55566232,
      "step": 95770
    },
    {
      "epoch": 14.264968722073279,
      "grad_norm": 11.375,
      "learning_rate": 1.1513018153290018e-05,
      "loss": 0.9227,
      "num_input_tokens_seen": 55569016,
      "step": 95775
    },
    {
      "epoch": 14.26571343461424,
      "grad_norm": 7.0625,
      "learning_rate": 1.15102822579979e-05,
      "loss": 0.8114,
      "num_input_tokens_seen": 55571832,
      "step": 95780
    },
    {
      "epoch": 14.266458147155198,
      "grad_norm": 13.125,
      "learning_rate": 1.1507546590600862e-05,
      "loss": 1.0561,
      "num_input_tokens_seen": 55574552,
      "step": 95785
    },
    {
      "epoch": 14.267202859696157,
      "grad_norm": 11.25,
      "learning_rate": 1.15048111511451e-05,
      "loss": 1.4107,
      "num_input_tokens_seen": 55577624,
      "step": 95790
    },
    {
      "epoch": 14.267947572237116,
      "grad_norm": 14.5625,
      "learning_rate": 1.1502075939676852e-05,
      "loss": 1.1008,
      "num_input_tokens_seen": 55580536,
      "step": 95795
    },
    {
      "epoch": 14.268692284778076,
      "grad_norm": 9.75,
      "learning_rate": 1.1499340956242307e-05,
      "loss": 0.9987,
      "num_input_tokens_seen": 55583096,
      "step": 95800
    },
    {
      "epoch": 14.269436997319035,
      "grad_norm": 10.5,
      "learning_rate": 1.1496606200887669e-05,
      "loss": 0.8764,
      "num_input_tokens_seen": 55586040,
      "step": 95805
    },
    {
      "epoch": 14.270181709859994,
      "grad_norm": 5.875,
      "learning_rate": 1.1493871673659155e-05,
      "loss": 1.0679,
      "num_input_tokens_seen": 55588728,
      "step": 95810
    },
    {
      "epoch": 14.270926422400953,
      "grad_norm": 7.90625,
      "learning_rate": 1.1491137374602939e-05,
      "loss": 0.7618,
      "num_input_tokens_seen": 55591544,
      "step": 95815
    },
    {
      "epoch": 14.271671134941913,
      "grad_norm": 11.75,
      "learning_rate": 1.1488403303765239e-05,
      "loss": 1.0723,
      "num_input_tokens_seen": 55594776,
      "step": 95820
    },
    {
      "epoch": 14.272415847482872,
      "grad_norm": 8.9375,
      "learning_rate": 1.1485669461192233e-05,
      "loss": 0.781,
      "num_input_tokens_seen": 55597944,
      "step": 95825
    },
    {
      "epoch": 14.27316056002383,
      "grad_norm": 8.5625,
      "learning_rate": 1.1482935846930104e-05,
      "loss": 1.4176,
      "num_input_tokens_seen": 55600568,
      "step": 95830
    },
    {
      "epoch": 14.27390527256479,
      "grad_norm": 7.09375,
      "learning_rate": 1.148020246102503e-05,
      "loss": 1.2007,
      "num_input_tokens_seen": 55603448,
      "step": 95835
    },
    {
      "epoch": 14.27464998510575,
      "grad_norm": 9.0,
      "learning_rate": 1.14774693035232e-05,
      "loss": 1.1273,
      "num_input_tokens_seen": 55606488,
      "step": 95840
    },
    {
      "epoch": 14.275394697646709,
      "grad_norm": 15.5625,
      "learning_rate": 1.1474736374470785e-05,
      "loss": 1.7899,
      "num_input_tokens_seen": 55609400,
      "step": 95845
    },
    {
      "epoch": 14.276139410187668,
      "grad_norm": 9.6875,
      "learning_rate": 1.1472003673913942e-05,
      "loss": 1.015,
      "num_input_tokens_seen": 55612440,
      "step": 95850
    },
    {
      "epoch": 14.276884122728626,
      "grad_norm": 10.4375,
      "learning_rate": 1.1469271201898857e-05,
      "loss": 1.0506,
      "num_input_tokens_seen": 55615352,
      "step": 95855
    },
    {
      "epoch": 14.277628835269585,
      "grad_norm": 10.125,
      "learning_rate": 1.1466538958471673e-05,
      "loss": 1.1715,
      "num_input_tokens_seen": 55618488,
      "step": 95860
    },
    {
      "epoch": 14.278373547810546,
      "grad_norm": 14.1875,
      "learning_rate": 1.1463806943678571e-05,
      "loss": 1.1254,
      "num_input_tokens_seen": 55621432,
      "step": 95865
    },
    {
      "epoch": 14.279118260351504,
      "grad_norm": 9.25,
      "learning_rate": 1.1461075157565681e-05,
      "loss": 0.8925,
      "num_input_tokens_seen": 55623960,
      "step": 95870
    },
    {
      "epoch": 14.279862972892463,
      "grad_norm": 5.21875,
      "learning_rate": 1.1458343600179175e-05,
      "loss": 1.2261,
      "num_input_tokens_seen": 55626712,
      "step": 95875
    },
    {
      "epoch": 14.280607685433422,
      "grad_norm": 9.6875,
      "learning_rate": 1.1455612271565192e-05,
      "loss": 1.2062,
      "num_input_tokens_seen": 55629784,
      "step": 95880
    },
    {
      "epoch": 14.281352397974382,
      "grad_norm": 6.6875,
      "learning_rate": 1.1452881171769872e-05,
      "loss": 1.0127,
      "num_input_tokens_seen": 55632856,
      "step": 95885
    },
    {
      "epoch": 14.282097110515341,
      "grad_norm": 13.8125,
      "learning_rate": 1.145015030083935e-05,
      "loss": 1.3308,
      "num_input_tokens_seen": 55636184,
      "step": 95890
    },
    {
      "epoch": 14.2828418230563,
      "grad_norm": 8.4375,
      "learning_rate": 1.1447419658819775e-05,
      "loss": 0.9646,
      "num_input_tokens_seen": 55638936,
      "step": 95895
    },
    {
      "epoch": 14.283586535597259,
      "grad_norm": 13.0625,
      "learning_rate": 1.1444689245757268e-05,
      "loss": 1.0233,
      "num_input_tokens_seen": 55641752,
      "step": 95900
    },
    {
      "epoch": 14.28433124813822,
      "grad_norm": 12.125,
      "learning_rate": 1.1441959061697952e-05,
      "loss": 0.9668,
      "num_input_tokens_seen": 55644472,
      "step": 95905
    },
    {
      "epoch": 14.285075960679178,
      "grad_norm": 13.6875,
      "learning_rate": 1.1439229106687969e-05,
      "loss": 1.4417,
      "num_input_tokens_seen": 55647384,
      "step": 95910
    },
    {
      "epoch": 14.285820673220137,
      "grad_norm": 12.375,
      "learning_rate": 1.1436499380773416e-05,
      "loss": 1.1975,
      "num_input_tokens_seen": 55650200,
      "step": 95915
    },
    {
      "epoch": 14.286565385761095,
      "grad_norm": 11.4375,
      "learning_rate": 1.1433769884000429e-05,
      "loss": 0.7854,
      "num_input_tokens_seen": 55653368,
      "step": 95920
    },
    {
      "epoch": 14.287310098302056,
      "grad_norm": 14.5,
      "learning_rate": 1.1431040616415114e-05,
      "loss": 1.5043,
      "num_input_tokens_seen": 55656152,
      "step": 95925
    },
    {
      "epoch": 14.288054810843015,
      "grad_norm": 3.28125,
      "learning_rate": 1.1428311578063566e-05,
      "loss": 0.9416,
      "num_input_tokens_seen": 55659032,
      "step": 95930
    },
    {
      "epoch": 14.288799523383974,
      "grad_norm": 10.3125,
      "learning_rate": 1.142558276899191e-05,
      "loss": 0.8297,
      "num_input_tokens_seen": 55661848,
      "step": 95935
    },
    {
      "epoch": 14.289544235924932,
      "grad_norm": 5.84375,
      "learning_rate": 1.142285418924623e-05,
      "loss": 1.1528,
      "num_input_tokens_seen": 55664568,
      "step": 95940
    },
    {
      "epoch": 14.290288948465893,
      "grad_norm": 6.96875,
      "learning_rate": 1.1420125838872633e-05,
      "loss": 1.0273,
      "num_input_tokens_seen": 55667352,
      "step": 95945
    },
    {
      "epoch": 14.291033661006852,
      "grad_norm": 7.5625,
      "learning_rate": 1.1417397717917213e-05,
      "loss": 0.904,
      "num_input_tokens_seen": 55670328,
      "step": 95950
    },
    {
      "epoch": 14.29177837354781,
      "grad_norm": 6.84375,
      "learning_rate": 1.1414669826426053e-05,
      "loss": 0.8477,
      "num_input_tokens_seen": 55673624,
      "step": 95955
    },
    {
      "epoch": 14.292523086088769,
      "grad_norm": 22.125,
      "learning_rate": 1.1411942164445228e-05,
      "loss": 1.1903,
      "num_input_tokens_seen": 55676408,
      "step": 95960
    },
    {
      "epoch": 14.29326779862973,
      "grad_norm": 8.6875,
      "learning_rate": 1.140921473202084e-05,
      "loss": 1.1707,
      "num_input_tokens_seen": 55679224,
      "step": 95965
    },
    {
      "epoch": 14.294012511170688,
      "grad_norm": 18.125,
      "learning_rate": 1.1406487529198956e-05,
      "loss": 1.4387,
      "num_input_tokens_seen": 55682360,
      "step": 95970
    },
    {
      "epoch": 14.294757223711647,
      "grad_norm": 7.625,
      "learning_rate": 1.1403760556025638e-05,
      "loss": 1.1121,
      "num_input_tokens_seen": 55684952,
      "step": 95975
    },
    {
      "epoch": 14.295501936252606,
      "grad_norm": 8.625,
      "learning_rate": 1.140103381254698e-05,
      "loss": 1.2499,
      "num_input_tokens_seen": 55687800,
      "step": 95980
    },
    {
      "epoch": 14.296246648793566,
      "grad_norm": 8.0625,
      "learning_rate": 1.1398307298809022e-05,
      "loss": 1.5546,
      "num_input_tokens_seen": 55690584,
      "step": 95985
    },
    {
      "epoch": 14.296991361334525,
      "grad_norm": 13.375,
      "learning_rate": 1.1395581014857848e-05,
      "loss": 1.1449,
      "num_input_tokens_seen": 55693880,
      "step": 95990
    },
    {
      "epoch": 14.297736073875484,
      "grad_norm": 12.0625,
      "learning_rate": 1.1392854960739497e-05,
      "loss": 1.0311,
      "num_input_tokens_seen": 55696792,
      "step": 95995
    },
    {
      "epoch": 14.298480786416443,
      "grad_norm": 13.375,
      "learning_rate": 1.1390129136500041e-05,
      "loss": 1.2526,
      "num_input_tokens_seen": 55699416,
      "step": 96000
    },
    {
      "epoch": 14.299225498957403,
      "grad_norm": 7.0625,
      "learning_rate": 1.138740354218552e-05,
      "loss": 1.0163,
      "num_input_tokens_seen": 55702840,
      "step": 96005
    },
    {
      "epoch": 14.299970211498362,
      "grad_norm": 9.1875,
      "learning_rate": 1.1384678177841973e-05,
      "loss": 1.0431,
      "num_input_tokens_seen": 55705656,
      "step": 96010
    },
    {
      "epoch": 14.30071492403932,
      "grad_norm": 13.4375,
      "learning_rate": 1.1381953043515459e-05,
      "loss": 1.0094,
      "num_input_tokens_seen": 55708248,
      "step": 96015
    },
    {
      "epoch": 14.30145963658028,
      "grad_norm": 8.375,
      "learning_rate": 1.1379228139252007e-05,
      "loss": 0.9432,
      "num_input_tokens_seen": 55710968,
      "step": 96020
    },
    {
      "epoch": 14.30220434912124,
      "grad_norm": 10.0,
      "learning_rate": 1.1376503465097651e-05,
      "loss": 1.0904,
      "num_input_tokens_seen": 55713944,
      "step": 96025
    },
    {
      "epoch": 14.302949061662199,
      "grad_norm": 10.5,
      "learning_rate": 1.1373779021098415e-05,
      "loss": 1.1939,
      "num_input_tokens_seen": 55716728,
      "step": 96030
    },
    {
      "epoch": 14.303693774203158,
      "grad_norm": 13.6875,
      "learning_rate": 1.1371054807300344e-05,
      "loss": 0.9291,
      "num_input_tokens_seen": 55719416,
      "step": 96035
    },
    {
      "epoch": 14.304438486744116,
      "grad_norm": 8.125,
      "learning_rate": 1.1368330823749441e-05,
      "loss": 1.3796,
      "num_input_tokens_seen": 55722328,
      "step": 96040
    },
    {
      "epoch": 14.305183199285075,
      "grad_norm": 12.3125,
      "learning_rate": 1.1365607070491741e-05,
      "loss": 1.2602,
      "num_input_tokens_seen": 55725304,
      "step": 96045
    },
    {
      "epoch": 14.305927911826036,
      "grad_norm": 10.5625,
      "learning_rate": 1.1362883547573252e-05,
      "loss": 1.2041,
      "num_input_tokens_seen": 55728088,
      "step": 96050
    },
    {
      "epoch": 14.306672624366994,
      "grad_norm": 8.5625,
      "learning_rate": 1.1360160255039976e-05,
      "loss": 0.7926,
      "num_input_tokens_seen": 55731064,
      "step": 96055
    },
    {
      "epoch": 14.307417336907953,
      "grad_norm": 7.90625,
      "learning_rate": 1.1357437192937943e-05,
      "loss": 1.124,
      "num_input_tokens_seen": 55733464,
      "step": 96060
    },
    {
      "epoch": 14.308162049448912,
      "grad_norm": 13.9375,
      "learning_rate": 1.1354714361313128e-05,
      "loss": 1.3494,
      "num_input_tokens_seen": 55736536,
      "step": 96065
    },
    {
      "epoch": 14.308906761989872,
      "grad_norm": 9.0,
      "learning_rate": 1.1351991760211558e-05,
      "loss": 1.0032,
      "num_input_tokens_seen": 55739480,
      "step": 96070
    },
    {
      "epoch": 14.309651474530831,
      "grad_norm": 6.40625,
      "learning_rate": 1.1349269389679203e-05,
      "loss": 1.0021,
      "num_input_tokens_seen": 55742520,
      "step": 96075
    },
    {
      "epoch": 14.31039618707179,
      "grad_norm": 9.8125,
      "learning_rate": 1.1346547249762082e-05,
      "loss": 1.3648,
      "num_input_tokens_seen": 55745272,
      "step": 96080
    },
    {
      "epoch": 14.311140899612749,
      "grad_norm": 25.375,
      "learning_rate": 1.1343825340506167e-05,
      "loss": 1.4065,
      "num_input_tokens_seen": 55748408,
      "step": 96085
    },
    {
      "epoch": 14.31188561215371,
      "grad_norm": 11.5,
      "learning_rate": 1.1341103661957441e-05,
      "loss": 1.1323,
      "num_input_tokens_seen": 55751448,
      "step": 96090
    },
    {
      "epoch": 14.312630324694668,
      "grad_norm": 4.21875,
      "learning_rate": 1.1338382214161888e-05,
      "loss": 0.8926,
      "num_input_tokens_seen": 55754136,
      "step": 96095
    },
    {
      "epoch": 14.313375037235627,
      "grad_norm": 8.375,
      "learning_rate": 1.1335660997165473e-05,
      "loss": 0.8304,
      "num_input_tokens_seen": 55757048,
      "step": 96100
    },
    {
      "epoch": 14.314119749776586,
      "grad_norm": 11.6875,
      "learning_rate": 1.133294001101419e-05,
      "loss": 0.9701,
      "num_input_tokens_seen": 55760088,
      "step": 96105
    },
    {
      "epoch": 14.314864462317546,
      "grad_norm": 12.75,
      "learning_rate": 1.1330219255753983e-05,
      "loss": 1.2412,
      "num_input_tokens_seen": 55762936,
      "step": 96110
    },
    {
      "epoch": 14.315609174858505,
      "grad_norm": 16.625,
      "learning_rate": 1.1327498731430835e-05,
      "loss": 1.2066,
      "num_input_tokens_seen": 55765976,
      "step": 96115
    },
    {
      "epoch": 14.316353887399464,
      "grad_norm": 6.625,
      "learning_rate": 1.1324778438090694e-05,
      "loss": 0.8891,
      "num_input_tokens_seen": 55768888,
      "step": 96120
    },
    {
      "epoch": 14.317098599940422,
      "grad_norm": 11.5625,
      "learning_rate": 1.132205837577953e-05,
      "loss": 1.284,
      "num_input_tokens_seen": 55771672,
      "step": 96125
    },
    {
      "epoch": 14.317843312481383,
      "grad_norm": 18.75,
      "learning_rate": 1.131933854454329e-05,
      "loss": 1.0548,
      "num_input_tokens_seen": 55774424,
      "step": 96130
    },
    {
      "epoch": 14.318588025022342,
      "grad_norm": 15.125,
      "learning_rate": 1.131661894442791e-05,
      "loss": 1.271,
      "num_input_tokens_seen": 55777240,
      "step": 96135
    },
    {
      "epoch": 14.3193327375633,
      "grad_norm": 4.65625,
      "learning_rate": 1.1313899575479355e-05,
      "loss": 0.9841,
      "num_input_tokens_seen": 55779832,
      "step": 96140
    },
    {
      "epoch": 14.32007745010426,
      "grad_norm": 22.5,
      "learning_rate": 1.1311180437743549e-05,
      "loss": 1.4036,
      "num_input_tokens_seen": 55783064,
      "step": 96145
    },
    {
      "epoch": 14.32082216264522,
      "grad_norm": 15.875,
      "learning_rate": 1.1308461531266442e-05,
      "loss": 1.4244,
      "num_input_tokens_seen": 55785848,
      "step": 96150
    },
    {
      "epoch": 14.321566875186178,
      "grad_norm": 11.125,
      "learning_rate": 1.1305742856093964e-05,
      "loss": 1.0824,
      "num_input_tokens_seen": 55788536,
      "step": 96155
    },
    {
      "epoch": 14.322311587727137,
      "grad_norm": 16.25,
      "learning_rate": 1.1303024412272046e-05,
      "loss": 0.9196,
      "num_input_tokens_seen": 55791448,
      "step": 96160
    },
    {
      "epoch": 14.323056300268096,
      "grad_norm": 20.875,
      "learning_rate": 1.1300306199846605e-05,
      "loss": 0.9814,
      "num_input_tokens_seen": 55794392,
      "step": 96165
    },
    {
      "epoch": 14.323801012809056,
      "grad_norm": 7.375,
      "learning_rate": 1.1297588218863561e-05,
      "loss": 1.197,
      "num_input_tokens_seen": 55797208,
      "step": 96170
    },
    {
      "epoch": 14.324545725350015,
      "grad_norm": 9.4375,
      "learning_rate": 1.1294870469368846e-05,
      "loss": 0.9646,
      "num_input_tokens_seen": 55800088,
      "step": 96175
    },
    {
      "epoch": 14.325290437890974,
      "grad_norm": 8.5,
      "learning_rate": 1.1292152951408356e-05,
      "loss": 0.9966,
      "num_input_tokens_seen": 55802872,
      "step": 96180
    },
    {
      "epoch": 14.326035150431933,
      "grad_norm": 12.5,
      "learning_rate": 1.1289435665028016e-05,
      "loss": 1.2579,
      "num_input_tokens_seen": 55805848,
      "step": 96185
    },
    {
      "epoch": 14.326779862972893,
      "grad_norm": 11.125,
      "learning_rate": 1.1286718610273719e-05,
      "loss": 1.1328,
      "num_input_tokens_seen": 55808568,
      "step": 96190
    },
    {
      "epoch": 14.327524575513852,
      "grad_norm": 6.84375,
      "learning_rate": 1.1284001787191381e-05,
      "loss": 0.9564,
      "num_input_tokens_seen": 55811640,
      "step": 96195
    },
    {
      "epoch": 14.32826928805481,
      "grad_norm": 6.90625,
      "learning_rate": 1.1281285195826884e-05,
      "loss": 1.0509,
      "num_input_tokens_seen": 55814392,
      "step": 96200
    },
    {
      "epoch": 14.32901400059577,
      "grad_norm": 5.875,
      "learning_rate": 1.1278568836226142e-05,
      "loss": 0.9258,
      "num_input_tokens_seen": 55817432,
      "step": 96205
    },
    {
      "epoch": 14.32975871313673,
      "grad_norm": 10.0625,
      "learning_rate": 1.1275852708435033e-05,
      "loss": 0.8847,
      "num_input_tokens_seen": 55820120,
      "step": 96210
    },
    {
      "epoch": 14.330503425677689,
      "grad_norm": 3.40625,
      "learning_rate": 1.127313681249944e-05,
      "loss": 1.0713,
      "num_input_tokens_seen": 55823096,
      "step": 96215
    },
    {
      "epoch": 14.331248138218648,
      "grad_norm": 9.375,
      "learning_rate": 1.1270421148465245e-05,
      "loss": 1.2173,
      "num_input_tokens_seen": 55826104,
      "step": 96220
    },
    {
      "epoch": 14.331992850759606,
      "grad_norm": 12.875,
      "learning_rate": 1.1267705716378338e-05,
      "loss": 1.5507,
      "num_input_tokens_seen": 55829112,
      "step": 96225
    },
    {
      "epoch": 14.332737563300565,
      "grad_norm": 12.6875,
      "learning_rate": 1.1264990516284585e-05,
      "loss": 0.8785,
      "num_input_tokens_seen": 55831864,
      "step": 96230
    },
    {
      "epoch": 14.333482275841526,
      "grad_norm": 8.6875,
      "learning_rate": 1.126227554822985e-05,
      "loss": 1.5409,
      "num_input_tokens_seen": 55834776,
      "step": 96235
    },
    {
      "epoch": 14.334226988382484,
      "grad_norm": 5.875,
      "learning_rate": 1.1259560812260014e-05,
      "loss": 0.8348,
      "num_input_tokens_seen": 55837784,
      "step": 96240
    },
    {
      "epoch": 14.334971700923443,
      "grad_norm": 9.1875,
      "learning_rate": 1.1256846308420935e-05,
      "loss": 1.0632,
      "num_input_tokens_seen": 55840888,
      "step": 96245
    },
    {
      "epoch": 14.335716413464402,
      "grad_norm": 5.15625,
      "learning_rate": 1.125413203675846e-05,
      "loss": 0.9834,
      "num_input_tokens_seen": 55843768,
      "step": 96250
    },
    {
      "epoch": 14.336461126005362,
      "grad_norm": 7.21875,
      "learning_rate": 1.1251417997318464e-05,
      "loss": 1.0814,
      "num_input_tokens_seen": 55846936,
      "step": 96255
    },
    {
      "epoch": 14.337205838546321,
      "grad_norm": 14.375,
      "learning_rate": 1.1248704190146778e-05,
      "loss": 1.1269,
      "num_input_tokens_seen": 55849656,
      "step": 96260
    },
    {
      "epoch": 14.33795055108728,
      "grad_norm": 9.125,
      "learning_rate": 1.1245990615289264e-05,
      "loss": 1.2513,
      "num_input_tokens_seen": 55852440,
      "step": 96265
    },
    {
      "epoch": 14.338695263628239,
      "grad_norm": 4.59375,
      "learning_rate": 1.1243277272791755e-05,
      "loss": 0.9328,
      "num_input_tokens_seen": 55855288,
      "step": 96270
    },
    {
      "epoch": 14.3394399761692,
      "grad_norm": 16.625,
      "learning_rate": 1.1240564162700101e-05,
      "loss": 1.4554,
      "num_input_tokens_seen": 55858200,
      "step": 96275
    },
    {
      "epoch": 14.340184688710158,
      "grad_norm": 10.6875,
      "learning_rate": 1.1237851285060133e-05,
      "loss": 1.265,
      "num_input_tokens_seen": 55861080,
      "step": 96280
    },
    {
      "epoch": 14.340929401251117,
      "grad_norm": 11.75,
      "learning_rate": 1.123513863991768e-05,
      "loss": 1.095,
      "num_input_tokens_seen": 55863896,
      "step": 96285
    },
    {
      "epoch": 14.341674113792076,
      "grad_norm": 9.25,
      "learning_rate": 1.1232426227318568e-05,
      "loss": 0.9654,
      "num_input_tokens_seen": 55867064,
      "step": 96290
    },
    {
      "epoch": 14.342418826333036,
      "grad_norm": 22.875,
      "learning_rate": 1.1229714047308615e-05,
      "loss": 1.0415,
      "num_input_tokens_seen": 55870136,
      "step": 96295
    },
    {
      "epoch": 14.343163538873995,
      "grad_norm": 8.0625,
      "learning_rate": 1.1227002099933657e-05,
      "loss": 0.9569,
      "num_input_tokens_seen": 55873144,
      "step": 96300
    },
    {
      "epoch": 14.343908251414954,
      "grad_norm": 5.96875,
      "learning_rate": 1.1224290385239488e-05,
      "loss": 1.0553,
      "num_input_tokens_seen": 55875896,
      "step": 96305
    },
    {
      "epoch": 14.344652963955912,
      "grad_norm": 7.34375,
      "learning_rate": 1.1221578903271943e-05,
      "loss": 0.9894,
      "num_input_tokens_seen": 55878648,
      "step": 96310
    },
    {
      "epoch": 14.345397676496873,
      "grad_norm": 27.625,
      "learning_rate": 1.1218867654076812e-05,
      "loss": 1.1031,
      "num_input_tokens_seen": 55881432,
      "step": 96315
    },
    {
      "epoch": 14.346142389037832,
      "grad_norm": 11.875,
      "learning_rate": 1.1216156637699909e-05,
      "loss": 1.1742,
      "num_input_tokens_seen": 55884248,
      "step": 96320
    },
    {
      "epoch": 14.34688710157879,
      "grad_norm": 9.6875,
      "learning_rate": 1.1213445854187035e-05,
      "loss": 1.1728,
      "num_input_tokens_seen": 55887096,
      "step": 96325
    },
    {
      "epoch": 14.34763181411975,
      "grad_norm": 6.5625,
      "learning_rate": 1.1210735303583972e-05,
      "loss": 1.1359,
      "num_input_tokens_seen": 55889912,
      "step": 96330
    },
    {
      "epoch": 14.34837652666071,
      "grad_norm": 12.75,
      "learning_rate": 1.1208024985936527e-05,
      "loss": 1.1167,
      "num_input_tokens_seen": 55893080,
      "step": 96335
    },
    {
      "epoch": 14.349121239201668,
      "grad_norm": 11.875,
      "learning_rate": 1.1205314901290475e-05,
      "loss": 1.0936,
      "num_input_tokens_seen": 55895928,
      "step": 96340
    },
    {
      "epoch": 14.349865951742627,
      "grad_norm": 5.9375,
      "learning_rate": 1.120260504969162e-05,
      "loss": 0.7827,
      "num_input_tokens_seen": 55898904,
      "step": 96345
    },
    {
      "epoch": 14.350610664283586,
      "grad_norm": 6.09375,
      "learning_rate": 1.1199895431185726e-05,
      "loss": 1.3937,
      "num_input_tokens_seen": 55901816,
      "step": 96350
    },
    {
      "epoch": 14.351355376824547,
      "grad_norm": 6.84375,
      "learning_rate": 1.1197186045818572e-05,
      "loss": 1.1686,
      "num_input_tokens_seen": 55904568,
      "step": 96355
    },
    {
      "epoch": 14.352100089365505,
      "grad_norm": 12.8125,
      "learning_rate": 1.1194476893635924e-05,
      "loss": 1.1308,
      "num_input_tokens_seen": 55907448,
      "step": 96360
    },
    {
      "epoch": 14.352844801906464,
      "grad_norm": 17.875,
      "learning_rate": 1.1191767974683567e-05,
      "loss": 1.1916,
      "num_input_tokens_seen": 55910392,
      "step": 96365
    },
    {
      "epoch": 14.353589514447423,
      "grad_norm": 19.125,
      "learning_rate": 1.1189059289007256e-05,
      "loss": 1.0141,
      "num_input_tokens_seen": 55913464,
      "step": 96370
    },
    {
      "epoch": 14.354334226988382,
      "grad_norm": 6.15625,
      "learning_rate": 1.118635083665274e-05,
      "loss": 1.0691,
      "num_input_tokens_seen": 55916280,
      "step": 96375
    },
    {
      "epoch": 14.355078939529342,
      "grad_norm": 17.375,
      "learning_rate": 1.1183642617665799e-05,
      "loss": 1.3654,
      "num_input_tokens_seen": 55919032,
      "step": 96380
    },
    {
      "epoch": 14.3558236520703,
      "grad_norm": 11.75,
      "learning_rate": 1.1180934632092163e-05,
      "loss": 0.9868,
      "num_input_tokens_seen": 55921944,
      "step": 96385
    },
    {
      "epoch": 14.35656836461126,
      "grad_norm": 13.0,
      "learning_rate": 1.11782268799776e-05,
      "loss": 0.9435,
      "num_input_tokens_seen": 55924856,
      "step": 96390
    },
    {
      "epoch": 14.357313077152218,
      "grad_norm": 8.25,
      "learning_rate": 1.1175519361367837e-05,
      "loss": 0.8496,
      "num_input_tokens_seen": 55927704,
      "step": 96395
    },
    {
      "epoch": 14.358057789693179,
      "grad_norm": 10.0,
      "learning_rate": 1.1172812076308634e-05,
      "loss": 1.2338,
      "num_input_tokens_seen": 55930456,
      "step": 96400
    },
    {
      "epoch": 14.358802502234138,
      "grad_norm": 14.1875,
      "learning_rate": 1.1170105024845718e-05,
      "loss": 0.9374,
      "num_input_tokens_seen": 55933400,
      "step": 96405
    },
    {
      "epoch": 14.359547214775096,
      "grad_norm": 20.0,
      "learning_rate": 1.1167398207024812e-05,
      "loss": 0.8992,
      "num_input_tokens_seen": 55936152,
      "step": 96410
    },
    {
      "epoch": 14.360291927316055,
      "grad_norm": 10.0625,
      "learning_rate": 1.1164691622891662e-05,
      "loss": 0.9949,
      "num_input_tokens_seen": 55939256,
      "step": 96415
    },
    {
      "epoch": 14.361036639857016,
      "grad_norm": 14.1875,
      "learning_rate": 1.1161985272491986e-05,
      "loss": 1.2724,
      "num_input_tokens_seen": 55941912,
      "step": 96420
    },
    {
      "epoch": 14.361781352397974,
      "grad_norm": 13.625,
      "learning_rate": 1.1159279155871507e-05,
      "loss": 0.9021,
      "num_input_tokens_seen": 55944856,
      "step": 96425
    },
    {
      "epoch": 14.362526064938933,
      "grad_norm": 12.875,
      "learning_rate": 1.115657327307593e-05,
      "loss": 1.2133,
      "num_input_tokens_seen": 55947704,
      "step": 96430
    },
    {
      "epoch": 14.363270777479892,
      "grad_norm": 11.5625,
      "learning_rate": 1.1153867624150986e-05,
      "loss": 0.966,
      "num_input_tokens_seen": 55950808,
      "step": 96435
    },
    {
      "epoch": 14.364015490020853,
      "grad_norm": 11.3125,
      "learning_rate": 1.1151162209142362e-05,
      "loss": 1.1878,
      "num_input_tokens_seen": 55953848,
      "step": 96440
    },
    {
      "epoch": 14.364760202561811,
      "grad_norm": 18.875,
      "learning_rate": 1.1148457028095794e-05,
      "loss": 1.3825,
      "num_input_tokens_seen": 55957016,
      "step": 96445
    },
    {
      "epoch": 14.36550491510277,
      "grad_norm": 4.9375,
      "learning_rate": 1.1145752081056961e-05,
      "loss": 0.7929,
      "num_input_tokens_seen": 55960024,
      "step": 96450
    },
    {
      "epoch": 14.366249627643729,
      "grad_norm": 11.5,
      "learning_rate": 1.114304736807156e-05,
      "loss": 1.2248,
      "num_input_tokens_seen": 55963224,
      "step": 96455
    },
    {
      "epoch": 14.36699434018469,
      "grad_norm": 6.0,
      "learning_rate": 1.1140342889185299e-05,
      "loss": 1.5025,
      "num_input_tokens_seen": 55966072,
      "step": 96460
    },
    {
      "epoch": 14.367739052725648,
      "grad_norm": 7.90625,
      "learning_rate": 1.1137638644443846e-05,
      "loss": 1.0859,
      "num_input_tokens_seen": 55969016,
      "step": 96465
    },
    {
      "epoch": 14.368483765266607,
      "grad_norm": 15.0625,
      "learning_rate": 1.113493463389291e-05,
      "loss": 1.1819,
      "num_input_tokens_seen": 55971896,
      "step": 96470
    },
    {
      "epoch": 14.369228477807566,
      "grad_norm": 11.25,
      "learning_rate": 1.1132230857578155e-05,
      "loss": 0.9955,
      "num_input_tokens_seen": 55974936,
      "step": 96475
    },
    {
      "epoch": 14.369973190348526,
      "grad_norm": 5.9375,
      "learning_rate": 1.1129527315545272e-05,
      "loss": 1.064,
      "num_input_tokens_seen": 55977752,
      "step": 96480
    },
    {
      "epoch": 14.370717902889485,
      "grad_norm": 8.3125,
      "learning_rate": 1.1126824007839927e-05,
      "loss": 1.0396,
      "num_input_tokens_seen": 55980440,
      "step": 96485
    },
    {
      "epoch": 14.371462615430444,
      "grad_norm": 9.1875,
      "learning_rate": 1.1124120934507792e-05,
      "loss": 1.1257,
      "num_input_tokens_seen": 55983160,
      "step": 96490
    },
    {
      "epoch": 14.372207327971402,
      "grad_norm": 14.0,
      "learning_rate": 1.112141809559453e-05,
      "loss": 1.0007,
      "num_input_tokens_seen": 55986136,
      "step": 96495
    },
    {
      "epoch": 14.372952040512363,
      "grad_norm": 10.4375,
      "learning_rate": 1.1118715491145795e-05,
      "loss": 1.2238,
      "num_input_tokens_seen": 55988728,
      "step": 96500
    },
    {
      "epoch": 14.373696753053322,
      "grad_norm": 9.8125,
      "learning_rate": 1.1116013121207261e-05,
      "loss": 1.1871,
      "num_input_tokens_seen": 55991512,
      "step": 96505
    },
    {
      "epoch": 14.37444146559428,
      "grad_norm": 5.84375,
      "learning_rate": 1.1113310985824566e-05,
      "loss": 0.8804,
      "num_input_tokens_seen": 55994328,
      "step": 96510
    },
    {
      "epoch": 14.37518617813524,
      "grad_norm": 8.75,
      "learning_rate": 1.1110609085043378e-05,
      "loss": 1.0307,
      "num_input_tokens_seen": 55997464,
      "step": 96515
    },
    {
      "epoch": 14.3759308906762,
      "grad_norm": 10.4375,
      "learning_rate": 1.1107907418909324e-05,
      "loss": 1.0865,
      "num_input_tokens_seen": 56000184,
      "step": 96520
    },
    {
      "epoch": 14.376675603217159,
      "grad_norm": 9.875,
      "learning_rate": 1.1105205987468064e-05,
      "loss": 1.2193,
      "num_input_tokens_seen": 56003096,
      "step": 96525
    },
    {
      "epoch": 14.377420315758117,
      "grad_norm": 9.6875,
      "learning_rate": 1.1102504790765225e-05,
      "loss": 1.0136,
      "num_input_tokens_seen": 56006008,
      "step": 96530
    },
    {
      "epoch": 14.378165028299076,
      "grad_norm": 6.25,
      "learning_rate": 1.1099803828846437e-05,
      "loss": 1.2065,
      "num_input_tokens_seen": 56008952,
      "step": 96535
    },
    {
      "epoch": 14.378909740840037,
      "grad_norm": 11.6875,
      "learning_rate": 1.1097103101757342e-05,
      "loss": 1.0378,
      "num_input_tokens_seen": 56011672,
      "step": 96540
    },
    {
      "epoch": 14.379654453380995,
      "grad_norm": 5.25,
      "learning_rate": 1.1094402609543561e-05,
      "loss": 0.9055,
      "num_input_tokens_seen": 56014456,
      "step": 96545
    },
    {
      "epoch": 14.380399165921954,
      "grad_norm": 14.5625,
      "learning_rate": 1.1091702352250704e-05,
      "loss": 1.1484,
      "num_input_tokens_seen": 56017464,
      "step": 96550
    },
    {
      "epoch": 14.381143878462913,
      "grad_norm": 19.5,
      "learning_rate": 1.108900232992441e-05,
      "loss": 1.0938,
      "num_input_tokens_seen": 56020440,
      "step": 96555
    },
    {
      "epoch": 14.381888591003872,
      "grad_norm": 16.875,
      "learning_rate": 1.1086302542610285e-05,
      "loss": 1.137,
      "num_input_tokens_seen": 56023480,
      "step": 96560
    },
    {
      "epoch": 14.382633303544832,
      "grad_norm": 11.9375,
      "learning_rate": 1.1083602990353928e-05,
      "loss": 0.853,
      "num_input_tokens_seen": 56026264,
      "step": 96565
    },
    {
      "epoch": 14.383378016085791,
      "grad_norm": 3.9375,
      "learning_rate": 1.1080903673200962e-05,
      "loss": 0.8066,
      "num_input_tokens_seen": 56029208,
      "step": 96570
    },
    {
      "epoch": 14.38412272862675,
      "grad_norm": 7.1875,
      "learning_rate": 1.107820459119698e-05,
      "loss": 1.0522,
      "num_input_tokens_seen": 56032056,
      "step": 96575
    },
    {
      "epoch": 14.384867441167708,
      "grad_norm": 12.875,
      "learning_rate": 1.1075505744387577e-05,
      "loss": 0.806,
      "num_input_tokens_seen": 56035000,
      "step": 96580
    },
    {
      "epoch": 14.385612153708669,
      "grad_norm": 12.125,
      "learning_rate": 1.1072807132818358e-05,
      "loss": 1.0505,
      "num_input_tokens_seen": 56037848,
      "step": 96585
    },
    {
      "epoch": 14.386356866249628,
      "grad_norm": 8.8125,
      "learning_rate": 1.10701087565349e-05,
      "loss": 0.8485,
      "num_input_tokens_seen": 56040536,
      "step": 96590
    },
    {
      "epoch": 14.387101578790586,
      "grad_norm": 13.5,
      "learning_rate": 1.1067410615582808e-05,
      "loss": 1.1264,
      "num_input_tokens_seen": 56043512,
      "step": 96595
    },
    {
      "epoch": 14.387846291331545,
      "grad_norm": 10.125,
      "learning_rate": 1.106471271000764e-05,
      "loss": 1.0487,
      "num_input_tokens_seen": 56046360,
      "step": 96600
    },
    {
      "epoch": 14.388591003872506,
      "grad_norm": 7.15625,
      "learning_rate": 1.1062015039854997e-05,
      "loss": 1.1892,
      "num_input_tokens_seen": 56049240,
      "step": 96605
    },
    {
      "epoch": 14.389335716413465,
      "grad_norm": 10.0625,
      "learning_rate": 1.1059317605170447e-05,
      "loss": 1.2603,
      "num_input_tokens_seen": 56052120,
      "step": 96610
    },
    {
      "epoch": 14.390080428954423,
      "grad_norm": 11.1875,
      "learning_rate": 1.1056620405999558e-05,
      "loss": 1.0813,
      "num_input_tokens_seen": 56055096,
      "step": 96615
    },
    {
      "epoch": 14.390825141495382,
      "grad_norm": 14.5625,
      "learning_rate": 1.1053923442387892e-05,
      "loss": 1.2232,
      "num_input_tokens_seen": 56058232,
      "step": 96620
    },
    {
      "epoch": 14.391569854036343,
      "grad_norm": 9.125,
      "learning_rate": 1.1051226714381008e-05,
      "loss": 1.4232,
      "num_input_tokens_seen": 56061080,
      "step": 96625
    },
    {
      "epoch": 14.392314566577301,
      "grad_norm": 12.125,
      "learning_rate": 1.1048530222024481e-05,
      "loss": 1.1779,
      "num_input_tokens_seen": 56064120,
      "step": 96630
    },
    {
      "epoch": 14.39305927911826,
      "grad_norm": 10.5625,
      "learning_rate": 1.1045833965363847e-05,
      "loss": 1.0826,
      "num_input_tokens_seen": 56067032,
      "step": 96635
    },
    {
      "epoch": 14.393803991659219,
      "grad_norm": 9.9375,
      "learning_rate": 1.1043137944444673e-05,
      "loss": 1.0557,
      "num_input_tokens_seen": 56070200,
      "step": 96640
    },
    {
      "epoch": 14.39454870420018,
      "grad_norm": 13.5,
      "learning_rate": 1.1040442159312491e-05,
      "loss": 1.1542,
      "num_input_tokens_seen": 56073112,
      "step": 96645
    },
    {
      "epoch": 14.395293416741138,
      "grad_norm": 22.375,
      "learning_rate": 1.1037746610012861e-05,
      "loss": 1.3721,
      "num_input_tokens_seen": 56075768,
      "step": 96650
    },
    {
      "epoch": 14.396038129282097,
      "grad_norm": 12.3125,
      "learning_rate": 1.1035051296591309e-05,
      "loss": 1.1626,
      "num_input_tokens_seen": 56078776,
      "step": 96655
    },
    {
      "epoch": 14.396782841823056,
      "grad_norm": 8.0,
      "learning_rate": 1.1032356219093365e-05,
      "loss": 0.9418,
      "num_input_tokens_seen": 56081592,
      "step": 96660
    },
    {
      "epoch": 14.397527554364016,
      "grad_norm": 7.1875,
      "learning_rate": 1.1029661377564576e-05,
      "loss": 1.0489,
      "num_input_tokens_seen": 56084600,
      "step": 96665
    },
    {
      "epoch": 14.398272266904975,
      "grad_norm": 7.34375,
      "learning_rate": 1.1026966772050448e-05,
      "loss": 1.3545,
      "num_input_tokens_seen": 56087416,
      "step": 96670
    },
    {
      "epoch": 14.399016979445934,
      "grad_norm": 13.9375,
      "learning_rate": 1.1024272402596526e-05,
      "loss": 1.1135,
      "num_input_tokens_seen": 56090168,
      "step": 96675
    },
    {
      "epoch": 14.399761691986892,
      "grad_norm": 11.0,
      "learning_rate": 1.1021578269248314e-05,
      "loss": 0.7834,
      "num_input_tokens_seen": 56092952,
      "step": 96680
    },
    {
      "epoch": 14.400506404527853,
      "grad_norm": 11.625,
      "learning_rate": 1.1018884372051333e-05,
      "loss": 1.1515,
      "num_input_tokens_seen": 56095800,
      "step": 96685
    },
    {
      "epoch": 14.401251117068812,
      "grad_norm": 7.1875,
      "learning_rate": 1.1016190711051092e-05,
      "loss": 1.1086,
      "num_input_tokens_seen": 56098712,
      "step": 96690
    },
    {
      "epoch": 14.40199582960977,
      "grad_norm": 4.78125,
      "learning_rate": 1.1013497286293085e-05,
      "loss": 0.6427,
      "num_input_tokens_seen": 56101624,
      "step": 96695
    },
    {
      "epoch": 14.40274054215073,
      "grad_norm": 17.75,
      "learning_rate": 1.1010804097822836e-05,
      "loss": 1.1891,
      "num_input_tokens_seen": 56104600,
      "step": 96700
    },
    {
      "epoch": 14.40348525469169,
      "grad_norm": 4.5625,
      "learning_rate": 1.1008111145685824e-05,
      "loss": 1.1204,
      "num_input_tokens_seen": 56107416,
      "step": 96705
    },
    {
      "epoch": 14.404229967232649,
      "grad_norm": 17.375,
      "learning_rate": 1.1005418429927563e-05,
      "loss": 1.1473,
      "num_input_tokens_seen": 56110392,
      "step": 96710
    },
    {
      "epoch": 14.404974679773607,
      "grad_norm": 7.8125,
      "learning_rate": 1.1002725950593525e-05,
      "loss": 1.1073,
      "num_input_tokens_seen": 56113112,
      "step": 96715
    },
    {
      "epoch": 14.405719392314566,
      "grad_norm": 11.3125,
      "learning_rate": 1.1000033707729216e-05,
      "loss": 0.8908,
      "num_input_tokens_seen": 56116216,
      "step": 96720
    },
    {
      "epoch": 14.406464104855527,
      "grad_norm": 6.125,
      "learning_rate": 1.0997341701380099e-05,
      "loss": 0.9156,
      "num_input_tokens_seen": 56119128,
      "step": 96725
    },
    {
      "epoch": 14.407208817396485,
      "grad_norm": 13.5625,
      "learning_rate": 1.0994649931591669e-05,
      "loss": 0.7049,
      "num_input_tokens_seen": 56121848,
      "step": 96730
    },
    {
      "epoch": 14.407953529937444,
      "grad_norm": 12.1875,
      "learning_rate": 1.0991958398409396e-05,
      "loss": 1.3106,
      "num_input_tokens_seen": 56124856,
      "step": 96735
    },
    {
      "epoch": 14.408698242478403,
      "grad_norm": 6.6875,
      "learning_rate": 1.0989267101878742e-05,
      "loss": 1.2886,
      "num_input_tokens_seen": 56127576,
      "step": 96740
    },
    {
      "epoch": 14.409442955019362,
      "grad_norm": 11.625,
      "learning_rate": 1.0986576042045186e-05,
      "loss": 1.064,
      "num_input_tokens_seen": 56130456,
      "step": 96745
    },
    {
      "epoch": 14.410187667560322,
      "grad_norm": 9.4375,
      "learning_rate": 1.0983885218954187e-05,
      "loss": 1.1288,
      "num_input_tokens_seen": 56133400,
      "step": 96750
    },
    {
      "epoch": 14.410932380101281,
      "grad_norm": 9.4375,
      "learning_rate": 1.0981194632651201e-05,
      "loss": 1.4245,
      "num_input_tokens_seen": 56136280,
      "step": 96755
    },
    {
      "epoch": 14.41167709264224,
      "grad_norm": 6.53125,
      "learning_rate": 1.0978504283181674e-05,
      "loss": 1.0727,
      "num_input_tokens_seen": 56139352,
      "step": 96760
    },
    {
      "epoch": 14.412421805183198,
      "grad_norm": 6.875,
      "learning_rate": 1.0975814170591076e-05,
      "loss": 0.8852,
      "num_input_tokens_seen": 56142520,
      "step": 96765
    },
    {
      "epoch": 14.413166517724159,
      "grad_norm": 6.125,
      "learning_rate": 1.0973124294924843e-05,
      "loss": 1.175,
      "num_input_tokens_seen": 56145208,
      "step": 96770
    },
    {
      "epoch": 14.413911230265118,
      "grad_norm": 5.375,
      "learning_rate": 1.0970434656228412e-05,
      "loss": 0.9728,
      "num_input_tokens_seen": 56148440,
      "step": 96775
    },
    {
      "epoch": 14.414655942806077,
      "grad_norm": 9.4375,
      "learning_rate": 1.0967745254547238e-05,
      "loss": 1.2203,
      "num_input_tokens_seen": 56151320,
      "step": 96780
    },
    {
      "epoch": 14.415400655347035,
      "grad_norm": 5.8125,
      "learning_rate": 1.0965056089926734e-05,
      "loss": 1.0005,
      "num_input_tokens_seen": 56154072,
      "step": 96785
    },
    {
      "epoch": 14.416145367887996,
      "grad_norm": 11.9375,
      "learning_rate": 1.0962367162412354e-05,
      "loss": 1.4033,
      "num_input_tokens_seen": 56156952,
      "step": 96790
    },
    {
      "epoch": 14.416890080428955,
      "grad_norm": 4.25,
      "learning_rate": 1.0959678472049502e-05,
      "loss": 1.0807,
      "num_input_tokens_seen": 56159480,
      "step": 96795
    },
    {
      "epoch": 14.417634792969913,
      "grad_norm": 6.875,
      "learning_rate": 1.0956990018883625e-05,
      "loss": 1.2507,
      "num_input_tokens_seen": 56162456,
      "step": 96800
    },
    {
      "epoch": 14.418379505510872,
      "grad_norm": 9.6875,
      "learning_rate": 1.0954301802960118e-05,
      "loss": 1.1724,
      "num_input_tokens_seen": 56165368,
      "step": 96805
    },
    {
      "epoch": 14.419124218051833,
      "grad_norm": 6.3125,
      "learning_rate": 1.0951613824324417e-05,
      "loss": 1.0117,
      "num_input_tokens_seen": 56168408,
      "step": 96810
    },
    {
      "epoch": 14.419868930592791,
      "grad_norm": 14.875,
      "learning_rate": 1.0948926083021921e-05,
      "loss": 1.2778,
      "num_input_tokens_seen": 56171224,
      "step": 96815
    },
    {
      "epoch": 14.42061364313375,
      "grad_norm": 11.8125,
      "learning_rate": 1.0946238579098036e-05,
      "loss": 0.998,
      "num_input_tokens_seen": 56173912,
      "step": 96820
    },
    {
      "epoch": 14.421358355674709,
      "grad_norm": 10.1875,
      "learning_rate": 1.0943551312598172e-05,
      "loss": 0.8422,
      "num_input_tokens_seen": 56176504,
      "step": 96825
    },
    {
      "epoch": 14.42210306821567,
      "grad_norm": 14.25,
      "learning_rate": 1.0940864283567708e-05,
      "loss": 1.2007,
      "num_input_tokens_seen": 56179320,
      "step": 96830
    },
    {
      "epoch": 14.422847780756628,
      "grad_norm": 6.875,
      "learning_rate": 1.0938177492052064e-05,
      "loss": 1.0778,
      "num_input_tokens_seen": 56182296,
      "step": 96835
    },
    {
      "epoch": 14.423592493297587,
      "grad_norm": 9.125,
      "learning_rate": 1.093549093809661e-05,
      "loss": 1.3974,
      "num_input_tokens_seen": 56185208,
      "step": 96840
    },
    {
      "epoch": 14.424337205838546,
      "grad_norm": 19.875,
      "learning_rate": 1.0932804621746751e-05,
      "loss": 1.3519,
      "num_input_tokens_seen": 56188024,
      "step": 96845
    },
    {
      "epoch": 14.425081918379506,
      "grad_norm": 17.125,
      "learning_rate": 1.0930118543047862e-05,
      "loss": 1.118,
      "num_input_tokens_seen": 56190968,
      "step": 96850
    },
    {
      "epoch": 14.425826630920465,
      "grad_norm": 9.5625,
      "learning_rate": 1.0927432702045309e-05,
      "loss": 1.1875,
      "num_input_tokens_seen": 56193688,
      "step": 96855
    },
    {
      "epoch": 14.426571343461424,
      "grad_norm": 11.75,
      "learning_rate": 1.0924747098784488e-05,
      "loss": 1.0175,
      "num_input_tokens_seen": 56196536,
      "step": 96860
    },
    {
      "epoch": 14.427316056002383,
      "grad_norm": 10.25,
      "learning_rate": 1.0922061733310751e-05,
      "loss": 1.1044,
      "num_input_tokens_seen": 56199640,
      "step": 96865
    },
    {
      "epoch": 14.428060768543343,
      "grad_norm": 9.125,
      "learning_rate": 1.0919376605669481e-05,
      "loss": 0.9622,
      "num_input_tokens_seen": 56202456,
      "step": 96870
    },
    {
      "epoch": 14.428805481084302,
      "grad_norm": 12.125,
      "learning_rate": 1.0916691715906034e-05,
      "loss": 1.4974,
      "num_input_tokens_seen": 56205336,
      "step": 96875
    },
    {
      "epoch": 14.42955019362526,
      "grad_norm": 8.6875,
      "learning_rate": 1.0914007064065754e-05,
      "loss": 1.0053,
      "num_input_tokens_seen": 56208184,
      "step": 96880
    },
    {
      "epoch": 14.43029490616622,
      "grad_norm": 6.84375,
      "learning_rate": 1.091132265019402e-05,
      "loss": 1.1126,
      "num_input_tokens_seen": 56211096,
      "step": 96885
    },
    {
      "epoch": 14.43103961870718,
      "grad_norm": 10.3125,
      "learning_rate": 1.0908638474336172e-05,
      "loss": 0.8603,
      "num_input_tokens_seen": 56213624,
      "step": 96890
    },
    {
      "epoch": 14.431784331248139,
      "grad_norm": 7.28125,
      "learning_rate": 1.0905954536537551e-05,
      "loss": 1.1262,
      "num_input_tokens_seen": 56216536,
      "step": 96895
    },
    {
      "epoch": 14.432529043789097,
      "grad_norm": 11.125,
      "learning_rate": 1.0903270836843499e-05,
      "loss": 1.3744,
      "num_input_tokens_seen": 56219064,
      "step": 96900
    },
    {
      "epoch": 14.433273756330056,
      "grad_norm": 9.8125,
      "learning_rate": 1.0900587375299365e-05,
      "loss": 1.0455,
      "num_input_tokens_seen": 56221944,
      "step": 96905
    },
    {
      "epoch": 14.434018468871017,
      "grad_norm": 25.875,
      "learning_rate": 1.0897904151950469e-05,
      "loss": 1.3246,
      "num_input_tokens_seen": 56224952,
      "step": 96910
    },
    {
      "epoch": 14.434763181411975,
      "grad_norm": 9.8125,
      "learning_rate": 1.089522116684216e-05,
      "loss": 1.006,
      "num_input_tokens_seen": 56228056,
      "step": 96915
    },
    {
      "epoch": 14.435507893952934,
      "grad_norm": 7.3125,
      "learning_rate": 1.0892538420019744e-05,
      "loss": 1.0028,
      "num_input_tokens_seen": 56230904,
      "step": 96920
    },
    {
      "epoch": 14.436252606493893,
      "grad_norm": 7.4375,
      "learning_rate": 1.0889855911528562e-05,
      "loss": 1.1264,
      "num_input_tokens_seen": 56233816,
      "step": 96925
    },
    {
      "epoch": 14.436997319034852,
      "grad_norm": 5.0,
      "learning_rate": 1.0887173641413923e-05,
      "loss": 0.9837,
      "num_input_tokens_seen": 56236984,
      "step": 96930
    },
    {
      "epoch": 14.437742031575812,
      "grad_norm": 6.65625,
      "learning_rate": 1.0884491609721133e-05,
      "loss": 0.8833,
      "num_input_tokens_seen": 56239928,
      "step": 96935
    },
    {
      "epoch": 14.438486744116771,
      "grad_norm": 12.5,
      "learning_rate": 1.088180981649552e-05,
      "loss": 1.033,
      "num_input_tokens_seen": 56242712,
      "step": 96940
    },
    {
      "epoch": 14.43923145665773,
      "grad_norm": 6.96875,
      "learning_rate": 1.0879128261782382e-05,
      "loss": 1.2208,
      "num_input_tokens_seen": 56245560,
      "step": 96945
    },
    {
      "epoch": 14.439976169198689,
      "grad_norm": 9.75,
      "learning_rate": 1.0876446945627019e-05,
      "loss": 1.2038,
      "num_input_tokens_seen": 56248440,
      "step": 96950
    },
    {
      "epoch": 14.440720881739649,
      "grad_norm": 12.25,
      "learning_rate": 1.0873765868074723e-05,
      "loss": 0.9839,
      "num_input_tokens_seen": 56251224,
      "step": 96955
    },
    {
      "epoch": 14.441465594280608,
      "grad_norm": 7.75,
      "learning_rate": 1.0871085029170802e-05,
      "loss": 0.9968,
      "num_input_tokens_seen": 56254296,
      "step": 96960
    },
    {
      "epoch": 14.442210306821567,
      "grad_norm": 6.53125,
      "learning_rate": 1.0868404428960532e-05,
      "loss": 1.0566,
      "num_input_tokens_seen": 56257112,
      "step": 96965
    },
    {
      "epoch": 14.442955019362525,
      "grad_norm": 6.15625,
      "learning_rate": 1.0865724067489214e-05,
      "loss": 1.0685,
      "num_input_tokens_seen": 56259896,
      "step": 96970
    },
    {
      "epoch": 14.443699731903486,
      "grad_norm": 8.75,
      "learning_rate": 1.0863043944802123e-05,
      "loss": 1.4276,
      "num_input_tokens_seen": 56262616,
      "step": 96975
    },
    {
      "epoch": 14.444444444444445,
      "grad_norm": 8.75,
      "learning_rate": 1.0860364060944527e-05,
      "loss": 1.0178,
      "num_input_tokens_seen": 56265592,
      "step": 96980
    },
    {
      "epoch": 14.445189156985403,
      "grad_norm": 9.5625,
      "learning_rate": 1.0857684415961721e-05,
      "loss": 0.8922,
      "num_input_tokens_seen": 56268920,
      "step": 96985
    },
    {
      "epoch": 14.445933869526362,
      "grad_norm": 15.1875,
      "learning_rate": 1.0855005009898953e-05,
      "loss": 1.3122,
      "num_input_tokens_seen": 56271896,
      "step": 96990
    },
    {
      "epoch": 14.446678582067323,
      "grad_norm": 9.125,
      "learning_rate": 1.0852325842801506e-05,
      "loss": 1.0544,
      "num_input_tokens_seen": 56274680,
      "step": 96995
    },
    {
      "epoch": 14.447423294608281,
      "grad_norm": 10.25,
      "learning_rate": 1.0849646914714628e-05,
      "loss": 0.8579,
      "num_input_tokens_seen": 56277432,
      "step": 97000
    },
    {
      "epoch": 14.44816800714924,
      "grad_norm": 9.4375,
      "learning_rate": 1.0846968225683591e-05,
      "loss": 1.163,
      "num_input_tokens_seen": 56280600,
      "step": 97005
    },
    {
      "epoch": 14.448912719690199,
      "grad_norm": 12.25,
      "learning_rate": 1.0844289775753645e-05,
      "loss": 1.1392,
      "num_input_tokens_seen": 56283384,
      "step": 97010
    },
    {
      "epoch": 14.44965743223116,
      "grad_norm": 12.5,
      "learning_rate": 1.084161156497003e-05,
      "loss": 1.1281,
      "num_input_tokens_seen": 56286232,
      "step": 97015
    },
    {
      "epoch": 14.450402144772118,
      "grad_norm": 13.375,
      "learning_rate": 1.0838933593378e-05,
      "loss": 1.1468,
      "num_input_tokens_seen": 56288824,
      "step": 97020
    },
    {
      "epoch": 14.451146857313077,
      "grad_norm": 8.875,
      "learning_rate": 1.0836255861022788e-05,
      "loss": 1.1686,
      "num_input_tokens_seen": 56291832,
      "step": 97025
    },
    {
      "epoch": 14.451891569854036,
      "grad_norm": 15.0,
      "learning_rate": 1.0833578367949646e-05,
      "loss": 1.2173,
      "num_input_tokens_seen": 56295128,
      "step": 97030
    },
    {
      "epoch": 14.452636282394996,
      "grad_norm": 14.3125,
      "learning_rate": 1.0830901114203786e-05,
      "loss": 0.7959,
      "num_input_tokens_seen": 56297880,
      "step": 97035
    },
    {
      "epoch": 14.453380994935955,
      "grad_norm": 18.5,
      "learning_rate": 1.0828224099830464e-05,
      "loss": 1.3247,
      "num_input_tokens_seen": 56300472,
      "step": 97040
    },
    {
      "epoch": 14.454125707476914,
      "grad_norm": 6.09375,
      "learning_rate": 1.0825547324874883e-05,
      "loss": 0.9718,
      "num_input_tokens_seen": 56303512,
      "step": 97045
    },
    {
      "epoch": 14.454870420017873,
      "grad_norm": 5.25,
      "learning_rate": 1.0822870789382283e-05,
      "loss": 0.9087,
      "num_input_tokens_seen": 56306456,
      "step": 97050
    },
    {
      "epoch": 14.455615132558833,
      "grad_norm": 12.125,
      "learning_rate": 1.082019449339787e-05,
      "loss": 0.9078,
      "num_input_tokens_seen": 56309560,
      "step": 97055
    },
    {
      "epoch": 14.456359845099792,
      "grad_norm": 8.125,
      "learning_rate": 1.0817518436966852e-05,
      "loss": 1.2844,
      "num_input_tokens_seen": 56312344,
      "step": 97060
    },
    {
      "epoch": 14.45710455764075,
      "grad_norm": 8.125,
      "learning_rate": 1.0814842620134456e-05,
      "loss": 0.8999,
      "num_input_tokens_seen": 56315160,
      "step": 97065
    },
    {
      "epoch": 14.45784927018171,
      "grad_norm": 10.6875,
      "learning_rate": 1.0812167042945864e-05,
      "loss": 0.9683,
      "num_input_tokens_seen": 56317944,
      "step": 97070
    },
    {
      "epoch": 14.458593982722668,
      "grad_norm": 7.25,
      "learning_rate": 1.08094917054463e-05,
      "loss": 1.0253,
      "num_input_tokens_seen": 56320856,
      "step": 97075
    },
    {
      "epoch": 14.459338695263629,
      "grad_norm": 8.0,
      "learning_rate": 1.0806816607680954e-05,
      "loss": 1.1196,
      "num_input_tokens_seen": 56323896,
      "step": 97080
    },
    {
      "epoch": 14.460083407804587,
      "grad_norm": 7.75,
      "learning_rate": 1.0804141749695012e-05,
      "loss": 1.2461,
      "num_input_tokens_seen": 56326808,
      "step": 97085
    },
    {
      "epoch": 14.460828120345546,
      "grad_norm": 7.0,
      "learning_rate": 1.0801467131533669e-05,
      "loss": 1.1958,
      "num_input_tokens_seen": 56329784,
      "step": 97090
    },
    {
      "epoch": 14.461572832886505,
      "grad_norm": 7.0625,
      "learning_rate": 1.0798792753242099e-05,
      "loss": 1.0117,
      "num_input_tokens_seen": 56332376,
      "step": 97095
    },
    {
      "epoch": 14.462317545427466,
      "grad_norm": 23.375,
      "learning_rate": 1.0796118614865503e-05,
      "loss": 0.9505,
      "num_input_tokens_seen": 56335512,
      "step": 97100
    },
    {
      "epoch": 14.463062257968424,
      "grad_norm": 9.1875,
      "learning_rate": 1.0793444716449033e-05,
      "loss": 1.3197,
      "num_input_tokens_seen": 56338520,
      "step": 97105
    },
    {
      "epoch": 14.463806970509383,
      "grad_norm": 10.5,
      "learning_rate": 1.0790771058037889e-05,
      "loss": 1.0678,
      "num_input_tokens_seen": 56341560,
      "step": 97110
    },
    {
      "epoch": 14.464551683050342,
      "grad_norm": 8.125,
      "learning_rate": 1.0788097639677216e-05,
      "loss": 1.4894,
      "num_input_tokens_seen": 56344504,
      "step": 97115
    },
    {
      "epoch": 14.465296395591302,
      "grad_norm": 7.78125,
      "learning_rate": 1.0785424461412197e-05,
      "loss": 0.9634,
      "num_input_tokens_seen": 56347448,
      "step": 97120
    },
    {
      "epoch": 14.466041108132261,
      "grad_norm": 12.6875,
      "learning_rate": 1.0782751523287977e-05,
      "loss": 1.2775,
      "num_input_tokens_seen": 56350232,
      "step": 97125
    },
    {
      "epoch": 14.46678582067322,
      "grad_norm": 11.1875,
      "learning_rate": 1.0780078825349729e-05,
      "loss": 1.2779,
      "num_input_tokens_seen": 56353144,
      "step": 97130
    },
    {
      "epoch": 14.467530533214179,
      "grad_norm": 13.75,
      "learning_rate": 1.0777406367642595e-05,
      "loss": 0.9199,
      "num_input_tokens_seen": 56355832,
      "step": 97135
    },
    {
      "epoch": 14.46827524575514,
      "grad_norm": 10.8125,
      "learning_rate": 1.0774734150211718e-05,
      "loss": 0.8887,
      "num_input_tokens_seen": 56358488,
      "step": 97140
    },
    {
      "epoch": 14.469019958296098,
      "grad_norm": 7.8125,
      "learning_rate": 1.077206217310226e-05,
      "loss": 1.319,
      "num_input_tokens_seen": 56361528,
      "step": 97145
    },
    {
      "epoch": 14.469764670837057,
      "grad_norm": 15.3125,
      "learning_rate": 1.0769390436359348e-05,
      "loss": 1.3765,
      "num_input_tokens_seen": 56364664,
      "step": 97150
    },
    {
      "epoch": 14.470509383378015,
      "grad_norm": 13.5625,
      "learning_rate": 1.0766718940028123e-05,
      "loss": 1.1992,
      "num_input_tokens_seen": 56367416,
      "step": 97155
    },
    {
      "epoch": 14.471254095918976,
      "grad_norm": 8.875,
      "learning_rate": 1.0764047684153705e-05,
      "loss": 1.1046,
      "num_input_tokens_seen": 56370072,
      "step": 97160
    },
    {
      "epoch": 14.471998808459935,
      "grad_norm": 6.09375,
      "learning_rate": 1.0761376668781244e-05,
      "loss": 0.9416,
      "num_input_tokens_seen": 56373080,
      "step": 97165
    },
    {
      "epoch": 14.472743521000893,
      "grad_norm": 4.375,
      "learning_rate": 1.0758705893955843e-05,
      "loss": 0.9196,
      "num_input_tokens_seen": 56375960,
      "step": 97170
    },
    {
      "epoch": 14.473488233541852,
      "grad_norm": 19.125,
      "learning_rate": 1.0756035359722639e-05,
      "loss": 1.0935,
      "num_input_tokens_seen": 56379128,
      "step": 97175
    },
    {
      "epoch": 14.474232946082813,
      "grad_norm": 9.0,
      "learning_rate": 1.0753365066126741e-05,
      "loss": 1.02,
      "num_input_tokens_seen": 56382072,
      "step": 97180
    },
    {
      "epoch": 14.474977658623772,
      "grad_norm": 15.75,
      "learning_rate": 1.0750695013213251e-05,
      "loss": 1.1815,
      "num_input_tokens_seen": 56384824,
      "step": 97185
    },
    {
      "epoch": 14.47572237116473,
      "grad_norm": 15.625,
      "learning_rate": 1.0748025201027298e-05,
      "loss": 0.9978,
      "num_input_tokens_seen": 56388280,
      "step": 97190
    },
    {
      "epoch": 14.476467083705689,
      "grad_norm": 3.9375,
      "learning_rate": 1.0745355629613965e-05,
      "loss": 0.8089,
      "num_input_tokens_seen": 56391000,
      "step": 97195
    },
    {
      "epoch": 14.47721179624665,
      "grad_norm": 6.5,
      "learning_rate": 1.0742686299018368e-05,
      "loss": 1.078,
      "num_input_tokens_seen": 56393848,
      "step": 97200
    },
    {
      "epoch": 14.477956508787608,
      "grad_norm": 8.8125,
      "learning_rate": 1.0740017209285597e-05,
      "loss": 0.9157,
      "num_input_tokens_seen": 56396536,
      "step": 97205
    },
    {
      "epoch": 14.478701221328567,
      "grad_norm": 18.375,
      "learning_rate": 1.0737348360460733e-05,
      "loss": 1.6343,
      "num_input_tokens_seen": 56399256,
      "step": 97210
    },
    {
      "epoch": 14.479445933869526,
      "grad_norm": 16.125,
      "learning_rate": 1.073467975258888e-05,
      "loss": 0.9517,
      "num_input_tokens_seen": 56402008,
      "step": 97215
    },
    {
      "epoch": 14.480190646410486,
      "grad_norm": 10.875,
      "learning_rate": 1.0732011385715116e-05,
      "loss": 1.1838,
      "num_input_tokens_seen": 56404952,
      "step": 97220
    },
    {
      "epoch": 14.480935358951445,
      "grad_norm": 4.90625,
      "learning_rate": 1.0729343259884516e-05,
      "loss": 0.9749,
      "num_input_tokens_seen": 56408312,
      "step": 97225
    },
    {
      "epoch": 14.481680071492404,
      "grad_norm": 6.34375,
      "learning_rate": 1.0726675375142151e-05,
      "loss": 1.0528,
      "num_input_tokens_seen": 56411096,
      "step": 97230
    },
    {
      "epoch": 14.482424784033363,
      "grad_norm": 11.625,
      "learning_rate": 1.0724007731533107e-05,
      "loss": 1.2127,
      "num_input_tokens_seen": 56413816,
      "step": 97235
    },
    {
      "epoch": 14.483169496574323,
      "grad_norm": 23.375,
      "learning_rate": 1.072134032910243e-05,
      "loss": 1.2151,
      "num_input_tokens_seen": 56416664,
      "step": 97240
    },
    {
      "epoch": 14.483914209115282,
      "grad_norm": 5.4375,
      "learning_rate": 1.071867316789521e-05,
      "loss": 1.282,
      "num_input_tokens_seen": 56419512,
      "step": 97245
    },
    {
      "epoch": 14.48465892165624,
      "grad_norm": 13.875,
      "learning_rate": 1.0716006247956481e-05,
      "loss": 0.9864,
      "num_input_tokens_seen": 56422424,
      "step": 97250
    },
    {
      "epoch": 14.4854036341972,
      "grad_norm": 18.375,
      "learning_rate": 1.0713339569331318e-05,
      "loss": 0.8391,
      "num_input_tokens_seen": 56425496,
      "step": 97255
    },
    {
      "epoch": 14.486148346738158,
      "grad_norm": 15.1875,
      "learning_rate": 1.0710673132064764e-05,
      "loss": 1.5676,
      "num_input_tokens_seen": 56428504,
      "step": 97260
    },
    {
      "epoch": 14.486893059279119,
      "grad_norm": 7.78125,
      "learning_rate": 1.0708006936201853e-05,
      "loss": 1.0077,
      "num_input_tokens_seen": 56431352,
      "step": 97265
    },
    {
      "epoch": 14.487637771820078,
      "grad_norm": 16.5,
      "learning_rate": 1.0705340981787648e-05,
      "loss": 1.0147,
      "num_input_tokens_seen": 56434040,
      "step": 97270
    },
    {
      "epoch": 14.488382484361036,
      "grad_norm": 13.0,
      "learning_rate": 1.070267526886718e-05,
      "loss": 1.0065,
      "num_input_tokens_seen": 56436792,
      "step": 97275
    },
    {
      "epoch": 14.489127196901995,
      "grad_norm": 9.125,
      "learning_rate": 1.0700009797485483e-05,
      "loss": 1.0093,
      "num_input_tokens_seen": 56439608,
      "step": 97280
    },
    {
      "epoch": 14.489871909442956,
      "grad_norm": 6.0,
      "learning_rate": 1.0697344567687575e-05,
      "loss": 1.1997,
      "num_input_tokens_seen": 56442328,
      "step": 97285
    },
    {
      "epoch": 14.490616621983914,
      "grad_norm": 12.8125,
      "learning_rate": 1.0694679579518508e-05,
      "loss": 1.1031,
      "num_input_tokens_seen": 56445400,
      "step": 97290
    },
    {
      "epoch": 14.491361334524873,
      "grad_norm": 9.4375,
      "learning_rate": 1.0692014833023283e-05,
      "loss": 1.0808,
      "num_input_tokens_seen": 56448504,
      "step": 97295
    },
    {
      "epoch": 14.492106047065832,
      "grad_norm": 10.75,
      "learning_rate": 1.0689350328246922e-05,
      "loss": 0.9235,
      "num_input_tokens_seen": 56451160,
      "step": 97300
    },
    {
      "epoch": 14.492850759606792,
      "grad_norm": 13.625,
      "learning_rate": 1.068668606523445e-05,
      "loss": 1.0336,
      "num_input_tokens_seen": 56453720,
      "step": 97305
    },
    {
      "epoch": 14.493595472147751,
      "grad_norm": 12.3125,
      "learning_rate": 1.0684022044030861e-05,
      "loss": 0.9342,
      "num_input_tokens_seen": 56456664,
      "step": 97310
    },
    {
      "epoch": 14.49434018468871,
      "grad_norm": 12.0,
      "learning_rate": 1.0681358264681176e-05,
      "loss": 1.0867,
      "num_input_tokens_seen": 56459384,
      "step": 97315
    },
    {
      "epoch": 14.495084897229669,
      "grad_norm": 9.3125,
      "learning_rate": 1.0678694727230384e-05,
      "loss": 1.3648,
      "num_input_tokens_seen": 56462520,
      "step": 97320
    },
    {
      "epoch": 14.49582960977063,
      "grad_norm": 10.8125,
      "learning_rate": 1.0676031431723497e-05,
      "loss": 1.2501,
      "num_input_tokens_seen": 56465304,
      "step": 97325
    },
    {
      "epoch": 14.496574322311588,
      "grad_norm": 6.375,
      "learning_rate": 1.0673368378205492e-05,
      "loss": 1.0422,
      "num_input_tokens_seen": 56467960,
      "step": 97330
    },
    {
      "epoch": 14.497319034852547,
      "grad_norm": 9.875,
      "learning_rate": 1.0670705566721376e-05,
      "loss": 1.4757,
      "num_input_tokens_seen": 56471128,
      "step": 97335
    },
    {
      "epoch": 14.498063747393505,
      "grad_norm": 6.625,
      "learning_rate": 1.0668042997316126e-05,
      "loss": 1.1598,
      "num_input_tokens_seen": 56473944,
      "step": 97340
    },
    {
      "epoch": 14.498808459934466,
      "grad_norm": 10.5625,
      "learning_rate": 1.0665380670034725e-05,
      "loss": 1.1761,
      "num_input_tokens_seen": 56476760,
      "step": 97345
    },
    {
      "epoch": 14.499553172475425,
      "grad_norm": 7.53125,
      "learning_rate": 1.0662718584922145e-05,
      "loss": 1.0772,
      "num_input_tokens_seen": 56479736,
      "step": 97350
    },
    {
      "epoch": 14.500297885016384,
      "grad_norm": 21.375,
      "learning_rate": 1.0660056742023355e-05,
      "loss": 1.1261,
      "num_input_tokens_seen": 56482776,
      "step": 97355
    },
    {
      "epoch": 14.501042597557342,
      "grad_norm": 5.28125,
      "learning_rate": 1.0657395141383342e-05,
      "loss": 1.0842,
      "num_input_tokens_seen": 56486168,
      "step": 97360
    },
    {
      "epoch": 14.501787310098303,
      "grad_norm": 4.3125,
      "learning_rate": 1.0654733783047052e-05,
      "loss": 0.7752,
      "num_input_tokens_seen": 56489016,
      "step": 97365
    },
    {
      "epoch": 14.502532022639262,
      "grad_norm": 9.625,
      "learning_rate": 1.0652072667059462e-05,
      "loss": 1.0859,
      "num_input_tokens_seen": 56491928,
      "step": 97370
    },
    {
      "epoch": 14.50327673518022,
      "grad_norm": 9.875,
      "learning_rate": 1.0649411793465525e-05,
      "loss": 0.9852,
      "num_input_tokens_seen": 56494840,
      "step": 97375
    },
    {
      "epoch": 14.504021447721179,
      "grad_norm": 8.25,
      "learning_rate": 1.0646751162310178e-05,
      "loss": 0.884,
      "num_input_tokens_seen": 56497944,
      "step": 97380
    },
    {
      "epoch": 14.50476616026214,
      "grad_norm": 7.53125,
      "learning_rate": 1.0644090773638394e-05,
      "loss": 1.001,
      "num_input_tokens_seen": 56500792,
      "step": 97385
    },
    {
      "epoch": 14.505510872803098,
      "grad_norm": 11.875,
      "learning_rate": 1.0641430627495094e-05,
      "loss": 1.1047,
      "num_input_tokens_seen": 56504024,
      "step": 97390
    },
    {
      "epoch": 14.506255585344057,
      "grad_norm": 10.25,
      "learning_rate": 1.063877072392524e-05,
      "loss": 0.9986,
      "num_input_tokens_seen": 56506776,
      "step": 97395
    },
    {
      "epoch": 14.507000297885016,
      "grad_norm": 17.5,
      "learning_rate": 1.063611106297375e-05,
      "loss": 1.4995,
      "num_input_tokens_seen": 56509592,
      "step": 97400
    },
    {
      "epoch": 14.507745010425975,
      "grad_norm": 9.25,
      "learning_rate": 1.0633451644685572e-05,
      "loss": 1.1261,
      "num_input_tokens_seen": 56512312,
      "step": 97405
    },
    {
      "epoch": 14.508489722966935,
      "grad_norm": 13.3125,
      "learning_rate": 1.063079246910563e-05,
      "loss": 1.4057,
      "num_input_tokens_seen": 56515352,
      "step": 97410
    },
    {
      "epoch": 14.509234435507894,
      "grad_norm": 13.5,
      "learning_rate": 1.0628133536278842e-05,
      "loss": 0.7816,
      "num_input_tokens_seen": 56518168,
      "step": 97415
    },
    {
      "epoch": 14.509979148048853,
      "grad_norm": 9.625,
      "learning_rate": 1.0625474846250134e-05,
      "loss": 1.2105,
      "num_input_tokens_seen": 56521144,
      "step": 97420
    },
    {
      "epoch": 14.510723860589813,
      "grad_norm": 5.75,
      "learning_rate": 1.062281639906441e-05,
      "loss": 0.9863,
      "num_input_tokens_seen": 56523864,
      "step": 97425
    },
    {
      "epoch": 14.511468573130772,
      "grad_norm": 13.5625,
      "learning_rate": 1.0620158194766597e-05,
      "loss": 0.8988,
      "num_input_tokens_seen": 56526456,
      "step": 97430
    },
    {
      "epoch": 14.51221328567173,
      "grad_norm": 9.0,
      "learning_rate": 1.0617500233401587e-05,
      "loss": 1.2719,
      "num_input_tokens_seen": 56529112,
      "step": 97435
    },
    {
      "epoch": 14.51295799821269,
      "grad_norm": 10.8125,
      "learning_rate": 1.0614842515014303e-05,
      "loss": 1.1924,
      "num_input_tokens_seen": 56531832,
      "step": 97440
    },
    {
      "epoch": 14.513702710753648,
      "grad_norm": 3.625,
      "learning_rate": 1.0612185039649625e-05,
      "loss": 0.8968,
      "num_input_tokens_seen": 56534680,
      "step": 97445
    },
    {
      "epoch": 14.514447423294609,
      "grad_norm": 9.8125,
      "learning_rate": 1.0609527807352469e-05,
      "loss": 1.2103,
      "num_input_tokens_seen": 56537432,
      "step": 97450
    },
    {
      "epoch": 14.515192135835568,
      "grad_norm": 17.875,
      "learning_rate": 1.060687081816771e-05,
      "loss": 1.0627,
      "num_input_tokens_seen": 56540280,
      "step": 97455
    },
    {
      "epoch": 14.515936848376526,
      "grad_norm": 14.75,
      "learning_rate": 1.0604214072140233e-05,
      "loss": 1.1988,
      "num_input_tokens_seen": 56543480,
      "step": 97460
    },
    {
      "epoch": 14.516681560917485,
      "grad_norm": 10.75,
      "learning_rate": 1.0601557569314941e-05,
      "loss": 1.1236,
      "num_input_tokens_seen": 56546424,
      "step": 97465
    },
    {
      "epoch": 14.517426273458446,
      "grad_norm": 12.125,
      "learning_rate": 1.0598901309736686e-05,
      "loss": 1.1615,
      "num_input_tokens_seen": 56549496,
      "step": 97470
    },
    {
      "epoch": 14.518170985999404,
      "grad_norm": 6.5625,
      "learning_rate": 1.0596245293450368e-05,
      "loss": 0.8415,
      "num_input_tokens_seen": 56552344,
      "step": 97475
    },
    {
      "epoch": 14.518915698540363,
      "grad_norm": 10.3125,
      "learning_rate": 1.0593589520500846e-05,
      "loss": 1.1047,
      "num_input_tokens_seen": 56555320,
      "step": 97480
    },
    {
      "epoch": 14.519660411081322,
      "grad_norm": 5.5,
      "learning_rate": 1.0590933990932989e-05,
      "loss": 1.061,
      "num_input_tokens_seen": 56558232,
      "step": 97485
    },
    {
      "epoch": 14.520405123622282,
      "grad_norm": 18.5,
      "learning_rate": 1.0588278704791646e-05,
      "loss": 1.3368,
      "num_input_tokens_seen": 56561368,
      "step": 97490
    },
    {
      "epoch": 14.521149836163241,
      "grad_norm": 9.8125,
      "learning_rate": 1.0585623662121696e-05,
      "loss": 0.7365,
      "num_input_tokens_seen": 56563992,
      "step": 97495
    },
    {
      "epoch": 14.5218945487042,
      "grad_norm": 11.75,
      "learning_rate": 1.0582968862967984e-05,
      "loss": 1.2811,
      "num_input_tokens_seen": 56566968,
      "step": 97500
    },
    {
      "epoch": 14.522639261245159,
      "grad_norm": 17.5,
      "learning_rate": 1.0580314307375352e-05,
      "loss": 1.0046,
      "num_input_tokens_seen": 56569784,
      "step": 97505
    },
    {
      "epoch": 14.52338397378612,
      "grad_norm": 8.25,
      "learning_rate": 1.0577659995388664e-05,
      "loss": 0.8431,
      "num_input_tokens_seen": 56572408,
      "step": 97510
    },
    {
      "epoch": 14.524128686327078,
      "grad_norm": 6.9375,
      "learning_rate": 1.0575005927052743e-05,
      "loss": 1.0275,
      "num_input_tokens_seen": 56575352,
      "step": 97515
    },
    {
      "epoch": 14.524873398868037,
      "grad_norm": 10.125,
      "learning_rate": 1.0572352102412442e-05,
      "loss": 1.2106,
      "num_input_tokens_seen": 56578424,
      "step": 97520
    },
    {
      "epoch": 14.525618111408996,
      "grad_norm": 14.0625,
      "learning_rate": 1.0569698521512583e-05,
      "loss": 1.2164,
      "num_input_tokens_seen": 56581496,
      "step": 97525
    },
    {
      "epoch": 14.526362823949956,
      "grad_norm": 9.5625,
      "learning_rate": 1.0567045184398009e-05,
      "loss": 1.2955,
      "num_input_tokens_seen": 56584088,
      "step": 97530
    },
    {
      "epoch": 14.527107536490915,
      "grad_norm": 6.53125,
      "learning_rate": 1.0564392091113537e-05,
      "loss": 0.7908,
      "num_input_tokens_seen": 56587160,
      "step": 97535
    },
    {
      "epoch": 14.527852249031874,
      "grad_norm": 14.3125,
      "learning_rate": 1.056173924170398e-05,
      "loss": 0.9639,
      "num_input_tokens_seen": 56589912,
      "step": 97540
    },
    {
      "epoch": 14.528596961572832,
      "grad_norm": 8.875,
      "learning_rate": 1.0559086636214174e-05,
      "loss": 1.1548,
      "num_input_tokens_seen": 56592792,
      "step": 97545
    },
    {
      "epoch": 14.529341674113793,
      "grad_norm": 9.1875,
      "learning_rate": 1.0556434274688923e-05,
      "loss": 0.9439,
      "num_input_tokens_seen": 56595896,
      "step": 97550
    },
    {
      "epoch": 14.530086386654752,
      "grad_norm": 10.4375,
      "learning_rate": 1.0553782157173034e-05,
      "loss": 0.9876,
      "num_input_tokens_seen": 56598776,
      "step": 97555
    },
    {
      "epoch": 14.53083109919571,
      "grad_norm": 9.875,
      "learning_rate": 1.0551130283711302e-05,
      "loss": 1.2191,
      "num_input_tokens_seen": 56601720,
      "step": 97560
    },
    {
      "epoch": 14.53157581173667,
      "grad_norm": 10.8125,
      "learning_rate": 1.0548478654348551e-05,
      "loss": 1.1615,
      "num_input_tokens_seen": 56604440,
      "step": 97565
    },
    {
      "epoch": 14.53232052427763,
      "grad_norm": 11.5,
      "learning_rate": 1.0545827269129554e-05,
      "loss": 1.2071,
      "num_input_tokens_seen": 56607352,
      "step": 97570
    },
    {
      "epoch": 14.533065236818588,
      "grad_norm": 10.5,
      "learning_rate": 1.0543176128099126e-05,
      "loss": 1.1915,
      "num_input_tokens_seen": 56610264,
      "step": 97575
    },
    {
      "epoch": 14.533809949359547,
      "grad_norm": 9.375,
      "learning_rate": 1.0540525231302043e-05,
      "loss": 1.1434,
      "num_input_tokens_seen": 56613176,
      "step": 97580
    },
    {
      "epoch": 14.534554661900506,
      "grad_norm": 15.75,
      "learning_rate": 1.053787457878308e-05,
      "loss": 1.1617,
      "num_input_tokens_seen": 56615800,
      "step": 97585
    },
    {
      "epoch": 14.535299374441465,
      "grad_norm": 7.53125,
      "learning_rate": 1.0535224170587038e-05,
      "loss": 0.851,
      "num_input_tokens_seen": 56619160,
      "step": 97590
    },
    {
      "epoch": 14.536044086982425,
      "grad_norm": 13.6875,
      "learning_rate": 1.0532574006758673e-05,
      "loss": 1.0798,
      "num_input_tokens_seen": 56622168,
      "step": 97595
    },
    {
      "epoch": 14.536788799523384,
      "grad_norm": 11.875,
      "learning_rate": 1.0529924087342774e-05,
      "loss": 0.866,
      "num_input_tokens_seen": 56625016,
      "step": 97600
    },
    {
      "epoch": 14.537533512064343,
      "grad_norm": 6.21875,
      "learning_rate": 1.05272744123841e-05,
      "loss": 0.7063,
      "num_input_tokens_seen": 56627864,
      "step": 97605
    },
    {
      "epoch": 14.538278224605303,
      "grad_norm": 9.25,
      "learning_rate": 1.0524624981927416e-05,
      "loss": 1.054,
      "num_input_tokens_seen": 56630808,
      "step": 97610
    },
    {
      "epoch": 14.539022937146262,
      "grad_norm": 10.9375,
      "learning_rate": 1.0521975796017483e-05,
      "loss": 1.2376,
      "num_input_tokens_seen": 56633688,
      "step": 97615
    },
    {
      "epoch": 14.53976764968722,
      "grad_norm": 7.71875,
      "learning_rate": 1.0519326854699043e-05,
      "loss": 0.8712,
      "num_input_tokens_seen": 56636504,
      "step": 97620
    },
    {
      "epoch": 14.54051236222818,
      "grad_norm": 14.9375,
      "learning_rate": 1.0516678158016868e-05,
      "loss": 1.0399,
      "num_input_tokens_seen": 56639224,
      "step": 97625
    },
    {
      "epoch": 14.541257074769138,
      "grad_norm": 6.59375,
      "learning_rate": 1.0514029706015687e-05,
      "loss": 0.9668,
      "num_input_tokens_seen": 56642040,
      "step": 97630
    },
    {
      "epoch": 14.542001787310099,
      "grad_norm": 11.8125,
      "learning_rate": 1.051138149874026e-05,
      "loss": 1.1357,
      "num_input_tokens_seen": 56645208,
      "step": 97635
    },
    {
      "epoch": 14.542746499851058,
      "grad_norm": 7.15625,
      "learning_rate": 1.0508733536235307e-05,
      "loss": 1.7224,
      "num_input_tokens_seen": 56648024,
      "step": 97640
    },
    {
      "epoch": 14.543491212392016,
      "grad_norm": 10.4375,
      "learning_rate": 1.0506085818545582e-05,
      "loss": 1.0716,
      "num_input_tokens_seen": 56650584,
      "step": 97645
    },
    {
      "epoch": 14.544235924932975,
      "grad_norm": 16.625,
      "learning_rate": 1.0503438345715798e-05,
      "loss": 1.0287,
      "num_input_tokens_seen": 56653400,
      "step": 97650
    },
    {
      "epoch": 14.544980637473936,
      "grad_norm": 8.8125,
      "learning_rate": 1.0500791117790699e-05,
      "loss": 1.2301,
      "num_input_tokens_seen": 56656056,
      "step": 97655
    },
    {
      "epoch": 14.545725350014894,
      "grad_norm": 7.96875,
      "learning_rate": 1.0498144134814996e-05,
      "loss": 0.9962,
      "num_input_tokens_seen": 56659064,
      "step": 97660
    },
    {
      "epoch": 14.546470062555853,
      "grad_norm": 12.625,
      "learning_rate": 1.04954973968334e-05,
      "loss": 1.0427,
      "num_input_tokens_seen": 56661880,
      "step": 97665
    },
    {
      "epoch": 14.547214775096812,
      "grad_norm": 9.375,
      "learning_rate": 1.0492850903890644e-05,
      "loss": 1.2233,
      "num_input_tokens_seen": 56664664,
      "step": 97670
    },
    {
      "epoch": 14.547959487637772,
      "grad_norm": 10.875,
      "learning_rate": 1.0490204656031427e-05,
      "loss": 1.3834,
      "num_input_tokens_seen": 56667928,
      "step": 97675
    },
    {
      "epoch": 14.548704200178731,
      "grad_norm": 13.875,
      "learning_rate": 1.0487558653300455e-05,
      "loss": 0.87,
      "num_input_tokens_seen": 56670808,
      "step": 97680
    },
    {
      "epoch": 14.54944891271969,
      "grad_norm": 23.75,
      "learning_rate": 1.0484912895742422e-05,
      "loss": 1.4912,
      "num_input_tokens_seen": 56673592,
      "step": 97685
    },
    {
      "epoch": 14.550193625260649,
      "grad_norm": 7.625,
      "learning_rate": 1.0482267383402041e-05,
      "loss": 1.0802,
      "num_input_tokens_seen": 56676504,
      "step": 97690
    },
    {
      "epoch": 14.55093833780161,
      "grad_norm": 10.25,
      "learning_rate": 1.0479622116323997e-05,
      "loss": 1.1954,
      "num_input_tokens_seen": 56679320,
      "step": 97695
    },
    {
      "epoch": 14.551683050342568,
      "grad_norm": 9.8125,
      "learning_rate": 1.047697709455297e-05,
      "loss": 0.9123,
      "num_input_tokens_seen": 56682168,
      "step": 97700
    },
    {
      "epoch": 14.552427762883527,
      "grad_norm": 4.96875,
      "learning_rate": 1.0474332318133664e-05,
      "loss": 1.2127,
      "num_input_tokens_seen": 56684984,
      "step": 97705
    },
    {
      "epoch": 14.553172475424486,
      "grad_norm": 12.6875,
      "learning_rate": 1.0471687787110743e-05,
      "loss": 1.2405,
      "num_input_tokens_seen": 56687896,
      "step": 97710
    },
    {
      "epoch": 14.553917187965446,
      "grad_norm": 10.4375,
      "learning_rate": 1.04690435015289e-05,
      "loss": 1.1723,
      "num_input_tokens_seen": 56690808,
      "step": 97715
    },
    {
      "epoch": 14.554661900506405,
      "grad_norm": 15.1875,
      "learning_rate": 1.0466399461432785e-05,
      "loss": 1.1913,
      "num_input_tokens_seen": 56693624,
      "step": 97720
    },
    {
      "epoch": 14.555406613047364,
      "grad_norm": 7.3125,
      "learning_rate": 1.0463755666867093e-05,
      "loss": 0.8609,
      "num_input_tokens_seen": 56696568,
      "step": 97725
    },
    {
      "epoch": 14.556151325588322,
      "grad_norm": 4.375,
      "learning_rate": 1.0461112117876464e-05,
      "loss": 1.0721,
      "num_input_tokens_seen": 56699352,
      "step": 97730
    },
    {
      "epoch": 14.556896038129283,
      "grad_norm": 5.625,
      "learning_rate": 1.0458468814505578e-05,
      "loss": 1.0099,
      "num_input_tokens_seen": 56702232,
      "step": 97735
    },
    {
      "epoch": 14.557640750670242,
      "grad_norm": 6.0,
      "learning_rate": 1.0455825756799084e-05,
      "loss": 1.1234,
      "num_input_tokens_seen": 56704984,
      "step": 97740
    },
    {
      "epoch": 14.5583854632112,
      "grad_norm": 13.8125,
      "learning_rate": 1.0453182944801631e-05,
      "loss": 1.3523,
      "num_input_tokens_seen": 56707864,
      "step": 97745
    },
    {
      "epoch": 14.55913017575216,
      "grad_norm": 8.3125,
      "learning_rate": 1.045054037855787e-05,
      "loss": 0.8256,
      "num_input_tokens_seen": 56711032,
      "step": 97750
    },
    {
      "epoch": 14.55987488829312,
      "grad_norm": 11.375,
      "learning_rate": 1.0447898058112427e-05,
      "loss": 1.3111,
      "num_input_tokens_seen": 56713592,
      "step": 97755
    },
    {
      "epoch": 14.560619600834078,
      "grad_norm": 10.125,
      "learning_rate": 1.0445255983509969e-05,
      "loss": 0.818,
      "num_input_tokens_seen": 56716984,
      "step": 97760
    },
    {
      "epoch": 14.561364313375037,
      "grad_norm": 14.0,
      "learning_rate": 1.0442614154795107e-05,
      "loss": 1.4249,
      "num_input_tokens_seen": 56719736,
      "step": 97765
    },
    {
      "epoch": 14.562109025915996,
      "grad_norm": 8.25,
      "learning_rate": 1.0439972572012496e-05,
      "loss": 0.962,
      "num_input_tokens_seen": 56722712,
      "step": 97770
    },
    {
      "epoch": 14.562853738456955,
      "grad_norm": 6.5,
      "learning_rate": 1.0437331235206737e-05,
      "loss": 1.0668,
      "num_input_tokens_seen": 56725560,
      "step": 97775
    },
    {
      "epoch": 14.563598450997915,
      "grad_norm": 7.09375,
      "learning_rate": 1.043469014442248e-05,
      "loss": 1.1315,
      "num_input_tokens_seen": 56728600,
      "step": 97780
    },
    {
      "epoch": 14.564343163538874,
      "grad_norm": 8.0625,
      "learning_rate": 1.0432049299704324e-05,
      "loss": 0.9834,
      "num_input_tokens_seen": 56731224,
      "step": 97785
    },
    {
      "epoch": 14.565087876079833,
      "grad_norm": 7.875,
      "learning_rate": 1.0429408701096884e-05,
      "loss": 0.8271,
      "num_input_tokens_seen": 56734040,
      "step": 97790
    },
    {
      "epoch": 14.565832588620792,
      "grad_norm": 7.875,
      "learning_rate": 1.0426768348644782e-05,
      "loss": 1.0672,
      "num_input_tokens_seen": 56736952,
      "step": 97795
    },
    {
      "epoch": 14.566577301161752,
      "grad_norm": 9.625,
      "learning_rate": 1.042412824239261e-05,
      "loss": 0.9029,
      "num_input_tokens_seen": 56739928,
      "step": 97800
    },
    {
      "epoch": 14.56732201370271,
      "grad_norm": 19.875,
      "learning_rate": 1.0421488382384986e-05,
      "loss": 0.9741,
      "num_input_tokens_seen": 56742584,
      "step": 97805
    },
    {
      "epoch": 14.56806672624367,
      "grad_norm": 9.625,
      "learning_rate": 1.0418848768666498e-05,
      "loss": 0.9875,
      "num_input_tokens_seen": 56745592,
      "step": 97810
    },
    {
      "epoch": 14.568811438784628,
      "grad_norm": 12.375,
      "learning_rate": 1.041620940128174e-05,
      "loss": 1.2897,
      "num_input_tokens_seen": 56748280,
      "step": 97815
    },
    {
      "epoch": 14.569556151325589,
      "grad_norm": 6.84375,
      "learning_rate": 1.04135702802753e-05,
      "loss": 1.0343,
      "num_input_tokens_seen": 56751000,
      "step": 97820
    },
    {
      "epoch": 14.570300863866548,
      "grad_norm": 11.1875,
      "learning_rate": 1.041093140569176e-05,
      "loss": 1.2518,
      "num_input_tokens_seen": 56753688,
      "step": 97825
    },
    {
      "epoch": 14.571045576407506,
      "grad_norm": 10.125,
      "learning_rate": 1.0408292777575712e-05,
      "loss": 0.8643,
      "num_input_tokens_seen": 56756600,
      "step": 97830
    },
    {
      "epoch": 14.571790288948465,
      "grad_norm": 9.75,
      "learning_rate": 1.040565439597172e-05,
      "loss": 1.2138,
      "num_input_tokens_seen": 56759640,
      "step": 97835
    },
    {
      "epoch": 14.572535001489426,
      "grad_norm": 15.375,
      "learning_rate": 1.0403016260924373e-05,
      "loss": 1.1045,
      "num_input_tokens_seen": 56762424,
      "step": 97840
    },
    {
      "epoch": 14.573279714030384,
      "grad_norm": 12.125,
      "learning_rate": 1.040037837247822e-05,
      "loss": 1.1278,
      "num_input_tokens_seen": 56765496,
      "step": 97845
    },
    {
      "epoch": 14.574024426571343,
      "grad_norm": 6.71875,
      "learning_rate": 1.0397740730677845e-05,
      "loss": 0.9058,
      "num_input_tokens_seen": 56768440,
      "step": 97850
    },
    {
      "epoch": 14.574769139112302,
      "grad_norm": 6.5625,
      "learning_rate": 1.0395103335567794e-05,
      "loss": 1.1212,
      "num_input_tokens_seen": 56771448,
      "step": 97855
    },
    {
      "epoch": 14.575513851653263,
      "grad_norm": 9.9375,
      "learning_rate": 1.0392466187192634e-05,
      "loss": 1.3063,
      "num_input_tokens_seen": 56774456,
      "step": 97860
    },
    {
      "epoch": 14.576258564194221,
      "grad_norm": 10.5,
      "learning_rate": 1.0389829285596914e-05,
      "loss": 0.8752,
      "num_input_tokens_seen": 56777304,
      "step": 97865
    },
    {
      "epoch": 14.57700327673518,
      "grad_norm": 12.1875,
      "learning_rate": 1.0387192630825168e-05,
      "loss": 1.4683,
      "num_input_tokens_seen": 56780248,
      "step": 97870
    },
    {
      "epoch": 14.577747989276139,
      "grad_norm": 11.625,
      "learning_rate": 1.038455622292196e-05,
      "loss": 1.3072,
      "num_input_tokens_seen": 56783000,
      "step": 97875
    },
    {
      "epoch": 14.5784927018171,
      "grad_norm": 4.875,
      "learning_rate": 1.0381920061931818e-05,
      "loss": 0.7034,
      "num_input_tokens_seen": 56786008,
      "step": 97880
    },
    {
      "epoch": 14.579237414358058,
      "grad_norm": 9.875,
      "learning_rate": 1.0379284147899281e-05,
      "loss": 0.9858,
      "num_input_tokens_seen": 56789112,
      "step": 97885
    },
    {
      "epoch": 14.579982126899017,
      "grad_norm": 8.75,
      "learning_rate": 1.0376648480868872e-05,
      "loss": 1.1734,
      "num_input_tokens_seen": 56791960,
      "step": 97890
    },
    {
      "epoch": 14.580726839439976,
      "grad_norm": 23.125,
      "learning_rate": 1.0374013060885133e-05,
      "loss": 1.0976,
      "num_input_tokens_seen": 56795192,
      "step": 97895
    },
    {
      "epoch": 14.581471551980936,
      "grad_norm": 9.25,
      "learning_rate": 1.0371377887992575e-05,
      "loss": 1.2413,
      "num_input_tokens_seen": 56798072,
      "step": 97900
    },
    {
      "epoch": 14.582216264521895,
      "grad_norm": 6.09375,
      "learning_rate": 1.036874296223571e-05,
      "loss": 0.9936,
      "num_input_tokens_seen": 56800760,
      "step": 97905
    },
    {
      "epoch": 14.582960977062854,
      "grad_norm": 14.0625,
      "learning_rate": 1.0366108283659077e-05,
      "loss": 1.0785,
      "num_input_tokens_seen": 56803544,
      "step": 97910
    },
    {
      "epoch": 14.583705689603812,
      "grad_norm": 4.65625,
      "learning_rate": 1.0363473852307157e-05,
      "loss": 0.8335,
      "num_input_tokens_seen": 56806168,
      "step": 97915
    },
    {
      "epoch": 14.584450402144771,
      "grad_norm": 14.1875,
      "learning_rate": 1.0360839668224481e-05,
      "loss": 1.2511,
      "num_input_tokens_seen": 56808920,
      "step": 97920
    },
    {
      "epoch": 14.585195114685732,
      "grad_norm": 11.9375,
      "learning_rate": 1.0358205731455531e-05,
      "loss": 1.1596,
      "num_input_tokens_seen": 56811736,
      "step": 97925
    },
    {
      "epoch": 14.58593982722669,
      "grad_norm": 10.75,
      "learning_rate": 1.0355572042044823e-05,
      "loss": 1.0491,
      "num_input_tokens_seen": 56814648,
      "step": 97930
    },
    {
      "epoch": 14.58668453976765,
      "grad_norm": 6.5,
      "learning_rate": 1.0352938600036843e-05,
      "loss": 0.8116,
      "num_input_tokens_seen": 56817592,
      "step": 97935
    },
    {
      "epoch": 14.58742925230861,
      "grad_norm": 20.0,
      "learning_rate": 1.0350305405476076e-05,
      "loss": 1.207,
      "num_input_tokens_seen": 56820504,
      "step": 97940
    },
    {
      "epoch": 14.588173964849569,
      "grad_norm": 12.5625,
      "learning_rate": 1.0347672458407012e-05,
      "loss": 0.9618,
      "num_input_tokens_seen": 56823512,
      "step": 97945
    },
    {
      "epoch": 14.588918677390527,
      "grad_norm": 7.125,
      "learning_rate": 1.034503975887412e-05,
      "loss": 0.978,
      "num_input_tokens_seen": 56826648,
      "step": 97950
    },
    {
      "epoch": 14.589663389931486,
      "grad_norm": 8.5,
      "learning_rate": 1.0342407306921894e-05,
      "loss": 1.1388,
      "num_input_tokens_seen": 56829688,
      "step": 97955
    },
    {
      "epoch": 14.590408102472445,
      "grad_norm": 10.3125,
      "learning_rate": 1.0339775102594793e-05,
      "loss": 1.5431,
      "num_input_tokens_seen": 56832536,
      "step": 97960
    },
    {
      "epoch": 14.591152815013405,
      "grad_norm": 22.5,
      "learning_rate": 1.0337143145937301e-05,
      "loss": 1.065,
      "num_input_tokens_seen": 56835672,
      "step": 97965
    },
    {
      "epoch": 14.591897527554364,
      "grad_norm": 7.15625,
      "learning_rate": 1.0334511436993863e-05,
      "loss": 0.8505,
      "num_input_tokens_seen": 56838712,
      "step": 97970
    },
    {
      "epoch": 14.592642240095323,
      "grad_norm": 19.0,
      "learning_rate": 1.0331879975808956e-05,
      "loss": 1.3034,
      "num_input_tokens_seen": 56841976,
      "step": 97975
    },
    {
      "epoch": 14.593386952636282,
      "grad_norm": 8.25,
      "learning_rate": 1.032924876242703e-05,
      "loss": 0.9839,
      "num_input_tokens_seen": 56844952,
      "step": 97980
    },
    {
      "epoch": 14.594131665177242,
      "grad_norm": 8.875,
      "learning_rate": 1.0326617796892527e-05,
      "loss": 1.0108,
      "num_input_tokens_seen": 56847800,
      "step": 97985
    },
    {
      "epoch": 14.594876377718201,
      "grad_norm": 12.0625,
      "learning_rate": 1.0323987079249911e-05,
      "loss": 0.9777,
      "num_input_tokens_seen": 56850680,
      "step": 97990
    },
    {
      "epoch": 14.59562109025916,
      "grad_norm": 17.25,
      "learning_rate": 1.0321356609543608e-05,
      "loss": 0.9573,
      "num_input_tokens_seen": 56853720,
      "step": 97995
    },
    {
      "epoch": 14.596365802800118,
      "grad_norm": 10.3125,
      "learning_rate": 1.0318726387818078e-05,
      "loss": 1.3266,
      "num_input_tokens_seen": 56856728,
      "step": 98000
    },
    {
      "epoch": 14.597110515341079,
      "grad_norm": 11.875,
      "learning_rate": 1.0316096414117744e-05,
      "loss": 0.8479,
      "num_input_tokens_seen": 56859512,
      "step": 98005
    },
    {
      "epoch": 14.597855227882038,
      "grad_norm": 6.875,
      "learning_rate": 1.0313466688487034e-05,
      "loss": 1.1945,
      "num_input_tokens_seen": 56862168,
      "step": 98010
    },
    {
      "epoch": 14.598599940422996,
      "grad_norm": 10.0,
      "learning_rate": 1.031083721097037e-05,
      "loss": 0.9939,
      "num_input_tokens_seen": 56864888,
      "step": 98015
    },
    {
      "epoch": 14.599344652963955,
      "grad_norm": 6.65625,
      "learning_rate": 1.0308207981612191e-05,
      "loss": 1.0148,
      "num_input_tokens_seen": 56867608,
      "step": 98020
    },
    {
      "epoch": 14.600089365504916,
      "grad_norm": 6.65625,
      "learning_rate": 1.0305579000456907e-05,
      "loss": 0.8796,
      "num_input_tokens_seen": 56870744,
      "step": 98025
    },
    {
      "epoch": 14.600834078045875,
      "grad_norm": 9.6875,
      "learning_rate": 1.0302950267548922e-05,
      "loss": 0.906,
      "num_input_tokens_seen": 56873496,
      "step": 98030
    },
    {
      "epoch": 14.601578790586833,
      "grad_norm": 10.75,
      "learning_rate": 1.0300321782932663e-05,
      "loss": 1.2423,
      "num_input_tokens_seen": 56876376,
      "step": 98035
    },
    {
      "epoch": 14.602323503127792,
      "grad_norm": 8.625,
      "learning_rate": 1.0297693546652518e-05,
      "loss": 1.0202,
      "num_input_tokens_seen": 56879416,
      "step": 98040
    },
    {
      "epoch": 14.603068215668753,
      "grad_norm": 9.8125,
      "learning_rate": 1.0295065558752905e-05,
      "loss": 1.1003,
      "num_input_tokens_seen": 56882328,
      "step": 98045
    },
    {
      "epoch": 14.603812928209711,
      "grad_norm": 15.9375,
      "learning_rate": 1.0292437819278208e-05,
      "loss": 1.1236,
      "num_input_tokens_seen": 56885112,
      "step": 98050
    },
    {
      "epoch": 14.60455764075067,
      "grad_norm": 10.0,
      "learning_rate": 1.0289810328272836e-05,
      "loss": 1.1759,
      "num_input_tokens_seen": 56887640,
      "step": 98055
    },
    {
      "epoch": 14.605302353291629,
      "grad_norm": 5.78125,
      "learning_rate": 1.0287183085781165e-05,
      "loss": 0.8673,
      "num_input_tokens_seen": 56890488,
      "step": 98060
    },
    {
      "epoch": 14.60604706583259,
      "grad_norm": 7.34375,
      "learning_rate": 1.0284556091847575e-05,
      "loss": 1.1098,
      "num_input_tokens_seen": 56893336,
      "step": 98065
    },
    {
      "epoch": 14.606791778373548,
      "grad_norm": 8.375,
      "learning_rate": 1.028192934651646e-05,
      "loss": 1.2531,
      "num_input_tokens_seen": 56896504,
      "step": 98070
    },
    {
      "epoch": 14.607536490914507,
      "grad_norm": 15.9375,
      "learning_rate": 1.0279302849832192e-05,
      "loss": 1.1548,
      "num_input_tokens_seen": 56899320,
      "step": 98075
    },
    {
      "epoch": 14.608281203455466,
      "grad_norm": 9.4375,
      "learning_rate": 1.027667660183914e-05,
      "loss": 0.7029,
      "num_input_tokens_seen": 56902040,
      "step": 98080
    },
    {
      "epoch": 14.609025915996426,
      "grad_norm": 12.375,
      "learning_rate": 1.0274050602581667e-05,
      "loss": 1.1253,
      "num_input_tokens_seen": 56905400,
      "step": 98085
    },
    {
      "epoch": 14.609770628537385,
      "grad_norm": 5.0,
      "learning_rate": 1.0271424852104147e-05,
      "loss": 1.35,
      "num_input_tokens_seen": 56908408,
      "step": 98090
    },
    {
      "epoch": 14.610515341078344,
      "grad_norm": 8.625,
      "learning_rate": 1.0268799350450928e-05,
      "loss": 1.4275,
      "num_input_tokens_seen": 56911288,
      "step": 98095
    },
    {
      "epoch": 14.611260053619302,
      "grad_norm": 15.6875,
      "learning_rate": 1.026617409766638e-05,
      "loss": 0.9756,
      "num_input_tokens_seen": 56914232,
      "step": 98100
    },
    {
      "epoch": 14.612004766160261,
      "grad_norm": 12.0625,
      "learning_rate": 1.0263549093794847e-05,
      "loss": 0.9663,
      "num_input_tokens_seen": 56916888,
      "step": 98105
    },
    {
      "epoch": 14.612749478701222,
      "grad_norm": 6.8125,
      "learning_rate": 1.0260924338880665e-05,
      "loss": 1.2033,
      "num_input_tokens_seen": 56920056,
      "step": 98110
    },
    {
      "epoch": 14.61349419124218,
      "grad_norm": 8.5,
      "learning_rate": 1.0258299832968196e-05,
      "loss": 0.9687,
      "num_input_tokens_seen": 56923192,
      "step": 98115
    },
    {
      "epoch": 14.61423890378314,
      "grad_norm": 15.625,
      "learning_rate": 1.0255675576101759e-05,
      "loss": 1.0702,
      "num_input_tokens_seen": 56926232,
      "step": 98120
    },
    {
      "epoch": 14.6149836163241,
      "grad_norm": 13.8125,
      "learning_rate": 1.0253051568325705e-05,
      "loss": 0.8983,
      "num_input_tokens_seen": 56929112,
      "step": 98125
    },
    {
      "epoch": 14.615728328865059,
      "grad_norm": 14.1875,
      "learning_rate": 1.0250427809684349e-05,
      "loss": 1.2128,
      "num_input_tokens_seen": 56931928,
      "step": 98130
    },
    {
      "epoch": 14.616473041406017,
      "grad_norm": 7.59375,
      "learning_rate": 1.0247804300222034e-05,
      "loss": 0.7922,
      "num_input_tokens_seen": 56934776,
      "step": 98135
    },
    {
      "epoch": 14.617217753946976,
      "grad_norm": 15.8125,
      "learning_rate": 1.0245181039983068e-05,
      "loss": 0.9584,
      "num_input_tokens_seen": 56937528,
      "step": 98140
    },
    {
      "epoch": 14.617962466487935,
      "grad_norm": 9.3125,
      "learning_rate": 1.0242558029011776e-05,
      "loss": 0.9422,
      "num_input_tokens_seen": 56940184,
      "step": 98145
    },
    {
      "epoch": 14.618707179028895,
      "grad_norm": 8.4375,
      "learning_rate": 1.0239935267352466e-05,
      "loss": 0.9588,
      "num_input_tokens_seen": 56943032,
      "step": 98150
    },
    {
      "epoch": 14.619451891569854,
      "grad_norm": 11.9375,
      "learning_rate": 1.023731275504944e-05,
      "loss": 0.7448,
      "num_input_tokens_seen": 56946040,
      "step": 98155
    },
    {
      "epoch": 14.620196604110813,
      "grad_norm": 8.625,
      "learning_rate": 1.023469049214702e-05,
      "loss": 1.1996,
      "num_input_tokens_seen": 56948856,
      "step": 98160
    },
    {
      "epoch": 14.620941316651772,
      "grad_norm": 11.625,
      "learning_rate": 1.0232068478689488e-05,
      "loss": 1.0067,
      "num_input_tokens_seen": 56951992,
      "step": 98165
    },
    {
      "epoch": 14.621686029192732,
      "grad_norm": 5.5625,
      "learning_rate": 1.0229446714721158e-05,
      "loss": 0.8082,
      "num_input_tokens_seen": 56954872,
      "step": 98170
    },
    {
      "epoch": 14.622430741733691,
      "grad_norm": 6.6875,
      "learning_rate": 1.0226825200286306e-05,
      "loss": 1.0065,
      "num_input_tokens_seen": 56957496,
      "step": 98175
    },
    {
      "epoch": 14.62317545427465,
      "grad_norm": 5.34375,
      "learning_rate": 1.0224203935429235e-05,
      "loss": 1.4459,
      "num_input_tokens_seen": 56960408,
      "step": 98180
    },
    {
      "epoch": 14.623920166815608,
      "grad_norm": 8.8125,
      "learning_rate": 1.0221582920194223e-05,
      "loss": 0.9272,
      "num_input_tokens_seen": 56963288,
      "step": 98185
    },
    {
      "epoch": 14.624664879356569,
      "grad_norm": 7.75,
      "learning_rate": 1.0218962154625535e-05,
      "loss": 0.8233,
      "num_input_tokens_seen": 56966360,
      "step": 98190
    },
    {
      "epoch": 14.625409591897528,
      "grad_norm": 7.375,
      "learning_rate": 1.021634163876747e-05,
      "loss": 1.0547,
      "num_input_tokens_seen": 56969208,
      "step": 98195
    },
    {
      "epoch": 14.626154304438487,
      "grad_norm": 6.28125,
      "learning_rate": 1.0213721372664279e-05,
      "loss": 1.1759,
      "num_input_tokens_seen": 56972152,
      "step": 98200
    },
    {
      "epoch": 14.626899016979445,
      "grad_norm": 5.03125,
      "learning_rate": 1.0211101356360245e-05,
      "loss": 0.9066,
      "num_input_tokens_seen": 56974872,
      "step": 98205
    },
    {
      "epoch": 14.627643729520406,
      "grad_norm": 6.21875,
      "learning_rate": 1.0208481589899623e-05,
      "loss": 1.2186,
      "num_input_tokens_seen": 56977816,
      "step": 98210
    },
    {
      "epoch": 14.628388442061365,
      "grad_norm": 8.5625,
      "learning_rate": 1.0205862073326673e-05,
      "loss": 0.79,
      "num_input_tokens_seen": 56980568,
      "step": 98215
    },
    {
      "epoch": 14.629133154602323,
      "grad_norm": 12.6875,
      "learning_rate": 1.0203242806685645e-05,
      "loss": 0.931,
      "num_input_tokens_seen": 56983256,
      "step": 98220
    },
    {
      "epoch": 14.629877867143282,
      "grad_norm": 8.5,
      "learning_rate": 1.0200623790020782e-05,
      "loss": 1.0547,
      "num_input_tokens_seen": 56986392,
      "step": 98225
    },
    {
      "epoch": 14.630622579684243,
      "grad_norm": 14.75,
      "learning_rate": 1.0198005023376347e-05,
      "loss": 1.3689,
      "num_input_tokens_seen": 56989176,
      "step": 98230
    },
    {
      "epoch": 14.631367292225201,
      "grad_norm": 11.25,
      "learning_rate": 1.0195386506796567e-05,
      "loss": 1.3389,
      "num_input_tokens_seen": 56991992,
      "step": 98235
    },
    {
      "epoch": 14.63211200476616,
      "grad_norm": 7.28125,
      "learning_rate": 1.0192768240325693e-05,
      "loss": 0.8266,
      "num_input_tokens_seen": 56994840,
      "step": 98240
    },
    {
      "epoch": 14.632856717307119,
      "grad_norm": 14.0,
      "learning_rate": 1.019015022400794e-05,
      "loss": 1.1527,
      "num_input_tokens_seen": 56997688,
      "step": 98245
    },
    {
      "epoch": 14.63360142984808,
      "grad_norm": 7.75,
      "learning_rate": 1.0187532457887557e-05,
      "loss": 1.2619,
      "num_input_tokens_seen": 57000536,
      "step": 98250
    },
    {
      "epoch": 14.634346142389038,
      "grad_norm": 6.9375,
      "learning_rate": 1.018491494200875e-05,
      "loss": 0.9919,
      "num_input_tokens_seen": 57003128,
      "step": 98255
    },
    {
      "epoch": 14.635090854929997,
      "grad_norm": 10.6875,
      "learning_rate": 1.0182297676415755e-05,
      "loss": 1.1326,
      "num_input_tokens_seen": 57006232,
      "step": 98260
    },
    {
      "epoch": 14.635835567470956,
      "grad_norm": 12.375,
      "learning_rate": 1.0179680661152782e-05,
      "loss": 1.2455,
      "num_input_tokens_seen": 57008856,
      "step": 98265
    },
    {
      "epoch": 14.636580280011916,
      "grad_norm": 7.1875,
      "learning_rate": 1.0177063896264042e-05,
      "loss": 0.8621,
      "num_input_tokens_seen": 57011512,
      "step": 98270
    },
    {
      "epoch": 14.637324992552875,
      "grad_norm": 8.0625,
      "learning_rate": 1.0174447381793739e-05,
      "loss": 1.1695,
      "num_input_tokens_seen": 57014264,
      "step": 98275
    },
    {
      "epoch": 14.638069705093834,
      "grad_norm": 9.6875,
      "learning_rate": 1.0171831117786074e-05,
      "loss": 0.8723,
      "num_input_tokens_seen": 57017048,
      "step": 98280
    },
    {
      "epoch": 14.638814417634793,
      "grad_norm": 13.4375,
      "learning_rate": 1.016921510428526e-05,
      "loss": 0.9515,
      "num_input_tokens_seen": 57019736,
      "step": 98285
    },
    {
      "epoch": 14.639559130175751,
      "grad_norm": 8.0,
      "learning_rate": 1.0166599341335473e-05,
      "loss": 1.1854,
      "num_input_tokens_seen": 57022392,
      "step": 98290
    },
    {
      "epoch": 14.640303842716712,
      "grad_norm": 23.625,
      "learning_rate": 1.0163983828980922e-05,
      "loss": 1.2726,
      "num_input_tokens_seen": 57025240,
      "step": 98295
    },
    {
      "epoch": 14.64104855525767,
      "grad_norm": 10.5,
      "learning_rate": 1.016136856726579e-05,
      "loss": 1.1764,
      "num_input_tokens_seen": 57028376,
      "step": 98300
    },
    {
      "epoch": 14.64179326779863,
      "grad_norm": 19.625,
      "learning_rate": 1.015875355623424e-05,
      "loss": 1.3752,
      "num_input_tokens_seen": 57031128,
      "step": 98305
    },
    {
      "epoch": 14.642537980339588,
      "grad_norm": 9.625,
      "learning_rate": 1.0156138795930479e-05,
      "loss": 1.3697,
      "num_input_tokens_seen": 57033976,
      "step": 98310
    },
    {
      "epoch": 14.643282692880549,
      "grad_norm": 15.0,
      "learning_rate": 1.0153524286398656e-05,
      "loss": 1.0143,
      "num_input_tokens_seen": 57036728,
      "step": 98315
    },
    {
      "epoch": 14.644027405421507,
      "grad_norm": 10.0,
      "learning_rate": 1.0150910027682958e-05,
      "loss": 0.8846,
      "num_input_tokens_seen": 57039448,
      "step": 98320
    },
    {
      "epoch": 14.644772117962466,
      "grad_norm": 6.5,
      "learning_rate": 1.0148296019827535e-05,
      "loss": 1.0413,
      "num_input_tokens_seen": 57042520,
      "step": 98325
    },
    {
      "epoch": 14.645516830503425,
      "grad_norm": 11.75,
      "learning_rate": 1.0145682262876566e-05,
      "loss": 0.9515,
      "num_input_tokens_seen": 57045592,
      "step": 98330
    },
    {
      "epoch": 14.646261543044385,
      "grad_norm": 16.625,
      "learning_rate": 1.0143068756874197e-05,
      "loss": 1.0088,
      "num_input_tokens_seen": 57048280,
      "step": 98335
    },
    {
      "epoch": 14.647006255585344,
      "grad_norm": 4.6875,
      "learning_rate": 1.0140455501864583e-05,
      "loss": 0.9099,
      "num_input_tokens_seen": 57051192,
      "step": 98340
    },
    {
      "epoch": 14.647750968126303,
      "grad_norm": 8.4375,
      "learning_rate": 1.013784249789187e-05,
      "loss": 1.0586,
      "num_input_tokens_seen": 57054072,
      "step": 98345
    },
    {
      "epoch": 14.648495680667262,
      "grad_norm": 20.75,
      "learning_rate": 1.013522974500019e-05,
      "loss": 1.0206,
      "num_input_tokens_seen": 57057016,
      "step": 98350
    },
    {
      "epoch": 14.649240393208222,
      "grad_norm": 7.78125,
      "learning_rate": 1.013261724323371e-05,
      "loss": 0.9743,
      "num_input_tokens_seen": 57059896,
      "step": 98355
    },
    {
      "epoch": 14.649985105749181,
      "grad_norm": 16.625,
      "learning_rate": 1.0130004992636541e-05,
      "loss": 0.8667,
      "num_input_tokens_seen": 57063032,
      "step": 98360
    },
    {
      "epoch": 14.65072981829014,
      "grad_norm": 7.09375,
      "learning_rate": 1.0127392993252832e-05,
      "loss": 1.0862,
      "num_input_tokens_seen": 57065816,
      "step": 98365
    },
    {
      "epoch": 14.651474530831099,
      "grad_norm": 11.875,
      "learning_rate": 1.0124781245126695e-05,
      "loss": 0.7212,
      "num_input_tokens_seen": 57068728,
      "step": 98370
    },
    {
      "epoch": 14.652219243372059,
      "grad_norm": 6.15625,
      "learning_rate": 1.0122169748302265e-05,
      "loss": 0.9775,
      "num_input_tokens_seen": 57071768,
      "step": 98375
    },
    {
      "epoch": 14.652963955913018,
      "grad_norm": 12.25,
      "learning_rate": 1.011955850282365e-05,
      "loss": 1.0489,
      "num_input_tokens_seen": 57075288,
      "step": 98380
    },
    {
      "epoch": 14.653708668453977,
      "grad_norm": 7.46875,
      "learning_rate": 1.0116947508734981e-05,
      "loss": 0.8236,
      "num_input_tokens_seen": 57078104,
      "step": 98385
    },
    {
      "epoch": 14.654453380994935,
      "grad_norm": 11.4375,
      "learning_rate": 1.0114336766080356e-05,
      "loss": 1.1763,
      "num_input_tokens_seen": 57081080,
      "step": 98390
    },
    {
      "epoch": 14.655198093535896,
      "grad_norm": 12.0,
      "learning_rate": 1.0111726274903873e-05,
      "loss": 1.1924,
      "num_input_tokens_seen": 57084376,
      "step": 98395
    },
    {
      "epoch": 14.655942806076855,
      "grad_norm": 4.65625,
      "learning_rate": 1.0109116035249652e-05,
      "loss": 1.2579,
      "num_input_tokens_seen": 57087320,
      "step": 98400
    },
    {
      "epoch": 14.656687518617813,
      "grad_norm": 9.75,
      "learning_rate": 1.0106506047161782e-05,
      "loss": 1.1291,
      "num_input_tokens_seen": 57090232,
      "step": 98405
    },
    {
      "epoch": 14.657432231158772,
      "grad_norm": 6.34375,
      "learning_rate": 1.0103896310684356e-05,
      "loss": 0.8444,
      "num_input_tokens_seen": 57092824,
      "step": 98410
    },
    {
      "epoch": 14.658176943699733,
      "grad_norm": 7.625,
      "learning_rate": 1.010128682586145e-05,
      "loss": 0.8558,
      "num_input_tokens_seen": 57095512,
      "step": 98415
    },
    {
      "epoch": 14.658921656240691,
      "grad_norm": 22.75,
      "learning_rate": 1.009867759273717e-05,
      "loss": 1.3982,
      "num_input_tokens_seen": 57098264,
      "step": 98420
    },
    {
      "epoch": 14.65966636878165,
      "grad_norm": 6.84375,
      "learning_rate": 1.0096068611355588e-05,
      "loss": 0.9923,
      "num_input_tokens_seen": 57101176,
      "step": 98425
    },
    {
      "epoch": 14.660411081322609,
      "grad_norm": 14.125,
      "learning_rate": 1.0093459881760772e-05,
      "loss": 1.36,
      "num_input_tokens_seen": 57104120,
      "step": 98430
    },
    {
      "epoch": 14.66115579386357,
      "grad_norm": 14.3125,
      "learning_rate": 1.0090851403996809e-05,
      "loss": 1.146,
      "num_input_tokens_seen": 57107000,
      "step": 98435
    },
    {
      "epoch": 14.661900506404528,
      "grad_norm": 5.5625,
      "learning_rate": 1.0088243178107748e-05,
      "loss": 1.1379,
      "num_input_tokens_seen": 57109816,
      "step": 98440
    },
    {
      "epoch": 14.662645218945487,
      "grad_norm": 4.96875,
      "learning_rate": 1.0085635204137672e-05,
      "loss": 0.8054,
      "num_input_tokens_seen": 57112792,
      "step": 98445
    },
    {
      "epoch": 14.663389931486446,
      "grad_norm": 7.3125,
      "learning_rate": 1.0083027482130625e-05,
      "loss": 0.8968,
      "num_input_tokens_seen": 57115832,
      "step": 98450
    },
    {
      "epoch": 14.664134644027406,
      "grad_norm": 13.1875,
      "learning_rate": 1.0080420012130673e-05,
      "loss": 1.0618,
      "num_input_tokens_seen": 57118552,
      "step": 98455
    },
    {
      "epoch": 14.664879356568365,
      "grad_norm": 12.5625,
      "learning_rate": 1.0077812794181854e-05,
      "loss": 0.9257,
      "num_input_tokens_seen": 57121432,
      "step": 98460
    },
    {
      "epoch": 14.665624069109324,
      "grad_norm": 16.625,
      "learning_rate": 1.0075205828328232e-05,
      "loss": 0.8157,
      "num_input_tokens_seen": 57124504,
      "step": 98465
    },
    {
      "epoch": 14.666368781650283,
      "grad_norm": 8.0625,
      "learning_rate": 1.0072599114613837e-05,
      "loss": 0.9785,
      "num_input_tokens_seen": 57127736,
      "step": 98470
    },
    {
      "epoch": 14.667113494191241,
      "grad_norm": 14.625,
      "learning_rate": 1.0069992653082707e-05,
      "loss": 1.093,
      "num_input_tokens_seen": 57130744,
      "step": 98475
    },
    {
      "epoch": 14.667858206732202,
      "grad_norm": 17.0,
      "learning_rate": 1.0067386443778879e-05,
      "loss": 1.0025,
      "num_input_tokens_seen": 57133656,
      "step": 98480
    },
    {
      "epoch": 14.66860291927316,
      "grad_norm": 9.25,
      "learning_rate": 1.006478048674637e-05,
      "loss": 1.1046,
      "num_input_tokens_seen": 57136472,
      "step": 98485
    },
    {
      "epoch": 14.66934763181412,
      "grad_norm": 13.9375,
      "learning_rate": 1.0062174782029227e-05,
      "loss": 1.0857,
      "num_input_tokens_seen": 57139448,
      "step": 98490
    },
    {
      "epoch": 14.670092344355078,
      "grad_norm": 7.25,
      "learning_rate": 1.0059569329671448e-05,
      "loss": 1.0423,
      "num_input_tokens_seen": 57142424,
      "step": 98495
    },
    {
      "epoch": 14.670837056896039,
      "grad_norm": 19.125,
      "learning_rate": 1.0056964129717067e-05,
      "loss": 1.0281,
      "num_input_tokens_seen": 57145112,
      "step": 98500
    },
    {
      "epoch": 14.671581769436997,
      "grad_norm": 11.0625,
      "learning_rate": 1.0054359182210093e-05,
      "loss": 1.0658,
      "num_input_tokens_seen": 57147960,
      "step": 98505
    },
    {
      "epoch": 14.672326481977956,
      "grad_norm": 9.5625,
      "learning_rate": 1.005175448719452e-05,
      "loss": 0.8379,
      "num_input_tokens_seen": 57150936,
      "step": 98510
    },
    {
      "epoch": 14.673071194518915,
      "grad_norm": 8.25,
      "learning_rate": 1.0049150044714373e-05,
      "loss": 1.31,
      "num_input_tokens_seen": 57153944,
      "step": 98515
    },
    {
      "epoch": 14.673815907059875,
      "grad_norm": 10.0,
      "learning_rate": 1.004654585481363e-05,
      "loss": 1.3385,
      "num_input_tokens_seen": 57156792,
      "step": 98520
    },
    {
      "epoch": 14.674560619600834,
      "grad_norm": 9.8125,
      "learning_rate": 1.0043941917536303e-05,
      "loss": 1.1904,
      "num_input_tokens_seen": 57159960,
      "step": 98525
    },
    {
      "epoch": 14.675305332141793,
      "grad_norm": 13.5,
      "learning_rate": 1.0041338232926373e-05,
      "loss": 1.6057,
      "num_input_tokens_seen": 57162776,
      "step": 98530
    },
    {
      "epoch": 14.676050044682752,
      "grad_norm": 13.5625,
      "learning_rate": 1.0038734801027836e-05,
      "loss": 1.099,
      "num_input_tokens_seen": 57165592,
      "step": 98535
    },
    {
      "epoch": 14.676794757223712,
      "grad_norm": 10.5625,
      "learning_rate": 1.003613162188467e-05,
      "loss": 1.2876,
      "num_input_tokens_seen": 57168696,
      "step": 98540
    },
    {
      "epoch": 14.677539469764671,
      "grad_norm": 10.25,
      "learning_rate": 1.003352869554085e-05,
      "loss": 1.3481,
      "num_input_tokens_seen": 57171768,
      "step": 98545
    },
    {
      "epoch": 14.67828418230563,
      "grad_norm": 5.90625,
      "learning_rate": 1.0030926022040355e-05,
      "loss": 1.2037,
      "num_input_tokens_seen": 57174392,
      "step": 98550
    },
    {
      "epoch": 14.679028894846589,
      "grad_norm": 11.5625,
      "learning_rate": 1.002832360142714e-05,
      "loss": 1.0216,
      "num_input_tokens_seen": 57177240,
      "step": 98555
    },
    {
      "epoch": 14.679773607387549,
      "grad_norm": 12.875,
      "learning_rate": 1.0025721433745188e-05,
      "loss": 1.098,
      "num_input_tokens_seen": 57179960,
      "step": 98560
    },
    {
      "epoch": 14.680518319928508,
      "grad_norm": 6.75,
      "learning_rate": 1.0023119519038445e-05,
      "loss": 1.0578,
      "num_input_tokens_seen": 57182616,
      "step": 98565
    },
    {
      "epoch": 14.681263032469467,
      "grad_norm": 18.875,
      "learning_rate": 1.0020517857350886e-05,
      "loss": 1.2651,
      "num_input_tokens_seen": 57185368,
      "step": 98570
    },
    {
      "epoch": 14.682007745010425,
      "grad_norm": 15.75,
      "learning_rate": 1.0017916448726444e-05,
      "loss": 1.1869,
      "num_input_tokens_seen": 57188632,
      "step": 98575
    },
    {
      "epoch": 14.682752457551386,
      "grad_norm": 7.46875,
      "learning_rate": 1.0015315293209087e-05,
      "loss": 1.2385,
      "num_input_tokens_seen": 57191416,
      "step": 98580
    },
    {
      "epoch": 14.683497170092345,
      "grad_norm": 6.75,
      "learning_rate": 1.0012714390842748e-05,
      "loss": 1.2613,
      "num_input_tokens_seen": 57194296,
      "step": 98585
    },
    {
      "epoch": 14.684241882633303,
      "grad_norm": 9.9375,
      "learning_rate": 1.0010113741671356e-05,
      "loss": 1.2323,
      "num_input_tokens_seen": 57197080,
      "step": 98590
    },
    {
      "epoch": 14.684986595174262,
      "grad_norm": 11.75,
      "learning_rate": 1.0007513345738867e-05,
      "loss": 1.1687,
      "num_input_tokens_seen": 57199672,
      "step": 98595
    },
    {
      "epoch": 14.685731307715223,
      "grad_norm": 6.84375,
      "learning_rate": 1.0004913203089202e-05,
      "loss": 0.9751,
      "num_input_tokens_seen": 57202552,
      "step": 98600
    },
    {
      "epoch": 14.686476020256181,
      "grad_norm": 11.625,
      "learning_rate": 1.000231331376629e-05,
      "loss": 1.0962,
      "num_input_tokens_seen": 57205464,
      "step": 98605
    },
    {
      "epoch": 14.68722073279714,
      "grad_norm": 13.125,
      "learning_rate": 9.99971367781404e-06,
      "loss": 0.8307,
      "num_input_tokens_seen": 57208664,
      "step": 98610
    },
    {
      "epoch": 14.687965445338099,
      "grad_norm": 25.375,
      "learning_rate": 9.997114295276395e-06,
      "loss": 1.5476,
      "num_input_tokens_seen": 57211608,
      "step": 98615
    },
    {
      "epoch": 14.688710157879058,
      "grad_norm": 6.125,
      "learning_rate": 9.994515166197241e-06,
      "loss": 0.7898,
      "num_input_tokens_seen": 57214456,
      "step": 98620
    },
    {
      "epoch": 14.689454870420018,
      "grad_norm": 8.375,
      "learning_rate": 9.991916290620515e-06,
      "loss": 0.8661,
      "num_input_tokens_seen": 57217336,
      "step": 98625
    },
    {
      "epoch": 14.690199582960977,
      "grad_norm": 11.9375,
      "learning_rate": 9.98931766859011e-06,
      "loss": 1.1105,
      "num_input_tokens_seen": 57220056,
      "step": 98630
    },
    {
      "epoch": 14.690944295501936,
      "grad_norm": 10.5625,
      "learning_rate": 9.986719300149915e-06,
      "loss": 1.3323,
      "num_input_tokens_seen": 57222808,
      "step": 98635
    },
    {
      "epoch": 14.691689008042896,
      "grad_norm": 12.0625,
      "learning_rate": 9.98412118534385e-06,
      "loss": 1.2344,
      "num_input_tokens_seen": 57225752,
      "step": 98640
    },
    {
      "epoch": 14.692433720583855,
      "grad_norm": 10.125,
      "learning_rate": 9.981523324215786e-06,
      "loss": 0.9229,
      "num_input_tokens_seen": 57228632,
      "step": 98645
    },
    {
      "epoch": 14.693178433124814,
      "grad_norm": 7.53125,
      "learning_rate": 9.978925716809631e-06,
      "loss": 0.97,
      "num_input_tokens_seen": 57231384,
      "step": 98650
    },
    {
      "epoch": 14.693923145665773,
      "grad_norm": 10.5,
      "learning_rate": 9.976328363169252e-06,
      "loss": 1.052,
      "num_input_tokens_seen": 57234360,
      "step": 98655
    },
    {
      "epoch": 14.694667858206731,
      "grad_norm": 14.75,
      "learning_rate": 9.973731263338542e-06,
      "loss": 1.4202,
      "num_input_tokens_seen": 57237144,
      "step": 98660
    },
    {
      "epoch": 14.695412570747692,
      "grad_norm": 9.3125,
      "learning_rate": 9.971134417361371e-06,
      "loss": 1.1095,
      "num_input_tokens_seen": 57239992,
      "step": 98665
    },
    {
      "epoch": 14.69615728328865,
      "grad_norm": 11.625,
      "learning_rate": 9.96853782528161e-06,
      "loss": 1.3079,
      "num_input_tokens_seen": 57242552,
      "step": 98670
    },
    {
      "epoch": 14.69690199582961,
      "grad_norm": 13.0625,
      "learning_rate": 9.965941487143123e-06,
      "loss": 1.0191,
      "num_input_tokens_seen": 57245656,
      "step": 98675
    },
    {
      "epoch": 14.697646708370568,
      "grad_norm": 12.0625,
      "learning_rate": 9.963345402989768e-06,
      "loss": 1.2382,
      "num_input_tokens_seen": 57248536,
      "step": 98680
    },
    {
      "epoch": 14.698391420911529,
      "grad_norm": 10.75,
      "learning_rate": 9.96074957286542e-06,
      "loss": 1.2166,
      "num_input_tokens_seen": 57251256,
      "step": 98685
    },
    {
      "epoch": 14.699136133452487,
      "grad_norm": 6.65625,
      "learning_rate": 9.958153996813912e-06,
      "loss": 0.8664,
      "num_input_tokens_seen": 57254232,
      "step": 98690
    },
    {
      "epoch": 14.699880845993446,
      "grad_norm": 18.125,
      "learning_rate": 9.955558674879115e-06,
      "loss": 1.1244,
      "num_input_tokens_seen": 57257240,
      "step": 98695
    },
    {
      "epoch": 14.700625558534405,
      "grad_norm": 12.6875,
      "learning_rate": 9.952963607104851e-06,
      "loss": 1.0534,
      "num_input_tokens_seen": 57260056,
      "step": 98700
    },
    {
      "epoch": 14.701370271075366,
      "grad_norm": 8.6875,
      "learning_rate": 9.950368793534986e-06,
      "loss": 0.9102,
      "num_input_tokens_seen": 57263256,
      "step": 98705
    },
    {
      "epoch": 14.702114983616324,
      "grad_norm": 7.25,
      "learning_rate": 9.947774234213342e-06,
      "loss": 1.1991,
      "num_input_tokens_seen": 57266008,
      "step": 98710
    },
    {
      "epoch": 14.702859696157283,
      "grad_norm": 4.875,
      "learning_rate": 9.945179929183749e-06,
      "loss": 1.0918,
      "num_input_tokens_seen": 57269208,
      "step": 98715
    },
    {
      "epoch": 14.703604408698242,
      "grad_norm": 12.0,
      "learning_rate": 9.942585878490046e-06,
      "loss": 0.8753,
      "num_input_tokens_seen": 57272024,
      "step": 98720
    },
    {
      "epoch": 14.704349121239202,
      "grad_norm": 9.75,
      "learning_rate": 9.939992082176041e-06,
      "loss": 1.2736,
      "num_input_tokens_seen": 57274808,
      "step": 98725
    },
    {
      "epoch": 14.705093833780161,
      "grad_norm": 11.0,
      "learning_rate": 9.937398540285575e-06,
      "loss": 0.8695,
      "num_input_tokens_seen": 57277816,
      "step": 98730
    },
    {
      "epoch": 14.70583854632112,
      "grad_norm": 5.53125,
      "learning_rate": 9.93480525286245e-06,
      "loss": 0.9858,
      "num_input_tokens_seen": 57280632,
      "step": 98735
    },
    {
      "epoch": 14.706583258862079,
      "grad_norm": 10.0625,
      "learning_rate": 9.93221221995048e-06,
      "loss": 0.7236,
      "num_input_tokens_seen": 57283544,
      "step": 98740
    },
    {
      "epoch": 14.70732797140304,
      "grad_norm": 14.0,
      "learning_rate": 9.929619441593469e-06,
      "loss": 1.3211,
      "num_input_tokens_seen": 57286744,
      "step": 98745
    },
    {
      "epoch": 14.708072683943998,
      "grad_norm": 7.09375,
      "learning_rate": 9.927026917835211e-06,
      "loss": 1.1402,
      "num_input_tokens_seen": 57289848,
      "step": 98750
    },
    {
      "epoch": 14.708817396484957,
      "grad_norm": 15.3125,
      "learning_rate": 9.924434648719525e-06,
      "loss": 1.0487,
      "num_input_tokens_seen": 57292888,
      "step": 98755
    },
    {
      "epoch": 14.709562109025915,
      "grad_norm": 9.8125,
      "learning_rate": 9.921842634290182e-06,
      "loss": 1.1847,
      "num_input_tokens_seen": 57295704,
      "step": 98760
    },
    {
      "epoch": 14.710306821566876,
      "grad_norm": 9.5625,
      "learning_rate": 9.919250874590993e-06,
      "loss": 1.0578,
      "num_input_tokens_seen": 57298712,
      "step": 98765
    },
    {
      "epoch": 14.711051534107835,
      "grad_norm": 10.375,
      "learning_rate": 9.916659369665726e-06,
      "loss": 1.2358,
      "num_input_tokens_seen": 57301720,
      "step": 98770
    },
    {
      "epoch": 14.711796246648793,
      "grad_norm": 8.25,
      "learning_rate": 9.914068119558177e-06,
      "loss": 1.1207,
      "num_input_tokens_seen": 57304760,
      "step": 98775
    },
    {
      "epoch": 14.712540959189752,
      "grad_norm": 10.4375,
      "learning_rate": 9.911477124312104e-06,
      "loss": 1.033,
      "num_input_tokens_seen": 57307256,
      "step": 98780
    },
    {
      "epoch": 14.713285671730713,
      "grad_norm": 13.5,
      "learning_rate": 9.9088863839713e-06,
      "loss": 1.3175,
      "num_input_tokens_seen": 57309848,
      "step": 98785
    },
    {
      "epoch": 14.714030384271672,
      "grad_norm": 11.625,
      "learning_rate": 9.90629589857952e-06,
      "loss": 0.8402,
      "num_input_tokens_seen": 57313080,
      "step": 98790
    },
    {
      "epoch": 14.71477509681263,
      "grad_norm": 10.25,
      "learning_rate": 9.903705668180524e-06,
      "loss": 1.0614,
      "num_input_tokens_seen": 57315896,
      "step": 98795
    },
    {
      "epoch": 14.715519809353589,
      "grad_norm": 7.875,
      "learning_rate": 9.901115692818085e-06,
      "loss": 1.1576,
      "num_input_tokens_seen": 57318872,
      "step": 98800
    },
    {
      "epoch": 14.716264521894548,
      "grad_norm": 8.5625,
      "learning_rate": 9.898525972535952e-06,
      "loss": 1.0425,
      "num_input_tokens_seen": 57321912,
      "step": 98805
    },
    {
      "epoch": 14.717009234435508,
      "grad_norm": 12.8125,
      "learning_rate": 9.895936507377873e-06,
      "loss": 1.2084,
      "num_input_tokens_seen": 57324824,
      "step": 98810
    },
    {
      "epoch": 14.717753946976467,
      "grad_norm": 11.125,
      "learning_rate": 9.89334729738759e-06,
      "loss": 1.3573,
      "num_input_tokens_seen": 57327800,
      "step": 98815
    },
    {
      "epoch": 14.718498659517426,
      "grad_norm": 10.375,
      "learning_rate": 9.890758342608856e-06,
      "loss": 0.8864,
      "num_input_tokens_seen": 57330456,
      "step": 98820
    },
    {
      "epoch": 14.719243372058386,
      "grad_norm": 11.9375,
      "learning_rate": 9.888169643085404e-06,
      "loss": 1.0248,
      "num_input_tokens_seen": 57333464,
      "step": 98825
    },
    {
      "epoch": 14.719988084599345,
      "grad_norm": 13.375,
      "learning_rate": 9.885581198860958e-06,
      "loss": 1.2659,
      "num_input_tokens_seen": 57336408,
      "step": 98830
    },
    {
      "epoch": 14.720732797140304,
      "grad_norm": 10.0,
      "learning_rate": 9.882993009979265e-06,
      "loss": 1.1477,
      "num_input_tokens_seen": 57339288,
      "step": 98835
    },
    {
      "epoch": 14.721477509681263,
      "grad_norm": 6.53125,
      "learning_rate": 9.880405076484034e-06,
      "loss": 1.2828,
      "num_input_tokens_seen": 57342200,
      "step": 98840
    },
    {
      "epoch": 14.722222222222221,
      "grad_norm": 10.8125,
      "learning_rate": 9.877817398418998e-06,
      "loss": 0.9775,
      "num_input_tokens_seen": 57345080,
      "step": 98845
    },
    {
      "epoch": 14.722966934763182,
      "grad_norm": 11.4375,
      "learning_rate": 9.87522997582786e-06,
      "loss": 0.7227,
      "num_input_tokens_seen": 57348056,
      "step": 98850
    },
    {
      "epoch": 14.72371164730414,
      "grad_norm": 7.65625,
      "learning_rate": 9.872642808754348e-06,
      "loss": 0.9688,
      "num_input_tokens_seen": 57351032,
      "step": 98855
    },
    {
      "epoch": 14.7244563598451,
      "grad_norm": 6.15625,
      "learning_rate": 9.870055897242152e-06,
      "loss": 0.8521,
      "num_input_tokens_seen": 57354040,
      "step": 98860
    },
    {
      "epoch": 14.725201072386058,
      "grad_norm": 7.78125,
      "learning_rate": 9.867469241334994e-06,
      "loss": 1.0496,
      "num_input_tokens_seen": 57356952,
      "step": 98865
    },
    {
      "epoch": 14.725945784927019,
      "grad_norm": 16.5,
      "learning_rate": 9.864882841076564e-06,
      "loss": 1.2562,
      "num_input_tokens_seen": 57359704,
      "step": 98870
    },
    {
      "epoch": 14.726690497467978,
      "grad_norm": 7.34375,
      "learning_rate": 9.862296696510557e-06,
      "loss": 1.1085,
      "num_input_tokens_seen": 57362552,
      "step": 98875
    },
    {
      "epoch": 14.727435210008936,
      "grad_norm": 7.9375,
      "learning_rate": 9.859710807680658e-06,
      "loss": 0.9279,
      "num_input_tokens_seen": 57365560,
      "step": 98880
    },
    {
      "epoch": 14.728179922549895,
      "grad_norm": 7.15625,
      "learning_rate": 9.85712517463055e-06,
      "loss": 0.9454,
      "num_input_tokens_seen": 57368216,
      "step": 98885
    },
    {
      "epoch": 14.728924635090856,
      "grad_norm": 8.4375,
      "learning_rate": 9.85453979740393e-06,
      "loss": 0.9247,
      "num_input_tokens_seen": 57370936,
      "step": 98890
    },
    {
      "epoch": 14.729669347631814,
      "grad_norm": 7.9375,
      "learning_rate": 9.851954676044458e-06,
      "loss": 1.0727,
      "num_input_tokens_seen": 57373624,
      "step": 98895
    },
    {
      "epoch": 14.730414060172773,
      "grad_norm": 8.9375,
      "learning_rate": 9.849369810595827e-06,
      "loss": 1.0487,
      "num_input_tokens_seen": 57376664,
      "step": 98900
    },
    {
      "epoch": 14.731158772713732,
      "grad_norm": 4.21875,
      "learning_rate": 9.846785201101691e-06,
      "loss": 1.003,
      "num_input_tokens_seen": 57379416,
      "step": 98905
    },
    {
      "epoch": 14.731903485254692,
      "grad_norm": 6.96875,
      "learning_rate": 9.84420084760571e-06,
      "loss": 1.3756,
      "num_input_tokens_seen": 57382200,
      "step": 98910
    },
    {
      "epoch": 14.732648197795651,
      "grad_norm": 25.125,
      "learning_rate": 9.841616750151565e-06,
      "loss": 1.5595,
      "num_input_tokens_seen": 57385144,
      "step": 98915
    },
    {
      "epoch": 14.73339291033661,
      "grad_norm": 9.6875,
      "learning_rate": 9.839032908782885e-06,
      "loss": 1.2154,
      "num_input_tokens_seen": 57388184,
      "step": 98920
    },
    {
      "epoch": 14.734137622877569,
      "grad_norm": 9.5625,
      "learning_rate": 9.836449323543345e-06,
      "loss": 1.0413,
      "num_input_tokens_seen": 57391000,
      "step": 98925
    },
    {
      "epoch": 14.73488233541853,
      "grad_norm": 9.125,
      "learning_rate": 9.833865994476584e-06,
      "loss": 1.1315,
      "num_input_tokens_seen": 57393944,
      "step": 98930
    },
    {
      "epoch": 14.735627047959488,
      "grad_norm": 7.21875,
      "learning_rate": 9.831282921626242e-06,
      "loss": 0.9172,
      "num_input_tokens_seen": 57396792,
      "step": 98935
    },
    {
      "epoch": 14.736371760500447,
      "grad_norm": 5.59375,
      "learning_rate": 9.82870010503595e-06,
      "loss": 1.041,
      "num_input_tokens_seen": 57399896,
      "step": 98940
    },
    {
      "epoch": 14.737116473041405,
      "grad_norm": 9.5625,
      "learning_rate": 9.826117544749357e-06,
      "loss": 0.8875,
      "num_input_tokens_seen": 57403096,
      "step": 98945
    },
    {
      "epoch": 14.737861185582366,
      "grad_norm": 12.8125,
      "learning_rate": 9.823535240810089e-06,
      "loss": 1.2635,
      "num_input_tokens_seen": 57406008,
      "step": 98950
    },
    {
      "epoch": 14.738605898123325,
      "grad_norm": 11.6875,
      "learning_rate": 9.820953193261756e-06,
      "loss": 1.1473,
      "num_input_tokens_seen": 57408888,
      "step": 98955
    },
    {
      "epoch": 14.739350610664284,
      "grad_norm": 7.5,
      "learning_rate": 9.818371402148002e-06,
      "loss": 1.3414,
      "num_input_tokens_seen": 57411608,
      "step": 98960
    },
    {
      "epoch": 14.740095323205242,
      "grad_norm": 6.96875,
      "learning_rate": 9.815789867512427e-06,
      "loss": 0.9387,
      "num_input_tokens_seen": 57414680,
      "step": 98965
    },
    {
      "epoch": 14.740840035746203,
      "grad_norm": 9.8125,
      "learning_rate": 9.813208589398654e-06,
      "loss": 1.1463,
      "num_input_tokens_seen": 57417688,
      "step": 98970
    },
    {
      "epoch": 14.741584748287162,
      "grad_norm": 11.875,
      "learning_rate": 9.81062756785028e-06,
      "loss": 1.4013,
      "num_input_tokens_seen": 57420408,
      "step": 98975
    },
    {
      "epoch": 14.74232946082812,
      "grad_norm": 8.0625,
      "learning_rate": 9.808046802910926e-06,
      "loss": 1.2088,
      "num_input_tokens_seen": 57423032,
      "step": 98980
    },
    {
      "epoch": 14.743074173369079,
      "grad_norm": 7.53125,
      "learning_rate": 9.80546629462417e-06,
      "loss": 0.8317,
      "num_input_tokens_seen": 57426136,
      "step": 98985
    },
    {
      "epoch": 14.743818885910038,
      "grad_norm": 10.5,
      "learning_rate": 9.802886043033626e-06,
      "loss": 1.1273,
      "num_input_tokens_seen": 57429176,
      "step": 98990
    },
    {
      "epoch": 14.744563598450998,
      "grad_norm": 14.375,
      "learning_rate": 9.800306048182878e-06,
      "loss": 0.9654,
      "num_input_tokens_seen": 57432120,
      "step": 98995
    },
    {
      "epoch": 14.745308310991957,
      "grad_norm": 6.71875,
      "learning_rate": 9.79772631011551e-06,
      "loss": 0.9763,
      "num_input_tokens_seen": 57435128,
      "step": 99000
    },
    {
      "epoch": 14.746053023532916,
      "grad_norm": 11.375,
      "learning_rate": 9.795146828875107e-06,
      "loss": 1.2154,
      "num_input_tokens_seen": 57437784,
      "step": 99005
    },
    {
      "epoch": 14.746797736073875,
      "grad_norm": 8.9375,
      "learning_rate": 9.792567604505234e-06,
      "loss": 0.6447,
      "num_input_tokens_seen": 57440856,
      "step": 99010
    },
    {
      "epoch": 14.747542448614835,
      "grad_norm": 12.0,
      "learning_rate": 9.789988637049485e-06,
      "loss": 1.0031,
      "num_input_tokens_seen": 57443864,
      "step": 99015
    },
    {
      "epoch": 14.748287161155794,
      "grad_norm": 10.625,
      "learning_rate": 9.787409926551411e-06,
      "loss": 1.3856,
      "num_input_tokens_seen": 57446488,
      "step": 99020
    },
    {
      "epoch": 14.749031873696753,
      "grad_norm": 18.25,
      "learning_rate": 9.784831473054592e-06,
      "loss": 1.1182,
      "num_input_tokens_seen": 57449400,
      "step": 99025
    },
    {
      "epoch": 14.749776586237711,
      "grad_norm": 9.6875,
      "learning_rate": 9.782253276602582e-06,
      "loss": 1.4109,
      "num_input_tokens_seen": 57452280,
      "step": 99030
    },
    {
      "epoch": 14.750521298778672,
      "grad_norm": 15.625,
      "learning_rate": 9.779675337238928e-06,
      "loss": 1.7862,
      "num_input_tokens_seen": 57455416,
      "step": 99035
    },
    {
      "epoch": 14.75126601131963,
      "grad_norm": 10.5625,
      "learning_rate": 9.777097655007197e-06,
      "loss": 1.0863,
      "num_input_tokens_seen": 57458424,
      "step": 99040
    },
    {
      "epoch": 14.75201072386059,
      "grad_norm": 9.5,
      "learning_rate": 9.774520229950923e-06,
      "loss": 1.0691,
      "num_input_tokens_seen": 57461848,
      "step": 99045
    },
    {
      "epoch": 14.752755436401548,
      "grad_norm": 7.3125,
      "learning_rate": 9.771943062113664e-06,
      "loss": 1.2479,
      "num_input_tokens_seen": 57464984,
      "step": 99050
    },
    {
      "epoch": 14.753500148942509,
      "grad_norm": 11.3125,
      "learning_rate": 9.76936615153894e-06,
      "loss": 0.8218,
      "num_input_tokens_seen": 57467896,
      "step": 99055
    },
    {
      "epoch": 14.754244861483468,
      "grad_norm": 8.0625,
      "learning_rate": 9.766789498270304e-06,
      "loss": 0.9273,
      "num_input_tokens_seen": 57470456,
      "step": 99060
    },
    {
      "epoch": 14.754989574024426,
      "grad_norm": 18.125,
      "learning_rate": 9.764213102351275e-06,
      "loss": 1.1389,
      "num_input_tokens_seen": 57473208,
      "step": 99065
    },
    {
      "epoch": 14.755734286565385,
      "grad_norm": 7.4375,
      "learning_rate": 9.761636963825382e-06,
      "loss": 0.9364,
      "num_input_tokens_seen": 57476184,
      "step": 99070
    },
    {
      "epoch": 14.756478999106346,
      "grad_norm": 12.625,
      "learning_rate": 9.759061082736145e-06,
      "loss": 1.2436,
      "num_input_tokens_seen": 57478776,
      "step": 99075
    },
    {
      "epoch": 14.757223711647304,
      "grad_norm": 10.9375,
      "learning_rate": 9.756485459127073e-06,
      "loss": 0.9749,
      "num_input_tokens_seen": 57481432,
      "step": 99080
    },
    {
      "epoch": 14.757968424188263,
      "grad_norm": 9.5625,
      "learning_rate": 9.753910093041696e-06,
      "loss": 0.9255,
      "num_input_tokens_seen": 57484408,
      "step": 99085
    },
    {
      "epoch": 14.758713136729222,
      "grad_norm": 10.8125,
      "learning_rate": 9.751334984523502e-06,
      "loss": 1.1348,
      "num_input_tokens_seen": 57487544,
      "step": 99090
    },
    {
      "epoch": 14.759457849270182,
      "grad_norm": 6.9375,
      "learning_rate": 9.748760133616015e-06,
      "loss": 1.0722,
      "num_input_tokens_seen": 57490232,
      "step": 99095
    },
    {
      "epoch": 14.760202561811141,
      "grad_norm": 12.0625,
      "learning_rate": 9.746185540362714e-06,
      "loss": 1.1172,
      "num_input_tokens_seen": 57492824,
      "step": 99100
    },
    {
      "epoch": 14.7609472743521,
      "grad_norm": 10.375,
      "learning_rate": 9.743611204807118e-06,
      "loss": 1.1783,
      "num_input_tokens_seen": 57495608,
      "step": 99105
    },
    {
      "epoch": 14.761691986893059,
      "grad_norm": 11.0,
      "learning_rate": 9.741037126992702e-06,
      "loss": 1.0527,
      "num_input_tokens_seen": 57498648,
      "step": 99110
    },
    {
      "epoch": 14.76243669943402,
      "grad_norm": 7.5,
      "learning_rate": 9.738463306962947e-06,
      "loss": 1.113,
      "num_input_tokens_seen": 57501592,
      "step": 99115
    },
    {
      "epoch": 14.763181411974978,
      "grad_norm": 8.4375,
      "learning_rate": 9.73588974476135e-06,
      "loss": 1.0998,
      "num_input_tokens_seen": 57504600,
      "step": 99120
    },
    {
      "epoch": 14.763926124515937,
      "grad_norm": 10.0625,
      "learning_rate": 9.733316440431375e-06,
      "loss": 0.9496,
      "num_input_tokens_seen": 57507512,
      "step": 99125
    },
    {
      "epoch": 14.764670837056896,
      "grad_norm": 14.25,
      "learning_rate": 9.730743394016512e-06,
      "loss": 0.8315,
      "num_input_tokens_seen": 57510584,
      "step": 99130
    },
    {
      "epoch": 14.765415549597854,
      "grad_norm": 12.0625,
      "learning_rate": 9.72817060556022e-06,
      "loss": 1.2874,
      "num_input_tokens_seen": 57513592,
      "step": 99135
    },
    {
      "epoch": 14.766160262138815,
      "grad_norm": 12.5625,
      "learning_rate": 9.725598075105963e-06,
      "loss": 1.0617,
      "num_input_tokens_seen": 57516504,
      "step": 99140
    },
    {
      "epoch": 14.766904974679774,
      "grad_norm": 8.3125,
      "learning_rate": 9.723025802697195e-06,
      "loss": 0.6605,
      "num_input_tokens_seen": 57519256,
      "step": 99145
    },
    {
      "epoch": 14.767649687220732,
      "grad_norm": 18.0,
      "learning_rate": 9.720453788377387e-06,
      "loss": 0.9781,
      "num_input_tokens_seen": 57522616,
      "step": 99150
    },
    {
      "epoch": 14.768394399761693,
      "grad_norm": 11.125,
      "learning_rate": 9.71788203218998e-06,
      "loss": 0.9402,
      "num_input_tokens_seen": 57525496,
      "step": 99155
    },
    {
      "epoch": 14.769139112302652,
      "grad_norm": 17.0,
      "learning_rate": 9.71531053417842e-06,
      "loss": 1.3594,
      "num_input_tokens_seen": 57528824,
      "step": 99160
    },
    {
      "epoch": 14.76988382484361,
      "grad_norm": 13.375,
      "learning_rate": 9.712739294386161e-06,
      "loss": 1.1727,
      "num_input_tokens_seen": 57531864,
      "step": 99165
    },
    {
      "epoch": 14.77062853738457,
      "grad_norm": 6.875,
      "learning_rate": 9.710168312856626e-06,
      "loss": 0.9797,
      "num_input_tokens_seen": 57534680,
      "step": 99170
    },
    {
      "epoch": 14.771373249925528,
      "grad_norm": 13.125,
      "learning_rate": 9.707597589633267e-06,
      "loss": 1.2165,
      "num_input_tokens_seen": 57537688,
      "step": 99175
    },
    {
      "epoch": 14.772117962466488,
      "grad_norm": 8.375,
      "learning_rate": 9.705027124759495e-06,
      "loss": 1.0753,
      "num_input_tokens_seen": 57540504,
      "step": 99180
    },
    {
      "epoch": 14.772862675007447,
      "grad_norm": 18.125,
      "learning_rate": 9.702456918278752e-06,
      "loss": 0.9922,
      "num_input_tokens_seen": 57543480,
      "step": 99185
    },
    {
      "epoch": 14.773607387548406,
      "grad_norm": 12.8125,
      "learning_rate": 9.69988697023445e-06,
      "loss": 1.0417,
      "num_input_tokens_seen": 57546328,
      "step": 99190
    },
    {
      "epoch": 14.774352100089365,
      "grad_norm": 4.3125,
      "learning_rate": 9.69731728067e-06,
      "loss": 1.1961,
      "num_input_tokens_seen": 57549016,
      "step": 99195
    },
    {
      "epoch": 14.775096812630325,
      "grad_norm": 11.0625,
      "learning_rate": 9.694747849628833e-06,
      "loss": 1.4401,
      "num_input_tokens_seen": 57551864,
      "step": 99200
    },
    {
      "epoch": 14.775841525171284,
      "grad_norm": 13.875,
      "learning_rate": 9.692178677154342e-06,
      "loss": 1.6057,
      "num_input_tokens_seen": 57554744,
      "step": 99205
    },
    {
      "epoch": 14.776586237712243,
      "grad_norm": 7.03125,
      "learning_rate": 9.689609763289936e-06,
      "loss": 1.2705,
      "num_input_tokens_seen": 57557752,
      "step": 99210
    },
    {
      "epoch": 14.777330950253202,
      "grad_norm": 14.4375,
      "learning_rate": 9.687041108079003e-06,
      "loss": 1.0964,
      "num_input_tokens_seen": 57560472,
      "step": 99215
    },
    {
      "epoch": 14.778075662794162,
      "grad_norm": 16.0,
      "learning_rate": 9.684472711564957e-06,
      "loss": 1.2173,
      "num_input_tokens_seen": 57563608,
      "step": 99220
    },
    {
      "epoch": 14.77882037533512,
      "grad_norm": 6.53125,
      "learning_rate": 9.681904573791168e-06,
      "loss": 1.1547,
      "num_input_tokens_seen": 57566392,
      "step": 99225
    },
    {
      "epoch": 14.77956508787608,
      "grad_norm": 12.0625,
      "learning_rate": 9.679336694801041e-06,
      "loss": 1.3157,
      "num_input_tokens_seen": 57569400,
      "step": 99230
    },
    {
      "epoch": 14.780309800417038,
      "grad_norm": 6.53125,
      "learning_rate": 9.67676907463795e-06,
      "loss": 0.9619,
      "num_input_tokens_seen": 57572440,
      "step": 99235
    },
    {
      "epoch": 14.781054512957999,
      "grad_norm": 4.25,
      "learning_rate": 9.674201713345265e-06,
      "loss": 1.083,
      "num_input_tokens_seen": 57575160,
      "step": 99240
    },
    {
      "epoch": 14.781799225498958,
      "grad_norm": 6.15625,
      "learning_rate": 9.671634610966373e-06,
      "loss": 0.8456,
      "num_input_tokens_seen": 57577976,
      "step": 99245
    },
    {
      "epoch": 14.782543938039916,
      "grad_norm": 6.4375,
      "learning_rate": 9.669067767544626e-06,
      "loss": 0.8897,
      "num_input_tokens_seen": 57580664,
      "step": 99250
    },
    {
      "epoch": 14.783288650580875,
      "grad_norm": 11.125,
      "learning_rate": 9.666501183123406e-06,
      "loss": 1.5325,
      "num_input_tokens_seen": 57583640,
      "step": 99255
    },
    {
      "epoch": 14.784033363121836,
      "grad_norm": 8.6875,
      "learning_rate": 9.663934857746065e-06,
      "loss": 1.2624,
      "num_input_tokens_seen": 57586264,
      "step": 99260
    },
    {
      "epoch": 14.784778075662794,
      "grad_norm": 11.0625,
      "learning_rate": 9.661368791455957e-06,
      "loss": 0.9623,
      "num_input_tokens_seen": 57588984,
      "step": 99265
    },
    {
      "epoch": 14.785522788203753,
      "grad_norm": 7.5,
      "learning_rate": 9.658802984296426e-06,
      "loss": 1.2909,
      "num_input_tokens_seen": 57591960,
      "step": 99270
    },
    {
      "epoch": 14.786267500744712,
      "grad_norm": 9.375,
      "learning_rate": 9.656237436310834e-06,
      "loss": 1.0205,
      "num_input_tokens_seen": 57595192,
      "step": 99275
    },
    {
      "epoch": 14.787012213285673,
      "grad_norm": 12.375,
      "learning_rate": 9.653672147542515e-06,
      "loss": 1.2059,
      "num_input_tokens_seen": 57597944,
      "step": 99280
    },
    {
      "epoch": 14.787756925826631,
      "grad_norm": 12.25,
      "learning_rate": 9.651107118034799e-06,
      "loss": 1.0476,
      "num_input_tokens_seen": 57601112,
      "step": 99285
    },
    {
      "epoch": 14.78850163836759,
      "grad_norm": 7.875,
      "learning_rate": 9.648542347831041e-06,
      "loss": 1.2339,
      "num_input_tokens_seen": 57603960,
      "step": 99290
    },
    {
      "epoch": 14.789246350908549,
      "grad_norm": 7.3125,
      "learning_rate": 9.645977836974545e-06,
      "loss": 1.1248,
      "num_input_tokens_seen": 57607000,
      "step": 99295
    },
    {
      "epoch": 14.78999106344951,
      "grad_norm": 12.0625,
      "learning_rate": 9.643413585508659e-06,
      "loss": 1.0995,
      "num_input_tokens_seen": 57609720,
      "step": 99300
    },
    {
      "epoch": 14.790735775990468,
      "grad_norm": 8.3125,
      "learning_rate": 9.640849593476684e-06,
      "loss": 1.0324,
      "num_input_tokens_seen": 57612568,
      "step": 99305
    },
    {
      "epoch": 14.791480488531427,
      "grad_norm": 10.5625,
      "learning_rate": 9.63828586092195e-06,
      "loss": 1.0177,
      "num_input_tokens_seen": 57615448,
      "step": 99310
    },
    {
      "epoch": 14.792225201072386,
      "grad_norm": 18.5,
      "learning_rate": 9.635722387887766e-06,
      "loss": 0.9983,
      "num_input_tokens_seen": 57618392,
      "step": 99315
    },
    {
      "epoch": 14.792969913613344,
      "grad_norm": 6.75,
      "learning_rate": 9.63315917441743e-06,
      "loss": 0.9758,
      "num_input_tokens_seen": 57621016,
      "step": 99320
    },
    {
      "epoch": 14.793714626154305,
      "grad_norm": 8.5,
      "learning_rate": 9.630596220554259e-06,
      "loss": 1.04,
      "num_input_tokens_seen": 57624504,
      "step": 99325
    },
    {
      "epoch": 14.794459338695264,
      "grad_norm": 10.4375,
      "learning_rate": 9.628033526341542e-06,
      "loss": 0.7746,
      "num_input_tokens_seen": 57627416,
      "step": 99330
    },
    {
      "epoch": 14.795204051236222,
      "grad_norm": 5.75,
      "learning_rate": 9.625471091822576e-06,
      "loss": 0.8304,
      "num_input_tokens_seen": 57630264,
      "step": 99335
    },
    {
      "epoch": 14.795948763777183,
      "grad_norm": 11.75,
      "learning_rate": 9.622908917040643e-06,
      "loss": 1.1815,
      "num_input_tokens_seen": 57633144,
      "step": 99340
    },
    {
      "epoch": 14.796693476318142,
      "grad_norm": 10.25,
      "learning_rate": 9.620347002039042e-06,
      "loss": 1.0608,
      "num_input_tokens_seen": 57635928,
      "step": 99345
    },
    {
      "epoch": 14.7974381888591,
      "grad_norm": 7.59375,
      "learning_rate": 9.61778534686105e-06,
      "loss": 1.2063,
      "num_input_tokens_seen": 57638712,
      "step": 99350
    },
    {
      "epoch": 14.79818290140006,
      "grad_norm": 7.875,
      "learning_rate": 9.615223951549929e-06,
      "loss": 0.9085,
      "num_input_tokens_seen": 57641400,
      "step": 99355
    },
    {
      "epoch": 14.798927613941018,
      "grad_norm": 8.6875,
      "learning_rate": 9.612662816148974e-06,
      "loss": 1.095,
      "num_input_tokens_seen": 57644216,
      "step": 99360
    },
    {
      "epoch": 14.799672326481979,
      "grad_norm": 18.625,
      "learning_rate": 9.61010194070143e-06,
      "loss": 1.3242,
      "num_input_tokens_seen": 57647224,
      "step": 99365
    },
    {
      "epoch": 14.800417039022937,
      "grad_norm": 22.0,
      "learning_rate": 9.607541325250582e-06,
      "loss": 1.3971,
      "num_input_tokens_seen": 57649944,
      "step": 99370
    },
    {
      "epoch": 14.801161751563896,
      "grad_norm": 9.0625,
      "learning_rate": 9.604980969839672e-06,
      "loss": 1.0133,
      "num_input_tokens_seen": 57652728,
      "step": 99375
    },
    {
      "epoch": 14.801906464104855,
      "grad_norm": 6.125,
      "learning_rate": 9.60242087451197e-06,
      "loss": 0.9364,
      "num_input_tokens_seen": 57655576,
      "step": 99380
    },
    {
      "epoch": 14.802651176645815,
      "grad_norm": 7.75,
      "learning_rate": 9.599861039310709e-06,
      "loss": 1.0105,
      "num_input_tokens_seen": 57658488,
      "step": 99385
    },
    {
      "epoch": 14.803395889186774,
      "grad_norm": 10.6875,
      "learning_rate": 9.597301464279151e-06,
      "loss": 0.8312,
      "num_input_tokens_seen": 57661272,
      "step": 99390
    },
    {
      "epoch": 14.804140601727733,
      "grad_norm": 15.125,
      "learning_rate": 9.59474214946053e-06,
      "loss": 1.0765,
      "num_input_tokens_seen": 57664280,
      "step": 99395
    },
    {
      "epoch": 14.804885314268692,
      "grad_norm": 10.5625,
      "learning_rate": 9.592183094898086e-06,
      "loss": 1.1656,
      "num_input_tokens_seen": 57667032,
      "step": 99400
    },
    {
      "epoch": 14.805630026809652,
      "grad_norm": 13.6875,
      "learning_rate": 9.589624300635047e-06,
      "loss": 0.8257,
      "num_input_tokens_seen": 57669816,
      "step": 99405
    },
    {
      "epoch": 14.80637473935061,
      "grad_norm": 10.3125,
      "learning_rate": 9.587065766714635e-06,
      "loss": 0.9861,
      "num_input_tokens_seen": 57672696,
      "step": 99410
    },
    {
      "epoch": 14.80711945189157,
      "grad_norm": 9.5,
      "learning_rate": 9.584507493180089e-06,
      "loss": 1.19,
      "num_input_tokens_seen": 57675672,
      "step": 99415
    },
    {
      "epoch": 14.807864164432528,
      "grad_norm": 8.1875,
      "learning_rate": 9.581949480074615e-06,
      "loss": 0.8875,
      "num_input_tokens_seen": 57678808,
      "step": 99420
    },
    {
      "epoch": 14.808608876973489,
      "grad_norm": 6.15625,
      "learning_rate": 9.579391727441442e-06,
      "loss": 0.9434,
      "num_input_tokens_seen": 57681688,
      "step": 99425
    },
    {
      "epoch": 14.809353589514448,
      "grad_norm": 12.0625,
      "learning_rate": 9.576834235323773e-06,
      "loss": 1.0365,
      "num_input_tokens_seen": 57684664,
      "step": 99430
    },
    {
      "epoch": 14.810098302055406,
      "grad_norm": 19.625,
      "learning_rate": 9.574277003764807e-06,
      "loss": 1.5611,
      "num_input_tokens_seen": 57687416,
      "step": 99435
    },
    {
      "epoch": 14.810843014596365,
      "grad_norm": 14.0,
      "learning_rate": 9.571720032807758e-06,
      "loss": 1.2405,
      "num_input_tokens_seen": 57690648,
      "step": 99440
    },
    {
      "epoch": 14.811587727137326,
      "grad_norm": 15.4375,
      "learning_rate": 9.569163322495811e-06,
      "loss": 1.0402,
      "num_input_tokens_seen": 57693720,
      "step": 99445
    },
    {
      "epoch": 14.812332439678285,
      "grad_norm": 11.4375,
      "learning_rate": 9.566606872872178e-06,
      "loss": 1.1432,
      "num_input_tokens_seen": 57696664,
      "step": 99450
    },
    {
      "epoch": 14.813077152219243,
      "grad_norm": 17.75,
      "learning_rate": 9.564050683980025e-06,
      "loss": 1.235,
      "num_input_tokens_seen": 57700056,
      "step": 99455
    },
    {
      "epoch": 14.813821864760202,
      "grad_norm": 9.125,
      "learning_rate": 9.561494755862554e-06,
      "loss": 1.296,
      "num_input_tokens_seen": 57702872,
      "step": 99460
    },
    {
      "epoch": 14.814566577301163,
      "grad_norm": 9.625,
      "learning_rate": 9.55893908856294e-06,
      "loss": 1.2419,
      "num_input_tokens_seen": 57705752,
      "step": 99465
    },
    {
      "epoch": 14.815311289842121,
      "grad_norm": 7.0625,
      "learning_rate": 9.55638368212436e-06,
      "loss": 0.9759,
      "num_input_tokens_seen": 57708376,
      "step": 99470
    },
    {
      "epoch": 14.81605600238308,
      "grad_norm": 7.78125,
      "learning_rate": 9.553828536589976e-06,
      "loss": 0.9714,
      "num_input_tokens_seen": 57711448,
      "step": 99475
    },
    {
      "epoch": 14.816800714924039,
      "grad_norm": 11.3125,
      "learning_rate": 9.551273652002955e-06,
      "loss": 0.8276,
      "num_input_tokens_seen": 57714520,
      "step": 99480
    },
    {
      "epoch": 14.817545427465,
      "grad_norm": 6.6875,
      "learning_rate": 9.548719028406472e-06,
      "loss": 0.8901,
      "num_input_tokens_seen": 57717400,
      "step": 99485
    },
    {
      "epoch": 14.818290140005958,
      "grad_norm": 8.0625,
      "learning_rate": 9.546164665843669e-06,
      "loss": 0.9888,
      "num_input_tokens_seen": 57720248,
      "step": 99490
    },
    {
      "epoch": 14.819034852546917,
      "grad_norm": 12.6875,
      "learning_rate": 9.543610564357714e-06,
      "loss": 1.1644,
      "num_input_tokens_seen": 57723480,
      "step": 99495
    },
    {
      "epoch": 14.819779565087876,
      "grad_norm": 7.71875,
      "learning_rate": 9.541056723991739e-06,
      "loss": 1.0472,
      "num_input_tokens_seen": 57726264,
      "step": 99500
    },
    {
      "epoch": 14.820524277628834,
      "grad_norm": 10.125,
      "learning_rate": 9.538503144788914e-06,
      "loss": 1.2343,
      "num_input_tokens_seen": 57729080,
      "step": 99505
    },
    {
      "epoch": 14.821268990169795,
      "grad_norm": 8.25,
      "learning_rate": 9.535949826792358e-06,
      "loss": 1.0152,
      "num_input_tokens_seen": 57731960,
      "step": 99510
    },
    {
      "epoch": 14.822013702710754,
      "grad_norm": 7.125,
      "learning_rate": 9.533396770045208e-06,
      "loss": 0.8561,
      "num_input_tokens_seen": 57734840,
      "step": 99515
    },
    {
      "epoch": 14.822758415251712,
      "grad_norm": 10.25,
      "learning_rate": 9.530843974590606e-06,
      "loss": 1.0576,
      "num_input_tokens_seen": 57737560,
      "step": 99520
    },
    {
      "epoch": 14.823503127792671,
      "grad_norm": 8.875,
      "learning_rate": 9.528291440471665e-06,
      "loss": 0.887,
      "num_input_tokens_seen": 57740184,
      "step": 99525
    },
    {
      "epoch": 14.824247840333632,
      "grad_norm": 11.875,
      "learning_rate": 9.525739167731527e-06,
      "loss": 1.3296,
      "num_input_tokens_seen": 57742936,
      "step": 99530
    },
    {
      "epoch": 14.82499255287459,
      "grad_norm": 10.125,
      "learning_rate": 9.523187156413294e-06,
      "loss": 0.9903,
      "num_input_tokens_seen": 57745912,
      "step": 99535
    },
    {
      "epoch": 14.82573726541555,
      "grad_norm": 14.375,
      "learning_rate": 9.520635406560086e-06,
      "loss": 1.1315,
      "num_input_tokens_seen": 57748728,
      "step": 99540
    },
    {
      "epoch": 14.826481977956508,
      "grad_norm": 15.125,
      "learning_rate": 9.518083918215e-06,
      "loss": 1.3123,
      "num_input_tokens_seen": 57751768,
      "step": 99545
    },
    {
      "epoch": 14.827226690497469,
      "grad_norm": 12.9375,
      "learning_rate": 9.515532691421162e-06,
      "loss": 0.9753,
      "num_input_tokens_seen": 57754552,
      "step": 99550
    },
    {
      "epoch": 14.827971403038427,
      "grad_norm": 18.625,
      "learning_rate": 9.512981726221661e-06,
      "loss": 1.1186,
      "num_input_tokens_seen": 57757592,
      "step": 99555
    },
    {
      "epoch": 14.828716115579386,
      "grad_norm": 12.25,
      "learning_rate": 9.510431022659586e-06,
      "loss": 1.2952,
      "num_input_tokens_seen": 57760248,
      "step": 99560
    },
    {
      "epoch": 14.829460828120345,
      "grad_norm": 9.25,
      "learning_rate": 9.507880580778042e-06,
      "loss": 1.1178,
      "num_input_tokens_seen": 57763128,
      "step": 99565
    },
    {
      "epoch": 14.830205540661305,
      "grad_norm": 8.375,
      "learning_rate": 9.505330400620101e-06,
      "loss": 0.9192,
      "num_input_tokens_seen": 57765784,
      "step": 99570
    },
    {
      "epoch": 14.830950253202264,
      "grad_norm": 7.6875,
      "learning_rate": 9.502780482228866e-06,
      "loss": 0.8282,
      "num_input_tokens_seen": 57768696,
      "step": 99575
    },
    {
      "epoch": 14.831694965743223,
      "grad_norm": 16.125,
      "learning_rate": 9.500230825647394e-06,
      "loss": 1.2694,
      "num_input_tokens_seen": 57771832,
      "step": 99580
    },
    {
      "epoch": 14.832439678284182,
      "grad_norm": 5.03125,
      "learning_rate": 9.497681430918778e-06,
      "loss": 1.0449,
      "num_input_tokens_seen": 57774520,
      "step": 99585
    },
    {
      "epoch": 14.833184390825142,
      "grad_norm": 6.96875,
      "learning_rate": 9.495132298086079e-06,
      "loss": 0.9613,
      "num_input_tokens_seen": 57777400,
      "step": 99590
    },
    {
      "epoch": 14.833929103366101,
      "grad_norm": 13.5,
      "learning_rate": 9.492583427192361e-06,
      "loss": 1.253,
      "num_input_tokens_seen": 57780408,
      "step": 99595
    },
    {
      "epoch": 14.83467381590706,
      "grad_norm": 11.5625,
      "learning_rate": 9.490034818280677e-06,
      "loss": 1.1802,
      "num_input_tokens_seen": 57783512,
      "step": 99600
    },
    {
      "epoch": 14.835418528448018,
      "grad_norm": 6.9375,
      "learning_rate": 9.487486471394096e-06,
      "loss": 1.0804,
      "num_input_tokens_seen": 57786520,
      "step": 99605
    },
    {
      "epoch": 14.836163240988979,
      "grad_norm": 18.75,
      "learning_rate": 9.48493838657567e-06,
      "loss": 0.9579,
      "num_input_tokens_seen": 57789368,
      "step": 99610
    },
    {
      "epoch": 14.836907953529938,
      "grad_norm": 17.25,
      "learning_rate": 9.482390563868429e-06,
      "loss": 0.969,
      "num_input_tokens_seen": 57792216,
      "step": 99615
    },
    {
      "epoch": 14.837652666070897,
      "grad_norm": 8.5,
      "learning_rate": 9.479843003315439e-06,
      "loss": 1.2032,
      "num_input_tokens_seen": 57794904,
      "step": 99620
    },
    {
      "epoch": 14.838397378611855,
      "grad_norm": 15.1875,
      "learning_rate": 9.477295704959718e-06,
      "loss": 1.0827,
      "num_input_tokens_seen": 57797528,
      "step": 99625
    },
    {
      "epoch": 14.839142091152816,
      "grad_norm": 7.1875,
      "learning_rate": 9.474748668844316e-06,
      "loss": 0.8543,
      "num_input_tokens_seen": 57800088,
      "step": 99630
    },
    {
      "epoch": 14.839886803693775,
      "grad_norm": 9.25,
      "learning_rate": 9.47220189501226e-06,
      "loss": 1.2031,
      "num_input_tokens_seen": 57803192,
      "step": 99635
    },
    {
      "epoch": 14.840631516234733,
      "grad_norm": 13.5625,
      "learning_rate": 9.46965538350656e-06,
      "loss": 1.3732,
      "num_input_tokens_seen": 57805816,
      "step": 99640
    },
    {
      "epoch": 14.841376228775692,
      "grad_norm": 11.0,
      "learning_rate": 9.467109134370255e-06,
      "loss": 1.2622,
      "num_input_tokens_seen": 57808920,
      "step": 99645
    },
    {
      "epoch": 14.842120941316653,
      "grad_norm": 8.5625,
      "learning_rate": 9.46456314764635e-06,
      "loss": 0.8135,
      "num_input_tokens_seen": 57811896,
      "step": 99650
    },
    {
      "epoch": 14.842865653857611,
      "grad_norm": 24.875,
      "learning_rate": 9.462017423377867e-06,
      "loss": 1.4877,
      "num_input_tokens_seen": 57814808,
      "step": 99655
    },
    {
      "epoch": 14.84361036639857,
      "grad_norm": 17.875,
      "learning_rate": 9.459471961607808e-06,
      "loss": 1.0609,
      "num_input_tokens_seen": 57817624,
      "step": 99660
    },
    {
      "epoch": 14.844355078939529,
      "grad_norm": 5.375,
      "learning_rate": 9.456926762379175e-06,
      "loss": 1.0715,
      "num_input_tokens_seen": 57820696,
      "step": 99665
    },
    {
      "epoch": 14.84509979148049,
      "grad_norm": 11.8125,
      "learning_rate": 9.45438182573496e-06,
      "loss": 1.1889,
      "num_input_tokens_seen": 57823544,
      "step": 99670
    },
    {
      "epoch": 14.845844504021448,
      "grad_norm": 13.875,
      "learning_rate": 9.451837151718171e-06,
      "loss": 1.2026,
      "num_input_tokens_seen": 57826520,
      "step": 99675
    },
    {
      "epoch": 14.846589216562407,
      "grad_norm": 10.125,
      "learning_rate": 9.449292740371793e-06,
      "loss": 1.2621,
      "num_input_tokens_seen": 57829784,
      "step": 99680
    },
    {
      "epoch": 14.847333929103366,
      "grad_norm": 9.375,
      "learning_rate": 9.4467485917388e-06,
      "loss": 1.1641,
      "num_input_tokens_seen": 57832376,
      "step": 99685
    },
    {
      "epoch": 14.848078641644324,
      "grad_norm": 11.5,
      "learning_rate": 9.444204705862189e-06,
      "loss": 1.0511,
      "num_input_tokens_seen": 57835480,
      "step": 99690
    },
    {
      "epoch": 14.848823354185285,
      "grad_norm": 20.375,
      "learning_rate": 9.441661082784919e-06,
      "loss": 1.2979,
      "num_input_tokens_seen": 57838232,
      "step": 99695
    },
    {
      "epoch": 14.849568066726244,
      "grad_norm": 7.09375,
      "learning_rate": 9.439117722549984e-06,
      "loss": 1.1807,
      "num_input_tokens_seen": 57840824,
      "step": 99700
    },
    {
      "epoch": 14.850312779267203,
      "grad_norm": 8.0,
      "learning_rate": 9.436574625200332e-06,
      "loss": 1.1121,
      "num_input_tokens_seen": 57843544,
      "step": 99705
    },
    {
      "epoch": 14.851057491808161,
      "grad_norm": 15.5,
      "learning_rate": 9.434031790778941e-06,
      "loss": 0.7483,
      "num_input_tokens_seen": 57846520,
      "step": 99710
    },
    {
      "epoch": 14.851802204349122,
      "grad_norm": 17.125,
      "learning_rate": 9.431489219328759e-06,
      "loss": 1.2033,
      "num_input_tokens_seen": 57849528,
      "step": 99715
    },
    {
      "epoch": 14.85254691689008,
      "grad_norm": 10.9375,
      "learning_rate": 9.42894691089274e-06,
      "loss": 0.9788,
      "num_input_tokens_seen": 57852312,
      "step": 99720
    },
    {
      "epoch": 14.85329162943104,
      "grad_norm": 14.6875,
      "learning_rate": 9.426404865513843e-06,
      "loss": 0.9235,
      "num_input_tokens_seen": 57855544,
      "step": 99725
    },
    {
      "epoch": 14.854036341971998,
      "grad_norm": 7.09375,
      "learning_rate": 9.42386308323501e-06,
      "loss": 0.841,
      "num_input_tokens_seen": 57858424,
      "step": 99730
    },
    {
      "epoch": 14.854781054512959,
      "grad_norm": 12.5,
      "learning_rate": 9.421321564099175e-06,
      "loss": 1.0369,
      "num_input_tokens_seen": 57861432,
      "step": 99735
    },
    {
      "epoch": 14.855525767053917,
      "grad_norm": 11.125,
      "learning_rate": 9.418780308149276e-06,
      "loss": 0.9191,
      "num_input_tokens_seen": 57864408,
      "step": 99740
    },
    {
      "epoch": 14.856270479594876,
      "grad_norm": 8.625,
      "learning_rate": 9.416239315428252e-06,
      "loss": 0.9932,
      "num_input_tokens_seen": 57867352,
      "step": 99745
    },
    {
      "epoch": 14.857015192135835,
      "grad_norm": 10.0,
      "learning_rate": 9.413698585979016e-06,
      "loss": 1.128,
      "num_input_tokens_seen": 57870488,
      "step": 99750
    },
    {
      "epoch": 14.857759904676795,
      "grad_norm": 9.4375,
      "learning_rate": 9.411158119844512e-06,
      "loss": 0.9959,
      "num_input_tokens_seen": 57873464,
      "step": 99755
    },
    {
      "epoch": 14.858504617217754,
      "grad_norm": 13.1875,
      "learning_rate": 9.40861791706765e-06,
      "loss": 1.2196,
      "num_input_tokens_seen": 57876344,
      "step": 99760
    },
    {
      "epoch": 14.859249329758713,
      "grad_norm": 5.96875,
      "learning_rate": 9.40607797769133e-06,
      "loss": 0.6924,
      "num_input_tokens_seen": 57878968,
      "step": 99765
    },
    {
      "epoch": 14.859994042299672,
      "grad_norm": 14.5625,
      "learning_rate": 9.403538301758486e-06,
      "loss": 1.2272,
      "num_input_tokens_seen": 57882328,
      "step": 99770
    },
    {
      "epoch": 14.860738754840632,
      "grad_norm": 7.25,
      "learning_rate": 9.400998889311999e-06,
      "loss": 0.9328,
      "num_input_tokens_seen": 57885144,
      "step": 99775
    },
    {
      "epoch": 14.861483467381591,
      "grad_norm": 7.875,
      "learning_rate": 9.398459740394792e-06,
      "loss": 1.1643,
      "num_input_tokens_seen": 57887896,
      "step": 99780
    },
    {
      "epoch": 14.86222817992255,
      "grad_norm": 20.375,
      "learning_rate": 9.395920855049739e-06,
      "loss": 0.9935,
      "num_input_tokens_seen": 57890680,
      "step": 99785
    },
    {
      "epoch": 14.862972892463509,
      "grad_norm": 7.59375,
      "learning_rate": 9.393382233319757e-06,
      "loss": 0.8948,
      "num_input_tokens_seen": 57893464,
      "step": 99790
    },
    {
      "epoch": 14.863717605004469,
      "grad_norm": 12.9375,
      "learning_rate": 9.390843875247717e-06,
      "loss": 1.3724,
      "num_input_tokens_seen": 57896408,
      "step": 99795
    },
    {
      "epoch": 14.864462317545428,
      "grad_norm": 7.53125,
      "learning_rate": 9.388305780876508e-06,
      "loss": 1.2202,
      "num_input_tokens_seen": 57899192,
      "step": 99800
    },
    {
      "epoch": 14.865207030086387,
      "grad_norm": 10.0,
      "learning_rate": 9.385767950249003e-06,
      "loss": 1.29,
      "num_input_tokens_seen": 57902136,
      "step": 99805
    },
    {
      "epoch": 14.865951742627345,
      "grad_norm": 9.0625,
      "learning_rate": 9.383230383408073e-06,
      "loss": 1.1365,
      "num_input_tokens_seen": 57904792,
      "step": 99810
    },
    {
      "epoch": 14.866696455168306,
      "grad_norm": 6.90625,
      "learning_rate": 9.380693080396599e-06,
      "loss": 0.7414,
      "num_input_tokens_seen": 57907992,
      "step": 99815
    },
    {
      "epoch": 14.867441167709265,
      "grad_norm": 13.375,
      "learning_rate": 9.378156041257436e-06,
      "loss": 1.0693,
      "num_input_tokens_seen": 57910840,
      "step": 99820
    },
    {
      "epoch": 14.868185880250223,
      "grad_norm": 13.625,
      "learning_rate": 9.375619266033456e-06,
      "loss": 1.3057,
      "num_input_tokens_seen": 57913400,
      "step": 99825
    },
    {
      "epoch": 14.868930592791182,
      "grad_norm": 7.1875,
      "learning_rate": 9.373082754767497e-06,
      "loss": 0.9069,
      "num_input_tokens_seen": 57916312,
      "step": 99830
    },
    {
      "epoch": 14.86967530533214,
      "grad_norm": 8.625,
      "learning_rate": 9.370546507502433e-06,
      "loss": 0.8551,
      "num_input_tokens_seen": 57919128,
      "step": 99835
    },
    {
      "epoch": 14.870420017873101,
      "grad_norm": 6.5625,
      "learning_rate": 9.368010524281104e-06,
      "loss": 0.9331,
      "num_input_tokens_seen": 57922104,
      "step": 99840
    },
    {
      "epoch": 14.87116473041406,
      "grad_norm": 10.625,
      "learning_rate": 9.365474805146337e-06,
      "loss": 0.9823,
      "num_input_tokens_seen": 57924920,
      "step": 99845
    },
    {
      "epoch": 14.871909442955019,
      "grad_norm": 11.375,
      "learning_rate": 9.362939350140992e-06,
      "loss": 0.8421,
      "num_input_tokens_seen": 57927768,
      "step": 99850
    },
    {
      "epoch": 14.87265415549598,
      "grad_norm": 11.4375,
      "learning_rate": 9.360404159307887e-06,
      "loss": 1.0675,
      "num_input_tokens_seen": 57930488,
      "step": 99855
    },
    {
      "epoch": 14.873398868036938,
      "grad_norm": 13.125,
      "learning_rate": 9.357869232689867e-06,
      "loss": 0.8911,
      "num_input_tokens_seen": 57933048,
      "step": 99860
    },
    {
      "epoch": 14.874143580577897,
      "grad_norm": 9.5625,
      "learning_rate": 9.355334570329746e-06,
      "loss": 1.0587,
      "num_input_tokens_seen": 57935896,
      "step": 99865
    },
    {
      "epoch": 14.874888293118856,
      "grad_norm": 10.5,
      "learning_rate": 9.352800172270352e-06,
      "loss": 1.1679,
      "num_input_tokens_seen": 57938840,
      "step": 99870
    },
    {
      "epoch": 14.875633005659815,
      "grad_norm": 10.875,
      "learning_rate": 9.35026603855449e-06,
      "loss": 0.9204,
      "num_input_tokens_seen": 57941560,
      "step": 99875
    },
    {
      "epoch": 14.876377718200775,
      "grad_norm": 11.625,
      "learning_rate": 9.347732169224972e-06,
      "loss": 1.0896,
      "num_input_tokens_seen": 57944344,
      "step": 99880
    },
    {
      "epoch": 14.877122430741734,
      "grad_norm": 6.28125,
      "learning_rate": 9.345198564324616e-06,
      "loss": 0.7624,
      "num_input_tokens_seen": 57948472,
      "step": 99885
    },
    {
      "epoch": 14.877867143282693,
      "grad_norm": 6.96875,
      "learning_rate": 9.342665223896216e-06,
      "loss": 1.0638,
      "num_input_tokens_seen": 57951352,
      "step": 99890
    },
    {
      "epoch": 14.878611855823651,
      "grad_norm": 13.9375,
      "learning_rate": 9.34013214798258e-06,
      "loss": 0.9402,
      "num_input_tokens_seen": 57954712,
      "step": 99895
    },
    {
      "epoch": 14.879356568364612,
      "grad_norm": 20.125,
      "learning_rate": 9.337599336626488e-06,
      "loss": 1.5517,
      "num_input_tokens_seen": 57957592,
      "step": 99900
    },
    {
      "epoch": 14.88010128090557,
      "grad_norm": 14.375,
      "learning_rate": 9.335066789870741e-06,
      "loss": 1.321,
      "num_input_tokens_seen": 57960632,
      "step": 99905
    },
    {
      "epoch": 14.88084599344653,
      "grad_norm": 7.5625,
      "learning_rate": 9.332534507758114e-06,
      "loss": 1.218,
      "num_input_tokens_seen": 57963768,
      "step": 99910
    },
    {
      "epoch": 14.881590705987488,
      "grad_norm": 10.5625,
      "learning_rate": 9.330002490331402e-06,
      "loss": 1.1644,
      "num_input_tokens_seen": 57966648,
      "step": 99915
    },
    {
      "epoch": 14.882335418528449,
      "grad_norm": 10.4375,
      "learning_rate": 9.32747073763337e-06,
      "loss": 1.335,
      "num_input_tokens_seen": 57969464,
      "step": 99920
    },
    {
      "epoch": 14.883080131069407,
      "grad_norm": 9.3125,
      "learning_rate": 9.324939249706793e-06,
      "loss": 1.0294,
      "num_input_tokens_seen": 57972792,
      "step": 99925
    },
    {
      "epoch": 14.883824843610366,
      "grad_norm": 12.3125,
      "learning_rate": 9.322408026594427e-06,
      "loss": 0.9037,
      "num_input_tokens_seen": 57975896,
      "step": 99930
    },
    {
      "epoch": 14.884569556151325,
      "grad_norm": 13.0625,
      "learning_rate": 9.319877068339051e-06,
      "loss": 1.035,
      "num_input_tokens_seen": 57979096,
      "step": 99935
    },
    {
      "epoch": 14.885314268692285,
      "grad_norm": 16.125,
      "learning_rate": 9.317346374983416e-06,
      "loss": 1.018,
      "num_input_tokens_seen": 57982008,
      "step": 99940
    },
    {
      "epoch": 14.886058981233244,
      "grad_norm": 15.6875,
      "learning_rate": 9.314815946570263e-06,
      "loss": 1.5332,
      "num_input_tokens_seen": 57984568,
      "step": 99945
    },
    {
      "epoch": 14.886803693774203,
      "grad_norm": 15.125,
      "learning_rate": 9.312285783142366e-06,
      "loss": 1.3921,
      "num_input_tokens_seen": 57987512,
      "step": 99950
    },
    {
      "epoch": 14.887548406315162,
      "grad_norm": 12.625,
      "learning_rate": 9.309755884742455e-06,
      "loss": 0.9853,
      "num_input_tokens_seen": 57990264,
      "step": 99955
    },
    {
      "epoch": 14.888293118856122,
      "grad_norm": 9.3125,
      "learning_rate": 9.307226251413262e-06,
      "loss": 1.2425,
      "num_input_tokens_seen": 57993272,
      "step": 99960
    },
    {
      "epoch": 14.889037831397081,
      "grad_norm": 6.0625,
      "learning_rate": 9.304696883197542e-06,
      "loss": 0.8066,
      "num_input_tokens_seen": 57996216,
      "step": 99965
    },
    {
      "epoch": 14.88978254393804,
      "grad_norm": 10.5625,
      "learning_rate": 9.302167780138005e-06,
      "loss": 1.147,
      "num_input_tokens_seen": 57999160,
      "step": 99970
    },
    {
      "epoch": 14.890527256478999,
      "grad_norm": 8.8125,
      "learning_rate": 9.2996389422774e-06,
      "loss": 1.2443,
      "num_input_tokens_seen": 58002040,
      "step": 99975
    },
    {
      "epoch": 14.891271969019959,
      "grad_norm": 11.875,
      "learning_rate": 9.297110369658426e-06,
      "loss": 1.1492,
      "num_input_tokens_seen": 58004824,
      "step": 99980
    },
    {
      "epoch": 14.892016681560918,
      "grad_norm": 7.875,
      "learning_rate": 9.294582062323825e-06,
      "loss": 1.2726,
      "num_input_tokens_seen": 58007672,
      "step": 99985
    },
    {
      "epoch": 14.892761394101877,
      "grad_norm": 14.5625,
      "learning_rate": 9.292054020316297e-06,
      "loss": 0.94,
      "num_input_tokens_seen": 58010808,
      "step": 99990
    },
    {
      "epoch": 14.893506106642835,
      "grad_norm": 7.96875,
      "learning_rate": 9.28952624367855e-06,
      "loss": 1.1131,
      "num_input_tokens_seen": 58013912,
      "step": 99995
    },
    {
      "epoch": 14.894250819183796,
      "grad_norm": 8.5,
      "learning_rate": 9.286998732453292e-06,
      "loss": 0.9431,
      "num_input_tokens_seen": 58016824,
      "step": 100000
    },
    {
      "epoch": 14.894995531724755,
      "grad_norm": 8.875,
      "learning_rate": 9.28447148668321e-06,
      "loss": 1.2424,
      "num_input_tokens_seen": 58019832,
      "step": 100005
    },
    {
      "epoch": 14.895740244265713,
      "grad_norm": 14.25,
      "learning_rate": 9.28194450641102e-06,
      "loss": 1.198,
      "num_input_tokens_seen": 58022552,
      "step": 100010
    },
    {
      "epoch": 14.896484956806672,
      "grad_norm": 9.3125,
      "learning_rate": 9.27941779167939e-06,
      "loss": 0.9956,
      "num_input_tokens_seen": 58025432,
      "step": 100015
    },
    {
      "epoch": 14.897229669347631,
      "grad_norm": 5.90625,
      "learning_rate": 9.27689134253103e-06,
      "loss": 0.8588,
      "num_input_tokens_seen": 58028376,
      "step": 100020
    },
    {
      "epoch": 14.897974381888591,
      "grad_norm": 11.6875,
      "learning_rate": 9.274365159008602e-06,
      "loss": 1.1311,
      "num_input_tokens_seen": 58031320,
      "step": 100025
    },
    {
      "epoch": 14.89871909442955,
      "grad_norm": 21.125,
      "learning_rate": 9.2718392411548e-06,
      "loss": 1.5189,
      "num_input_tokens_seen": 58034232,
      "step": 100030
    },
    {
      "epoch": 14.899463806970509,
      "grad_norm": 9.0625,
      "learning_rate": 9.26931358901229e-06,
      "loss": 1.2591,
      "num_input_tokens_seen": 58037240,
      "step": 100035
    },
    {
      "epoch": 14.900208519511468,
      "grad_norm": 11.6875,
      "learning_rate": 9.26678820262373e-06,
      "loss": 1.0573,
      "num_input_tokens_seen": 58040088,
      "step": 100040
    },
    {
      "epoch": 14.900953232052428,
      "grad_norm": 12.6875,
      "learning_rate": 9.2642630820318e-06,
      "loss": 1.1639,
      "num_input_tokens_seen": 58042680,
      "step": 100045
    },
    {
      "epoch": 14.901697944593387,
      "grad_norm": 10.625,
      "learning_rate": 9.261738227279144e-06,
      "loss": 1.0127,
      "num_input_tokens_seen": 58045400,
      "step": 100050
    },
    {
      "epoch": 14.902442657134346,
      "grad_norm": 7.625,
      "learning_rate": 9.259213638408434e-06,
      "loss": 1.0178,
      "num_input_tokens_seen": 58048216,
      "step": 100055
    },
    {
      "epoch": 14.903187369675305,
      "grad_norm": 8.6875,
      "learning_rate": 9.25668931546231e-06,
      "loss": 1.296,
      "num_input_tokens_seen": 58051064,
      "step": 100060
    },
    {
      "epoch": 14.903932082216265,
      "grad_norm": 8.3125,
      "learning_rate": 9.254165258483421e-06,
      "loss": 1.0167,
      "num_input_tokens_seen": 58053848,
      "step": 100065
    },
    {
      "epoch": 14.904676794757224,
      "grad_norm": 14.3125,
      "learning_rate": 9.251641467514399e-06,
      "loss": 1.1311,
      "num_input_tokens_seen": 58057144,
      "step": 100070
    },
    {
      "epoch": 14.905421507298183,
      "grad_norm": 11.5625,
      "learning_rate": 9.249117942597895e-06,
      "loss": 1.1725,
      "num_input_tokens_seen": 58060344,
      "step": 100075
    },
    {
      "epoch": 14.906166219839141,
      "grad_norm": 5.875,
      "learning_rate": 9.246594683776536e-06,
      "loss": 0.9426,
      "num_input_tokens_seen": 58063448,
      "step": 100080
    },
    {
      "epoch": 14.906910932380102,
      "grad_norm": 7.625,
      "learning_rate": 9.244071691092937e-06,
      "loss": 1.1056,
      "num_input_tokens_seen": 58066488,
      "step": 100085
    },
    {
      "epoch": 14.90765564492106,
      "grad_norm": 5.15625,
      "learning_rate": 9.241548964589747e-06,
      "loss": 1.1802,
      "num_input_tokens_seen": 58069592,
      "step": 100090
    },
    {
      "epoch": 14.90840035746202,
      "grad_norm": 9.6875,
      "learning_rate": 9.239026504309558e-06,
      "loss": 0.9502,
      "num_input_tokens_seen": 58072504,
      "step": 100095
    },
    {
      "epoch": 14.909145070002978,
      "grad_norm": 6.0,
      "learning_rate": 9.236504310295007e-06,
      "loss": 0.9064,
      "num_input_tokens_seen": 58075384,
      "step": 100100
    },
    {
      "epoch": 14.909889782543939,
      "grad_norm": 10.875,
      "learning_rate": 9.233982382588688e-06,
      "loss": 0.9852,
      "num_input_tokens_seen": 58078104,
      "step": 100105
    },
    {
      "epoch": 14.910634495084897,
      "grad_norm": 6.28125,
      "learning_rate": 9.23146072123322e-06,
      "loss": 0.8488,
      "num_input_tokens_seen": 58081048,
      "step": 100110
    },
    {
      "epoch": 14.911379207625856,
      "grad_norm": 7.625,
      "learning_rate": 9.228939326271197e-06,
      "loss": 0.8146,
      "num_input_tokens_seen": 58084056,
      "step": 100115
    },
    {
      "epoch": 14.912123920166815,
      "grad_norm": 13.75,
      "learning_rate": 9.226418197745206e-06,
      "loss": 1.0994,
      "num_input_tokens_seen": 58086840,
      "step": 100120
    },
    {
      "epoch": 14.912868632707776,
      "grad_norm": 10.8125,
      "learning_rate": 9.223897335697856e-06,
      "loss": 1.1579,
      "num_input_tokens_seen": 58089656,
      "step": 100125
    },
    {
      "epoch": 14.913613345248734,
      "grad_norm": 8.125,
      "learning_rate": 9.221376740171727e-06,
      "loss": 1.2197,
      "num_input_tokens_seen": 58092440,
      "step": 100130
    },
    {
      "epoch": 14.914358057789693,
      "grad_norm": 7.59375,
      "learning_rate": 9.2188564112094e-06,
      "loss": 0.9214,
      "num_input_tokens_seen": 58095352,
      "step": 100135
    },
    {
      "epoch": 14.915102770330652,
      "grad_norm": 9.75,
      "learning_rate": 9.216336348853449e-06,
      "loss": 1.2012,
      "num_input_tokens_seen": 58098328,
      "step": 100140
    },
    {
      "epoch": 14.915847482871612,
      "grad_norm": 7.71875,
      "learning_rate": 9.213816553146462e-06,
      "loss": 0.7674,
      "num_input_tokens_seen": 58101208,
      "step": 100145
    },
    {
      "epoch": 14.916592195412571,
      "grad_norm": 17.5,
      "learning_rate": 9.211297024130989e-06,
      "loss": 0.96,
      "num_input_tokens_seen": 58104120,
      "step": 100150
    },
    {
      "epoch": 14.91733690795353,
      "grad_norm": 14.9375,
      "learning_rate": 9.208777761849616e-06,
      "loss": 1.4272,
      "num_input_tokens_seen": 58107448,
      "step": 100155
    },
    {
      "epoch": 14.918081620494489,
      "grad_norm": 16.625,
      "learning_rate": 9.20625876634489e-06,
      "loss": 1.2717,
      "num_input_tokens_seen": 58110296,
      "step": 100160
    },
    {
      "epoch": 14.91882633303545,
      "grad_norm": 17.75,
      "learning_rate": 9.203740037659367e-06,
      "loss": 1.2396,
      "num_input_tokens_seen": 58112984,
      "step": 100165
    },
    {
      "epoch": 14.919571045576408,
      "grad_norm": 15.4375,
      "learning_rate": 9.201221575835608e-06,
      "loss": 1.143,
      "num_input_tokens_seen": 58116120,
      "step": 100170
    },
    {
      "epoch": 14.920315758117367,
      "grad_norm": 16.75,
      "learning_rate": 9.198703380916143e-06,
      "loss": 1.2302,
      "num_input_tokens_seen": 58118872,
      "step": 100175
    },
    {
      "epoch": 14.921060470658325,
      "grad_norm": 13.75,
      "learning_rate": 9.196185452943534e-06,
      "loss": 1.1152,
      "num_input_tokens_seen": 58121720,
      "step": 100180
    },
    {
      "epoch": 14.921805183199286,
      "grad_norm": 7.46875,
      "learning_rate": 9.193667791960303e-06,
      "loss": 1.0729,
      "num_input_tokens_seen": 58124632,
      "step": 100185
    },
    {
      "epoch": 14.922549895740245,
      "grad_norm": 15.4375,
      "learning_rate": 9.191150398008996e-06,
      "loss": 1.0422,
      "num_input_tokens_seen": 58127512,
      "step": 100190
    },
    {
      "epoch": 14.923294608281203,
      "grad_norm": 5.8125,
      "learning_rate": 9.188633271132135e-06,
      "loss": 0.8389,
      "num_input_tokens_seen": 58130488,
      "step": 100195
    },
    {
      "epoch": 14.924039320822162,
      "grad_norm": 12.3125,
      "learning_rate": 9.186116411372248e-06,
      "loss": 1.347,
      "num_input_tokens_seen": 58133720,
      "step": 100200
    },
    {
      "epoch": 14.924784033363121,
      "grad_norm": 19.125,
      "learning_rate": 9.183599818771849e-06,
      "loss": 1.0953,
      "num_input_tokens_seen": 58136408,
      "step": 100205
    },
    {
      "epoch": 14.925528745904082,
      "grad_norm": 13.0,
      "learning_rate": 9.181083493373449e-06,
      "loss": 0.9148,
      "num_input_tokens_seen": 58139480,
      "step": 100210
    },
    {
      "epoch": 14.92627345844504,
      "grad_norm": 8.375,
      "learning_rate": 9.178567435219574e-06,
      "loss": 1.0046,
      "num_input_tokens_seen": 58142424,
      "step": 100215
    },
    {
      "epoch": 14.927018170985999,
      "grad_norm": 30.625,
      "learning_rate": 9.176051644352713e-06,
      "loss": 1.2432,
      "num_input_tokens_seen": 58145176,
      "step": 100220
    },
    {
      "epoch": 14.927762883526958,
      "grad_norm": 10.5625,
      "learning_rate": 9.173536120815385e-06,
      "loss": 1.3646,
      "num_input_tokens_seen": 58147800,
      "step": 100225
    },
    {
      "epoch": 14.928507596067918,
      "grad_norm": 10.375,
      "learning_rate": 9.171020864650071e-06,
      "loss": 0.9122,
      "num_input_tokens_seen": 58150648,
      "step": 100230
    },
    {
      "epoch": 14.929252308608877,
      "grad_norm": 10.625,
      "learning_rate": 9.16850587589928e-06,
      "loss": 1.4035,
      "num_input_tokens_seen": 58153432,
      "step": 100235
    },
    {
      "epoch": 14.929997021149836,
      "grad_norm": 10.8125,
      "learning_rate": 9.16599115460549e-06,
      "loss": 0.8332,
      "num_input_tokens_seen": 58156280,
      "step": 100240
    },
    {
      "epoch": 14.930741733690795,
      "grad_norm": 9.0625,
      "learning_rate": 9.16347670081118e-06,
      "loss": 1.2002,
      "num_input_tokens_seen": 58158872,
      "step": 100245
    },
    {
      "epoch": 14.931486446231755,
      "grad_norm": 14.3125,
      "learning_rate": 9.160962514558843e-06,
      "loss": 1.3466,
      "num_input_tokens_seen": 58161976,
      "step": 100250
    },
    {
      "epoch": 14.932231158772714,
      "grad_norm": 7.8125,
      "learning_rate": 9.158448595890948e-06,
      "loss": 1.1993,
      "num_input_tokens_seen": 58164664,
      "step": 100255
    },
    {
      "epoch": 14.932975871313673,
      "grad_norm": 16.875,
      "learning_rate": 9.155934944849953e-06,
      "loss": 1.1561,
      "num_input_tokens_seen": 58167672,
      "step": 100260
    },
    {
      "epoch": 14.933720583854631,
      "grad_norm": 14.125,
      "learning_rate": 9.153421561478346e-06,
      "loss": 1.1514,
      "num_input_tokens_seen": 58170584,
      "step": 100265
    },
    {
      "epoch": 14.934465296395592,
      "grad_norm": 10.6875,
      "learning_rate": 9.150908445818571e-06,
      "loss": 1.1721,
      "num_input_tokens_seen": 58173528,
      "step": 100270
    },
    {
      "epoch": 14.93521000893655,
      "grad_norm": 15.875,
      "learning_rate": 9.148395597913085e-06,
      "loss": 1.2163,
      "num_input_tokens_seen": 58176728,
      "step": 100275
    },
    {
      "epoch": 14.93595472147751,
      "grad_norm": 9.1875,
      "learning_rate": 9.14588301780435e-06,
      "loss": 0.8071,
      "num_input_tokens_seen": 58179480,
      "step": 100280
    },
    {
      "epoch": 14.936699434018468,
      "grad_norm": 11.0625,
      "learning_rate": 9.14337070553481e-06,
      "loss": 0.9233,
      "num_input_tokens_seen": 58182392,
      "step": 100285
    },
    {
      "epoch": 14.937444146559429,
      "grad_norm": 3.421875,
      "learning_rate": 9.140858661146897e-06,
      "loss": 0.6615,
      "num_input_tokens_seen": 58185272,
      "step": 100290
    },
    {
      "epoch": 14.938188859100388,
      "grad_norm": 7.9375,
      "learning_rate": 9.138346884683066e-06,
      "loss": 1.059,
      "num_input_tokens_seen": 58188344,
      "step": 100295
    },
    {
      "epoch": 14.938933571641346,
      "grad_norm": 12.125,
      "learning_rate": 9.135835376185737e-06,
      "loss": 1.1285,
      "num_input_tokens_seen": 58191256,
      "step": 100300
    },
    {
      "epoch": 14.939678284182305,
      "grad_norm": 9.1875,
      "learning_rate": 9.133324135697351e-06,
      "loss": 1.1652,
      "num_input_tokens_seen": 58194456,
      "step": 100305
    },
    {
      "epoch": 14.940422996723266,
      "grad_norm": 9.0,
      "learning_rate": 9.130813163260321e-06,
      "loss": 1.0038,
      "num_input_tokens_seen": 58197464,
      "step": 100310
    },
    {
      "epoch": 14.941167709264224,
      "grad_norm": 7.125,
      "learning_rate": 9.128302458917081e-06,
      "loss": 0.8837,
      "num_input_tokens_seen": 58200152,
      "step": 100315
    },
    {
      "epoch": 14.941912421805183,
      "grad_norm": 16.75,
      "learning_rate": 9.125792022710042e-06,
      "loss": 0.7903,
      "num_input_tokens_seen": 58202936,
      "step": 100320
    },
    {
      "epoch": 14.942657134346142,
      "grad_norm": 15.375,
      "learning_rate": 9.123281854681612e-06,
      "loss": 1.5148,
      "num_input_tokens_seen": 58205880,
      "step": 100325
    },
    {
      "epoch": 14.943401846887102,
      "grad_norm": 9.25,
      "learning_rate": 9.120771954874199e-06,
      "loss": 0.943,
      "num_input_tokens_seen": 58208792,
      "step": 100330
    },
    {
      "epoch": 14.944146559428061,
      "grad_norm": 12.8125,
      "learning_rate": 9.118262323330196e-06,
      "loss": 1.0106,
      "num_input_tokens_seen": 58211576,
      "step": 100335
    },
    {
      "epoch": 14.94489127196902,
      "grad_norm": 12.25,
      "learning_rate": 9.115752960092017e-06,
      "loss": 1.0288,
      "num_input_tokens_seen": 58214552,
      "step": 100340
    },
    {
      "epoch": 14.945635984509979,
      "grad_norm": 12.9375,
      "learning_rate": 9.11324386520204e-06,
      "loss": 1.1802,
      "num_input_tokens_seen": 58217272,
      "step": 100345
    },
    {
      "epoch": 14.946380697050937,
      "grad_norm": 21.375,
      "learning_rate": 9.11073503870267e-06,
      "loss": 1.1729,
      "num_input_tokens_seen": 58220216,
      "step": 100350
    },
    {
      "epoch": 14.947125409591898,
      "grad_norm": 17.0,
      "learning_rate": 9.108226480636276e-06,
      "loss": 1.1001,
      "num_input_tokens_seen": 58223160,
      "step": 100355
    },
    {
      "epoch": 14.947870122132857,
      "grad_norm": 9.75,
      "learning_rate": 9.105718191045248e-06,
      "loss": 1.4759,
      "num_input_tokens_seen": 58225848,
      "step": 100360
    },
    {
      "epoch": 14.948614834673815,
      "grad_norm": 7.40625,
      "learning_rate": 9.10321016997196e-06,
      "loss": 0.9889,
      "num_input_tokens_seen": 58228664,
      "step": 100365
    },
    {
      "epoch": 14.949359547214776,
      "grad_norm": 14.375,
      "learning_rate": 9.10070241745877e-06,
      "loss": 1.1607,
      "num_input_tokens_seen": 58231576,
      "step": 100370
    },
    {
      "epoch": 14.950104259755735,
      "grad_norm": 10.625,
      "learning_rate": 9.098194933548063e-06,
      "loss": 1.1615,
      "num_input_tokens_seen": 58234680,
      "step": 100375
    },
    {
      "epoch": 14.950848972296694,
      "grad_norm": 6.34375,
      "learning_rate": 9.09568771828218e-06,
      "loss": 0.8728,
      "num_input_tokens_seen": 58237464,
      "step": 100380
    },
    {
      "epoch": 14.951593684837652,
      "grad_norm": 5.6875,
      "learning_rate": 9.0931807717035e-06,
      "loss": 0.8792,
      "num_input_tokens_seen": 58240472,
      "step": 100385
    },
    {
      "epoch": 14.952338397378611,
      "grad_norm": 13.3125,
      "learning_rate": 9.090674093854362e-06,
      "loss": 1.2276,
      "num_input_tokens_seen": 58243160,
      "step": 100390
    },
    {
      "epoch": 14.953083109919572,
      "grad_norm": 6.5625,
      "learning_rate": 9.088167684777115e-06,
      "loss": 1.2235,
      "num_input_tokens_seen": 58245880,
      "step": 100395
    },
    {
      "epoch": 14.95382782246053,
      "grad_norm": 7.46875,
      "learning_rate": 9.085661544514104e-06,
      "loss": 0.9958,
      "num_input_tokens_seen": 58248888,
      "step": 100400
    },
    {
      "epoch": 14.954572535001489,
      "grad_norm": 7.15625,
      "learning_rate": 9.083155673107657e-06,
      "loss": 1.2492,
      "num_input_tokens_seen": 58251672,
      "step": 100405
    },
    {
      "epoch": 14.955317247542448,
      "grad_norm": 12.1875,
      "learning_rate": 9.080650070600128e-06,
      "loss": 0.908,
      "num_input_tokens_seen": 58254936,
      "step": 100410
    },
    {
      "epoch": 14.956061960083408,
      "grad_norm": 12.0,
      "learning_rate": 9.078144737033827e-06,
      "loss": 0.8698,
      "num_input_tokens_seen": 58257624,
      "step": 100415
    },
    {
      "epoch": 14.956806672624367,
      "grad_norm": 13.4375,
      "learning_rate": 9.075639672451097e-06,
      "loss": 1.2631,
      "num_input_tokens_seen": 58260568,
      "step": 100420
    },
    {
      "epoch": 14.957551385165326,
      "grad_norm": 11.5,
      "learning_rate": 9.073134876894241e-06,
      "loss": 1.0772,
      "num_input_tokens_seen": 58263800,
      "step": 100425
    },
    {
      "epoch": 14.958296097706285,
      "grad_norm": 11.3125,
      "learning_rate": 9.070630350405593e-06,
      "loss": 0.857,
      "num_input_tokens_seen": 58266584,
      "step": 100430
    },
    {
      "epoch": 14.959040810247245,
      "grad_norm": 10.4375,
      "learning_rate": 9.068126093027447e-06,
      "loss": 1.139,
      "num_input_tokens_seen": 58269624,
      "step": 100435
    },
    {
      "epoch": 14.959785522788204,
      "grad_norm": 11.9375,
      "learning_rate": 9.065622104802126e-06,
      "loss": 0.8595,
      "num_input_tokens_seen": 58272600,
      "step": 100440
    },
    {
      "epoch": 14.960530235329163,
      "grad_norm": 7.34375,
      "learning_rate": 9.063118385771924e-06,
      "loss": 1.0498,
      "num_input_tokens_seen": 58275352,
      "step": 100445
    },
    {
      "epoch": 14.961274947870121,
      "grad_norm": 6.34375,
      "learning_rate": 9.060614935979131e-06,
      "loss": 0.8664,
      "num_input_tokens_seen": 58278136,
      "step": 100450
    },
    {
      "epoch": 14.962019660411082,
      "grad_norm": 10.0,
      "learning_rate": 9.058111755466059e-06,
      "loss": 1.2649,
      "num_input_tokens_seen": 58281080,
      "step": 100455
    },
    {
      "epoch": 14.96276437295204,
      "grad_norm": 17.0,
      "learning_rate": 9.055608844274985e-06,
      "loss": 1.504,
      "num_input_tokens_seen": 58284632,
      "step": 100460
    },
    {
      "epoch": 14.963509085493,
      "grad_norm": 6.40625,
      "learning_rate": 9.053106202448194e-06,
      "loss": 0.8765,
      "num_input_tokens_seen": 58287544,
      "step": 100465
    },
    {
      "epoch": 14.964253798033958,
      "grad_norm": 12.25,
      "learning_rate": 9.050603830027959e-06,
      "loss": 1.0953,
      "num_input_tokens_seen": 58290232,
      "step": 100470
    },
    {
      "epoch": 14.964998510574919,
      "grad_norm": 14.5,
      "learning_rate": 9.048101727056568e-06,
      "loss": 1.1752,
      "num_input_tokens_seen": 58292984,
      "step": 100475
    },
    {
      "epoch": 14.965743223115878,
      "grad_norm": 8.875,
      "learning_rate": 9.045599893576287e-06,
      "loss": 1.1432,
      "num_input_tokens_seen": 58295896,
      "step": 100480
    },
    {
      "epoch": 14.966487935656836,
      "grad_norm": 7.0625,
      "learning_rate": 9.043098329629374e-06,
      "loss": 1.0212,
      "num_input_tokens_seen": 58298680,
      "step": 100485
    },
    {
      "epoch": 14.967232648197795,
      "grad_norm": 8.0,
      "learning_rate": 9.040597035258103e-06,
      "loss": 0.7812,
      "num_input_tokens_seen": 58301592,
      "step": 100490
    },
    {
      "epoch": 14.967977360738756,
      "grad_norm": 5.375,
      "learning_rate": 9.038096010504714e-06,
      "loss": 1.2852,
      "num_input_tokens_seen": 58304696,
      "step": 100495
    },
    {
      "epoch": 14.968722073279714,
      "grad_norm": 9.3125,
      "learning_rate": 9.035595255411482e-06,
      "loss": 1.2412,
      "num_input_tokens_seen": 58307736,
      "step": 100500
    },
    {
      "epoch": 14.969466785820673,
      "grad_norm": 6.75,
      "learning_rate": 9.033094770020634e-06,
      "loss": 0.919,
      "num_input_tokens_seen": 58310712,
      "step": 100505
    },
    {
      "epoch": 14.970211498361632,
      "grad_norm": 17.875,
      "learning_rate": 9.03059455437443e-06,
      "loss": 1.2726,
      "num_input_tokens_seen": 58313784,
      "step": 100510
    },
    {
      "epoch": 14.970956210902592,
      "grad_norm": 7.0625,
      "learning_rate": 9.028094608515093e-06,
      "loss": 0.9024,
      "num_input_tokens_seen": 58316696,
      "step": 100515
    },
    {
      "epoch": 14.971700923443551,
      "grad_norm": 9.75,
      "learning_rate": 9.02559493248487e-06,
      "loss": 0.9558,
      "num_input_tokens_seen": 58319448,
      "step": 100520
    },
    {
      "epoch": 14.97244563598451,
      "grad_norm": 12.75,
      "learning_rate": 9.023095526325987e-06,
      "loss": 0.8698,
      "num_input_tokens_seen": 58322392,
      "step": 100525
    },
    {
      "epoch": 14.973190348525469,
      "grad_norm": 12.75,
      "learning_rate": 9.020596390080665e-06,
      "loss": 1.2391,
      "num_input_tokens_seen": 58325400,
      "step": 100530
    },
    {
      "epoch": 14.973935061066427,
      "grad_norm": 4.96875,
      "learning_rate": 9.018097523791127e-06,
      "loss": 1.2643,
      "num_input_tokens_seen": 58328184,
      "step": 100535
    },
    {
      "epoch": 14.974679773607388,
      "grad_norm": 7.875,
      "learning_rate": 9.01559892749958e-06,
      "loss": 0.9462,
      "num_input_tokens_seen": 58331032,
      "step": 100540
    },
    {
      "epoch": 14.975424486148347,
      "grad_norm": 13.9375,
      "learning_rate": 9.013100601248254e-06,
      "loss": 1.0657,
      "num_input_tokens_seen": 58333592,
      "step": 100545
    },
    {
      "epoch": 14.976169198689306,
      "grad_norm": 15.8125,
      "learning_rate": 9.010602545079332e-06,
      "loss": 0.8397,
      "num_input_tokens_seen": 58336184,
      "step": 100550
    },
    {
      "epoch": 14.976913911230266,
      "grad_norm": 7.6875,
      "learning_rate": 9.00810475903504e-06,
      "loss": 0.8799,
      "num_input_tokens_seen": 58339480,
      "step": 100555
    },
    {
      "epoch": 14.977658623771225,
      "grad_norm": 8.0625,
      "learning_rate": 9.005607243157565e-06,
      "loss": 1.1123,
      "num_input_tokens_seen": 58342232,
      "step": 100560
    },
    {
      "epoch": 14.978403336312184,
      "grad_norm": 10.9375,
      "learning_rate": 9.003109997489092e-06,
      "loss": 1.0876,
      "num_input_tokens_seen": 58344888,
      "step": 100565
    },
    {
      "epoch": 14.979148048853142,
      "grad_norm": 9.8125,
      "learning_rate": 9.000613022071824e-06,
      "loss": 1.067,
      "num_input_tokens_seen": 58347896,
      "step": 100570
    },
    {
      "epoch": 14.979892761394101,
      "grad_norm": 17.375,
      "learning_rate": 8.99811631694793e-06,
      "loss": 1.126,
      "num_input_tokens_seen": 58350616,
      "step": 100575
    },
    {
      "epoch": 14.980637473935062,
      "grad_norm": 9.375,
      "learning_rate": 8.995619882159606e-06,
      "loss": 1.1457,
      "num_input_tokens_seen": 58353464,
      "step": 100580
    },
    {
      "epoch": 14.98138218647602,
      "grad_norm": 9.0625,
      "learning_rate": 8.993123717749016e-06,
      "loss": 1.0936,
      "num_input_tokens_seen": 58356600,
      "step": 100585
    },
    {
      "epoch": 14.98212689901698,
      "grad_norm": 5.09375,
      "learning_rate": 8.990627823758327e-06,
      "loss": 0.8971,
      "num_input_tokens_seen": 58359448,
      "step": 100590
    },
    {
      "epoch": 14.982871611557938,
      "grad_norm": 7.40625,
      "learning_rate": 8.988132200229716e-06,
      "loss": 1.0958,
      "num_input_tokens_seen": 58362328,
      "step": 100595
    },
    {
      "epoch": 14.983616324098898,
      "grad_norm": 12.625,
      "learning_rate": 8.985636847205336e-06,
      "loss": 1.139,
      "num_input_tokens_seen": 58365400,
      "step": 100600
    },
    {
      "epoch": 14.984361036639857,
      "grad_norm": 11.125,
      "learning_rate": 8.983141764727348e-06,
      "loss": 1.0311,
      "num_input_tokens_seen": 58368376,
      "step": 100605
    },
    {
      "epoch": 14.985105749180816,
      "grad_norm": 4.28125,
      "learning_rate": 8.980646952837894e-06,
      "loss": 0.9747,
      "num_input_tokens_seen": 58371160,
      "step": 100610
    },
    {
      "epoch": 14.985850461721775,
      "grad_norm": 19.5,
      "learning_rate": 8.978152411579133e-06,
      "loss": 1.2316,
      "num_input_tokens_seen": 58374008,
      "step": 100615
    },
    {
      "epoch": 14.986595174262735,
      "grad_norm": 8.6875,
      "learning_rate": 8.975658140993196e-06,
      "loss": 0.9567,
      "num_input_tokens_seen": 58376952,
      "step": 100620
    },
    {
      "epoch": 14.987339886803694,
      "grad_norm": 6.0625,
      "learning_rate": 8.973164141122237e-06,
      "loss": 0.8514,
      "num_input_tokens_seen": 58379736,
      "step": 100625
    },
    {
      "epoch": 14.988084599344653,
      "grad_norm": 12.5,
      "learning_rate": 8.970670412008372e-06,
      "loss": 1.0921,
      "num_input_tokens_seen": 58382840,
      "step": 100630
    },
    {
      "epoch": 14.988829311885612,
      "grad_norm": 4.09375,
      "learning_rate": 8.96817695369375e-06,
      "loss": 1.1086,
      "num_input_tokens_seen": 58385688,
      "step": 100635
    },
    {
      "epoch": 14.989574024426572,
      "grad_norm": 7.03125,
      "learning_rate": 8.965683766220481e-06,
      "loss": 0.8483,
      "num_input_tokens_seen": 58388600,
      "step": 100640
    },
    {
      "epoch": 14.99031873696753,
      "grad_norm": 18.75,
      "learning_rate": 8.963190849630682e-06,
      "loss": 0.9201,
      "num_input_tokens_seen": 58391384,
      "step": 100645
    },
    {
      "epoch": 14.99106344950849,
      "grad_norm": 7.34375,
      "learning_rate": 8.96069820396648e-06,
      "loss": 1.3151,
      "num_input_tokens_seen": 58394264,
      "step": 100650
    },
    {
      "epoch": 14.991808162049448,
      "grad_norm": 6.6875,
      "learning_rate": 8.958205829269984e-06,
      "loss": 1.0979,
      "num_input_tokens_seen": 58397240,
      "step": 100655
    },
    {
      "epoch": 14.992552874590409,
      "grad_norm": 11.75,
      "learning_rate": 8.955713725583295e-06,
      "loss": 1.3999,
      "num_input_tokens_seen": 58400024,
      "step": 100660
    },
    {
      "epoch": 14.993297587131368,
      "grad_norm": 5.25,
      "learning_rate": 8.953221892948508e-06,
      "loss": 1.2042,
      "num_input_tokens_seen": 58403160,
      "step": 100665
    },
    {
      "epoch": 14.994042299672326,
      "grad_norm": 7.0625,
      "learning_rate": 8.950730331407733e-06,
      "loss": 0.8485,
      "num_input_tokens_seen": 58406104,
      "step": 100670
    },
    {
      "epoch": 14.994787012213285,
      "grad_norm": 12.875,
      "learning_rate": 8.94823904100305e-06,
      "loss": 1.1381,
      "num_input_tokens_seen": 58408920,
      "step": 100675
    },
    {
      "epoch": 14.995531724754246,
      "grad_norm": 8.4375,
      "learning_rate": 8.945748021776564e-06,
      "loss": 0.961,
      "num_input_tokens_seen": 58411896,
      "step": 100680
    },
    {
      "epoch": 14.996276437295204,
      "grad_norm": 8.125,
      "learning_rate": 8.943257273770351e-06,
      "loss": 0.9291,
      "num_input_tokens_seen": 58414744,
      "step": 100685
    },
    {
      "epoch": 14.997021149836163,
      "grad_norm": 10.625,
      "learning_rate": 8.940766797026476e-06,
      "loss": 0.8563,
      "num_input_tokens_seen": 58417656,
      "step": 100690
    },
    {
      "epoch": 14.997765862377122,
      "grad_norm": 24.875,
      "learning_rate": 8.938276591587031e-06,
      "loss": 0.9748,
      "num_input_tokens_seen": 58420632,
      "step": 100695
    },
    {
      "epoch": 14.998510574918082,
      "grad_norm": 9.5,
      "learning_rate": 8.935786657494072e-06,
      "loss": 1.0324,
      "num_input_tokens_seen": 58423672,
      "step": 100700
    },
    {
      "epoch": 14.999255287459041,
      "grad_norm": 12.875,
      "learning_rate": 8.933296994789678e-06,
      "loss": 0.9616,
      "num_input_tokens_seen": 58426904,
      "step": 100705
    },
    {
      "epoch": 15.0,
      "grad_norm": 25.375,
      "learning_rate": 8.930807603515895e-06,
      "loss": 1.1255,
      "num_input_tokens_seen": 58429624,
      "step": 100710
    },
    {
      "epoch": 15.0,
      "eval_loss": 1.0822354555130005,
      "eval_runtime": 43.8306,
      "eval_samples_per_second": 68.08,
      "eval_steps_per_second": 17.02,
      "num_input_tokens_seen": 58429624,
      "step": 100710
    },
    {
      "epoch": 15.000744712540959,
      "grad_norm": 5.0625,
      "learning_rate": 8.928318483714793e-06,
      "loss": 0.929,
      "num_input_tokens_seen": 58433016,
      "step": 100715
    },
    {
      "epoch": 15.001489425081918,
      "grad_norm": 20.625,
      "learning_rate": 8.925829635428414e-06,
      "loss": 1.001,
      "num_input_tokens_seen": 58435928,
      "step": 100720
    },
    {
      "epoch": 15.002234137622878,
      "grad_norm": 6.78125,
      "learning_rate": 8.92334105869881e-06,
      "loss": 0.9841,
      "num_input_tokens_seen": 58438776,
      "step": 100725
    },
    {
      "epoch": 15.002978850163837,
      "grad_norm": 7.84375,
      "learning_rate": 8.920852753568015e-06,
      "loss": 1.0018,
      "num_input_tokens_seen": 58441848,
      "step": 100730
    },
    {
      "epoch": 15.003723562704796,
      "grad_norm": 11.0625,
      "learning_rate": 8.918364720078063e-06,
      "loss": 1.0473,
      "num_input_tokens_seen": 58445112,
      "step": 100735
    },
    {
      "epoch": 15.004468275245754,
      "grad_norm": 14.4375,
      "learning_rate": 8.915876958271006e-06,
      "loss": 1.334,
      "num_input_tokens_seen": 58447832,
      "step": 100740
    },
    {
      "epoch": 15.005212987786715,
      "grad_norm": 15.6875,
      "learning_rate": 8.913389468188849e-06,
      "loss": 1.1489,
      "num_input_tokens_seen": 58450776,
      "step": 100745
    },
    {
      "epoch": 15.005957700327674,
      "grad_norm": 10.8125,
      "learning_rate": 8.910902249873637e-06,
      "loss": 1.1264,
      "num_input_tokens_seen": 58453752,
      "step": 100750
    },
    {
      "epoch": 15.006702412868632,
      "grad_norm": 3.5625,
      "learning_rate": 8.908415303367371e-06,
      "loss": 1.0162,
      "num_input_tokens_seen": 58456696,
      "step": 100755
    },
    {
      "epoch": 15.007447125409591,
      "grad_norm": 16.125,
      "learning_rate": 8.905928628712083e-06,
      "loss": 1.008,
      "num_input_tokens_seen": 58459416,
      "step": 100760
    },
    {
      "epoch": 15.008191837950552,
      "grad_norm": 10.125,
      "learning_rate": 8.90344222594977e-06,
      "loss": 0.9713,
      "num_input_tokens_seen": 58462264,
      "step": 100765
    },
    {
      "epoch": 15.00893655049151,
      "grad_norm": 12.3125,
      "learning_rate": 8.900956095122435e-06,
      "loss": 1.61,
      "num_input_tokens_seen": 58465176,
      "step": 100770
    },
    {
      "epoch": 15.00968126303247,
      "grad_norm": 5.84375,
      "learning_rate": 8.898470236272091e-06,
      "loss": 1.2115,
      "num_input_tokens_seen": 58468088,
      "step": 100775
    },
    {
      "epoch": 15.010425975573428,
      "grad_norm": 13.9375,
      "learning_rate": 8.895984649440722e-06,
      "loss": 1.0007,
      "num_input_tokens_seen": 58471064,
      "step": 100780
    },
    {
      "epoch": 15.011170688114388,
      "grad_norm": 7.96875,
      "learning_rate": 8.89349933467033e-06,
      "loss": 1.0481,
      "num_input_tokens_seen": 58474104,
      "step": 100785
    },
    {
      "epoch": 15.011915400655347,
      "grad_norm": 18.0,
      "learning_rate": 8.8910142920029e-06,
      "loss": 1.413,
      "num_input_tokens_seen": 58476952,
      "step": 100790
    },
    {
      "epoch": 15.012660113196306,
      "grad_norm": 12.25,
      "learning_rate": 8.88852952148041e-06,
      "loss": 1.4152,
      "num_input_tokens_seen": 58479928,
      "step": 100795
    },
    {
      "epoch": 15.013404825737265,
      "grad_norm": 9.8125,
      "learning_rate": 8.886045023144829e-06,
      "loss": 1.2743,
      "num_input_tokens_seen": 58482968,
      "step": 100800
    },
    {
      "epoch": 15.014149538278225,
      "grad_norm": 20.875,
      "learning_rate": 8.883560797038152e-06,
      "loss": 1.2198,
      "num_input_tokens_seen": 58485816,
      "step": 100805
    },
    {
      "epoch": 15.014894250819184,
      "grad_norm": 8.0,
      "learning_rate": 8.881076843202332e-06,
      "loss": 0.8812,
      "num_input_tokens_seen": 58488888,
      "step": 100810
    },
    {
      "epoch": 15.015638963360143,
      "grad_norm": 12.25,
      "learning_rate": 8.878593161679327e-06,
      "loss": 1.0101,
      "num_input_tokens_seen": 58492056,
      "step": 100815
    },
    {
      "epoch": 15.016383675901102,
      "grad_norm": 3.21875,
      "learning_rate": 8.876109752511117e-06,
      "loss": 1.0518,
      "num_input_tokens_seen": 58495128,
      "step": 100820
    },
    {
      "epoch": 15.017128388442062,
      "grad_norm": 17.75,
      "learning_rate": 8.873626615739632e-06,
      "loss": 0.8938,
      "num_input_tokens_seen": 58497592,
      "step": 100825
    },
    {
      "epoch": 15.01787310098302,
      "grad_norm": 11.1875,
      "learning_rate": 8.871143751406849e-06,
      "loss": 0.9063,
      "num_input_tokens_seen": 58500408,
      "step": 100830
    },
    {
      "epoch": 15.01861781352398,
      "grad_norm": 11.4375,
      "learning_rate": 8.868661159554689e-06,
      "loss": 1.0067,
      "num_input_tokens_seen": 58503768,
      "step": 100835
    },
    {
      "epoch": 15.019362526064938,
      "grad_norm": 10.875,
      "learning_rate": 8.866178840225111e-06,
      "loss": 0.989,
      "num_input_tokens_seen": 58506552,
      "step": 100840
    },
    {
      "epoch": 15.020107238605899,
      "grad_norm": 9.9375,
      "learning_rate": 8.863696793460047e-06,
      "loss": 1.1752,
      "num_input_tokens_seen": 58509432,
      "step": 100845
    },
    {
      "epoch": 15.020851951146858,
      "grad_norm": 5.96875,
      "learning_rate": 8.861215019301414e-06,
      "loss": 1.264,
      "num_input_tokens_seen": 58512440,
      "step": 100850
    },
    {
      "epoch": 15.021596663687816,
      "grad_norm": 15.9375,
      "learning_rate": 8.85873351779116e-06,
      "loss": 1.0405,
      "num_input_tokens_seen": 58515064,
      "step": 100855
    },
    {
      "epoch": 15.022341376228775,
      "grad_norm": 8.875,
      "learning_rate": 8.856252288971198e-06,
      "loss": 0.9741,
      "num_input_tokens_seen": 58518040,
      "step": 100860
    },
    {
      "epoch": 15.023086088769736,
      "grad_norm": 17.5,
      "learning_rate": 8.853771332883446e-06,
      "loss": 1.2215,
      "num_input_tokens_seen": 58520856,
      "step": 100865
    },
    {
      "epoch": 15.023830801310694,
      "grad_norm": 7.46875,
      "learning_rate": 8.851290649569808e-06,
      "loss": 0.9914,
      "num_input_tokens_seen": 58523896,
      "step": 100870
    },
    {
      "epoch": 15.024575513851653,
      "grad_norm": 9.0,
      "learning_rate": 8.848810239072208e-06,
      "loss": 0.8881,
      "num_input_tokens_seen": 58526968,
      "step": 100875
    },
    {
      "epoch": 15.025320226392612,
      "grad_norm": 8.9375,
      "learning_rate": 8.84633010143254e-06,
      "loss": 1.1995,
      "num_input_tokens_seen": 58529912,
      "step": 100880
    },
    {
      "epoch": 15.02606493893357,
      "grad_norm": 17.0,
      "learning_rate": 8.84385023669271e-06,
      "loss": 1.4731,
      "num_input_tokens_seen": 58532856,
      "step": 100885
    },
    {
      "epoch": 15.026809651474531,
      "grad_norm": 7.5,
      "learning_rate": 8.841370644894614e-06,
      "loss": 0.9499,
      "num_input_tokens_seen": 58536184,
      "step": 100890
    },
    {
      "epoch": 15.02755436401549,
      "grad_norm": 18.375,
      "learning_rate": 8.838891326080129e-06,
      "loss": 0.8328,
      "num_input_tokens_seen": 58539256,
      "step": 100895
    },
    {
      "epoch": 15.028299076556449,
      "grad_norm": 13.875,
      "learning_rate": 8.83641228029116e-06,
      "loss": 1.1883,
      "num_input_tokens_seen": 58542200,
      "step": 100900
    },
    {
      "epoch": 15.029043789097408,
      "grad_norm": 5.1875,
      "learning_rate": 8.833933507569564e-06,
      "loss": 0.9608,
      "num_input_tokens_seen": 58545272,
      "step": 100905
    },
    {
      "epoch": 15.029788501638368,
      "grad_norm": 9.3125,
      "learning_rate": 8.831455007957243e-06,
      "loss": 1.135,
      "num_input_tokens_seen": 58548120,
      "step": 100910
    },
    {
      "epoch": 15.030533214179327,
      "grad_norm": 8.75,
      "learning_rate": 8.828976781496057e-06,
      "loss": 1.1036,
      "num_input_tokens_seen": 58551256,
      "step": 100915
    },
    {
      "epoch": 15.031277926720286,
      "grad_norm": 15.875,
      "learning_rate": 8.826498828227861e-06,
      "loss": 1.3292,
      "num_input_tokens_seen": 58554104,
      "step": 100920
    },
    {
      "epoch": 15.032022639261244,
      "grad_norm": 14.9375,
      "learning_rate": 8.824021148194541e-06,
      "loss": 1.0586,
      "num_input_tokens_seen": 58557176,
      "step": 100925
    },
    {
      "epoch": 15.032767351802205,
      "grad_norm": 17.375,
      "learning_rate": 8.82154374143794e-06,
      "loss": 1.4266,
      "num_input_tokens_seen": 58560088,
      "step": 100930
    },
    {
      "epoch": 15.033512064343164,
      "grad_norm": 11.8125,
      "learning_rate": 8.819066607999918e-06,
      "loss": 0.9663,
      "num_input_tokens_seen": 58562680,
      "step": 100935
    },
    {
      "epoch": 15.034256776884122,
      "grad_norm": 8.625,
      "learning_rate": 8.816589747922311e-06,
      "loss": 1.1641,
      "num_input_tokens_seen": 58565496,
      "step": 100940
    },
    {
      "epoch": 15.035001489425081,
      "grad_norm": 11.0,
      "learning_rate": 8.814113161246979e-06,
      "loss": 0.9109,
      "num_input_tokens_seen": 58568472,
      "step": 100945
    },
    {
      "epoch": 15.035746201966042,
      "grad_norm": 10.8125,
      "learning_rate": 8.811636848015747e-06,
      "loss": 1.2329,
      "num_input_tokens_seen": 58571544,
      "step": 100950
    },
    {
      "epoch": 15.036490914507,
      "grad_norm": 11.375,
      "learning_rate": 8.809160808270464e-06,
      "loss": 0.8962,
      "num_input_tokens_seen": 58574488,
      "step": 100955
    },
    {
      "epoch": 15.03723562704796,
      "grad_norm": 9.1875,
      "learning_rate": 8.806685042052949e-06,
      "loss": 1.2249,
      "num_input_tokens_seen": 58577272,
      "step": 100960
    },
    {
      "epoch": 15.037980339588918,
      "grad_norm": 20.875,
      "learning_rate": 8.804209549405037e-06,
      "loss": 1.2928,
      "num_input_tokens_seen": 58580312,
      "step": 100965
    },
    {
      "epoch": 15.038725052129879,
      "grad_norm": 10.0625,
      "learning_rate": 8.801734330368544e-06,
      "loss": 1.2326,
      "num_input_tokens_seen": 58583256,
      "step": 100970
    },
    {
      "epoch": 15.039469764670837,
      "grad_norm": 11.5,
      "learning_rate": 8.79925938498528e-06,
      "loss": 1.0392,
      "num_input_tokens_seen": 58586200,
      "step": 100975
    },
    {
      "epoch": 15.040214477211796,
      "grad_norm": 9.6875,
      "learning_rate": 8.796784713297072e-06,
      "loss": 1.1422,
      "num_input_tokens_seen": 58589176,
      "step": 100980
    },
    {
      "epoch": 15.040959189752755,
      "grad_norm": 11.8125,
      "learning_rate": 8.794310315345713e-06,
      "loss": 1.1826,
      "num_input_tokens_seen": 58592184,
      "step": 100985
    },
    {
      "epoch": 15.041703902293715,
      "grad_norm": 7.84375,
      "learning_rate": 8.791836191173017e-06,
      "loss": 1.2299,
      "num_input_tokens_seen": 58595448,
      "step": 100990
    },
    {
      "epoch": 15.042448614834674,
      "grad_norm": 6.5625,
      "learning_rate": 8.78936234082076e-06,
      "loss": 0.9111,
      "num_input_tokens_seen": 58598392,
      "step": 100995
    },
    {
      "epoch": 15.043193327375633,
      "grad_norm": 5.03125,
      "learning_rate": 8.786888764330767e-06,
      "loss": 1.2097,
      "num_input_tokens_seen": 58601240,
      "step": 101000
    },
    {
      "epoch": 15.043938039916592,
      "grad_norm": 15.4375,
      "learning_rate": 8.784415461744805e-06,
      "loss": 1.3907,
      "num_input_tokens_seen": 58604152,
      "step": 101005
    },
    {
      "epoch": 15.044682752457552,
      "grad_norm": 10.5,
      "learning_rate": 8.781942433104654e-06,
      "loss": 1.3205,
      "num_input_tokens_seen": 58607032,
      "step": 101010
    },
    {
      "epoch": 15.045427464998511,
      "grad_norm": 11.4375,
      "learning_rate": 8.779469678452113e-06,
      "loss": 1.2852,
      "num_input_tokens_seen": 58609848,
      "step": 101015
    },
    {
      "epoch": 15.04617217753947,
      "grad_norm": 6.5625,
      "learning_rate": 8.776997197828937e-06,
      "loss": 1.1115,
      "num_input_tokens_seen": 58612568,
      "step": 101020
    },
    {
      "epoch": 15.046916890080428,
      "grad_norm": 6.46875,
      "learning_rate": 8.774524991276911e-06,
      "loss": 1.2354,
      "num_input_tokens_seen": 58615320,
      "step": 101025
    },
    {
      "epoch": 15.047661602621389,
      "grad_norm": 14.75,
      "learning_rate": 8.77205305883779e-06,
      "loss": 1.3526,
      "num_input_tokens_seen": 58618296,
      "step": 101030
    },
    {
      "epoch": 15.048406315162348,
      "grad_norm": 10.25,
      "learning_rate": 8.769581400553346e-06,
      "loss": 1.0453,
      "num_input_tokens_seen": 58621240,
      "step": 101035
    },
    {
      "epoch": 15.049151027703306,
      "grad_norm": 7.78125,
      "learning_rate": 8.767110016465318e-06,
      "loss": 1.0586,
      "num_input_tokens_seen": 58623928,
      "step": 101040
    },
    {
      "epoch": 15.049895740244265,
      "grad_norm": 15.8125,
      "learning_rate": 8.76463890661548e-06,
      "loss": 1.013,
      "num_input_tokens_seen": 58626904,
      "step": 101045
    },
    {
      "epoch": 15.050640452785226,
      "grad_norm": 13.0,
      "learning_rate": 8.762168071045566e-06,
      "loss": 1.2315,
      "num_input_tokens_seen": 58629624,
      "step": 101050
    },
    {
      "epoch": 15.051385165326185,
      "grad_norm": 4.59375,
      "learning_rate": 8.759697509797315e-06,
      "loss": 0.8501,
      "num_input_tokens_seen": 58632408,
      "step": 101055
    },
    {
      "epoch": 15.052129877867143,
      "grad_norm": 9.125,
      "learning_rate": 8.757227222912473e-06,
      "loss": 1.3373,
      "num_input_tokens_seen": 58635608,
      "step": 101060
    },
    {
      "epoch": 15.052874590408102,
      "grad_norm": 13.8125,
      "learning_rate": 8.754757210432758e-06,
      "loss": 1.2312,
      "num_input_tokens_seen": 58638616,
      "step": 101065
    },
    {
      "epoch": 15.05361930294906,
      "grad_norm": 13.1875,
      "learning_rate": 8.752287472399918e-06,
      "loss": 1.3046,
      "num_input_tokens_seen": 58641432,
      "step": 101070
    },
    {
      "epoch": 15.054364015490021,
      "grad_norm": 8.4375,
      "learning_rate": 8.74981800885566e-06,
      "loss": 1.0657,
      "num_input_tokens_seen": 58644536,
      "step": 101075
    },
    {
      "epoch": 15.05510872803098,
      "grad_norm": 15.9375,
      "learning_rate": 8.747348819841719e-06,
      "loss": 0.9241,
      "num_input_tokens_seen": 58647896,
      "step": 101080
    },
    {
      "epoch": 15.055853440571939,
      "grad_norm": 12.6875,
      "learning_rate": 8.7448799053998e-06,
      "loss": 1.1558,
      "num_input_tokens_seen": 58650808,
      "step": 101085
    },
    {
      "epoch": 15.056598153112898,
      "grad_norm": 10.375,
      "learning_rate": 8.742411265571607e-06,
      "loss": 1.2931,
      "num_input_tokens_seen": 58653432,
      "step": 101090
    },
    {
      "epoch": 15.057342865653858,
      "grad_norm": 10.75,
      "learning_rate": 8.73994290039886e-06,
      "loss": 1.0656,
      "num_input_tokens_seen": 58656216,
      "step": 101095
    },
    {
      "epoch": 15.058087578194817,
      "grad_norm": 11.0,
      "learning_rate": 8.737474809923244e-06,
      "loss": 1.1672,
      "num_input_tokens_seen": 58659160,
      "step": 101100
    },
    {
      "epoch": 15.058832290735776,
      "grad_norm": 24.5,
      "learning_rate": 8.73500699418647e-06,
      "loss": 1.3311,
      "num_input_tokens_seen": 58662136,
      "step": 101105
    },
    {
      "epoch": 15.059577003276734,
      "grad_norm": 13.125,
      "learning_rate": 8.732539453230215e-06,
      "loss": 1.0727,
      "num_input_tokens_seen": 58664984,
      "step": 101110
    },
    {
      "epoch": 15.060321715817695,
      "grad_norm": 10.4375,
      "learning_rate": 8.730072187096178e-06,
      "loss": 1.6128,
      "num_input_tokens_seen": 58667800,
      "step": 101115
    },
    {
      "epoch": 15.061066428358654,
      "grad_norm": 10.3125,
      "learning_rate": 8.727605195826038e-06,
      "loss": 1.3407,
      "num_input_tokens_seen": 58670648,
      "step": 101120
    },
    {
      "epoch": 15.061811140899612,
      "grad_norm": 16.625,
      "learning_rate": 8.72513847946147e-06,
      "loss": 1.036,
      "num_input_tokens_seen": 58673368,
      "step": 101125
    },
    {
      "epoch": 15.062555853440571,
      "grad_norm": 3.890625,
      "learning_rate": 8.722672038044145e-06,
      "loss": 0.7636,
      "num_input_tokens_seen": 58676472,
      "step": 101130
    },
    {
      "epoch": 15.063300565981532,
      "grad_norm": 6.59375,
      "learning_rate": 8.720205871615722e-06,
      "loss": 1.148,
      "num_input_tokens_seen": 58679288,
      "step": 101135
    },
    {
      "epoch": 15.06404527852249,
      "grad_norm": 14.3125,
      "learning_rate": 8.717739980217887e-06,
      "loss": 1.2462,
      "num_input_tokens_seen": 58682168,
      "step": 101140
    },
    {
      "epoch": 15.06478999106345,
      "grad_norm": 18.0,
      "learning_rate": 8.715274363892276e-06,
      "loss": 1.0502,
      "num_input_tokens_seen": 58685368,
      "step": 101145
    },
    {
      "epoch": 15.065534703604408,
      "grad_norm": 6.90625,
      "learning_rate": 8.712809022680563e-06,
      "loss": 1.134,
      "num_input_tokens_seen": 58688280,
      "step": 101150
    },
    {
      "epoch": 15.066279416145369,
      "grad_norm": 12.6875,
      "learning_rate": 8.710343956624379e-06,
      "loss": 1.0816,
      "num_input_tokens_seen": 58691256,
      "step": 101155
    },
    {
      "epoch": 15.067024128686327,
      "grad_norm": 8.125,
      "learning_rate": 8.707879165765384e-06,
      "loss": 1.0829,
      "num_input_tokens_seen": 58694520,
      "step": 101160
    },
    {
      "epoch": 15.067768841227286,
      "grad_norm": 11.5,
      "learning_rate": 8.705414650145215e-06,
      "loss": 0.7816,
      "num_input_tokens_seen": 58697336,
      "step": 101165
    },
    {
      "epoch": 15.068513553768245,
      "grad_norm": 11.125,
      "learning_rate": 8.702950409805493e-06,
      "loss": 1.1172,
      "num_input_tokens_seen": 58700280,
      "step": 101170
    },
    {
      "epoch": 15.069258266309205,
      "grad_norm": 6.03125,
      "learning_rate": 8.700486444787872e-06,
      "loss": 1.0121,
      "num_input_tokens_seen": 58703064,
      "step": 101175
    },
    {
      "epoch": 15.070002978850164,
      "grad_norm": 12.6875,
      "learning_rate": 8.698022755133957e-06,
      "loss": 1.3264,
      "num_input_tokens_seen": 58705880,
      "step": 101180
    },
    {
      "epoch": 15.070747691391123,
      "grad_norm": 4.5,
      "learning_rate": 8.695559340885387e-06,
      "loss": 1.0558,
      "num_input_tokens_seen": 58709208,
      "step": 101185
    },
    {
      "epoch": 15.071492403932082,
      "grad_norm": 12.6875,
      "learning_rate": 8.693096202083773e-06,
      "loss": 0.9553,
      "num_input_tokens_seen": 58711896,
      "step": 101190
    },
    {
      "epoch": 15.072237116473042,
      "grad_norm": 9.3125,
      "learning_rate": 8.69063333877072e-06,
      "loss": 1.1286,
      "num_input_tokens_seen": 58715352,
      "step": 101195
    },
    {
      "epoch": 15.072981829014001,
      "grad_norm": 12.4375,
      "learning_rate": 8.688170750987836e-06,
      "loss": 1.0052,
      "num_input_tokens_seen": 58717784,
      "step": 101200
    },
    {
      "epoch": 15.07372654155496,
      "grad_norm": 14.625,
      "learning_rate": 8.685708438776739e-06,
      "loss": 1.4319,
      "num_input_tokens_seen": 58720536,
      "step": 101205
    },
    {
      "epoch": 15.074471254095918,
      "grad_norm": 8.875,
      "learning_rate": 8.683246402179013e-06,
      "loss": 0.785,
      "num_input_tokens_seen": 58723544,
      "step": 101210
    },
    {
      "epoch": 15.075215966636879,
      "grad_norm": 5.65625,
      "learning_rate": 8.680784641236248e-06,
      "loss": 0.6412,
      "num_input_tokens_seen": 58726648,
      "step": 101215
    },
    {
      "epoch": 15.075960679177838,
      "grad_norm": 15.0,
      "learning_rate": 8.678323155990047e-06,
      "loss": 1.4651,
      "num_input_tokens_seen": 58729848,
      "step": 101220
    },
    {
      "epoch": 15.076705391718797,
      "grad_norm": 12.25,
      "learning_rate": 8.67586194648198e-06,
      "loss": 1.0326,
      "num_input_tokens_seen": 58732696,
      "step": 101225
    },
    {
      "epoch": 15.077450104259755,
      "grad_norm": 9.5625,
      "learning_rate": 8.673401012753646e-06,
      "loss": 1.3033,
      "num_input_tokens_seen": 58735640,
      "step": 101230
    },
    {
      "epoch": 15.078194816800714,
      "grad_norm": 7.5625,
      "learning_rate": 8.670940354846596e-06,
      "loss": 1.2919,
      "num_input_tokens_seen": 58738456,
      "step": 101235
    },
    {
      "epoch": 15.078939529341675,
      "grad_norm": 7.96875,
      "learning_rate": 8.668479972802423e-06,
      "loss": 1.0586,
      "num_input_tokens_seen": 58741496,
      "step": 101240
    },
    {
      "epoch": 15.079684241882633,
      "grad_norm": 7.625,
      "learning_rate": 8.666019866662683e-06,
      "loss": 1.0083,
      "num_input_tokens_seen": 58744376,
      "step": 101245
    },
    {
      "epoch": 15.080428954423592,
      "grad_norm": 23.0,
      "learning_rate": 8.663560036468926e-06,
      "loss": 1.2533,
      "num_input_tokens_seen": 58747224,
      "step": 101250
    },
    {
      "epoch": 15.08117366696455,
      "grad_norm": 4.90625,
      "learning_rate": 8.661100482262729e-06,
      "loss": 1.183,
      "num_input_tokens_seen": 58750136,
      "step": 101255
    },
    {
      "epoch": 15.081918379505511,
      "grad_norm": 10.625,
      "learning_rate": 8.658641204085632e-06,
      "loss": 1.0531,
      "num_input_tokens_seen": 58752824,
      "step": 101260
    },
    {
      "epoch": 15.08266309204647,
      "grad_norm": 9.6875,
      "learning_rate": 8.656182201979181e-06,
      "loss": 1.0099,
      "num_input_tokens_seen": 58755768,
      "step": 101265
    },
    {
      "epoch": 15.083407804587429,
      "grad_norm": 8.9375,
      "learning_rate": 8.653723475984916e-06,
      "loss": 1.0322,
      "num_input_tokens_seen": 58758392,
      "step": 101270
    },
    {
      "epoch": 15.084152517128388,
      "grad_norm": 5.5625,
      "learning_rate": 8.651265026144387e-06,
      "loss": 0.8804,
      "num_input_tokens_seen": 58760984,
      "step": 101275
    },
    {
      "epoch": 15.084897229669348,
      "grad_norm": 9.8125,
      "learning_rate": 8.648806852499109e-06,
      "loss": 1.0215,
      "num_input_tokens_seen": 58763832,
      "step": 101280
    },
    {
      "epoch": 15.085641942210307,
      "grad_norm": 18.125,
      "learning_rate": 8.64634895509063e-06,
      "loss": 1.1102,
      "num_input_tokens_seen": 58766648,
      "step": 101285
    },
    {
      "epoch": 15.086386654751266,
      "grad_norm": 8.5625,
      "learning_rate": 8.643891333960464e-06,
      "loss": 1.0591,
      "num_input_tokens_seen": 58769848,
      "step": 101290
    },
    {
      "epoch": 15.087131367292224,
      "grad_norm": 12.625,
      "learning_rate": 8.641433989150123e-06,
      "loss": 1.2369,
      "num_input_tokens_seen": 58772824,
      "step": 101295
    },
    {
      "epoch": 15.087876079833185,
      "grad_norm": 10.5625,
      "learning_rate": 8.638976920701137e-06,
      "loss": 1.0348,
      "num_input_tokens_seen": 58775384,
      "step": 101300
    },
    {
      "epoch": 15.088620792374144,
      "grad_norm": 12.8125,
      "learning_rate": 8.636520128654995e-06,
      "loss": 1.1573,
      "num_input_tokens_seen": 58778296,
      "step": 101305
    },
    {
      "epoch": 15.089365504915103,
      "grad_norm": 14.125,
      "learning_rate": 8.634063613053228e-06,
      "loss": 1.1268,
      "num_input_tokens_seen": 58781208,
      "step": 101310
    },
    {
      "epoch": 15.090110217456061,
      "grad_norm": 8.6875,
      "learning_rate": 8.631607373937319e-06,
      "loss": 1.3432,
      "num_input_tokens_seen": 58783800,
      "step": 101315
    },
    {
      "epoch": 15.090854929997022,
      "grad_norm": 11.875,
      "learning_rate": 8.62915141134877e-06,
      "loss": 1.1276,
      "num_input_tokens_seen": 58787128,
      "step": 101320
    },
    {
      "epoch": 15.09159964253798,
      "grad_norm": 6.46875,
      "learning_rate": 8.626695725329059e-06,
      "loss": 1.0522,
      "num_input_tokens_seen": 58790264,
      "step": 101325
    },
    {
      "epoch": 15.09234435507894,
      "grad_norm": 7.3125,
      "learning_rate": 8.624240315919693e-06,
      "loss": 0.9825,
      "num_input_tokens_seen": 58793464,
      "step": 101330
    },
    {
      "epoch": 15.093089067619898,
      "grad_norm": 11.0625,
      "learning_rate": 8.62178518316214e-06,
      "loss": 0.8943,
      "num_input_tokens_seen": 58796632,
      "step": 101335
    },
    {
      "epoch": 15.093833780160859,
      "grad_norm": 10.6875,
      "learning_rate": 8.619330327097874e-06,
      "loss": 1.1061,
      "num_input_tokens_seen": 58799608,
      "step": 101340
    },
    {
      "epoch": 15.094578492701817,
      "grad_norm": 10.0625,
      "learning_rate": 8.616875747768382e-06,
      "loss": 1.1386,
      "num_input_tokens_seen": 58802584,
      "step": 101345
    },
    {
      "epoch": 15.095323205242776,
      "grad_norm": 17.625,
      "learning_rate": 8.614421445215116e-06,
      "loss": 0.9491,
      "num_input_tokens_seen": 58805336,
      "step": 101350
    },
    {
      "epoch": 15.096067917783735,
      "grad_norm": 11.375,
      "learning_rate": 8.611967419479553e-06,
      "loss": 1.2778,
      "num_input_tokens_seen": 58808408,
      "step": 101355
    },
    {
      "epoch": 15.096812630324695,
      "grad_norm": 17.625,
      "learning_rate": 8.609513670603137e-06,
      "loss": 1.3096,
      "num_input_tokens_seen": 58811096,
      "step": 101360
    },
    {
      "epoch": 15.097557342865654,
      "grad_norm": 15.1875,
      "learning_rate": 8.607060198627337e-06,
      "loss": 0.9892,
      "num_input_tokens_seen": 58813784,
      "step": 101365
    },
    {
      "epoch": 15.098302055406613,
      "grad_norm": 12.3125,
      "learning_rate": 8.604607003593593e-06,
      "loss": 1.045,
      "num_input_tokens_seen": 58816664,
      "step": 101370
    },
    {
      "epoch": 15.099046767947572,
      "grad_norm": 7.375,
      "learning_rate": 8.602154085543341e-06,
      "loss": 1.2756,
      "num_input_tokens_seen": 58819448,
      "step": 101375
    },
    {
      "epoch": 15.099791480488532,
      "grad_norm": 9.1875,
      "learning_rate": 8.59970144451804e-06,
      "loss": 0.8898,
      "num_input_tokens_seen": 58822200,
      "step": 101380
    },
    {
      "epoch": 15.100536193029491,
      "grad_norm": 12.1875,
      "learning_rate": 8.597249080559114e-06,
      "loss": 1.0953,
      "num_input_tokens_seen": 58824824,
      "step": 101385
    },
    {
      "epoch": 15.10128090557045,
      "grad_norm": 8.4375,
      "learning_rate": 8.594796993707993e-06,
      "loss": 1.0275,
      "num_input_tokens_seen": 58827480,
      "step": 101390
    },
    {
      "epoch": 15.102025618111409,
      "grad_norm": 8.6875,
      "learning_rate": 8.592345184006096e-06,
      "loss": 1.0571,
      "num_input_tokens_seen": 58830200,
      "step": 101395
    },
    {
      "epoch": 15.102770330652369,
      "grad_norm": 3.234375,
      "learning_rate": 8.58989365149486e-06,
      "loss": 0.9211,
      "num_input_tokens_seen": 58833016,
      "step": 101400
    },
    {
      "epoch": 15.103515043193328,
      "grad_norm": 12.3125,
      "learning_rate": 8.58744239621568e-06,
      "loss": 1.0129,
      "num_input_tokens_seen": 58835544,
      "step": 101405
    },
    {
      "epoch": 15.104259755734287,
      "grad_norm": 9.125,
      "learning_rate": 8.584991418209992e-06,
      "loss": 1.0801,
      "num_input_tokens_seen": 58838328,
      "step": 101410
    },
    {
      "epoch": 15.105004468275245,
      "grad_norm": 11.9375,
      "learning_rate": 8.582540717519191e-06,
      "loss": 0.9234,
      "num_input_tokens_seen": 58841336,
      "step": 101415
    },
    {
      "epoch": 15.105749180816204,
      "grad_norm": 7.59375,
      "learning_rate": 8.580090294184667e-06,
      "loss": 0.9486,
      "num_input_tokens_seen": 58844216,
      "step": 101420
    },
    {
      "epoch": 15.106493893357165,
      "grad_norm": 13.0,
      "learning_rate": 8.57764014824784e-06,
      "loss": 1.328,
      "num_input_tokens_seen": 58847160,
      "step": 101425
    },
    {
      "epoch": 15.107238605898123,
      "grad_norm": 3.28125,
      "learning_rate": 8.575190279750085e-06,
      "loss": 0.9142,
      "num_input_tokens_seen": 58850104,
      "step": 101430
    },
    {
      "epoch": 15.107983318439082,
      "grad_norm": 10.5,
      "learning_rate": 8.5727406887328e-06,
      "loss": 0.8769,
      "num_input_tokens_seen": 58852920,
      "step": 101435
    },
    {
      "epoch": 15.108728030980041,
      "grad_norm": 11.1875,
      "learning_rate": 8.570291375237361e-06,
      "loss": 0.8195,
      "num_input_tokens_seen": 58856120,
      "step": 101440
    },
    {
      "epoch": 15.109472743521001,
      "grad_norm": 10.6875,
      "learning_rate": 8.567842339305157e-06,
      "loss": 1.1307,
      "num_input_tokens_seen": 58858840,
      "step": 101445
    },
    {
      "epoch": 15.11021745606196,
      "grad_norm": 18.0,
      "learning_rate": 8.565393580977558e-06,
      "loss": 1.2858,
      "num_input_tokens_seen": 58861688,
      "step": 101450
    },
    {
      "epoch": 15.110962168602919,
      "grad_norm": 7.875,
      "learning_rate": 8.562945100295927e-06,
      "loss": 1.2979,
      "num_input_tokens_seen": 58864696,
      "step": 101455
    },
    {
      "epoch": 15.111706881143878,
      "grad_norm": 17.0,
      "learning_rate": 8.560496897301637e-06,
      "loss": 1.0773,
      "num_input_tokens_seen": 58867576,
      "step": 101460
    },
    {
      "epoch": 15.112451593684838,
      "grad_norm": 9.0625,
      "learning_rate": 8.558048972036031e-06,
      "loss": 1.3168,
      "num_input_tokens_seen": 58870488,
      "step": 101465
    },
    {
      "epoch": 15.113196306225797,
      "grad_norm": 7.21875,
      "learning_rate": 8.555601324540488e-06,
      "loss": 1.0681,
      "num_input_tokens_seen": 58873496,
      "step": 101470
    },
    {
      "epoch": 15.113941018766756,
      "grad_norm": 12.6875,
      "learning_rate": 8.553153954856338e-06,
      "loss": 0.9708,
      "num_input_tokens_seen": 58876600,
      "step": 101475
    },
    {
      "epoch": 15.114685731307715,
      "grad_norm": 12.5625,
      "learning_rate": 8.550706863024945e-06,
      "loss": 1.1408,
      "num_input_tokens_seen": 58879320,
      "step": 101480
    },
    {
      "epoch": 15.115430443848675,
      "grad_norm": 11.5625,
      "learning_rate": 8.548260049087634e-06,
      "loss": 1.0622,
      "num_input_tokens_seen": 58882424,
      "step": 101485
    },
    {
      "epoch": 15.116175156389634,
      "grad_norm": 9.25,
      "learning_rate": 8.545813513085757e-06,
      "loss": 1.1994,
      "num_input_tokens_seen": 58885400,
      "step": 101490
    },
    {
      "epoch": 15.116919868930593,
      "grad_norm": 20.625,
      "learning_rate": 8.543367255060636e-06,
      "loss": 1.3606,
      "num_input_tokens_seen": 58888536,
      "step": 101495
    },
    {
      "epoch": 15.117664581471551,
      "grad_norm": 10.4375,
      "learning_rate": 8.54092127505359e-06,
      "loss": 1.1676,
      "num_input_tokens_seen": 58891256,
      "step": 101500
    },
    {
      "epoch": 15.118409294012512,
      "grad_norm": 7.25,
      "learning_rate": 8.538475573105961e-06,
      "loss": 0.5969,
      "num_input_tokens_seen": 58894232,
      "step": 101505
    },
    {
      "epoch": 15.11915400655347,
      "grad_norm": 3.1875,
      "learning_rate": 8.536030149259046e-06,
      "loss": 0.762,
      "num_input_tokens_seen": 58897208,
      "step": 101510
    },
    {
      "epoch": 15.11989871909443,
      "grad_norm": 6.125,
      "learning_rate": 8.533585003554179e-06,
      "loss": 1.1341,
      "num_input_tokens_seen": 58899864,
      "step": 101515
    },
    {
      "epoch": 15.120643431635388,
      "grad_norm": 12.75,
      "learning_rate": 8.53114013603266e-06,
      "loss": 1.1868,
      "num_input_tokens_seen": 58902936,
      "step": 101520
    },
    {
      "epoch": 15.121388144176349,
      "grad_norm": 12.5,
      "learning_rate": 8.528695546735784e-06,
      "loss": 0.9825,
      "num_input_tokens_seen": 58905880,
      "step": 101525
    },
    {
      "epoch": 15.122132856717307,
      "grad_norm": 7.8125,
      "learning_rate": 8.52625123570486e-06,
      "loss": 1.0533,
      "num_input_tokens_seen": 58908792,
      "step": 101530
    },
    {
      "epoch": 15.122877569258266,
      "grad_norm": 14.3125,
      "learning_rate": 8.523807202981168e-06,
      "loss": 1.1514,
      "num_input_tokens_seen": 58911576,
      "step": 101535
    },
    {
      "epoch": 15.123622281799225,
      "grad_norm": 12.25,
      "learning_rate": 8.521363448606018e-06,
      "loss": 1.2309,
      "num_input_tokens_seen": 58914520,
      "step": 101540
    },
    {
      "epoch": 15.124366994340185,
      "grad_norm": 10.625,
      "learning_rate": 8.518919972620675e-06,
      "loss": 0.9438,
      "num_input_tokens_seen": 58917336,
      "step": 101545
    },
    {
      "epoch": 15.125111706881144,
      "grad_norm": 9.6875,
      "learning_rate": 8.516476775066438e-06,
      "loss": 0.9341,
      "num_input_tokens_seen": 58920024,
      "step": 101550
    },
    {
      "epoch": 15.125856419422103,
      "grad_norm": 11.375,
      "learning_rate": 8.514033855984563e-06,
      "loss": 0.9824,
      "num_input_tokens_seen": 58922872,
      "step": 101555
    },
    {
      "epoch": 15.126601131963062,
      "grad_norm": 8.5,
      "learning_rate": 8.51159121541634e-06,
      "loss": 0.9475,
      "num_input_tokens_seen": 58926072,
      "step": 101560
    },
    {
      "epoch": 15.127345844504022,
      "grad_norm": 22.25,
      "learning_rate": 8.509148853403015e-06,
      "loss": 1.3207,
      "num_input_tokens_seen": 58929144,
      "step": 101565
    },
    {
      "epoch": 15.128090557044981,
      "grad_norm": 10.1875,
      "learning_rate": 8.50670676998587e-06,
      "loss": 1.2846,
      "num_input_tokens_seen": 58931928,
      "step": 101570
    },
    {
      "epoch": 15.12883526958594,
      "grad_norm": 29.125,
      "learning_rate": 8.504264965206148e-06,
      "loss": 1.0065,
      "num_input_tokens_seen": 58934712,
      "step": 101575
    },
    {
      "epoch": 15.129579982126899,
      "grad_norm": 8.5,
      "learning_rate": 8.5018234391051e-06,
      "loss": 1.0436,
      "num_input_tokens_seen": 58937560,
      "step": 101580
    },
    {
      "epoch": 15.130324694667857,
      "grad_norm": 14.1875,
      "learning_rate": 8.499382191723981e-06,
      "loss": 1.2526,
      "num_input_tokens_seen": 58940440,
      "step": 101585
    },
    {
      "epoch": 15.131069407208818,
      "grad_norm": 8.9375,
      "learning_rate": 8.496941223104032e-06,
      "loss": 0.826,
      "num_input_tokens_seen": 58943448,
      "step": 101590
    },
    {
      "epoch": 15.131814119749777,
      "grad_norm": 10.1875,
      "learning_rate": 8.494500533286487e-06,
      "loss": 1.2471,
      "num_input_tokens_seen": 58946296,
      "step": 101595
    },
    {
      "epoch": 15.132558832290735,
      "grad_norm": 6.4375,
      "learning_rate": 8.492060122312572e-06,
      "loss": 0.84,
      "num_input_tokens_seen": 58949240,
      "step": 101600
    },
    {
      "epoch": 15.133303544831694,
      "grad_norm": 8.5,
      "learning_rate": 8.489619990223533e-06,
      "loss": 0.8551,
      "num_input_tokens_seen": 58952152,
      "step": 101605
    },
    {
      "epoch": 15.134048257372655,
      "grad_norm": 12.4375,
      "learning_rate": 8.487180137060582e-06,
      "loss": 0.9376,
      "num_input_tokens_seen": 58955064,
      "step": 101610
    },
    {
      "epoch": 15.134792969913613,
      "grad_norm": 11.375,
      "learning_rate": 8.484740562864931e-06,
      "loss": 1.0259,
      "num_input_tokens_seen": 58958200,
      "step": 101615
    },
    {
      "epoch": 15.135537682454572,
      "grad_norm": 10.125,
      "learning_rate": 8.482301267677813e-06,
      "loss": 0.7597,
      "num_input_tokens_seen": 58960888,
      "step": 101620
    },
    {
      "epoch": 15.136282394995531,
      "grad_norm": 9.0,
      "learning_rate": 8.47986225154042e-06,
      "loss": 0.777,
      "num_input_tokens_seen": 58963832,
      "step": 101625
    },
    {
      "epoch": 15.137027107536491,
      "grad_norm": 7.625,
      "learning_rate": 8.477423514493967e-06,
      "loss": 1.1372,
      "num_input_tokens_seen": 58966456,
      "step": 101630
    },
    {
      "epoch": 15.13777182007745,
      "grad_norm": 14.6875,
      "learning_rate": 8.474985056579648e-06,
      "loss": 1.0879,
      "num_input_tokens_seen": 58969144,
      "step": 101635
    },
    {
      "epoch": 15.138516532618409,
      "grad_norm": 4.4375,
      "learning_rate": 8.47254687783867e-06,
      "loss": 0.8991,
      "num_input_tokens_seen": 58971864,
      "step": 101640
    },
    {
      "epoch": 15.139261245159368,
      "grad_norm": 8.875,
      "learning_rate": 8.470108978312211e-06,
      "loss": 1.221,
      "num_input_tokens_seen": 58974584,
      "step": 101645
    },
    {
      "epoch": 15.140005957700328,
      "grad_norm": 14.1875,
      "learning_rate": 8.46767135804146e-06,
      "loss": 1.086,
      "num_input_tokens_seen": 58977528,
      "step": 101650
    },
    {
      "epoch": 15.140750670241287,
      "grad_norm": 7.59375,
      "learning_rate": 8.465234017067595e-06,
      "loss": 0.9555,
      "num_input_tokens_seen": 58980216,
      "step": 101655
    },
    {
      "epoch": 15.141495382782246,
      "grad_norm": 15.0,
      "learning_rate": 8.462796955431801e-06,
      "loss": 0.9836,
      "num_input_tokens_seen": 58983032,
      "step": 101660
    },
    {
      "epoch": 15.142240095323205,
      "grad_norm": 8.375,
      "learning_rate": 8.460360173175244e-06,
      "loss": 0.7551,
      "num_input_tokens_seen": 58985752,
      "step": 101665
    },
    {
      "epoch": 15.142984807864165,
      "grad_norm": 13.25,
      "learning_rate": 8.457923670339085e-06,
      "loss": 1.081,
      "num_input_tokens_seen": 58988440,
      "step": 101670
    },
    {
      "epoch": 15.143729520405124,
      "grad_norm": 8.25,
      "learning_rate": 8.455487446964502e-06,
      "loss": 1.1503,
      "num_input_tokens_seen": 58991672,
      "step": 101675
    },
    {
      "epoch": 15.144474232946083,
      "grad_norm": 7.46875,
      "learning_rate": 8.453051503092632e-06,
      "loss": 1.1857,
      "num_input_tokens_seen": 58994360,
      "step": 101680
    },
    {
      "epoch": 15.145218945487041,
      "grad_norm": 9.4375,
      "learning_rate": 8.450615838764653e-06,
      "loss": 0.9981,
      "num_input_tokens_seen": 58997560,
      "step": 101685
    },
    {
      "epoch": 15.145963658028002,
      "grad_norm": 12.4375,
      "learning_rate": 8.448180454021695e-06,
      "loss": 1.0683,
      "num_input_tokens_seen": 59000792,
      "step": 101690
    },
    {
      "epoch": 15.14670837056896,
      "grad_norm": 8.375,
      "learning_rate": 8.445745348904898e-06,
      "loss": 1.3511,
      "num_input_tokens_seen": 59003768,
      "step": 101695
    },
    {
      "epoch": 15.14745308310992,
      "grad_norm": 10.0,
      "learning_rate": 8.443310523455416e-06,
      "loss": 1.0651,
      "num_input_tokens_seen": 59007032,
      "step": 101700
    },
    {
      "epoch": 15.148197795650878,
      "grad_norm": 11.3125,
      "learning_rate": 8.440875977714368e-06,
      "loss": 1.3242,
      "num_input_tokens_seen": 59010040,
      "step": 101705
    },
    {
      "epoch": 15.148942508191839,
      "grad_norm": 12.125,
      "learning_rate": 8.4384417117229e-06,
      "loss": 0.9961,
      "num_input_tokens_seen": 59013240,
      "step": 101710
    },
    {
      "epoch": 15.149687220732797,
      "grad_norm": 7.5,
      "learning_rate": 8.436007725522127e-06,
      "loss": 1.2405,
      "num_input_tokens_seen": 59016408,
      "step": 101715
    },
    {
      "epoch": 15.150431933273756,
      "grad_norm": 10.3125,
      "learning_rate": 8.433574019153167e-06,
      "loss": 0.7654,
      "num_input_tokens_seen": 59019448,
      "step": 101720
    },
    {
      "epoch": 15.151176645814715,
      "grad_norm": 8.75,
      "learning_rate": 8.43114059265713e-06,
      "loss": 1.1407,
      "num_input_tokens_seen": 59022584,
      "step": 101725
    },
    {
      "epoch": 15.151921358355676,
      "grad_norm": 14.75,
      "learning_rate": 8.428707446075138e-06,
      "loss": 1.1987,
      "num_input_tokens_seen": 59025304,
      "step": 101730
    },
    {
      "epoch": 15.152666070896634,
      "grad_norm": 8.75,
      "learning_rate": 8.426274579448293e-06,
      "loss": 0.9065,
      "num_input_tokens_seen": 59027896,
      "step": 101735
    },
    {
      "epoch": 15.153410783437593,
      "grad_norm": 11.625,
      "learning_rate": 8.423841992817688e-06,
      "loss": 1.111,
      "num_input_tokens_seen": 59030584,
      "step": 101740
    },
    {
      "epoch": 15.154155495978552,
      "grad_norm": 10.5,
      "learning_rate": 8.42140968622443e-06,
      "loss": 1.1027,
      "num_input_tokens_seen": 59033496,
      "step": 101745
    },
    {
      "epoch": 15.15490020851951,
      "grad_norm": 7.0,
      "learning_rate": 8.4189776597096e-06,
      "loss": 1.0224,
      "num_input_tokens_seen": 59036344,
      "step": 101750
    },
    {
      "epoch": 15.155644921060471,
      "grad_norm": 8.125,
      "learning_rate": 8.416545913314296e-06,
      "loss": 0.9728,
      "num_input_tokens_seen": 59039704,
      "step": 101755
    },
    {
      "epoch": 15.15638963360143,
      "grad_norm": 14.9375,
      "learning_rate": 8.414114447079588e-06,
      "loss": 1.2039,
      "num_input_tokens_seen": 59042840,
      "step": 101760
    },
    {
      "epoch": 15.157134346142389,
      "grad_norm": 16.375,
      "learning_rate": 8.411683261046569e-06,
      "loss": 0.8833,
      "num_input_tokens_seen": 59045912,
      "step": 101765
    },
    {
      "epoch": 15.157879058683347,
      "grad_norm": 5.6875,
      "learning_rate": 8.409252355256297e-06,
      "loss": 1.3638,
      "num_input_tokens_seen": 59048696,
      "step": 101770
    },
    {
      "epoch": 15.158623771224308,
      "grad_norm": 14.8125,
      "learning_rate": 8.40682172974984e-06,
      "loss": 1.1715,
      "num_input_tokens_seen": 59051416,
      "step": 101775
    },
    {
      "epoch": 15.159368483765267,
      "grad_norm": 11.9375,
      "learning_rate": 8.404391384568271e-06,
      "loss": 1.1349,
      "num_input_tokens_seen": 59054392,
      "step": 101780
    },
    {
      "epoch": 15.160113196306225,
      "grad_norm": 5.90625,
      "learning_rate": 8.401961319752646e-06,
      "loss": 0.9853,
      "num_input_tokens_seen": 59057400,
      "step": 101785
    },
    {
      "epoch": 15.160857908847184,
      "grad_norm": 7.75,
      "learning_rate": 8.399531535344013e-06,
      "loss": 0.9607,
      "num_input_tokens_seen": 59060600,
      "step": 101790
    },
    {
      "epoch": 15.161602621388145,
      "grad_norm": 13.875,
      "learning_rate": 8.397102031383414e-06,
      "loss": 1.3391,
      "num_input_tokens_seen": 59063480,
      "step": 101795
    },
    {
      "epoch": 15.162347333929103,
      "grad_norm": 9.5,
      "learning_rate": 8.39467280791191e-06,
      "loss": 0.8724,
      "num_input_tokens_seen": 59066264,
      "step": 101800
    },
    {
      "epoch": 15.163092046470062,
      "grad_norm": 10.0625,
      "learning_rate": 8.392243864970525e-06,
      "loss": 1.0997,
      "num_input_tokens_seen": 59069112,
      "step": 101805
    },
    {
      "epoch": 15.163836759011021,
      "grad_norm": 6.78125,
      "learning_rate": 8.389815202600306e-06,
      "loss": 0.9913,
      "num_input_tokens_seen": 59071928,
      "step": 101810
    },
    {
      "epoch": 15.164581471551982,
      "grad_norm": 7.21875,
      "learning_rate": 8.38738682084228e-06,
      "loss": 1.0168,
      "num_input_tokens_seen": 59074648,
      "step": 101815
    },
    {
      "epoch": 15.16532618409294,
      "grad_norm": 9.6875,
      "learning_rate": 8.38495871973746e-06,
      "loss": 0.7209,
      "num_input_tokens_seen": 59077304,
      "step": 101820
    },
    {
      "epoch": 15.166070896633899,
      "grad_norm": 9.0,
      "learning_rate": 8.382530899326885e-06,
      "loss": 0.8884,
      "num_input_tokens_seen": 59080024,
      "step": 101825
    },
    {
      "epoch": 15.166815609174858,
      "grad_norm": 7.59375,
      "learning_rate": 8.380103359651553e-06,
      "loss": 0.9072,
      "num_input_tokens_seen": 59082936,
      "step": 101830
    },
    {
      "epoch": 15.167560321715818,
      "grad_norm": 8.125,
      "learning_rate": 8.377676100752491e-06,
      "loss": 1.0972,
      "num_input_tokens_seen": 59085752,
      "step": 101835
    },
    {
      "epoch": 15.168305034256777,
      "grad_norm": 10.25,
      "learning_rate": 8.375249122670686e-06,
      "loss": 1.0207,
      "num_input_tokens_seen": 59088760,
      "step": 101840
    },
    {
      "epoch": 15.169049746797736,
      "grad_norm": 12.5625,
      "learning_rate": 8.372822425447164e-06,
      "loss": 0.7877,
      "num_input_tokens_seen": 59091448,
      "step": 101845
    },
    {
      "epoch": 15.169794459338695,
      "grad_norm": 11.0,
      "learning_rate": 8.370396009122902e-06,
      "loss": 1.0925,
      "num_input_tokens_seen": 59094328,
      "step": 101850
    },
    {
      "epoch": 15.170539171879655,
      "grad_norm": 9.5625,
      "learning_rate": 8.3679698737389e-06,
      "loss": 1.0475,
      "num_input_tokens_seen": 59097336,
      "step": 101855
    },
    {
      "epoch": 15.171283884420614,
      "grad_norm": 7.5,
      "learning_rate": 8.365544019336146e-06,
      "loss": 1.1801,
      "num_input_tokens_seen": 59100312,
      "step": 101860
    },
    {
      "epoch": 15.172028596961573,
      "grad_norm": 7.40625,
      "learning_rate": 8.363118445955609e-06,
      "loss": 1.0516,
      "num_input_tokens_seen": 59103192,
      "step": 101865
    },
    {
      "epoch": 15.172773309502531,
      "grad_norm": 11.375,
      "learning_rate": 8.360693153638285e-06,
      "loss": 0.7266,
      "num_input_tokens_seen": 59106264,
      "step": 101870
    },
    {
      "epoch": 15.173518022043492,
      "grad_norm": 13.25,
      "learning_rate": 8.35826814242513e-06,
      "loss": 0.9135,
      "num_input_tokens_seen": 59109016,
      "step": 101875
    },
    {
      "epoch": 15.17426273458445,
      "grad_norm": 5.9375,
      "learning_rate": 8.355843412357131e-06,
      "loss": 1.0855,
      "num_input_tokens_seen": 59112024,
      "step": 101880
    },
    {
      "epoch": 15.17500744712541,
      "grad_norm": 13.125,
      "learning_rate": 8.353418963475232e-06,
      "loss": 1.1412,
      "num_input_tokens_seen": 59114936,
      "step": 101885
    },
    {
      "epoch": 15.175752159666368,
      "grad_norm": 11.0,
      "learning_rate": 8.350994795820407e-06,
      "loss": 1.2196,
      "num_input_tokens_seen": 59118040,
      "step": 101890
    },
    {
      "epoch": 15.176496872207329,
      "grad_norm": 6.5625,
      "learning_rate": 8.348570909433607e-06,
      "loss": 0.9045,
      "num_input_tokens_seen": 59121112,
      "step": 101895
    },
    {
      "epoch": 15.177241584748288,
      "grad_norm": 6.90625,
      "learning_rate": 8.346147304355767e-06,
      "loss": 0.9939,
      "num_input_tokens_seen": 59123960,
      "step": 101900
    },
    {
      "epoch": 15.177986297289246,
      "grad_norm": 10.0625,
      "learning_rate": 8.343723980627848e-06,
      "loss": 0.9924,
      "num_input_tokens_seen": 59127160,
      "step": 101905
    },
    {
      "epoch": 15.178731009830205,
      "grad_norm": 18.625,
      "learning_rate": 8.34130093829078e-06,
      "loss": 0.9908,
      "num_input_tokens_seen": 59130200,
      "step": 101910
    },
    {
      "epoch": 15.179475722371166,
      "grad_norm": 7.8125,
      "learning_rate": 8.338878177385508e-06,
      "loss": 0.8511,
      "num_input_tokens_seen": 59133592,
      "step": 101915
    },
    {
      "epoch": 15.180220434912124,
      "grad_norm": 10.5625,
      "learning_rate": 8.336455697952956e-06,
      "loss": 1.1302,
      "num_input_tokens_seen": 59136536,
      "step": 101920
    },
    {
      "epoch": 15.180965147453083,
      "grad_norm": 16.875,
      "learning_rate": 8.33403350003405e-06,
      "loss": 1.0141,
      "num_input_tokens_seen": 59139256,
      "step": 101925
    },
    {
      "epoch": 15.181709859994042,
      "grad_norm": 5.71875,
      "learning_rate": 8.3316115836697e-06,
      "loss": 1.1588,
      "num_input_tokens_seen": 59142328,
      "step": 101930
    },
    {
      "epoch": 15.182454572535,
      "grad_norm": 10.5625,
      "learning_rate": 8.32918994890084e-06,
      "loss": 1.174,
      "num_input_tokens_seen": 59145656,
      "step": 101935
    },
    {
      "epoch": 15.183199285075961,
      "grad_norm": 9.5,
      "learning_rate": 8.32676859576837e-06,
      "loss": 0.823,
      "num_input_tokens_seen": 59148600,
      "step": 101940
    },
    {
      "epoch": 15.18394399761692,
      "grad_norm": 7.5625,
      "learning_rate": 8.324347524313192e-06,
      "loss": 1.2658,
      "num_input_tokens_seen": 59151352,
      "step": 101945
    },
    {
      "epoch": 15.184688710157879,
      "grad_norm": 5.125,
      "learning_rate": 8.321926734576223e-06,
      "loss": 0.8454,
      "num_input_tokens_seen": 59154072,
      "step": 101950
    },
    {
      "epoch": 15.185433422698837,
      "grad_norm": 12.25,
      "learning_rate": 8.319506226598342e-06,
      "loss": 1.1994,
      "num_input_tokens_seen": 59156920,
      "step": 101955
    },
    {
      "epoch": 15.186178135239798,
      "grad_norm": 14.875,
      "learning_rate": 8.317086000420459e-06,
      "loss": 1.3159,
      "num_input_tokens_seen": 59159672,
      "step": 101960
    },
    {
      "epoch": 15.186922847780757,
      "grad_norm": 7.21875,
      "learning_rate": 8.314666056083444e-06,
      "loss": 0.6814,
      "num_input_tokens_seen": 59162328,
      "step": 101965
    },
    {
      "epoch": 15.187667560321715,
      "grad_norm": 12.0625,
      "learning_rate": 8.312246393628195e-06,
      "loss": 1.0234,
      "num_input_tokens_seen": 59165400,
      "step": 101970
    },
    {
      "epoch": 15.188412272862674,
      "grad_norm": 14.5,
      "learning_rate": 8.309827013095584e-06,
      "loss": 1.1533,
      "num_input_tokens_seen": 59168376,
      "step": 101975
    },
    {
      "epoch": 15.189156985403635,
      "grad_norm": 9.4375,
      "learning_rate": 8.30740791452648e-06,
      "loss": 1.1567,
      "num_input_tokens_seen": 59171352,
      "step": 101980
    },
    {
      "epoch": 15.189901697944594,
      "grad_norm": 7.0625,
      "learning_rate": 8.304989097961748e-06,
      "loss": 1.1719,
      "num_input_tokens_seen": 59174328,
      "step": 101985
    },
    {
      "epoch": 15.190646410485552,
      "grad_norm": 18.5,
      "learning_rate": 8.302570563442263e-06,
      "loss": 1.3257,
      "num_input_tokens_seen": 59177144,
      "step": 101990
    },
    {
      "epoch": 15.191391123026511,
      "grad_norm": 12.375,
      "learning_rate": 8.300152311008883e-06,
      "loss": 1.2723,
      "num_input_tokens_seen": 59180024,
      "step": 101995
    },
    {
      "epoch": 15.192135835567472,
      "grad_norm": 13.5,
      "learning_rate": 8.297734340702443e-06,
      "loss": 1.4278,
      "num_input_tokens_seen": 59183160,
      "step": 102000
    },
    {
      "epoch": 15.19288054810843,
      "grad_norm": 18.625,
      "learning_rate": 8.295316652563817e-06,
      "loss": 1.0712,
      "num_input_tokens_seen": 59185784,
      "step": 102005
    },
    {
      "epoch": 15.19362526064939,
      "grad_norm": 13.875,
      "learning_rate": 8.292899246633828e-06,
      "loss": 1.1672,
      "num_input_tokens_seen": 59188568,
      "step": 102010
    },
    {
      "epoch": 15.194369973190348,
      "grad_norm": 7.59375,
      "learning_rate": 8.290482122953336e-06,
      "loss": 1.116,
      "num_input_tokens_seen": 59191480,
      "step": 102015
    },
    {
      "epoch": 15.195114685731308,
      "grad_norm": 18.375,
      "learning_rate": 8.288065281563164e-06,
      "loss": 1.5365,
      "num_input_tokens_seen": 59194328,
      "step": 102020
    },
    {
      "epoch": 15.195859398272267,
      "grad_norm": 7.96875,
      "learning_rate": 8.285648722504136e-06,
      "loss": 0.9082,
      "num_input_tokens_seen": 59197560,
      "step": 102025
    },
    {
      "epoch": 15.196604110813226,
      "grad_norm": 6.21875,
      "learning_rate": 8.283232445817094e-06,
      "loss": 0.7491,
      "num_input_tokens_seen": 59200440,
      "step": 102030
    },
    {
      "epoch": 15.197348823354185,
      "grad_norm": 18.875,
      "learning_rate": 8.280816451542841e-06,
      "loss": 1.1895,
      "num_input_tokens_seen": 59203352,
      "step": 102035
    },
    {
      "epoch": 15.198093535895145,
      "grad_norm": 8.9375,
      "learning_rate": 8.278400739722211e-06,
      "loss": 1.1292,
      "num_input_tokens_seen": 59206232,
      "step": 102040
    },
    {
      "epoch": 15.198838248436104,
      "grad_norm": 14.0,
      "learning_rate": 8.275985310396003e-06,
      "loss": 1.0826,
      "num_input_tokens_seen": 59209240,
      "step": 102045
    },
    {
      "epoch": 15.199582960977063,
      "grad_norm": 10.6875,
      "learning_rate": 8.273570163605026e-06,
      "loss": 1.1761,
      "num_input_tokens_seen": 59211864,
      "step": 102050
    },
    {
      "epoch": 15.200327673518021,
      "grad_norm": 4.5,
      "learning_rate": 8.271155299390082e-06,
      "loss": 0.8642,
      "num_input_tokens_seen": 59214680,
      "step": 102055
    },
    {
      "epoch": 15.201072386058982,
      "grad_norm": 11.5625,
      "learning_rate": 8.26874071779196e-06,
      "loss": 0.9457,
      "num_input_tokens_seen": 59217528,
      "step": 102060
    },
    {
      "epoch": 15.20181709859994,
      "grad_norm": 7.65625,
      "learning_rate": 8.266326418851467e-06,
      "loss": 1.0246,
      "num_input_tokens_seen": 59220248,
      "step": 102065
    },
    {
      "epoch": 15.2025618111409,
      "grad_norm": 8.0625,
      "learning_rate": 8.26391240260937e-06,
      "loss": 1.3197,
      "num_input_tokens_seen": 59223384,
      "step": 102070
    },
    {
      "epoch": 15.203306523681858,
      "grad_norm": 12.375,
      "learning_rate": 8.261498669106473e-06,
      "loss": 1.0183,
      "num_input_tokens_seen": 59226584,
      "step": 102075
    },
    {
      "epoch": 15.204051236222819,
      "grad_norm": 7.53125,
      "learning_rate": 8.259085218383536e-06,
      "loss": 0.9226,
      "num_input_tokens_seen": 59229336,
      "step": 102080
    },
    {
      "epoch": 15.204795948763778,
      "grad_norm": 13.5625,
      "learning_rate": 8.256672050481348e-06,
      "loss": 1.2263,
      "num_input_tokens_seen": 59232312,
      "step": 102085
    },
    {
      "epoch": 15.205540661304736,
      "grad_norm": 6.96875,
      "learning_rate": 8.254259165440662e-06,
      "loss": 0.7374,
      "num_input_tokens_seen": 59235000,
      "step": 102090
    },
    {
      "epoch": 15.206285373845695,
      "grad_norm": 11.125,
      "learning_rate": 8.251846563302253e-06,
      "loss": 1.2097,
      "num_input_tokens_seen": 59238040,
      "step": 102095
    },
    {
      "epoch": 15.207030086386654,
      "grad_norm": 14.6875,
      "learning_rate": 8.249434244106875e-06,
      "loss": 0.9891,
      "num_input_tokens_seen": 59240824,
      "step": 102100
    },
    {
      "epoch": 15.207774798927614,
      "grad_norm": 5.59375,
      "learning_rate": 8.247022207895271e-06,
      "loss": 1.131,
      "num_input_tokens_seen": 59244056,
      "step": 102105
    },
    {
      "epoch": 15.208519511468573,
      "grad_norm": 12.0625,
      "learning_rate": 8.244610454708213e-06,
      "loss": 1.4221,
      "num_input_tokens_seen": 59246744,
      "step": 102110
    },
    {
      "epoch": 15.209264224009532,
      "grad_norm": 5.375,
      "learning_rate": 8.242198984586427e-06,
      "loss": 1.0652,
      "num_input_tokens_seen": 59249496,
      "step": 102115
    },
    {
      "epoch": 15.21000893655049,
      "grad_norm": 10.0,
      "learning_rate": 8.239787797570661e-06,
      "loss": 0.9637,
      "num_input_tokens_seen": 59252216,
      "step": 102120
    },
    {
      "epoch": 15.210753649091451,
      "grad_norm": 8.9375,
      "learning_rate": 8.237376893701635e-06,
      "loss": 1.1646,
      "num_input_tokens_seen": 59254776,
      "step": 102125
    },
    {
      "epoch": 15.21149836163241,
      "grad_norm": 9.625,
      "learning_rate": 8.2349662730201e-06,
      "loss": 1.5958,
      "num_input_tokens_seen": 59257624,
      "step": 102130
    },
    {
      "epoch": 15.212243074173369,
      "grad_norm": 11.5625,
      "learning_rate": 8.232555935566769e-06,
      "loss": 1.2502,
      "num_input_tokens_seen": 59260568,
      "step": 102135
    },
    {
      "epoch": 15.212987786714327,
      "grad_norm": 10.5625,
      "learning_rate": 8.230145881382357e-06,
      "loss": 1.385,
      "num_input_tokens_seen": 59263416,
      "step": 102140
    },
    {
      "epoch": 15.213732499255288,
      "grad_norm": 9.4375,
      "learning_rate": 8.227736110507592e-06,
      "loss": 1.0666,
      "num_input_tokens_seen": 59266360,
      "step": 102145
    },
    {
      "epoch": 15.214477211796247,
      "grad_norm": 10.375,
      "learning_rate": 8.225326622983173e-06,
      "loss": 0.9841,
      "num_input_tokens_seen": 59268952,
      "step": 102150
    },
    {
      "epoch": 15.215221924337206,
      "grad_norm": 12.875,
      "learning_rate": 8.222917418849819e-06,
      "loss": 1.2678,
      "num_input_tokens_seen": 59271672,
      "step": 102155
    },
    {
      "epoch": 15.215966636878164,
      "grad_norm": 11.125,
      "learning_rate": 8.220508498148213e-06,
      "loss": 0.9843,
      "num_input_tokens_seen": 59274424,
      "step": 102160
    },
    {
      "epoch": 15.216711349419125,
      "grad_norm": 7.78125,
      "learning_rate": 8.218099860919074e-06,
      "loss": 0.9519,
      "num_input_tokens_seen": 59277144,
      "step": 102165
    },
    {
      "epoch": 15.217456061960084,
      "grad_norm": 12.5625,
      "learning_rate": 8.215691507203072e-06,
      "loss": 1.3062,
      "num_input_tokens_seen": 59279896,
      "step": 102170
    },
    {
      "epoch": 15.218200774501042,
      "grad_norm": 6.0625,
      "learning_rate": 8.213283437040911e-06,
      "loss": 1.1774,
      "num_input_tokens_seen": 59283032,
      "step": 102175
    },
    {
      "epoch": 15.218945487042001,
      "grad_norm": 27.375,
      "learning_rate": 8.210875650473266e-06,
      "loss": 1.407,
      "num_input_tokens_seen": 59285912,
      "step": 102180
    },
    {
      "epoch": 15.219690199582962,
      "grad_norm": 15.6875,
      "learning_rate": 8.208468147540812e-06,
      "loss": 1.0648,
      "num_input_tokens_seen": 59288856,
      "step": 102185
    },
    {
      "epoch": 15.22043491212392,
      "grad_norm": 10.5,
      "learning_rate": 8.206060928284223e-06,
      "loss": 1.4223,
      "num_input_tokens_seen": 59291672,
      "step": 102190
    },
    {
      "epoch": 15.22117962466488,
      "grad_norm": 18.125,
      "learning_rate": 8.20365399274416e-06,
      "loss": 1.3956,
      "num_input_tokens_seen": 59294616,
      "step": 102195
    },
    {
      "epoch": 15.221924337205838,
      "grad_norm": 15.3125,
      "learning_rate": 8.201247340961296e-06,
      "loss": 1.1199,
      "num_input_tokens_seen": 59297304,
      "step": 102200
    },
    {
      "epoch": 15.222669049746798,
      "grad_norm": 7.75,
      "learning_rate": 8.19884097297628e-06,
      "loss": 0.9377,
      "num_input_tokens_seen": 59300216,
      "step": 102205
    },
    {
      "epoch": 15.223413762287757,
      "grad_norm": 5.1875,
      "learning_rate": 8.196434888829774e-06,
      "loss": 1.3263,
      "num_input_tokens_seen": 59302968,
      "step": 102210
    },
    {
      "epoch": 15.224158474828716,
      "grad_norm": 17.0,
      "learning_rate": 8.194029088562425e-06,
      "loss": 1.2929,
      "num_input_tokens_seen": 59305656,
      "step": 102215
    },
    {
      "epoch": 15.224903187369675,
      "grad_norm": 12.3125,
      "learning_rate": 8.191623572214865e-06,
      "loss": 0.9847,
      "num_input_tokens_seen": 59308344,
      "step": 102220
    },
    {
      "epoch": 15.225647899910635,
      "grad_norm": 6.125,
      "learning_rate": 8.18921833982775e-06,
      "loss": 0.8683,
      "num_input_tokens_seen": 59311320,
      "step": 102225
    },
    {
      "epoch": 15.226392612451594,
      "grad_norm": 12.4375,
      "learning_rate": 8.186813391441697e-06,
      "loss": 1.1195,
      "num_input_tokens_seen": 59314360,
      "step": 102230
    },
    {
      "epoch": 15.227137324992553,
      "grad_norm": 13.4375,
      "learning_rate": 8.184408727097354e-06,
      "loss": 1.3158,
      "num_input_tokens_seen": 59317272,
      "step": 102235
    },
    {
      "epoch": 15.227882037533512,
      "grad_norm": 21.5,
      "learning_rate": 8.182004346835323e-06,
      "loss": 1.0703,
      "num_input_tokens_seen": 59320088,
      "step": 102240
    },
    {
      "epoch": 15.228626750074472,
      "grad_norm": 14.6875,
      "learning_rate": 8.179600250696245e-06,
      "loss": 1.5019,
      "num_input_tokens_seen": 59323448,
      "step": 102245
    },
    {
      "epoch": 15.22937146261543,
      "grad_norm": 4.84375,
      "learning_rate": 8.177196438720724e-06,
      "loss": 1.3057,
      "num_input_tokens_seen": 59326296,
      "step": 102250
    },
    {
      "epoch": 15.23011617515639,
      "grad_norm": 13.0,
      "learning_rate": 8.174792910949376e-06,
      "loss": 1.1859,
      "num_input_tokens_seen": 59329176,
      "step": 102255
    },
    {
      "epoch": 15.230860887697348,
      "grad_norm": 12.8125,
      "learning_rate": 8.172389667422797e-06,
      "loss": 0.8254,
      "num_input_tokens_seen": 59332152,
      "step": 102260
    },
    {
      "epoch": 15.231605600238307,
      "grad_norm": 5.5,
      "learning_rate": 8.169986708181584e-06,
      "loss": 1.0227,
      "num_input_tokens_seen": 59334968,
      "step": 102265
    },
    {
      "epoch": 15.232350312779268,
      "grad_norm": 10.6875,
      "learning_rate": 8.167584033266349e-06,
      "loss": 1.0245,
      "num_input_tokens_seen": 59337816,
      "step": 102270
    },
    {
      "epoch": 15.233095025320226,
      "grad_norm": 13.75,
      "learning_rate": 8.165181642717668e-06,
      "loss": 1.0738,
      "num_input_tokens_seen": 59340728,
      "step": 102275
    },
    {
      "epoch": 15.233839737861185,
      "grad_norm": 12.8125,
      "learning_rate": 8.162779536576138e-06,
      "loss": 0.7691,
      "num_input_tokens_seen": 59343352,
      "step": 102280
    },
    {
      "epoch": 15.234584450402144,
      "grad_norm": 7.46875,
      "learning_rate": 8.160377714882327e-06,
      "loss": 0.9721,
      "num_input_tokens_seen": 59346456,
      "step": 102285
    },
    {
      "epoch": 15.235329162943104,
      "grad_norm": 10.25,
      "learning_rate": 8.15797617767683e-06,
      "loss": 0.9724,
      "num_input_tokens_seen": 59349112,
      "step": 102290
    },
    {
      "epoch": 15.236073875484063,
      "grad_norm": 14.6875,
      "learning_rate": 8.155574925000207e-06,
      "loss": 0.8732,
      "num_input_tokens_seen": 59351960,
      "step": 102295
    },
    {
      "epoch": 15.236818588025022,
      "grad_norm": 16.125,
      "learning_rate": 8.153173956893018e-06,
      "loss": 0.9873,
      "num_input_tokens_seen": 59354712,
      "step": 102300
    },
    {
      "epoch": 15.23756330056598,
      "grad_norm": 8.8125,
      "learning_rate": 8.15077327339584e-06,
      "loss": 0.9939,
      "num_input_tokens_seen": 59357496,
      "step": 102305
    },
    {
      "epoch": 15.238308013106941,
      "grad_norm": 9.3125,
      "learning_rate": 8.148372874549224e-06,
      "loss": 1.2671,
      "num_input_tokens_seen": 59360248,
      "step": 102310
    },
    {
      "epoch": 15.2390527256479,
      "grad_norm": 7.6875,
      "learning_rate": 8.145972760393711e-06,
      "loss": 1.0502,
      "num_input_tokens_seen": 59363384,
      "step": 102315
    },
    {
      "epoch": 15.239797438188859,
      "grad_norm": 8.125,
      "learning_rate": 8.143572930969866e-06,
      "loss": 1.0939,
      "num_input_tokens_seen": 59366424,
      "step": 102320
    },
    {
      "epoch": 15.240542150729818,
      "grad_norm": 15.25,
      "learning_rate": 8.141173386318226e-06,
      "loss": 1.2761,
      "num_input_tokens_seen": 59369528,
      "step": 102325
    },
    {
      "epoch": 15.241286863270778,
      "grad_norm": 4.375,
      "learning_rate": 8.13877412647932e-06,
      "loss": 1.0147,
      "num_input_tokens_seen": 59373304,
      "step": 102330
    },
    {
      "epoch": 15.242031575811737,
      "grad_norm": 10.875,
      "learning_rate": 8.136375151493695e-06,
      "loss": 1.4439,
      "num_input_tokens_seen": 59376216,
      "step": 102335
    },
    {
      "epoch": 15.242776288352696,
      "grad_norm": 7.90625,
      "learning_rate": 8.13397646140187e-06,
      "loss": 1.1242,
      "num_input_tokens_seen": 59379160,
      "step": 102340
    },
    {
      "epoch": 15.243521000893654,
      "grad_norm": 7.21875,
      "learning_rate": 8.131578056244365e-06,
      "loss": 1.2258,
      "num_input_tokens_seen": 59382456,
      "step": 102345
    },
    {
      "epoch": 15.244265713434615,
      "grad_norm": 6.28125,
      "learning_rate": 8.129179936061715e-06,
      "loss": 1.0184,
      "num_input_tokens_seen": 59385400,
      "step": 102350
    },
    {
      "epoch": 15.245010425975574,
      "grad_norm": 8.6875,
      "learning_rate": 8.126782100894411e-06,
      "loss": 0.982,
      "num_input_tokens_seen": 59388472,
      "step": 102355
    },
    {
      "epoch": 15.245755138516532,
      "grad_norm": 14.5,
      "learning_rate": 8.124384550782985e-06,
      "loss": 1.2139,
      "num_input_tokens_seen": 59391448,
      "step": 102360
    },
    {
      "epoch": 15.246499851057491,
      "grad_norm": 7.90625,
      "learning_rate": 8.12198728576792e-06,
      "loss": 0.9939,
      "num_input_tokens_seen": 59394200,
      "step": 102365
    },
    {
      "epoch": 15.247244563598452,
      "grad_norm": 7.8125,
      "learning_rate": 8.119590305889737e-06,
      "loss": 0.9666,
      "num_input_tokens_seen": 59396856,
      "step": 102370
    },
    {
      "epoch": 15.24798927613941,
      "grad_norm": 6.4375,
      "learning_rate": 8.117193611188917e-06,
      "loss": 1.1201,
      "num_input_tokens_seen": 59399608,
      "step": 102375
    },
    {
      "epoch": 15.24873398868037,
      "grad_norm": 11.9375,
      "learning_rate": 8.114797201705954e-06,
      "loss": 1.0993,
      "num_input_tokens_seen": 59402584,
      "step": 102380
    },
    {
      "epoch": 15.249478701221328,
      "grad_norm": 16.875,
      "learning_rate": 8.112401077481329e-06,
      "loss": 0.9246,
      "num_input_tokens_seen": 59406136,
      "step": 102385
    },
    {
      "epoch": 15.250223413762289,
      "grad_norm": 10.6875,
      "learning_rate": 8.110005238555517e-06,
      "loss": 0.9034,
      "num_input_tokens_seen": 59408856,
      "step": 102390
    },
    {
      "epoch": 15.250968126303247,
      "grad_norm": 10.5,
      "learning_rate": 8.107609684969008e-06,
      "loss": 1.201,
      "num_input_tokens_seen": 59411608,
      "step": 102395
    },
    {
      "epoch": 15.251712838844206,
      "grad_norm": 12.625,
      "learning_rate": 8.105214416762255e-06,
      "loss": 1.0184,
      "num_input_tokens_seen": 59414360,
      "step": 102400
    },
    {
      "epoch": 15.252457551385165,
      "grad_norm": 10.3125,
      "learning_rate": 8.102819433975745e-06,
      "loss": 0.9933,
      "num_input_tokens_seen": 59417112,
      "step": 102405
    },
    {
      "epoch": 15.253202263926125,
      "grad_norm": 10.8125,
      "learning_rate": 8.100424736649918e-06,
      "loss": 1.1271,
      "num_input_tokens_seen": 59419832,
      "step": 102410
    },
    {
      "epoch": 15.253946976467084,
      "grad_norm": 10.125,
      "learning_rate": 8.098030324825246e-06,
      "loss": 1.0403,
      "num_input_tokens_seen": 59423000,
      "step": 102415
    },
    {
      "epoch": 15.254691689008043,
      "grad_norm": 16.25,
      "learning_rate": 8.095636198542173e-06,
      "loss": 1.1722,
      "num_input_tokens_seen": 59425880,
      "step": 102420
    },
    {
      "epoch": 15.255436401549002,
      "grad_norm": 6.75,
      "learning_rate": 8.093242357841136e-06,
      "loss": 1.0571,
      "num_input_tokens_seen": 59428856,
      "step": 102425
    },
    {
      "epoch": 15.256181114089962,
      "grad_norm": 10.3125,
      "learning_rate": 8.090848802762596e-06,
      "loss": 1.1584,
      "num_input_tokens_seen": 59431896,
      "step": 102430
    },
    {
      "epoch": 15.256925826630921,
      "grad_norm": 15.625,
      "learning_rate": 8.088455533346973e-06,
      "loss": 1.3603,
      "num_input_tokens_seen": 59434744,
      "step": 102435
    },
    {
      "epoch": 15.25767053917188,
      "grad_norm": 9.375,
      "learning_rate": 8.086062549634712e-06,
      "loss": 0.8982,
      "num_input_tokens_seen": 59437432,
      "step": 102440
    },
    {
      "epoch": 15.258415251712838,
      "grad_norm": 10.6875,
      "learning_rate": 8.083669851666235e-06,
      "loss": 1.0252,
      "num_input_tokens_seen": 59440632,
      "step": 102445
    },
    {
      "epoch": 15.259159964253797,
      "grad_norm": 4.8125,
      "learning_rate": 8.081277439481961e-06,
      "loss": 0.9619,
      "num_input_tokens_seen": 59443224,
      "step": 102450
    },
    {
      "epoch": 15.259904676794758,
      "grad_norm": 10.625,
      "learning_rate": 8.078885313122311e-06,
      "loss": 0.9559,
      "num_input_tokens_seen": 59446456,
      "step": 102455
    },
    {
      "epoch": 15.260649389335716,
      "grad_norm": 4.8125,
      "learning_rate": 8.076493472627687e-06,
      "loss": 0.9807,
      "num_input_tokens_seen": 59449688,
      "step": 102460
    },
    {
      "epoch": 15.261394101876675,
      "grad_norm": 13.25,
      "learning_rate": 8.074101918038512e-06,
      "loss": 1.2728,
      "num_input_tokens_seen": 59452632,
      "step": 102465
    },
    {
      "epoch": 15.262138814417634,
      "grad_norm": 8.0,
      "learning_rate": 8.071710649395178e-06,
      "loss": 0.848,
      "num_input_tokens_seen": 59455704,
      "step": 102470
    },
    {
      "epoch": 15.262883526958595,
      "grad_norm": 6.1875,
      "learning_rate": 8.069319666738093e-06,
      "loss": 0.7347,
      "num_input_tokens_seen": 59458520,
      "step": 102475
    },
    {
      "epoch": 15.263628239499553,
      "grad_norm": 6.375,
      "learning_rate": 8.066928970107638e-06,
      "loss": 0.8789,
      "num_input_tokens_seen": 59461496,
      "step": 102480
    },
    {
      "epoch": 15.264372952040512,
      "grad_norm": 7.9375,
      "learning_rate": 8.064538559544213e-06,
      "loss": 0.852,
      "num_input_tokens_seen": 59464280,
      "step": 102485
    },
    {
      "epoch": 15.26511766458147,
      "grad_norm": 13.4375,
      "learning_rate": 8.06214843508819e-06,
      "loss": 1.04,
      "num_input_tokens_seen": 59466968,
      "step": 102490
    },
    {
      "epoch": 15.265862377122431,
      "grad_norm": 22.875,
      "learning_rate": 8.059758596779965e-06,
      "loss": 0.9511,
      "num_input_tokens_seen": 59469976,
      "step": 102495
    },
    {
      "epoch": 15.26660708966339,
      "grad_norm": 10.3125,
      "learning_rate": 8.057369044659899e-06,
      "loss": 1.0423,
      "num_input_tokens_seen": 59472792,
      "step": 102500
    },
    {
      "epoch": 15.267351802204349,
      "grad_norm": 6.4375,
      "learning_rate": 8.054979778768354e-06,
      "loss": 0.8438,
      "num_input_tokens_seen": 59475480,
      "step": 102505
    },
    {
      "epoch": 15.268096514745308,
      "grad_norm": 10.6875,
      "learning_rate": 8.052590799145715e-06,
      "loss": 1.4766,
      "num_input_tokens_seen": 59478584,
      "step": 102510
    },
    {
      "epoch": 15.268841227286268,
      "grad_norm": 13.125,
      "learning_rate": 8.050202105832327e-06,
      "loss": 0.9079,
      "num_input_tokens_seen": 59481720,
      "step": 102515
    },
    {
      "epoch": 15.269585939827227,
      "grad_norm": 10.125,
      "learning_rate": 8.047813698868548e-06,
      "loss": 1.3607,
      "num_input_tokens_seen": 59484472,
      "step": 102520
    },
    {
      "epoch": 15.270330652368186,
      "grad_norm": 9.6875,
      "learning_rate": 8.045425578294719e-06,
      "loss": 1.1311,
      "num_input_tokens_seen": 59487192,
      "step": 102525
    },
    {
      "epoch": 15.271075364909144,
      "grad_norm": 8.75,
      "learning_rate": 8.043037744151203e-06,
      "loss": 1.344,
      "num_input_tokens_seen": 59490232,
      "step": 102530
    },
    {
      "epoch": 15.271820077450105,
      "grad_norm": 21.25,
      "learning_rate": 8.040650196478319e-06,
      "loss": 1.21,
      "num_input_tokens_seen": 59493304,
      "step": 102535
    },
    {
      "epoch": 15.272564789991064,
      "grad_norm": 13.125,
      "learning_rate": 8.038262935316423e-06,
      "loss": 1.2575,
      "num_input_tokens_seen": 59496312,
      "step": 102540
    },
    {
      "epoch": 15.273309502532022,
      "grad_norm": 13.3125,
      "learning_rate": 8.035875960705835e-06,
      "loss": 1.0284,
      "num_input_tokens_seen": 59499416,
      "step": 102545
    },
    {
      "epoch": 15.274054215072981,
      "grad_norm": 11.0,
      "learning_rate": 8.033489272686872e-06,
      "loss": 1.0595,
      "num_input_tokens_seen": 59502488,
      "step": 102550
    },
    {
      "epoch": 15.274798927613942,
      "grad_norm": 14.375,
      "learning_rate": 8.031102871299876e-06,
      "loss": 1.1447,
      "num_input_tokens_seen": 59505240,
      "step": 102555
    },
    {
      "epoch": 15.2755436401549,
      "grad_norm": 6.4375,
      "learning_rate": 8.02871675658514e-06,
      "loss": 0.9437,
      "num_input_tokens_seen": 59507960,
      "step": 102560
    },
    {
      "epoch": 15.27628835269586,
      "grad_norm": 9.1875,
      "learning_rate": 8.026330928582993e-06,
      "loss": 1.1896,
      "num_input_tokens_seen": 59510904,
      "step": 102565
    },
    {
      "epoch": 15.277033065236818,
      "grad_norm": 18.125,
      "learning_rate": 8.023945387333722e-06,
      "loss": 1.3036,
      "num_input_tokens_seen": 59513688,
      "step": 102570
    },
    {
      "epoch": 15.277777777777779,
      "grad_norm": 13.0,
      "learning_rate": 8.021560132877653e-06,
      "loss": 1.2559,
      "num_input_tokens_seen": 59516600,
      "step": 102575
    },
    {
      "epoch": 15.278522490318737,
      "grad_norm": 5.40625,
      "learning_rate": 8.019175165255069e-06,
      "loss": 0.7616,
      "num_input_tokens_seen": 59519224,
      "step": 102580
    },
    {
      "epoch": 15.279267202859696,
      "grad_norm": 9.6875,
      "learning_rate": 8.016790484506261e-06,
      "loss": 1.4591,
      "num_input_tokens_seen": 59522488,
      "step": 102585
    },
    {
      "epoch": 15.280011915400655,
      "grad_norm": 15.6875,
      "learning_rate": 8.014406090671516e-06,
      "loss": 1.2517,
      "num_input_tokens_seen": 59525432,
      "step": 102590
    },
    {
      "epoch": 15.280756627941615,
      "grad_norm": 6.28125,
      "learning_rate": 8.012021983791112e-06,
      "loss": 0.9526,
      "num_input_tokens_seen": 59528184,
      "step": 102595
    },
    {
      "epoch": 15.281501340482574,
      "grad_norm": 14.125,
      "learning_rate": 8.009638163905337e-06,
      "loss": 1.1773,
      "num_input_tokens_seen": 59530904,
      "step": 102600
    },
    {
      "epoch": 15.282246053023533,
      "grad_norm": 11.625,
      "learning_rate": 8.007254631054448e-06,
      "loss": 0.9944,
      "num_input_tokens_seen": 59533784,
      "step": 102605
    },
    {
      "epoch": 15.282990765564492,
      "grad_norm": 8.0,
      "learning_rate": 8.00487138527873e-06,
      "loss": 1.3393,
      "num_input_tokens_seen": 59536728,
      "step": 102610
    },
    {
      "epoch": 15.283735478105452,
      "grad_norm": 8.3125,
      "learning_rate": 8.002488426618429e-06,
      "loss": 0.9713,
      "num_input_tokens_seen": 59539672,
      "step": 102615
    },
    {
      "epoch": 15.284480190646411,
      "grad_norm": 8.4375,
      "learning_rate": 8.000105755113818e-06,
      "loss": 0.9828,
      "num_input_tokens_seen": 59542744,
      "step": 102620
    },
    {
      "epoch": 15.28522490318737,
      "grad_norm": 10.5,
      "learning_rate": 7.997723370805143e-06,
      "loss": 1.111,
      "num_input_tokens_seen": 59545528,
      "step": 102625
    },
    {
      "epoch": 15.285969615728328,
      "grad_norm": 8.0625,
      "learning_rate": 7.995341273732642e-06,
      "loss": 1.1415,
      "num_input_tokens_seen": 59548536,
      "step": 102630
    },
    {
      "epoch": 15.286714328269287,
      "grad_norm": 11.8125,
      "learning_rate": 7.992959463936578e-06,
      "loss": 1.0312,
      "num_input_tokens_seen": 59551480,
      "step": 102635
    },
    {
      "epoch": 15.287459040810248,
      "grad_norm": 4.96875,
      "learning_rate": 7.990577941457175e-06,
      "loss": 0.8631,
      "num_input_tokens_seen": 59554136,
      "step": 102640
    },
    {
      "epoch": 15.288203753351207,
      "grad_norm": 11.9375,
      "learning_rate": 7.988196706334666e-06,
      "loss": 1.2379,
      "num_input_tokens_seen": 59556920,
      "step": 102645
    },
    {
      "epoch": 15.288948465892165,
      "grad_norm": 10.1875,
      "learning_rate": 7.985815758609289e-06,
      "loss": 0.9366,
      "num_input_tokens_seen": 59559768,
      "step": 102650
    },
    {
      "epoch": 15.289693178433124,
      "grad_norm": 18.25,
      "learning_rate": 7.983435098321263e-06,
      "loss": 0.9421,
      "num_input_tokens_seen": 59562872,
      "step": 102655
    },
    {
      "epoch": 15.290437890974085,
      "grad_norm": 5.59375,
      "learning_rate": 7.981054725510805e-06,
      "loss": 0.6811,
      "num_input_tokens_seen": 59565912,
      "step": 102660
    },
    {
      "epoch": 15.291182603515043,
      "grad_norm": 9.1875,
      "learning_rate": 7.978674640218126e-06,
      "loss": 1.1745,
      "num_input_tokens_seen": 59568600,
      "step": 102665
    },
    {
      "epoch": 15.291927316056002,
      "grad_norm": 13.6875,
      "learning_rate": 7.976294842483446e-06,
      "loss": 1.4434,
      "num_input_tokens_seen": 59571576,
      "step": 102670
    },
    {
      "epoch": 15.29267202859696,
      "grad_norm": 9.5625,
      "learning_rate": 7.97391533234695e-06,
      "loss": 0.9715,
      "num_input_tokens_seen": 59574744,
      "step": 102675
    },
    {
      "epoch": 15.293416741137921,
      "grad_norm": 10.125,
      "learning_rate": 7.971536109848862e-06,
      "loss": 1.2925,
      "num_input_tokens_seen": 59577912,
      "step": 102680
    },
    {
      "epoch": 15.29416145367888,
      "grad_norm": 4.53125,
      "learning_rate": 7.969157175029354e-06,
      "loss": 0.9674,
      "num_input_tokens_seen": 59580696,
      "step": 102685
    },
    {
      "epoch": 15.294906166219839,
      "grad_norm": 11.1875,
      "learning_rate": 7.966778527928637e-06,
      "loss": 1.2964,
      "num_input_tokens_seen": 59583256,
      "step": 102690
    },
    {
      "epoch": 15.295650878760798,
      "grad_norm": 10.125,
      "learning_rate": 7.964400168586875e-06,
      "loss": 1.1702,
      "num_input_tokens_seen": 59585976,
      "step": 102695
    },
    {
      "epoch": 15.296395591301758,
      "grad_norm": 14.25,
      "learning_rate": 7.962022097044266e-06,
      "loss": 1.1385,
      "num_input_tokens_seen": 59588984,
      "step": 102700
    },
    {
      "epoch": 15.297140303842717,
      "grad_norm": 8.9375,
      "learning_rate": 7.959644313340978e-06,
      "loss": 1.0662,
      "num_input_tokens_seen": 59592088,
      "step": 102705
    },
    {
      "epoch": 15.297885016383676,
      "grad_norm": 11.4375,
      "learning_rate": 7.95726681751718e-06,
      "loss": 0.767,
      "num_input_tokens_seen": 59594840,
      "step": 102710
    },
    {
      "epoch": 15.298629728924634,
      "grad_norm": 12.125,
      "learning_rate": 7.95488960961304e-06,
      "loss": 1.0069,
      "num_input_tokens_seen": 59597464,
      "step": 102715
    },
    {
      "epoch": 15.299374441465595,
      "grad_norm": 10.9375,
      "learning_rate": 7.952512689668703e-06,
      "loss": 1.2896,
      "num_input_tokens_seen": 59599992,
      "step": 102720
    },
    {
      "epoch": 15.300119154006554,
      "grad_norm": 5.53125,
      "learning_rate": 7.95013605772435e-06,
      "loss": 1.0025,
      "num_input_tokens_seen": 59602936,
      "step": 102725
    },
    {
      "epoch": 15.300863866547513,
      "grad_norm": 11.4375,
      "learning_rate": 7.947759713820111e-06,
      "loss": 1.3241,
      "num_input_tokens_seen": 59605656,
      "step": 102730
    },
    {
      "epoch": 15.301608579088471,
      "grad_norm": 8.6875,
      "learning_rate": 7.945383657996148e-06,
      "loss": 0.8938,
      "num_input_tokens_seen": 59608472,
      "step": 102735
    },
    {
      "epoch": 15.302353291629432,
      "grad_norm": 11.625,
      "learning_rate": 7.943007890292593e-06,
      "loss": 1.1398,
      "num_input_tokens_seen": 59610968,
      "step": 102740
    },
    {
      "epoch": 15.30309800417039,
      "grad_norm": 9.4375,
      "learning_rate": 7.940632410749577e-06,
      "loss": 1.0185,
      "num_input_tokens_seen": 59613912,
      "step": 102745
    },
    {
      "epoch": 15.30384271671135,
      "grad_norm": 9.5,
      "learning_rate": 7.938257219407246e-06,
      "loss": 1.2878,
      "num_input_tokens_seen": 59616696,
      "step": 102750
    },
    {
      "epoch": 15.304587429252308,
      "grad_norm": 11.0625,
      "learning_rate": 7.93588231630571e-06,
      "loss": 1.0933,
      "num_input_tokens_seen": 59619704,
      "step": 102755
    },
    {
      "epoch": 15.305332141793269,
      "grad_norm": 11.5,
      "learning_rate": 7.933507701485108e-06,
      "loss": 1.1501,
      "num_input_tokens_seen": 59622232,
      "step": 102760
    },
    {
      "epoch": 15.306076854334227,
      "grad_norm": 8.25,
      "learning_rate": 7.93113337498554e-06,
      "loss": 0.9953,
      "num_input_tokens_seen": 59625304,
      "step": 102765
    },
    {
      "epoch": 15.306821566875186,
      "grad_norm": 17.0,
      "learning_rate": 7.928759336847133e-06,
      "loss": 1.3662,
      "num_input_tokens_seen": 59628408,
      "step": 102770
    },
    {
      "epoch": 15.307566279416145,
      "grad_norm": 13.625,
      "learning_rate": 7.926385587109986e-06,
      "loss": 1.2989,
      "num_input_tokens_seen": 59630936,
      "step": 102775
    },
    {
      "epoch": 15.308310991957104,
      "grad_norm": 7.625,
      "learning_rate": 7.924012125814203e-06,
      "loss": 1.1976,
      "num_input_tokens_seen": 59633752,
      "step": 102780
    },
    {
      "epoch": 15.309055704498064,
      "grad_norm": 12.375,
      "learning_rate": 7.92163895299988e-06,
      "loss": 0.8562,
      "num_input_tokens_seen": 59636600,
      "step": 102785
    },
    {
      "epoch": 15.309800417039023,
      "grad_norm": 9.4375,
      "learning_rate": 7.919266068707099e-06,
      "loss": 0.9365,
      "num_input_tokens_seen": 59639448,
      "step": 102790
    },
    {
      "epoch": 15.310545129579982,
      "grad_norm": 9.4375,
      "learning_rate": 7.916893472975967e-06,
      "loss": 0.6419,
      "num_input_tokens_seen": 59642264,
      "step": 102795
    },
    {
      "epoch": 15.31128984212094,
      "grad_norm": 11.875,
      "learning_rate": 7.91452116584655e-06,
      "loss": 1.0931,
      "num_input_tokens_seen": 59645272,
      "step": 102800
    },
    {
      "epoch": 15.312034554661901,
      "grad_norm": 7.875,
      "learning_rate": 7.912149147358938e-06,
      "loss": 1.1925,
      "num_input_tokens_seen": 59648184,
      "step": 102805
    },
    {
      "epoch": 15.31277926720286,
      "grad_norm": 8.125,
      "learning_rate": 7.909777417553193e-06,
      "loss": 0.9386,
      "num_input_tokens_seen": 59651352,
      "step": 102810
    },
    {
      "epoch": 15.313523979743819,
      "grad_norm": 6.0625,
      "learning_rate": 7.907405976469397e-06,
      "loss": 1.0416,
      "num_input_tokens_seen": 59654520,
      "step": 102815
    },
    {
      "epoch": 15.314268692284777,
      "grad_norm": 12.9375,
      "learning_rate": 7.905034824147605e-06,
      "loss": 1.0213,
      "num_input_tokens_seen": 59657560,
      "step": 102820
    },
    {
      "epoch": 15.315013404825738,
      "grad_norm": 15.25,
      "learning_rate": 7.902663960627869e-06,
      "loss": 1.095,
      "num_input_tokens_seen": 59660472,
      "step": 102825
    },
    {
      "epoch": 15.315758117366697,
      "grad_norm": 10.125,
      "learning_rate": 7.900293385950253e-06,
      "loss": 1.1583,
      "num_input_tokens_seen": 59663288,
      "step": 102830
    },
    {
      "epoch": 15.316502829907655,
      "grad_norm": 14.875,
      "learning_rate": 7.897923100154794e-06,
      "loss": 1.0852,
      "num_input_tokens_seen": 59666296,
      "step": 102835
    },
    {
      "epoch": 15.317247542448614,
      "grad_norm": 22.75,
      "learning_rate": 7.895553103281552e-06,
      "loss": 1.0893,
      "num_input_tokens_seen": 59669080,
      "step": 102840
    },
    {
      "epoch": 15.317992254989575,
      "grad_norm": 12.0,
      "learning_rate": 7.893183395370554e-06,
      "loss": 1.1256,
      "num_input_tokens_seen": 59671704,
      "step": 102845
    },
    {
      "epoch": 15.318736967530533,
      "grad_norm": 12.8125,
      "learning_rate": 7.890813976461836e-06,
      "loss": 1.1966,
      "num_input_tokens_seen": 59674328,
      "step": 102850
    },
    {
      "epoch": 15.319481680071492,
      "grad_norm": 10.8125,
      "learning_rate": 7.888444846595422e-06,
      "loss": 1.0643,
      "num_input_tokens_seen": 59677176,
      "step": 102855
    },
    {
      "epoch": 15.320226392612451,
      "grad_norm": 5.0625,
      "learning_rate": 7.886076005811346e-06,
      "loss": 0.864,
      "num_input_tokens_seen": 59679960,
      "step": 102860
    },
    {
      "epoch": 15.320971105153411,
      "grad_norm": 11.625,
      "learning_rate": 7.883707454149621e-06,
      "loss": 1.1181,
      "num_input_tokens_seen": 59682584,
      "step": 102865
    },
    {
      "epoch": 15.32171581769437,
      "grad_norm": 4.625,
      "learning_rate": 7.881339191650256e-06,
      "loss": 1.0208,
      "num_input_tokens_seen": 59685464,
      "step": 102870
    },
    {
      "epoch": 15.322460530235329,
      "grad_norm": 11.9375,
      "learning_rate": 7.878971218353275e-06,
      "loss": 0.9731,
      "num_input_tokens_seen": 59688472,
      "step": 102875
    },
    {
      "epoch": 15.323205242776288,
      "grad_norm": 9.1875,
      "learning_rate": 7.876603534298666e-06,
      "loss": 1.0114,
      "num_input_tokens_seen": 59691160,
      "step": 102880
    },
    {
      "epoch": 15.323949955317248,
      "grad_norm": 8.75,
      "learning_rate": 7.874236139526445e-06,
      "loss": 1.2616,
      "num_input_tokens_seen": 59693880,
      "step": 102885
    },
    {
      "epoch": 15.324694667858207,
      "grad_norm": 12.8125,
      "learning_rate": 7.87186903407659e-06,
      "loss": 0.8632,
      "num_input_tokens_seen": 59696920,
      "step": 102890
    },
    {
      "epoch": 15.325439380399166,
      "grad_norm": 8.1875,
      "learning_rate": 7.869502217989108e-06,
      "loss": 1.1314,
      "num_input_tokens_seen": 59699704,
      "step": 102895
    },
    {
      "epoch": 15.326184092940125,
      "grad_norm": 9.0625,
      "learning_rate": 7.867135691303975e-06,
      "loss": 1.1189,
      "num_input_tokens_seen": 59702584,
      "step": 102900
    },
    {
      "epoch": 15.326928805481085,
      "grad_norm": 5.59375,
      "learning_rate": 7.864769454061163e-06,
      "loss": 1.1081,
      "num_input_tokens_seen": 59705592,
      "step": 102905
    },
    {
      "epoch": 15.327673518022044,
      "grad_norm": 9.0625,
      "learning_rate": 7.862403506300664e-06,
      "loss": 1.0328,
      "num_input_tokens_seen": 59708600,
      "step": 102910
    },
    {
      "epoch": 15.328418230563003,
      "grad_norm": 13.125,
      "learning_rate": 7.86003784806244e-06,
      "loss": 1.1589,
      "num_input_tokens_seen": 59711320,
      "step": 102915
    },
    {
      "epoch": 15.329162943103961,
      "grad_norm": 12.9375,
      "learning_rate": 7.857672479386458e-06,
      "loss": 0.7618,
      "num_input_tokens_seen": 59714136,
      "step": 102920
    },
    {
      "epoch": 15.329907655644922,
      "grad_norm": 5.0,
      "learning_rate": 7.855307400312667e-06,
      "loss": 1.2136,
      "num_input_tokens_seen": 59717176,
      "step": 102925
    },
    {
      "epoch": 15.33065236818588,
      "grad_norm": 9.4375,
      "learning_rate": 7.85294261088104e-06,
      "loss": 0.8899,
      "num_input_tokens_seen": 59720056,
      "step": 102930
    },
    {
      "epoch": 15.33139708072684,
      "grad_norm": 5.40625,
      "learning_rate": 7.850578111131513e-06,
      "loss": 0.9501,
      "num_input_tokens_seen": 59722968,
      "step": 102935
    },
    {
      "epoch": 15.332141793267798,
      "grad_norm": 10.125,
      "learning_rate": 7.848213901104045e-06,
      "loss": 0.8695,
      "num_input_tokens_seen": 59725784,
      "step": 102940
    },
    {
      "epoch": 15.332886505808759,
      "grad_norm": 10.375,
      "learning_rate": 7.845849980838574e-06,
      "loss": 0.7066,
      "num_input_tokens_seen": 59728472,
      "step": 102945
    },
    {
      "epoch": 15.333631218349717,
      "grad_norm": 9.375,
      "learning_rate": 7.843486350375023e-06,
      "loss": 1.0175,
      "num_input_tokens_seen": 59731320,
      "step": 102950
    },
    {
      "epoch": 15.334375930890676,
      "grad_norm": 11.125,
      "learning_rate": 7.84112300975334e-06,
      "loss": 0.8701,
      "num_input_tokens_seen": 59734168,
      "step": 102955
    },
    {
      "epoch": 15.335120643431635,
      "grad_norm": 4.84375,
      "learning_rate": 7.838759959013439e-06,
      "loss": 0.7901,
      "num_input_tokens_seen": 59737016,
      "step": 102960
    },
    {
      "epoch": 15.335865355972594,
      "grad_norm": 8.375,
      "learning_rate": 7.836397198195252e-06,
      "loss": 0.9454,
      "num_input_tokens_seen": 59739960,
      "step": 102965
    },
    {
      "epoch": 15.336610068513554,
      "grad_norm": 21.375,
      "learning_rate": 7.83403472733869e-06,
      "loss": 1.1169,
      "num_input_tokens_seen": 59742904,
      "step": 102970
    },
    {
      "epoch": 15.337354781054513,
      "grad_norm": 3.28125,
      "learning_rate": 7.83167254648366e-06,
      "loss": 0.9249,
      "num_input_tokens_seen": 59745560,
      "step": 102975
    },
    {
      "epoch": 15.338099493595472,
      "grad_norm": 9.625,
      "learning_rate": 7.829310655670077e-06,
      "loss": 1.5561,
      "num_input_tokens_seen": 59748600,
      "step": 102980
    },
    {
      "epoch": 15.33884420613643,
      "grad_norm": 13.0,
      "learning_rate": 7.82694905493784e-06,
      "loss": 1.2777,
      "num_input_tokens_seen": 59751512,
      "step": 102985
    },
    {
      "epoch": 15.339588918677391,
      "grad_norm": 18.5,
      "learning_rate": 7.824587744326847e-06,
      "loss": 1.0123,
      "num_input_tokens_seen": 59754392,
      "step": 102990
    },
    {
      "epoch": 15.34033363121835,
      "grad_norm": 8.625,
      "learning_rate": 7.822226723876976e-06,
      "loss": 1.0915,
      "num_input_tokens_seen": 59757240,
      "step": 102995
    },
    {
      "epoch": 15.341078343759309,
      "grad_norm": 10.1875,
      "learning_rate": 7.819865993628139e-06,
      "loss": 0.9595,
      "num_input_tokens_seen": 59760280,
      "step": 103000
    },
    {
      "epoch": 15.341823056300267,
      "grad_norm": 10.4375,
      "learning_rate": 7.817505553620194e-06,
      "loss": 1.0108,
      "num_input_tokens_seen": 59763000,
      "step": 103005
    },
    {
      "epoch": 15.342567768841228,
      "grad_norm": 7.75,
      "learning_rate": 7.815145403893037e-06,
      "loss": 1.0732,
      "num_input_tokens_seen": 59765880,
      "step": 103010
    },
    {
      "epoch": 15.343312481382187,
      "grad_norm": 8.5,
      "learning_rate": 7.812785544486526e-06,
      "loss": 1.1379,
      "num_input_tokens_seen": 59768920,
      "step": 103015
    },
    {
      "epoch": 15.344057193923145,
      "grad_norm": 11.0625,
      "learning_rate": 7.81042597544054e-06,
      "loss": 1.2017,
      "num_input_tokens_seen": 59771896,
      "step": 103020
    },
    {
      "epoch": 15.344801906464104,
      "grad_norm": 6.59375,
      "learning_rate": 7.808066696794938e-06,
      "loss": 0.8891,
      "num_input_tokens_seen": 59775064,
      "step": 103025
    },
    {
      "epoch": 15.345546619005065,
      "grad_norm": 7.125,
      "learning_rate": 7.805707708589569e-06,
      "loss": 1.2109,
      "num_input_tokens_seen": 59777912,
      "step": 103030
    },
    {
      "epoch": 15.346291331546023,
      "grad_norm": 13.125,
      "learning_rate": 7.8033490108643e-06,
      "loss": 0.9959,
      "num_input_tokens_seen": 59780824,
      "step": 103035
    },
    {
      "epoch": 15.347036044086982,
      "grad_norm": 16.5,
      "learning_rate": 7.80099060365897e-06,
      "loss": 1.1638,
      "num_input_tokens_seen": 59783640,
      "step": 103040
    },
    {
      "epoch": 15.347780756627941,
      "grad_norm": 8.625,
      "learning_rate": 7.798632487013427e-06,
      "loss": 1.1921,
      "num_input_tokens_seen": 59786776,
      "step": 103045
    },
    {
      "epoch": 15.348525469168901,
      "grad_norm": 13.625,
      "learning_rate": 7.796274660967496e-06,
      "loss": 1.2343,
      "num_input_tokens_seen": 59790296,
      "step": 103050
    },
    {
      "epoch": 15.34927018170986,
      "grad_norm": 7.46875,
      "learning_rate": 7.793917125561027e-06,
      "loss": 0.7861,
      "num_input_tokens_seen": 59793112,
      "step": 103055
    },
    {
      "epoch": 15.350014894250819,
      "grad_norm": 14.5,
      "learning_rate": 7.79155988083384e-06,
      "loss": 0.8386,
      "num_input_tokens_seen": 59796120,
      "step": 103060
    },
    {
      "epoch": 15.350759606791778,
      "grad_norm": 6.03125,
      "learning_rate": 7.78920292682575e-06,
      "loss": 0.8897,
      "num_input_tokens_seen": 59799128,
      "step": 103065
    },
    {
      "epoch": 15.351504319332738,
      "grad_norm": 19.0,
      "learning_rate": 7.786846263576594e-06,
      "loss": 0.9106,
      "num_input_tokens_seen": 59801976,
      "step": 103070
    },
    {
      "epoch": 15.352249031873697,
      "grad_norm": 6.21875,
      "learning_rate": 7.784489891126167e-06,
      "loss": 0.9493,
      "num_input_tokens_seen": 59805176,
      "step": 103075
    },
    {
      "epoch": 15.352993744414656,
      "grad_norm": 9.1875,
      "learning_rate": 7.782133809514297e-06,
      "loss": 1.1948,
      "num_input_tokens_seen": 59807832,
      "step": 103080
    },
    {
      "epoch": 15.353738456955615,
      "grad_norm": 14.9375,
      "learning_rate": 7.779778018780765e-06,
      "loss": 1.063,
      "num_input_tokens_seen": 59810680,
      "step": 103085
    },
    {
      "epoch": 15.354483169496575,
      "grad_norm": 9.5,
      "learning_rate": 7.77742251896539e-06,
      "loss": 1.3307,
      "num_input_tokens_seen": 59813592,
      "step": 103090
    },
    {
      "epoch": 15.355227882037534,
      "grad_norm": 10.3125,
      "learning_rate": 7.775067310107953e-06,
      "loss": 1.1568,
      "num_input_tokens_seen": 59816408,
      "step": 103095
    },
    {
      "epoch": 15.355972594578493,
      "grad_norm": 8.875,
      "learning_rate": 7.772712392248251e-06,
      "loss": 0.8411,
      "num_input_tokens_seen": 59819544,
      "step": 103100
    },
    {
      "epoch": 15.356717307119451,
      "grad_norm": 8.125,
      "learning_rate": 7.770357765426068e-06,
      "loss": 1.1748,
      "num_input_tokens_seen": 59822616,
      "step": 103105
    },
    {
      "epoch": 15.357462019660412,
      "grad_norm": 4.84375,
      "learning_rate": 7.768003429681175e-06,
      "loss": 0.9173,
      "num_input_tokens_seen": 59825528,
      "step": 103110
    },
    {
      "epoch": 15.35820673220137,
      "grad_norm": 9.4375,
      "learning_rate": 7.765649385053353e-06,
      "loss": 0.8536,
      "num_input_tokens_seen": 59828376,
      "step": 103115
    },
    {
      "epoch": 15.35895144474233,
      "grad_norm": 16.875,
      "learning_rate": 7.76329563158236e-06,
      "loss": 1.3086,
      "num_input_tokens_seen": 59831192,
      "step": 103120
    },
    {
      "epoch": 15.359696157283288,
      "grad_norm": 13.0,
      "learning_rate": 7.760942169307975e-06,
      "loss": 1.0651,
      "num_input_tokens_seen": 59834168,
      "step": 103125
    },
    {
      "epoch": 15.360440869824249,
      "grad_norm": 8.0,
      "learning_rate": 7.758588998269944e-06,
      "loss": 0.8658,
      "num_input_tokens_seen": 59837432,
      "step": 103130
    },
    {
      "epoch": 15.361185582365207,
      "grad_norm": 17.0,
      "learning_rate": 7.756236118508036e-06,
      "loss": 1.0903,
      "num_input_tokens_seen": 59840184,
      "step": 103135
    },
    {
      "epoch": 15.361930294906166,
      "grad_norm": 10.875,
      "learning_rate": 7.753883530061987e-06,
      "loss": 0.9395,
      "num_input_tokens_seen": 59843224,
      "step": 103140
    },
    {
      "epoch": 15.362675007447125,
      "grad_norm": 8.875,
      "learning_rate": 7.751531232971554e-06,
      "loss": 1.297,
      "num_input_tokens_seen": 59846264,
      "step": 103145
    },
    {
      "epoch": 15.363419719988084,
      "grad_norm": 9.3125,
      "learning_rate": 7.749179227276471e-06,
      "loss": 0.8935,
      "num_input_tokens_seen": 59849144,
      "step": 103150
    },
    {
      "epoch": 15.364164432529044,
      "grad_norm": 11.5625,
      "learning_rate": 7.746827513016464e-06,
      "loss": 0.8063,
      "num_input_tokens_seen": 59852024,
      "step": 103155
    },
    {
      "epoch": 15.364909145070003,
      "grad_norm": 11.4375,
      "learning_rate": 7.744476090231275e-06,
      "loss": 0.9909,
      "num_input_tokens_seen": 59855032,
      "step": 103160
    },
    {
      "epoch": 15.365653857610962,
      "grad_norm": 22.75,
      "learning_rate": 7.742124958960622e-06,
      "loss": 0.9671,
      "num_input_tokens_seen": 59857816,
      "step": 103165
    },
    {
      "epoch": 15.36639857015192,
      "grad_norm": 18.25,
      "learning_rate": 7.739774119244233e-06,
      "loss": 1.3128,
      "num_input_tokens_seen": 59860728,
      "step": 103170
    },
    {
      "epoch": 15.367143282692881,
      "grad_norm": 12.125,
      "learning_rate": 7.737423571121818e-06,
      "loss": 0.8657,
      "num_input_tokens_seen": 59863448,
      "step": 103175
    },
    {
      "epoch": 15.36788799523384,
      "grad_norm": 14.125,
      "learning_rate": 7.73507331463309e-06,
      "loss": 1.443,
      "num_input_tokens_seen": 59866264,
      "step": 103180
    },
    {
      "epoch": 15.368632707774799,
      "grad_norm": 11.375,
      "learning_rate": 7.732723349817747e-06,
      "loss": 0.9328,
      "num_input_tokens_seen": 59869176,
      "step": 103185
    },
    {
      "epoch": 15.369377420315757,
      "grad_norm": 6.875,
      "learning_rate": 7.730373676715488e-06,
      "loss": 0.8214,
      "num_input_tokens_seen": 59871960,
      "step": 103190
    },
    {
      "epoch": 15.370122132856718,
      "grad_norm": 14.625,
      "learning_rate": 7.728024295366018e-06,
      "loss": 1.0162,
      "num_input_tokens_seen": 59874872,
      "step": 103195
    },
    {
      "epoch": 15.370866845397677,
      "grad_norm": 9.4375,
      "learning_rate": 7.725675205809019e-06,
      "loss": 0.8341,
      "num_input_tokens_seen": 59877464,
      "step": 103200
    },
    {
      "epoch": 15.371611557938635,
      "grad_norm": 9.1875,
      "learning_rate": 7.723326408084186e-06,
      "loss": 1.5836,
      "num_input_tokens_seen": 59880504,
      "step": 103205
    },
    {
      "epoch": 15.372356270479594,
      "grad_norm": 6.59375,
      "learning_rate": 7.720977902231189e-06,
      "loss": 1.1101,
      "num_input_tokens_seen": 59883320,
      "step": 103210
    },
    {
      "epoch": 15.373100983020555,
      "grad_norm": 7.5,
      "learning_rate": 7.718629688289713e-06,
      "loss": 0.7118,
      "num_input_tokens_seen": 59886136,
      "step": 103215
    },
    {
      "epoch": 15.373845695561513,
      "grad_norm": 12.9375,
      "learning_rate": 7.716281766299419e-06,
      "loss": 1.2645,
      "num_input_tokens_seen": 59889304,
      "step": 103220
    },
    {
      "epoch": 15.374590408102472,
      "grad_norm": 9.8125,
      "learning_rate": 7.713934136299985e-06,
      "loss": 1.0213,
      "num_input_tokens_seen": 59892088,
      "step": 103225
    },
    {
      "epoch": 15.375335120643431,
      "grad_norm": 11.5625,
      "learning_rate": 7.711586798331066e-06,
      "loss": 1.2601,
      "num_input_tokens_seen": 59895000,
      "step": 103230
    },
    {
      "epoch": 15.376079833184392,
      "grad_norm": 7.03125,
      "learning_rate": 7.70923975243231e-06,
      "loss": 0.8541,
      "num_input_tokens_seen": 59897752,
      "step": 103235
    },
    {
      "epoch": 15.37682454572535,
      "grad_norm": 9.125,
      "learning_rate": 7.70689299864338e-06,
      "loss": 1.2189,
      "num_input_tokens_seen": 59900984,
      "step": 103240
    },
    {
      "epoch": 15.377569258266309,
      "grad_norm": 4.9375,
      "learning_rate": 7.704546537003918e-06,
      "loss": 0.8628,
      "num_input_tokens_seen": 59903928,
      "step": 103245
    },
    {
      "epoch": 15.378313970807268,
      "grad_norm": 9.4375,
      "learning_rate": 7.702200367553563e-06,
      "loss": 1.2506,
      "num_input_tokens_seen": 59906616,
      "step": 103250
    },
    {
      "epoch": 15.379058683348228,
      "grad_norm": 11.6875,
      "learning_rate": 7.699854490331948e-06,
      "loss": 1.1271,
      "num_input_tokens_seen": 59909400,
      "step": 103255
    },
    {
      "epoch": 15.379803395889187,
      "grad_norm": 10.875,
      "learning_rate": 7.697508905378712e-06,
      "loss": 1.0377,
      "num_input_tokens_seen": 59912504,
      "step": 103260
    },
    {
      "epoch": 15.380548108430146,
      "grad_norm": 10.9375,
      "learning_rate": 7.69516361273348e-06,
      "loss": 1.3587,
      "num_input_tokens_seen": 59915224,
      "step": 103265
    },
    {
      "epoch": 15.381292820971105,
      "grad_norm": 11.875,
      "learning_rate": 7.692818612435862e-06,
      "loss": 1.0946,
      "num_input_tokens_seen": 59918456,
      "step": 103270
    },
    {
      "epoch": 15.382037533512065,
      "grad_norm": 9.375,
      "learning_rate": 7.690473904525491e-06,
      "loss": 1.0529,
      "num_input_tokens_seen": 59921272,
      "step": 103275
    },
    {
      "epoch": 15.382782246053024,
      "grad_norm": 10.0,
      "learning_rate": 7.688129489041963e-06,
      "loss": 1.0042,
      "num_input_tokens_seen": 59924120,
      "step": 103280
    },
    {
      "epoch": 15.383526958593983,
      "grad_norm": 9.0625,
      "learning_rate": 7.685785366024901e-06,
      "loss": 1.1071,
      "num_input_tokens_seen": 59927064,
      "step": 103285
    },
    {
      "epoch": 15.384271671134941,
      "grad_norm": 6.1875,
      "learning_rate": 7.683441535513888e-06,
      "loss": 1.3026,
      "num_input_tokens_seen": 59929848,
      "step": 103290
    },
    {
      "epoch": 15.3850163836759,
      "grad_norm": 7.5625,
      "learning_rate": 7.681097997548539e-06,
      "loss": 1.3344,
      "num_input_tokens_seen": 59932856,
      "step": 103295
    },
    {
      "epoch": 15.38576109621686,
      "grad_norm": 11.75,
      "learning_rate": 7.678754752168438e-06,
      "loss": 1.3316,
      "num_input_tokens_seen": 59935864,
      "step": 103300
    },
    {
      "epoch": 15.38650580875782,
      "grad_norm": 13.5625,
      "learning_rate": 7.676411799413163e-06,
      "loss": 0.9787,
      "num_input_tokens_seen": 59938904,
      "step": 103305
    },
    {
      "epoch": 15.387250521298778,
      "grad_norm": 8.875,
      "learning_rate": 7.674069139322312e-06,
      "loss": 1.0473,
      "num_input_tokens_seen": 59941816,
      "step": 103310
    },
    {
      "epoch": 15.387995233839739,
      "grad_norm": 14.125,
      "learning_rate": 7.671726771935453e-06,
      "loss": 1.1893,
      "num_input_tokens_seen": 59944792,
      "step": 103315
    },
    {
      "epoch": 15.388739946380698,
      "grad_norm": 13.5625,
      "learning_rate": 7.669384697292158e-06,
      "loss": 1.2514,
      "num_input_tokens_seen": 59947896,
      "step": 103320
    },
    {
      "epoch": 15.389484658921656,
      "grad_norm": 12.625,
      "learning_rate": 7.667042915431987e-06,
      "loss": 1.1772,
      "num_input_tokens_seen": 59950712,
      "step": 103325
    },
    {
      "epoch": 15.390229371462615,
      "grad_norm": 12.0625,
      "learning_rate": 7.66470142639452e-06,
      "loss": 1.2371,
      "num_input_tokens_seen": 59953752,
      "step": 103330
    },
    {
      "epoch": 15.390974084003574,
      "grad_norm": 13.3125,
      "learning_rate": 7.662360230219293e-06,
      "loss": 1.052,
      "num_input_tokens_seen": 59956824,
      "step": 103335
    },
    {
      "epoch": 15.391718796544534,
      "grad_norm": 8.3125,
      "learning_rate": 7.660019326945874e-06,
      "loss": 1.1673,
      "num_input_tokens_seen": 59959576,
      "step": 103340
    },
    {
      "epoch": 15.392463509085493,
      "grad_norm": 17.375,
      "learning_rate": 7.657678716613808e-06,
      "loss": 0.7293,
      "num_input_tokens_seen": 59962232,
      "step": 103345
    },
    {
      "epoch": 15.393208221626452,
      "grad_norm": 3.6875,
      "learning_rate": 7.655338399262627e-06,
      "loss": 1.1727,
      "num_input_tokens_seen": 59964856,
      "step": 103350
    },
    {
      "epoch": 15.39395293416741,
      "grad_norm": 13.6875,
      "learning_rate": 7.652998374931882e-06,
      "loss": 1.0692,
      "num_input_tokens_seen": 59967960,
      "step": 103355
    },
    {
      "epoch": 15.394697646708371,
      "grad_norm": 7.03125,
      "learning_rate": 7.65065864366109e-06,
      "loss": 1.2107,
      "num_input_tokens_seen": 59970808,
      "step": 103360
    },
    {
      "epoch": 15.39544235924933,
      "grad_norm": 9.75,
      "learning_rate": 7.648319205489798e-06,
      "loss": 1.1153,
      "num_input_tokens_seen": 59973656,
      "step": 103365
    },
    {
      "epoch": 15.396187071790289,
      "grad_norm": 8.5625,
      "learning_rate": 7.645980060457517e-06,
      "loss": 1.2595,
      "num_input_tokens_seen": 59977048,
      "step": 103370
    },
    {
      "epoch": 15.396931784331247,
      "grad_norm": 5.5625,
      "learning_rate": 7.643641208603764e-06,
      "loss": 1.2683,
      "num_input_tokens_seen": 59980024,
      "step": 103375
    },
    {
      "epoch": 15.397676496872208,
      "grad_norm": 11.375,
      "learning_rate": 7.641302649968043e-06,
      "loss": 0.7836,
      "num_input_tokens_seen": 59983000,
      "step": 103380
    },
    {
      "epoch": 15.398421209413167,
      "grad_norm": 8.5625,
      "learning_rate": 7.638964384589881e-06,
      "loss": 1.1242,
      "num_input_tokens_seen": 59985848,
      "step": 103385
    },
    {
      "epoch": 15.399165921954125,
      "grad_norm": 14.625,
      "learning_rate": 7.63662641250877e-06,
      "loss": 1.2831,
      "num_input_tokens_seen": 59988664,
      "step": 103390
    },
    {
      "epoch": 15.399910634495084,
      "grad_norm": 11.3125,
      "learning_rate": 7.6342887337642e-06,
      "loss": 1.133,
      "num_input_tokens_seen": 59991672,
      "step": 103395
    },
    {
      "epoch": 15.400655347036045,
      "grad_norm": 16.375,
      "learning_rate": 7.631951348395683e-06,
      "loss": 1.3197,
      "num_input_tokens_seen": 59994360,
      "step": 103400
    },
    {
      "epoch": 15.401400059577004,
      "grad_norm": 4.6875,
      "learning_rate": 7.629614256442685e-06,
      "loss": 0.8708,
      "num_input_tokens_seen": 59997464,
      "step": 103405
    },
    {
      "epoch": 15.402144772117962,
      "grad_norm": 21.0,
      "learning_rate": 7.6272774579447065e-06,
      "loss": 1.0551,
      "num_input_tokens_seen": 60000472,
      "step": 103410
    },
    {
      "epoch": 15.402889484658921,
      "grad_norm": 11.0,
      "learning_rate": 7.6249409529412145e-06,
      "loss": 1.0151,
      "num_input_tokens_seen": 60003448,
      "step": 103415
    },
    {
      "epoch": 15.403634197199882,
      "grad_norm": 6.8125,
      "learning_rate": 7.622604741471692e-06,
      "loss": 1.231,
      "num_input_tokens_seen": 60006552,
      "step": 103420
    },
    {
      "epoch": 15.40437890974084,
      "grad_norm": 13.3125,
      "learning_rate": 7.620268823575599e-06,
      "loss": 1.3345,
      "num_input_tokens_seen": 60009176,
      "step": 103425
    },
    {
      "epoch": 15.405123622281799,
      "grad_norm": 8.3125,
      "learning_rate": 7.617933199292396e-06,
      "loss": 0.8675,
      "num_input_tokens_seen": 60011704,
      "step": 103430
    },
    {
      "epoch": 15.405868334822758,
      "grad_norm": 11.4375,
      "learning_rate": 7.61559786866155e-06,
      "loss": 1.1595,
      "num_input_tokens_seen": 60014776,
      "step": 103435
    },
    {
      "epoch": 15.406613047363718,
      "grad_norm": 15.875,
      "learning_rate": 7.613262831722509e-06,
      "loss": 1.6102,
      "num_input_tokens_seen": 60017848,
      "step": 103440
    },
    {
      "epoch": 15.407357759904677,
      "grad_norm": 10.3125,
      "learning_rate": 7.610928088514724e-06,
      "loss": 1.7282,
      "num_input_tokens_seen": 60020792,
      "step": 103445
    },
    {
      "epoch": 15.408102472445636,
      "grad_norm": 5.125,
      "learning_rate": 7.608593639077627e-06,
      "loss": 1.556,
      "num_input_tokens_seen": 60023544,
      "step": 103450
    },
    {
      "epoch": 15.408847184986595,
      "grad_norm": 13.75,
      "learning_rate": 7.60625948345067e-06,
      "loss": 1.1159,
      "num_input_tokens_seen": 60026424,
      "step": 103455
    },
    {
      "epoch": 15.409591897527555,
      "grad_norm": 13.0,
      "learning_rate": 7.603925621673275e-06,
      "loss": 1.2011,
      "num_input_tokens_seen": 60029720,
      "step": 103460
    },
    {
      "epoch": 15.410336610068514,
      "grad_norm": 7.21875,
      "learning_rate": 7.601592053784884e-06,
      "loss": 1.0942,
      "num_input_tokens_seen": 60032728,
      "step": 103465
    },
    {
      "epoch": 15.411081322609473,
      "grad_norm": 8.8125,
      "learning_rate": 7.599258779824911e-06,
      "loss": 1.0702,
      "num_input_tokens_seen": 60035448,
      "step": 103470
    },
    {
      "epoch": 15.411826035150431,
      "grad_norm": 7.0,
      "learning_rate": 7.596925799832769e-06,
      "loss": 1.1665,
      "num_input_tokens_seen": 60038104,
      "step": 103475
    },
    {
      "epoch": 15.41257074769139,
      "grad_norm": 13.4375,
      "learning_rate": 7.594593113847887e-06,
      "loss": 0.9662,
      "num_input_tokens_seen": 60040760,
      "step": 103480
    },
    {
      "epoch": 15.41331546023235,
      "grad_norm": 12.5625,
      "learning_rate": 7.592260721909655e-06,
      "loss": 1.076,
      "num_input_tokens_seen": 60043640,
      "step": 103485
    },
    {
      "epoch": 15.41406017277331,
      "grad_norm": 19.75,
      "learning_rate": 7.589928624057494e-06,
      "loss": 1.4425,
      "num_input_tokens_seen": 60046392,
      "step": 103490
    },
    {
      "epoch": 15.414804885314268,
      "grad_norm": 8.6875,
      "learning_rate": 7.587596820330783e-06,
      "loss": 1.1616,
      "num_input_tokens_seen": 60049336,
      "step": 103495
    },
    {
      "epoch": 15.415549597855227,
      "grad_norm": 9.5625,
      "learning_rate": 7.585265310768938e-06,
      "loss": 0.8125,
      "num_input_tokens_seen": 60052568,
      "step": 103500
    },
    {
      "epoch": 15.416294310396188,
      "grad_norm": 10.1875,
      "learning_rate": 7.582934095411337e-06,
      "loss": 1.0541,
      "num_input_tokens_seen": 60055416,
      "step": 103505
    },
    {
      "epoch": 15.417039022937146,
      "grad_norm": 6.78125,
      "learning_rate": 7.58060317429736e-06,
      "loss": 1.2153,
      "num_input_tokens_seen": 60058232,
      "step": 103510
    },
    {
      "epoch": 15.417783735478105,
      "grad_norm": 9.5625,
      "learning_rate": 7.5782725474663894e-06,
      "loss": 1.0102,
      "num_input_tokens_seen": 60061080,
      "step": 103515
    },
    {
      "epoch": 15.418528448019064,
      "grad_norm": 7.6875,
      "learning_rate": 7.575942214957787e-06,
      "loss": 0.955,
      "num_input_tokens_seen": 60064024,
      "step": 103520
    },
    {
      "epoch": 15.419273160560024,
      "grad_norm": 7.625,
      "learning_rate": 7.573612176810943e-06,
      "loss": 1.116,
      "num_input_tokens_seen": 60066840,
      "step": 103525
    },
    {
      "epoch": 15.420017873100983,
      "grad_norm": 6.96875,
      "learning_rate": 7.5712824330651995e-06,
      "loss": 1.4552,
      "num_input_tokens_seen": 60069592,
      "step": 103530
    },
    {
      "epoch": 15.420762585641942,
      "grad_norm": 10.25,
      "learning_rate": 7.568952983759936e-06,
      "loss": 1.0879,
      "num_input_tokens_seen": 60072344,
      "step": 103535
    },
    {
      "epoch": 15.4215072981829,
      "grad_norm": 9.8125,
      "learning_rate": 7.566623828934485e-06,
      "loss": 1.1089,
      "num_input_tokens_seen": 60075256,
      "step": 103540
    },
    {
      "epoch": 15.422252010723861,
      "grad_norm": 15.625,
      "learning_rate": 7.5642949686282165e-06,
      "loss": 1.1952,
      "num_input_tokens_seen": 60078072,
      "step": 103545
    },
    {
      "epoch": 15.42299672326482,
      "grad_norm": 12.125,
      "learning_rate": 7.561966402880461e-06,
      "loss": 1.0044,
      "num_input_tokens_seen": 60080856,
      "step": 103550
    },
    {
      "epoch": 15.423741435805779,
      "grad_norm": 17.625,
      "learning_rate": 7.559638131730554e-06,
      "loss": 1.031,
      "num_input_tokens_seen": 60083672,
      "step": 103555
    },
    {
      "epoch": 15.424486148346737,
      "grad_norm": 7.125,
      "learning_rate": 7.557310155217842e-06,
      "loss": 1.2507,
      "num_input_tokens_seen": 60086776,
      "step": 103560
    },
    {
      "epoch": 15.425230860887698,
      "grad_norm": 4.4375,
      "learning_rate": 7.554982473381639e-06,
      "loss": 0.8482,
      "num_input_tokens_seen": 60090104,
      "step": 103565
    },
    {
      "epoch": 15.425975573428657,
      "grad_norm": 15.1875,
      "learning_rate": 7.552655086261287e-06,
      "loss": 0.9995,
      "num_input_tokens_seen": 60092600,
      "step": 103570
    },
    {
      "epoch": 15.426720285969616,
      "grad_norm": 6.46875,
      "learning_rate": 7.550327993896092e-06,
      "loss": 1.2876,
      "num_input_tokens_seen": 60095192,
      "step": 103575
    },
    {
      "epoch": 15.427464998510574,
      "grad_norm": 7.09375,
      "learning_rate": 7.548001196325372e-06,
      "loss": 1.5483,
      "num_input_tokens_seen": 60098136,
      "step": 103580
    },
    {
      "epoch": 15.428209711051535,
      "grad_norm": 12.5,
      "learning_rate": 7.545674693588434e-06,
      "loss": 0.9176,
      "num_input_tokens_seen": 60100696,
      "step": 103585
    },
    {
      "epoch": 15.428954423592494,
      "grad_norm": 14.4375,
      "learning_rate": 7.543348485724572e-06,
      "loss": 1.1561,
      "num_input_tokens_seen": 60103288,
      "step": 103590
    },
    {
      "epoch": 15.429699136133452,
      "grad_norm": 12.125,
      "learning_rate": 7.541022572773107e-06,
      "loss": 0.9192,
      "num_input_tokens_seen": 60106424,
      "step": 103595
    },
    {
      "epoch": 15.430443848674411,
      "grad_norm": 14.0625,
      "learning_rate": 7.538696954773311e-06,
      "loss": 0.8677,
      "num_input_tokens_seen": 60109304,
      "step": 103600
    },
    {
      "epoch": 15.431188561215372,
      "grad_norm": 6.59375,
      "learning_rate": 7.536371631764491e-06,
      "loss": 0.8659,
      "num_input_tokens_seen": 60112440,
      "step": 103605
    },
    {
      "epoch": 15.43193327375633,
      "grad_norm": 13.5,
      "learning_rate": 7.534046603785916e-06,
      "loss": 1.2674,
      "num_input_tokens_seen": 60115160,
      "step": 103610
    },
    {
      "epoch": 15.43267798629729,
      "grad_norm": 7.34375,
      "learning_rate": 7.531721870876879e-06,
      "loss": 1.1264,
      "num_input_tokens_seen": 60118264,
      "step": 103615
    },
    {
      "epoch": 15.433422698838248,
      "grad_norm": 14.0,
      "learning_rate": 7.529397433076638e-06,
      "loss": 1.2321,
      "num_input_tokens_seen": 60120888,
      "step": 103620
    },
    {
      "epoch": 15.434167411379208,
      "grad_norm": 11.0625,
      "learning_rate": 7.52707329042448e-06,
      "loss": 1.0288,
      "num_input_tokens_seen": 60123608,
      "step": 103625
    },
    {
      "epoch": 15.434912123920167,
      "grad_norm": 13.9375,
      "learning_rate": 7.524749442959661e-06,
      "loss": 1.4663,
      "num_input_tokens_seen": 60126744,
      "step": 103630
    },
    {
      "epoch": 15.435656836461126,
      "grad_norm": 6.71875,
      "learning_rate": 7.522425890721432e-06,
      "loss": 1.2522,
      "num_input_tokens_seen": 60129592,
      "step": 103635
    },
    {
      "epoch": 15.436401549002085,
      "grad_norm": 12.75,
      "learning_rate": 7.52010263374906e-06,
      "loss": 1.3885,
      "num_input_tokens_seen": 60132472,
      "step": 103640
    },
    {
      "epoch": 15.437146261543045,
      "grad_norm": 5.90625,
      "learning_rate": 7.51777967208179e-06,
      "loss": 1.1846,
      "num_input_tokens_seen": 60135224,
      "step": 103645
    },
    {
      "epoch": 15.437890974084004,
      "grad_norm": 10.8125,
      "learning_rate": 7.515457005758864e-06,
      "loss": 0.9971,
      "num_input_tokens_seen": 60137848,
      "step": 103650
    },
    {
      "epoch": 15.438635686624963,
      "grad_norm": 9.75,
      "learning_rate": 7.5131346348195105e-06,
      "loss": 0.9099,
      "num_input_tokens_seen": 60140600,
      "step": 103655
    },
    {
      "epoch": 15.439380399165922,
      "grad_norm": 12.125,
      "learning_rate": 7.510812559302985e-06,
      "loss": 1.5316,
      "num_input_tokens_seen": 60143640,
      "step": 103660
    },
    {
      "epoch": 15.44012511170688,
      "grad_norm": 8.5625,
      "learning_rate": 7.508490779248506e-06,
      "loss": 0.9204,
      "num_input_tokens_seen": 60146392,
      "step": 103665
    },
    {
      "epoch": 15.44086982424784,
      "grad_norm": 11.1875,
      "learning_rate": 7.5061692946952896e-06,
      "loss": 1.2591,
      "num_input_tokens_seen": 60148984,
      "step": 103670
    },
    {
      "epoch": 15.4416145367888,
      "grad_norm": 6.09375,
      "learning_rate": 7.503848105682571e-06,
      "loss": 1.0416,
      "num_input_tokens_seen": 60152056,
      "step": 103675
    },
    {
      "epoch": 15.442359249329758,
      "grad_norm": 21.875,
      "learning_rate": 7.501527212249549e-06,
      "loss": 1.4009,
      "num_input_tokens_seen": 60155384,
      "step": 103680
    },
    {
      "epoch": 15.443103961870717,
      "grad_norm": 8.3125,
      "learning_rate": 7.4992066144354475e-06,
      "loss": 1.0617,
      "num_input_tokens_seen": 60158104,
      "step": 103685
    },
    {
      "epoch": 15.443848674411678,
      "grad_norm": 10.5,
      "learning_rate": 7.496886312279455e-06,
      "loss": 1.273,
      "num_input_tokens_seen": 60161048,
      "step": 103690
    },
    {
      "epoch": 15.444593386952636,
      "grad_norm": 14.0625,
      "learning_rate": 7.494566305820788e-06,
      "loss": 1.4042,
      "num_input_tokens_seen": 60164088,
      "step": 103695
    },
    {
      "epoch": 15.445338099493595,
      "grad_norm": 19.0,
      "learning_rate": 7.492246595098629e-06,
      "loss": 1.1822,
      "num_input_tokens_seen": 60166872,
      "step": 103700
    },
    {
      "epoch": 15.446082812034554,
      "grad_norm": 13.25,
      "learning_rate": 7.489927180152173e-06,
      "loss": 1.0205,
      "num_input_tokens_seen": 60169624,
      "step": 103705
    },
    {
      "epoch": 15.446827524575514,
      "grad_norm": 6.5,
      "learning_rate": 7.487608061020599e-06,
      "loss": 1.2041,
      "num_input_tokens_seen": 60172344,
      "step": 103710
    },
    {
      "epoch": 15.447572237116473,
      "grad_norm": 11.9375,
      "learning_rate": 7.485289237743079e-06,
      "loss": 0.9513,
      "num_input_tokens_seen": 60175160,
      "step": 103715
    },
    {
      "epoch": 15.448316949657432,
      "grad_norm": 10.9375,
      "learning_rate": 7.482970710358806e-06,
      "loss": 1.0715,
      "num_input_tokens_seen": 60177944,
      "step": 103720
    },
    {
      "epoch": 15.44906166219839,
      "grad_norm": 14.3125,
      "learning_rate": 7.48065247890693e-06,
      "loss": 1.4164,
      "num_input_tokens_seen": 60180664,
      "step": 103725
    },
    {
      "epoch": 15.449806374739351,
      "grad_norm": 9.6875,
      "learning_rate": 7.478334543426632e-06,
      "loss": 0.9381,
      "num_input_tokens_seen": 60183576,
      "step": 103730
    },
    {
      "epoch": 15.45055108728031,
      "grad_norm": 6.78125,
      "learning_rate": 7.476016903957058e-06,
      "loss": 0.8647,
      "num_input_tokens_seen": 60186520,
      "step": 103735
    },
    {
      "epoch": 15.451295799821269,
      "grad_norm": 8.5,
      "learning_rate": 7.473699560537376e-06,
      "loss": 1.1204,
      "num_input_tokens_seen": 60189464,
      "step": 103740
    },
    {
      "epoch": 15.452040512362228,
      "grad_norm": 6.25,
      "learning_rate": 7.471382513206718e-06,
      "loss": 0.9563,
      "num_input_tokens_seen": 60192088,
      "step": 103745
    },
    {
      "epoch": 15.452785224903188,
      "grad_norm": 7.34375,
      "learning_rate": 7.469065762004243e-06,
      "loss": 1.0356,
      "num_input_tokens_seen": 60195032,
      "step": 103750
    },
    {
      "epoch": 15.453529937444147,
      "grad_norm": 13.6875,
      "learning_rate": 7.466749306969087e-06,
      "loss": 1.163,
      "num_input_tokens_seen": 60198040,
      "step": 103755
    },
    {
      "epoch": 15.454274649985106,
      "grad_norm": 5.09375,
      "learning_rate": 7.464433148140371e-06,
      "loss": 1.0321,
      "num_input_tokens_seen": 60200920,
      "step": 103760
    },
    {
      "epoch": 15.455019362526064,
      "grad_norm": 8.25,
      "learning_rate": 7.462117285557246e-06,
      "loss": 1.0789,
      "num_input_tokens_seen": 60203896,
      "step": 103765
    },
    {
      "epoch": 15.455764075067025,
      "grad_norm": 11.5625,
      "learning_rate": 7.459801719258821e-06,
      "loss": 1.1854,
      "num_input_tokens_seen": 60206488,
      "step": 103770
    },
    {
      "epoch": 15.456508787607984,
      "grad_norm": 6.5,
      "learning_rate": 7.457486449284221e-06,
      "loss": 1.0117,
      "num_input_tokens_seen": 60209400,
      "step": 103775
    },
    {
      "epoch": 15.457253500148942,
      "grad_norm": 8.3125,
      "learning_rate": 7.455171475672551e-06,
      "loss": 0.9456,
      "num_input_tokens_seen": 60212696,
      "step": 103780
    },
    {
      "epoch": 15.457998212689901,
      "grad_norm": 10.5625,
      "learning_rate": 7.4528567984629344e-06,
      "loss": 1.02,
      "num_input_tokens_seen": 60215640,
      "step": 103785
    },
    {
      "epoch": 15.458742925230862,
      "grad_norm": 5.875,
      "learning_rate": 7.450542417694467e-06,
      "loss": 1.2188,
      "num_input_tokens_seen": 60218200,
      "step": 103790
    },
    {
      "epoch": 15.45948763777182,
      "grad_norm": 18.0,
      "learning_rate": 7.448228333406241e-06,
      "loss": 1.0857,
      "num_input_tokens_seen": 60221336,
      "step": 103795
    },
    {
      "epoch": 15.46023235031278,
      "grad_norm": 15.3125,
      "learning_rate": 7.445914545637367e-06,
      "loss": 1.4396,
      "num_input_tokens_seen": 60225368,
      "step": 103800
    },
    {
      "epoch": 15.460977062853738,
      "grad_norm": 9.625,
      "learning_rate": 7.443601054426919e-06,
      "loss": 1.3639,
      "num_input_tokens_seen": 60228376,
      "step": 103805
    },
    {
      "epoch": 15.461721775394698,
      "grad_norm": 7.65625,
      "learning_rate": 7.441287859813995e-06,
      "loss": 0.9613,
      "num_input_tokens_seen": 60231000,
      "step": 103810
    },
    {
      "epoch": 15.462466487935657,
      "grad_norm": 18.0,
      "learning_rate": 7.438974961837655e-06,
      "loss": 1.2073,
      "num_input_tokens_seen": 60233880,
      "step": 103815
    },
    {
      "epoch": 15.463211200476616,
      "grad_norm": 6.3125,
      "learning_rate": 7.436662360536997e-06,
      "loss": 0.9912,
      "num_input_tokens_seen": 60236824,
      "step": 103820
    },
    {
      "epoch": 15.463955913017575,
      "grad_norm": 4.03125,
      "learning_rate": 7.43435005595107e-06,
      "loss": 1.0182,
      "num_input_tokens_seen": 60239864,
      "step": 103825
    },
    {
      "epoch": 15.464700625558535,
      "grad_norm": 7.1875,
      "learning_rate": 7.432038048118953e-06,
      "loss": 0.7199,
      "num_input_tokens_seen": 60242584,
      "step": 103830
    },
    {
      "epoch": 15.465445338099494,
      "grad_norm": 8.9375,
      "learning_rate": 7.429726337079695e-06,
      "loss": 0.9518,
      "num_input_tokens_seen": 60245400,
      "step": 103835
    },
    {
      "epoch": 15.466190050640453,
      "grad_norm": 11.875,
      "learning_rate": 7.427414922872356e-06,
      "loss": 1.2354,
      "num_input_tokens_seen": 60248440,
      "step": 103840
    },
    {
      "epoch": 15.466934763181412,
      "grad_norm": 8.5,
      "learning_rate": 7.4251038055359825e-06,
      "loss": 1.5905,
      "num_input_tokens_seen": 60251160,
      "step": 103845
    },
    {
      "epoch": 15.46767947572237,
      "grad_norm": 5.25,
      "learning_rate": 7.422792985109608e-06,
      "loss": 1.0279,
      "num_input_tokens_seen": 60254424,
      "step": 103850
    },
    {
      "epoch": 15.46842418826333,
      "grad_norm": 8.9375,
      "learning_rate": 7.420482461632289e-06,
      "loss": 1.087,
      "num_input_tokens_seen": 60257496,
      "step": 103855
    },
    {
      "epoch": 15.46916890080429,
      "grad_norm": 13.875,
      "learning_rate": 7.418172235143045e-06,
      "loss": 1.6872,
      "num_input_tokens_seen": 60260344,
      "step": 103860
    },
    {
      "epoch": 15.469913613345248,
      "grad_norm": 7.8125,
      "learning_rate": 7.415862305680921e-06,
      "loss": 0.8851,
      "num_input_tokens_seen": 60263064,
      "step": 103865
    },
    {
      "epoch": 15.470658325886207,
      "grad_norm": 11.5625,
      "learning_rate": 7.413552673284929e-06,
      "loss": 1.3596,
      "num_input_tokens_seen": 60265848,
      "step": 103870
    },
    {
      "epoch": 15.471403038427168,
      "grad_norm": 12.25,
      "learning_rate": 7.411243337994084e-06,
      "loss": 1.3207,
      "num_input_tokens_seen": 60268504,
      "step": 103875
    },
    {
      "epoch": 15.472147750968126,
      "grad_norm": 12.5,
      "learning_rate": 7.4089342998474145e-06,
      "loss": 1.4372,
      "num_input_tokens_seen": 60271576,
      "step": 103880
    },
    {
      "epoch": 15.472892463509085,
      "grad_norm": 14.0625,
      "learning_rate": 7.406625558883912e-06,
      "loss": 1.1875,
      "num_input_tokens_seen": 60274648,
      "step": 103885
    },
    {
      "epoch": 15.473637176050044,
      "grad_norm": 12.6875,
      "learning_rate": 7.404317115142598e-06,
      "loss": 0.758,
      "num_input_tokens_seen": 60277688,
      "step": 103890
    },
    {
      "epoch": 15.474381888591004,
      "grad_norm": 6.5625,
      "learning_rate": 7.402008968662455e-06,
      "loss": 1.1784,
      "num_input_tokens_seen": 60280632,
      "step": 103895
    },
    {
      "epoch": 15.475126601131963,
      "grad_norm": 9.1875,
      "learning_rate": 7.399701119482494e-06,
      "loss": 1.0737,
      "num_input_tokens_seen": 60283704,
      "step": 103900
    },
    {
      "epoch": 15.475871313672922,
      "grad_norm": 12.4375,
      "learning_rate": 7.397393567641694e-06,
      "loss": 0.9628,
      "num_input_tokens_seen": 60286680,
      "step": 103905
    },
    {
      "epoch": 15.47661602621388,
      "grad_norm": 11.125,
      "learning_rate": 7.395086313179037e-06,
      "loss": 0.9665,
      "num_input_tokens_seen": 60289592,
      "step": 103910
    },
    {
      "epoch": 15.477360738754841,
      "grad_norm": 10.4375,
      "learning_rate": 7.392779356133506e-06,
      "loss": 0.9424,
      "num_input_tokens_seen": 60292280,
      "step": 103915
    },
    {
      "epoch": 15.4781054512958,
      "grad_norm": 13.4375,
      "learning_rate": 7.390472696544065e-06,
      "loss": 1.0434,
      "num_input_tokens_seen": 60295000,
      "step": 103920
    },
    {
      "epoch": 15.478850163836759,
      "grad_norm": 21.75,
      "learning_rate": 7.388166334449697e-06,
      "loss": 1.2329,
      "num_input_tokens_seen": 60297976,
      "step": 103925
    },
    {
      "epoch": 15.479594876377718,
      "grad_norm": 19.125,
      "learning_rate": 7.3858602698893495e-06,
      "loss": 1.6293,
      "num_input_tokens_seen": 60300760,
      "step": 103930
    },
    {
      "epoch": 15.480339588918678,
      "grad_norm": 12.75,
      "learning_rate": 7.383554502902001e-06,
      "loss": 0.8493,
      "num_input_tokens_seen": 60303960,
      "step": 103935
    },
    {
      "epoch": 15.481084301459637,
      "grad_norm": 11.125,
      "learning_rate": 7.381249033526585e-06,
      "loss": 0.9433,
      "num_input_tokens_seen": 60306456,
      "step": 103940
    },
    {
      "epoch": 15.481829014000596,
      "grad_norm": 10.8125,
      "learning_rate": 7.37894386180207e-06,
      "loss": 0.9072,
      "num_input_tokens_seen": 60309144,
      "step": 103945
    },
    {
      "epoch": 15.482573726541554,
      "grad_norm": 8.875,
      "learning_rate": 7.376638987767387e-06,
      "loss": 1.3082,
      "num_input_tokens_seen": 60312216,
      "step": 103950
    },
    {
      "epoch": 15.483318439082515,
      "grad_norm": 11.625,
      "learning_rate": 7.37433441146147e-06,
      "loss": 1.4377,
      "num_input_tokens_seen": 60315096,
      "step": 103955
    },
    {
      "epoch": 15.484063151623474,
      "grad_norm": 10.5625,
      "learning_rate": 7.372030132923266e-06,
      "loss": 1.0283,
      "num_input_tokens_seen": 60317912,
      "step": 103960
    },
    {
      "epoch": 15.484807864164432,
      "grad_norm": 12.875,
      "learning_rate": 7.369726152191692e-06,
      "loss": 1.0613,
      "num_input_tokens_seen": 60320920,
      "step": 103965
    },
    {
      "epoch": 15.485552576705391,
      "grad_norm": 5.875,
      "learning_rate": 7.367422469305679e-06,
      "loss": 0.7772,
      "num_input_tokens_seen": 60323704,
      "step": 103970
    },
    {
      "epoch": 15.486297289246352,
      "grad_norm": 8.9375,
      "learning_rate": 7.365119084304145e-06,
      "loss": 0.9705,
      "num_input_tokens_seen": 60326328,
      "step": 103975
    },
    {
      "epoch": 15.48704200178731,
      "grad_norm": 7.5,
      "learning_rate": 7.362815997226e-06,
      "loss": 1.2301,
      "num_input_tokens_seen": 60329336,
      "step": 103980
    },
    {
      "epoch": 15.48778671432827,
      "grad_norm": 7.65625,
      "learning_rate": 7.360513208110148e-06,
      "loss": 1.0172,
      "num_input_tokens_seen": 60332472,
      "step": 103985
    },
    {
      "epoch": 15.488531426869228,
      "grad_norm": 9.375,
      "learning_rate": 7.3582107169955005e-06,
      "loss": 0.8786,
      "num_input_tokens_seen": 60335256,
      "step": 103990
    },
    {
      "epoch": 15.489276139410187,
      "grad_norm": 11.125,
      "learning_rate": 7.355908523920957e-06,
      "loss": 0.6679,
      "num_input_tokens_seen": 60338328,
      "step": 103995
    },
    {
      "epoch": 15.490020851951147,
      "grad_norm": 16.75,
      "learning_rate": 7.353606628925397e-06,
      "loss": 0.9129,
      "num_input_tokens_seen": 60341432,
      "step": 104000
    },
    {
      "epoch": 15.490765564492106,
      "grad_norm": 16.75,
      "learning_rate": 7.351305032047726e-06,
      "loss": 1.1908,
      "num_input_tokens_seen": 60344376,
      "step": 104005
    },
    {
      "epoch": 15.491510277033065,
      "grad_norm": 11.0,
      "learning_rate": 7.349003733326809e-06,
      "loss": 1.2528,
      "num_input_tokens_seen": 60347224,
      "step": 104010
    },
    {
      "epoch": 15.492254989574024,
      "grad_norm": 7.9375,
      "learning_rate": 7.346702732801544e-06,
      "loss": 1.2267,
      "num_input_tokens_seen": 60350040,
      "step": 104015
    },
    {
      "epoch": 15.492999702114984,
      "grad_norm": 8.0625,
      "learning_rate": 7.344402030510786e-06,
      "loss": 1.0796,
      "num_input_tokens_seen": 60352920,
      "step": 104020
    },
    {
      "epoch": 15.493744414655943,
      "grad_norm": 7.03125,
      "learning_rate": 7.34210162649342e-06,
      "loss": 1.142,
      "num_input_tokens_seen": 60355864,
      "step": 104025
    },
    {
      "epoch": 15.494489127196902,
      "grad_norm": 8.75,
      "learning_rate": 7.3398015207883006e-06,
      "loss": 1.363,
      "num_input_tokens_seen": 60358840,
      "step": 104030
    },
    {
      "epoch": 15.49523383973786,
      "grad_norm": 14.0,
      "learning_rate": 7.337501713434283e-06,
      "loss": 1.2449,
      "num_input_tokens_seen": 60361656,
      "step": 104035
    },
    {
      "epoch": 15.495978552278821,
      "grad_norm": 8.375,
      "learning_rate": 7.3352022044702266e-06,
      "loss": 0.9178,
      "num_input_tokens_seen": 60364440,
      "step": 104040
    },
    {
      "epoch": 15.49672326481978,
      "grad_norm": 11.375,
      "learning_rate": 7.332902993934965e-06,
      "loss": 0.98,
      "num_input_tokens_seen": 60367224,
      "step": 104045
    },
    {
      "epoch": 15.497467977360738,
      "grad_norm": 9.8125,
      "learning_rate": 7.33060408186736e-06,
      "loss": 1.3037,
      "num_input_tokens_seen": 60370264,
      "step": 104050
    },
    {
      "epoch": 15.498212689901697,
      "grad_norm": 7.0,
      "learning_rate": 7.328305468306229e-06,
      "loss": 0.8911,
      "num_input_tokens_seen": 60373144,
      "step": 104055
    },
    {
      "epoch": 15.498957402442658,
      "grad_norm": 9.6875,
      "learning_rate": 7.326007153290429e-06,
      "loss": 0.9375,
      "num_input_tokens_seen": 60375992,
      "step": 104060
    },
    {
      "epoch": 15.499702114983616,
      "grad_norm": 14.6875,
      "learning_rate": 7.323709136858764e-06,
      "loss": 1.4174,
      "num_input_tokens_seen": 60378968,
      "step": 104065
    },
    {
      "epoch": 15.500446827524575,
      "grad_norm": 10.5625,
      "learning_rate": 7.321411419050078e-06,
      "loss": 1.0336,
      "num_input_tokens_seen": 60381784,
      "step": 104070
    },
    {
      "epoch": 15.501191540065534,
      "grad_norm": 9.75,
      "learning_rate": 7.319113999903176e-06,
      "loss": 1.1207,
      "num_input_tokens_seen": 60384600,
      "step": 104075
    },
    {
      "epoch": 15.501936252606495,
      "grad_norm": 6.25,
      "learning_rate": 7.31681687945687e-06,
      "loss": 1.0109,
      "num_input_tokens_seen": 60387672,
      "step": 104080
    },
    {
      "epoch": 15.502680965147453,
      "grad_norm": 10.75,
      "learning_rate": 7.314520057749974e-06,
      "loss": 1.1327,
      "num_input_tokens_seen": 60390456,
      "step": 104085
    },
    {
      "epoch": 15.503425677688412,
      "grad_norm": 8.3125,
      "learning_rate": 7.312223534821281e-06,
      "loss": 0.674,
      "num_input_tokens_seen": 60393272,
      "step": 104090
    },
    {
      "epoch": 15.50417039022937,
      "grad_norm": 19.125,
      "learning_rate": 7.3099273107096e-06,
      "loss": 0.9774,
      "num_input_tokens_seen": 60396088,
      "step": 104095
    },
    {
      "epoch": 15.504915102770331,
      "grad_norm": 9.75,
      "learning_rate": 7.307631385453717e-06,
      "loss": 1.2029,
      "num_input_tokens_seen": 60399032,
      "step": 104100
    },
    {
      "epoch": 15.50565981531129,
      "grad_norm": 10.25,
      "learning_rate": 7.305335759092424e-06,
      "loss": 0.8769,
      "num_input_tokens_seen": 60401784,
      "step": 104105
    },
    {
      "epoch": 15.506404527852249,
      "grad_norm": 5.96875,
      "learning_rate": 7.303040431664496e-06,
      "loss": 0.8608,
      "num_input_tokens_seen": 60404856,
      "step": 104110
    },
    {
      "epoch": 15.507149240393208,
      "grad_norm": 8.375,
      "learning_rate": 7.300745403208705e-06,
      "loss": 1.6976,
      "num_input_tokens_seen": 60407800,
      "step": 104115
    },
    {
      "epoch": 15.507893952934168,
      "grad_norm": 9.625,
      "learning_rate": 7.298450673763843e-06,
      "loss": 1.2535,
      "num_input_tokens_seen": 60410616,
      "step": 104120
    },
    {
      "epoch": 15.508638665475127,
      "grad_norm": 8.9375,
      "learning_rate": 7.296156243368657e-06,
      "loss": 1.1663,
      "num_input_tokens_seen": 60413720,
      "step": 104125
    },
    {
      "epoch": 15.509383378016086,
      "grad_norm": 12.125,
      "learning_rate": 7.293862112061925e-06,
      "loss": 0.8196,
      "num_input_tokens_seen": 60416504,
      "step": 104130
    },
    {
      "epoch": 15.510128090557044,
      "grad_norm": 9.875,
      "learning_rate": 7.291568279882388e-06,
      "loss": 1.0744,
      "num_input_tokens_seen": 60419192,
      "step": 104135
    },
    {
      "epoch": 15.510872803098005,
      "grad_norm": 10.3125,
      "learning_rate": 7.289274746868818e-06,
      "loss": 0.9361,
      "num_input_tokens_seen": 60422104,
      "step": 104140
    },
    {
      "epoch": 15.511617515638964,
      "grad_norm": 5.03125,
      "learning_rate": 7.28698151305994e-06,
      "loss": 1.463,
      "num_input_tokens_seen": 60424888,
      "step": 104145
    },
    {
      "epoch": 15.512362228179922,
      "grad_norm": 8.375,
      "learning_rate": 7.284688578494514e-06,
      "loss": 1.0548,
      "num_input_tokens_seen": 60428792,
      "step": 104150
    },
    {
      "epoch": 15.513106940720881,
      "grad_norm": 7.6875,
      "learning_rate": 7.2823959432112705e-06,
      "loss": 0.8022,
      "num_input_tokens_seen": 60431736,
      "step": 104155
    },
    {
      "epoch": 15.513851653261842,
      "grad_norm": 9.0,
      "learning_rate": 7.280103607248934e-06,
      "loss": 1.2172,
      "num_input_tokens_seen": 60434616,
      "step": 104160
    },
    {
      "epoch": 15.5145963658028,
      "grad_norm": 4.96875,
      "learning_rate": 7.277811570646242e-06,
      "loss": 0.7818,
      "num_input_tokens_seen": 60437432,
      "step": 104165
    },
    {
      "epoch": 15.51534107834376,
      "grad_norm": 7.9375,
      "learning_rate": 7.275519833441915e-06,
      "loss": 1.2732,
      "num_input_tokens_seen": 60440120,
      "step": 104170
    },
    {
      "epoch": 15.516085790884718,
      "grad_norm": 15.125,
      "learning_rate": 7.273228395674664e-06,
      "loss": 1.3467,
      "num_input_tokens_seen": 60442776,
      "step": 104175
    },
    {
      "epoch": 15.516830503425677,
      "grad_norm": 14.0,
      "learning_rate": 7.270937257383195e-06,
      "loss": 1.2852,
      "num_input_tokens_seen": 60445784,
      "step": 104180
    },
    {
      "epoch": 15.517575215966637,
      "grad_norm": 14.75,
      "learning_rate": 7.268646418606229e-06,
      "loss": 1.3337,
      "num_input_tokens_seen": 60448504,
      "step": 104185
    },
    {
      "epoch": 15.518319928507596,
      "grad_norm": 11.4375,
      "learning_rate": 7.266355879382461e-06,
      "loss": 1.165,
      "num_input_tokens_seen": 60451192,
      "step": 104190
    },
    {
      "epoch": 15.519064641048555,
      "grad_norm": 5.84375,
      "learning_rate": 7.2640656397505805e-06,
      "loss": 1.2665,
      "num_input_tokens_seen": 60454360,
      "step": 104195
    },
    {
      "epoch": 15.519809353589514,
      "grad_norm": 12.3125,
      "learning_rate": 7.26177569974929e-06,
      "loss": 1.1212,
      "num_input_tokens_seen": 60457016,
      "step": 104200
    },
    {
      "epoch": 15.520554066130474,
      "grad_norm": 19.375,
      "learning_rate": 7.259486059417265e-06,
      "loss": 1.4998,
      "num_input_tokens_seen": 60459960,
      "step": 104205
    },
    {
      "epoch": 15.521298778671433,
      "grad_norm": 10.0625,
      "learning_rate": 7.2571967187932e-06,
      "loss": 1.2429,
      "num_input_tokens_seen": 60462712,
      "step": 104210
    },
    {
      "epoch": 15.522043491212392,
      "grad_norm": 10.1875,
      "learning_rate": 7.2549076779157565e-06,
      "loss": 1.0027,
      "num_input_tokens_seen": 60465368,
      "step": 104215
    },
    {
      "epoch": 15.52278820375335,
      "grad_norm": 17.375,
      "learning_rate": 7.252618936823618e-06,
      "loss": 0.9028,
      "num_input_tokens_seen": 60468184,
      "step": 104220
    },
    {
      "epoch": 15.523532916294311,
      "grad_norm": 9.9375,
      "learning_rate": 7.250330495555438e-06,
      "loss": 0.9521,
      "num_input_tokens_seen": 60470968,
      "step": 104225
    },
    {
      "epoch": 15.52427762883527,
      "grad_norm": 18.875,
      "learning_rate": 7.248042354149892e-06,
      "loss": 1.3063,
      "num_input_tokens_seen": 60473784,
      "step": 104230
    },
    {
      "epoch": 15.525022341376228,
      "grad_norm": 6.78125,
      "learning_rate": 7.2457545126456275e-06,
      "loss": 1.0745,
      "num_input_tokens_seen": 60476888,
      "step": 104235
    },
    {
      "epoch": 15.525767053917187,
      "grad_norm": 9.5,
      "learning_rate": 7.243466971081297e-06,
      "loss": 1.3328,
      "num_input_tokens_seen": 60479768,
      "step": 104240
    },
    {
      "epoch": 15.526511766458148,
      "grad_norm": 7.1875,
      "learning_rate": 7.2411797294955455e-06,
      "loss": 1.2935,
      "num_input_tokens_seen": 60482456,
      "step": 104245
    },
    {
      "epoch": 15.527256478999107,
      "grad_norm": 5.78125,
      "learning_rate": 7.238892787927004e-06,
      "loss": 1.1197,
      "num_input_tokens_seen": 60485464,
      "step": 104250
    },
    {
      "epoch": 15.528001191540065,
      "grad_norm": 9.125,
      "learning_rate": 7.2366061464143265e-06,
      "loss": 1.0109,
      "num_input_tokens_seen": 60488408,
      "step": 104255
    },
    {
      "epoch": 15.528745904081024,
      "grad_norm": 11.125,
      "learning_rate": 7.234319804996126e-06,
      "loss": 1.3412,
      "num_input_tokens_seen": 60491256,
      "step": 104260
    },
    {
      "epoch": 15.529490616621985,
      "grad_norm": 17.875,
      "learning_rate": 7.232033763711044e-06,
      "loss": 1.2858,
      "num_input_tokens_seen": 60494200,
      "step": 104265
    },
    {
      "epoch": 15.530235329162943,
      "grad_norm": 13.0,
      "learning_rate": 7.229748022597693e-06,
      "loss": 1.1752,
      "num_input_tokens_seen": 60497400,
      "step": 104270
    },
    {
      "epoch": 15.530980041703902,
      "grad_norm": 9.9375,
      "learning_rate": 7.22746258169468e-06,
      "loss": 0.7562,
      "num_input_tokens_seen": 60500216,
      "step": 104275
    },
    {
      "epoch": 15.53172475424486,
      "grad_norm": 13.9375,
      "learning_rate": 7.225177441040632e-06,
      "loss": 1.4564,
      "num_input_tokens_seen": 60503128,
      "step": 104280
    },
    {
      "epoch": 15.532469466785821,
      "grad_norm": 15.9375,
      "learning_rate": 7.2228926006741385e-06,
      "loss": 1.2831,
      "num_input_tokens_seen": 60506072,
      "step": 104285
    },
    {
      "epoch": 15.53321417932678,
      "grad_norm": 9.625,
      "learning_rate": 7.220608060633813e-06,
      "loss": 1.3367,
      "num_input_tokens_seen": 60508920,
      "step": 104290
    },
    {
      "epoch": 15.533958891867739,
      "grad_norm": 11.0,
      "learning_rate": 7.218323820958237e-06,
      "loss": 0.9835,
      "num_input_tokens_seen": 60511896,
      "step": 104295
    },
    {
      "epoch": 15.534703604408698,
      "grad_norm": 14.75,
      "learning_rate": 7.2160398816860155e-06,
      "loss": 0.9988,
      "num_input_tokens_seen": 60514968,
      "step": 104300
    },
    {
      "epoch": 15.535448316949658,
      "grad_norm": 17.25,
      "learning_rate": 7.213756242855724e-06,
      "loss": 1.2684,
      "num_input_tokens_seen": 60517912,
      "step": 104305
    },
    {
      "epoch": 15.536193029490617,
      "grad_norm": 7.25,
      "learning_rate": 7.211472904505945e-06,
      "loss": 0.7863,
      "num_input_tokens_seen": 60520664,
      "step": 104310
    },
    {
      "epoch": 15.536937742031576,
      "grad_norm": 10.5,
      "learning_rate": 7.20918986667525e-06,
      "loss": 1.1877,
      "num_input_tokens_seen": 60523448,
      "step": 104315
    },
    {
      "epoch": 15.537682454572534,
      "grad_norm": 9.375,
      "learning_rate": 7.206907129402205e-06,
      "loss": 0.9162,
      "num_input_tokens_seen": 60526360,
      "step": 104320
    },
    {
      "epoch": 15.538427167113493,
      "grad_norm": 5.34375,
      "learning_rate": 7.204624692725387e-06,
      "loss": 0.9638,
      "num_input_tokens_seen": 60529304,
      "step": 104325
    },
    {
      "epoch": 15.539171879654454,
      "grad_norm": 19.5,
      "learning_rate": 7.202342556683339e-06,
      "loss": 1.27,
      "num_input_tokens_seen": 60532120,
      "step": 104330
    },
    {
      "epoch": 15.539916592195413,
      "grad_norm": 12.0,
      "learning_rate": 7.200060721314636e-06,
      "loss": 1.2401,
      "num_input_tokens_seen": 60535000,
      "step": 104335
    },
    {
      "epoch": 15.540661304736371,
      "grad_norm": 6.1875,
      "learning_rate": 7.1977791866578045e-06,
      "loss": 0.7849,
      "num_input_tokens_seen": 60538136,
      "step": 104340
    },
    {
      "epoch": 15.541406017277332,
      "grad_norm": 8.875,
      "learning_rate": 7.195497952751409e-06,
      "loss": 1.0853,
      "num_input_tokens_seen": 60541144,
      "step": 104345
    },
    {
      "epoch": 15.54215072981829,
      "grad_norm": 9.875,
      "learning_rate": 7.1932170196339745e-06,
      "loss": 0.9437,
      "num_input_tokens_seen": 60544376,
      "step": 104350
    },
    {
      "epoch": 15.54289544235925,
      "grad_norm": 14.6875,
      "learning_rate": 7.190936387344047e-06,
      "loss": 1.0688,
      "num_input_tokens_seen": 60547128,
      "step": 104355
    },
    {
      "epoch": 15.543640154900208,
      "grad_norm": 8.75,
      "learning_rate": 7.188656055920149e-06,
      "loss": 1.2427,
      "num_input_tokens_seen": 60550072,
      "step": 104360
    },
    {
      "epoch": 15.544384867441167,
      "grad_norm": 10.25,
      "learning_rate": 7.186376025400804e-06,
      "loss": 0.8388,
      "num_input_tokens_seen": 60552760,
      "step": 104365
    },
    {
      "epoch": 15.545129579982127,
      "grad_norm": 20.25,
      "learning_rate": 7.18409629582453e-06,
      "loss": 1.4454,
      "num_input_tokens_seen": 60555544,
      "step": 104370
    },
    {
      "epoch": 15.545874292523086,
      "grad_norm": 10.75,
      "learning_rate": 7.181816867229835e-06,
      "loss": 1.21,
      "num_input_tokens_seen": 60558392,
      "step": 104375
    },
    {
      "epoch": 15.546619005064045,
      "grad_norm": 20.25,
      "learning_rate": 7.179537739655243e-06,
      "loss": 0.9075,
      "num_input_tokens_seen": 60561464,
      "step": 104380
    },
    {
      "epoch": 15.547363717605004,
      "grad_norm": 5.65625,
      "learning_rate": 7.17725891313924e-06,
      "loss": 1.135,
      "num_input_tokens_seen": 60564216,
      "step": 104385
    },
    {
      "epoch": 15.548108430145964,
      "grad_norm": 9.0,
      "learning_rate": 7.17498038772034e-06,
      "loss": 0.9444,
      "num_input_tokens_seen": 60566936,
      "step": 104390
    },
    {
      "epoch": 15.548853142686923,
      "grad_norm": 5.5625,
      "learning_rate": 7.172702163437034e-06,
      "loss": 0.8421,
      "num_input_tokens_seen": 60569976,
      "step": 104395
    },
    {
      "epoch": 15.549597855227882,
      "grad_norm": 8.4375,
      "learning_rate": 7.170424240327794e-06,
      "loss": 1.1765,
      "num_input_tokens_seen": 60572920,
      "step": 104400
    },
    {
      "epoch": 15.55034256776884,
      "grad_norm": 9.25,
      "learning_rate": 7.168146618431127e-06,
      "loss": 1.042,
      "num_input_tokens_seen": 60576248,
      "step": 104405
    },
    {
      "epoch": 15.551087280309801,
      "grad_norm": 10.5625,
      "learning_rate": 7.165869297785488e-06,
      "loss": 0.7231,
      "num_input_tokens_seen": 60579000,
      "step": 104410
    },
    {
      "epoch": 15.55183199285076,
      "grad_norm": 13.3125,
      "learning_rate": 7.163592278429371e-06,
      "loss": 0.925,
      "num_input_tokens_seen": 60582040,
      "step": 104415
    },
    {
      "epoch": 15.552576705391719,
      "grad_norm": 7.59375,
      "learning_rate": 7.161315560401224e-06,
      "loss": 1.1612,
      "num_input_tokens_seen": 60585176,
      "step": 104420
    },
    {
      "epoch": 15.553321417932677,
      "grad_norm": 7.25,
      "learning_rate": 7.159039143739532e-06,
      "loss": 0.8944,
      "num_input_tokens_seen": 60588088,
      "step": 104425
    },
    {
      "epoch": 15.554066130473638,
      "grad_norm": 15.125,
      "learning_rate": 7.1567630284827384e-06,
      "loss": 1.0242,
      "num_input_tokens_seen": 60590840,
      "step": 104430
    },
    {
      "epoch": 15.554810843014597,
      "grad_norm": 6.3125,
      "learning_rate": 7.1544872146693e-06,
      "loss": 1.0099,
      "num_input_tokens_seen": 60593688,
      "step": 104435
    },
    {
      "epoch": 15.555555555555555,
      "grad_norm": 10.5,
      "learning_rate": 7.1522117023376606e-06,
      "loss": 0.873,
      "num_input_tokens_seen": 60596440,
      "step": 104440
    },
    {
      "epoch": 15.556300268096514,
      "grad_norm": 14.3125,
      "learning_rate": 7.149936491526258e-06,
      "loss": 1.0939,
      "num_input_tokens_seen": 60599096,
      "step": 104445
    },
    {
      "epoch": 15.557044980637475,
      "grad_norm": 5.375,
      "learning_rate": 7.147661582273546e-06,
      "loss": 0.9078,
      "num_input_tokens_seen": 60601912,
      "step": 104450
    },
    {
      "epoch": 15.557789693178433,
      "grad_norm": 14.1875,
      "learning_rate": 7.145386974617937e-06,
      "loss": 1.2781,
      "num_input_tokens_seen": 60604984,
      "step": 104455
    },
    {
      "epoch": 15.558534405719392,
      "grad_norm": 10.625,
      "learning_rate": 7.143112668597876e-06,
      "loss": 1.2462,
      "num_input_tokens_seen": 60607704,
      "step": 104460
    },
    {
      "epoch": 15.559279118260351,
      "grad_norm": 8.8125,
      "learning_rate": 7.140838664251773e-06,
      "loss": 1.3556,
      "num_input_tokens_seen": 60610584,
      "step": 104465
    },
    {
      "epoch": 15.560023830801311,
      "grad_norm": 6.1875,
      "learning_rate": 7.138564961618055e-06,
      "loss": 1.044,
      "num_input_tokens_seen": 60613400,
      "step": 104470
    },
    {
      "epoch": 15.56076854334227,
      "grad_norm": 14.4375,
      "learning_rate": 7.1362915607351285e-06,
      "loss": 1.0272,
      "num_input_tokens_seen": 60616344,
      "step": 104475
    },
    {
      "epoch": 15.561513255883229,
      "grad_norm": 15.625,
      "learning_rate": 7.1340184616413926e-06,
      "loss": 1.289,
      "num_input_tokens_seen": 60619224,
      "step": 104480
    },
    {
      "epoch": 15.562257968424188,
      "grad_norm": 14.75,
      "learning_rate": 7.131745664375264e-06,
      "loss": 0.9364,
      "num_input_tokens_seen": 60622680,
      "step": 104485
    },
    {
      "epoch": 15.563002680965148,
      "grad_norm": 8.6875,
      "learning_rate": 7.129473168975123e-06,
      "loss": 0.9143,
      "num_input_tokens_seen": 60625560,
      "step": 104490
    },
    {
      "epoch": 15.563747393506107,
      "grad_norm": 4.84375,
      "learning_rate": 7.127200975479381e-06,
      "loss": 1.2836,
      "num_input_tokens_seen": 60628344,
      "step": 104495
    },
    {
      "epoch": 15.564492106047066,
      "grad_norm": 7.8125,
      "learning_rate": 7.12492908392641e-06,
      "loss": 0.8071,
      "num_input_tokens_seen": 60631224,
      "step": 104500
    },
    {
      "epoch": 15.565236818588025,
      "grad_norm": 13.375,
      "learning_rate": 7.122657494354596e-06,
      "loss": 1.2216,
      "num_input_tokens_seen": 60634168,
      "step": 104505
    },
    {
      "epoch": 15.565981531128983,
      "grad_norm": 13.0,
      "learning_rate": 7.120386206802307e-06,
      "loss": 1.2013,
      "num_input_tokens_seen": 60637048,
      "step": 104510
    },
    {
      "epoch": 15.566726243669944,
      "grad_norm": 16.25,
      "learning_rate": 7.1181152213079275e-06,
      "loss": 0.9438,
      "num_input_tokens_seen": 60639960,
      "step": 104515
    },
    {
      "epoch": 15.567470956210903,
      "grad_norm": 10.4375,
      "learning_rate": 7.115844537909819e-06,
      "loss": 0.9887,
      "num_input_tokens_seen": 60642680,
      "step": 104520
    },
    {
      "epoch": 15.568215668751861,
      "grad_norm": 19.0,
      "learning_rate": 7.11357415664633e-06,
      "loss": 0.8649,
      "num_input_tokens_seen": 60645880,
      "step": 104525
    },
    {
      "epoch": 15.568960381292822,
      "grad_norm": 10.5,
      "learning_rate": 7.111304077555836e-06,
      "loss": 1.1038,
      "num_input_tokens_seen": 60648760,
      "step": 104530
    },
    {
      "epoch": 15.56970509383378,
      "grad_norm": 12.9375,
      "learning_rate": 7.1090343006766704e-06,
      "loss": 1.1943,
      "num_input_tokens_seen": 60651736,
      "step": 104535
    },
    {
      "epoch": 15.57044980637474,
      "grad_norm": 17.0,
      "learning_rate": 7.106764826047196e-06,
      "loss": 1.0495,
      "num_input_tokens_seen": 60654712,
      "step": 104540
    },
    {
      "epoch": 15.571194518915698,
      "grad_norm": 8.5,
      "learning_rate": 7.104495653705734e-06,
      "loss": 1.0175,
      "num_input_tokens_seen": 60657976,
      "step": 104545
    },
    {
      "epoch": 15.571939231456657,
      "grad_norm": 7.5,
      "learning_rate": 7.102226783690638e-06,
      "loss": 0.7096,
      "num_input_tokens_seen": 60660664,
      "step": 104550
    },
    {
      "epoch": 15.572683943997617,
      "grad_norm": 14.6875,
      "learning_rate": 7.099958216040231e-06,
      "loss": 1.0887,
      "num_input_tokens_seen": 60663608,
      "step": 104555
    },
    {
      "epoch": 15.573428656538576,
      "grad_norm": 8.5625,
      "learning_rate": 7.09768995079283e-06,
      "loss": 1.0078,
      "num_input_tokens_seen": 60666552,
      "step": 104560
    },
    {
      "epoch": 15.574173369079535,
      "grad_norm": 12.4375,
      "learning_rate": 7.095421987986766e-06,
      "loss": 0.9013,
      "num_input_tokens_seen": 60669368,
      "step": 104565
    },
    {
      "epoch": 15.574918081620494,
      "grad_norm": 7.25,
      "learning_rate": 7.093154327660354e-06,
      "loss": 0.9139,
      "num_input_tokens_seen": 60672216,
      "step": 104570
    },
    {
      "epoch": 15.575662794161454,
      "grad_norm": 11.875,
      "learning_rate": 7.090886969851898e-06,
      "loss": 1.1086,
      "num_input_tokens_seen": 60675032,
      "step": 104575
    },
    {
      "epoch": 15.576407506702413,
      "grad_norm": 9.9375,
      "learning_rate": 7.088619914599698e-06,
      "loss": 1.0774,
      "num_input_tokens_seen": 60677944,
      "step": 104580
    },
    {
      "epoch": 15.577152219243372,
      "grad_norm": 12.25,
      "learning_rate": 7.086353161942066e-06,
      "loss": 1.2688,
      "num_input_tokens_seen": 60680984,
      "step": 104585
    },
    {
      "epoch": 15.57789693178433,
      "grad_norm": 9.6875,
      "learning_rate": 7.084086711917287e-06,
      "loss": 0.9005,
      "num_input_tokens_seen": 60683736,
      "step": 104590
    },
    {
      "epoch": 15.578641644325291,
      "grad_norm": 8.1875,
      "learning_rate": 7.081820564563657e-06,
      "loss": 0.9178,
      "num_input_tokens_seen": 60686840,
      "step": 104595
    },
    {
      "epoch": 15.57938635686625,
      "grad_norm": 5.3125,
      "learning_rate": 7.0795547199194624e-06,
      "loss": 1.1035,
      "num_input_tokens_seen": 60689656,
      "step": 104600
    },
    {
      "epoch": 15.580131069407209,
      "grad_norm": 6.34375,
      "learning_rate": 7.077289178022967e-06,
      "loss": 1.0383,
      "num_input_tokens_seen": 60692632,
      "step": 104605
    },
    {
      "epoch": 15.580875781948167,
      "grad_norm": 12.25,
      "learning_rate": 7.075023938912461e-06,
      "loss": 1.3846,
      "num_input_tokens_seen": 60695512,
      "step": 104610
    },
    {
      "epoch": 15.581620494489128,
      "grad_norm": 5.4375,
      "learning_rate": 7.0727590026262e-06,
      "loss": 0.8604,
      "num_input_tokens_seen": 60698200,
      "step": 104615
    },
    {
      "epoch": 15.582365207030087,
      "grad_norm": 7.25,
      "learning_rate": 7.070494369202465e-06,
      "loss": 0.9271,
      "num_input_tokens_seen": 60700856,
      "step": 104620
    },
    {
      "epoch": 15.583109919571045,
      "grad_norm": 10.875,
      "learning_rate": 7.068230038679496e-06,
      "loss": 0.9739,
      "num_input_tokens_seen": 60703864,
      "step": 104625
    },
    {
      "epoch": 15.583854632112004,
      "grad_norm": 14.375,
      "learning_rate": 7.065966011095565e-06,
      "loss": 1.4733,
      "num_input_tokens_seen": 60706776,
      "step": 104630
    },
    {
      "epoch": 15.584599344652965,
      "grad_norm": 6.71875,
      "learning_rate": 7.06370228648891e-06,
      "loss": 1.1076,
      "num_input_tokens_seen": 60709720,
      "step": 104635
    },
    {
      "epoch": 15.585344057193923,
      "grad_norm": 13.3125,
      "learning_rate": 7.061438864897774e-06,
      "loss": 1.1528,
      "num_input_tokens_seen": 60712792,
      "step": 104640
    },
    {
      "epoch": 15.586088769734882,
      "grad_norm": 4.6875,
      "learning_rate": 7.059175746360397e-06,
      "loss": 1.0978,
      "num_input_tokens_seen": 60715832,
      "step": 104645
    },
    {
      "epoch": 15.586833482275841,
      "grad_norm": 6.5625,
      "learning_rate": 7.056912930915005e-06,
      "loss": 0.823,
      "num_input_tokens_seen": 60718808,
      "step": 104650
    },
    {
      "epoch": 15.587578194816802,
      "grad_norm": 4.8125,
      "learning_rate": 7.054650418599837e-06,
      "loss": 1.0434,
      "num_input_tokens_seen": 60721912,
      "step": 104655
    },
    {
      "epoch": 15.58832290735776,
      "grad_norm": 12.0,
      "learning_rate": 7.052388209453106e-06,
      "loss": 1.0619,
      "num_input_tokens_seen": 60724696,
      "step": 104660
    },
    {
      "epoch": 15.589067619898719,
      "grad_norm": 30.25,
      "learning_rate": 7.0501263035130435e-06,
      "loss": 1.1531,
      "num_input_tokens_seen": 60728024,
      "step": 104665
    },
    {
      "epoch": 15.589812332439678,
      "grad_norm": 10.3125,
      "learning_rate": 7.0478647008178435e-06,
      "loss": 1.357,
      "num_input_tokens_seen": 60731160,
      "step": 104670
    },
    {
      "epoch": 15.590557044980638,
      "grad_norm": 13.0625,
      "learning_rate": 7.045603401405735e-06,
      "loss": 1.2095,
      "num_input_tokens_seen": 60734200,
      "step": 104675
    },
    {
      "epoch": 15.591301757521597,
      "grad_norm": 7.5625,
      "learning_rate": 7.043342405314907e-06,
      "loss": 0.953,
      "num_input_tokens_seen": 60736920,
      "step": 104680
    },
    {
      "epoch": 15.592046470062556,
      "grad_norm": 13.625,
      "learning_rate": 7.04108171258355e-06,
      "loss": 1.2701,
      "num_input_tokens_seen": 60739608,
      "step": 104685
    },
    {
      "epoch": 15.592791182603515,
      "grad_norm": 10.5625,
      "learning_rate": 7.038821323249875e-06,
      "loss": 1.0417,
      "num_input_tokens_seen": 60742936,
      "step": 104690
    },
    {
      "epoch": 15.593535895144473,
      "grad_norm": 4.8125,
      "learning_rate": 7.036561237352057e-06,
      "loss": 0.998,
      "num_input_tokens_seen": 60746168,
      "step": 104695
    },
    {
      "epoch": 15.594280607685434,
      "grad_norm": 9.4375,
      "learning_rate": 7.03430145492828e-06,
      "loss": 1.2008,
      "num_input_tokens_seen": 60749368,
      "step": 104700
    },
    {
      "epoch": 15.595025320226393,
      "grad_norm": 13.1875,
      "learning_rate": 7.032041976016712e-06,
      "loss": 1.252,
      "num_input_tokens_seen": 60752152,
      "step": 104705
    },
    {
      "epoch": 15.595770032767351,
      "grad_norm": 10.0625,
      "learning_rate": 7.02978280065554e-06,
      "loss": 0.9276,
      "num_input_tokens_seen": 60755128,
      "step": 104710
    },
    {
      "epoch": 15.59651474530831,
      "grad_norm": 13.25,
      "learning_rate": 7.027523928882926e-06,
      "loss": 1.278,
      "num_input_tokens_seen": 60757752,
      "step": 104715
    },
    {
      "epoch": 15.59725945784927,
      "grad_norm": 11.25,
      "learning_rate": 7.025265360737021e-06,
      "loss": 1.2186,
      "num_input_tokens_seen": 60761016,
      "step": 104720
    },
    {
      "epoch": 15.59800417039023,
      "grad_norm": 6.46875,
      "learning_rate": 7.023007096255996e-06,
      "loss": 1.0212,
      "num_input_tokens_seen": 60764376,
      "step": 104725
    },
    {
      "epoch": 15.598748882931188,
      "grad_norm": 14.0,
      "learning_rate": 7.020749135477986e-06,
      "loss": 0.6976,
      "num_input_tokens_seen": 60767064,
      "step": 104730
    },
    {
      "epoch": 15.599493595472147,
      "grad_norm": 10.3125,
      "learning_rate": 7.0184914784411555e-06,
      "loss": 1.3333,
      "num_input_tokens_seen": 60769752,
      "step": 104735
    },
    {
      "epoch": 15.600238308013108,
      "grad_norm": 9.5625,
      "learning_rate": 7.0162341251836264e-06,
      "loss": 1.3227,
      "num_input_tokens_seen": 60772760,
      "step": 104740
    },
    {
      "epoch": 15.600983020554066,
      "grad_norm": 7.9375,
      "learning_rate": 7.013977075743553e-06,
      "loss": 1.3819,
      "num_input_tokens_seen": 60775640,
      "step": 104745
    },
    {
      "epoch": 15.601727733095025,
      "grad_norm": 6.4375,
      "learning_rate": 7.01172033015905e-06,
      "loss": 1.1331,
      "num_input_tokens_seen": 60778360,
      "step": 104750
    },
    {
      "epoch": 15.602472445635984,
      "grad_norm": 14.125,
      "learning_rate": 7.009463888468254e-06,
      "loss": 0.7764,
      "num_input_tokens_seen": 60781080,
      "step": 104755
    },
    {
      "epoch": 15.603217158176944,
      "grad_norm": 16.25,
      "learning_rate": 7.0072077507092825e-06,
      "loss": 1.2345,
      "num_input_tokens_seen": 60784024,
      "step": 104760
    },
    {
      "epoch": 15.603961870717903,
      "grad_norm": 7.96875,
      "learning_rate": 7.004951916920249e-06,
      "loss": 1.0116,
      "num_input_tokens_seen": 60786904,
      "step": 104765
    },
    {
      "epoch": 15.604706583258862,
      "grad_norm": 7.6875,
      "learning_rate": 7.002696387139265e-06,
      "loss": 1.0782,
      "num_input_tokens_seen": 60790040,
      "step": 104770
    },
    {
      "epoch": 15.60545129579982,
      "grad_norm": 6.46875,
      "learning_rate": 7.000441161404425e-06,
      "loss": 1.184,
      "num_input_tokens_seen": 60792824,
      "step": 104775
    },
    {
      "epoch": 15.606196008340781,
      "grad_norm": 9.0,
      "learning_rate": 6.998186239753846e-06,
      "loss": 0.9463,
      "num_input_tokens_seen": 60795640,
      "step": 104780
    },
    {
      "epoch": 15.60694072088174,
      "grad_norm": 5.84375,
      "learning_rate": 6.995931622225605e-06,
      "loss": 0.9375,
      "num_input_tokens_seen": 60798488,
      "step": 104785
    },
    {
      "epoch": 15.607685433422699,
      "grad_norm": 11.625,
      "learning_rate": 6.99367730885781e-06,
      "loss": 1.1117,
      "num_input_tokens_seen": 60801464,
      "step": 104790
    },
    {
      "epoch": 15.608430145963657,
      "grad_norm": 7.46875,
      "learning_rate": 6.991423299688535e-06,
      "loss": 1.0905,
      "num_input_tokens_seen": 60804248,
      "step": 104795
    },
    {
      "epoch": 15.609174858504618,
      "grad_norm": 11.875,
      "learning_rate": 6.989169594755854e-06,
      "loss": 1.0528,
      "num_input_tokens_seen": 60807032,
      "step": 104800
    },
    {
      "epoch": 15.609919571045577,
      "grad_norm": 15.9375,
      "learning_rate": 6.9869161940978535e-06,
      "loss": 1.0805,
      "num_input_tokens_seen": 60809688,
      "step": 104805
    },
    {
      "epoch": 15.610664283586535,
      "grad_norm": 11.5625,
      "learning_rate": 6.984663097752589e-06,
      "loss": 1.0198,
      "num_input_tokens_seen": 60812472,
      "step": 104810
    },
    {
      "epoch": 15.611408996127494,
      "grad_norm": 9.9375,
      "learning_rate": 6.982410305758138e-06,
      "loss": 1.029,
      "num_input_tokens_seen": 60815640,
      "step": 104815
    },
    {
      "epoch": 15.612153708668455,
      "grad_norm": 15.5625,
      "learning_rate": 6.980157818152547e-06,
      "loss": 0.9682,
      "num_input_tokens_seen": 60818424,
      "step": 104820
    },
    {
      "epoch": 15.612898421209414,
      "grad_norm": 13.125,
      "learning_rate": 6.97790563497388e-06,
      "loss": 1.2049,
      "num_input_tokens_seen": 60821432,
      "step": 104825
    },
    {
      "epoch": 15.613643133750372,
      "grad_norm": 12.5625,
      "learning_rate": 6.9756537562601835e-06,
      "loss": 1.3109,
      "num_input_tokens_seen": 60824440,
      "step": 104830
    },
    {
      "epoch": 15.614387846291331,
      "grad_norm": 19.875,
      "learning_rate": 6.973402182049496e-06,
      "loss": 1.0936,
      "num_input_tokens_seen": 60827192,
      "step": 104835
    },
    {
      "epoch": 15.615132558832292,
      "grad_norm": 7.28125,
      "learning_rate": 6.971150912379859e-06,
      "loss": 0.934,
      "num_input_tokens_seen": 60830360,
      "step": 104840
    },
    {
      "epoch": 15.61587727137325,
      "grad_norm": 21.75,
      "learning_rate": 6.968899947289295e-06,
      "loss": 1.419,
      "num_input_tokens_seen": 60833368,
      "step": 104845
    },
    {
      "epoch": 15.616621983914209,
      "grad_norm": 7.375,
      "learning_rate": 6.966649286815846e-06,
      "loss": 0.966,
      "num_input_tokens_seen": 60836408,
      "step": 104850
    },
    {
      "epoch": 15.617366696455168,
      "grad_norm": 17.75,
      "learning_rate": 6.9643989309975235e-06,
      "loss": 1.0347,
      "num_input_tokens_seen": 60839384,
      "step": 104855
    },
    {
      "epoch": 15.618111408996128,
      "grad_norm": 11.5625,
      "learning_rate": 6.962148879872357e-06,
      "loss": 1.0586,
      "num_input_tokens_seen": 60842104,
      "step": 104860
    },
    {
      "epoch": 15.618856121537087,
      "grad_norm": 7.40625,
      "learning_rate": 6.9598991334783485e-06,
      "loss": 0.8443,
      "num_input_tokens_seen": 60844920,
      "step": 104865
    },
    {
      "epoch": 15.619600834078046,
      "grad_norm": 5.15625,
      "learning_rate": 6.957649691853513e-06,
      "loss": 1.0364,
      "num_input_tokens_seen": 60847896,
      "step": 104870
    },
    {
      "epoch": 15.620345546619005,
      "grad_norm": 8.75,
      "learning_rate": 6.955400555035849e-06,
      "loss": 1.0851,
      "num_input_tokens_seen": 60850872,
      "step": 104875
    },
    {
      "epoch": 15.621090259159963,
      "grad_norm": 16.375,
      "learning_rate": 6.953151723063345e-06,
      "loss": 0.969,
      "num_input_tokens_seen": 60853784,
      "step": 104880
    },
    {
      "epoch": 15.621834971700924,
      "grad_norm": 13.25,
      "learning_rate": 6.95090319597401e-06,
      "loss": 1.1403,
      "num_input_tokens_seen": 60856536,
      "step": 104885
    },
    {
      "epoch": 15.622579684241883,
      "grad_norm": 10.6875,
      "learning_rate": 6.94865497380581e-06,
      "loss": 1.0516,
      "num_input_tokens_seen": 60859224,
      "step": 104890
    },
    {
      "epoch": 15.623324396782841,
      "grad_norm": 5.4375,
      "learning_rate": 6.9464070565967486e-06,
      "loss": 0.8941,
      "num_input_tokens_seen": 60861848,
      "step": 104895
    },
    {
      "epoch": 15.6240691093238,
      "grad_norm": 13.1875,
      "learning_rate": 6.94415944438479e-06,
      "loss": 1.1701,
      "num_input_tokens_seen": 60864824,
      "step": 104900
    },
    {
      "epoch": 15.62481382186476,
      "grad_norm": 10.3125,
      "learning_rate": 6.941912137207907e-06,
      "loss": 1.0702,
      "num_input_tokens_seen": 60867832,
      "step": 104905
    },
    {
      "epoch": 15.62555853440572,
      "grad_norm": 8.375,
      "learning_rate": 6.939665135104056e-06,
      "loss": 1.2476,
      "num_input_tokens_seen": 60870744,
      "step": 104910
    },
    {
      "epoch": 15.626303246946678,
      "grad_norm": 13.1875,
      "learning_rate": 6.9374184381112155e-06,
      "loss": 0.914,
      "num_input_tokens_seen": 60873592,
      "step": 104915
    },
    {
      "epoch": 15.627047959487637,
      "grad_norm": 10.5625,
      "learning_rate": 6.935172046267333e-06,
      "loss": 1.0654,
      "num_input_tokens_seen": 60876344,
      "step": 104920
    },
    {
      "epoch": 15.627792672028598,
      "grad_norm": 10.875,
      "learning_rate": 6.932925959610351e-06,
      "loss": 1.0984,
      "num_input_tokens_seen": 60878904,
      "step": 104925
    },
    {
      "epoch": 15.628537384569556,
      "grad_norm": 17.875,
      "learning_rate": 6.930680178178228e-06,
      "loss": 1.0046,
      "num_input_tokens_seen": 60882168,
      "step": 104930
    },
    {
      "epoch": 15.629282097110515,
      "grad_norm": 7.625,
      "learning_rate": 6.928434702008893e-06,
      "loss": 0.9656,
      "num_input_tokens_seen": 60885208,
      "step": 104935
    },
    {
      "epoch": 15.630026809651474,
      "grad_norm": 6.71875,
      "learning_rate": 6.9261895311402925e-06,
      "loss": 0.8849,
      "num_input_tokens_seen": 60888152,
      "step": 104940
    },
    {
      "epoch": 15.630771522192434,
      "grad_norm": 7.3125,
      "learning_rate": 6.923944665610344e-06,
      "loss": 0.8194,
      "num_input_tokens_seen": 60891192,
      "step": 104945
    },
    {
      "epoch": 15.631516234733393,
      "grad_norm": 18.375,
      "learning_rate": 6.921700105456985e-06,
      "loss": 0.9679,
      "num_input_tokens_seen": 60894232,
      "step": 104950
    },
    {
      "epoch": 15.632260947274352,
      "grad_norm": 12.1875,
      "learning_rate": 6.919455850718123e-06,
      "loss": 1.3286,
      "num_input_tokens_seen": 60897016,
      "step": 104955
    },
    {
      "epoch": 15.63300565981531,
      "grad_norm": 34.0,
      "learning_rate": 6.917211901431683e-06,
      "loss": 1.312,
      "num_input_tokens_seen": 60899704,
      "step": 104960
    },
    {
      "epoch": 15.633750372356271,
      "grad_norm": 7.5625,
      "learning_rate": 6.914968257635573e-06,
      "loss": 0.8964,
      "num_input_tokens_seen": 60902424,
      "step": 104965
    },
    {
      "epoch": 15.63449508489723,
      "grad_norm": 13.25,
      "learning_rate": 6.912724919367691e-06,
      "loss": 1.0298,
      "num_input_tokens_seen": 60905368,
      "step": 104970
    },
    {
      "epoch": 15.635239797438189,
      "grad_norm": 14.375,
      "learning_rate": 6.91048188666594e-06,
      "loss": 1.4385,
      "num_input_tokens_seen": 60908248,
      "step": 104975
    },
    {
      "epoch": 15.635984509979147,
      "grad_norm": 7.625,
      "learning_rate": 6.908239159568203e-06,
      "loss": 1.2979,
      "num_input_tokens_seen": 60911064,
      "step": 104980
    },
    {
      "epoch": 15.636729222520108,
      "grad_norm": 14.0625,
      "learning_rate": 6.9059967381123854e-06,
      "loss": 1.493,
      "num_input_tokens_seen": 60913816,
      "step": 104985
    },
    {
      "epoch": 15.637473935061067,
      "grad_norm": 17.75,
      "learning_rate": 6.903754622336358e-06,
      "loss": 0.9315,
      "num_input_tokens_seen": 60916504,
      "step": 104990
    },
    {
      "epoch": 15.638218647602026,
      "grad_norm": 6.21875,
      "learning_rate": 6.90151281227801e-06,
      "loss": 0.5893,
      "num_input_tokens_seen": 60919352,
      "step": 104995
    },
    {
      "epoch": 15.638963360142984,
      "grad_norm": 12.3125,
      "learning_rate": 6.899271307975208e-06,
      "loss": 0.9659,
      "num_input_tokens_seen": 60922264,
      "step": 105000
    },
    {
      "epoch": 15.639708072683945,
      "grad_norm": 11.1875,
      "learning_rate": 6.897030109465813e-06,
      "loss": 1.2763,
      "num_input_tokens_seen": 60925112,
      "step": 105005
    },
    {
      "epoch": 15.640452785224904,
      "grad_norm": 9.0,
      "learning_rate": 6.894789216787703e-06,
      "loss": 1.121,
      "num_input_tokens_seen": 60927896,
      "step": 105010
    },
    {
      "epoch": 15.641197497765862,
      "grad_norm": 25.25,
      "learning_rate": 6.892548629978721e-06,
      "loss": 0.9798,
      "num_input_tokens_seen": 60930872,
      "step": 105015
    },
    {
      "epoch": 15.641942210306821,
      "grad_norm": 11.8125,
      "learning_rate": 6.890308349076732e-06,
      "loss": 0.9728,
      "num_input_tokens_seen": 60933944,
      "step": 105020
    },
    {
      "epoch": 15.64268692284778,
      "grad_norm": 15.75,
      "learning_rate": 6.88806837411958e-06,
      "loss": 1.2676,
      "num_input_tokens_seen": 60936856,
      "step": 105025
    },
    {
      "epoch": 15.64343163538874,
      "grad_norm": 11.9375,
      "learning_rate": 6.885828705145103e-06,
      "loss": 1.4394,
      "num_input_tokens_seen": 60939768,
      "step": 105030
    },
    {
      "epoch": 15.6441763479297,
      "grad_norm": 8.0,
      "learning_rate": 6.883589342191132e-06,
      "loss": 0.9655,
      "num_input_tokens_seen": 60942424,
      "step": 105035
    },
    {
      "epoch": 15.644921060470658,
      "grad_norm": 14.125,
      "learning_rate": 6.881350285295515e-06,
      "loss": 1.2925,
      "num_input_tokens_seen": 60945400,
      "step": 105040
    },
    {
      "epoch": 15.645665773011618,
      "grad_norm": 10.6875,
      "learning_rate": 6.879111534496069e-06,
      "loss": 1.1601,
      "num_input_tokens_seen": 60948440,
      "step": 105045
    },
    {
      "epoch": 15.646410485552577,
      "grad_norm": 6.21875,
      "learning_rate": 6.87687308983061e-06,
      "loss": 1.0841,
      "num_input_tokens_seen": 60951448,
      "step": 105050
    },
    {
      "epoch": 15.647155198093536,
      "grad_norm": 7.0625,
      "learning_rate": 6.874634951336967e-06,
      "loss": 0.8894,
      "num_input_tokens_seen": 60954328,
      "step": 105055
    },
    {
      "epoch": 15.647899910634495,
      "grad_norm": 21.0,
      "learning_rate": 6.872397119052937e-06,
      "loss": 1.3073,
      "num_input_tokens_seen": 60957368,
      "step": 105060
    },
    {
      "epoch": 15.648644623175453,
      "grad_norm": 8.875,
      "learning_rate": 6.870159593016343e-06,
      "loss": 1.4088,
      "num_input_tokens_seen": 60960600,
      "step": 105065
    },
    {
      "epoch": 15.649389335716414,
      "grad_norm": 10.8125,
      "learning_rate": 6.867922373264968e-06,
      "loss": 0.8303,
      "num_input_tokens_seen": 60963576,
      "step": 105070
    },
    {
      "epoch": 15.650134048257373,
      "grad_norm": 8.0,
      "learning_rate": 6.865685459836621e-06,
      "loss": 0.7619,
      "num_input_tokens_seen": 60966392,
      "step": 105075
    },
    {
      "epoch": 15.650878760798332,
      "grad_norm": 5.15625,
      "learning_rate": 6.8634488527690915e-06,
      "loss": 0.8839,
      "num_input_tokens_seen": 60969272,
      "step": 105080
    },
    {
      "epoch": 15.65162347333929,
      "grad_norm": 11.125,
      "learning_rate": 6.861212552100149e-06,
      "loss": 0.7193,
      "num_input_tokens_seen": 60972056,
      "step": 105085
    },
    {
      "epoch": 15.65236818588025,
      "grad_norm": 10.75,
      "learning_rate": 6.858976557867594e-06,
      "loss": 1.0573,
      "num_input_tokens_seen": 60975032,
      "step": 105090
    },
    {
      "epoch": 15.65311289842121,
      "grad_norm": 12.125,
      "learning_rate": 6.856740870109194e-06,
      "loss": 1.0979,
      "num_input_tokens_seen": 60977912,
      "step": 105095
    },
    {
      "epoch": 15.653857610962168,
      "grad_norm": 5.78125,
      "learning_rate": 6.854505488862714e-06,
      "loss": 1.0512,
      "num_input_tokens_seen": 60980792,
      "step": 105100
    },
    {
      "epoch": 15.654602323503127,
      "grad_norm": 11.625,
      "learning_rate": 6.852270414165915e-06,
      "loss": 0.9686,
      "num_input_tokens_seen": 60983736,
      "step": 105105
    },
    {
      "epoch": 15.655347036044088,
      "grad_norm": 13.875,
      "learning_rate": 6.850035646056571e-06,
      "loss": 1.2764,
      "num_input_tokens_seen": 60986840,
      "step": 105110
    },
    {
      "epoch": 15.656091748585046,
      "grad_norm": 9.1875,
      "learning_rate": 6.847801184572422e-06,
      "loss": 0.8921,
      "num_input_tokens_seen": 60989720,
      "step": 105115
    },
    {
      "epoch": 15.656836461126005,
      "grad_norm": 10.4375,
      "learning_rate": 6.845567029751229e-06,
      "loss": 1.2422,
      "num_input_tokens_seen": 60992824,
      "step": 105120
    },
    {
      "epoch": 15.657581173666964,
      "grad_norm": 17.625,
      "learning_rate": 6.843333181630729e-06,
      "loss": 0.9248,
      "num_input_tokens_seen": 60995704,
      "step": 105125
    },
    {
      "epoch": 15.658325886207924,
      "grad_norm": 9.0,
      "learning_rate": 6.841099640248655e-06,
      "loss": 0.9216,
      "num_input_tokens_seen": 60998744,
      "step": 105130
    },
    {
      "epoch": 15.659070598748883,
      "grad_norm": 9.75,
      "learning_rate": 6.838866405642752e-06,
      "loss": 1.2408,
      "num_input_tokens_seen": 61001368,
      "step": 105135
    },
    {
      "epoch": 15.659815311289842,
      "grad_norm": 8.75,
      "learning_rate": 6.836633477850737e-06,
      "loss": 1.0361,
      "num_input_tokens_seen": 61004696,
      "step": 105140
    },
    {
      "epoch": 15.6605600238308,
      "grad_norm": 6.71875,
      "learning_rate": 6.834400856910348e-06,
      "loss": 1.2337,
      "num_input_tokens_seen": 61007736,
      "step": 105145
    },
    {
      "epoch": 15.661304736371761,
      "grad_norm": 10.625,
      "learning_rate": 6.832168542859283e-06,
      "loss": 1.248,
      "num_input_tokens_seen": 61010712,
      "step": 105150
    },
    {
      "epoch": 15.66204944891272,
      "grad_norm": 14.0625,
      "learning_rate": 6.829936535735273e-06,
      "loss": 1.364,
      "num_input_tokens_seen": 61013432,
      "step": 105155
    },
    {
      "epoch": 15.662794161453679,
      "grad_norm": 15.75,
      "learning_rate": 6.827704835576021e-06,
      "loss": 1.1178,
      "num_input_tokens_seen": 61016120,
      "step": 105160
    },
    {
      "epoch": 15.663538873994638,
      "grad_norm": 9.1875,
      "learning_rate": 6.8254734424192255e-06,
      "loss": 1.0772,
      "num_input_tokens_seen": 61019032,
      "step": 105165
    },
    {
      "epoch": 15.664283586535598,
      "grad_norm": 7.09375,
      "learning_rate": 6.823242356302584e-06,
      "loss": 1.3641,
      "num_input_tokens_seen": 61021912,
      "step": 105170
    },
    {
      "epoch": 15.665028299076557,
      "grad_norm": 8.75,
      "learning_rate": 6.821011577263781e-06,
      "loss": 1.2586,
      "num_input_tokens_seen": 61024888,
      "step": 105175
    },
    {
      "epoch": 15.665773011617516,
      "grad_norm": 14.9375,
      "learning_rate": 6.8187811053405185e-06,
      "loss": 0.8748,
      "num_input_tokens_seen": 61027544,
      "step": 105180
    },
    {
      "epoch": 15.666517724158474,
      "grad_norm": 5.6875,
      "learning_rate": 6.816550940570463e-06,
      "loss": 0.7454,
      "num_input_tokens_seen": 61030520,
      "step": 105185
    },
    {
      "epoch": 15.667262436699435,
      "grad_norm": 14.1875,
      "learning_rate": 6.8143210829913065e-06,
      "loss": 1.2622,
      "num_input_tokens_seen": 61033528,
      "step": 105190
    },
    {
      "epoch": 15.668007149240394,
      "grad_norm": 11.8125,
      "learning_rate": 6.812091532640705e-06,
      "loss": 1.0329,
      "num_input_tokens_seen": 61036568,
      "step": 105195
    },
    {
      "epoch": 15.668751861781352,
      "grad_norm": 10.0625,
      "learning_rate": 6.80986228955634e-06,
      "loss": 1.3308,
      "num_input_tokens_seen": 61039352,
      "step": 105200
    },
    {
      "epoch": 15.669496574322311,
      "grad_norm": 12.375,
      "learning_rate": 6.807633353775861e-06,
      "loss": 1.0563,
      "num_input_tokens_seen": 61042296,
      "step": 105205
    },
    {
      "epoch": 15.67024128686327,
      "grad_norm": 19.5,
      "learning_rate": 6.805404725336923e-06,
      "loss": 1.3968,
      "num_input_tokens_seen": 61045432,
      "step": 105210
    },
    {
      "epoch": 15.67098599940423,
      "grad_norm": 12.5625,
      "learning_rate": 6.803176404277184e-06,
      "loss": 0.6515,
      "num_input_tokens_seen": 61048056,
      "step": 105215
    },
    {
      "epoch": 15.67173071194519,
      "grad_norm": 15.875,
      "learning_rate": 6.800948390634279e-06,
      "loss": 1.1413,
      "num_input_tokens_seen": 61051288,
      "step": 105220
    },
    {
      "epoch": 15.672475424486148,
      "grad_norm": 12.3125,
      "learning_rate": 6.798720684445861e-06,
      "loss": 1.3918,
      "num_input_tokens_seen": 61054072,
      "step": 105225
    },
    {
      "epoch": 15.673220137027108,
      "grad_norm": 12.4375,
      "learning_rate": 6.796493285749561e-06,
      "loss": 1.0113,
      "num_input_tokens_seen": 61056952,
      "step": 105230
    },
    {
      "epoch": 15.673964849568067,
      "grad_norm": 8.9375,
      "learning_rate": 6.794266194583005e-06,
      "loss": 1.0515,
      "num_input_tokens_seen": 61059576,
      "step": 105235
    },
    {
      "epoch": 15.674709562109026,
      "grad_norm": 5.875,
      "learning_rate": 6.792039410983817e-06,
      "loss": 1.4498,
      "num_input_tokens_seen": 61062584,
      "step": 105240
    },
    {
      "epoch": 15.675454274649985,
      "grad_norm": 8.375,
      "learning_rate": 6.789812934989609e-06,
      "loss": 0.8677,
      "num_input_tokens_seen": 61065624,
      "step": 105245
    },
    {
      "epoch": 15.676198987190944,
      "grad_norm": 10.0,
      "learning_rate": 6.78758676663801e-06,
      "loss": 0.9135,
      "num_input_tokens_seen": 61068600,
      "step": 105250
    },
    {
      "epoch": 15.676943699731904,
      "grad_norm": 20.25,
      "learning_rate": 6.785360905966617e-06,
      "loss": 1.378,
      "num_input_tokens_seen": 61071512,
      "step": 105255
    },
    {
      "epoch": 15.677688412272863,
      "grad_norm": 10.8125,
      "learning_rate": 6.783135353013045e-06,
      "loss": 1.0492,
      "num_input_tokens_seen": 61074680,
      "step": 105260
    },
    {
      "epoch": 15.678433124813822,
      "grad_norm": 13.0625,
      "learning_rate": 6.7809101078148805e-06,
      "loss": 1.2284,
      "num_input_tokens_seen": 61077784,
      "step": 105265
    },
    {
      "epoch": 15.67917783735478,
      "grad_norm": 17.375,
      "learning_rate": 6.7786851704097295e-06,
      "loss": 1.2339,
      "num_input_tokens_seen": 61080824,
      "step": 105270
    },
    {
      "epoch": 15.67992254989574,
      "grad_norm": 10.8125,
      "learning_rate": 6.776460540835167e-06,
      "loss": 1.0606,
      "num_input_tokens_seen": 61084024,
      "step": 105275
    },
    {
      "epoch": 15.6806672624367,
      "grad_norm": 16.25,
      "learning_rate": 6.774236219128788e-06,
      "loss": 1.1281,
      "num_input_tokens_seen": 61087032,
      "step": 105280
    },
    {
      "epoch": 15.681411974977658,
      "grad_norm": 17.875,
      "learning_rate": 6.772012205328166e-06,
      "loss": 1.64,
      "num_input_tokens_seen": 61089784,
      "step": 105285
    },
    {
      "epoch": 15.682156687518617,
      "grad_norm": 9.75,
      "learning_rate": 6.769788499470861e-06,
      "loss": 0.8841,
      "num_input_tokens_seen": 61092856,
      "step": 105290
    },
    {
      "epoch": 15.682901400059578,
      "grad_norm": 8.625,
      "learning_rate": 6.767565101594461e-06,
      "loss": 0.961,
      "num_input_tokens_seen": 61095576,
      "step": 105295
    },
    {
      "epoch": 15.683646112600536,
      "grad_norm": 13.5625,
      "learning_rate": 6.765342011736517e-06,
      "loss": 0.9781,
      "num_input_tokens_seen": 61098424,
      "step": 105300
    },
    {
      "epoch": 15.684390825141495,
      "grad_norm": 12.25,
      "learning_rate": 6.763119229934589e-06,
      "loss": 0.9263,
      "num_input_tokens_seen": 61101624,
      "step": 105305
    },
    {
      "epoch": 15.685135537682454,
      "grad_norm": 8.5625,
      "learning_rate": 6.760896756226215e-06,
      "loss": 0.8983,
      "num_input_tokens_seen": 61104376,
      "step": 105310
    },
    {
      "epoch": 15.685880250223414,
      "grad_norm": 4.75,
      "learning_rate": 6.758674590648964e-06,
      "loss": 1.105,
      "num_input_tokens_seen": 61107032,
      "step": 105315
    },
    {
      "epoch": 15.686624962764373,
      "grad_norm": 6.71875,
      "learning_rate": 6.756452733240365e-06,
      "loss": 0.8875,
      "num_input_tokens_seen": 61110392,
      "step": 105320
    },
    {
      "epoch": 15.687369675305332,
      "grad_norm": 10.25,
      "learning_rate": 6.7542311840379455e-06,
      "loss": 0.9523,
      "num_input_tokens_seen": 61113272,
      "step": 105325
    },
    {
      "epoch": 15.68811438784629,
      "grad_norm": 13.875,
      "learning_rate": 6.752009943079257e-06,
      "loss": 1.0354,
      "num_input_tokens_seen": 61115928,
      "step": 105330
    },
    {
      "epoch": 15.688859100387251,
      "grad_norm": 6.8125,
      "learning_rate": 6.749789010401805e-06,
      "loss": 1.0544,
      "num_input_tokens_seen": 61118712,
      "step": 105335
    },
    {
      "epoch": 15.68960381292821,
      "grad_norm": 9.125,
      "learning_rate": 6.7475683860431266e-06,
      "loss": 1.0107,
      "num_input_tokens_seen": 61121496,
      "step": 105340
    },
    {
      "epoch": 15.690348525469169,
      "grad_norm": 8.0625,
      "learning_rate": 6.745348070040722e-06,
      "loss": 0.8995,
      "num_input_tokens_seen": 61124408,
      "step": 105345
    },
    {
      "epoch": 15.691093238010128,
      "grad_norm": 10.3125,
      "learning_rate": 6.743128062432113e-06,
      "loss": 1.2735,
      "num_input_tokens_seen": 61127544,
      "step": 105350
    },
    {
      "epoch": 15.691837950551088,
      "grad_norm": 13.3125,
      "learning_rate": 6.740908363254805e-06,
      "loss": 0.9681,
      "num_input_tokens_seen": 61130296,
      "step": 105355
    },
    {
      "epoch": 15.692582663092047,
      "grad_norm": 9.3125,
      "learning_rate": 6.7386889725462894e-06,
      "loss": 1.188,
      "num_input_tokens_seen": 61133304,
      "step": 105360
    },
    {
      "epoch": 15.693327375633006,
      "grad_norm": 7.53125,
      "learning_rate": 6.736469890344058e-06,
      "loss": 0.8823,
      "num_input_tokens_seen": 61136344,
      "step": 105365
    },
    {
      "epoch": 15.694072088173964,
      "grad_norm": 12.1875,
      "learning_rate": 6.734251116685611e-06,
      "loss": 1.0636,
      "num_input_tokens_seen": 61139224,
      "step": 105370
    },
    {
      "epoch": 15.694816800714925,
      "grad_norm": 8.125,
      "learning_rate": 6.732032651608427e-06,
      "loss": 1.3038,
      "num_input_tokens_seen": 61142264,
      "step": 105375
    },
    {
      "epoch": 15.695561513255884,
      "grad_norm": 10.0,
      "learning_rate": 6.7298144951499774e-06,
      "loss": 0.7446,
      "num_input_tokens_seen": 61145144,
      "step": 105380
    },
    {
      "epoch": 15.696306225796842,
      "grad_norm": 15.5,
      "learning_rate": 6.727596647347753e-06,
      "loss": 0.9993,
      "num_input_tokens_seen": 61148216,
      "step": 105385
    },
    {
      "epoch": 15.697050938337801,
      "grad_norm": 17.25,
      "learning_rate": 6.725379108239202e-06,
      "loss": 1.1814,
      "num_input_tokens_seen": 61151160,
      "step": 105390
    },
    {
      "epoch": 15.69779565087876,
      "grad_norm": 11.125,
      "learning_rate": 6.723161877861805e-06,
      "loss": 0.8458,
      "num_input_tokens_seen": 61154008,
      "step": 105395
    },
    {
      "epoch": 15.69854036341972,
      "grad_norm": 10.875,
      "learning_rate": 6.720944956253012e-06,
      "loss": 1.1115,
      "num_input_tokens_seen": 61156920,
      "step": 105400
    },
    {
      "epoch": 15.69928507596068,
      "grad_norm": 6.78125,
      "learning_rate": 6.71872834345027e-06,
      "loss": 1.2222,
      "num_input_tokens_seen": 61159608,
      "step": 105405
    },
    {
      "epoch": 15.700029788501638,
      "grad_norm": 11.875,
      "learning_rate": 6.716512039491038e-06,
      "loss": 1.2789,
      "num_input_tokens_seen": 61162488,
      "step": 105410
    },
    {
      "epoch": 15.700774501042597,
      "grad_norm": 13.0,
      "learning_rate": 6.714296044412746e-06,
      "loss": 1.1431,
      "num_input_tokens_seen": 61165464,
      "step": 105415
    },
    {
      "epoch": 15.701519213583557,
      "grad_norm": 11.5625,
      "learning_rate": 6.712080358252845e-06,
      "loss": 1.1661,
      "num_input_tokens_seen": 61168376,
      "step": 105420
    },
    {
      "epoch": 15.702263926124516,
      "grad_norm": 8.0625,
      "learning_rate": 6.709864981048761e-06,
      "loss": 0.9054,
      "num_input_tokens_seen": 61171320,
      "step": 105425
    },
    {
      "epoch": 15.703008638665475,
      "grad_norm": 3.875,
      "learning_rate": 6.707649912837919e-06,
      "loss": 1.1228,
      "num_input_tokens_seen": 61174232,
      "step": 105430
    },
    {
      "epoch": 15.703753351206434,
      "grad_norm": 15.0,
      "learning_rate": 6.70543515365773e-06,
      "loss": 1.1571,
      "num_input_tokens_seen": 61177368,
      "step": 105435
    },
    {
      "epoch": 15.704498063747394,
      "grad_norm": 5.8125,
      "learning_rate": 6.703220703545629e-06,
      "loss": 0.9677,
      "num_input_tokens_seen": 61180280,
      "step": 105440
    },
    {
      "epoch": 15.705242776288353,
      "grad_norm": 5.0,
      "learning_rate": 6.701006562539019e-06,
      "loss": 1.5354,
      "num_input_tokens_seen": 61183480,
      "step": 105445
    },
    {
      "epoch": 15.705987488829312,
      "grad_norm": 9.75,
      "learning_rate": 6.698792730675296e-06,
      "loss": 0.9557,
      "num_input_tokens_seen": 61186392,
      "step": 105450
    },
    {
      "epoch": 15.70673220137027,
      "grad_norm": 8.875,
      "learning_rate": 6.6965792079918765e-06,
      "loss": 1.2257,
      "num_input_tokens_seen": 61189336,
      "step": 105455
    },
    {
      "epoch": 15.707476913911231,
      "grad_norm": 14.375,
      "learning_rate": 6.694365994526142e-06,
      "loss": 1.4015,
      "num_input_tokens_seen": 61192216,
      "step": 105460
    },
    {
      "epoch": 15.70822162645219,
      "grad_norm": 6.28125,
      "learning_rate": 6.692153090315498e-06,
      "loss": 0.9131,
      "num_input_tokens_seen": 61194840,
      "step": 105465
    },
    {
      "epoch": 15.708966338993148,
      "grad_norm": 4.625,
      "learning_rate": 6.689940495397309e-06,
      "loss": 0.8443,
      "num_input_tokens_seen": 61197720,
      "step": 105470
    },
    {
      "epoch": 15.709711051534107,
      "grad_norm": 6.40625,
      "learning_rate": 6.687728209808977e-06,
      "loss": 0.9185,
      "num_input_tokens_seen": 61200824,
      "step": 105475
    },
    {
      "epoch": 15.710455764075068,
      "grad_norm": 21.0,
      "learning_rate": 6.6855162335878626e-06,
      "loss": 1.0471,
      "num_input_tokens_seen": 61204216,
      "step": 105480
    },
    {
      "epoch": 15.711200476616026,
      "grad_norm": 10.3125,
      "learning_rate": 6.683304566771331e-06,
      "loss": 1.0536,
      "num_input_tokens_seen": 61207064,
      "step": 105485
    },
    {
      "epoch": 15.711945189156985,
      "grad_norm": 5.4375,
      "learning_rate": 6.68109320939676e-06,
      "loss": 1.0927,
      "num_input_tokens_seen": 61209944,
      "step": 105490
    },
    {
      "epoch": 15.712689901697944,
      "grad_norm": 10.0,
      "learning_rate": 6.678882161501502e-06,
      "loss": 0.7571,
      "num_input_tokens_seen": 61212664,
      "step": 105495
    },
    {
      "epoch": 15.713434614238905,
      "grad_norm": 10.3125,
      "learning_rate": 6.676671423122907e-06,
      "loss": 0.7989,
      "num_input_tokens_seen": 61215416,
      "step": 105500
    },
    {
      "epoch": 15.714179326779863,
      "grad_norm": 4.15625,
      "learning_rate": 6.674460994298317e-06,
      "loss": 0.8922,
      "num_input_tokens_seen": 61218232,
      "step": 105505
    },
    {
      "epoch": 15.714924039320822,
      "grad_norm": 11.8125,
      "learning_rate": 6.672250875065095e-06,
      "loss": 0.7644,
      "num_input_tokens_seen": 61221400,
      "step": 105510
    },
    {
      "epoch": 15.71566875186178,
      "grad_norm": 7.40625,
      "learning_rate": 6.670041065460555e-06,
      "loss": 0.9059,
      "num_input_tokens_seen": 61224248,
      "step": 105515
    },
    {
      "epoch": 15.716413464402741,
      "grad_norm": 9.9375,
      "learning_rate": 6.667831565522051e-06,
      "loss": 0.9361,
      "num_input_tokens_seen": 61227480,
      "step": 105520
    },
    {
      "epoch": 15.7171581769437,
      "grad_norm": 17.5,
      "learning_rate": 6.665622375286901e-06,
      "loss": 1.0265,
      "num_input_tokens_seen": 61230296,
      "step": 105525
    },
    {
      "epoch": 15.717902889484659,
      "grad_norm": 7.1875,
      "learning_rate": 6.66341349479242e-06,
      "loss": 0.8887,
      "num_input_tokens_seen": 61233016,
      "step": 105530
    },
    {
      "epoch": 15.718647602025618,
      "grad_norm": 6.6875,
      "learning_rate": 6.661204924075937e-06,
      "loss": 1.0307,
      "num_input_tokens_seen": 61236184,
      "step": 105535
    },
    {
      "epoch": 15.719392314566576,
      "grad_norm": 13.75,
      "learning_rate": 6.658996663174752e-06,
      "loss": 1.2662,
      "num_input_tokens_seen": 61238744,
      "step": 105540
    },
    {
      "epoch": 15.720137027107537,
      "grad_norm": 10.875,
      "learning_rate": 6.656788712126183e-06,
      "loss": 0.9969,
      "num_input_tokens_seen": 61241560,
      "step": 105545
    },
    {
      "epoch": 15.720881739648496,
      "grad_norm": 13.875,
      "learning_rate": 6.654581070967519e-06,
      "loss": 1.1361,
      "num_input_tokens_seen": 61244312,
      "step": 105550
    },
    {
      "epoch": 15.721626452189454,
      "grad_norm": 13.25,
      "learning_rate": 6.6523737397360705e-06,
      "loss": 1.0681,
      "num_input_tokens_seen": 61247160,
      "step": 105555
    },
    {
      "epoch": 15.722371164730415,
      "grad_norm": 11.6875,
      "learning_rate": 6.65016671846912e-06,
      "loss": 1.1775,
      "num_input_tokens_seen": 61250040,
      "step": 105560
    },
    {
      "epoch": 15.723115877271374,
      "grad_norm": 13.0625,
      "learning_rate": 6.647960007203952e-06,
      "loss": 0.9364,
      "num_input_tokens_seen": 61252664,
      "step": 105565
    },
    {
      "epoch": 15.723860589812332,
      "grad_norm": 12.625,
      "learning_rate": 6.645753605977847e-06,
      "loss": 1.168,
      "num_input_tokens_seen": 61255352,
      "step": 105570
    },
    {
      "epoch": 15.724605302353291,
      "grad_norm": 8.75,
      "learning_rate": 6.643547514828075e-06,
      "loss": 1.1255,
      "num_input_tokens_seen": 61258232,
      "step": 105575
    },
    {
      "epoch": 15.72535001489425,
      "grad_norm": 10.6875,
      "learning_rate": 6.641341733791917e-06,
      "loss": 0.8949,
      "num_input_tokens_seen": 61261176,
      "step": 105580
    },
    {
      "epoch": 15.72609472743521,
      "grad_norm": 8.75,
      "learning_rate": 6.639136262906625e-06,
      "loss": 1.0142,
      "num_input_tokens_seen": 61263896,
      "step": 105585
    },
    {
      "epoch": 15.72683943997617,
      "grad_norm": 11.6875,
      "learning_rate": 6.636931102209471e-06,
      "loss": 1.2075,
      "num_input_tokens_seen": 61266744,
      "step": 105590
    },
    {
      "epoch": 15.727584152517128,
      "grad_norm": 17.125,
      "learning_rate": 6.634726251737697e-06,
      "loss": 1.3727,
      "num_input_tokens_seen": 61269400,
      "step": 105595
    },
    {
      "epoch": 15.728328865058087,
      "grad_norm": 7.6875,
      "learning_rate": 6.632521711528564e-06,
      "loss": 1.1594,
      "num_input_tokens_seen": 61272280,
      "step": 105600
    },
    {
      "epoch": 15.729073577599047,
      "grad_norm": 6.4375,
      "learning_rate": 6.630317481619308e-06,
      "loss": 1.2204,
      "num_input_tokens_seen": 61275288,
      "step": 105605
    },
    {
      "epoch": 15.729818290140006,
      "grad_norm": 13.875,
      "learning_rate": 6.628113562047161e-06,
      "loss": 1.1888,
      "num_input_tokens_seen": 61278104,
      "step": 105610
    },
    {
      "epoch": 15.730563002680965,
      "grad_norm": 15.5625,
      "learning_rate": 6.625909952849368e-06,
      "loss": 1.057,
      "num_input_tokens_seen": 61280856,
      "step": 105615
    },
    {
      "epoch": 15.731307715221924,
      "grad_norm": 7.875,
      "learning_rate": 6.62370665406315e-06,
      "loss": 1.1499,
      "num_input_tokens_seen": 61283832,
      "step": 105620
    },
    {
      "epoch": 15.732052427762884,
      "grad_norm": 12.8125,
      "learning_rate": 6.621503665725734e-06,
      "loss": 1.0319,
      "num_input_tokens_seen": 61286776,
      "step": 105625
    },
    {
      "epoch": 15.732797140303843,
      "grad_norm": 14.375,
      "learning_rate": 6.619300987874336e-06,
      "loss": 0.9026,
      "num_input_tokens_seen": 61289400,
      "step": 105630
    },
    {
      "epoch": 15.733541852844802,
      "grad_norm": 8.0,
      "learning_rate": 6.617098620546166e-06,
      "loss": 1.2569,
      "num_input_tokens_seen": 61291992,
      "step": 105635
    },
    {
      "epoch": 15.73428656538576,
      "grad_norm": 11.4375,
      "learning_rate": 6.614896563778425e-06,
      "loss": 1.0119,
      "num_input_tokens_seen": 61294648,
      "step": 105640
    },
    {
      "epoch": 15.735031277926721,
      "grad_norm": 6.53125,
      "learning_rate": 6.6126948176083284e-06,
      "loss": 0.8964,
      "num_input_tokens_seen": 61297528,
      "step": 105645
    },
    {
      "epoch": 15.73577599046768,
      "grad_norm": 13.9375,
      "learning_rate": 6.610493382073063e-06,
      "loss": 0.9074,
      "num_input_tokens_seen": 61300152,
      "step": 105650
    },
    {
      "epoch": 15.736520703008638,
      "grad_norm": 8.5,
      "learning_rate": 6.6082922572098135e-06,
      "loss": 0.8157,
      "num_input_tokens_seen": 61303032,
      "step": 105655
    },
    {
      "epoch": 15.737265415549597,
      "grad_norm": 10.5,
      "learning_rate": 6.6060914430557845e-06,
      "loss": 0.909,
      "num_input_tokens_seen": 61306072,
      "step": 105660
    },
    {
      "epoch": 15.738010128090558,
      "grad_norm": 11.375,
      "learning_rate": 6.603890939648136e-06,
      "loss": 0.9446,
      "num_input_tokens_seen": 61309176,
      "step": 105665
    },
    {
      "epoch": 15.738754840631517,
      "grad_norm": 5.3125,
      "learning_rate": 6.601690747024061e-06,
      "loss": 1.0015,
      "num_input_tokens_seen": 61312184,
      "step": 105670
    },
    {
      "epoch": 15.739499553172475,
      "grad_norm": 18.75,
      "learning_rate": 6.599490865220714e-06,
      "loss": 1.3264,
      "num_input_tokens_seen": 61315672,
      "step": 105675
    },
    {
      "epoch": 15.740244265713434,
      "grad_norm": 13.9375,
      "learning_rate": 6.597291294275276e-06,
      "loss": 1.1118,
      "num_input_tokens_seen": 61318488,
      "step": 105680
    },
    {
      "epoch": 15.740988978254395,
      "grad_norm": 6.375,
      "learning_rate": 6.595092034224898e-06,
      "loss": 0.9725,
      "num_input_tokens_seen": 61321176,
      "step": 105685
    },
    {
      "epoch": 15.741733690795353,
      "grad_norm": 20.875,
      "learning_rate": 6.592893085106733e-06,
      "loss": 1.0739,
      "num_input_tokens_seen": 61324152,
      "step": 105690
    },
    {
      "epoch": 15.742478403336312,
      "grad_norm": 4.59375,
      "learning_rate": 6.590694446957924e-06,
      "loss": 0.7922,
      "num_input_tokens_seen": 61326904,
      "step": 105695
    },
    {
      "epoch": 15.74322311587727,
      "grad_norm": 8.3125,
      "learning_rate": 6.588496119815629e-06,
      "loss": 1.0201,
      "num_input_tokens_seen": 61329784,
      "step": 105700
    },
    {
      "epoch": 15.743967828418231,
      "grad_norm": 12.0625,
      "learning_rate": 6.5862981037169816e-06,
      "loss": 0.8676,
      "num_input_tokens_seen": 61332760,
      "step": 105705
    },
    {
      "epoch": 15.74471254095919,
      "grad_norm": 6.21875,
      "learning_rate": 6.584100398699103e-06,
      "loss": 0.8596,
      "num_input_tokens_seen": 61335544,
      "step": 105710
    },
    {
      "epoch": 15.745457253500149,
      "grad_norm": 13.625,
      "learning_rate": 6.581903004799139e-06,
      "loss": 0.6542,
      "num_input_tokens_seen": 61338168,
      "step": 105715
    },
    {
      "epoch": 15.746201966041108,
      "grad_norm": 9.9375,
      "learning_rate": 6.5797059220541965e-06,
      "loss": 0.9048,
      "num_input_tokens_seen": 61341240,
      "step": 105720
    },
    {
      "epoch": 15.746946678582066,
      "grad_norm": 12.6875,
      "learning_rate": 6.57750915050141e-06,
      "loss": 1.5547,
      "num_input_tokens_seen": 61343896,
      "step": 105725
    },
    {
      "epoch": 15.747691391123027,
      "grad_norm": 4.1875,
      "learning_rate": 6.575312690177882e-06,
      "loss": 1.012,
      "num_input_tokens_seen": 61346488,
      "step": 105730
    },
    {
      "epoch": 15.748436103663986,
      "grad_norm": 7.5625,
      "learning_rate": 6.573116541120714e-06,
      "loss": 1.0787,
      "num_input_tokens_seen": 61349368,
      "step": 105735
    },
    {
      "epoch": 15.749180816204944,
      "grad_norm": 10.0625,
      "learning_rate": 6.570920703367017e-06,
      "loss": 1.1231,
      "num_input_tokens_seen": 61352344,
      "step": 105740
    },
    {
      "epoch": 15.749925528745905,
      "grad_norm": 19.0,
      "learning_rate": 6.5687251769538795e-06,
      "loss": 1.0374,
      "num_input_tokens_seen": 61354968,
      "step": 105745
    },
    {
      "epoch": 15.750670241286864,
      "grad_norm": 11.6875,
      "learning_rate": 6.566529961918405e-06,
      "loss": 1.4685,
      "num_input_tokens_seen": 61357944,
      "step": 105750
    },
    {
      "epoch": 15.751414953827823,
      "grad_norm": 11.3125,
      "learning_rate": 6.564335058297674e-06,
      "loss": 1.4313,
      "num_input_tokens_seen": 61360728,
      "step": 105755
    },
    {
      "epoch": 15.752159666368781,
      "grad_norm": 13.8125,
      "learning_rate": 6.56214046612876e-06,
      "loss": 0.9729,
      "num_input_tokens_seen": 61363512,
      "step": 105760
    },
    {
      "epoch": 15.75290437890974,
      "grad_norm": 14.75,
      "learning_rate": 6.5599461854487485e-06,
      "loss": 1.3528,
      "num_input_tokens_seen": 61366392,
      "step": 105765
    },
    {
      "epoch": 15.7536490914507,
      "grad_norm": 12.4375,
      "learning_rate": 6.557752216294693e-06,
      "loss": 0.9704,
      "num_input_tokens_seen": 61369304,
      "step": 105770
    },
    {
      "epoch": 15.75439380399166,
      "grad_norm": 7.125,
      "learning_rate": 6.555558558703681e-06,
      "loss": 1.0498,
      "num_input_tokens_seen": 61372056,
      "step": 105775
    },
    {
      "epoch": 15.755138516532618,
      "grad_norm": 13.9375,
      "learning_rate": 6.5533652127127505e-06,
      "loss": 1.3245,
      "num_input_tokens_seen": 61375160,
      "step": 105780
    },
    {
      "epoch": 15.755883229073577,
      "grad_norm": 9.375,
      "learning_rate": 6.551172178358975e-06,
      "loss": 0.6358,
      "num_input_tokens_seen": 61378168,
      "step": 105785
    },
    {
      "epoch": 15.756627941614537,
      "grad_norm": 11.0625,
      "learning_rate": 6.548979455679388e-06,
      "loss": 0.9725,
      "num_input_tokens_seen": 61381016,
      "step": 105790
    },
    {
      "epoch": 15.757372654155496,
      "grad_norm": 5.40625,
      "learning_rate": 6.5467870447110475e-06,
      "loss": 1.0059,
      "num_input_tokens_seen": 61383928,
      "step": 105795
    },
    {
      "epoch": 15.758117366696455,
      "grad_norm": 12.25,
      "learning_rate": 6.544594945490978e-06,
      "loss": 1.254,
      "num_input_tokens_seen": 61386808,
      "step": 105800
    },
    {
      "epoch": 15.758862079237414,
      "grad_norm": 13.3125,
      "learning_rate": 6.542403158056226e-06,
      "loss": 1.6094,
      "num_input_tokens_seen": 61389528,
      "step": 105805
    },
    {
      "epoch": 15.759606791778374,
      "grad_norm": 5.78125,
      "learning_rate": 6.540211682443814e-06,
      "loss": 0.8366,
      "num_input_tokens_seen": 61392536,
      "step": 105810
    },
    {
      "epoch": 15.760351504319333,
      "grad_norm": 15.4375,
      "learning_rate": 6.5380205186907545e-06,
      "loss": 1.3058,
      "num_input_tokens_seen": 61395448,
      "step": 105815
    },
    {
      "epoch": 15.761096216860292,
      "grad_norm": 10.4375,
      "learning_rate": 6.535829666834084e-06,
      "loss": 1.0744,
      "num_input_tokens_seen": 61398008,
      "step": 105820
    },
    {
      "epoch": 15.76184092940125,
      "grad_norm": 17.125,
      "learning_rate": 6.533639126910804e-06,
      "loss": 1.1892,
      "num_input_tokens_seen": 61400792,
      "step": 105825
    },
    {
      "epoch": 15.762585641942211,
      "grad_norm": 4.625,
      "learning_rate": 6.531448898957923e-06,
      "loss": 1.0411,
      "num_input_tokens_seen": 61403512,
      "step": 105830
    },
    {
      "epoch": 15.76333035448317,
      "grad_norm": 10.25,
      "learning_rate": 6.529258983012437e-06,
      "loss": 1.2501,
      "num_input_tokens_seen": 61406232,
      "step": 105835
    },
    {
      "epoch": 15.764075067024129,
      "grad_norm": 9.0625,
      "learning_rate": 6.527069379111353e-06,
      "loss": 1.2054,
      "num_input_tokens_seen": 61409080,
      "step": 105840
    },
    {
      "epoch": 15.764819779565087,
      "grad_norm": 17.375,
      "learning_rate": 6.5248800872916574e-06,
      "loss": 1.1552,
      "num_input_tokens_seen": 61411960,
      "step": 105845
    },
    {
      "epoch": 15.765564492106048,
      "grad_norm": 12.0625,
      "learning_rate": 6.5226911075903255e-06,
      "loss": 1.0315,
      "num_input_tokens_seen": 61414904,
      "step": 105850
    },
    {
      "epoch": 15.766309204647007,
      "grad_norm": 4.125,
      "learning_rate": 6.5205024400443575e-06,
      "loss": 1.0052,
      "num_input_tokens_seen": 61417496,
      "step": 105855
    },
    {
      "epoch": 15.767053917187965,
      "grad_norm": 12.25,
      "learning_rate": 6.51831408469071e-06,
      "loss": 1.0931,
      "num_input_tokens_seen": 61420504,
      "step": 105860
    },
    {
      "epoch": 15.767798629728924,
      "grad_norm": 13.8125,
      "learning_rate": 6.516126041566373e-06,
      "loss": 1.0831,
      "num_input_tokens_seen": 61423608,
      "step": 105865
    },
    {
      "epoch": 15.768543342269885,
      "grad_norm": 11.125,
      "learning_rate": 6.51393831070829e-06,
      "loss": 1.0755,
      "num_input_tokens_seen": 61426520,
      "step": 105870
    },
    {
      "epoch": 15.769288054810843,
      "grad_norm": 6.84375,
      "learning_rate": 6.511750892153439e-06,
      "loss": 1.2767,
      "num_input_tokens_seen": 61429176,
      "step": 105875
    },
    {
      "epoch": 15.770032767351802,
      "grad_norm": 13.125,
      "learning_rate": 6.50956378593876e-06,
      "loss": 1.0079,
      "num_input_tokens_seen": 61432120,
      "step": 105880
    },
    {
      "epoch": 15.770777479892761,
      "grad_norm": 7.625,
      "learning_rate": 6.507376992101214e-06,
      "loss": 0.9999,
      "num_input_tokens_seen": 61434904,
      "step": 105885
    },
    {
      "epoch": 15.771522192433721,
      "grad_norm": 14.3125,
      "learning_rate": 6.505190510677739e-06,
      "loss": 1.4035,
      "num_input_tokens_seen": 61437752,
      "step": 105890
    },
    {
      "epoch": 15.77226690497468,
      "grad_norm": 10.8125,
      "learning_rate": 6.503004341705271e-06,
      "loss": 1.0374,
      "num_input_tokens_seen": 61440664,
      "step": 105895
    },
    {
      "epoch": 15.773011617515639,
      "grad_norm": 10.0,
      "learning_rate": 6.5008184852207475e-06,
      "loss": 1.2321,
      "num_input_tokens_seen": 61443512,
      "step": 105900
    },
    {
      "epoch": 15.773756330056598,
      "grad_norm": 10.0625,
      "learning_rate": 6.498632941261088e-06,
      "loss": 0.5317,
      "num_input_tokens_seen": 61446456,
      "step": 105905
    },
    {
      "epoch": 15.774501042597556,
      "grad_norm": 9.1875,
      "learning_rate": 6.496447709863227e-06,
      "loss": 1.4469,
      "num_input_tokens_seen": 61449368,
      "step": 105910
    },
    {
      "epoch": 15.775245755138517,
      "grad_norm": 13.0625,
      "learning_rate": 6.494262791064065e-06,
      "loss": 0.8954,
      "num_input_tokens_seen": 61452248,
      "step": 105915
    },
    {
      "epoch": 15.775990467679476,
      "grad_norm": 7.15625,
      "learning_rate": 6.4920781849005355e-06,
      "loss": 0.9418,
      "num_input_tokens_seen": 61455064,
      "step": 105920
    },
    {
      "epoch": 15.776735180220435,
      "grad_norm": 9.9375,
      "learning_rate": 6.489893891409535e-06,
      "loss": 1.2785,
      "num_input_tokens_seen": 61457944,
      "step": 105925
    },
    {
      "epoch": 15.777479892761393,
      "grad_norm": 12.625,
      "learning_rate": 6.487709910627957e-06,
      "loss": 1.047,
      "num_input_tokens_seen": 61460632,
      "step": 105930
    },
    {
      "epoch": 15.778224605302354,
      "grad_norm": 7.375,
      "learning_rate": 6.48552624259271e-06,
      "loss": 1.3235,
      "num_input_tokens_seen": 61463288,
      "step": 105935
    },
    {
      "epoch": 15.778969317843313,
      "grad_norm": 13.9375,
      "learning_rate": 6.483342887340674e-06,
      "loss": 1.3057,
      "num_input_tokens_seen": 61466392,
      "step": 105940
    },
    {
      "epoch": 15.779714030384271,
      "grad_norm": 10.625,
      "learning_rate": 6.481159844908746e-06,
      "loss": 1.2561,
      "num_input_tokens_seen": 61469400,
      "step": 105945
    },
    {
      "epoch": 15.78045874292523,
      "grad_norm": 8.4375,
      "learning_rate": 6.478977115333796e-06,
      "loss": 0.9888,
      "num_input_tokens_seen": 61472344,
      "step": 105950
    },
    {
      "epoch": 15.78120345546619,
      "grad_norm": 6.0625,
      "learning_rate": 6.4767946986527105e-06,
      "loss": 1.3375,
      "num_input_tokens_seen": 61475128,
      "step": 105955
    },
    {
      "epoch": 15.78194816800715,
      "grad_norm": 18.625,
      "learning_rate": 6.474612594902351e-06,
      "loss": 1.3176,
      "num_input_tokens_seen": 61478360,
      "step": 105960
    },
    {
      "epoch": 15.782692880548108,
      "grad_norm": 11.5,
      "learning_rate": 6.472430804119584e-06,
      "loss": 0.8565,
      "num_input_tokens_seen": 61481400,
      "step": 105965
    },
    {
      "epoch": 15.783437593089067,
      "grad_norm": 6.40625,
      "learning_rate": 6.470249326341269e-06,
      "loss": 1.0084,
      "num_input_tokens_seen": 61484184,
      "step": 105970
    },
    {
      "epoch": 15.784182305630027,
      "grad_norm": 6.0,
      "learning_rate": 6.468068161604251e-06,
      "loss": 1.0049,
      "num_input_tokens_seen": 61487256,
      "step": 105975
    },
    {
      "epoch": 15.784927018170986,
      "grad_norm": 4.96875,
      "learning_rate": 6.465887309945393e-06,
      "loss": 0.9689,
      "num_input_tokens_seen": 61490200,
      "step": 105980
    },
    {
      "epoch": 15.785671730711945,
      "grad_norm": 8.0,
      "learning_rate": 6.4637067714015255e-06,
      "loss": 0.976,
      "num_input_tokens_seen": 61493496,
      "step": 105985
    },
    {
      "epoch": 15.786416443252904,
      "grad_norm": 13.625,
      "learning_rate": 6.4615265460095e-06,
      "loss": 0.982,
      "num_input_tokens_seen": 61496088,
      "step": 105990
    },
    {
      "epoch": 15.787161155793864,
      "grad_norm": 7.03125,
      "learning_rate": 6.459346633806132e-06,
      "loss": 1.048,
      "num_input_tokens_seen": 61499000,
      "step": 105995
    },
    {
      "epoch": 15.787905868334823,
      "grad_norm": 4.28125,
      "learning_rate": 6.457167034828268e-06,
      "loss": 0.9644,
      "num_input_tokens_seen": 61501784,
      "step": 106000
    },
    {
      "epoch": 15.788650580875782,
      "grad_norm": 6.34375,
      "learning_rate": 6.45498774911272e-06,
      "loss": 1.0439,
      "num_input_tokens_seen": 61504664,
      "step": 106005
    },
    {
      "epoch": 15.78939529341674,
      "grad_norm": 15.125,
      "learning_rate": 6.452808776696301e-06,
      "loss": 1.2052,
      "num_input_tokens_seen": 61507480,
      "step": 106010
    },
    {
      "epoch": 15.790140005957701,
      "grad_norm": 12.5625,
      "learning_rate": 6.450630117615833e-06,
      "loss": 1.2142,
      "num_input_tokens_seen": 61510360,
      "step": 106015
    },
    {
      "epoch": 15.79088471849866,
      "grad_norm": 10.125,
      "learning_rate": 6.448451771908115e-06,
      "loss": 0.8685,
      "num_input_tokens_seen": 61513336,
      "step": 106020
    },
    {
      "epoch": 15.791629431039619,
      "grad_norm": 7.1875,
      "learning_rate": 6.446273739609943e-06,
      "loss": 1.424,
      "num_input_tokens_seen": 61516472,
      "step": 106025
    },
    {
      "epoch": 15.792374143580577,
      "grad_norm": 3.921875,
      "learning_rate": 6.444096020758125e-06,
      "loss": 0.9888,
      "num_input_tokens_seen": 61519480,
      "step": 106030
    },
    {
      "epoch": 15.793118856121538,
      "grad_norm": 10.8125,
      "learning_rate": 6.4419186153894475e-06,
      "loss": 1.1861,
      "num_input_tokens_seen": 61521976,
      "step": 106035
    },
    {
      "epoch": 15.793863568662497,
      "grad_norm": 4.625,
      "learning_rate": 6.439741523540685e-06,
      "loss": 0.8337,
      "num_input_tokens_seen": 61524760,
      "step": 106040
    },
    {
      "epoch": 15.794608281203455,
      "grad_norm": 5.75,
      "learning_rate": 6.437564745248634e-06,
      "loss": 0.8658,
      "num_input_tokens_seen": 61527352,
      "step": 106045
    },
    {
      "epoch": 15.795352993744414,
      "grad_norm": 9.4375,
      "learning_rate": 6.435388280550062e-06,
      "loss": 1.2309,
      "num_input_tokens_seen": 61530232,
      "step": 106050
    },
    {
      "epoch": 15.796097706285373,
      "grad_norm": 10.625,
      "learning_rate": 6.4332121294817245e-06,
      "loss": 1.3049,
      "num_input_tokens_seen": 61532888,
      "step": 106055
    },
    {
      "epoch": 15.796842418826333,
      "grad_norm": 11.6875,
      "learning_rate": 6.431036292080409e-06,
      "loss": 1.0205,
      "num_input_tokens_seen": 61535768,
      "step": 106060
    },
    {
      "epoch": 15.797587131367292,
      "grad_norm": 8.5625,
      "learning_rate": 6.428860768382855e-06,
      "loss": 0.9715,
      "num_input_tokens_seen": 61538712,
      "step": 106065
    },
    {
      "epoch": 15.798331843908251,
      "grad_norm": 8.0,
      "learning_rate": 6.42668555842583e-06,
      "loss": 0.9798,
      "num_input_tokens_seen": 61541432,
      "step": 106070
    },
    {
      "epoch": 15.799076556449211,
      "grad_norm": 15.375,
      "learning_rate": 6.4245106622460665e-06,
      "loss": 1.2323,
      "num_input_tokens_seen": 61544024,
      "step": 106075
    },
    {
      "epoch": 15.79982126899017,
      "grad_norm": 18.875,
      "learning_rate": 6.422336079880325e-06,
      "loss": 1.0433,
      "num_input_tokens_seen": 61546616,
      "step": 106080
    },
    {
      "epoch": 15.800565981531129,
      "grad_norm": 13.8125,
      "learning_rate": 6.420161811365336e-06,
      "loss": 1.0522,
      "num_input_tokens_seen": 61549848,
      "step": 106085
    },
    {
      "epoch": 15.801310694072088,
      "grad_norm": 18.25,
      "learning_rate": 6.417987856737825e-06,
      "loss": 1.297,
      "num_input_tokens_seen": 61552536,
      "step": 106090
    },
    {
      "epoch": 15.802055406613047,
      "grad_norm": 7.15625,
      "learning_rate": 6.415814216034527e-06,
      "loss": 1.0091,
      "num_input_tokens_seen": 61555320,
      "step": 106095
    },
    {
      "epoch": 15.802800119154007,
      "grad_norm": 14.25,
      "learning_rate": 6.41364088929215e-06,
      "loss": 1.1757,
      "num_input_tokens_seen": 61558360,
      "step": 106100
    },
    {
      "epoch": 15.803544831694966,
      "grad_norm": 8.9375,
      "learning_rate": 6.4114678765474275e-06,
      "loss": 0.9424,
      "num_input_tokens_seen": 61561176,
      "step": 106105
    },
    {
      "epoch": 15.804289544235925,
      "grad_norm": 17.5,
      "learning_rate": 6.409295177837058e-06,
      "loss": 1.3803,
      "num_input_tokens_seen": 61564024,
      "step": 106110
    },
    {
      "epoch": 15.805034256776883,
      "grad_norm": 7.78125,
      "learning_rate": 6.407122793197756e-06,
      "loss": 1.3548,
      "num_input_tokens_seen": 61566808,
      "step": 106115
    },
    {
      "epoch": 15.805778969317844,
      "grad_norm": 8.625,
      "learning_rate": 6.404950722666211e-06,
      "loss": 0.9874,
      "num_input_tokens_seen": 61569528,
      "step": 106120
    },
    {
      "epoch": 15.806523681858803,
      "grad_norm": 9.1875,
      "learning_rate": 6.402778966279133e-06,
      "loss": 1.1314,
      "num_input_tokens_seen": 61572184,
      "step": 106125
    },
    {
      "epoch": 15.807268394399761,
      "grad_norm": 6.65625,
      "learning_rate": 6.400607524073201e-06,
      "loss": 1.0021,
      "num_input_tokens_seen": 61575352,
      "step": 106130
    },
    {
      "epoch": 15.80801310694072,
      "grad_norm": 10.375,
      "learning_rate": 6.3984363960850945e-06,
      "loss": 1.1931,
      "num_input_tokens_seen": 61578328,
      "step": 106135
    },
    {
      "epoch": 15.80875781948168,
      "grad_norm": 6.34375,
      "learning_rate": 6.396265582351508e-06,
      "loss": 1.1822,
      "num_input_tokens_seen": 61581208,
      "step": 106140
    },
    {
      "epoch": 15.80950253202264,
      "grad_norm": 12.5,
      "learning_rate": 6.394095082909099e-06,
      "loss": 0.8358,
      "num_input_tokens_seen": 61584120,
      "step": 106145
    },
    {
      "epoch": 15.810247244563598,
      "grad_norm": 5.03125,
      "learning_rate": 6.391924897794549e-06,
      "loss": 1.085,
      "num_input_tokens_seen": 61586744,
      "step": 106150
    },
    {
      "epoch": 15.810991957104557,
      "grad_norm": 12.8125,
      "learning_rate": 6.3897550270445165e-06,
      "loss": 0.857,
      "num_input_tokens_seen": 61589912,
      "step": 106155
    },
    {
      "epoch": 15.811736669645517,
      "grad_norm": 9.6875,
      "learning_rate": 6.387585470695659e-06,
      "loss": 0.8754,
      "num_input_tokens_seen": 61592760,
      "step": 106160
    },
    {
      "epoch": 15.812481382186476,
      "grad_norm": 12.875,
      "learning_rate": 6.385416228784618e-06,
      "loss": 1.0531,
      "num_input_tokens_seen": 61595736,
      "step": 106165
    },
    {
      "epoch": 15.813226094727435,
      "grad_norm": 9.0625,
      "learning_rate": 6.383247301348061e-06,
      "loss": 0.9978,
      "num_input_tokens_seen": 61598616,
      "step": 106170
    },
    {
      "epoch": 15.813970807268394,
      "grad_norm": 9.125,
      "learning_rate": 6.381078688422617e-06,
      "loss": 1.3802,
      "num_input_tokens_seen": 61601752,
      "step": 106175
    },
    {
      "epoch": 15.814715519809354,
      "grad_norm": 7.09375,
      "learning_rate": 6.3789103900449205e-06,
      "loss": 0.6626,
      "num_input_tokens_seen": 61604664,
      "step": 106180
    },
    {
      "epoch": 15.815460232350313,
      "grad_norm": 9.5625,
      "learning_rate": 6.3767424062516155e-06,
      "loss": 1.123,
      "num_input_tokens_seen": 61607672,
      "step": 106185
    },
    {
      "epoch": 15.816204944891272,
      "grad_norm": 17.125,
      "learning_rate": 6.374574737079309e-06,
      "loss": 1.1355,
      "num_input_tokens_seen": 61610904,
      "step": 106190
    },
    {
      "epoch": 15.81694965743223,
      "grad_norm": 7.90625,
      "learning_rate": 6.372407382564641e-06,
      "loss": 0.9999,
      "num_input_tokens_seen": 61613720,
      "step": 106195
    },
    {
      "epoch": 15.817694369973191,
      "grad_norm": 14.375,
      "learning_rate": 6.37024034274421e-06,
      "loss": 1.1647,
      "num_input_tokens_seen": 61616504,
      "step": 106200
    },
    {
      "epoch": 15.81843908251415,
      "grad_norm": 6.71875,
      "learning_rate": 6.368073617654643e-06,
      "loss": 1.3016,
      "num_input_tokens_seen": 61619608,
      "step": 106205
    },
    {
      "epoch": 15.819183795055109,
      "grad_norm": 8.375,
      "learning_rate": 6.365907207332536e-06,
      "loss": 0.9228,
      "num_input_tokens_seen": 61622520,
      "step": 106210
    },
    {
      "epoch": 15.819928507596067,
      "grad_norm": 9.125,
      "learning_rate": 6.3637411118144776e-06,
      "loss": 0.7243,
      "num_input_tokens_seen": 61625464,
      "step": 106215
    },
    {
      "epoch": 15.820673220137028,
      "grad_norm": 10.8125,
      "learning_rate": 6.361575331137082e-06,
      "loss": 1.1979,
      "num_input_tokens_seen": 61628184,
      "step": 106220
    },
    {
      "epoch": 15.821417932677987,
      "grad_norm": 22.75,
      "learning_rate": 6.359409865336927e-06,
      "loss": 1.1469,
      "num_input_tokens_seen": 61631000,
      "step": 106225
    },
    {
      "epoch": 15.822162645218945,
      "grad_norm": 6.09375,
      "learning_rate": 6.357244714450597e-06,
      "loss": 1.1945,
      "num_input_tokens_seen": 61633752,
      "step": 106230
    },
    {
      "epoch": 15.822907357759904,
      "grad_norm": 9.5,
      "learning_rate": 6.355079878514661e-06,
      "loss": 1.6371,
      "num_input_tokens_seen": 61636600,
      "step": 106235
    },
    {
      "epoch": 15.823652070300863,
      "grad_norm": 12.3125,
      "learning_rate": 6.352915357565712e-06,
      "loss": 1.2524,
      "num_input_tokens_seen": 61639288,
      "step": 106240
    },
    {
      "epoch": 15.824396782841823,
      "grad_norm": 11.6875,
      "learning_rate": 6.350751151640294e-06,
      "loss": 1.3119,
      "num_input_tokens_seen": 61642424,
      "step": 106245
    },
    {
      "epoch": 15.825141495382782,
      "grad_norm": 7.90625,
      "learning_rate": 6.348587260774991e-06,
      "loss": 1.08,
      "num_input_tokens_seen": 61645176,
      "step": 106250
    },
    {
      "epoch": 15.825886207923741,
      "grad_norm": 7.125,
      "learning_rate": 6.346423685006348e-06,
      "loss": 1.2006,
      "num_input_tokens_seen": 61648472,
      "step": 106255
    },
    {
      "epoch": 15.826630920464702,
      "grad_norm": 14.3125,
      "learning_rate": 6.344260424370912e-06,
      "loss": 0.8641,
      "num_input_tokens_seen": 61651512,
      "step": 106260
    },
    {
      "epoch": 15.82737563300566,
      "grad_norm": 10.0625,
      "learning_rate": 6.342097478905243e-06,
      "loss": 1.4054,
      "num_input_tokens_seen": 61654360,
      "step": 106265
    },
    {
      "epoch": 15.828120345546619,
      "grad_norm": 15.375,
      "learning_rate": 6.339934848645868e-06,
      "loss": 1.0695,
      "num_input_tokens_seen": 61657144,
      "step": 106270
    },
    {
      "epoch": 15.828865058087578,
      "grad_norm": 15.6875,
      "learning_rate": 6.337772533629333e-06,
      "loss": 0.9259,
      "num_input_tokens_seen": 61659992,
      "step": 106275
    },
    {
      "epoch": 15.829609770628537,
      "grad_norm": 14.0,
      "learning_rate": 6.335610533892156e-06,
      "loss": 0.8582,
      "num_input_tokens_seen": 61662552,
      "step": 106280
    },
    {
      "epoch": 15.830354483169497,
      "grad_norm": 5.5625,
      "learning_rate": 6.33344884947088e-06,
      "loss": 0.9645,
      "num_input_tokens_seen": 61665560,
      "step": 106285
    },
    {
      "epoch": 15.831099195710456,
      "grad_norm": 6.1875,
      "learning_rate": 6.331287480402012e-06,
      "loss": 1.1139,
      "num_input_tokens_seen": 61668216,
      "step": 106290
    },
    {
      "epoch": 15.831843908251415,
      "grad_norm": 6.125,
      "learning_rate": 6.329126426722068e-06,
      "loss": 0.8142,
      "num_input_tokens_seen": 61671192,
      "step": 106295
    },
    {
      "epoch": 15.832588620792373,
      "grad_norm": 14.5625,
      "learning_rate": 6.326965688467557e-06,
      "loss": 1.3299,
      "num_input_tokens_seen": 61674104,
      "step": 106300
    },
    {
      "epoch": 15.833333333333334,
      "grad_norm": 13.125,
      "learning_rate": 6.324805265674974e-06,
      "loss": 1.1641,
      "num_input_tokens_seen": 61677176,
      "step": 106305
    },
    {
      "epoch": 15.834078045874293,
      "grad_norm": 14.75,
      "learning_rate": 6.322645158380833e-06,
      "loss": 1.1699,
      "num_input_tokens_seen": 61679928,
      "step": 106310
    },
    {
      "epoch": 15.834822758415251,
      "grad_norm": 7.8125,
      "learning_rate": 6.320485366621612e-06,
      "loss": 0.9605,
      "num_input_tokens_seen": 61682840,
      "step": 106315
    },
    {
      "epoch": 15.83556747095621,
      "grad_norm": 20.875,
      "learning_rate": 6.318325890433813e-06,
      "loss": 0.9405,
      "num_input_tokens_seen": 61685656,
      "step": 106320
    },
    {
      "epoch": 15.83631218349717,
      "grad_norm": 11.4375,
      "learning_rate": 6.316166729853906e-06,
      "loss": 1.2599,
      "num_input_tokens_seen": 61688312,
      "step": 106325
    },
    {
      "epoch": 15.83705689603813,
      "grad_norm": 5.65625,
      "learning_rate": 6.314007884918377e-06,
      "loss": 1.0357,
      "num_input_tokens_seen": 61691288,
      "step": 106330
    },
    {
      "epoch": 15.837801608579088,
      "grad_norm": 22.5,
      "learning_rate": 6.311849355663693e-06,
      "loss": 1.0789,
      "num_input_tokens_seen": 61694296,
      "step": 106335
    },
    {
      "epoch": 15.838546321120047,
      "grad_norm": 14.125,
      "learning_rate": 6.309691142126315e-06,
      "loss": 1.3248,
      "num_input_tokens_seen": 61697336,
      "step": 106340
    },
    {
      "epoch": 15.839291033661008,
      "grad_norm": 8.6875,
      "learning_rate": 6.307533244342717e-06,
      "loss": 1.3212,
      "num_input_tokens_seen": 61700504,
      "step": 106345
    },
    {
      "epoch": 15.840035746201966,
      "grad_norm": 10.75,
      "learning_rate": 6.305375662349344e-06,
      "loss": 1.1439,
      "num_input_tokens_seen": 61703416,
      "step": 106350
    },
    {
      "epoch": 15.840780458742925,
      "grad_norm": 13.6875,
      "learning_rate": 6.303218396182644e-06,
      "loss": 0.9819,
      "num_input_tokens_seen": 61706232,
      "step": 106355
    },
    {
      "epoch": 15.841525171283884,
      "grad_norm": 5.625,
      "learning_rate": 6.301061445879072e-06,
      "loss": 0.9421,
      "num_input_tokens_seen": 61709304,
      "step": 106360
    },
    {
      "epoch": 15.842269883824844,
      "grad_norm": 7.5,
      "learning_rate": 6.298904811475062e-06,
      "loss": 1.0671,
      "num_input_tokens_seen": 61712152,
      "step": 106365
    },
    {
      "epoch": 15.843014596365803,
      "grad_norm": 9.375,
      "learning_rate": 6.296748493007051e-06,
      "loss": 0.8315,
      "num_input_tokens_seen": 61714904,
      "step": 106370
    },
    {
      "epoch": 15.843759308906762,
      "grad_norm": 9.0625,
      "learning_rate": 6.294592490511456e-06,
      "loss": 0.8451,
      "num_input_tokens_seen": 61717784,
      "step": 106375
    },
    {
      "epoch": 15.84450402144772,
      "grad_norm": 12.625,
      "learning_rate": 6.292436804024715e-06,
      "loss": 1.1374,
      "num_input_tokens_seen": 61720568,
      "step": 106380
    },
    {
      "epoch": 15.845248733988681,
      "grad_norm": 9.9375,
      "learning_rate": 6.290281433583237e-06,
      "loss": 1.3273,
      "num_input_tokens_seen": 61723384,
      "step": 106385
    },
    {
      "epoch": 15.84599344652964,
      "grad_norm": 8.0,
      "learning_rate": 6.288126379223444e-06,
      "loss": 0.8791,
      "num_input_tokens_seen": 61726264,
      "step": 106390
    },
    {
      "epoch": 15.846738159070599,
      "grad_norm": 6.3125,
      "learning_rate": 6.285971640981731e-06,
      "loss": 1.0205,
      "num_input_tokens_seen": 61729144,
      "step": 106395
    },
    {
      "epoch": 15.847482871611557,
      "grad_norm": 10.5,
      "learning_rate": 6.283817218894514e-06,
      "loss": 0.9119,
      "num_input_tokens_seen": 61732088,
      "step": 106400
    },
    {
      "epoch": 15.848227584152518,
      "grad_norm": 11.0,
      "learning_rate": 6.281663112998174e-06,
      "loss": 0.9629,
      "num_input_tokens_seen": 61735288,
      "step": 106405
    },
    {
      "epoch": 15.848972296693477,
      "grad_norm": 7.4375,
      "learning_rate": 6.2795093233291195e-06,
      "loss": 0.9987,
      "num_input_tokens_seen": 61738488,
      "step": 106410
    },
    {
      "epoch": 15.849717009234435,
      "grad_norm": 5.9375,
      "learning_rate": 6.27735584992373e-06,
      "loss": 0.9063,
      "num_input_tokens_seen": 61741336,
      "step": 106415
    },
    {
      "epoch": 15.850461721775394,
      "grad_norm": 13.875,
      "learning_rate": 6.275202692818383e-06,
      "loss": 1.0121,
      "num_input_tokens_seen": 61744344,
      "step": 106420
    },
    {
      "epoch": 15.851206434316353,
      "grad_norm": 10.6875,
      "learning_rate": 6.2730498520494565e-06,
      "loss": 1.2598,
      "num_input_tokens_seen": 61747192,
      "step": 106425
    },
    {
      "epoch": 15.851951146857314,
      "grad_norm": 6.59375,
      "learning_rate": 6.27089732765331e-06,
      "loss": 0.9134,
      "num_input_tokens_seen": 61749816,
      "step": 106430
    },
    {
      "epoch": 15.852695859398272,
      "grad_norm": 8.5,
      "learning_rate": 6.2687451196663275e-06,
      "loss": 0.8155,
      "num_input_tokens_seen": 61752632,
      "step": 106435
    },
    {
      "epoch": 15.853440571939231,
      "grad_norm": 10.8125,
      "learning_rate": 6.266593228124851e-06,
      "loss": 0.8772,
      "num_input_tokens_seen": 61755512,
      "step": 106440
    },
    {
      "epoch": 15.85418528448019,
      "grad_norm": 11.5,
      "learning_rate": 6.264441653065248e-06,
      "loss": 1.494,
      "num_input_tokens_seen": 61758040,
      "step": 106445
    },
    {
      "epoch": 15.85492999702115,
      "grad_norm": 5.875,
      "learning_rate": 6.262290394523862e-06,
      "loss": 1.1407,
      "num_input_tokens_seen": 61760824,
      "step": 106450
    },
    {
      "epoch": 15.85567470956211,
      "grad_norm": 13.375,
      "learning_rate": 6.260139452537028e-06,
      "loss": 1.2354,
      "num_input_tokens_seen": 61763672,
      "step": 106455
    },
    {
      "epoch": 15.856419422103068,
      "grad_norm": 5.3125,
      "learning_rate": 6.257988827141101e-06,
      "loss": 0.917,
      "num_input_tokens_seen": 61766424,
      "step": 106460
    },
    {
      "epoch": 15.857164134644027,
      "grad_norm": 17.0,
      "learning_rate": 6.255838518372395e-06,
      "loss": 1.2196,
      "num_input_tokens_seen": 61769272,
      "step": 106465
    },
    {
      "epoch": 15.857908847184987,
      "grad_norm": 14.5,
      "learning_rate": 6.253688526267254e-06,
      "loss": 0.9399,
      "num_input_tokens_seen": 61772344,
      "step": 106470
    },
    {
      "epoch": 15.858653559725946,
      "grad_norm": 9.8125,
      "learning_rate": 6.251538850861985e-06,
      "loss": 0.9878,
      "num_input_tokens_seen": 61774968,
      "step": 106475
    },
    {
      "epoch": 15.859398272266905,
      "grad_norm": 5.40625,
      "learning_rate": 6.24938949219292e-06,
      "loss": 1.0624,
      "num_input_tokens_seen": 61778136,
      "step": 106480
    },
    {
      "epoch": 15.860142984807863,
      "grad_norm": 8.5,
      "learning_rate": 6.2472404502963625e-06,
      "loss": 1.0913,
      "num_input_tokens_seen": 61780952,
      "step": 106485
    },
    {
      "epoch": 15.860887697348824,
      "grad_norm": 5.125,
      "learning_rate": 6.245091725208616e-06,
      "loss": 0.8413,
      "num_input_tokens_seen": 61783992,
      "step": 106490
    },
    {
      "epoch": 15.861632409889783,
      "grad_norm": 2.796875,
      "learning_rate": 6.242943316965985e-06,
      "loss": 1.059,
      "num_input_tokens_seen": 61786744,
      "step": 106495
    },
    {
      "epoch": 15.862377122430741,
      "grad_norm": 10.375,
      "learning_rate": 6.2407952256047565e-06,
      "loss": 1.3778,
      "num_input_tokens_seen": 61789720,
      "step": 106500
    },
    {
      "epoch": 15.8631218349717,
      "grad_norm": 16.75,
      "learning_rate": 6.238647451161231e-06,
      "loss": 1.6521,
      "num_input_tokens_seen": 61792376,
      "step": 106505
    },
    {
      "epoch": 15.86386654751266,
      "grad_norm": 5.34375,
      "learning_rate": 6.2364999936716825e-06,
      "loss": 0.96,
      "num_input_tokens_seen": 61795192,
      "step": 106510
    },
    {
      "epoch": 15.86461126005362,
      "grad_norm": 9.625,
      "learning_rate": 6.234352853172404e-06,
      "loss": 1.1021,
      "num_input_tokens_seen": 61798168,
      "step": 106515
    },
    {
      "epoch": 15.865355972594578,
      "grad_norm": 5.40625,
      "learning_rate": 6.232206029699655e-06,
      "loss": 0.9845,
      "num_input_tokens_seen": 61801240,
      "step": 106520
    },
    {
      "epoch": 15.866100685135537,
      "grad_norm": 7.53125,
      "learning_rate": 6.230059523289716e-06,
      "loss": 1.2241,
      "num_input_tokens_seen": 61804184,
      "step": 106525
    },
    {
      "epoch": 15.866845397676498,
      "grad_norm": 8.9375,
      "learning_rate": 6.227913333978847e-06,
      "loss": 1.27,
      "num_input_tokens_seen": 61807352,
      "step": 106530
    },
    {
      "epoch": 15.867590110217456,
      "grad_norm": 18.5,
      "learning_rate": 6.225767461803295e-06,
      "loss": 1.4172,
      "num_input_tokens_seen": 61810488,
      "step": 106535
    },
    {
      "epoch": 15.868334822758415,
      "grad_norm": 4.59375,
      "learning_rate": 6.223621906799326e-06,
      "loss": 0.9473,
      "num_input_tokens_seen": 61813176,
      "step": 106540
    },
    {
      "epoch": 15.869079535299374,
      "grad_norm": 19.0,
      "learning_rate": 6.221476669003176e-06,
      "loss": 0.8431,
      "num_input_tokens_seen": 61816216,
      "step": 106545
    },
    {
      "epoch": 15.869824247840334,
      "grad_norm": 8.4375,
      "learning_rate": 6.219331748451096e-06,
      "loss": 1.1548,
      "num_input_tokens_seen": 61819032,
      "step": 106550
    },
    {
      "epoch": 15.870568960381293,
      "grad_norm": 6.5625,
      "learning_rate": 6.217187145179321e-06,
      "loss": 1.1796,
      "num_input_tokens_seen": 61822040,
      "step": 106555
    },
    {
      "epoch": 15.871313672922252,
      "grad_norm": 8.875,
      "learning_rate": 6.2150428592240795e-06,
      "loss": 1.1176,
      "num_input_tokens_seen": 61824920,
      "step": 106560
    },
    {
      "epoch": 15.87205838546321,
      "grad_norm": 12.125,
      "learning_rate": 6.212898890621588e-06,
      "loss": 0.902,
      "num_input_tokens_seen": 61827928,
      "step": 106565
    },
    {
      "epoch": 15.872803098004171,
      "grad_norm": 3.90625,
      "learning_rate": 6.210755239408083e-06,
      "loss": 0.987,
      "num_input_tokens_seen": 61830936,
      "step": 106570
    },
    {
      "epoch": 15.87354781054513,
      "grad_norm": 14.75,
      "learning_rate": 6.208611905619774e-06,
      "loss": 0.9353,
      "num_input_tokens_seen": 61833976,
      "step": 106575
    },
    {
      "epoch": 15.874292523086089,
      "grad_norm": 10.0,
      "learning_rate": 6.20646888929286e-06,
      "loss": 1.2208,
      "num_input_tokens_seen": 61836728,
      "step": 106580
    },
    {
      "epoch": 15.875037235627047,
      "grad_norm": 6.46875,
      "learning_rate": 6.204326190463558e-06,
      "loss": 1.0129,
      "num_input_tokens_seen": 61839608,
      "step": 106585
    },
    {
      "epoch": 15.875781948168008,
      "grad_norm": 13.3125,
      "learning_rate": 6.202183809168058e-06,
      "loss": 0.9375,
      "num_input_tokens_seen": 61842520,
      "step": 106590
    },
    {
      "epoch": 15.876526660708967,
      "grad_norm": 4.8125,
      "learning_rate": 6.200041745442561e-06,
      "loss": 0.7268,
      "num_input_tokens_seen": 61845496,
      "step": 106595
    },
    {
      "epoch": 15.877271373249926,
      "grad_norm": 14.25,
      "learning_rate": 6.197899999323245e-06,
      "loss": 1.2038,
      "num_input_tokens_seen": 61848408,
      "step": 106600
    },
    {
      "epoch": 15.878016085790884,
      "grad_norm": 6.6875,
      "learning_rate": 6.195758570846308e-06,
      "loss": 1.3378,
      "num_input_tokens_seen": 61851576,
      "step": 106605
    },
    {
      "epoch": 15.878760798331843,
      "grad_norm": 10.1875,
      "learning_rate": 6.193617460047918e-06,
      "loss": 1.4014,
      "num_input_tokens_seen": 61854424,
      "step": 106610
    },
    {
      "epoch": 15.879505510872804,
      "grad_norm": 8.875,
      "learning_rate": 6.191476666964238e-06,
      "loss": 0.939,
      "num_input_tokens_seen": 61857336,
      "step": 106615
    },
    {
      "epoch": 15.880250223413762,
      "grad_norm": 14.25,
      "learning_rate": 6.189336191631451e-06,
      "loss": 0.9983,
      "num_input_tokens_seen": 61860024,
      "step": 106620
    },
    {
      "epoch": 15.880994935954721,
      "grad_norm": 8.75,
      "learning_rate": 6.1871960340857126e-06,
      "loss": 1.1763,
      "num_input_tokens_seen": 61863032,
      "step": 106625
    },
    {
      "epoch": 15.88173964849568,
      "grad_norm": 7.96875,
      "learning_rate": 6.185056194363176e-06,
      "loss": 1.3332,
      "num_input_tokens_seen": 61865880,
      "step": 106630
    },
    {
      "epoch": 15.88248436103664,
      "grad_norm": 14.375,
      "learning_rate": 6.182916672499983e-06,
      "loss": 1.3231,
      "num_input_tokens_seen": 61868824,
      "step": 106635
    },
    {
      "epoch": 15.8832290735776,
      "grad_norm": 13.6875,
      "learning_rate": 6.180777468532298e-06,
      "loss": 0.9486,
      "num_input_tokens_seen": 61871704,
      "step": 106640
    },
    {
      "epoch": 15.883973786118558,
      "grad_norm": 9.9375,
      "learning_rate": 6.1786385824962436e-06,
      "loss": 0.9378,
      "num_input_tokens_seen": 61874616,
      "step": 106645
    },
    {
      "epoch": 15.884718498659517,
      "grad_norm": 13.625,
      "learning_rate": 6.176500014427966e-06,
      "loss": 1.458,
      "num_input_tokens_seen": 61877464,
      "step": 106650
    },
    {
      "epoch": 15.885463211200477,
      "grad_norm": 5.875,
      "learning_rate": 6.17436176436359e-06,
      "loss": 1.1251,
      "num_input_tokens_seen": 61880440,
      "step": 106655
    },
    {
      "epoch": 15.886207923741436,
      "grad_norm": 7.78125,
      "learning_rate": 6.1722238323392325e-06,
      "loss": 0.8481,
      "num_input_tokens_seen": 61883384,
      "step": 106660
    },
    {
      "epoch": 15.886952636282395,
      "grad_norm": 12.3125,
      "learning_rate": 6.1700862183910245e-06,
      "loss": 0.8966,
      "num_input_tokens_seen": 61886232,
      "step": 106665
    },
    {
      "epoch": 15.887697348823353,
      "grad_norm": 13.125,
      "learning_rate": 6.167948922555064e-06,
      "loss": 1.2943,
      "num_input_tokens_seen": 61888824,
      "step": 106670
    },
    {
      "epoch": 15.888442061364314,
      "grad_norm": 5.65625,
      "learning_rate": 6.165811944867475e-06,
      "loss": 0.9859,
      "num_input_tokens_seen": 61891416,
      "step": 106675
    },
    {
      "epoch": 15.889186773905273,
      "grad_norm": 10.6875,
      "learning_rate": 6.163675285364348e-06,
      "loss": 1.0118,
      "num_input_tokens_seen": 61894360,
      "step": 106680
    },
    {
      "epoch": 15.889931486446232,
      "grad_norm": 7.34375,
      "learning_rate": 6.161538944081779e-06,
      "loss": 1.0064,
      "num_input_tokens_seen": 61897048,
      "step": 106685
    },
    {
      "epoch": 15.89067619898719,
      "grad_norm": 15.375,
      "learning_rate": 6.1594029210558675e-06,
      "loss": 0.913,
      "num_input_tokens_seen": 61899736,
      "step": 106690
    },
    {
      "epoch": 15.89142091152815,
      "grad_norm": 7.8125,
      "learning_rate": 6.157267216322696e-06,
      "loss": 1.2398,
      "num_input_tokens_seen": 61902552,
      "step": 106695
    },
    {
      "epoch": 15.89216562406911,
      "grad_norm": 9.625,
      "learning_rate": 6.155131829918345e-06,
      "loss": 0.9647,
      "num_input_tokens_seen": 61905464,
      "step": 106700
    },
    {
      "epoch": 15.892910336610068,
      "grad_norm": 7.59375,
      "learning_rate": 6.1529967618788795e-06,
      "loss": 1.1861,
      "num_input_tokens_seen": 61908344,
      "step": 106705
    },
    {
      "epoch": 15.893655049151027,
      "grad_norm": 12.625,
      "learning_rate": 6.1508620122403885e-06,
      "loss": 0.9439,
      "num_input_tokens_seen": 61910968,
      "step": 106710
    },
    {
      "epoch": 15.894399761691988,
      "grad_norm": 12.3125,
      "learning_rate": 6.148727581038915e-06,
      "loss": 0.8747,
      "num_input_tokens_seen": 61913944,
      "step": 106715
    },
    {
      "epoch": 15.895144474232946,
      "grad_norm": 18.875,
      "learning_rate": 6.146593468310541e-06,
      "loss": 0.9099,
      "num_input_tokens_seen": 61916504,
      "step": 106720
    },
    {
      "epoch": 15.895889186773905,
      "grad_norm": 15.0625,
      "learning_rate": 6.144459674091299e-06,
      "loss": 1.3008,
      "num_input_tokens_seen": 61919160,
      "step": 106725
    },
    {
      "epoch": 15.896633899314864,
      "grad_norm": 10.5,
      "learning_rate": 6.1423261984172535e-06,
      "loss": 1.198,
      "num_input_tokens_seen": 61922168,
      "step": 106730
    },
    {
      "epoch": 15.897378611855824,
      "grad_norm": 14.1875,
      "learning_rate": 6.140193041324444e-06,
      "loss": 1.0841,
      "num_input_tokens_seen": 61925176,
      "step": 106735
    },
    {
      "epoch": 15.898123324396783,
      "grad_norm": 12.25,
      "learning_rate": 6.138060202848894e-06,
      "loss": 0.9577,
      "num_input_tokens_seen": 61927864,
      "step": 106740
    },
    {
      "epoch": 15.898868036937742,
      "grad_norm": 12.25,
      "learning_rate": 6.135927683026654e-06,
      "loss": 1.1116,
      "num_input_tokens_seen": 61930648,
      "step": 106745
    },
    {
      "epoch": 15.8996127494787,
      "grad_norm": 9.75,
      "learning_rate": 6.133795481893745e-06,
      "loss": 1.078,
      "num_input_tokens_seen": 61933464,
      "step": 106750
    },
    {
      "epoch": 15.90035746201966,
      "grad_norm": 13.5625,
      "learning_rate": 6.1316635994861875e-06,
      "loss": 1.4154,
      "num_input_tokens_seen": 61936568,
      "step": 106755
    },
    {
      "epoch": 15.90110217456062,
      "grad_norm": 12.6875,
      "learning_rate": 6.129532035839985e-06,
      "loss": 1.1006,
      "num_input_tokens_seen": 61939320,
      "step": 106760
    },
    {
      "epoch": 15.901846887101579,
      "grad_norm": 9.4375,
      "learning_rate": 6.12740079099117e-06,
      "loss": 0.737,
      "num_input_tokens_seen": 61942072,
      "step": 106765
    },
    {
      "epoch": 15.902591599642538,
      "grad_norm": 8.3125,
      "learning_rate": 6.12526986497573e-06,
      "loss": 1.3483,
      "num_input_tokens_seen": 61944952,
      "step": 106770
    },
    {
      "epoch": 15.903336312183498,
      "grad_norm": 17.0,
      "learning_rate": 6.12313925782968e-06,
      "loss": 0.9964,
      "num_input_tokens_seen": 61947640,
      "step": 106775
    },
    {
      "epoch": 15.904081024724457,
      "grad_norm": 8.1875,
      "learning_rate": 6.1210089695890065e-06,
      "loss": 1.3659,
      "num_input_tokens_seen": 61950584,
      "step": 106780
    },
    {
      "epoch": 15.904825737265416,
      "grad_norm": 9.4375,
      "learning_rate": 6.11887900028969e-06,
      "loss": 1.1257,
      "num_input_tokens_seen": 61953464,
      "step": 106785
    },
    {
      "epoch": 15.905570449806374,
      "grad_norm": 6.40625,
      "learning_rate": 6.116749349967732e-06,
      "loss": 1.0302,
      "num_input_tokens_seen": 61956728,
      "step": 106790
    },
    {
      "epoch": 15.906315162347333,
      "grad_norm": 7.625,
      "learning_rate": 6.114620018659093e-06,
      "loss": 1.1524,
      "num_input_tokens_seen": 61959960,
      "step": 106795
    },
    {
      "epoch": 15.907059874888294,
      "grad_norm": 17.5,
      "learning_rate": 6.112491006399762e-06,
      "loss": 1.5354,
      "num_input_tokens_seen": 61962872,
      "step": 106800
    },
    {
      "epoch": 15.907804587429252,
      "grad_norm": 15.4375,
      "learning_rate": 6.110362313225693e-06,
      "loss": 1.1408,
      "num_input_tokens_seen": 61965560,
      "step": 106805
    },
    {
      "epoch": 15.908549299970211,
      "grad_norm": 13.0,
      "learning_rate": 6.108233939172858e-06,
      "loss": 1.4527,
      "num_input_tokens_seen": 61968664,
      "step": 106810
    },
    {
      "epoch": 15.90929401251117,
      "grad_norm": 11.75,
      "learning_rate": 6.106105884277213e-06,
      "loss": 1.0089,
      "num_input_tokens_seen": 61971416,
      "step": 106815
    },
    {
      "epoch": 15.91003872505213,
      "grad_norm": 12.75,
      "learning_rate": 6.1039781485747045e-06,
      "loss": 1.2044,
      "num_input_tokens_seen": 61974264,
      "step": 106820
    },
    {
      "epoch": 15.91078343759309,
      "grad_norm": 9.0,
      "learning_rate": 6.101850732101283e-06,
      "loss": 0.8848,
      "num_input_tokens_seen": 61977048,
      "step": 106825
    },
    {
      "epoch": 15.911528150134048,
      "grad_norm": 7.875,
      "learning_rate": 6.099723634892876e-06,
      "loss": 1.1334,
      "num_input_tokens_seen": 61979928,
      "step": 106830
    },
    {
      "epoch": 15.912272862675007,
      "grad_norm": 9.875,
      "learning_rate": 6.097596856985435e-06,
      "loss": 1.4423,
      "num_input_tokens_seen": 61983160,
      "step": 106835
    },
    {
      "epoch": 15.913017575215967,
      "grad_norm": 8.8125,
      "learning_rate": 6.095470398414879e-06,
      "loss": 1.3282,
      "num_input_tokens_seen": 61985976,
      "step": 106840
    },
    {
      "epoch": 15.913762287756926,
      "grad_norm": 13.25,
      "learning_rate": 6.093344259217143e-06,
      "loss": 1.2845,
      "num_input_tokens_seen": 61988824,
      "step": 106845
    },
    {
      "epoch": 15.914507000297885,
      "grad_norm": 5.53125,
      "learning_rate": 6.091218439428134e-06,
      "loss": 0.8756,
      "num_input_tokens_seen": 61991704,
      "step": 106850
    },
    {
      "epoch": 15.915251712838844,
      "grad_norm": 11.3125,
      "learning_rate": 6.089092939083779e-06,
      "loss": 1.1071,
      "num_input_tokens_seen": 61994680,
      "step": 106855
    },
    {
      "epoch": 15.915996425379804,
      "grad_norm": 13.5625,
      "learning_rate": 6.0869677582199775e-06,
      "loss": 0.8929,
      "num_input_tokens_seen": 61997944,
      "step": 106860
    },
    {
      "epoch": 15.916741137920763,
      "grad_norm": 7.625,
      "learning_rate": 6.084842896872625e-06,
      "loss": 1.2324,
      "num_input_tokens_seen": 62000504,
      "step": 106865
    },
    {
      "epoch": 15.917485850461722,
      "grad_norm": 12.4375,
      "learning_rate": 6.082718355077635e-06,
      "loss": 0.98,
      "num_input_tokens_seen": 62003320,
      "step": 106870
    },
    {
      "epoch": 15.91823056300268,
      "grad_norm": 9.375,
      "learning_rate": 6.080594132870885e-06,
      "loss": 1.0504,
      "num_input_tokens_seen": 62006232,
      "step": 106875
    },
    {
      "epoch": 15.918975275543641,
      "grad_norm": 7.0625,
      "learning_rate": 6.078470230288274e-06,
      "loss": 1.0181,
      "num_input_tokens_seen": 62009112,
      "step": 106880
    },
    {
      "epoch": 15.9197199880846,
      "grad_norm": 12.375,
      "learning_rate": 6.07634664736568e-06,
      "loss": 1.1785,
      "num_input_tokens_seen": 62011992,
      "step": 106885
    },
    {
      "epoch": 15.920464700625558,
      "grad_norm": 10.3125,
      "learning_rate": 6.074223384138977e-06,
      "loss": 1.1144,
      "num_input_tokens_seen": 62015160,
      "step": 106890
    },
    {
      "epoch": 15.921209413166517,
      "grad_norm": 12.8125,
      "learning_rate": 6.072100440644033e-06,
      "loss": 1.106,
      "num_input_tokens_seen": 62017848,
      "step": 106895
    },
    {
      "epoch": 15.921954125707478,
      "grad_norm": 17.0,
      "learning_rate": 6.069977816916705e-06,
      "loss": 1.447,
      "num_input_tokens_seen": 62020824,
      "step": 106900
    },
    {
      "epoch": 15.922698838248436,
      "grad_norm": 12.125,
      "learning_rate": 6.067855512992873e-06,
      "loss": 0.8031,
      "num_input_tokens_seen": 62023800,
      "step": 106905
    },
    {
      "epoch": 15.923443550789395,
      "grad_norm": 9.4375,
      "learning_rate": 6.065733528908371e-06,
      "loss": 1.0986,
      "num_input_tokens_seen": 62026712,
      "step": 106910
    },
    {
      "epoch": 15.924188263330354,
      "grad_norm": 17.875,
      "learning_rate": 6.0636118646990644e-06,
      "loss": 0.6508,
      "num_input_tokens_seen": 62029560,
      "step": 106915
    },
    {
      "epoch": 15.924932975871315,
      "grad_norm": 10.0625,
      "learning_rate": 6.061490520400784e-06,
      "loss": 1.1337,
      "num_input_tokens_seen": 62032152,
      "step": 106920
    },
    {
      "epoch": 15.925677688412273,
      "grad_norm": 14.75,
      "learning_rate": 6.059369496049377e-06,
      "loss": 0.9375,
      "num_input_tokens_seen": 62034936,
      "step": 106925
    },
    {
      "epoch": 15.926422400953232,
      "grad_norm": 15.4375,
      "learning_rate": 6.05724879168067e-06,
      "loss": 1.0751,
      "num_input_tokens_seen": 62037816,
      "step": 106930
    },
    {
      "epoch": 15.92716711349419,
      "grad_norm": 5.0,
      "learning_rate": 6.055128407330493e-06,
      "loss": 0.8207,
      "num_input_tokens_seen": 62040632,
      "step": 106935
    },
    {
      "epoch": 15.92791182603515,
      "grad_norm": 7.96875,
      "learning_rate": 6.053008343034671e-06,
      "loss": 1.2603,
      "num_input_tokens_seen": 62043736,
      "step": 106940
    },
    {
      "epoch": 15.92865653857611,
      "grad_norm": 8.4375,
      "learning_rate": 6.0508885988290075e-06,
      "loss": 1.0694,
      "num_input_tokens_seen": 62046456,
      "step": 106945
    },
    {
      "epoch": 15.929401251117069,
      "grad_norm": 3.703125,
      "learning_rate": 6.04876917474933e-06,
      "loss": 0.6847,
      "num_input_tokens_seen": 62049304,
      "step": 106950
    },
    {
      "epoch": 15.930145963658028,
      "grad_norm": 11.3125,
      "learning_rate": 6.046650070831436e-06,
      "loss": 0.9906,
      "num_input_tokens_seen": 62052056,
      "step": 106955
    },
    {
      "epoch": 15.930890676198988,
      "grad_norm": 14.1875,
      "learning_rate": 6.044531287111124e-06,
      "loss": 1.0563,
      "num_input_tokens_seen": 62054840,
      "step": 106960
    },
    {
      "epoch": 15.931635388739947,
      "grad_norm": 13.3125,
      "learning_rate": 6.042412823624186e-06,
      "loss": 0.9715,
      "num_input_tokens_seen": 62057784,
      "step": 106965
    },
    {
      "epoch": 15.932380101280906,
      "grad_norm": 8.6875,
      "learning_rate": 6.040294680406422e-06,
      "loss": 0.9891,
      "num_input_tokens_seen": 62060920,
      "step": 106970
    },
    {
      "epoch": 15.933124813821864,
      "grad_norm": 9.125,
      "learning_rate": 6.0381768574936104e-06,
      "loss": 1.0607,
      "num_input_tokens_seen": 62063608,
      "step": 106975
    },
    {
      "epoch": 15.933869526362823,
      "grad_norm": 7.6875,
      "learning_rate": 6.03605935492152e-06,
      "loss": 0.898,
      "num_input_tokens_seen": 62066552,
      "step": 106980
    },
    {
      "epoch": 15.934614238903784,
      "grad_norm": 7.125,
      "learning_rate": 6.0339421727259395e-06,
      "loss": 1.3337,
      "num_input_tokens_seen": 62069720,
      "step": 106985
    },
    {
      "epoch": 15.935358951444742,
      "grad_norm": 16.75,
      "learning_rate": 6.031825310942624e-06,
      "loss": 1.3292,
      "num_input_tokens_seen": 62072632,
      "step": 106990
    },
    {
      "epoch": 15.936103663985701,
      "grad_norm": 11.375,
      "learning_rate": 6.029708769607348e-06,
      "loss": 0.9175,
      "num_input_tokens_seen": 62075576,
      "step": 106995
    },
    {
      "epoch": 15.93684837652666,
      "grad_norm": 12.25,
      "learning_rate": 6.027592548755853e-06,
      "loss": 1.0025,
      "num_input_tokens_seen": 62078328,
      "step": 107000
    },
    {
      "epoch": 15.93759308906762,
      "grad_norm": 11.625,
      "learning_rate": 6.025476648423908e-06,
      "loss": 1.2312,
      "num_input_tokens_seen": 62081208,
      "step": 107005
    },
    {
      "epoch": 15.93833780160858,
      "grad_norm": 13.0625,
      "learning_rate": 6.023361068647251e-06,
      "loss": 1.1425,
      "num_input_tokens_seen": 62084248,
      "step": 107010
    },
    {
      "epoch": 15.939082514149538,
      "grad_norm": 7.53125,
      "learning_rate": 6.021245809461615e-06,
      "loss": 0.8739,
      "num_input_tokens_seen": 62087096,
      "step": 107015
    },
    {
      "epoch": 15.939827226690497,
      "grad_norm": 9.125,
      "learning_rate": 6.0191308709027475e-06,
      "loss": 1.0352,
      "num_input_tokens_seen": 62090232,
      "step": 107020
    },
    {
      "epoch": 15.940571939231457,
      "grad_norm": 17.125,
      "learning_rate": 6.017016253006372e-06,
      "loss": 1.0261,
      "num_input_tokens_seen": 62093240,
      "step": 107025
    },
    {
      "epoch": 15.941316651772416,
      "grad_norm": 7.0625,
      "learning_rate": 6.014901955808216e-06,
      "loss": 1.4209,
      "num_input_tokens_seen": 62095896,
      "step": 107030
    },
    {
      "epoch": 15.942061364313375,
      "grad_norm": 10.4375,
      "learning_rate": 6.012787979343987e-06,
      "loss": 0.956,
      "num_input_tokens_seen": 62098776,
      "step": 107035
    },
    {
      "epoch": 15.942806076854334,
      "grad_norm": 8.8125,
      "learning_rate": 6.010674323649415e-06,
      "loss": 0.7434,
      "num_input_tokens_seen": 62101784,
      "step": 107040
    },
    {
      "epoch": 15.943550789395294,
      "grad_norm": 12.3125,
      "learning_rate": 6.0085609887601925e-06,
      "loss": 1.1733,
      "num_input_tokens_seen": 62104760,
      "step": 107045
    },
    {
      "epoch": 15.944295501936253,
      "grad_norm": 13.125,
      "learning_rate": 6.0064479747120375e-06,
      "loss": 0.7034,
      "num_input_tokens_seen": 62107928,
      "step": 107050
    },
    {
      "epoch": 15.945040214477212,
      "grad_norm": 23.375,
      "learning_rate": 6.004335281540641e-06,
      "loss": 1.212,
      "num_input_tokens_seen": 62110520,
      "step": 107055
    },
    {
      "epoch": 15.94578492701817,
      "grad_norm": 13.875,
      "learning_rate": 6.002222909281685e-06,
      "loss": 1.1398,
      "num_input_tokens_seen": 62113656,
      "step": 107060
    },
    {
      "epoch": 15.946529639559131,
      "grad_norm": 10.0625,
      "learning_rate": 6.000110857970873e-06,
      "loss": 1.0909,
      "num_input_tokens_seen": 62116952,
      "step": 107065
    },
    {
      "epoch": 15.94727435210009,
      "grad_norm": 15.6875,
      "learning_rate": 5.9979991276438695e-06,
      "loss": 1.3165,
      "num_input_tokens_seen": 62119640,
      "step": 107070
    },
    {
      "epoch": 15.948019064641048,
      "grad_norm": 12.375,
      "learning_rate": 5.995887718336363e-06,
      "loss": 1.1089,
      "num_input_tokens_seen": 62122296,
      "step": 107075
    },
    {
      "epoch": 15.948763777182007,
      "grad_norm": 8.125,
      "learning_rate": 5.993776630084022e-06,
      "loss": 1.0549,
      "num_input_tokens_seen": 62125144,
      "step": 107080
    },
    {
      "epoch": 15.949508489722968,
      "grad_norm": 6.96875,
      "learning_rate": 5.991665862922505e-06,
      "loss": 0.8078,
      "num_input_tokens_seen": 62128088,
      "step": 107085
    },
    {
      "epoch": 15.950253202263927,
      "grad_norm": 15.125,
      "learning_rate": 5.989555416887469e-06,
      "loss": 1.1361,
      "num_input_tokens_seen": 62130936,
      "step": 107090
    },
    {
      "epoch": 15.950997914804885,
      "grad_norm": 6.59375,
      "learning_rate": 5.987445292014579e-06,
      "loss": 1.197,
      "num_input_tokens_seen": 62133848,
      "step": 107095
    },
    {
      "epoch": 15.951742627345844,
      "grad_norm": 5.8125,
      "learning_rate": 5.985335488339477e-06,
      "loss": 0.8373,
      "num_input_tokens_seen": 62136792,
      "step": 107100
    },
    {
      "epoch": 15.952487339886805,
      "grad_norm": 9.5625,
      "learning_rate": 5.983226005897799e-06,
      "loss": 1.4325,
      "num_input_tokens_seen": 62139640,
      "step": 107105
    },
    {
      "epoch": 15.953232052427763,
      "grad_norm": 7.8125,
      "learning_rate": 5.981116844725199e-06,
      "loss": 0.7935,
      "num_input_tokens_seen": 62142520,
      "step": 107110
    },
    {
      "epoch": 15.953976764968722,
      "grad_norm": 8.1875,
      "learning_rate": 5.979008004857292e-06,
      "loss": 0.7815,
      "num_input_tokens_seen": 62145240,
      "step": 107115
    },
    {
      "epoch": 15.95472147750968,
      "grad_norm": 10.25,
      "learning_rate": 5.976899486329718e-06,
      "loss": 1.1283,
      "num_input_tokens_seen": 62148312,
      "step": 107120
    },
    {
      "epoch": 15.95546619005064,
      "grad_norm": 13.0,
      "learning_rate": 5.974791289178089e-06,
      "loss": 1.0476,
      "num_input_tokens_seen": 62151064,
      "step": 107125
    },
    {
      "epoch": 15.9562109025916,
      "grad_norm": 14.625,
      "learning_rate": 5.972683413438029e-06,
      "loss": 1.1731,
      "num_input_tokens_seen": 62153848,
      "step": 107130
    },
    {
      "epoch": 15.956955615132559,
      "grad_norm": 7.125,
      "learning_rate": 5.970575859145144e-06,
      "loss": 1.3988,
      "num_input_tokens_seen": 62156760,
      "step": 107135
    },
    {
      "epoch": 15.957700327673518,
      "grad_norm": 11.5,
      "learning_rate": 5.968468626335033e-06,
      "loss": 1.0447,
      "num_input_tokens_seen": 62159544,
      "step": 107140
    },
    {
      "epoch": 15.958445040214476,
      "grad_norm": 8.6875,
      "learning_rate": 5.966361715043312e-06,
      "loss": 0.7958,
      "num_input_tokens_seen": 62162456,
      "step": 107145
    },
    {
      "epoch": 15.959189752755437,
      "grad_norm": 12.5,
      "learning_rate": 5.96425512530556e-06,
      "loss": 1.2783,
      "num_input_tokens_seen": 62165400,
      "step": 107150
    },
    {
      "epoch": 15.959934465296396,
      "grad_norm": 8.1875,
      "learning_rate": 5.962148857157373e-06,
      "loss": 1.3189,
      "num_input_tokens_seen": 62168152,
      "step": 107155
    },
    {
      "epoch": 15.960679177837354,
      "grad_norm": 7.53125,
      "learning_rate": 5.960042910634325e-06,
      "loss": 1.1323,
      "num_input_tokens_seen": 62170840,
      "step": 107160
    },
    {
      "epoch": 15.961423890378313,
      "grad_norm": 12.6875,
      "learning_rate": 5.9579372857720085e-06,
      "loss": 0.9059,
      "num_input_tokens_seen": 62173592,
      "step": 107165
    },
    {
      "epoch": 15.962168602919274,
      "grad_norm": 5.09375,
      "learning_rate": 5.9558319826059775e-06,
      "loss": 1.1306,
      "num_input_tokens_seen": 62176568,
      "step": 107170
    },
    {
      "epoch": 15.962913315460233,
      "grad_norm": 9.625,
      "learning_rate": 5.953727001171819e-06,
      "loss": 1.1671,
      "num_input_tokens_seen": 62179448,
      "step": 107175
    },
    {
      "epoch": 15.963658028001191,
      "grad_norm": 9.8125,
      "learning_rate": 5.951622341505086e-06,
      "loss": 0.9795,
      "num_input_tokens_seen": 62182424,
      "step": 107180
    },
    {
      "epoch": 15.96440274054215,
      "grad_norm": 14.5625,
      "learning_rate": 5.949518003641325e-06,
      "loss": 1.1751,
      "num_input_tokens_seen": 62185464,
      "step": 107185
    },
    {
      "epoch": 15.96514745308311,
      "grad_norm": 9.4375,
      "learning_rate": 5.947413987616105e-06,
      "loss": 1.2338,
      "num_input_tokens_seen": 62188248,
      "step": 107190
    },
    {
      "epoch": 15.96589216562407,
      "grad_norm": 6.40625,
      "learning_rate": 5.94531029346495e-06,
      "loss": 1.1076,
      "num_input_tokens_seen": 62191032,
      "step": 107195
    },
    {
      "epoch": 15.966636878165028,
      "grad_norm": 6.0,
      "learning_rate": 5.943206921223421e-06,
      "loss": 0.9581,
      "num_input_tokens_seen": 62193944,
      "step": 107200
    },
    {
      "epoch": 15.967381590705987,
      "grad_norm": 15.0,
      "learning_rate": 5.9411038709270365e-06,
      "loss": 1.1845,
      "num_input_tokens_seen": 62197048,
      "step": 107205
    },
    {
      "epoch": 15.968126303246947,
      "grad_norm": 8.5,
      "learning_rate": 5.939001142611336e-06,
      "loss": 0.9418,
      "num_input_tokens_seen": 62199928,
      "step": 107210
    },
    {
      "epoch": 15.968871015787906,
      "grad_norm": 7.4375,
      "learning_rate": 5.9368987363118415e-06,
      "loss": 1.3816,
      "num_input_tokens_seen": 62202840,
      "step": 107215
    },
    {
      "epoch": 15.969615728328865,
      "grad_norm": 7.59375,
      "learning_rate": 5.934796652064065e-06,
      "loss": 1.0524,
      "num_input_tokens_seen": 62205720,
      "step": 107220
    },
    {
      "epoch": 15.970360440869824,
      "grad_norm": 9.6875,
      "learning_rate": 5.932694889903523e-06,
      "loss": 0.962,
      "num_input_tokens_seen": 62208696,
      "step": 107225
    },
    {
      "epoch": 15.971105153410784,
      "grad_norm": 13.0,
      "learning_rate": 5.930593449865715e-06,
      "loss": 1.4525,
      "num_input_tokens_seen": 62211544,
      "step": 107230
    },
    {
      "epoch": 15.971849865951743,
      "grad_norm": 9.5625,
      "learning_rate": 5.928492331986155e-06,
      "loss": 1.3341,
      "num_input_tokens_seen": 62214232,
      "step": 107235
    },
    {
      "epoch": 15.972594578492702,
      "grad_norm": 13.3125,
      "learning_rate": 5.9263915363003294e-06,
      "loss": 0.9088,
      "num_input_tokens_seen": 62217624,
      "step": 107240
    },
    {
      "epoch": 15.97333929103366,
      "grad_norm": 6.78125,
      "learning_rate": 5.924291062843737e-06,
      "loss": 1.1768,
      "num_input_tokens_seen": 62220632,
      "step": 107245
    },
    {
      "epoch": 15.974084003574621,
      "grad_norm": 9.0,
      "learning_rate": 5.922190911651857e-06,
      "loss": 0.8219,
      "num_input_tokens_seen": 62223576,
      "step": 107250
    },
    {
      "epoch": 15.97482871611558,
      "grad_norm": 11.4375,
      "learning_rate": 5.920091082760174e-06,
      "loss": 0.9818,
      "num_input_tokens_seen": 62226424,
      "step": 107255
    },
    {
      "epoch": 15.975573428656539,
      "grad_norm": 19.625,
      "learning_rate": 5.917991576204163e-06,
      "loss": 1.1809,
      "num_input_tokens_seen": 62229240,
      "step": 107260
    },
    {
      "epoch": 15.976318141197497,
      "grad_norm": 13.9375,
      "learning_rate": 5.915892392019282e-06,
      "loss": 1.3663,
      "num_input_tokens_seen": 62232344,
      "step": 107265
    },
    {
      "epoch": 15.977062853738456,
      "grad_norm": 4.5,
      "learning_rate": 5.913793530241011e-06,
      "loss": 0.9915,
      "num_input_tokens_seen": 62235736,
      "step": 107270
    },
    {
      "epoch": 15.977807566279417,
      "grad_norm": 5.09375,
      "learning_rate": 5.91169499090479e-06,
      "loss": 0.8491,
      "num_input_tokens_seen": 62238488,
      "step": 107275
    },
    {
      "epoch": 15.978552278820375,
      "grad_norm": 7.53125,
      "learning_rate": 5.909596774046092e-06,
      "loss": 1.1655,
      "num_input_tokens_seen": 62241368,
      "step": 107280
    },
    {
      "epoch": 15.979296991361334,
      "grad_norm": 12.3125,
      "learning_rate": 5.907498879700352e-06,
      "loss": 0.6496,
      "num_input_tokens_seen": 62244280,
      "step": 107285
    },
    {
      "epoch": 15.980041703902295,
      "grad_norm": 11.1875,
      "learning_rate": 5.905401307903013e-06,
      "loss": 0.946,
      "num_input_tokens_seen": 62246968,
      "step": 107290
    },
    {
      "epoch": 15.980786416443253,
      "grad_norm": 9.8125,
      "learning_rate": 5.903304058689507e-06,
      "loss": 1.021,
      "num_input_tokens_seen": 62249720,
      "step": 107295
    },
    {
      "epoch": 15.981531128984212,
      "grad_norm": 12.6875,
      "learning_rate": 5.901207132095276e-06,
      "loss": 1.0519,
      "num_input_tokens_seen": 62252792,
      "step": 107300
    },
    {
      "epoch": 15.982275841525171,
      "grad_norm": 10.3125,
      "learning_rate": 5.899110528155741e-06,
      "loss": 0.8949,
      "num_input_tokens_seen": 62255416,
      "step": 107305
    },
    {
      "epoch": 15.98302055406613,
      "grad_norm": 14.875,
      "learning_rate": 5.897014246906312e-06,
      "loss": 1.1159,
      "num_input_tokens_seen": 62258200,
      "step": 107310
    },
    {
      "epoch": 15.98376526660709,
      "grad_norm": 6.6875,
      "learning_rate": 5.894918288382417e-06,
      "loss": 1.0007,
      "num_input_tokens_seen": 62260920,
      "step": 107315
    },
    {
      "epoch": 15.984509979148049,
      "grad_norm": 11.9375,
      "learning_rate": 5.8928226526194565e-06,
      "loss": 1.3575,
      "num_input_tokens_seen": 62263960,
      "step": 107320
    },
    {
      "epoch": 15.985254691689008,
      "grad_norm": 9.4375,
      "learning_rate": 5.890727339652843e-06,
      "loss": 1.291,
      "num_input_tokens_seen": 62266552,
      "step": 107325
    },
    {
      "epoch": 15.985999404229966,
      "grad_norm": 9.0,
      "learning_rate": 5.888632349517962e-06,
      "loss": 1.1267,
      "num_input_tokens_seen": 62269624,
      "step": 107330
    },
    {
      "epoch": 15.986744116770927,
      "grad_norm": 11.5625,
      "learning_rate": 5.886537682250221e-06,
      "loss": 1.1853,
      "num_input_tokens_seen": 62272536,
      "step": 107335
    },
    {
      "epoch": 15.987488829311886,
      "grad_norm": 6.0625,
      "learning_rate": 5.8844433378849986e-06,
      "loss": 1.1211,
      "num_input_tokens_seen": 62275384,
      "step": 107340
    },
    {
      "epoch": 15.988233541852845,
      "grad_norm": 8.5,
      "learning_rate": 5.882349316457672e-06,
      "loss": 1.0677,
      "num_input_tokens_seen": 62278136,
      "step": 107345
    },
    {
      "epoch": 15.988978254393803,
      "grad_norm": 11.25,
      "learning_rate": 5.88025561800363e-06,
      "loss": 1.1224,
      "num_input_tokens_seen": 62280792,
      "step": 107350
    },
    {
      "epoch": 15.989722966934764,
      "grad_norm": 12.375,
      "learning_rate": 5.878162242558239e-06,
      "loss": 0.9673,
      "num_input_tokens_seen": 62283512,
      "step": 107355
    },
    {
      "epoch": 15.990467679475723,
      "grad_norm": 6.28125,
      "learning_rate": 5.87606919015686e-06,
      "loss": 1.1677,
      "num_input_tokens_seen": 62286360,
      "step": 107360
    },
    {
      "epoch": 15.991212392016681,
      "grad_norm": 15.125,
      "learning_rate": 5.873976460834848e-06,
      "loss": 0.9268,
      "num_input_tokens_seen": 62289144,
      "step": 107365
    },
    {
      "epoch": 15.99195710455764,
      "grad_norm": 16.375,
      "learning_rate": 5.871884054627571e-06,
      "loss": 1.284,
      "num_input_tokens_seen": 62291736,
      "step": 107370
    },
    {
      "epoch": 15.9927018170986,
      "grad_norm": 6.1875,
      "learning_rate": 5.869791971570368e-06,
      "loss": 1.0399,
      "num_input_tokens_seen": 62294904,
      "step": 107375
    },
    {
      "epoch": 15.99344652963956,
      "grad_norm": 3.625,
      "learning_rate": 5.867700211698593e-06,
      "loss": 0.7465,
      "num_input_tokens_seen": 62297784,
      "step": 107380
    },
    {
      "epoch": 15.994191242180518,
      "grad_norm": 6.875,
      "learning_rate": 5.8656087750475765e-06,
      "loss": 1.0533,
      "num_input_tokens_seen": 62300504,
      "step": 107385
    },
    {
      "epoch": 15.994935954721477,
      "grad_norm": 11.8125,
      "learning_rate": 5.863517661652645e-06,
      "loss": 1.0817,
      "num_input_tokens_seen": 62303768,
      "step": 107390
    },
    {
      "epoch": 15.995680667262437,
      "grad_norm": 8.9375,
      "learning_rate": 5.861426871549142e-06,
      "loss": 0.9659,
      "num_input_tokens_seen": 62307032,
      "step": 107395
    },
    {
      "epoch": 15.996425379803396,
      "grad_norm": 9.125,
      "learning_rate": 5.859336404772372e-06,
      "loss": 1.2236,
      "num_input_tokens_seen": 62309816,
      "step": 107400
    },
    {
      "epoch": 15.997170092344355,
      "grad_norm": 9.9375,
      "learning_rate": 5.857246261357666e-06,
      "loss": 1.1673,
      "num_input_tokens_seen": 62312888,
      "step": 107405
    },
    {
      "epoch": 15.997914804885314,
      "grad_norm": 13.625,
      "learning_rate": 5.855156441340331e-06,
      "loss": 0.9404,
      "num_input_tokens_seen": 62316056,
      "step": 107410
    },
    {
      "epoch": 15.998659517426274,
      "grad_norm": 3.765625,
      "learning_rate": 5.853066944755667e-06,
      "loss": 0.7288,
      "num_input_tokens_seen": 62318936,
      "step": 107415
    },
    {
      "epoch": 15.999404229967233,
      "grad_norm": 8.375,
      "learning_rate": 5.8509777716389715e-06,
      "loss": 0.9647,
      "num_input_tokens_seen": 62322040,
      "step": 107420
    },
    {
      "epoch": 16.0,
      "eval_loss": 1.0798132419586182,
      "eval_runtime": 43.8098,
      "eval_samples_per_second": 68.113,
      "eval_steps_per_second": 17.028,
      "num_input_tokens_seen": 62323952,
      "step": 107424
    },
    {
      "epoch": 16.000148942508194,
      "grad_norm": 7.25,
      "learning_rate": 5.848888922025553e-06,
      "loss": 0.4783,
      "num_input_tokens_seen": 62324624,
      "step": 107425
    },
    {
      "epoch": 16.00089365504915,
      "grad_norm": 18.75,
      "learning_rate": 5.8468003959506915e-06,
      "loss": 1.0128,
      "num_input_tokens_seen": 62327344,
      "step": 107430
    },
    {
      "epoch": 16.00163836759011,
      "grad_norm": 7.71875,
      "learning_rate": 5.844712193449662e-06,
      "loss": 1.2113,
      "num_input_tokens_seen": 62330128,
      "step": 107435
    },
    {
      "epoch": 16.002383080131068,
      "grad_norm": 9.0625,
      "learning_rate": 5.842624314557757e-06,
      "loss": 1.2137,
      "num_input_tokens_seen": 62332944,
      "step": 107440
    },
    {
      "epoch": 16.00312779267203,
      "grad_norm": 10.125,
      "learning_rate": 5.840536759310239e-06,
      "loss": 0.8031,
      "num_input_tokens_seen": 62335792,
      "step": 107445
    },
    {
      "epoch": 16.00387250521299,
      "grad_norm": 5.71875,
      "learning_rate": 5.838449527742388e-06,
      "loss": 1.0141,
      "num_input_tokens_seen": 62338736,
      "step": 107450
    },
    {
      "epoch": 16.004617217753946,
      "grad_norm": 5.1875,
      "learning_rate": 5.836362619889446e-06,
      "loss": 0.7817,
      "num_input_tokens_seen": 62341616,
      "step": 107455
    },
    {
      "epoch": 16.005361930294907,
      "grad_norm": 3.578125,
      "learning_rate": 5.83427603578669e-06,
      "loss": 0.8984,
      "num_input_tokens_seen": 62344368,
      "step": 107460
    },
    {
      "epoch": 16.006106642835864,
      "grad_norm": 5.0625,
      "learning_rate": 5.832189775469363e-06,
      "loss": 1.1122,
      "num_input_tokens_seen": 62347440,
      "step": 107465
    },
    {
      "epoch": 16.006851355376824,
      "grad_norm": 14.6875,
      "learning_rate": 5.8301038389727005e-06,
      "loss": 1.0499,
      "num_input_tokens_seen": 62350736,
      "step": 107470
    },
    {
      "epoch": 16.007596067917785,
      "grad_norm": 10.875,
      "learning_rate": 5.8280182263319545e-06,
      "loss": 1.366,
      "num_input_tokens_seen": 62353680,
      "step": 107475
    },
    {
      "epoch": 16.00834078045874,
      "grad_norm": 7.625,
      "learning_rate": 5.825932937582357e-06,
      "loss": 1.3353,
      "num_input_tokens_seen": 62356656,
      "step": 107480
    },
    {
      "epoch": 16.009085492999702,
      "grad_norm": 13.5,
      "learning_rate": 5.823847972759136e-06,
      "loss": 1.0575,
      "num_input_tokens_seen": 62359664,
      "step": 107485
    },
    {
      "epoch": 16.009830205540663,
      "grad_norm": 10.9375,
      "learning_rate": 5.821763331897503e-06,
      "loss": 1.0012,
      "num_input_tokens_seen": 62362512,
      "step": 107490
    },
    {
      "epoch": 16.01057491808162,
      "grad_norm": 7.46875,
      "learning_rate": 5.819679015032697e-06,
      "loss": 1.2277,
      "num_input_tokens_seen": 62365616,
      "step": 107495
    },
    {
      "epoch": 16.01131963062258,
      "grad_norm": 13.0625,
      "learning_rate": 5.81759502219992e-06,
      "loss": 1.1286,
      "num_input_tokens_seen": 62368528,
      "step": 107500
    },
    {
      "epoch": 16.012064343163537,
      "grad_norm": 15.4375,
      "learning_rate": 5.815511353434372e-06,
      "loss": 1.2254,
      "num_input_tokens_seen": 62371504,
      "step": 107505
    },
    {
      "epoch": 16.012809055704498,
      "grad_norm": 14.3125,
      "learning_rate": 5.813428008771266e-06,
      "loss": 0.937,
      "num_input_tokens_seen": 62374384,
      "step": 107510
    },
    {
      "epoch": 16.01355376824546,
      "grad_norm": 11.6875,
      "learning_rate": 5.811344988245787e-06,
      "loss": 1.0336,
      "num_input_tokens_seen": 62377008,
      "step": 107515
    },
    {
      "epoch": 16.014298480786415,
      "grad_norm": 9.125,
      "learning_rate": 5.809262291893141e-06,
      "loss": 0.9706,
      "num_input_tokens_seen": 62380208,
      "step": 107520
    },
    {
      "epoch": 16.015043193327376,
      "grad_norm": 15.5,
      "learning_rate": 5.807179919748496e-06,
      "loss": 1.1224,
      "num_input_tokens_seen": 62382960,
      "step": 107525
    },
    {
      "epoch": 16.015787905868336,
      "grad_norm": 10.3125,
      "learning_rate": 5.805097871847046e-06,
      "loss": 1.1599,
      "num_input_tokens_seen": 62385648,
      "step": 107530
    },
    {
      "epoch": 16.016532618409293,
      "grad_norm": 10.9375,
      "learning_rate": 5.803016148223953e-06,
      "loss": 0.9979,
      "num_input_tokens_seen": 62388496,
      "step": 107535
    },
    {
      "epoch": 16.017277330950254,
      "grad_norm": 7.78125,
      "learning_rate": 5.800934748914397e-06,
      "loss": 1.3986,
      "num_input_tokens_seen": 62391536,
      "step": 107540
    },
    {
      "epoch": 16.01802204349121,
      "grad_norm": 8.8125,
      "learning_rate": 5.798853673953536e-06,
      "loss": 1.029,
      "num_input_tokens_seen": 62394320,
      "step": 107545
    },
    {
      "epoch": 16.01876675603217,
      "grad_norm": 12.125,
      "learning_rate": 5.796772923376526e-06,
      "loss": 1.1112,
      "num_input_tokens_seen": 62397072,
      "step": 107550
    },
    {
      "epoch": 16.019511468573132,
      "grad_norm": 11.5625,
      "learning_rate": 5.794692497218521e-06,
      "loss": 1.0114,
      "num_input_tokens_seen": 62400016,
      "step": 107555
    },
    {
      "epoch": 16.02025618111409,
      "grad_norm": 12.9375,
      "learning_rate": 5.79261239551466e-06,
      "loss": 1.0398,
      "num_input_tokens_seen": 62402864,
      "step": 107560
    },
    {
      "epoch": 16.02100089365505,
      "grad_norm": 7.84375,
      "learning_rate": 5.790532618300099e-06,
      "loss": 0.9093,
      "num_input_tokens_seen": 62405968,
      "step": 107565
    },
    {
      "epoch": 16.02174560619601,
      "grad_norm": 12.25,
      "learning_rate": 5.788453165609955e-06,
      "loss": 1.4345,
      "num_input_tokens_seen": 62408880,
      "step": 107570
    },
    {
      "epoch": 16.022490318736967,
      "grad_norm": 7.75,
      "learning_rate": 5.786374037479381e-06,
      "loss": 0.8307,
      "num_input_tokens_seen": 62411952,
      "step": 107575
    },
    {
      "epoch": 16.023235031277927,
      "grad_norm": 14.8125,
      "learning_rate": 5.784295233943488e-06,
      "loss": 1.1941,
      "num_input_tokens_seen": 62414800,
      "step": 107580
    },
    {
      "epoch": 16.023979743818884,
      "grad_norm": 13.125,
      "learning_rate": 5.7822167550373865e-06,
      "loss": 0.8017,
      "num_input_tokens_seen": 62417616,
      "step": 107585
    },
    {
      "epoch": 16.024724456359845,
      "grad_norm": 14.0625,
      "learning_rate": 5.780138600796212e-06,
      "loss": 1.1186,
      "num_input_tokens_seen": 62420496,
      "step": 107590
    },
    {
      "epoch": 16.025469168900806,
      "grad_norm": 12.5,
      "learning_rate": 5.778060771255053e-06,
      "loss": 0.9131,
      "num_input_tokens_seen": 62423312,
      "step": 107595
    },
    {
      "epoch": 16.026213881441763,
      "grad_norm": 9.75,
      "learning_rate": 5.775983266449029e-06,
      "loss": 1.1152,
      "num_input_tokens_seen": 62426544,
      "step": 107600
    },
    {
      "epoch": 16.026958593982723,
      "grad_norm": 14.125,
      "learning_rate": 5.773906086413222e-06,
      "loss": 1.1419,
      "num_input_tokens_seen": 62429488,
      "step": 107605
    },
    {
      "epoch": 16.027703306523684,
      "grad_norm": 23.75,
      "learning_rate": 5.771829231182737e-06,
      "loss": 1.4153,
      "num_input_tokens_seen": 62432272,
      "step": 107610
    },
    {
      "epoch": 16.02844801906464,
      "grad_norm": 16.25,
      "learning_rate": 5.769752700792655e-06,
      "loss": 1.1566,
      "num_input_tokens_seen": 62435216,
      "step": 107615
    },
    {
      "epoch": 16.0291927316056,
      "grad_norm": 6.53125,
      "learning_rate": 5.767676495278057e-06,
      "loss": 0.9583,
      "num_input_tokens_seen": 62438032,
      "step": 107620
    },
    {
      "epoch": 16.029937444146558,
      "grad_norm": 4.96875,
      "learning_rate": 5.765600614674019e-06,
      "loss": 1.0706,
      "num_input_tokens_seen": 62440848,
      "step": 107625
    },
    {
      "epoch": 16.03068215668752,
      "grad_norm": 8.375,
      "learning_rate": 5.763525059015601e-06,
      "loss": 1.0159,
      "num_input_tokens_seen": 62443952,
      "step": 107630
    },
    {
      "epoch": 16.03142686922848,
      "grad_norm": 17.125,
      "learning_rate": 5.761449828337881e-06,
      "loss": 1.1025,
      "num_input_tokens_seen": 62446960,
      "step": 107635
    },
    {
      "epoch": 16.032171581769436,
      "grad_norm": 20.125,
      "learning_rate": 5.759374922675908e-06,
      "loss": 1.2513,
      "num_input_tokens_seen": 62449584,
      "step": 107640
    },
    {
      "epoch": 16.032916294310397,
      "grad_norm": 6.75,
      "learning_rate": 5.757300342064748e-06,
      "loss": 0.9623,
      "num_input_tokens_seen": 62452432,
      "step": 107645
    },
    {
      "epoch": 16.033661006851354,
      "grad_norm": 9.5625,
      "learning_rate": 5.755226086539433e-06,
      "loss": 1.2645,
      "num_input_tokens_seen": 62455664,
      "step": 107650
    },
    {
      "epoch": 16.034405719392314,
      "grad_norm": 8.4375,
      "learning_rate": 5.753152156135022e-06,
      "loss": 0.9302,
      "num_input_tokens_seen": 62458640,
      "step": 107655
    },
    {
      "epoch": 16.035150431933275,
      "grad_norm": 14.6875,
      "learning_rate": 5.751078550886543e-06,
      "loss": 1.273,
      "num_input_tokens_seen": 62461680,
      "step": 107660
    },
    {
      "epoch": 16.03589514447423,
      "grad_norm": 13.5625,
      "learning_rate": 5.749005270829022e-06,
      "loss": 0.9362,
      "num_input_tokens_seen": 62464816,
      "step": 107665
    },
    {
      "epoch": 16.036639857015192,
      "grad_norm": 14.625,
      "learning_rate": 5.746932315997497e-06,
      "loss": 1.7431,
      "num_input_tokens_seen": 62467376,
      "step": 107670
    },
    {
      "epoch": 16.037384569556153,
      "grad_norm": 6.875,
      "learning_rate": 5.744859686426976e-06,
      "loss": 1.1148,
      "num_input_tokens_seen": 62470288,
      "step": 107675
    },
    {
      "epoch": 16.03812928209711,
      "grad_norm": 15.8125,
      "learning_rate": 5.742787382152489e-06,
      "loss": 1.3949,
      "num_input_tokens_seen": 62473520,
      "step": 107680
    },
    {
      "epoch": 16.03887399463807,
      "grad_norm": 10.3125,
      "learning_rate": 5.740715403209035e-06,
      "loss": 1.0067,
      "num_input_tokens_seen": 62476560,
      "step": 107685
    },
    {
      "epoch": 16.039618707179027,
      "grad_norm": 18.875,
      "learning_rate": 5.738643749631623e-06,
      "loss": 0.9927,
      "num_input_tokens_seen": 62479600,
      "step": 107690
    },
    {
      "epoch": 16.040363419719988,
      "grad_norm": 8.625,
      "learning_rate": 5.736572421455239e-06,
      "loss": 0.9775,
      "num_input_tokens_seen": 62482416,
      "step": 107695
    },
    {
      "epoch": 16.04110813226095,
      "grad_norm": 12.0625,
      "learning_rate": 5.734501418714891e-06,
      "loss": 0.9923,
      "num_input_tokens_seen": 62485360,
      "step": 107700
    },
    {
      "epoch": 16.041852844801905,
      "grad_norm": 10.3125,
      "learning_rate": 5.732430741445563e-06,
      "loss": 1.1593,
      "num_input_tokens_seen": 62488208,
      "step": 107705
    },
    {
      "epoch": 16.042597557342866,
      "grad_norm": 8.4375,
      "learning_rate": 5.730360389682227e-06,
      "loss": 0.9632,
      "num_input_tokens_seen": 62491184,
      "step": 107710
    },
    {
      "epoch": 16.043342269883826,
      "grad_norm": 17.875,
      "learning_rate": 5.728290363459876e-06,
      "loss": 1.0433,
      "num_input_tokens_seen": 62493776,
      "step": 107715
    },
    {
      "epoch": 16.044086982424783,
      "grad_norm": 10.3125,
      "learning_rate": 5.726220662813464e-06,
      "loss": 1.0667,
      "num_input_tokens_seen": 62496624,
      "step": 107720
    },
    {
      "epoch": 16.044831694965744,
      "grad_norm": 15.375,
      "learning_rate": 5.72415128777797e-06,
      "loss": 1.0516,
      "num_input_tokens_seen": 62499376,
      "step": 107725
    },
    {
      "epoch": 16.0455764075067,
      "grad_norm": 10.0625,
      "learning_rate": 5.722082238388346e-06,
      "loss": 0.9108,
      "num_input_tokens_seen": 62502320,
      "step": 107730
    },
    {
      "epoch": 16.04632112004766,
      "grad_norm": 14.125,
      "learning_rate": 5.720013514679553e-06,
      "loss": 1.0109,
      "num_input_tokens_seen": 62505456,
      "step": 107735
    },
    {
      "epoch": 16.047065832588622,
      "grad_norm": 7.3125,
      "learning_rate": 5.71794511668654e-06,
      "loss": 1.0729,
      "num_input_tokens_seen": 62508528,
      "step": 107740
    },
    {
      "epoch": 16.04781054512958,
      "grad_norm": 10.5,
      "learning_rate": 5.7158770444442425e-06,
      "loss": 0.8755,
      "num_input_tokens_seen": 62511440,
      "step": 107745
    },
    {
      "epoch": 16.04855525767054,
      "grad_norm": 7.625,
      "learning_rate": 5.713809297987599e-06,
      "loss": 1.0465,
      "num_input_tokens_seen": 62514224,
      "step": 107750
    },
    {
      "epoch": 16.0492999702115,
      "grad_norm": 6.4375,
      "learning_rate": 5.7117418773515535e-06,
      "loss": 1.2395,
      "num_input_tokens_seen": 62517104,
      "step": 107755
    },
    {
      "epoch": 16.050044682752457,
      "grad_norm": 10.0,
      "learning_rate": 5.709674782571023e-06,
      "loss": 1.3319,
      "num_input_tokens_seen": 62520016,
      "step": 107760
    },
    {
      "epoch": 16.050789395293418,
      "grad_norm": 13.5,
      "learning_rate": 5.707608013680923e-06,
      "loss": 0.8893,
      "num_input_tokens_seen": 62522864,
      "step": 107765
    },
    {
      "epoch": 16.051534107834375,
      "grad_norm": 13.875,
      "learning_rate": 5.705541570716189e-06,
      "loss": 1.3232,
      "num_input_tokens_seen": 62525968,
      "step": 107770
    },
    {
      "epoch": 16.052278820375335,
      "grad_norm": 10.25,
      "learning_rate": 5.70347545371171e-06,
      "loss": 0.8548,
      "num_input_tokens_seen": 62528848,
      "step": 107775
    },
    {
      "epoch": 16.053023532916296,
      "grad_norm": 7.375,
      "learning_rate": 5.701409662702409e-06,
      "loss": 0.9658,
      "num_input_tokens_seen": 62531824,
      "step": 107780
    },
    {
      "epoch": 16.053768245457253,
      "grad_norm": 9.1875,
      "learning_rate": 5.699344197723178e-06,
      "loss": 1.0406,
      "num_input_tokens_seen": 62534672,
      "step": 107785
    },
    {
      "epoch": 16.054512957998213,
      "grad_norm": 11.0625,
      "learning_rate": 5.697279058808902e-06,
      "loss": 1.0553,
      "num_input_tokens_seen": 62537520,
      "step": 107790
    },
    {
      "epoch": 16.055257670539174,
      "grad_norm": 8.5,
      "learning_rate": 5.6952142459944845e-06,
      "loss": 0.6724,
      "num_input_tokens_seen": 62540272,
      "step": 107795
    },
    {
      "epoch": 16.05600238308013,
      "grad_norm": 14.6875,
      "learning_rate": 5.693149759314798e-06,
      "loss": 0.9767,
      "num_input_tokens_seen": 62543056,
      "step": 107800
    },
    {
      "epoch": 16.05674709562109,
      "grad_norm": 10.75,
      "learning_rate": 5.691085598804727e-06,
      "loss": 1.0348,
      "num_input_tokens_seen": 62545968,
      "step": 107805
    },
    {
      "epoch": 16.057491808162048,
      "grad_norm": 8.625,
      "learning_rate": 5.689021764499142e-06,
      "loss": 1.073,
      "num_input_tokens_seen": 62548528,
      "step": 107810
    },
    {
      "epoch": 16.05823652070301,
      "grad_norm": 14.0,
      "learning_rate": 5.6869582564329085e-06,
      "loss": 0.8797,
      "num_input_tokens_seen": 62551312,
      "step": 107815
    },
    {
      "epoch": 16.05898123324397,
      "grad_norm": 15.5,
      "learning_rate": 5.684895074640884e-06,
      "loss": 1.0027,
      "num_input_tokens_seen": 62554448,
      "step": 107820
    },
    {
      "epoch": 16.059725945784926,
      "grad_norm": 11.25,
      "learning_rate": 5.682832219157922e-06,
      "loss": 1.0263,
      "num_input_tokens_seen": 62557424,
      "step": 107825
    },
    {
      "epoch": 16.060470658325887,
      "grad_norm": 11.5,
      "learning_rate": 5.68076969001888e-06,
      "loss": 1.2022,
      "num_input_tokens_seen": 62560720,
      "step": 107830
    },
    {
      "epoch": 16.061215370866844,
      "grad_norm": 12.3125,
      "learning_rate": 5.678707487258594e-06,
      "loss": 1.2547,
      "num_input_tokens_seen": 62563952,
      "step": 107835
    },
    {
      "epoch": 16.061960083407804,
      "grad_norm": 12.25,
      "learning_rate": 5.676645610911916e-06,
      "loss": 1.1058,
      "num_input_tokens_seen": 62567152,
      "step": 107840
    },
    {
      "epoch": 16.062704795948765,
      "grad_norm": 10.875,
      "learning_rate": 5.674584061013663e-06,
      "loss": 1.1012,
      "num_input_tokens_seen": 62570000,
      "step": 107845
    },
    {
      "epoch": 16.06344950848972,
      "grad_norm": 9.0625,
      "learning_rate": 5.672522837598676e-06,
      "loss": 1.2727,
      "num_input_tokens_seen": 62572848,
      "step": 107850
    },
    {
      "epoch": 16.064194221030682,
      "grad_norm": 18.875,
      "learning_rate": 5.670461940701768e-06,
      "loss": 1.0364,
      "num_input_tokens_seen": 62575664,
      "step": 107855
    },
    {
      "epoch": 16.064938933571643,
      "grad_norm": 8.0625,
      "learning_rate": 5.668401370357765e-06,
      "loss": 0.7705,
      "num_input_tokens_seen": 62578608,
      "step": 107860
    },
    {
      "epoch": 16.0656836461126,
      "grad_norm": 7.0625,
      "learning_rate": 5.666341126601474e-06,
      "loss": 1.0095,
      "num_input_tokens_seen": 62581424,
      "step": 107865
    },
    {
      "epoch": 16.06642835865356,
      "grad_norm": 11.3125,
      "learning_rate": 5.664281209467692e-06,
      "loss": 1.0265,
      "num_input_tokens_seen": 62584240,
      "step": 107870
    },
    {
      "epoch": 16.067173071194517,
      "grad_norm": 10.625,
      "learning_rate": 5.662221618991234e-06,
      "loss": 1.2116,
      "num_input_tokens_seen": 62587216,
      "step": 107875
    },
    {
      "epoch": 16.067917783735478,
      "grad_norm": 12.5,
      "learning_rate": 5.660162355206888e-06,
      "loss": 1.4792,
      "num_input_tokens_seen": 62590160,
      "step": 107880
    },
    {
      "epoch": 16.06866249627644,
      "grad_norm": 10.8125,
      "learning_rate": 5.658103418149443e-06,
      "loss": 1.3222,
      "num_input_tokens_seen": 62592848,
      "step": 107885
    },
    {
      "epoch": 16.069407208817395,
      "grad_norm": 11.375,
      "learning_rate": 5.656044807853675e-06,
      "loss": 1.1889,
      "num_input_tokens_seen": 62595728,
      "step": 107890
    },
    {
      "epoch": 16.070151921358356,
      "grad_norm": 12.1875,
      "learning_rate": 5.653986524354377e-06,
      "loss": 1.1226,
      "num_input_tokens_seen": 62598512,
      "step": 107895
    },
    {
      "epoch": 16.070896633899316,
      "grad_norm": 14.5,
      "learning_rate": 5.651928567686307e-06,
      "loss": 1.3078,
      "num_input_tokens_seen": 62601232,
      "step": 107900
    },
    {
      "epoch": 16.071641346440273,
      "grad_norm": 15.3125,
      "learning_rate": 5.649870937884247e-06,
      "loss": 1.3313,
      "num_input_tokens_seen": 62604080,
      "step": 107905
    },
    {
      "epoch": 16.072386058981234,
      "grad_norm": 6.96875,
      "learning_rate": 5.647813634982952e-06,
      "loss": 1.3009,
      "num_input_tokens_seen": 62606992,
      "step": 107910
    },
    {
      "epoch": 16.07313077152219,
      "grad_norm": 9.5625,
      "learning_rate": 5.6457566590171675e-06,
      "loss": 0.8081,
      "num_input_tokens_seen": 62609616,
      "step": 107915
    },
    {
      "epoch": 16.07387548406315,
      "grad_norm": 6.09375,
      "learning_rate": 5.64370001002166e-06,
      "loss": 0.8997,
      "num_input_tokens_seen": 62612368,
      "step": 107920
    },
    {
      "epoch": 16.074620196604112,
      "grad_norm": 17.125,
      "learning_rate": 5.641643688031162e-06,
      "loss": 1.3946,
      "num_input_tokens_seen": 62615184,
      "step": 107925
    },
    {
      "epoch": 16.07536490914507,
      "grad_norm": 7.75,
      "learning_rate": 5.639587693080428e-06,
      "loss": 1.1383,
      "num_input_tokens_seen": 62618192,
      "step": 107930
    },
    {
      "epoch": 16.07610962168603,
      "grad_norm": 6.65625,
      "learning_rate": 5.637532025204173e-06,
      "loss": 1.3291,
      "num_input_tokens_seen": 62621168,
      "step": 107935
    },
    {
      "epoch": 16.07685433422699,
      "grad_norm": 4.3125,
      "learning_rate": 5.635476684437144e-06,
      "loss": 1.089,
      "num_input_tokens_seen": 62623856,
      "step": 107940
    },
    {
      "epoch": 16.077599046767947,
      "grad_norm": 5.96875,
      "learning_rate": 5.633421670814054e-06,
      "loss": 0.6725,
      "num_input_tokens_seen": 62626864,
      "step": 107945
    },
    {
      "epoch": 16.078343759308908,
      "grad_norm": 11.0625,
      "learning_rate": 5.631366984369624e-06,
      "loss": 1.1235,
      "num_input_tokens_seen": 62629840,
      "step": 107950
    },
    {
      "epoch": 16.079088471849865,
      "grad_norm": 12.125,
      "learning_rate": 5.629312625138561e-06,
      "loss": 1.2107,
      "num_input_tokens_seen": 62632848,
      "step": 107955
    },
    {
      "epoch": 16.079833184390825,
      "grad_norm": 17.125,
      "learning_rate": 5.627258593155568e-06,
      "loss": 1.1277,
      "num_input_tokens_seen": 62636080,
      "step": 107960
    },
    {
      "epoch": 16.080577896931786,
      "grad_norm": 9.0625,
      "learning_rate": 5.625204888455357e-06,
      "loss": 1.1979,
      "num_input_tokens_seen": 62639056,
      "step": 107965
    },
    {
      "epoch": 16.081322609472743,
      "grad_norm": 9.625,
      "learning_rate": 5.623151511072613e-06,
      "loss": 1.1011,
      "num_input_tokens_seen": 62641840,
      "step": 107970
    },
    {
      "epoch": 16.082067322013703,
      "grad_norm": 12.25,
      "learning_rate": 5.6210984610420345e-06,
      "loss": 1.1026,
      "num_input_tokens_seen": 62644720,
      "step": 107975
    },
    {
      "epoch": 16.082812034554664,
      "grad_norm": 7.09375,
      "learning_rate": 5.619045738398299e-06,
      "loss": 1.1903,
      "num_input_tokens_seen": 62647856,
      "step": 107980
    },
    {
      "epoch": 16.08355674709562,
      "grad_norm": 12.0625,
      "learning_rate": 5.616993343176091e-06,
      "loss": 1.043,
      "num_input_tokens_seen": 62650736,
      "step": 107985
    },
    {
      "epoch": 16.08430145963658,
      "grad_norm": 17.25,
      "learning_rate": 5.614941275410082e-06,
      "loss": 1.4521,
      "num_input_tokens_seen": 62653872,
      "step": 107990
    },
    {
      "epoch": 16.085046172177538,
      "grad_norm": 8.4375,
      "learning_rate": 5.61288953513493e-06,
      "loss": 1.1159,
      "num_input_tokens_seen": 62656912,
      "step": 107995
    },
    {
      "epoch": 16.0857908847185,
      "grad_norm": 11.25,
      "learning_rate": 5.610838122385312e-06,
      "loss": 0.9696,
      "num_input_tokens_seen": 62659824,
      "step": 108000
    },
    {
      "epoch": 16.08653559725946,
      "grad_norm": 5.59375,
      "learning_rate": 5.608787037195873e-06,
      "loss": 1.0349,
      "num_input_tokens_seen": 62662352,
      "step": 108005
    },
    {
      "epoch": 16.087280309800416,
      "grad_norm": 14.0625,
      "learning_rate": 5.606736279601274e-06,
      "loss": 0.9874,
      "num_input_tokens_seen": 62665136,
      "step": 108010
    },
    {
      "epoch": 16.088025022341377,
      "grad_norm": 15.8125,
      "learning_rate": 5.6046858496361545e-06,
      "loss": 1.0919,
      "num_input_tokens_seen": 62668112,
      "step": 108015
    },
    {
      "epoch": 16.088769734882334,
      "grad_norm": 17.75,
      "learning_rate": 5.602635747335155e-06,
      "loss": 1.4446,
      "num_input_tokens_seen": 62670736,
      "step": 108020
    },
    {
      "epoch": 16.089514447423294,
      "grad_norm": 9.1875,
      "learning_rate": 5.600585972732911e-06,
      "loss": 1.2481,
      "num_input_tokens_seen": 62673584,
      "step": 108025
    },
    {
      "epoch": 16.090259159964255,
      "grad_norm": 9.625,
      "learning_rate": 5.598536525864042e-06,
      "loss": 1.0919,
      "num_input_tokens_seen": 62676432,
      "step": 108030
    },
    {
      "epoch": 16.091003872505212,
      "grad_norm": 14.9375,
      "learning_rate": 5.59648740676319e-06,
      "loss": 1.1477,
      "num_input_tokens_seen": 62679312,
      "step": 108035
    },
    {
      "epoch": 16.091748585046172,
      "grad_norm": 9.9375,
      "learning_rate": 5.594438615464953e-06,
      "loss": 1.0497,
      "num_input_tokens_seen": 62682352,
      "step": 108040
    },
    {
      "epoch": 16.092493297587133,
      "grad_norm": 4.75,
      "learning_rate": 5.59239015200396e-06,
      "loss": 0.841,
      "num_input_tokens_seen": 62685424,
      "step": 108045
    },
    {
      "epoch": 16.09323801012809,
      "grad_norm": 14.1875,
      "learning_rate": 5.590342016414801e-06,
      "loss": 1.0307,
      "num_input_tokens_seen": 62688432,
      "step": 108050
    },
    {
      "epoch": 16.09398272266905,
      "grad_norm": 17.625,
      "learning_rate": 5.588294208732098e-06,
      "loss": 1.263,
      "num_input_tokens_seen": 62691568,
      "step": 108055
    },
    {
      "epoch": 16.094727435210007,
      "grad_norm": 12.375,
      "learning_rate": 5.586246728990424e-06,
      "loss": 1.1874,
      "num_input_tokens_seen": 62694288,
      "step": 108060
    },
    {
      "epoch": 16.095472147750968,
      "grad_norm": 6.40625,
      "learning_rate": 5.584199577224389e-06,
      "loss": 1.0885,
      "num_input_tokens_seen": 62696944,
      "step": 108065
    },
    {
      "epoch": 16.09621686029193,
      "grad_norm": 13.3125,
      "learning_rate": 5.58215275346857e-06,
      "loss": 1.1932,
      "num_input_tokens_seen": 62699792,
      "step": 108070
    },
    {
      "epoch": 16.096961572832885,
      "grad_norm": 20.125,
      "learning_rate": 5.580106257757542e-06,
      "loss": 0.8838,
      "num_input_tokens_seen": 62702704,
      "step": 108075
    },
    {
      "epoch": 16.097706285373846,
      "grad_norm": 12.125,
      "learning_rate": 5.5780600901258774e-06,
      "loss": 1.5175,
      "num_input_tokens_seen": 62705520,
      "step": 108080
    },
    {
      "epoch": 16.098450997914806,
      "grad_norm": 15.5625,
      "learning_rate": 5.576014250608152e-06,
      "loss": 1.215,
      "num_input_tokens_seen": 62708400,
      "step": 108085
    },
    {
      "epoch": 16.099195710455763,
      "grad_norm": 6.3125,
      "learning_rate": 5.573968739238927e-06,
      "loss": 0.8065,
      "num_input_tokens_seen": 62711376,
      "step": 108090
    },
    {
      "epoch": 16.099940422996724,
      "grad_norm": 18.5,
      "learning_rate": 5.571923556052749e-06,
      "loss": 1.2527,
      "num_input_tokens_seen": 62714384,
      "step": 108095
    },
    {
      "epoch": 16.10068513553768,
      "grad_norm": 9.0,
      "learning_rate": 5.569878701084183e-06,
      "loss": 0.989,
      "num_input_tokens_seen": 62717008,
      "step": 108100
    },
    {
      "epoch": 16.10142984807864,
      "grad_norm": 14.625,
      "learning_rate": 5.567834174367767e-06,
      "loss": 0.9576,
      "num_input_tokens_seen": 62720112,
      "step": 108105
    },
    {
      "epoch": 16.102174560619602,
      "grad_norm": 21.0,
      "learning_rate": 5.565789975938038e-06,
      "loss": 1.2712,
      "num_input_tokens_seen": 62723152,
      "step": 108110
    },
    {
      "epoch": 16.10291927316056,
      "grad_norm": 10.4375,
      "learning_rate": 5.56374610582954e-06,
      "loss": 1.5521,
      "num_input_tokens_seen": 62726128,
      "step": 108115
    },
    {
      "epoch": 16.10366398570152,
      "grad_norm": 11.3125,
      "learning_rate": 5.561702564076793e-06,
      "loss": 0.868,
      "num_input_tokens_seen": 62729072,
      "step": 108120
    },
    {
      "epoch": 16.10440869824248,
      "grad_norm": 11.8125,
      "learning_rate": 5.5596593507143304e-06,
      "loss": 0.943,
      "num_input_tokens_seen": 62731632,
      "step": 108125
    },
    {
      "epoch": 16.105153410783437,
      "grad_norm": 8.75,
      "learning_rate": 5.557616465776658e-06,
      "loss": 1.4381,
      "num_input_tokens_seen": 62734384,
      "step": 108130
    },
    {
      "epoch": 16.105898123324398,
      "grad_norm": 7.5,
      "learning_rate": 5.5555739092983e-06,
      "loss": 0.8437,
      "num_input_tokens_seen": 62737264,
      "step": 108135
    },
    {
      "epoch": 16.106642835865355,
      "grad_norm": 12.625,
      "learning_rate": 5.553531681313762e-06,
      "loss": 1.0849,
      "num_input_tokens_seen": 62740016,
      "step": 108140
    },
    {
      "epoch": 16.107387548406315,
      "grad_norm": 12.625,
      "learning_rate": 5.5514897818575415e-06,
      "loss": 1.3571,
      "num_input_tokens_seen": 62742928,
      "step": 108145
    },
    {
      "epoch": 16.108132260947276,
      "grad_norm": 17.5,
      "learning_rate": 5.549448210964131e-06,
      "loss": 1.0024,
      "num_input_tokens_seen": 62745648,
      "step": 108150
    },
    {
      "epoch": 16.108876973488233,
      "grad_norm": 9.5,
      "learning_rate": 5.5474069686680205e-06,
      "loss": 1.2309,
      "num_input_tokens_seen": 62748848,
      "step": 108155
    },
    {
      "epoch": 16.109621686029193,
      "grad_norm": 6.0625,
      "learning_rate": 5.545366055003706e-06,
      "loss": 0.8822,
      "num_input_tokens_seen": 62751760,
      "step": 108160
    },
    {
      "epoch": 16.11036639857015,
      "grad_norm": 10.8125,
      "learning_rate": 5.54332547000565e-06,
      "loss": 0.9246,
      "num_input_tokens_seen": 62754928,
      "step": 108165
    },
    {
      "epoch": 16.11111111111111,
      "grad_norm": 4.0,
      "learning_rate": 5.541285213708342e-06,
      "loss": 1.0308,
      "num_input_tokens_seen": 62757744,
      "step": 108170
    },
    {
      "epoch": 16.11185582365207,
      "grad_norm": 12.4375,
      "learning_rate": 5.539245286146238e-06,
      "loss": 1.0405,
      "num_input_tokens_seen": 62760752,
      "step": 108175
    },
    {
      "epoch": 16.11260053619303,
      "grad_norm": 7.21875,
      "learning_rate": 5.537205687353813e-06,
      "loss": 0.9647,
      "num_input_tokens_seen": 62763824,
      "step": 108180
    },
    {
      "epoch": 16.11334524873399,
      "grad_norm": 15.5,
      "learning_rate": 5.535166417365517e-06,
      "loss": 1.0303,
      "num_input_tokens_seen": 62766448,
      "step": 108185
    },
    {
      "epoch": 16.11408996127495,
      "grad_norm": 8.5625,
      "learning_rate": 5.533127476215791e-06,
      "loss": 1.4182,
      "num_input_tokens_seen": 62769328,
      "step": 108190
    },
    {
      "epoch": 16.114834673815906,
      "grad_norm": 14.125,
      "learning_rate": 5.531088863939101e-06,
      "loss": 1.2684,
      "num_input_tokens_seen": 62772144,
      "step": 108195
    },
    {
      "epoch": 16.115579386356867,
      "grad_norm": 17.625,
      "learning_rate": 5.529050580569869e-06,
      "loss": 1.2176,
      "num_input_tokens_seen": 62774992,
      "step": 108200
    },
    {
      "epoch": 16.116324098897824,
      "grad_norm": 17.0,
      "learning_rate": 5.527012626142547e-06,
      "loss": 1.3043,
      "num_input_tokens_seen": 62777904,
      "step": 108205
    },
    {
      "epoch": 16.117068811438784,
      "grad_norm": 5.625,
      "learning_rate": 5.524975000691554e-06,
      "loss": 1.0371,
      "num_input_tokens_seen": 62780720,
      "step": 108210
    },
    {
      "epoch": 16.117813523979745,
      "grad_norm": 21.875,
      "learning_rate": 5.522937704251316e-06,
      "loss": 1.197,
      "num_input_tokens_seen": 62783536,
      "step": 108215
    },
    {
      "epoch": 16.118558236520702,
      "grad_norm": 12.4375,
      "learning_rate": 5.520900736856241e-06,
      "loss": 0.9264,
      "num_input_tokens_seen": 62786384,
      "step": 108220
    },
    {
      "epoch": 16.119302949061662,
      "grad_norm": 11.0,
      "learning_rate": 5.5188640985407575e-06,
      "loss": 1.1869,
      "num_input_tokens_seen": 62789136,
      "step": 108225
    },
    {
      "epoch": 16.120047661602623,
      "grad_norm": 10.25,
      "learning_rate": 5.516827789339266e-06,
      "loss": 1.0369,
      "num_input_tokens_seen": 62792112,
      "step": 108230
    },
    {
      "epoch": 16.12079237414358,
      "grad_norm": 9.75,
      "learning_rate": 5.51479180928616e-06,
      "loss": 0.984,
      "num_input_tokens_seen": 62795152,
      "step": 108235
    },
    {
      "epoch": 16.12153708668454,
      "grad_norm": 10.375,
      "learning_rate": 5.5127561584158495e-06,
      "loss": 1.4051,
      "num_input_tokens_seen": 62798160,
      "step": 108240
    },
    {
      "epoch": 16.122281799225497,
      "grad_norm": 3.984375,
      "learning_rate": 5.510720836762712e-06,
      "loss": 1.0594,
      "num_input_tokens_seen": 62801008,
      "step": 108245
    },
    {
      "epoch": 16.123026511766458,
      "grad_norm": 11.625,
      "learning_rate": 5.508685844361142e-06,
      "loss": 1.1148,
      "num_input_tokens_seen": 62804080,
      "step": 108250
    },
    {
      "epoch": 16.12377122430742,
      "grad_norm": 14.0,
      "learning_rate": 5.506651181245509e-06,
      "loss": 1.2245,
      "num_input_tokens_seen": 62807120,
      "step": 108255
    },
    {
      "epoch": 16.124515936848375,
      "grad_norm": 15.1875,
      "learning_rate": 5.5046168474502e-06,
      "loss": 1.1027,
      "num_input_tokens_seen": 62810352,
      "step": 108260
    },
    {
      "epoch": 16.125260649389336,
      "grad_norm": 10.5625,
      "learning_rate": 5.502582843009577e-06,
      "loss": 1.2134,
      "num_input_tokens_seen": 62813232,
      "step": 108265
    },
    {
      "epoch": 16.126005361930297,
      "grad_norm": 6.375,
      "learning_rate": 5.500549167957989e-06,
      "loss": 0.9153,
      "num_input_tokens_seen": 62816112,
      "step": 108270
    },
    {
      "epoch": 16.126750074471254,
      "grad_norm": 4.875,
      "learning_rate": 5.498515822329814e-06,
      "loss": 1.0216,
      "num_input_tokens_seen": 62819056,
      "step": 108275
    },
    {
      "epoch": 16.127494787012214,
      "grad_norm": 10.625,
      "learning_rate": 5.496482806159395e-06,
      "loss": 1.0835,
      "num_input_tokens_seen": 62821808,
      "step": 108280
    },
    {
      "epoch": 16.12823949955317,
      "grad_norm": 6.65625,
      "learning_rate": 5.494450119481073e-06,
      "loss": 1.0049,
      "num_input_tokens_seen": 62824592,
      "step": 108285
    },
    {
      "epoch": 16.12898421209413,
      "grad_norm": 13.0,
      "learning_rate": 5.492417762329188e-06,
      "loss": 1.1687,
      "num_input_tokens_seen": 62827376,
      "step": 108290
    },
    {
      "epoch": 16.129728924635092,
      "grad_norm": 15.25,
      "learning_rate": 5.490385734738082e-06,
      "loss": 1.0098,
      "num_input_tokens_seen": 62830256,
      "step": 108295
    },
    {
      "epoch": 16.13047363717605,
      "grad_norm": 4.78125,
      "learning_rate": 5.4883540367420775e-06,
      "loss": 0.9284,
      "num_input_tokens_seen": 62833008,
      "step": 108300
    },
    {
      "epoch": 16.13121834971701,
      "grad_norm": 12.75,
      "learning_rate": 5.486322668375504e-06,
      "loss": 0.9485,
      "num_input_tokens_seen": 62835824,
      "step": 108305
    },
    {
      "epoch": 16.13196306225797,
      "grad_norm": 12.75,
      "learning_rate": 5.484291629672677e-06,
      "loss": 1.1469,
      "num_input_tokens_seen": 62839024,
      "step": 108310
    },
    {
      "epoch": 16.132707774798927,
      "grad_norm": 20.125,
      "learning_rate": 5.482260920667903e-06,
      "loss": 0.8258,
      "num_input_tokens_seen": 62842160,
      "step": 108315
    },
    {
      "epoch": 16.133452487339888,
      "grad_norm": 10.1875,
      "learning_rate": 5.480230541395501e-06,
      "loss": 1.1062,
      "num_input_tokens_seen": 62845040,
      "step": 108320
    },
    {
      "epoch": 16.134197199880845,
      "grad_norm": 14.25,
      "learning_rate": 5.478200491889754e-06,
      "loss": 1.0188,
      "num_input_tokens_seen": 62847824,
      "step": 108325
    },
    {
      "epoch": 16.134941912421805,
      "grad_norm": 10.4375,
      "learning_rate": 5.47617077218498e-06,
      "loss": 0.903,
      "num_input_tokens_seen": 62850736,
      "step": 108330
    },
    {
      "epoch": 16.135686624962766,
      "grad_norm": 8.1875,
      "learning_rate": 5.474141382315448e-06,
      "loss": 1.4927,
      "num_input_tokens_seen": 62853744,
      "step": 108335
    },
    {
      "epoch": 16.136431337503723,
      "grad_norm": 8.875,
      "learning_rate": 5.472112322315459e-06,
      "loss": 1.2105,
      "num_input_tokens_seen": 62856464,
      "step": 108340
    },
    {
      "epoch": 16.137176050044683,
      "grad_norm": 12.625,
      "learning_rate": 5.4700835922192885e-06,
      "loss": 1.0935,
      "num_input_tokens_seen": 62859344,
      "step": 108345
    },
    {
      "epoch": 16.13792076258564,
      "grad_norm": 14.5625,
      "learning_rate": 5.468055192061203e-06,
      "loss": 1.1144,
      "num_input_tokens_seen": 62862672,
      "step": 108350
    },
    {
      "epoch": 16.1386654751266,
      "grad_norm": 7.0,
      "learning_rate": 5.466027121875475e-06,
      "loss": 0.8974,
      "num_input_tokens_seen": 62865808,
      "step": 108355
    },
    {
      "epoch": 16.13941018766756,
      "grad_norm": 7.84375,
      "learning_rate": 5.463999381696358e-06,
      "loss": 1.2366,
      "num_input_tokens_seen": 62868656,
      "step": 108360
    },
    {
      "epoch": 16.14015490020852,
      "grad_norm": 10.0625,
      "learning_rate": 5.4619719715581215e-06,
      "loss": 1.3147,
      "num_input_tokens_seen": 62871664,
      "step": 108365
    },
    {
      "epoch": 16.14089961274948,
      "grad_norm": 10.125,
      "learning_rate": 5.4599448914950055e-06,
      "loss": 0.8655,
      "num_input_tokens_seen": 62874704,
      "step": 108370
    },
    {
      "epoch": 16.14164432529044,
      "grad_norm": 8.9375,
      "learning_rate": 5.457918141541268e-06,
      "loss": 0.9111,
      "num_input_tokens_seen": 62877808,
      "step": 108375
    },
    {
      "epoch": 16.142389037831396,
      "grad_norm": 8.25,
      "learning_rate": 5.455891721731135e-06,
      "loss": 0.7956,
      "num_input_tokens_seen": 62880656,
      "step": 108380
    },
    {
      "epoch": 16.143133750372357,
      "grad_norm": 6.0625,
      "learning_rate": 5.453865632098853e-06,
      "loss": 1.197,
      "num_input_tokens_seen": 62883664,
      "step": 108385
    },
    {
      "epoch": 16.143878462913314,
      "grad_norm": 7.96875,
      "learning_rate": 5.451839872678646e-06,
      "loss": 1.2969,
      "num_input_tokens_seen": 62886608,
      "step": 108390
    },
    {
      "epoch": 16.144623175454274,
      "grad_norm": 7.84375,
      "learning_rate": 5.449814443504731e-06,
      "loss": 0.9308,
      "num_input_tokens_seen": 62889488,
      "step": 108395
    },
    {
      "epoch": 16.145367887995235,
      "grad_norm": 12.4375,
      "learning_rate": 5.447789344611337e-06,
      "loss": 1.0854,
      "num_input_tokens_seen": 62892464,
      "step": 108400
    },
    {
      "epoch": 16.146112600536192,
      "grad_norm": 9.0625,
      "learning_rate": 5.445764576032672e-06,
      "loss": 1.5062,
      "num_input_tokens_seen": 62895408,
      "step": 108405
    },
    {
      "epoch": 16.146857313077152,
      "grad_norm": 6.4375,
      "learning_rate": 5.44374013780293e-06,
      "loss": 1.11,
      "num_input_tokens_seen": 62898512,
      "step": 108410
    },
    {
      "epoch": 16.147602025618113,
      "grad_norm": 8.5625,
      "learning_rate": 5.441716029956331e-06,
      "loss": 0.9457,
      "num_input_tokens_seen": 62902256,
      "step": 108415
    },
    {
      "epoch": 16.14834673815907,
      "grad_norm": 14.25,
      "learning_rate": 5.439692252527062e-06,
      "loss": 0.9993,
      "num_input_tokens_seen": 62904944,
      "step": 108420
    },
    {
      "epoch": 16.14909145070003,
      "grad_norm": 10.375,
      "learning_rate": 5.437668805549312e-06,
      "loss": 0.7808,
      "num_input_tokens_seen": 62907856,
      "step": 108425
    },
    {
      "epoch": 16.149836163240987,
      "grad_norm": 9.0,
      "learning_rate": 5.435645689057256e-06,
      "loss": 1.1759,
      "num_input_tokens_seen": 62910736,
      "step": 108430
    },
    {
      "epoch": 16.150580875781948,
      "grad_norm": 12.625,
      "learning_rate": 5.433622903085092e-06,
      "loss": 0.9446,
      "num_input_tokens_seen": 62913648,
      "step": 108435
    },
    {
      "epoch": 16.15132558832291,
      "grad_norm": 7.65625,
      "learning_rate": 5.4316004476669735e-06,
      "loss": 1.0435,
      "num_input_tokens_seen": 62916784,
      "step": 108440
    },
    {
      "epoch": 16.152070300863866,
      "grad_norm": 3.78125,
      "learning_rate": 5.429578322837084e-06,
      "loss": 0.786,
      "num_input_tokens_seen": 62919632,
      "step": 108445
    },
    {
      "epoch": 16.152815013404826,
      "grad_norm": 10.5625,
      "learning_rate": 5.4275565286295735e-06,
      "loss": 1.2455,
      "num_input_tokens_seen": 62922896,
      "step": 108450
    },
    {
      "epoch": 16.153559725945787,
      "grad_norm": 5.78125,
      "learning_rate": 5.425535065078608e-06,
      "loss": 0.9773,
      "num_input_tokens_seen": 62925936,
      "step": 108455
    },
    {
      "epoch": 16.154304438486744,
      "grad_norm": 14.4375,
      "learning_rate": 5.423513932218327e-06,
      "loss": 1.0336,
      "num_input_tokens_seen": 62928784,
      "step": 108460
    },
    {
      "epoch": 16.155049151027704,
      "grad_norm": 9.625,
      "learning_rate": 5.421493130082889e-06,
      "loss": 1.1221,
      "num_input_tokens_seen": 62931792,
      "step": 108465
    },
    {
      "epoch": 16.15579386356866,
      "grad_norm": 8.1875,
      "learning_rate": 5.419472658706423e-06,
      "loss": 1.1042,
      "num_input_tokens_seen": 62934768,
      "step": 108470
    },
    {
      "epoch": 16.15653857610962,
      "grad_norm": 13.3125,
      "learning_rate": 5.417452518123067e-06,
      "loss": 1.0908,
      "num_input_tokens_seen": 62937456,
      "step": 108475
    },
    {
      "epoch": 16.157283288650582,
      "grad_norm": 18.0,
      "learning_rate": 5.415432708366949e-06,
      "loss": 1.2395,
      "num_input_tokens_seen": 62940464,
      "step": 108480
    },
    {
      "epoch": 16.15802800119154,
      "grad_norm": 10.125,
      "learning_rate": 5.413413229472184e-06,
      "loss": 0.6723,
      "num_input_tokens_seen": 62943376,
      "step": 108485
    },
    {
      "epoch": 16.1587727137325,
      "grad_norm": 6.8125,
      "learning_rate": 5.411394081472901e-06,
      "loss": 1.038,
      "num_input_tokens_seen": 62946160,
      "step": 108490
    },
    {
      "epoch": 16.15951742627346,
      "grad_norm": 7.40625,
      "learning_rate": 5.409375264403199e-06,
      "loss": 0.8172,
      "num_input_tokens_seen": 62948944,
      "step": 108495
    },
    {
      "epoch": 16.160262138814417,
      "grad_norm": 16.25,
      "learning_rate": 5.407356778297198e-06,
      "loss": 1.1992,
      "num_input_tokens_seen": 62951824,
      "step": 108500
    },
    {
      "epoch": 16.161006851355378,
      "grad_norm": 21.0,
      "learning_rate": 5.4053386231889855e-06,
      "loss": 1.2718,
      "num_input_tokens_seen": 62954800,
      "step": 108505
    },
    {
      "epoch": 16.161751563896335,
      "grad_norm": 8.5,
      "learning_rate": 5.403320799112666e-06,
      "loss": 1.0583,
      "num_input_tokens_seen": 62957456,
      "step": 108510
    },
    {
      "epoch": 16.162496276437295,
      "grad_norm": 10.1875,
      "learning_rate": 5.401303306102326e-06,
      "loss": 1.2662,
      "num_input_tokens_seen": 62960272,
      "step": 108515
    },
    {
      "epoch": 16.163240988978256,
      "grad_norm": 6.0,
      "learning_rate": 5.3992861441920425e-06,
      "loss": 0.8274,
      "num_input_tokens_seen": 62963024,
      "step": 108520
    },
    {
      "epoch": 16.163985701519213,
      "grad_norm": 18.125,
      "learning_rate": 5.397269313415903e-06,
      "loss": 1.2034,
      "num_input_tokens_seen": 62966128,
      "step": 108525
    },
    {
      "epoch": 16.164730414060173,
      "grad_norm": 19.0,
      "learning_rate": 5.395252813807969e-06,
      "loss": 1.2774,
      "num_input_tokens_seen": 62969008,
      "step": 108530
    },
    {
      "epoch": 16.16547512660113,
      "grad_norm": 7.4375,
      "learning_rate": 5.39323664540232e-06,
      "loss": 0.9613,
      "num_input_tokens_seen": 62971888,
      "step": 108535
    },
    {
      "epoch": 16.16621983914209,
      "grad_norm": 12.1875,
      "learning_rate": 5.391220808233008e-06,
      "loss": 1.1271,
      "num_input_tokens_seen": 62974736,
      "step": 108540
    },
    {
      "epoch": 16.16696455168305,
      "grad_norm": 14.5625,
      "learning_rate": 5.3892053023340935e-06,
      "loss": 1.4621,
      "num_input_tokens_seen": 62977360,
      "step": 108545
    },
    {
      "epoch": 16.16770926422401,
      "grad_norm": 15.9375,
      "learning_rate": 5.387190127739625e-06,
      "loss": 1.3663,
      "num_input_tokens_seen": 62980016,
      "step": 108550
    },
    {
      "epoch": 16.16845397676497,
      "grad_norm": 11.3125,
      "learning_rate": 5.3851752844836374e-06,
      "loss": 1.2192,
      "num_input_tokens_seen": 62982960,
      "step": 108555
    },
    {
      "epoch": 16.16919868930593,
      "grad_norm": 6.75,
      "learning_rate": 5.383160772600185e-06,
      "loss": 1.0542,
      "num_input_tokens_seen": 62985904,
      "step": 108560
    },
    {
      "epoch": 16.169943401846886,
      "grad_norm": 11.9375,
      "learning_rate": 5.381146592123287e-06,
      "loss": 1.2788,
      "num_input_tokens_seen": 62988816,
      "step": 108565
    },
    {
      "epoch": 16.170688114387847,
      "grad_norm": 12.0,
      "learning_rate": 5.379132743086984e-06,
      "loss": 1.2178,
      "num_input_tokens_seen": 62991888,
      "step": 108570
    },
    {
      "epoch": 16.171432826928804,
      "grad_norm": 10.8125,
      "learning_rate": 5.377119225525284e-06,
      "loss": 1.3972,
      "num_input_tokens_seen": 62994608,
      "step": 108575
    },
    {
      "epoch": 16.172177539469764,
      "grad_norm": 11.5625,
      "learning_rate": 5.375106039472219e-06,
      "loss": 1.0516,
      "num_input_tokens_seen": 62997392,
      "step": 108580
    },
    {
      "epoch": 16.172922252010725,
      "grad_norm": 14.4375,
      "learning_rate": 5.373093184961783e-06,
      "loss": 1.6088,
      "num_input_tokens_seen": 63000432,
      "step": 108585
    },
    {
      "epoch": 16.173666964551682,
      "grad_norm": 12.5,
      "learning_rate": 5.371080662028e-06,
      "loss": 1.1496,
      "num_input_tokens_seen": 63003280,
      "step": 108590
    },
    {
      "epoch": 16.174411677092642,
      "grad_norm": 9.0,
      "learning_rate": 5.369068470704855e-06,
      "loss": 1.2817,
      "num_input_tokens_seen": 63005904,
      "step": 108595
    },
    {
      "epoch": 16.175156389633603,
      "grad_norm": 7.59375,
      "learning_rate": 5.367056611026341e-06,
      "loss": 1.437,
      "num_input_tokens_seen": 63008912,
      "step": 108600
    },
    {
      "epoch": 16.17590110217456,
      "grad_norm": 12.1875,
      "learning_rate": 5.36504508302646e-06,
      "loss": 1.1875,
      "num_input_tokens_seen": 63011760,
      "step": 108605
    },
    {
      "epoch": 16.17664581471552,
      "grad_norm": 14.8125,
      "learning_rate": 5.363033886739186e-06,
      "loss": 0.839,
      "num_input_tokens_seen": 63014512,
      "step": 108610
    },
    {
      "epoch": 16.177390527256478,
      "grad_norm": 6.9375,
      "learning_rate": 5.361023022198494e-06,
      "loss": 1.2401,
      "num_input_tokens_seen": 63017264,
      "step": 108615
    },
    {
      "epoch": 16.178135239797438,
      "grad_norm": 5.46875,
      "learning_rate": 5.359012489438353e-06,
      "loss": 0.7313,
      "num_input_tokens_seen": 63020144,
      "step": 108620
    },
    {
      "epoch": 16.1788799523384,
      "grad_norm": 11.875,
      "learning_rate": 5.357002288492741e-06,
      "loss": 0.7835,
      "num_input_tokens_seen": 63023184,
      "step": 108625
    },
    {
      "epoch": 16.179624664879356,
      "grad_norm": 6.28125,
      "learning_rate": 5.35499241939561e-06,
      "loss": 0.8925,
      "num_input_tokens_seen": 63025936,
      "step": 108630
    },
    {
      "epoch": 16.180369377420316,
      "grad_norm": 5.9375,
      "learning_rate": 5.3529828821809065e-06,
      "loss": 1.1552,
      "num_input_tokens_seen": 63028848,
      "step": 108635
    },
    {
      "epoch": 16.181114089961277,
      "grad_norm": 11.0,
      "learning_rate": 5.350973676882601e-06,
      "loss": 1.1263,
      "num_input_tokens_seen": 63031984,
      "step": 108640
    },
    {
      "epoch": 16.181858802502234,
      "grad_norm": 6.875,
      "learning_rate": 5.3489648035346144e-06,
      "loss": 1.7736,
      "num_input_tokens_seen": 63034768,
      "step": 108645
    },
    {
      "epoch": 16.182603515043194,
      "grad_norm": 19.875,
      "learning_rate": 5.346956262170902e-06,
      "loss": 1.2987,
      "num_input_tokens_seen": 63037616,
      "step": 108650
    },
    {
      "epoch": 16.18334822758415,
      "grad_norm": 7.78125,
      "learning_rate": 5.3449480528253825e-06,
      "loss": 1.077,
      "num_input_tokens_seen": 63040336,
      "step": 108655
    },
    {
      "epoch": 16.18409294012511,
      "grad_norm": 8.125,
      "learning_rate": 5.342940175531999e-06,
      "loss": 1.2145,
      "num_input_tokens_seen": 63043280,
      "step": 108660
    },
    {
      "epoch": 16.184837652666072,
      "grad_norm": 6.28125,
      "learning_rate": 5.3409326303246524e-06,
      "loss": 1.1002,
      "num_input_tokens_seen": 63045936,
      "step": 108665
    },
    {
      "epoch": 16.18558236520703,
      "grad_norm": 10.4375,
      "learning_rate": 5.338925417237275e-06,
      "loss": 1.1034,
      "num_input_tokens_seen": 63048784,
      "step": 108670
    },
    {
      "epoch": 16.18632707774799,
      "grad_norm": 7.3125,
      "learning_rate": 5.336918536303773e-06,
      "loss": 1.0479,
      "num_input_tokens_seen": 63051760,
      "step": 108675
    },
    {
      "epoch": 16.187071790288947,
      "grad_norm": 10.5,
      "learning_rate": 5.334911987558045e-06,
      "loss": 0.9561,
      "num_input_tokens_seen": 63054384,
      "step": 108680
    },
    {
      "epoch": 16.187816502829907,
      "grad_norm": 11.3125,
      "learning_rate": 5.332905771033994e-06,
      "loss": 1.4093,
      "num_input_tokens_seen": 63057040,
      "step": 108685
    },
    {
      "epoch": 16.188561215370868,
      "grad_norm": 5.53125,
      "learning_rate": 5.330899886765503e-06,
      "loss": 1.1578,
      "num_input_tokens_seen": 63059728,
      "step": 108690
    },
    {
      "epoch": 16.189305927911825,
      "grad_norm": 13.3125,
      "learning_rate": 5.328894334786474e-06,
      "loss": 0.9883,
      "num_input_tokens_seen": 63062544,
      "step": 108695
    },
    {
      "epoch": 16.190050640452785,
      "grad_norm": 11.0,
      "learning_rate": 5.326889115130779e-06,
      "loss": 1.2891,
      "num_input_tokens_seen": 63065712,
      "step": 108700
    },
    {
      "epoch": 16.190795352993746,
      "grad_norm": 18.125,
      "learning_rate": 5.324884227832302e-06,
      "loss": 1.1484,
      "num_input_tokens_seen": 63068336,
      "step": 108705
    },
    {
      "epoch": 16.191540065534703,
      "grad_norm": 10.875,
      "learning_rate": 5.322879672924908e-06,
      "loss": 1.1761,
      "num_input_tokens_seen": 63071376,
      "step": 108710
    },
    {
      "epoch": 16.192284778075663,
      "grad_norm": 7.96875,
      "learning_rate": 5.3208754504424585e-06,
      "loss": 1.0142,
      "num_input_tokens_seen": 63074160,
      "step": 108715
    },
    {
      "epoch": 16.19302949061662,
      "grad_norm": 9.875,
      "learning_rate": 5.318871560418822e-06,
      "loss": 1.2938,
      "num_input_tokens_seen": 63076880,
      "step": 108720
    },
    {
      "epoch": 16.19377420315758,
      "grad_norm": 5.84375,
      "learning_rate": 5.316868002887843e-06,
      "loss": 1.2732,
      "num_input_tokens_seen": 63079600,
      "step": 108725
    },
    {
      "epoch": 16.19451891569854,
      "grad_norm": 14.6875,
      "learning_rate": 5.314864777883377e-06,
      "loss": 1.195,
      "num_input_tokens_seen": 63082544,
      "step": 108730
    },
    {
      "epoch": 16.1952636282395,
      "grad_norm": 12.4375,
      "learning_rate": 5.3128618854392655e-06,
      "loss": 0.7835,
      "num_input_tokens_seen": 63085840,
      "step": 108735
    },
    {
      "epoch": 16.19600834078046,
      "grad_norm": 6.46875,
      "learning_rate": 5.3108593255893376e-06,
      "loss": 0.997,
      "num_input_tokens_seen": 63088400,
      "step": 108740
    },
    {
      "epoch": 16.19675305332142,
      "grad_norm": 11.9375,
      "learning_rate": 5.308857098367437e-06,
      "loss": 0.9905,
      "num_input_tokens_seen": 63091120,
      "step": 108745
    },
    {
      "epoch": 16.197497765862376,
      "grad_norm": 17.5,
      "learning_rate": 5.306855203807382e-06,
      "loss": 0.95,
      "num_input_tokens_seen": 63094096,
      "step": 108750
    },
    {
      "epoch": 16.198242478403337,
      "grad_norm": 9.1875,
      "learning_rate": 5.304853641942995e-06,
      "loss": 0.9916,
      "num_input_tokens_seen": 63096976,
      "step": 108755
    },
    {
      "epoch": 16.198987190944294,
      "grad_norm": 8.5,
      "learning_rate": 5.302852412808079e-06,
      "loss": 1.2705,
      "num_input_tokens_seen": 63099568,
      "step": 108760
    },
    {
      "epoch": 16.199731903485254,
      "grad_norm": 14.8125,
      "learning_rate": 5.3008515164364585e-06,
      "loss": 1.1993,
      "num_input_tokens_seen": 63102480,
      "step": 108765
    },
    {
      "epoch": 16.200476616026215,
      "grad_norm": 5.9375,
      "learning_rate": 5.298850952861925e-06,
      "loss": 0.718,
      "num_input_tokens_seen": 63105392,
      "step": 108770
    },
    {
      "epoch": 16.201221328567172,
      "grad_norm": 13.125,
      "learning_rate": 5.296850722118288e-06,
      "loss": 1.1049,
      "num_input_tokens_seen": 63108240,
      "step": 108775
    },
    {
      "epoch": 16.201966041108133,
      "grad_norm": 10.1875,
      "learning_rate": 5.294850824239325e-06,
      "loss": 0.9008,
      "num_input_tokens_seen": 63111344,
      "step": 108780
    },
    {
      "epoch": 16.202710753649093,
      "grad_norm": 13.125,
      "learning_rate": 5.292851259258838e-06,
      "loss": 0.7036,
      "num_input_tokens_seen": 63114384,
      "step": 108785
    },
    {
      "epoch": 16.20345546619005,
      "grad_norm": 15.9375,
      "learning_rate": 5.2908520272106e-06,
      "loss": 1.3744,
      "num_input_tokens_seen": 63117296,
      "step": 108790
    },
    {
      "epoch": 16.20420017873101,
      "grad_norm": 8.1875,
      "learning_rate": 5.288853128128377e-06,
      "loss": 1.0856,
      "num_input_tokens_seen": 63120016,
      "step": 108795
    },
    {
      "epoch": 16.204944891271968,
      "grad_norm": 33.0,
      "learning_rate": 5.2868545620459535e-06,
      "loss": 1.295,
      "num_input_tokens_seen": 63122800,
      "step": 108800
    },
    {
      "epoch": 16.205689603812928,
      "grad_norm": 11.4375,
      "learning_rate": 5.284856328997087e-06,
      "loss": 1.5545,
      "num_input_tokens_seen": 63125552,
      "step": 108805
    },
    {
      "epoch": 16.20643431635389,
      "grad_norm": 11.5625,
      "learning_rate": 5.282858429015536e-06,
      "loss": 0.8593,
      "num_input_tokens_seen": 63128400,
      "step": 108810
    },
    {
      "epoch": 16.207179028894846,
      "grad_norm": 16.0,
      "learning_rate": 5.280860862135045e-06,
      "loss": 0.8188,
      "num_input_tokens_seen": 63131376,
      "step": 108815
    },
    {
      "epoch": 16.207923741435806,
      "grad_norm": 12.625,
      "learning_rate": 5.278863628389377e-06,
      "loss": 0.9869,
      "num_input_tokens_seen": 63134320,
      "step": 108820
    },
    {
      "epoch": 16.208668453976767,
      "grad_norm": 15.8125,
      "learning_rate": 5.276866727812255e-06,
      "loss": 1.3833,
      "num_input_tokens_seen": 63137584,
      "step": 108825
    },
    {
      "epoch": 16.209413166517724,
      "grad_norm": 9.9375,
      "learning_rate": 5.274870160437431e-06,
      "loss": 1.1945,
      "num_input_tokens_seen": 63140272,
      "step": 108830
    },
    {
      "epoch": 16.210157879058684,
      "grad_norm": 14.75,
      "learning_rate": 5.272873926298627e-06,
      "loss": 1.1835,
      "num_input_tokens_seen": 63143376,
      "step": 108835
    },
    {
      "epoch": 16.21090259159964,
      "grad_norm": 9.8125,
      "learning_rate": 5.270878025429565e-06,
      "loss": 1.2355,
      "num_input_tokens_seen": 63145936,
      "step": 108840
    },
    {
      "epoch": 16.2116473041406,
      "grad_norm": 8.375,
      "learning_rate": 5.268882457863972e-06,
      "loss": 1.2672,
      "num_input_tokens_seen": 63148912,
      "step": 108845
    },
    {
      "epoch": 16.212392016681562,
      "grad_norm": 6.28125,
      "learning_rate": 5.266887223635547e-06,
      "loss": 0.8718,
      "num_input_tokens_seen": 63151600,
      "step": 108850
    },
    {
      "epoch": 16.21313672922252,
      "grad_norm": 5.4375,
      "learning_rate": 5.264892322778014e-06,
      "loss": 0.9657,
      "num_input_tokens_seen": 63154512,
      "step": 108855
    },
    {
      "epoch": 16.21388144176348,
      "grad_norm": 8.9375,
      "learning_rate": 5.262897755325064e-06,
      "loss": 1.0254,
      "num_input_tokens_seen": 63157328,
      "step": 108860
    },
    {
      "epoch": 16.214626154304437,
      "grad_norm": 12.9375,
      "learning_rate": 5.260903521310401e-06,
      "loss": 1.2007,
      "num_input_tokens_seen": 63160592,
      "step": 108865
    },
    {
      "epoch": 16.215370866845397,
      "grad_norm": 15.375,
      "learning_rate": 5.25890962076771e-06,
      "loss": 1.451,
      "num_input_tokens_seen": 63163696,
      "step": 108870
    },
    {
      "epoch": 16.216115579386358,
      "grad_norm": 6.8125,
      "learning_rate": 5.256916053730679e-06,
      "loss": 0.8357,
      "num_input_tokens_seen": 63166576,
      "step": 108875
    },
    {
      "epoch": 16.216860291927315,
      "grad_norm": 4.25,
      "learning_rate": 5.254922820232983e-06,
      "loss": 0.9948,
      "num_input_tokens_seen": 63169392,
      "step": 108880
    },
    {
      "epoch": 16.217605004468275,
      "grad_norm": 9.5,
      "learning_rate": 5.2529299203082914e-06,
      "loss": 0.9791,
      "num_input_tokens_seen": 63172176,
      "step": 108885
    },
    {
      "epoch": 16.218349717009236,
      "grad_norm": 10.75,
      "learning_rate": 5.250937353990288e-06,
      "loss": 1.1185,
      "num_input_tokens_seen": 63175120,
      "step": 108890
    },
    {
      "epoch": 16.219094429550193,
      "grad_norm": 12.0625,
      "learning_rate": 5.248945121312618e-06,
      "loss": 0.9016,
      "num_input_tokens_seen": 63178032,
      "step": 108895
    },
    {
      "epoch": 16.219839142091153,
      "grad_norm": 9.8125,
      "learning_rate": 5.246953222308953e-06,
      "loss": 0.9939,
      "num_input_tokens_seen": 63181072,
      "step": 108900
    },
    {
      "epoch": 16.22058385463211,
      "grad_norm": 9.25,
      "learning_rate": 5.244961657012928e-06,
      "loss": 0.7976,
      "num_input_tokens_seen": 63183728,
      "step": 108905
    },
    {
      "epoch": 16.22132856717307,
      "grad_norm": 8.625,
      "learning_rate": 5.242970425458208e-06,
      "loss": 1.0731,
      "num_input_tokens_seen": 63187024,
      "step": 108910
    },
    {
      "epoch": 16.22207327971403,
      "grad_norm": 9.3125,
      "learning_rate": 5.240979527678422e-06,
      "loss": 1.219,
      "num_input_tokens_seen": 63190000,
      "step": 108915
    },
    {
      "epoch": 16.22281799225499,
      "grad_norm": 13.125,
      "learning_rate": 5.238988963707195e-06,
      "loss": 1.0523,
      "num_input_tokens_seen": 63192944,
      "step": 108920
    },
    {
      "epoch": 16.22356270479595,
      "grad_norm": 9.8125,
      "learning_rate": 5.236998733578175e-06,
      "loss": 0.8907,
      "num_input_tokens_seen": 63195760,
      "step": 108925
    },
    {
      "epoch": 16.22430741733691,
      "grad_norm": 9.0,
      "learning_rate": 5.235008837324967e-06,
      "loss": 0.934,
      "num_input_tokens_seen": 63198864,
      "step": 108930
    },
    {
      "epoch": 16.225052129877866,
      "grad_norm": 12.75,
      "learning_rate": 5.233019274981205e-06,
      "loss": 1.1424,
      "num_input_tokens_seen": 63201680,
      "step": 108935
    },
    {
      "epoch": 16.225796842418827,
      "grad_norm": 13.25,
      "learning_rate": 5.23103004658049e-06,
      "loss": 0.9467,
      "num_input_tokens_seen": 63204560,
      "step": 108940
    },
    {
      "epoch": 16.226541554959784,
      "grad_norm": 9.5625,
      "learning_rate": 5.2290411521564305e-06,
      "loss": 1.1004,
      "num_input_tokens_seen": 63207216,
      "step": 108945
    },
    {
      "epoch": 16.227286267500745,
      "grad_norm": 11.9375,
      "learning_rate": 5.227052591742626e-06,
      "loss": 0.9497,
      "num_input_tokens_seen": 63210288,
      "step": 108950
    },
    {
      "epoch": 16.228030980041705,
      "grad_norm": 18.75,
      "learning_rate": 5.225064365372667e-06,
      "loss": 1.4746,
      "num_input_tokens_seen": 63213168,
      "step": 108955
    },
    {
      "epoch": 16.228775692582662,
      "grad_norm": 26.625,
      "learning_rate": 5.223076473080152e-06,
      "loss": 1.2122,
      "num_input_tokens_seen": 63216080,
      "step": 108960
    },
    {
      "epoch": 16.229520405123623,
      "grad_norm": 8.6875,
      "learning_rate": 5.221088914898653e-06,
      "loss": 1.1646,
      "num_input_tokens_seen": 63218960,
      "step": 108965
    },
    {
      "epoch": 16.230265117664583,
      "grad_norm": 9.3125,
      "learning_rate": 5.219101690861763e-06,
      "loss": 1.0687,
      "num_input_tokens_seen": 63222160,
      "step": 108970
    },
    {
      "epoch": 16.23100983020554,
      "grad_norm": 13.0625,
      "learning_rate": 5.217114801003037e-06,
      "loss": 0.8951,
      "num_input_tokens_seen": 63224912,
      "step": 108975
    },
    {
      "epoch": 16.2317545427465,
      "grad_norm": 20.75,
      "learning_rate": 5.215128245356057e-06,
      "loss": 1.025,
      "num_input_tokens_seen": 63227920,
      "step": 108980
    },
    {
      "epoch": 16.232499255287458,
      "grad_norm": 6.625,
      "learning_rate": 5.2131420239543704e-06,
      "loss": 0.8645,
      "num_input_tokens_seen": 63230768,
      "step": 108985
    },
    {
      "epoch": 16.233243967828418,
      "grad_norm": 11.0,
      "learning_rate": 5.211156136831546e-06,
      "loss": 1.2304,
      "num_input_tokens_seen": 63233648,
      "step": 108990
    },
    {
      "epoch": 16.23398868036938,
      "grad_norm": 12.1875,
      "learning_rate": 5.209170584021125e-06,
      "loss": 1.1347,
      "num_input_tokens_seen": 63236496,
      "step": 108995
    },
    {
      "epoch": 16.234733392910336,
      "grad_norm": 27.5,
      "learning_rate": 5.207185365556646e-06,
      "loss": 1.6009,
      "num_input_tokens_seen": 63239280,
      "step": 109000
    },
    {
      "epoch": 16.235478105451296,
      "grad_norm": 8.6875,
      "learning_rate": 5.205200481471662e-06,
      "loss": 0.8522,
      "num_input_tokens_seen": 63241840,
      "step": 109005
    },
    {
      "epoch": 16.236222817992257,
      "grad_norm": 16.875,
      "learning_rate": 5.2032159317996955e-06,
      "loss": 1.1781,
      "num_input_tokens_seen": 63244848,
      "step": 109010
    },
    {
      "epoch": 16.236967530533214,
      "grad_norm": 15.9375,
      "learning_rate": 5.201231716574276e-06,
      "loss": 1.2422,
      "num_input_tokens_seen": 63247696,
      "step": 109015
    },
    {
      "epoch": 16.237712243074174,
      "grad_norm": 16.0,
      "learning_rate": 5.199247835828916e-06,
      "loss": 0.8675,
      "num_input_tokens_seen": 63250704,
      "step": 109020
    },
    {
      "epoch": 16.23845695561513,
      "grad_norm": 4.28125,
      "learning_rate": 5.197264289597148e-06,
      "loss": 1.5181,
      "num_input_tokens_seen": 63253456,
      "step": 109025
    },
    {
      "epoch": 16.239201668156092,
      "grad_norm": 10.75,
      "learning_rate": 5.195281077912473e-06,
      "loss": 1.1828,
      "num_input_tokens_seen": 63256336,
      "step": 109030
    },
    {
      "epoch": 16.239946380697052,
      "grad_norm": 15.625,
      "learning_rate": 5.193298200808389e-06,
      "loss": 1.0127,
      "num_input_tokens_seen": 63259312,
      "step": 109035
    },
    {
      "epoch": 16.24069109323801,
      "grad_norm": 15.1875,
      "learning_rate": 5.191315658318408e-06,
      "loss": 0.7684,
      "num_input_tokens_seen": 63262256,
      "step": 109040
    },
    {
      "epoch": 16.24143580577897,
      "grad_norm": 13.9375,
      "learning_rate": 5.189333450476008e-06,
      "loss": 0.8798,
      "num_input_tokens_seen": 63264976,
      "step": 109045
    },
    {
      "epoch": 16.242180518319927,
      "grad_norm": 12.8125,
      "learning_rate": 5.187351577314692e-06,
      "loss": 0.8542,
      "num_input_tokens_seen": 63267888,
      "step": 109050
    },
    {
      "epoch": 16.242925230860887,
      "grad_norm": 12.75,
      "learning_rate": 5.185370038867929e-06,
      "loss": 1.1909,
      "num_input_tokens_seen": 63270832,
      "step": 109055
    },
    {
      "epoch": 16.243669943401848,
      "grad_norm": 13.875,
      "learning_rate": 5.183388835169206e-06,
      "loss": 1.3973,
      "num_input_tokens_seen": 63273488,
      "step": 109060
    },
    {
      "epoch": 16.244414655942805,
      "grad_norm": 10.375,
      "learning_rate": 5.181407966251986e-06,
      "loss": 1.4397,
      "num_input_tokens_seen": 63276368,
      "step": 109065
    },
    {
      "epoch": 16.245159368483765,
      "grad_norm": 15.875,
      "learning_rate": 5.179427432149733e-06,
      "loss": 1.3154,
      "num_input_tokens_seen": 63279504,
      "step": 109070
    },
    {
      "epoch": 16.245904081024726,
      "grad_norm": 7.25,
      "learning_rate": 5.177447232895913e-06,
      "loss": 1.25,
      "num_input_tokens_seen": 63282352,
      "step": 109075
    },
    {
      "epoch": 16.246648793565683,
      "grad_norm": 12.125,
      "learning_rate": 5.1754673685239755e-06,
      "loss": 0.9437,
      "num_input_tokens_seen": 63285072,
      "step": 109080
    },
    {
      "epoch": 16.247393506106643,
      "grad_norm": 20.125,
      "learning_rate": 5.173487839067371e-06,
      "loss": 0.978,
      "num_input_tokens_seen": 63287760,
      "step": 109085
    },
    {
      "epoch": 16.2481382186476,
      "grad_norm": 14.5,
      "learning_rate": 5.171508644559528e-06,
      "loss": 1.2751,
      "num_input_tokens_seen": 63290896,
      "step": 109090
    },
    {
      "epoch": 16.24888293118856,
      "grad_norm": 11.6875,
      "learning_rate": 5.169529785033903e-06,
      "loss": 1.099,
      "num_input_tokens_seen": 63293648,
      "step": 109095
    },
    {
      "epoch": 16.24962764372952,
      "grad_norm": 16.25,
      "learning_rate": 5.167551260523909e-06,
      "loss": 1.0121,
      "num_input_tokens_seen": 63296336,
      "step": 109100
    },
    {
      "epoch": 16.25037235627048,
      "grad_norm": 14.75,
      "learning_rate": 5.165573071062985e-06,
      "loss": 0.7603,
      "num_input_tokens_seen": 63298864,
      "step": 109105
    },
    {
      "epoch": 16.25111706881144,
      "grad_norm": 12.625,
      "learning_rate": 5.163595216684541e-06,
      "loss": 1.1518,
      "num_input_tokens_seen": 63302032,
      "step": 109110
    },
    {
      "epoch": 16.2518617813524,
      "grad_norm": 11.0625,
      "learning_rate": 5.161617697422003e-06,
      "loss": 0.778,
      "num_input_tokens_seen": 63304784,
      "step": 109115
    },
    {
      "epoch": 16.252606493893357,
      "grad_norm": 6.71875,
      "learning_rate": 5.159640513308767e-06,
      "loss": 0.9827,
      "num_input_tokens_seen": 63307728,
      "step": 109120
    },
    {
      "epoch": 16.253351206434317,
      "grad_norm": 30.375,
      "learning_rate": 5.1576636643782376e-06,
      "loss": 1.2375,
      "num_input_tokens_seen": 63310448,
      "step": 109125
    },
    {
      "epoch": 16.254095918975274,
      "grad_norm": 8.4375,
      "learning_rate": 5.155687150663815e-06,
      "loss": 1.6137,
      "num_input_tokens_seen": 63313232,
      "step": 109130
    },
    {
      "epoch": 16.254840631516235,
      "grad_norm": 7.59375,
      "learning_rate": 5.153710972198894e-06,
      "loss": 0.8455,
      "num_input_tokens_seen": 63316272,
      "step": 109135
    },
    {
      "epoch": 16.255585344057195,
      "grad_norm": 7.1875,
      "learning_rate": 5.151735129016855e-06,
      "loss": 0.8053,
      "num_input_tokens_seen": 63319280,
      "step": 109140
    },
    {
      "epoch": 16.256330056598152,
      "grad_norm": 12.3125,
      "learning_rate": 5.149759621151068e-06,
      "loss": 0.936,
      "num_input_tokens_seen": 63322256,
      "step": 109145
    },
    {
      "epoch": 16.257074769139113,
      "grad_norm": 6.5,
      "learning_rate": 5.147784448634926e-06,
      "loss": 1.1891,
      "num_input_tokens_seen": 63325360,
      "step": 109150
    },
    {
      "epoch": 16.257819481680073,
      "grad_norm": 10.4375,
      "learning_rate": 5.145809611501789e-06,
      "loss": 0.9323,
      "num_input_tokens_seen": 63328304,
      "step": 109155
    },
    {
      "epoch": 16.25856419422103,
      "grad_norm": 9.0625,
      "learning_rate": 5.143835109785014e-06,
      "loss": 1.0565,
      "num_input_tokens_seen": 63331088,
      "step": 109160
    },
    {
      "epoch": 16.25930890676199,
      "grad_norm": 6.90625,
      "learning_rate": 5.1418609435179676e-06,
      "loss": 1.1073,
      "num_input_tokens_seen": 63333872,
      "step": 109165
    },
    {
      "epoch": 16.260053619302948,
      "grad_norm": 8.75,
      "learning_rate": 5.139887112733993e-06,
      "loss": 0.8748,
      "num_input_tokens_seen": 63336400,
      "step": 109170
    },
    {
      "epoch": 16.260798331843908,
      "grad_norm": 15.25,
      "learning_rate": 5.137913617466447e-06,
      "loss": 1.7161,
      "num_input_tokens_seen": 63339184,
      "step": 109175
    },
    {
      "epoch": 16.26154304438487,
      "grad_norm": 3.796875,
      "learning_rate": 5.1359404577486585e-06,
      "loss": 0.7637,
      "num_input_tokens_seen": 63342032,
      "step": 109180
    },
    {
      "epoch": 16.262287756925826,
      "grad_norm": 5.125,
      "learning_rate": 5.13396763361397e-06,
      "loss": 0.935,
      "num_input_tokens_seen": 63345232,
      "step": 109185
    },
    {
      "epoch": 16.263032469466786,
      "grad_norm": 21.75,
      "learning_rate": 5.131995145095705e-06,
      "loss": 1.0555,
      "num_input_tokens_seen": 63348016,
      "step": 109190
    },
    {
      "epoch": 16.263777182007743,
      "grad_norm": 12.75,
      "learning_rate": 5.130022992227193e-06,
      "loss": 1.049,
      "num_input_tokens_seen": 63350672,
      "step": 109195
    },
    {
      "epoch": 16.264521894548704,
      "grad_norm": 13.1875,
      "learning_rate": 5.128051175041748e-06,
      "loss": 1.1337,
      "num_input_tokens_seen": 63353584,
      "step": 109200
    },
    {
      "epoch": 16.265266607089664,
      "grad_norm": 9.125,
      "learning_rate": 5.126079693572683e-06,
      "loss": 1.1034,
      "num_input_tokens_seen": 63356720,
      "step": 109205
    },
    {
      "epoch": 16.26601131963062,
      "grad_norm": 19.875,
      "learning_rate": 5.124108547853301e-06,
      "loss": 1.4037,
      "num_input_tokens_seen": 63359472,
      "step": 109210
    },
    {
      "epoch": 16.266756032171582,
      "grad_norm": 7.4375,
      "learning_rate": 5.122137737916896e-06,
      "loss": 1.1119,
      "num_input_tokens_seen": 63362832,
      "step": 109215
    },
    {
      "epoch": 16.267500744712542,
      "grad_norm": 10.8125,
      "learning_rate": 5.120167263796779e-06,
      "loss": 0.8258,
      "num_input_tokens_seen": 63365712,
      "step": 109220
    },
    {
      "epoch": 16.2682454572535,
      "grad_norm": 21.5,
      "learning_rate": 5.118197125526228e-06,
      "loss": 1.5018,
      "num_input_tokens_seen": 63368560,
      "step": 109225
    },
    {
      "epoch": 16.26899016979446,
      "grad_norm": 12.125,
      "learning_rate": 5.116227323138531e-06,
      "loss": 1.0057,
      "num_input_tokens_seen": 63371600,
      "step": 109230
    },
    {
      "epoch": 16.269734882335417,
      "grad_norm": 10.75,
      "learning_rate": 5.114257856666968e-06,
      "loss": 0.8872,
      "num_input_tokens_seen": 63374320,
      "step": 109235
    },
    {
      "epoch": 16.270479594876377,
      "grad_norm": 13.5,
      "learning_rate": 5.112288726144798e-06,
      "loss": 1.2073,
      "num_input_tokens_seen": 63377392,
      "step": 109240
    },
    {
      "epoch": 16.271224307417338,
      "grad_norm": 6.375,
      "learning_rate": 5.110319931605306e-06,
      "loss": 0.985,
      "num_input_tokens_seen": 63380464,
      "step": 109245
    },
    {
      "epoch": 16.271969019958295,
      "grad_norm": 9.75,
      "learning_rate": 5.1083514730817375e-06,
      "loss": 1.0986,
      "num_input_tokens_seen": 63383696,
      "step": 109250
    },
    {
      "epoch": 16.272713732499255,
      "grad_norm": 8.125,
      "learning_rate": 5.106383350607358e-06,
      "loss": 0.9423,
      "num_input_tokens_seen": 63386384,
      "step": 109255
    },
    {
      "epoch": 16.273458445040216,
      "grad_norm": 14.6875,
      "learning_rate": 5.104415564215409e-06,
      "loss": 1.4138,
      "num_input_tokens_seen": 63389104,
      "step": 109260
    },
    {
      "epoch": 16.274203157581173,
      "grad_norm": 8.875,
      "learning_rate": 5.102448113939143e-06,
      "loss": 1.2105,
      "num_input_tokens_seen": 63391856,
      "step": 109265
    },
    {
      "epoch": 16.274947870122134,
      "grad_norm": 5.4375,
      "learning_rate": 5.100480999811794e-06,
      "loss": 0.9993,
      "num_input_tokens_seen": 63394544,
      "step": 109270
    },
    {
      "epoch": 16.27569258266309,
      "grad_norm": 8.875,
      "learning_rate": 5.09851422186659e-06,
      "loss": 1.1501,
      "num_input_tokens_seen": 63397808,
      "step": 109275
    },
    {
      "epoch": 16.27643729520405,
      "grad_norm": 6.65625,
      "learning_rate": 5.096547780136765e-06,
      "loss": 1.0751,
      "num_input_tokens_seen": 63401040,
      "step": 109280
    },
    {
      "epoch": 16.27718200774501,
      "grad_norm": 12.5,
      "learning_rate": 5.0945816746555295e-06,
      "loss": 0.9722,
      "num_input_tokens_seen": 63403888,
      "step": 109285
    },
    {
      "epoch": 16.27792672028597,
      "grad_norm": 4.4375,
      "learning_rate": 5.092615905456111e-06,
      "loss": 1.0859,
      "num_input_tokens_seen": 63406608,
      "step": 109290
    },
    {
      "epoch": 16.27867143282693,
      "grad_norm": 7.5,
      "learning_rate": 5.090650472571709e-06,
      "loss": 1.3325,
      "num_input_tokens_seen": 63409840,
      "step": 109295
    },
    {
      "epoch": 16.27941614536789,
      "grad_norm": 9.9375,
      "learning_rate": 5.088685376035538e-06,
      "loss": 0.849,
      "num_input_tokens_seen": 63412528,
      "step": 109300
    },
    {
      "epoch": 16.280160857908847,
      "grad_norm": 6.53125,
      "learning_rate": 5.086720615880783e-06,
      "loss": 0.7579,
      "num_input_tokens_seen": 63415376,
      "step": 109305
    },
    {
      "epoch": 16.280905570449807,
      "grad_norm": 14.4375,
      "learning_rate": 5.084756192140652e-06,
      "loss": 0.9689,
      "num_input_tokens_seen": 63418096,
      "step": 109310
    },
    {
      "epoch": 16.281650282990764,
      "grad_norm": 18.5,
      "learning_rate": 5.082792104848325e-06,
      "loss": 1.1783,
      "num_input_tokens_seen": 63421328,
      "step": 109315
    },
    {
      "epoch": 16.282394995531725,
      "grad_norm": 7.59375,
      "learning_rate": 5.080828354036974e-06,
      "loss": 1.0977,
      "num_input_tokens_seen": 63424464,
      "step": 109320
    },
    {
      "epoch": 16.283139708072685,
      "grad_norm": 13.25,
      "learning_rate": 5.078864939739789e-06,
      "loss": 1.2914,
      "num_input_tokens_seen": 63427184,
      "step": 109325
    },
    {
      "epoch": 16.283884420613642,
      "grad_norm": 7.625,
      "learning_rate": 5.076901861989927e-06,
      "loss": 0.8389,
      "num_input_tokens_seen": 63430128,
      "step": 109330
    },
    {
      "epoch": 16.284629133154603,
      "grad_norm": 2.890625,
      "learning_rate": 5.074939120820568e-06,
      "loss": 1.1445,
      "num_input_tokens_seen": 63433008,
      "step": 109335
    },
    {
      "epoch": 16.285373845695563,
      "grad_norm": 13.4375,
      "learning_rate": 5.072976716264863e-06,
      "loss": 1.0696,
      "num_input_tokens_seen": 63435888,
      "step": 109340
    },
    {
      "epoch": 16.28611855823652,
      "grad_norm": 7.65625,
      "learning_rate": 5.07101464835596e-06,
      "loss": 1.1099,
      "num_input_tokens_seen": 63438864,
      "step": 109345
    },
    {
      "epoch": 16.28686327077748,
      "grad_norm": 16.5,
      "learning_rate": 5.069052917127004e-06,
      "loss": 1.1796,
      "num_input_tokens_seen": 63442064,
      "step": 109350
    },
    {
      "epoch": 16.287607983318438,
      "grad_norm": 9.9375,
      "learning_rate": 5.06709152261115e-06,
      "loss": 0.9619,
      "num_input_tokens_seen": 63445264,
      "step": 109355
    },
    {
      "epoch": 16.2883526958594,
      "grad_norm": 17.75,
      "learning_rate": 5.065130464841525e-06,
      "loss": 1.0248,
      "num_input_tokens_seen": 63448208,
      "step": 109360
    },
    {
      "epoch": 16.28909740840036,
      "grad_norm": 5.0625,
      "learning_rate": 5.063169743851251e-06,
      "loss": 1.2044,
      "num_input_tokens_seen": 63451152,
      "step": 109365
    },
    {
      "epoch": 16.289842120941316,
      "grad_norm": 14.9375,
      "learning_rate": 5.061209359673471e-06,
      "loss": 1.0057,
      "num_input_tokens_seen": 63454160,
      "step": 109370
    },
    {
      "epoch": 16.290586833482276,
      "grad_norm": 9.25,
      "learning_rate": 5.059249312341286e-06,
      "loss": 1.1008,
      "num_input_tokens_seen": 63456944,
      "step": 109375
    },
    {
      "epoch": 16.291331546023233,
      "grad_norm": 11.5625,
      "learning_rate": 5.057289601887824e-06,
      "loss": 0.9352,
      "num_input_tokens_seen": 63459632,
      "step": 109380
    },
    {
      "epoch": 16.292076258564194,
      "grad_norm": 6.8125,
      "learning_rate": 5.055330228346178e-06,
      "loss": 0.914,
      "num_input_tokens_seen": 63462608,
      "step": 109385
    },
    {
      "epoch": 16.292820971105154,
      "grad_norm": 18.625,
      "learning_rate": 5.053371191749465e-06,
      "loss": 1.2764,
      "num_input_tokens_seen": 63465712,
      "step": 109390
    },
    {
      "epoch": 16.29356568364611,
      "grad_norm": 6.9375,
      "learning_rate": 5.051412492130772e-06,
      "loss": 0.8798,
      "num_input_tokens_seen": 63468432,
      "step": 109395
    },
    {
      "epoch": 16.294310396187072,
      "grad_norm": 8.5,
      "learning_rate": 5.049454129523185e-06,
      "loss": 1.219,
      "num_input_tokens_seen": 63471248,
      "step": 109400
    },
    {
      "epoch": 16.295055108728032,
      "grad_norm": 9.625,
      "learning_rate": 5.047496103959798e-06,
      "loss": 1.1427,
      "num_input_tokens_seen": 63473776,
      "step": 109405
    },
    {
      "epoch": 16.29579982126899,
      "grad_norm": 9.25,
      "learning_rate": 5.045538415473686e-06,
      "loss": 0.9531,
      "num_input_tokens_seen": 63476592,
      "step": 109410
    },
    {
      "epoch": 16.29654453380995,
      "grad_norm": 12.9375,
      "learning_rate": 5.0435810640979215e-06,
      "loss": 0.9763,
      "num_input_tokens_seen": 63479472,
      "step": 109415
    },
    {
      "epoch": 16.297289246350907,
      "grad_norm": 13.875,
      "learning_rate": 5.041624049865567e-06,
      "loss": 1.083,
      "num_input_tokens_seen": 63482576,
      "step": 109420
    },
    {
      "epoch": 16.298033958891867,
      "grad_norm": 6.03125,
      "learning_rate": 5.039667372809695e-06,
      "loss": 1.3078,
      "num_input_tokens_seen": 63485392,
      "step": 109425
    },
    {
      "epoch": 16.298778671432828,
      "grad_norm": 22.0,
      "learning_rate": 5.0377110329633495e-06,
      "loss": 1.0781,
      "num_input_tokens_seen": 63488304,
      "step": 109430
    },
    {
      "epoch": 16.299523383973785,
      "grad_norm": 7.25,
      "learning_rate": 5.035755030359593e-06,
      "loss": 0.8676,
      "num_input_tokens_seen": 63491312,
      "step": 109435
    },
    {
      "epoch": 16.300268096514746,
      "grad_norm": 14.625,
      "learning_rate": 5.0337993650314665e-06,
      "loss": 0.9687,
      "num_input_tokens_seen": 63494256,
      "step": 109440
    },
    {
      "epoch": 16.301012809055706,
      "grad_norm": 7.1875,
      "learning_rate": 5.0318440370119985e-06,
      "loss": 1.2656,
      "num_input_tokens_seen": 63497168,
      "step": 109445
    },
    {
      "epoch": 16.301757521596663,
      "grad_norm": 11.25,
      "learning_rate": 5.029889046334238e-06,
      "loss": 0.8882,
      "num_input_tokens_seen": 63500048,
      "step": 109450
    },
    {
      "epoch": 16.302502234137624,
      "grad_norm": 11.8125,
      "learning_rate": 5.0279343930312e-06,
      "loss": 1.0693,
      "num_input_tokens_seen": 63503216,
      "step": 109455
    },
    {
      "epoch": 16.30324694667858,
      "grad_norm": 9.5625,
      "learning_rate": 5.025980077135917e-06,
      "loss": 1.0199,
      "num_input_tokens_seen": 63506256,
      "step": 109460
    },
    {
      "epoch": 16.30399165921954,
      "grad_norm": 14.5,
      "learning_rate": 5.0240260986814e-06,
      "loss": 1.2532,
      "num_input_tokens_seen": 63509136,
      "step": 109465
    },
    {
      "epoch": 16.3047363717605,
      "grad_norm": 11.0625,
      "learning_rate": 5.022072457700658e-06,
      "loss": 1.0694,
      "num_input_tokens_seen": 63511952,
      "step": 109470
    },
    {
      "epoch": 16.30548108430146,
      "grad_norm": 8.5625,
      "learning_rate": 5.020119154226699e-06,
      "loss": 0.9511,
      "num_input_tokens_seen": 63515024,
      "step": 109475
    },
    {
      "epoch": 16.30622579684242,
      "grad_norm": 7.28125,
      "learning_rate": 5.018166188292514e-06,
      "loss": 1.1418,
      "num_input_tokens_seen": 63518000,
      "step": 109480
    },
    {
      "epoch": 16.30697050938338,
      "grad_norm": 6.4375,
      "learning_rate": 5.016213559931107e-06,
      "loss": 1.2947,
      "num_input_tokens_seen": 63520880,
      "step": 109485
    },
    {
      "epoch": 16.307715221924337,
      "grad_norm": 10.4375,
      "learning_rate": 5.014261269175457e-06,
      "loss": 1.0018,
      "num_input_tokens_seen": 63524048,
      "step": 109490
    },
    {
      "epoch": 16.308459934465297,
      "grad_norm": 3.828125,
      "learning_rate": 5.012309316058555e-06,
      "loss": 1.1451,
      "num_input_tokens_seen": 63526736,
      "step": 109495
    },
    {
      "epoch": 16.309204647006254,
      "grad_norm": 9.5,
      "learning_rate": 5.0103577006133685e-06,
      "loss": 0.687,
      "num_input_tokens_seen": 63529680,
      "step": 109500
    },
    {
      "epoch": 16.309949359547215,
      "grad_norm": 4.8125,
      "learning_rate": 5.008406422872878e-06,
      "loss": 1.0356,
      "num_input_tokens_seen": 63532752,
      "step": 109505
    },
    {
      "epoch": 16.310694072088175,
      "grad_norm": 10.9375,
      "learning_rate": 5.0064554828700345e-06,
      "loss": 1.0925,
      "num_input_tokens_seen": 63535280,
      "step": 109510
    },
    {
      "epoch": 16.311438784629132,
      "grad_norm": 8.0,
      "learning_rate": 5.004504880637812e-06,
      "loss": 0.8804,
      "num_input_tokens_seen": 63538704,
      "step": 109515
    },
    {
      "epoch": 16.312183497170093,
      "grad_norm": 10.8125,
      "learning_rate": 5.002554616209157e-06,
      "loss": 0.9619,
      "num_input_tokens_seen": 63541424,
      "step": 109520
    },
    {
      "epoch": 16.312928209711053,
      "grad_norm": 6.1875,
      "learning_rate": 5.000604689617011e-06,
      "loss": 0.8588,
      "num_input_tokens_seen": 63544304,
      "step": 109525
    },
    {
      "epoch": 16.31367292225201,
      "grad_norm": 12.125,
      "learning_rate": 4.998655100894328e-06,
      "loss": 1.0872,
      "num_input_tokens_seen": 63547280,
      "step": 109530
    },
    {
      "epoch": 16.31441763479297,
      "grad_norm": 9.5,
      "learning_rate": 4.996705850074041e-06,
      "loss": 1.0915,
      "num_input_tokens_seen": 63549776,
      "step": 109535
    },
    {
      "epoch": 16.315162347333928,
      "grad_norm": 8.4375,
      "learning_rate": 4.994756937189076e-06,
      "loss": 0.8978,
      "num_input_tokens_seen": 63553104,
      "step": 109540
    },
    {
      "epoch": 16.31590705987489,
      "grad_norm": 9.1875,
      "learning_rate": 4.992808362272353e-06,
      "loss": 1.4267,
      "num_input_tokens_seen": 63556336,
      "step": 109545
    },
    {
      "epoch": 16.31665177241585,
      "grad_norm": 5.9375,
      "learning_rate": 4.990860125356806e-06,
      "loss": 0.9749,
      "num_input_tokens_seen": 63559152,
      "step": 109550
    },
    {
      "epoch": 16.317396484956806,
      "grad_norm": 14.1875,
      "learning_rate": 4.988912226475342e-06,
      "loss": 1.1109,
      "num_input_tokens_seen": 63561680,
      "step": 109555
    },
    {
      "epoch": 16.318141197497766,
      "grad_norm": 6.78125,
      "learning_rate": 4.986964665660859e-06,
      "loss": 0.8459,
      "num_input_tokens_seen": 63564368,
      "step": 109560
    },
    {
      "epoch": 16.318885910038723,
      "grad_norm": 7.25,
      "learning_rate": 4.985017442946274e-06,
      "loss": 0.9084,
      "num_input_tokens_seen": 63567312,
      "step": 109565
    },
    {
      "epoch": 16.319630622579684,
      "grad_norm": 8.1875,
      "learning_rate": 4.983070558364472e-06,
      "loss": 0.8961,
      "num_input_tokens_seen": 63569904,
      "step": 109570
    },
    {
      "epoch": 16.320375335120644,
      "grad_norm": 7.0625,
      "learning_rate": 4.981124011948355e-06,
      "loss": 1.1458,
      "num_input_tokens_seen": 63572688,
      "step": 109575
    },
    {
      "epoch": 16.3211200476616,
      "grad_norm": 14.25,
      "learning_rate": 4.979177803730794e-06,
      "loss": 1.3631,
      "num_input_tokens_seen": 63575536,
      "step": 109580
    },
    {
      "epoch": 16.321864760202562,
      "grad_norm": 8.0625,
      "learning_rate": 4.9772319337446835e-06,
      "loss": 1.1278,
      "num_input_tokens_seen": 63578192,
      "step": 109585
    },
    {
      "epoch": 16.322609472743522,
      "grad_norm": 25.25,
      "learning_rate": 4.975286402022883e-06,
      "loss": 1.4831,
      "num_input_tokens_seen": 63581200,
      "step": 109590
    },
    {
      "epoch": 16.32335418528448,
      "grad_norm": 14.6875,
      "learning_rate": 4.973341208598273e-06,
      "loss": 1.5282,
      "num_input_tokens_seen": 63584048,
      "step": 109595
    },
    {
      "epoch": 16.32409889782544,
      "grad_norm": 17.375,
      "learning_rate": 4.971396353503707e-06,
      "loss": 1.3276,
      "num_input_tokens_seen": 63587152,
      "step": 109600
    },
    {
      "epoch": 16.324843610366397,
      "grad_norm": 6.90625,
      "learning_rate": 4.969451836772046e-06,
      "loss": 1.1864,
      "num_input_tokens_seen": 63589744,
      "step": 109605
    },
    {
      "epoch": 16.325588322907358,
      "grad_norm": 13.25,
      "learning_rate": 4.9675076584361355e-06,
      "loss": 1.1016,
      "num_input_tokens_seen": 63592528,
      "step": 109610
    },
    {
      "epoch": 16.326333035448318,
      "grad_norm": 12.5,
      "learning_rate": 4.965563818528818e-06,
      "loss": 1.1028,
      "num_input_tokens_seen": 63595088,
      "step": 109615
    },
    {
      "epoch": 16.327077747989275,
      "grad_norm": 6.03125,
      "learning_rate": 4.9636203170829424e-06,
      "loss": 0.7751,
      "num_input_tokens_seen": 63598064,
      "step": 109620
    },
    {
      "epoch": 16.327822460530236,
      "grad_norm": 6.46875,
      "learning_rate": 4.9616771541313335e-06,
      "loss": 1.0944,
      "num_input_tokens_seen": 63601040,
      "step": 109625
    },
    {
      "epoch": 16.328567173071196,
      "grad_norm": 8.0625,
      "learning_rate": 4.9597343297068274e-06,
      "loss": 0.7867,
      "num_input_tokens_seen": 63603696,
      "step": 109630
    },
    {
      "epoch": 16.329311885612153,
      "grad_norm": 14.0625,
      "learning_rate": 4.957791843842244e-06,
      "loss": 0.9567,
      "num_input_tokens_seen": 63606352,
      "step": 109635
    },
    {
      "epoch": 16.330056598153114,
      "grad_norm": 5.8125,
      "learning_rate": 4.955849696570392e-06,
      "loss": 0.9347,
      "num_input_tokens_seen": 63609168,
      "step": 109640
    },
    {
      "epoch": 16.33080131069407,
      "grad_norm": 10.5,
      "learning_rate": 4.953907887924089e-06,
      "loss": 1.4497,
      "num_input_tokens_seen": 63611984,
      "step": 109645
    },
    {
      "epoch": 16.33154602323503,
      "grad_norm": 7.6875,
      "learning_rate": 4.9519664179361355e-06,
      "loss": 1.0958,
      "num_input_tokens_seen": 63614864,
      "step": 109650
    },
    {
      "epoch": 16.33229073577599,
      "grad_norm": 15.0,
      "learning_rate": 4.95002528663934e-06,
      "loss": 1.1979,
      "num_input_tokens_seen": 63617904,
      "step": 109655
    },
    {
      "epoch": 16.33303544831695,
      "grad_norm": 13.9375,
      "learning_rate": 4.948084494066482e-06,
      "loss": 1.412,
      "num_input_tokens_seen": 63620880,
      "step": 109660
    },
    {
      "epoch": 16.33378016085791,
      "grad_norm": 6.0,
      "learning_rate": 4.946144040250361e-06,
      "loss": 0.9519,
      "num_input_tokens_seen": 63623728,
      "step": 109665
    },
    {
      "epoch": 16.33452487339887,
      "grad_norm": 12.0,
      "learning_rate": 4.944203925223759e-06,
      "loss": 1.1319,
      "num_input_tokens_seen": 63626736,
      "step": 109670
    },
    {
      "epoch": 16.335269585939827,
      "grad_norm": 11.75,
      "learning_rate": 4.942264149019446e-06,
      "loss": 1.2756,
      "num_input_tokens_seen": 63629712,
      "step": 109675
    },
    {
      "epoch": 16.336014298480787,
      "grad_norm": 10.8125,
      "learning_rate": 4.940324711670194e-06,
      "loss": 0.9888,
      "num_input_tokens_seen": 63632528,
      "step": 109680
    },
    {
      "epoch": 16.336759011021744,
      "grad_norm": 12.8125,
      "learning_rate": 4.93838561320876e-06,
      "loss": 1.072,
      "num_input_tokens_seen": 63635376,
      "step": 109685
    },
    {
      "epoch": 16.337503723562705,
      "grad_norm": 9.0,
      "learning_rate": 4.93644685366792e-06,
      "loss": 1.3466,
      "num_input_tokens_seen": 63637968,
      "step": 109690
    },
    {
      "epoch": 16.338248436103665,
      "grad_norm": 9.5625,
      "learning_rate": 4.934508433080412e-06,
      "loss": 0.961,
      "num_input_tokens_seen": 63640816,
      "step": 109695
    },
    {
      "epoch": 16.338993148644622,
      "grad_norm": 11.3125,
      "learning_rate": 4.932570351478996e-06,
      "loss": 1.2139,
      "num_input_tokens_seen": 63643408,
      "step": 109700
    },
    {
      "epoch": 16.339737861185583,
      "grad_norm": 8.8125,
      "learning_rate": 4.930632608896402e-06,
      "loss": 1.208,
      "num_input_tokens_seen": 63646416,
      "step": 109705
    },
    {
      "epoch": 16.34048257372654,
      "grad_norm": 19.0,
      "learning_rate": 4.92869520536538e-06,
      "loss": 0.918,
      "num_input_tokens_seen": 63649136,
      "step": 109710
    },
    {
      "epoch": 16.3412272862675,
      "grad_norm": 14.8125,
      "learning_rate": 4.926758140918647e-06,
      "loss": 0.8702,
      "num_input_tokens_seen": 63652432,
      "step": 109715
    },
    {
      "epoch": 16.34197199880846,
      "grad_norm": 5.28125,
      "learning_rate": 4.924821415588937e-06,
      "loss": 0.9184,
      "num_input_tokens_seen": 63655248,
      "step": 109720
    },
    {
      "epoch": 16.342716711349418,
      "grad_norm": 9.75,
      "learning_rate": 4.922885029408969e-06,
      "loss": 1.1252,
      "num_input_tokens_seen": 63658448,
      "step": 109725
    },
    {
      "epoch": 16.34346142389038,
      "grad_norm": 14.25,
      "learning_rate": 4.920948982411444e-06,
      "loss": 1.2577,
      "num_input_tokens_seen": 63661104,
      "step": 109730
    },
    {
      "epoch": 16.34420613643134,
      "grad_norm": 6.75,
      "learning_rate": 4.919013274629087e-06,
      "loss": 0.7959,
      "num_input_tokens_seen": 63664112,
      "step": 109735
    },
    {
      "epoch": 16.344950848972296,
      "grad_norm": 10.125,
      "learning_rate": 4.9170779060945916e-06,
      "loss": 0.945,
      "num_input_tokens_seen": 63666736,
      "step": 109740
    },
    {
      "epoch": 16.345695561513256,
      "grad_norm": 9.375,
      "learning_rate": 4.915142876840653e-06,
      "loss": 1.0395,
      "num_input_tokens_seen": 63669296,
      "step": 109745
    },
    {
      "epoch": 16.346440274054213,
      "grad_norm": 11.9375,
      "learning_rate": 4.9132081868999535e-06,
      "loss": 1.0227,
      "num_input_tokens_seen": 63672528,
      "step": 109750
    },
    {
      "epoch": 16.347184986595174,
      "grad_norm": 13.4375,
      "learning_rate": 4.911273836305194e-06,
      "loss": 1.1606,
      "num_input_tokens_seen": 63675376,
      "step": 109755
    },
    {
      "epoch": 16.347929699136134,
      "grad_norm": 3.359375,
      "learning_rate": 4.909339825089049e-06,
      "loss": 0.9471,
      "num_input_tokens_seen": 63677904,
      "step": 109760
    },
    {
      "epoch": 16.34867441167709,
      "grad_norm": 7.78125,
      "learning_rate": 4.9074061532841774e-06,
      "loss": 1.0024,
      "num_input_tokens_seen": 63680784,
      "step": 109765
    },
    {
      "epoch": 16.349419124218052,
      "grad_norm": 8.6875,
      "learning_rate": 4.905472820923265e-06,
      "loss": 1.1922,
      "num_input_tokens_seen": 63683568,
      "step": 109770
    },
    {
      "epoch": 16.350163836759013,
      "grad_norm": 13.75,
      "learning_rate": 4.903539828038961e-06,
      "loss": 1.4679,
      "num_input_tokens_seen": 63686448,
      "step": 109775
    },
    {
      "epoch": 16.35090854929997,
      "grad_norm": 11.3125,
      "learning_rate": 4.901607174663933e-06,
      "loss": 1.0184,
      "num_input_tokens_seen": 63689264,
      "step": 109780
    },
    {
      "epoch": 16.35165326184093,
      "grad_norm": 9.3125,
      "learning_rate": 4.899674860830819e-06,
      "loss": 1.0307,
      "num_input_tokens_seen": 63692208,
      "step": 109785
    },
    {
      "epoch": 16.352397974381887,
      "grad_norm": 9.9375,
      "learning_rate": 4.897742886572274e-06,
      "loss": 0.8571,
      "num_input_tokens_seen": 63695024,
      "step": 109790
    },
    {
      "epoch": 16.353142686922848,
      "grad_norm": 6.65625,
      "learning_rate": 4.8958112519209315e-06,
      "loss": 0.8668,
      "num_input_tokens_seen": 63698128,
      "step": 109795
    },
    {
      "epoch": 16.353887399463808,
      "grad_norm": 6.8125,
      "learning_rate": 4.8938799569094275e-06,
      "loss": 1.1025,
      "num_input_tokens_seen": 63701008,
      "step": 109800
    },
    {
      "epoch": 16.354632112004765,
      "grad_norm": 18.125,
      "learning_rate": 4.891949001570384e-06,
      "loss": 1.1303,
      "num_input_tokens_seen": 63703888,
      "step": 109805
    },
    {
      "epoch": 16.355376824545726,
      "grad_norm": 5.125,
      "learning_rate": 4.890018385936421e-06,
      "loss": 0.9764,
      "num_input_tokens_seen": 63706672,
      "step": 109810
    },
    {
      "epoch": 16.356121537086686,
      "grad_norm": 7.90625,
      "learning_rate": 4.888088110040162e-06,
      "loss": 0.8879,
      "num_input_tokens_seen": 63709456,
      "step": 109815
    },
    {
      "epoch": 16.356866249627643,
      "grad_norm": 11.8125,
      "learning_rate": 4.88615817391421e-06,
      "loss": 1.3912,
      "num_input_tokens_seen": 63712336,
      "step": 109820
    },
    {
      "epoch": 16.357610962168604,
      "grad_norm": 8.5625,
      "learning_rate": 4.884228577591177e-06,
      "loss": 1.1261,
      "num_input_tokens_seen": 63715216,
      "step": 109825
    },
    {
      "epoch": 16.35835567470956,
      "grad_norm": 6.96875,
      "learning_rate": 4.882299321103653e-06,
      "loss": 0.918,
      "num_input_tokens_seen": 63718032,
      "step": 109830
    },
    {
      "epoch": 16.35910038725052,
      "grad_norm": 8.6875,
      "learning_rate": 4.880370404484242e-06,
      "loss": 1.2664,
      "num_input_tokens_seen": 63721072,
      "step": 109835
    },
    {
      "epoch": 16.35984509979148,
      "grad_norm": 11.125,
      "learning_rate": 4.87844182776552e-06,
      "loss": 0.82,
      "num_input_tokens_seen": 63724048,
      "step": 109840
    },
    {
      "epoch": 16.36058981233244,
      "grad_norm": 7.1875,
      "learning_rate": 4.87651359098007e-06,
      "loss": 1.1062,
      "num_input_tokens_seen": 63726640,
      "step": 109845
    },
    {
      "epoch": 16.3613345248734,
      "grad_norm": 7.375,
      "learning_rate": 4.874585694160477e-06,
      "loss": 1.1247,
      "num_input_tokens_seen": 63729456,
      "step": 109850
    },
    {
      "epoch": 16.36207923741436,
      "grad_norm": 11.0625,
      "learning_rate": 4.872658137339295e-06,
      "loss": 1.2037,
      "num_input_tokens_seen": 63732560,
      "step": 109855
    },
    {
      "epoch": 16.362823949955317,
      "grad_norm": 15.0625,
      "learning_rate": 4.870730920549108e-06,
      "loss": 0.885,
      "num_input_tokens_seen": 63735376,
      "step": 109860
    },
    {
      "epoch": 16.363568662496277,
      "grad_norm": 12.25,
      "learning_rate": 4.868804043822458e-06,
      "loss": 1.1376,
      "num_input_tokens_seen": 63738640,
      "step": 109865
    },
    {
      "epoch": 16.364313375037234,
      "grad_norm": 9.5625,
      "learning_rate": 4.866877507191908e-06,
      "loss": 0.9757,
      "num_input_tokens_seen": 63741456,
      "step": 109870
    },
    {
      "epoch": 16.365058087578195,
      "grad_norm": 10.9375,
      "learning_rate": 4.864951310689991e-06,
      "loss": 1.0267,
      "num_input_tokens_seen": 63744432,
      "step": 109875
    },
    {
      "epoch": 16.365802800119155,
      "grad_norm": 10.9375,
      "learning_rate": 4.863025454349266e-06,
      "loss": 0.9545,
      "num_input_tokens_seen": 63747440,
      "step": 109880
    },
    {
      "epoch": 16.366547512660112,
      "grad_norm": 8.1875,
      "learning_rate": 4.861099938202257e-06,
      "loss": 0.8679,
      "num_input_tokens_seen": 63750320,
      "step": 109885
    },
    {
      "epoch": 16.367292225201073,
      "grad_norm": 10.6875,
      "learning_rate": 4.859174762281493e-06,
      "loss": 1.0805,
      "num_input_tokens_seen": 63753136,
      "step": 109890
    },
    {
      "epoch": 16.36803693774203,
      "grad_norm": 17.5,
      "learning_rate": 4.857249926619506e-06,
      "loss": 1.0716,
      "num_input_tokens_seen": 63756016,
      "step": 109895
    },
    {
      "epoch": 16.36878165028299,
      "grad_norm": 13.9375,
      "learning_rate": 4.855325431248803e-06,
      "loss": 1.086,
      "num_input_tokens_seen": 63758800,
      "step": 109900
    },
    {
      "epoch": 16.36952636282395,
      "grad_norm": 9.75,
      "learning_rate": 4.853401276201908e-06,
      "loss": 0.8494,
      "num_input_tokens_seen": 63761680,
      "step": 109905
    },
    {
      "epoch": 16.370271075364908,
      "grad_norm": 7.5625,
      "learning_rate": 4.851477461511317e-06,
      "loss": 1.3493,
      "num_input_tokens_seen": 63764400,
      "step": 109910
    },
    {
      "epoch": 16.37101578790587,
      "grad_norm": 12.75,
      "learning_rate": 4.84955398720954e-06,
      "loss": 1.2675,
      "num_input_tokens_seen": 63767248,
      "step": 109915
    },
    {
      "epoch": 16.37176050044683,
      "grad_norm": 7.03125,
      "learning_rate": 4.8476308533290714e-06,
      "loss": 0.6596,
      "num_input_tokens_seen": 63770000,
      "step": 109920
    },
    {
      "epoch": 16.372505212987786,
      "grad_norm": 9.1875,
      "learning_rate": 4.8457080599023905e-06,
      "loss": 1.2714,
      "num_input_tokens_seen": 63772816,
      "step": 109925
    },
    {
      "epoch": 16.373249925528746,
      "grad_norm": 14.3125,
      "learning_rate": 4.843785606961995e-06,
      "loss": 1.177,
      "num_input_tokens_seen": 63775856,
      "step": 109930
    },
    {
      "epoch": 16.373994638069703,
      "grad_norm": 13.375,
      "learning_rate": 4.8418634945403555e-06,
      "loss": 1.6698,
      "num_input_tokens_seen": 63778832,
      "step": 109935
    },
    {
      "epoch": 16.374739350610664,
      "grad_norm": 17.625,
      "learning_rate": 4.839941722669944e-06,
      "loss": 1.09,
      "num_input_tokens_seen": 63781840,
      "step": 109940
    },
    {
      "epoch": 16.375484063151625,
      "grad_norm": 19.625,
      "learning_rate": 4.8380202913832215e-06,
      "loss": 1.0508,
      "num_input_tokens_seen": 63784592,
      "step": 109945
    },
    {
      "epoch": 16.37622877569258,
      "grad_norm": 8.5625,
      "learning_rate": 4.83609920071266e-06,
      "loss": 0.9129,
      "num_input_tokens_seen": 63787568,
      "step": 109950
    },
    {
      "epoch": 16.376973488233542,
      "grad_norm": 11.9375,
      "learning_rate": 4.834178450690704e-06,
      "loss": 1.1299,
      "num_input_tokens_seen": 63790352,
      "step": 109955
    },
    {
      "epoch": 16.377718200774503,
      "grad_norm": 11.125,
      "learning_rate": 4.832258041349813e-06,
      "loss": 1.0233,
      "num_input_tokens_seen": 63793520,
      "step": 109960
    },
    {
      "epoch": 16.37846291331546,
      "grad_norm": 10.75,
      "learning_rate": 4.830337972722424e-06,
      "loss": 1.0659,
      "num_input_tokens_seen": 63796368,
      "step": 109965
    },
    {
      "epoch": 16.37920762585642,
      "grad_norm": 11.1875,
      "learning_rate": 4.828418244840968e-06,
      "loss": 1.2605,
      "num_input_tokens_seen": 63799280,
      "step": 109970
    },
    {
      "epoch": 16.379952338397377,
      "grad_norm": 8.75,
      "learning_rate": 4.8264988577378934e-06,
      "loss": 0.9948,
      "num_input_tokens_seen": 63802096,
      "step": 109975
    },
    {
      "epoch": 16.380697050938338,
      "grad_norm": 16.0,
      "learning_rate": 4.824579811445609e-06,
      "loss": 1.0004,
      "num_input_tokens_seen": 63804688,
      "step": 109980
    },
    {
      "epoch": 16.381441763479298,
      "grad_norm": 8.8125,
      "learning_rate": 4.822661105996551e-06,
      "loss": 0.9449,
      "num_input_tokens_seen": 63807664,
      "step": 109985
    },
    {
      "epoch": 16.382186476020255,
      "grad_norm": 9.8125,
      "learning_rate": 4.82074274142312e-06,
      "loss": 1.26,
      "num_input_tokens_seen": 63810704,
      "step": 109990
    },
    {
      "epoch": 16.382931188561216,
      "grad_norm": 4.65625,
      "learning_rate": 4.818824717757736e-06,
      "loss": 1.007,
      "num_input_tokens_seen": 63813392,
      "step": 109995
    },
    {
      "epoch": 16.383675901102176,
      "grad_norm": 6.46875,
      "learning_rate": 4.816907035032797e-06,
      "loss": 1.1571,
      "num_input_tokens_seen": 63816496,
      "step": 110000
    },
    {
      "epoch": 16.384420613643133,
      "grad_norm": 7.59375,
      "learning_rate": 4.814989693280703e-06,
      "loss": 0.6771,
      "num_input_tokens_seen": 63819344,
      "step": 110005
    },
    {
      "epoch": 16.385165326184094,
      "grad_norm": 9.0625,
      "learning_rate": 4.81307269253384e-06,
      "loss": 1.1184,
      "num_input_tokens_seen": 63822128,
      "step": 110010
    },
    {
      "epoch": 16.38591003872505,
      "grad_norm": 11.1875,
      "learning_rate": 4.811156032824593e-06,
      "loss": 1.339,
      "num_input_tokens_seen": 63825168,
      "step": 110015
    },
    {
      "epoch": 16.38665475126601,
      "grad_norm": 6.375,
      "learning_rate": 4.8092397141853515e-06,
      "loss": 0.9387,
      "num_input_tokens_seen": 63828080,
      "step": 110020
    },
    {
      "epoch": 16.38739946380697,
      "grad_norm": 16.375,
      "learning_rate": 4.807323736648475e-06,
      "loss": 1.0387,
      "num_input_tokens_seen": 63830928,
      "step": 110025
    },
    {
      "epoch": 16.38814417634793,
      "grad_norm": 5.5625,
      "learning_rate": 4.80540810024635e-06,
      "loss": 1.1017,
      "num_input_tokens_seen": 63833872,
      "step": 110030
    },
    {
      "epoch": 16.38888888888889,
      "grad_norm": 12.625,
      "learning_rate": 4.8034928050113256e-06,
      "loss": 1.335,
      "num_input_tokens_seen": 63836688,
      "step": 110035
    },
    {
      "epoch": 16.38963360142985,
      "grad_norm": 16.375,
      "learning_rate": 4.8015778509757665e-06,
      "loss": 1.0434,
      "num_input_tokens_seen": 63839376,
      "step": 110040
    },
    {
      "epoch": 16.390378313970807,
      "grad_norm": 11.5625,
      "learning_rate": 4.799663238172022e-06,
      "loss": 1.2434,
      "num_input_tokens_seen": 63842160,
      "step": 110045
    },
    {
      "epoch": 16.391123026511767,
      "grad_norm": 11.25,
      "learning_rate": 4.7977489666324285e-06,
      "loss": 1.1211,
      "num_input_tokens_seen": 63844976,
      "step": 110050
    },
    {
      "epoch": 16.391867739052724,
      "grad_norm": 7.53125,
      "learning_rate": 4.7958350363893424e-06,
      "loss": 0.884,
      "num_input_tokens_seen": 63847760,
      "step": 110055
    },
    {
      "epoch": 16.392612451593685,
      "grad_norm": 15.5,
      "learning_rate": 4.793921447475083e-06,
      "loss": 1.1674,
      "num_input_tokens_seen": 63850704,
      "step": 110060
    },
    {
      "epoch": 16.393357164134645,
      "grad_norm": 8.1875,
      "learning_rate": 4.7920081999219875e-06,
      "loss": 1.1012,
      "num_input_tokens_seen": 63854096,
      "step": 110065
    },
    {
      "epoch": 16.394101876675602,
      "grad_norm": 8.0625,
      "learning_rate": 4.790095293762379e-06,
      "loss": 0.9485,
      "num_input_tokens_seen": 63856848,
      "step": 110070
    },
    {
      "epoch": 16.394846589216563,
      "grad_norm": 7.6875,
      "learning_rate": 4.788182729028565e-06,
      "loss": 1.0119,
      "num_input_tokens_seen": 63859984,
      "step": 110075
    },
    {
      "epoch": 16.39559130175752,
      "grad_norm": 6.59375,
      "learning_rate": 4.786270505752866e-06,
      "loss": 1.1137,
      "num_input_tokens_seen": 63862768,
      "step": 110080
    },
    {
      "epoch": 16.39633601429848,
      "grad_norm": 10.8125,
      "learning_rate": 4.784358623967572e-06,
      "loss": 1.2639,
      "num_input_tokens_seen": 63865712,
      "step": 110085
    },
    {
      "epoch": 16.39708072683944,
      "grad_norm": 9.875,
      "learning_rate": 4.782447083705002e-06,
      "loss": 1.2411,
      "num_input_tokens_seen": 63868688,
      "step": 110090
    },
    {
      "epoch": 16.397825439380398,
      "grad_norm": 6.125,
      "learning_rate": 4.780535884997433e-06,
      "loss": 0.728,
      "num_input_tokens_seen": 63871952,
      "step": 110095
    },
    {
      "epoch": 16.39857015192136,
      "grad_norm": 10.5,
      "learning_rate": 4.7786250278771675e-06,
      "loss": 1.1548,
      "num_input_tokens_seen": 63874608,
      "step": 110100
    },
    {
      "epoch": 16.39931486446232,
      "grad_norm": 11.25,
      "learning_rate": 4.776714512376474e-06,
      "loss": 1.2505,
      "num_input_tokens_seen": 63877616,
      "step": 110105
    },
    {
      "epoch": 16.400059577003276,
      "grad_norm": 8.1875,
      "learning_rate": 4.774804338527639e-06,
      "loss": 1.223,
      "num_input_tokens_seen": 63880560,
      "step": 110110
    },
    {
      "epoch": 16.400804289544237,
      "grad_norm": 10.5625,
      "learning_rate": 4.772894506362924e-06,
      "loss": 1.1451,
      "num_input_tokens_seen": 63883504,
      "step": 110115
    },
    {
      "epoch": 16.401549002085194,
      "grad_norm": 4.46875,
      "learning_rate": 4.770985015914603e-06,
      "loss": 0.7784,
      "num_input_tokens_seen": 63886352,
      "step": 110120
    },
    {
      "epoch": 16.402293714626154,
      "grad_norm": 10.0,
      "learning_rate": 4.769075867214931e-06,
      "loss": 1.4956,
      "num_input_tokens_seen": 63889264,
      "step": 110125
    },
    {
      "epoch": 16.403038427167115,
      "grad_norm": 9.6875,
      "learning_rate": 4.767167060296163e-06,
      "loss": 0.7565,
      "num_input_tokens_seen": 63891824,
      "step": 110130
    },
    {
      "epoch": 16.40378313970807,
      "grad_norm": 7.8125,
      "learning_rate": 4.7652585951905415e-06,
      "loss": 1.1222,
      "num_input_tokens_seen": 63894608,
      "step": 110135
    },
    {
      "epoch": 16.404527852249032,
      "grad_norm": 14.0625,
      "learning_rate": 4.763350471930303e-06,
      "loss": 0.9044,
      "num_input_tokens_seen": 63897552,
      "step": 110140
    },
    {
      "epoch": 16.405272564789993,
      "grad_norm": 13.625,
      "learning_rate": 4.761442690547699e-06,
      "loss": 1.3498,
      "num_input_tokens_seen": 63900400,
      "step": 110145
    },
    {
      "epoch": 16.40601727733095,
      "grad_norm": 8.1875,
      "learning_rate": 4.759535251074942e-06,
      "loss": 1.1614,
      "num_input_tokens_seen": 63903248,
      "step": 110150
    },
    {
      "epoch": 16.40676198987191,
      "grad_norm": 10.875,
      "learning_rate": 4.7576281535442745e-06,
      "loss": 1.2911,
      "num_input_tokens_seen": 63906320,
      "step": 110155
    },
    {
      "epoch": 16.407506702412867,
      "grad_norm": 15.8125,
      "learning_rate": 4.755721397987906e-06,
      "loss": 1.1231,
      "num_input_tokens_seen": 63909360,
      "step": 110160
    },
    {
      "epoch": 16.408251414953828,
      "grad_norm": 7.1875,
      "learning_rate": 4.753814984438043e-06,
      "loss": 1.0645,
      "num_input_tokens_seen": 63912112,
      "step": 110165
    },
    {
      "epoch": 16.408996127494788,
      "grad_norm": 11.875,
      "learning_rate": 4.7519089129269026e-06,
      "loss": 0.9602,
      "num_input_tokens_seen": 63915088,
      "step": 110170
    },
    {
      "epoch": 16.409740840035745,
      "grad_norm": 5.4375,
      "learning_rate": 4.750003183486676e-06,
      "loss": 0.9947,
      "num_input_tokens_seen": 63918000,
      "step": 110175
    },
    {
      "epoch": 16.410485552576706,
      "grad_norm": 11.3125,
      "learning_rate": 4.748097796149573e-06,
      "loss": 1.0986,
      "num_input_tokens_seen": 63920816,
      "step": 110180
    },
    {
      "epoch": 16.411230265117666,
      "grad_norm": 8.0,
      "learning_rate": 4.746192750947767e-06,
      "loss": 1.0611,
      "num_input_tokens_seen": 63923664,
      "step": 110185
    },
    {
      "epoch": 16.411974977658623,
      "grad_norm": 18.375,
      "learning_rate": 4.744288047913456e-06,
      "loss": 0.9869,
      "num_input_tokens_seen": 63926704,
      "step": 110190
    },
    {
      "epoch": 16.412719690199584,
      "grad_norm": 8.5,
      "learning_rate": 4.742383687078811e-06,
      "loss": 0.9377,
      "num_input_tokens_seen": 63929392,
      "step": 110195
    },
    {
      "epoch": 16.41346440274054,
      "grad_norm": 20.125,
      "learning_rate": 4.7404796684760055e-06,
      "loss": 1.3641,
      "num_input_tokens_seen": 63932336,
      "step": 110200
    },
    {
      "epoch": 16.4142091152815,
      "grad_norm": 7.3125,
      "learning_rate": 4.738575992137203e-06,
      "loss": 1.1663,
      "num_input_tokens_seen": 63935280,
      "step": 110205
    },
    {
      "epoch": 16.414953827822462,
      "grad_norm": 4.25,
      "learning_rate": 4.736672658094562e-06,
      "loss": 0.7391,
      "num_input_tokens_seen": 63938192,
      "step": 110210
    },
    {
      "epoch": 16.41569854036342,
      "grad_norm": 9.125,
      "learning_rate": 4.734769666380248e-06,
      "loss": 1.0257,
      "num_input_tokens_seen": 63941168,
      "step": 110215
    },
    {
      "epoch": 16.41644325290438,
      "grad_norm": 6.09375,
      "learning_rate": 4.732867017026396e-06,
      "loss": 1.3068,
      "num_input_tokens_seen": 63944112,
      "step": 110220
    },
    {
      "epoch": 16.417187965445336,
      "grad_norm": 15.875,
      "learning_rate": 4.730964710065164e-06,
      "loss": 1.1743,
      "num_input_tokens_seen": 63946960,
      "step": 110225
    },
    {
      "epoch": 16.417932677986297,
      "grad_norm": 7.15625,
      "learning_rate": 4.729062745528678e-06,
      "loss": 1.2417,
      "num_input_tokens_seen": 63950192,
      "step": 110230
    },
    {
      "epoch": 16.418677390527257,
      "grad_norm": 10.0625,
      "learning_rate": 4.727161123449078e-06,
      "loss": 1.2513,
      "num_input_tokens_seen": 63953200,
      "step": 110235
    },
    {
      "epoch": 16.419422103068214,
      "grad_norm": 7.8125,
      "learning_rate": 4.72525984385849e-06,
      "loss": 0.8021,
      "num_input_tokens_seen": 63956048,
      "step": 110240
    },
    {
      "epoch": 16.420166815609175,
      "grad_norm": 14.25,
      "learning_rate": 4.7233589067890215e-06,
      "loss": 1.0842,
      "num_input_tokens_seen": 63959216,
      "step": 110245
    },
    {
      "epoch": 16.420911528150135,
      "grad_norm": 7.53125,
      "learning_rate": 4.721458312272803e-06,
      "loss": 0.9314,
      "num_input_tokens_seen": 63962224,
      "step": 110250
    },
    {
      "epoch": 16.421656240691092,
      "grad_norm": 6.84375,
      "learning_rate": 4.719558060341931e-06,
      "loss": 0.8239,
      "num_input_tokens_seen": 63965296,
      "step": 110255
    },
    {
      "epoch": 16.422400953232053,
      "grad_norm": 13.25,
      "learning_rate": 4.717658151028517e-06,
      "loss": 1.1286,
      "num_input_tokens_seen": 63968592,
      "step": 110260
    },
    {
      "epoch": 16.42314566577301,
      "grad_norm": 9.0,
      "learning_rate": 4.715758584364657e-06,
      "loss": 1.1085,
      "num_input_tokens_seen": 63971504,
      "step": 110265
    },
    {
      "epoch": 16.42389037831397,
      "grad_norm": 6.0,
      "learning_rate": 4.713859360382439e-06,
      "loss": 1.0917,
      "num_input_tokens_seen": 63974224,
      "step": 110270
    },
    {
      "epoch": 16.42463509085493,
      "grad_norm": 15.625,
      "learning_rate": 4.7119604791139414e-06,
      "loss": 1.2263,
      "num_input_tokens_seen": 63977392,
      "step": 110275
    },
    {
      "epoch": 16.425379803395888,
      "grad_norm": 11.1875,
      "learning_rate": 4.7100619405912625e-06,
      "loss": 1.0951,
      "num_input_tokens_seen": 63980368,
      "step": 110280
    },
    {
      "epoch": 16.42612451593685,
      "grad_norm": 8.6875,
      "learning_rate": 4.708163744846461e-06,
      "loss": 0.9404,
      "num_input_tokens_seen": 63983056,
      "step": 110285
    },
    {
      "epoch": 16.42686922847781,
      "grad_norm": 11.375,
      "learning_rate": 4.706265891911604e-06,
      "loss": 1.1448,
      "num_input_tokens_seen": 63985840,
      "step": 110290
    },
    {
      "epoch": 16.427613941018766,
      "grad_norm": 17.0,
      "learning_rate": 4.704368381818766e-06,
      "loss": 1.0959,
      "num_input_tokens_seen": 63988816,
      "step": 110295
    },
    {
      "epoch": 16.428358653559727,
      "grad_norm": 6.34375,
      "learning_rate": 4.70247121459999e-06,
      "loss": 0.7673,
      "num_input_tokens_seen": 63991568,
      "step": 110300
    },
    {
      "epoch": 16.429103366100684,
      "grad_norm": 5.46875,
      "learning_rate": 4.700574390287341e-06,
      "loss": 0.7744,
      "num_input_tokens_seen": 63994352,
      "step": 110305
    },
    {
      "epoch": 16.429848078641644,
      "grad_norm": 6.5,
      "learning_rate": 4.698677908912846e-06,
      "loss": 0.9412,
      "num_input_tokens_seen": 63997296,
      "step": 110310
    },
    {
      "epoch": 16.430592791182605,
      "grad_norm": 11.9375,
      "learning_rate": 4.696781770508566e-06,
      "loss": 0.9863,
      "num_input_tokens_seen": 63999984,
      "step": 110315
    },
    {
      "epoch": 16.43133750372356,
      "grad_norm": 17.625,
      "learning_rate": 4.694885975106511e-06,
      "loss": 1.4247,
      "num_input_tokens_seen": 64002992,
      "step": 110320
    },
    {
      "epoch": 16.432082216264522,
      "grad_norm": 5.875,
      "learning_rate": 4.6929905227387295e-06,
      "loss": 0.8356,
      "num_input_tokens_seen": 64005840,
      "step": 110325
    },
    {
      "epoch": 16.432826928805483,
      "grad_norm": 8.75,
      "learning_rate": 4.691095413437235e-06,
      "loss": 1.2121,
      "num_input_tokens_seen": 64008688,
      "step": 110330
    },
    {
      "epoch": 16.43357164134644,
      "grad_norm": 7.96875,
      "learning_rate": 4.6892006472340405e-06,
      "loss": 0.811,
      "num_input_tokens_seen": 64011472,
      "step": 110335
    },
    {
      "epoch": 16.4343163538874,
      "grad_norm": 13.75,
      "learning_rate": 4.687306224161159e-06,
      "loss": 0.7979,
      "num_input_tokens_seen": 64014672,
      "step": 110340
    },
    {
      "epoch": 16.435061066428357,
      "grad_norm": 10.625,
      "learning_rate": 4.685412144250586e-06,
      "loss": 1.1697,
      "num_input_tokens_seen": 64017360,
      "step": 110345
    },
    {
      "epoch": 16.435805778969318,
      "grad_norm": 15.125,
      "learning_rate": 4.683518407534338e-06,
      "loss": 1.5137,
      "num_input_tokens_seen": 64020304,
      "step": 110350
    },
    {
      "epoch": 16.43655049151028,
      "grad_norm": 10.4375,
      "learning_rate": 4.6816250140443884e-06,
      "loss": 1.0775,
      "num_input_tokens_seen": 64023504,
      "step": 110355
    },
    {
      "epoch": 16.437295204051235,
      "grad_norm": 7.71875,
      "learning_rate": 4.679731963812742e-06,
      "loss": 0.7795,
      "num_input_tokens_seen": 64026544,
      "step": 110360
    },
    {
      "epoch": 16.438039916592196,
      "grad_norm": 9.0,
      "learning_rate": 4.6778392568713695e-06,
      "loss": 1.012,
      "num_input_tokens_seen": 64029232,
      "step": 110365
    },
    {
      "epoch": 16.438784629133156,
      "grad_norm": 12.5625,
      "learning_rate": 4.675946893252242e-06,
      "loss": 1.2203,
      "num_input_tokens_seen": 64032432,
      "step": 110370
    },
    {
      "epoch": 16.439529341674113,
      "grad_norm": 7.46875,
      "learning_rate": 4.674054872987344e-06,
      "loss": 1.1076,
      "num_input_tokens_seen": 64035376,
      "step": 110375
    },
    {
      "epoch": 16.440274054215074,
      "grad_norm": 13.8125,
      "learning_rate": 4.67216319610862e-06,
      "loss": 1.0066,
      "num_input_tokens_seen": 64038384,
      "step": 110380
    },
    {
      "epoch": 16.44101876675603,
      "grad_norm": 13.3125,
      "learning_rate": 4.670271862648049e-06,
      "loss": 1.3596,
      "num_input_tokens_seen": 64041520,
      "step": 110385
    },
    {
      "epoch": 16.44176347929699,
      "grad_norm": 13.6875,
      "learning_rate": 4.668380872637562e-06,
      "loss": 1.1355,
      "num_input_tokens_seen": 64044080,
      "step": 110390
    },
    {
      "epoch": 16.442508191837952,
      "grad_norm": 8.3125,
      "learning_rate": 4.666490226109127e-06,
      "loss": 1.0357,
      "num_input_tokens_seen": 64046832,
      "step": 110395
    },
    {
      "epoch": 16.44325290437891,
      "grad_norm": 4.90625,
      "learning_rate": 4.66459992309467e-06,
      "loss": 0.7565,
      "num_input_tokens_seen": 64049392,
      "step": 110400
    },
    {
      "epoch": 16.44399761691987,
      "grad_norm": 11.3125,
      "learning_rate": 4.662709963626133e-06,
      "loss": 1.0417,
      "num_input_tokens_seen": 64052144,
      "step": 110405
    },
    {
      "epoch": 16.44474232946083,
      "grad_norm": 8.0625,
      "learning_rate": 4.660820347735437e-06,
      "loss": 0.8894,
      "num_input_tokens_seen": 64055024,
      "step": 110410
    },
    {
      "epoch": 16.445487042001787,
      "grad_norm": 21.375,
      "learning_rate": 4.658931075454507e-06,
      "loss": 1.5368,
      "num_input_tokens_seen": 64057936,
      "step": 110415
    },
    {
      "epoch": 16.446231754542747,
      "grad_norm": 7.625,
      "learning_rate": 4.657042146815266e-06,
      "loss": 1.045,
      "num_input_tokens_seen": 64060656,
      "step": 110420
    },
    {
      "epoch": 16.446976467083704,
      "grad_norm": 12.0625,
      "learning_rate": 4.655153561849618e-06,
      "loss": 1.1067,
      "num_input_tokens_seen": 64063568,
      "step": 110425
    },
    {
      "epoch": 16.447721179624665,
      "grad_norm": 8.8125,
      "learning_rate": 4.6532653205894786e-06,
      "loss": 1.2712,
      "num_input_tokens_seen": 64066288,
      "step": 110430
    },
    {
      "epoch": 16.448465892165625,
      "grad_norm": 12.1875,
      "learning_rate": 4.651377423066736e-06,
      "loss": 1.1769,
      "num_input_tokens_seen": 64069392,
      "step": 110435
    },
    {
      "epoch": 16.449210604706582,
      "grad_norm": 11.6875,
      "learning_rate": 4.649489869313295e-06,
      "loss": 1.1024,
      "num_input_tokens_seen": 64072400,
      "step": 110440
    },
    {
      "epoch": 16.449955317247543,
      "grad_norm": 9.0625,
      "learning_rate": 4.647602659361042e-06,
      "loss": 0.8663,
      "num_input_tokens_seen": 64075440,
      "step": 110445
    },
    {
      "epoch": 16.4507000297885,
      "grad_norm": 7.96875,
      "learning_rate": 4.645715793241848e-06,
      "loss": 1.3604,
      "num_input_tokens_seen": 64078192,
      "step": 110450
    },
    {
      "epoch": 16.45144474232946,
      "grad_norm": 19.875,
      "learning_rate": 4.6438292709876065e-06,
      "loss": 1.0734,
      "num_input_tokens_seen": 64081264,
      "step": 110455
    },
    {
      "epoch": 16.45218945487042,
      "grad_norm": 14.875,
      "learning_rate": 4.64194309263018e-06,
      "loss": 1.2174,
      "num_input_tokens_seen": 64084208,
      "step": 110460
    },
    {
      "epoch": 16.452934167411378,
      "grad_norm": 10.8125,
      "learning_rate": 4.6400572582014325e-06,
      "loss": 1.2932,
      "num_input_tokens_seen": 64086896,
      "step": 110465
    },
    {
      "epoch": 16.45367887995234,
      "grad_norm": 10.875,
      "learning_rate": 4.638171767733221e-06,
      "loss": 0.8748,
      "num_input_tokens_seen": 64089936,
      "step": 110470
    },
    {
      "epoch": 16.4544235924933,
      "grad_norm": 8.4375,
      "learning_rate": 4.636286621257407e-06,
      "loss": 1.229,
      "num_input_tokens_seen": 64092848,
      "step": 110475
    },
    {
      "epoch": 16.455168305034256,
      "grad_norm": 10.375,
      "learning_rate": 4.634401818805828e-06,
      "loss": 0.9787,
      "num_input_tokens_seen": 64096112,
      "step": 110480
    },
    {
      "epoch": 16.455913017575217,
      "grad_norm": 6.625,
      "learning_rate": 4.632517360410338e-06,
      "loss": 1.1609,
      "num_input_tokens_seen": 64098928,
      "step": 110485
    },
    {
      "epoch": 16.456657730116174,
      "grad_norm": 8.6875,
      "learning_rate": 4.630633246102767e-06,
      "loss": 0.8851,
      "num_input_tokens_seen": 64101776,
      "step": 110490
    },
    {
      "epoch": 16.457402442657134,
      "grad_norm": 5.96875,
      "learning_rate": 4.62874947591494e-06,
      "loss": 1.2312,
      "num_input_tokens_seen": 64104784,
      "step": 110495
    },
    {
      "epoch": 16.458147155198095,
      "grad_norm": 13.8125,
      "learning_rate": 4.62686604987869e-06,
      "loss": 1.0853,
      "num_input_tokens_seen": 64107920,
      "step": 110500
    },
    {
      "epoch": 16.45889186773905,
      "grad_norm": 20.0,
      "learning_rate": 4.624982968025826e-06,
      "loss": 1.5805,
      "num_input_tokens_seen": 64110864,
      "step": 110505
    },
    {
      "epoch": 16.459636580280012,
      "grad_norm": 11.5,
      "learning_rate": 4.623100230388172e-06,
      "loss": 1.0799,
      "num_input_tokens_seen": 64113584,
      "step": 110510
    },
    {
      "epoch": 16.460381292820973,
      "grad_norm": 9.1875,
      "learning_rate": 4.621217836997524e-06,
      "loss": 1.3075,
      "num_input_tokens_seen": 64116624,
      "step": 110515
    },
    {
      "epoch": 16.46112600536193,
      "grad_norm": 10.8125,
      "learning_rate": 4.619335787885695e-06,
      "loss": 1.0296,
      "num_input_tokens_seen": 64119408,
      "step": 110520
    },
    {
      "epoch": 16.46187071790289,
      "grad_norm": 10.625,
      "learning_rate": 4.617454083084474e-06,
      "loss": 1.0536,
      "num_input_tokens_seen": 64122256,
      "step": 110525
    },
    {
      "epoch": 16.462615430443847,
      "grad_norm": 7.6875,
      "learning_rate": 4.615572722625649e-06,
      "loss": 1.3174,
      "num_input_tokens_seen": 64124944,
      "step": 110530
    },
    {
      "epoch": 16.463360142984808,
      "grad_norm": 3.9375,
      "learning_rate": 4.6136917065410065e-06,
      "loss": 0.7905,
      "num_input_tokens_seen": 64127824,
      "step": 110535
    },
    {
      "epoch": 16.46410485552577,
      "grad_norm": 9.3125,
      "learning_rate": 4.611811034862318e-06,
      "loss": 0.8496,
      "num_input_tokens_seen": 64130640,
      "step": 110540
    },
    {
      "epoch": 16.464849568066725,
      "grad_norm": 8.4375,
      "learning_rate": 4.609930707621366e-06,
      "loss": 1.3227,
      "num_input_tokens_seen": 64133584,
      "step": 110545
    },
    {
      "epoch": 16.465594280607686,
      "grad_norm": 9.0,
      "learning_rate": 4.608050724849902e-06,
      "loss": 1.3566,
      "num_input_tokens_seen": 64136624,
      "step": 110550
    },
    {
      "epoch": 16.466338993148646,
      "grad_norm": 8.0625,
      "learning_rate": 4.6061710865797055e-06,
      "loss": 1.018,
      "num_input_tokens_seen": 64139824,
      "step": 110555
    },
    {
      "epoch": 16.467083705689603,
      "grad_norm": 10.8125,
      "learning_rate": 4.604291792842513e-06,
      "loss": 1.2817,
      "num_input_tokens_seen": 64142608,
      "step": 110560
    },
    {
      "epoch": 16.467828418230564,
      "grad_norm": 24.875,
      "learning_rate": 4.602412843670087e-06,
      "loss": 1.3394,
      "num_input_tokens_seen": 64145392,
      "step": 110565
    },
    {
      "epoch": 16.46857313077152,
      "grad_norm": 6.9375,
      "learning_rate": 4.600534239094165e-06,
      "loss": 1.1928,
      "num_input_tokens_seen": 64148176,
      "step": 110570
    },
    {
      "epoch": 16.46931784331248,
      "grad_norm": 9.0,
      "learning_rate": 4.598655979146479e-06,
      "loss": 1.1766,
      "num_input_tokens_seen": 64150992,
      "step": 110575
    },
    {
      "epoch": 16.470062555853442,
      "grad_norm": 18.0,
      "learning_rate": 4.59677806385877e-06,
      "loss": 1.1406,
      "num_input_tokens_seen": 64153936,
      "step": 110580
    },
    {
      "epoch": 16.4708072683944,
      "grad_norm": 9.625,
      "learning_rate": 4.5949004932627545e-06,
      "loss": 1.4152,
      "num_input_tokens_seen": 64156848,
      "step": 110585
    },
    {
      "epoch": 16.47155198093536,
      "grad_norm": 9.875,
      "learning_rate": 4.593023267390162e-06,
      "loss": 1.0182,
      "num_input_tokens_seen": 64160016,
      "step": 110590
    },
    {
      "epoch": 16.472296693476316,
      "grad_norm": 10.0625,
      "learning_rate": 4.591146386272699e-06,
      "loss": 0.964,
      "num_input_tokens_seen": 64162864,
      "step": 110595
    },
    {
      "epoch": 16.473041406017277,
      "grad_norm": 11.3125,
      "learning_rate": 4.5892698499420764e-06,
      "loss": 1.1851,
      "num_input_tokens_seen": 64165840,
      "step": 110600
    },
    {
      "epoch": 16.473786118558237,
      "grad_norm": 15.6875,
      "learning_rate": 4.5873936584299946e-06,
      "loss": 1.1742,
      "num_input_tokens_seen": 64168784,
      "step": 110605
    },
    {
      "epoch": 16.474530831099194,
      "grad_norm": 10.1875,
      "learning_rate": 4.5855178117681444e-06,
      "loss": 1.0194,
      "num_input_tokens_seen": 64171728,
      "step": 110610
    },
    {
      "epoch": 16.475275543640155,
      "grad_norm": 10.25,
      "learning_rate": 4.583642309988229e-06,
      "loss": 0.9685,
      "num_input_tokens_seen": 64174672,
      "step": 110615
    },
    {
      "epoch": 16.476020256181116,
      "grad_norm": 9.125,
      "learning_rate": 4.581767153121922e-06,
      "loss": 0.8823,
      "num_input_tokens_seen": 64177648,
      "step": 110620
    },
    {
      "epoch": 16.476764968722073,
      "grad_norm": 10.0,
      "learning_rate": 4.579892341200911e-06,
      "loss": 0.9428,
      "num_input_tokens_seen": 64180528,
      "step": 110625
    },
    {
      "epoch": 16.477509681263033,
      "grad_norm": 8.9375,
      "learning_rate": 4.578017874256857e-06,
      "loss": 1.0414,
      "num_input_tokens_seen": 64183312,
      "step": 110630
    },
    {
      "epoch": 16.47825439380399,
      "grad_norm": 6.53125,
      "learning_rate": 4.5761437523214435e-06,
      "loss": 1.0268,
      "num_input_tokens_seen": 64186160,
      "step": 110635
    },
    {
      "epoch": 16.47899910634495,
      "grad_norm": 6.5,
      "learning_rate": 4.574269975426318e-06,
      "loss": 1.0862,
      "num_input_tokens_seen": 64189008,
      "step": 110640
    },
    {
      "epoch": 16.47974381888591,
      "grad_norm": 7.65625,
      "learning_rate": 4.572396543603147e-06,
      "loss": 1.0347,
      "num_input_tokens_seen": 64191824,
      "step": 110645
    },
    {
      "epoch": 16.480488531426868,
      "grad_norm": 10.0625,
      "learning_rate": 4.570523456883574e-06,
      "loss": 1.2258,
      "num_input_tokens_seen": 64194736,
      "step": 110650
    },
    {
      "epoch": 16.48123324396783,
      "grad_norm": 22.5,
      "learning_rate": 4.568650715299236e-06,
      "loss": 1.143,
      "num_input_tokens_seen": 64197392,
      "step": 110655
    },
    {
      "epoch": 16.48197795650879,
      "grad_norm": 11.4375,
      "learning_rate": 4.566778318881787e-06,
      "loss": 0.8863,
      "num_input_tokens_seen": 64200432,
      "step": 110660
    },
    {
      "epoch": 16.482722669049746,
      "grad_norm": 8.6875,
      "learning_rate": 4.56490626766285e-06,
      "loss": 0.9743,
      "num_input_tokens_seen": 64203568,
      "step": 110665
    },
    {
      "epoch": 16.483467381590707,
      "grad_norm": 8.6875,
      "learning_rate": 4.563034561674054e-06,
      "loss": 1.4973,
      "num_input_tokens_seen": 64206960,
      "step": 110670
    },
    {
      "epoch": 16.484212094131664,
      "grad_norm": 7.6875,
      "learning_rate": 4.561163200947008e-06,
      "loss": 1.1725,
      "num_input_tokens_seen": 64209936,
      "step": 110675
    },
    {
      "epoch": 16.484956806672624,
      "grad_norm": 11.0625,
      "learning_rate": 4.559292185513347e-06,
      "loss": 1.1606,
      "num_input_tokens_seen": 64212784,
      "step": 110680
    },
    {
      "epoch": 16.485701519213585,
      "grad_norm": 12.875,
      "learning_rate": 4.557421515404667e-06,
      "loss": 1.2286,
      "num_input_tokens_seen": 64215568,
      "step": 110685
    },
    {
      "epoch": 16.48644623175454,
      "grad_norm": 5.34375,
      "learning_rate": 4.555551190652568e-06,
      "loss": 0.7017,
      "num_input_tokens_seen": 64218448,
      "step": 110690
    },
    {
      "epoch": 16.487190944295502,
      "grad_norm": 15.9375,
      "learning_rate": 4.55368121128866e-06,
      "loss": 1.2436,
      "num_input_tokens_seen": 64221360,
      "step": 110695
    },
    {
      "epoch": 16.487935656836463,
      "grad_norm": 13.1875,
      "learning_rate": 4.55181157734452e-06,
      "loss": 1.4024,
      "num_input_tokens_seen": 64224208,
      "step": 110700
    },
    {
      "epoch": 16.48868036937742,
      "grad_norm": 10.625,
      "learning_rate": 4.549942288851747e-06,
      "loss": 0.9799,
      "num_input_tokens_seen": 64226992,
      "step": 110705
    },
    {
      "epoch": 16.48942508191838,
      "grad_norm": 11.875,
      "learning_rate": 4.5480733458419074e-06,
      "loss": 1.1801,
      "num_input_tokens_seen": 64230032,
      "step": 110710
    },
    {
      "epoch": 16.490169794459337,
      "grad_norm": 8.4375,
      "learning_rate": 4.5462047483465886e-06,
      "loss": 0.9485,
      "num_input_tokens_seen": 64233040,
      "step": 110715
    },
    {
      "epoch": 16.490914507000298,
      "grad_norm": 7.40625,
      "learning_rate": 4.5443364963973475e-06,
      "loss": 1.0715,
      "num_input_tokens_seen": 64236016,
      "step": 110720
    },
    {
      "epoch": 16.49165921954126,
      "grad_norm": 9.5625,
      "learning_rate": 4.542468590025756e-06,
      "loss": 0.8218,
      "num_input_tokens_seen": 64238640,
      "step": 110725
    },
    {
      "epoch": 16.492403932082215,
      "grad_norm": 6.3125,
      "learning_rate": 4.540601029263367e-06,
      "loss": 0.7971,
      "num_input_tokens_seen": 64241744,
      "step": 110730
    },
    {
      "epoch": 16.493148644623176,
      "grad_norm": 12.6875,
      "learning_rate": 4.538733814141729e-06,
      "loss": 1.0223,
      "num_input_tokens_seen": 64244336,
      "step": 110735
    },
    {
      "epoch": 16.493893357164133,
      "grad_norm": 9.0625,
      "learning_rate": 4.536866944692386e-06,
      "loss": 0.76,
      "num_input_tokens_seen": 64247408,
      "step": 110740
    },
    {
      "epoch": 16.494638069705093,
      "grad_norm": 7.90625,
      "learning_rate": 4.535000420946875e-06,
      "loss": 1.0827,
      "num_input_tokens_seen": 64250448,
      "step": 110745
    },
    {
      "epoch": 16.495382782246054,
      "grad_norm": 6.40625,
      "learning_rate": 4.533134242936735e-06,
      "loss": 0.9932,
      "num_input_tokens_seen": 64253008,
      "step": 110750
    },
    {
      "epoch": 16.49612749478701,
      "grad_norm": 14.125,
      "learning_rate": 4.531268410693488e-06,
      "loss": 1.1092,
      "num_input_tokens_seen": 64255856,
      "step": 110755
    },
    {
      "epoch": 16.49687220732797,
      "grad_norm": 12.25,
      "learning_rate": 4.52940292424866e-06,
      "loss": 1.0242,
      "num_input_tokens_seen": 64258704,
      "step": 110760
    },
    {
      "epoch": 16.497616919868932,
      "grad_norm": 5.40625,
      "learning_rate": 4.527537783633764e-06,
      "loss": 1.0032,
      "num_input_tokens_seen": 64262000,
      "step": 110765
    },
    {
      "epoch": 16.49836163240989,
      "grad_norm": 7.4375,
      "learning_rate": 4.525672988880308e-06,
      "loss": 0.9897,
      "num_input_tokens_seen": 64264944,
      "step": 110770
    },
    {
      "epoch": 16.49910634495085,
      "grad_norm": 10.375,
      "learning_rate": 4.5238085400198e-06,
      "loss": 1.0388,
      "num_input_tokens_seen": 64268080,
      "step": 110775
    },
    {
      "epoch": 16.499851057491806,
      "grad_norm": 7.75,
      "learning_rate": 4.521944437083731e-06,
      "loss": 1.0925,
      "num_input_tokens_seen": 64271216,
      "step": 110780
    },
    {
      "epoch": 16.500595770032767,
      "grad_norm": 6.71875,
      "learning_rate": 4.520080680103603e-06,
      "loss": 1.1124,
      "num_input_tokens_seen": 64273904,
      "step": 110785
    },
    {
      "epoch": 16.501340482573728,
      "grad_norm": 11.25,
      "learning_rate": 4.5182172691108996e-06,
      "loss": 0.9805,
      "num_input_tokens_seen": 64276880,
      "step": 110790
    },
    {
      "epoch": 16.502085195114685,
      "grad_norm": 7.1875,
      "learning_rate": 4.5163542041370965e-06,
      "loss": 0.7911,
      "num_input_tokens_seen": 64279472,
      "step": 110795
    },
    {
      "epoch": 16.502829907655645,
      "grad_norm": 10.4375,
      "learning_rate": 4.514491485213665e-06,
      "loss": 1.1965,
      "num_input_tokens_seen": 64282384,
      "step": 110800
    },
    {
      "epoch": 16.503574620196606,
      "grad_norm": 17.125,
      "learning_rate": 4.512629112372085e-06,
      "loss": 1.2852,
      "num_input_tokens_seen": 64285392,
      "step": 110805
    },
    {
      "epoch": 16.504319332737563,
      "grad_norm": 12.375,
      "learning_rate": 4.510767085643814e-06,
      "loss": 0.9008,
      "num_input_tokens_seen": 64288368,
      "step": 110810
    },
    {
      "epoch": 16.505064045278523,
      "grad_norm": 8.375,
      "learning_rate": 4.508905405060301e-06,
      "loss": 1.1178,
      "num_input_tokens_seen": 64291120,
      "step": 110815
    },
    {
      "epoch": 16.50580875781948,
      "grad_norm": 16.75,
      "learning_rate": 4.5070440706530135e-06,
      "loss": 0.8332,
      "num_input_tokens_seen": 64293808,
      "step": 110820
    },
    {
      "epoch": 16.50655347036044,
      "grad_norm": 16.375,
      "learning_rate": 4.505183082453382e-06,
      "loss": 1.3875,
      "num_input_tokens_seen": 64296720,
      "step": 110825
    },
    {
      "epoch": 16.5072981829014,
      "grad_norm": 7.78125,
      "learning_rate": 4.503322440492858e-06,
      "loss": 1.2201,
      "num_input_tokens_seen": 64299856,
      "step": 110830
    },
    {
      "epoch": 16.508042895442358,
      "grad_norm": 15.0625,
      "learning_rate": 4.501462144802862e-06,
      "loss": 1.2296,
      "num_input_tokens_seen": 64302640,
      "step": 110835
    },
    {
      "epoch": 16.50878760798332,
      "grad_norm": 3.234375,
      "learning_rate": 4.4996021954148375e-06,
      "loss": 0.8155,
      "num_input_tokens_seen": 64305296,
      "step": 110840
    },
    {
      "epoch": 16.50953232052428,
      "grad_norm": 10.1875,
      "learning_rate": 4.497742592360196e-06,
      "loss": 1.2433,
      "num_input_tokens_seen": 64308368,
      "step": 110845
    },
    {
      "epoch": 16.510277033065236,
      "grad_norm": 10.1875,
      "learning_rate": 4.495883335670351e-06,
      "loss": 1.2155,
      "num_input_tokens_seen": 64311440,
      "step": 110850
    },
    {
      "epoch": 16.511021745606197,
      "grad_norm": 12.5,
      "learning_rate": 4.494024425376722e-06,
      "loss": 1.0194,
      "num_input_tokens_seen": 64314352,
      "step": 110855
    },
    {
      "epoch": 16.511766458147154,
      "grad_norm": 9.6875,
      "learning_rate": 4.4921658615107106e-06,
      "loss": 1.1988,
      "num_input_tokens_seen": 64317328,
      "step": 110860
    },
    {
      "epoch": 16.512511170688114,
      "grad_norm": 14.0,
      "learning_rate": 4.490307644103717e-06,
      "loss": 1.2558,
      "num_input_tokens_seen": 64320560,
      "step": 110865
    },
    {
      "epoch": 16.513255883229075,
      "grad_norm": 13.5625,
      "learning_rate": 4.48844977318712e-06,
      "loss": 0.9317,
      "num_input_tokens_seen": 64323536,
      "step": 110870
    },
    {
      "epoch": 16.51400059577003,
      "grad_norm": 9.4375,
      "learning_rate": 4.486592248792323e-06,
      "loss": 1.1115,
      "num_input_tokens_seen": 64326192,
      "step": 110875
    },
    {
      "epoch": 16.514745308310992,
      "grad_norm": 7.8125,
      "learning_rate": 4.484735070950696e-06,
      "loss": 1.2381,
      "num_input_tokens_seen": 64329040,
      "step": 110880
    },
    {
      "epoch": 16.515490020851953,
      "grad_norm": 11.1875,
      "learning_rate": 4.482878239693628e-06,
      "loss": 1.0936,
      "num_input_tokens_seen": 64331984,
      "step": 110885
    },
    {
      "epoch": 16.51623473339291,
      "grad_norm": 16.5,
      "learning_rate": 4.481021755052476e-06,
      "loss": 1.3721,
      "num_input_tokens_seen": 64334960,
      "step": 110890
    },
    {
      "epoch": 16.51697944593387,
      "grad_norm": 3.578125,
      "learning_rate": 4.479165617058603e-06,
      "loss": 0.9223,
      "num_input_tokens_seen": 64338288,
      "step": 110895
    },
    {
      "epoch": 16.517724158474827,
      "grad_norm": 7.5625,
      "learning_rate": 4.4773098257433754e-06,
      "loss": 1.4426,
      "num_input_tokens_seen": 64341328,
      "step": 110900
    },
    {
      "epoch": 16.518468871015788,
      "grad_norm": 10.3125,
      "learning_rate": 4.4754543811381335e-06,
      "loss": 1.0594,
      "num_input_tokens_seen": 64344080,
      "step": 110905
    },
    {
      "epoch": 16.51921358355675,
      "grad_norm": 8.125,
      "learning_rate": 4.473599283274235e-06,
      "loss": 0.968,
      "num_input_tokens_seen": 64346896,
      "step": 110910
    },
    {
      "epoch": 16.519958296097705,
      "grad_norm": 6.34375,
      "learning_rate": 4.471744532183012e-06,
      "loss": 1.227,
      "num_input_tokens_seen": 64349584,
      "step": 110915
    },
    {
      "epoch": 16.520703008638666,
      "grad_norm": 10.3125,
      "learning_rate": 4.469890127895804e-06,
      "loss": 1.3139,
      "num_input_tokens_seen": 64352336,
      "step": 110920
    },
    {
      "epoch": 16.521447721179626,
      "grad_norm": 4.78125,
      "learning_rate": 4.468036070443938e-06,
      "loss": 1.1328,
      "num_input_tokens_seen": 64355408,
      "step": 110925
    },
    {
      "epoch": 16.522192433720583,
      "grad_norm": 17.875,
      "learning_rate": 4.466182359858734e-06,
      "loss": 1.9698,
      "num_input_tokens_seen": 64358512,
      "step": 110930
    },
    {
      "epoch": 16.522937146261544,
      "grad_norm": 12.5625,
      "learning_rate": 4.4643289961715076e-06,
      "loss": 1.1414,
      "num_input_tokens_seen": 64361648,
      "step": 110935
    },
    {
      "epoch": 16.5236818588025,
      "grad_norm": 6.78125,
      "learning_rate": 4.462475979413569e-06,
      "loss": 1.0442,
      "num_input_tokens_seen": 64364464,
      "step": 110940
    },
    {
      "epoch": 16.52442657134346,
      "grad_norm": 8.75,
      "learning_rate": 4.46062330961623e-06,
      "loss": 1.1589,
      "num_input_tokens_seen": 64367024,
      "step": 110945
    },
    {
      "epoch": 16.525171283884422,
      "grad_norm": 5.5625,
      "learning_rate": 4.458770986810776e-06,
      "loss": 0.8194,
      "num_input_tokens_seen": 64370096,
      "step": 110950
    },
    {
      "epoch": 16.52591599642538,
      "grad_norm": 8.9375,
      "learning_rate": 4.456919011028518e-06,
      "loss": 1.2848,
      "num_input_tokens_seen": 64373168,
      "step": 110955
    },
    {
      "epoch": 16.52666070896634,
      "grad_norm": 7.3125,
      "learning_rate": 4.4550673823007284e-06,
      "loss": 1.0635,
      "num_input_tokens_seen": 64375888,
      "step": 110960
    },
    {
      "epoch": 16.527405421507297,
      "grad_norm": 11.875,
      "learning_rate": 4.4532161006587e-06,
      "loss": 1.0354,
      "num_input_tokens_seen": 64378896,
      "step": 110965
    },
    {
      "epoch": 16.528150134048257,
      "grad_norm": 14.125,
      "learning_rate": 4.4513651661337e-06,
      "loss": 1.2271,
      "num_input_tokens_seen": 64381520,
      "step": 110970
    },
    {
      "epoch": 16.528894846589218,
      "grad_norm": 12.25,
      "learning_rate": 4.449514578757e-06,
      "loss": 1.1958,
      "num_input_tokens_seen": 64384240,
      "step": 110975
    },
    {
      "epoch": 16.529639559130175,
      "grad_norm": 9.5625,
      "learning_rate": 4.447664338559867e-06,
      "loss": 1.4056,
      "num_input_tokens_seen": 64387248,
      "step": 110980
    },
    {
      "epoch": 16.530384271671135,
      "grad_norm": 5.75,
      "learning_rate": 4.445814445573551e-06,
      "loss": 1.3545,
      "num_input_tokens_seen": 64390000,
      "step": 110985
    },
    {
      "epoch": 16.531128984212096,
      "grad_norm": 4.40625,
      "learning_rate": 4.443964899829317e-06,
      "loss": 1.1846,
      "num_input_tokens_seen": 64392848,
      "step": 110990
    },
    {
      "epoch": 16.531873696753053,
      "grad_norm": 19.875,
      "learning_rate": 4.442115701358401e-06,
      "loss": 1.0083,
      "num_input_tokens_seen": 64395696,
      "step": 110995
    },
    {
      "epoch": 16.532618409294013,
      "grad_norm": 11.0,
      "learning_rate": 4.440266850192049e-06,
      "loss": 0.99,
      "num_input_tokens_seen": 64398608,
      "step": 111000
    },
    {
      "epoch": 16.53336312183497,
      "grad_norm": 19.625,
      "learning_rate": 4.4384183463614865e-06,
      "loss": 1.2202,
      "num_input_tokens_seen": 64401520,
      "step": 111005
    },
    {
      "epoch": 16.53410783437593,
      "grad_norm": 14.4375,
      "learning_rate": 4.436570189897951e-06,
      "loss": 0.9625,
      "num_input_tokens_seen": 64404528,
      "step": 111010
    },
    {
      "epoch": 16.53485254691689,
      "grad_norm": 6.0625,
      "learning_rate": 4.434722380832665e-06,
      "loss": 0.8933,
      "num_input_tokens_seen": 64407536,
      "step": 111015
    },
    {
      "epoch": 16.535597259457848,
      "grad_norm": 4.5625,
      "learning_rate": 4.432874919196836e-06,
      "loss": 1.0168,
      "num_input_tokens_seen": 64410512,
      "step": 111020
    },
    {
      "epoch": 16.53634197199881,
      "grad_norm": 16.375,
      "learning_rate": 4.4310278050216895e-06,
      "loss": 1.0254,
      "num_input_tokens_seen": 64413488,
      "step": 111025
    },
    {
      "epoch": 16.53708668453977,
      "grad_norm": 4.40625,
      "learning_rate": 4.429181038338415e-06,
      "loss": 1.113,
      "num_input_tokens_seen": 64416272,
      "step": 111030
    },
    {
      "epoch": 16.537831397080726,
      "grad_norm": 8.375,
      "learning_rate": 4.427334619178225e-06,
      "loss": 1.0878,
      "num_input_tokens_seen": 64418736,
      "step": 111035
    },
    {
      "epoch": 16.538576109621687,
      "grad_norm": 18.25,
      "learning_rate": 4.425488547572304e-06,
      "loss": 1.2725,
      "num_input_tokens_seen": 64421616,
      "step": 111040
    },
    {
      "epoch": 16.539320822162644,
      "grad_norm": 13.3125,
      "learning_rate": 4.4236428235518465e-06,
      "loss": 1.5336,
      "num_input_tokens_seen": 64424528,
      "step": 111045
    },
    {
      "epoch": 16.540065534703604,
      "grad_norm": 8.1875,
      "learning_rate": 4.421797447148032e-06,
      "loss": 0.9095,
      "num_input_tokens_seen": 64427440,
      "step": 111050
    },
    {
      "epoch": 16.540810247244565,
      "grad_norm": 9.625,
      "learning_rate": 4.419952418392029e-06,
      "loss": 1.0525,
      "num_input_tokens_seen": 64430480,
      "step": 111055
    },
    {
      "epoch": 16.541554959785522,
      "grad_norm": 9.5,
      "learning_rate": 4.418107737315019e-06,
      "loss": 1.4467,
      "num_input_tokens_seen": 64433296,
      "step": 111060
    },
    {
      "epoch": 16.542299672326482,
      "grad_norm": 10.75,
      "learning_rate": 4.416263403948159e-06,
      "loss": 1.1168,
      "num_input_tokens_seen": 64436176,
      "step": 111065
    },
    {
      "epoch": 16.543044384867443,
      "grad_norm": 7.28125,
      "learning_rate": 4.41441941832261e-06,
      "loss": 1.0866,
      "num_input_tokens_seen": 64439440,
      "step": 111070
    },
    {
      "epoch": 16.5437890974084,
      "grad_norm": 5.28125,
      "learning_rate": 4.412575780469516e-06,
      "loss": 0.7245,
      "num_input_tokens_seen": 64442320,
      "step": 111075
    },
    {
      "epoch": 16.54453380994936,
      "grad_norm": 17.25,
      "learning_rate": 4.410732490420036e-06,
      "loss": 1.5117,
      "num_input_tokens_seen": 64445456,
      "step": 111080
    },
    {
      "epoch": 16.545278522490317,
      "grad_norm": 16.0,
      "learning_rate": 4.4088895482053e-06,
      "loss": 1.2695,
      "num_input_tokens_seen": 64448432,
      "step": 111085
    },
    {
      "epoch": 16.546023235031278,
      "grad_norm": 8.625,
      "learning_rate": 4.40704695385645e-06,
      "loss": 0.9865,
      "num_input_tokens_seen": 64451024,
      "step": 111090
    },
    {
      "epoch": 16.54676794757224,
      "grad_norm": 7.71875,
      "learning_rate": 4.405204707404614e-06,
      "loss": 0.8126,
      "num_input_tokens_seen": 64453936,
      "step": 111095
    },
    {
      "epoch": 16.547512660113195,
      "grad_norm": 10.75,
      "learning_rate": 4.403362808880909e-06,
      "loss": 1.0065,
      "num_input_tokens_seen": 64456720,
      "step": 111100
    },
    {
      "epoch": 16.548257372654156,
      "grad_norm": 8.125,
      "learning_rate": 4.40152125831646e-06,
      "loss": 1.1006,
      "num_input_tokens_seen": 64459728,
      "step": 111105
    },
    {
      "epoch": 16.549002085195113,
      "grad_norm": 13.3125,
      "learning_rate": 4.3996800557423665e-06,
      "loss": 1.391,
      "num_input_tokens_seen": 64462480,
      "step": 111110
    },
    {
      "epoch": 16.549746797736073,
      "grad_norm": 10.0,
      "learning_rate": 4.397839201189749e-06,
      "loss": 1.1061,
      "num_input_tokens_seen": 64465488,
      "step": 111115
    },
    {
      "epoch": 16.550491510277034,
      "grad_norm": 11.875,
      "learning_rate": 4.395998694689699e-06,
      "loss": 1.1778,
      "num_input_tokens_seen": 64468464,
      "step": 111120
    },
    {
      "epoch": 16.55123622281799,
      "grad_norm": 11.9375,
      "learning_rate": 4.39415853627331e-06,
      "loss": 1.0641,
      "num_input_tokens_seen": 64471088,
      "step": 111125
    },
    {
      "epoch": 16.55198093535895,
      "grad_norm": 23.125,
      "learning_rate": 4.3923187259716615e-06,
      "loss": 0.9624,
      "num_input_tokens_seen": 64473808,
      "step": 111130
    },
    {
      "epoch": 16.552725647899912,
      "grad_norm": 11.5,
      "learning_rate": 4.390479263815852e-06,
      "loss": 1.2171,
      "num_input_tokens_seen": 64476560,
      "step": 111135
    },
    {
      "epoch": 16.55347036044087,
      "grad_norm": 10.0,
      "learning_rate": 4.388640149836948e-06,
      "loss": 0.7006,
      "num_input_tokens_seen": 64479376,
      "step": 111140
    },
    {
      "epoch": 16.55421507298183,
      "grad_norm": 11.5,
      "learning_rate": 4.3868013840660135e-06,
      "loss": 0.8989,
      "num_input_tokens_seen": 64482032,
      "step": 111145
    },
    {
      "epoch": 16.554959785522787,
      "grad_norm": 5.59375,
      "learning_rate": 4.3849629665341255e-06,
      "loss": 1.0345,
      "num_input_tokens_seen": 64484752,
      "step": 111150
    },
    {
      "epoch": 16.555704498063747,
      "grad_norm": 6.78125,
      "learning_rate": 4.383124897272331e-06,
      "loss": 1.0799,
      "num_input_tokens_seen": 64487664,
      "step": 111155
    },
    {
      "epoch": 16.556449210604708,
      "grad_norm": 11.5625,
      "learning_rate": 4.381287176311694e-06,
      "loss": 0.9029,
      "num_input_tokens_seen": 64490448,
      "step": 111160
    },
    {
      "epoch": 16.557193923145665,
      "grad_norm": 21.75,
      "learning_rate": 4.379449803683247e-06,
      "loss": 0.9848,
      "num_input_tokens_seen": 64493168,
      "step": 111165
    },
    {
      "epoch": 16.557938635686625,
      "grad_norm": 16.5,
      "learning_rate": 4.377612779418041e-06,
      "loss": 1.3872,
      "num_input_tokens_seen": 64495856,
      "step": 111170
    },
    {
      "epoch": 16.558683348227586,
      "grad_norm": 14.625,
      "learning_rate": 4.375776103547114e-06,
      "loss": 1.0781,
      "num_input_tokens_seen": 64498672,
      "step": 111175
    },
    {
      "epoch": 16.559428060768543,
      "grad_norm": 12.25,
      "learning_rate": 4.373939776101476e-06,
      "loss": 0.9345,
      "num_input_tokens_seen": 64501552,
      "step": 111180
    },
    {
      "epoch": 16.560172773309503,
      "grad_norm": 6.6875,
      "learning_rate": 4.37210379711217e-06,
      "loss": 1.1599,
      "num_input_tokens_seen": 64505040,
      "step": 111185
    },
    {
      "epoch": 16.56091748585046,
      "grad_norm": 11.9375,
      "learning_rate": 4.370268166610206e-06,
      "loss": 1.191,
      "num_input_tokens_seen": 64507856,
      "step": 111190
    },
    {
      "epoch": 16.56166219839142,
      "grad_norm": 7.78125,
      "learning_rate": 4.368432884626594e-06,
      "loss": 1.1321,
      "num_input_tokens_seen": 64510928,
      "step": 111195
    },
    {
      "epoch": 16.56240691093238,
      "grad_norm": 9.3125,
      "learning_rate": 4.366597951192333e-06,
      "loss": 1.0196,
      "num_input_tokens_seen": 64513680,
      "step": 111200
    },
    {
      "epoch": 16.56315162347334,
      "grad_norm": 11.0,
      "learning_rate": 4.364763366338437e-06,
      "loss": 1.0667,
      "num_input_tokens_seen": 64516528,
      "step": 111205
    },
    {
      "epoch": 16.5638963360143,
      "grad_norm": 10.0,
      "learning_rate": 4.362929130095888e-06,
      "loss": 1.2679,
      "num_input_tokens_seen": 64519344,
      "step": 111210
    },
    {
      "epoch": 16.56464104855526,
      "grad_norm": 7.3125,
      "learning_rate": 4.361095242495672e-06,
      "loss": 0.7025,
      "num_input_tokens_seen": 64522000,
      "step": 111215
    },
    {
      "epoch": 16.565385761096216,
      "grad_norm": 8.9375,
      "learning_rate": 4.359261703568781e-06,
      "loss": 0.8913,
      "num_input_tokens_seen": 64524400,
      "step": 111220
    },
    {
      "epoch": 16.566130473637177,
      "grad_norm": 10.5625,
      "learning_rate": 4.357428513346179e-06,
      "loss": 1.0844,
      "num_input_tokens_seen": 64527184,
      "step": 111225
    },
    {
      "epoch": 16.566875186178134,
      "grad_norm": 11.25,
      "learning_rate": 4.35559567185885e-06,
      "loss": 1.1092,
      "num_input_tokens_seen": 64529968,
      "step": 111230
    },
    {
      "epoch": 16.567619898719094,
      "grad_norm": 5.8125,
      "learning_rate": 4.353763179137743e-06,
      "loss": 1.1764,
      "num_input_tokens_seen": 64532976,
      "step": 111235
    },
    {
      "epoch": 16.568364611260055,
      "grad_norm": 10.25,
      "learning_rate": 4.351931035213827e-06,
      "loss": 1.3093,
      "num_input_tokens_seen": 64535856,
      "step": 111240
    },
    {
      "epoch": 16.569109323801012,
      "grad_norm": 9.5625,
      "learning_rate": 4.350099240118047e-06,
      "loss": 0.9501,
      "num_input_tokens_seen": 64538768,
      "step": 111245
    },
    {
      "epoch": 16.569854036341972,
      "grad_norm": 11.875,
      "learning_rate": 4.348267793881358e-06,
      "loss": 1.4371,
      "num_input_tokens_seen": 64541616,
      "step": 111250
    },
    {
      "epoch": 16.57059874888293,
      "grad_norm": 6.71875,
      "learning_rate": 4.346436696534698e-06,
      "loss": 1.2845,
      "num_input_tokens_seen": 64544496,
      "step": 111255
    },
    {
      "epoch": 16.57134346142389,
      "grad_norm": 6.3125,
      "learning_rate": 4.344605948108993e-06,
      "loss": 0.9469,
      "num_input_tokens_seen": 64547408,
      "step": 111260
    },
    {
      "epoch": 16.57208817396485,
      "grad_norm": 9.1875,
      "learning_rate": 4.342775548635181e-06,
      "loss": 1.2572,
      "num_input_tokens_seen": 64550448,
      "step": 111265
    },
    {
      "epoch": 16.572832886505807,
      "grad_norm": 7.9375,
      "learning_rate": 4.340945498144175e-06,
      "loss": 1.0851,
      "num_input_tokens_seen": 64553552,
      "step": 111270
    },
    {
      "epoch": 16.573577599046768,
      "grad_norm": 12.0,
      "learning_rate": 4.3391157966669036e-06,
      "loss": 0.9868,
      "num_input_tokens_seen": 64556336,
      "step": 111275
    },
    {
      "epoch": 16.57432231158773,
      "grad_norm": 15.5625,
      "learning_rate": 4.337286444234265e-06,
      "loss": 1.2962,
      "num_input_tokens_seen": 64559408,
      "step": 111280
    },
    {
      "epoch": 16.575067024128685,
      "grad_norm": 11.5,
      "learning_rate": 4.335457440877177e-06,
      "loss": 1.1047,
      "num_input_tokens_seen": 64562096,
      "step": 111285
    },
    {
      "epoch": 16.575811736669646,
      "grad_norm": 7.71875,
      "learning_rate": 4.333628786626534e-06,
      "loss": 0.912,
      "num_input_tokens_seen": 64565264,
      "step": 111290
    },
    {
      "epoch": 16.576556449210603,
      "grad_norm": 15.125,
      "learning_rate": 4.331800481513223e-06,
      "loss": 1.0902,
      "num_input_tokens_seen": 64568144,
      "step": 111295
    },
    {
      "epoch": 16.577301161751564,
      "grad_norm": 9.0625,
      "learning_rate": 4.329972525568141e-06,
      "loss": 1.0384,
      "num_input_tokens_seen": 64570672,
      "step": 111300
    },
    {
      "epoch": 16.578045874292524,
      "grad_norm": 13.4375,
      "learning_rate": 4.3281449188221605e-06,
      "loss": 1.1326,
      "num_input_tokens_seen": 64573232,
      "step": 111305
    },
    {
      "epoch": 16.57879058683348,
      "grad_norm": 13.75,
      "learning_rate": 4.326317661306168e-06,
      "loss": 1.3256,
      "num_input_tokens_seen": 64575984,
      "step": 111310
    },
    {
      "epoch": 16.57953529937444,
      "grad_norm": 14.1875,
      "learning_rate": 4.324490753051019e-06,
      "loss": 1.1761,
      "num_input_tokens_seen": 64578832,
      "step": 111315
    },
    {
      "epoch": 16.580280011915402,
      "grad_norm": 18.5,
      "learning_rate": 4.322664194087591e-06,
      "loss": 1.3793,
      "num_input_tokens_seen": 64581712,
      "step": 111320
    },
    {
      "epoch": 16.58102472445636,
      "grad_norm": 8.625,
      "learning_rate": 4.320837984446738e-06,
      "loss": 1.1721,
      "num_input_tokens_seen": 64584528,
      "step": 111325
    },
    {
      "epoch": 16.58176943699732,
      "grad_norm": 29.375,
      "learning_rate": 4.319012124159308e-06,
      "loss": 1.4399,
      "num_input_tokens_seen": 64587376,
      "step": 111330
    },
    {
      "epoch": 16.582514149538277,
      "grad_norm": 13.0625,
      "learning_rate": 4.317186613256149e-06,
      "loss": 0.9869,
      "num_input_tokens_seen": 64590160,
      "step": 111335
    },
    {
      "epoch": 16.583258862079237,
      "grad_norm": 16.875,
      "learning_rate": 4.3153614517680965e-06,
      "loss": 1.2182,
      "num_input_tokens_seen": 64593072,
      "step": 111340
    },
    {
      "epoch": 16.584003574620198,
      "grad_norm": 9.5625,
      "learning_rate": 4.313536639725996e-06,
      "loss": 1.4162,
      "num_input_tokens_seen": 64596016,
      "step": 111345
    },
    {
      "epoch": 16.584748287161155,
      "grad_norm": 11.8125,
      "learning_rate": 4.311712177160662e-06,
      "loss": 1.001,
      "num_input_tokens_seen": 64598832,
      "step": 111350
    },
    {
      "epoch": 16.585492999702115,
      "grad_norm": 10.625,
      "learning_rate": 4.30988806410293e-06,
      "loss": 1.208,
      "num_input_tokens_seen": 64601872,
      "step": 111355
    },
    {
      "epoch": 16.586237712243076,
      "grad_norm": 7.03125,
      "learning_rate": 4.308064300583603e-06,
      "loss": 1.017,
      "num_input_tokens_seen": 64604848,
      "step": 111360
    },
    {
      "epoch": 16.586982424784033,
      "grad_norm": 14.6875,
      "learning_rate": 4.3062408866335085e-06,
      "loss": 1.3354,
      "num_input_tokens_seen": 64607792,
      "step": 111365
    },
    {
      "epoch": 16.587727137324993,
      "grad_norm": 12.5,
      "learning_rate": 4.30441782228344e-06,
      "loss": 0.9567,
      "num_input_tokens_seen": 64610576,
      "step": 111370
    },
    {
      "epoch": 16.58847184986595,
      "grad_norm": 13.3125,
      "learning_rate": 4.302595107564192e-06,
      "loss": 1.0497,
      "num_input_tokens_seen": 64613584,
      "step": 111375
    },
    {
      "epoch": 16.58921656240691,
      "grad_norm": 7.40625,
      "learning_rate": 4.300772742506571e-06,
      "loss": 0.9869,
      "num_input_tokens_seen": 64616624,
      "step": 111380
    },
    {
      "epoch": 16.58996127494787,
      "grad_norm": 6.4375,
      "learning_rate": 4.2989507271413515e-06,
      "loss": 0.7681,
      "num_input_tokens_seen": 64619760,
      "step": 111385
    },
    {
      "epoch": 16.59070598748883,
      "grad_norm": 9.25,
      "learning_rate": 4.297129061499324e-06,
      "loss": 1.0824,
      "num_input_tokens_seen": 64622896,
      "step": 111390
    },
    {
      "epoch": 16.59145070002979,
      "grad_norm": 6.53125,
      "learning_rate": 4.29530774561126e-06,
      "loss": 0.9487,
      "num_input_tokens_seen": 64625616,
      "step": 111395
    },
    {
      "epoch": 16.59219541257075,
      "grad_norm": 15.375,
      "learning_rate": 4.29348677950793e-06,
      "loss": 1.0786,
      "num_input_tokens_seen": 64628528,
      "step": 111400
    },
    {
      "epoch": 16.592940125111706,
      "grad_norm": 11.625,
      "learning_rate": 4.291666163220087e-06,
      "loss": 1.098,
      "num_input_tokens_seen": 64631376,
      "step": 111405
    },
    {
      "epoch": 16.593684837652667,
      "grad_norm": 14.25,
      "learning_rate": 4.289845896778505e-06,
      "loss": 0.9387,
      "num_input_tokens_seen": 64633968,
      "step": 111410
    },
    {
      "epoch": 16.594429550193624,
      "grad_norm": 9.1875,
      "learning_rate": 4.2880259802139276e-06,
      "loss": 1.1809,
      "num_input_tokens_seen": 64636784,
      "step": 111415
    },
    {
      "epoch": 16.595174262734584,
      "grad_norm": 8.0,
      "learning_rate": 4.286206413557092e-06,
      "loss": 0.8761,
      "num_input_tokens_seen": 64639760,
      "step": 111420
    },
    {
      "epoch": 16.595918975275545,
      "grad_norm": 6.53125,
      "learning_rate": 4.284387196838755e-06,
      "loss": 0.8542,
      "num_input_tokens_seen": 64642480,
      "step": 111425
    },
    {
      "epoch": 16.596663687816502,
      "grad_norm": 20.25,
      "learning_rate": 4.282568330089637e-06,
      "loss": 0.8706,
      "num_input_tokens_seen": 64645360,
      "step": 111430
    },
    {
      "epoch": 16.597408400357462,
      "grad_norm": 3.171875,
      "learning_rate": 4.280749813340473e-06,
      "loss": 0.9971,
      "num_input_tokens_seen": 64647952,
      "step": 111435
    },
    {
      "epoch": 16.598153112898423,
      "grad_norm": 10.8125,
      "learning_rate": 4.278931646621981e-06,
      "loss": 1.35,
      "num_input_tokens_seen": 64651152,
      "step": 111440
    },
    {
      "epoch": 16.59889782543938,
      "grad_norm": 9.1875,
      "learning_rate": 4.2771138299648825e-06,
      "loss": 0.9852,
      "num_input_tokens_seen": 64654224,
      "step": 111445
    },
    {
      "epoch": 16.59964253798034,
      "grad_norm": 11.0625,
      "learning_rate": 4.275296363399883e-06,
      "loss": 0.9895,
      "num_input_tokens_seen": 64657232,
      "step": 111450
    },
    {
      "epoch": 16.600387250521297,
      "grad_norm": 13.75,
      "learning_rate": 4.27347924695769e-06,
      "loss": 1.0293,
      "num_input_tokens_seen": 64660208,
      "step": 111455
    },
    {
      "epoch": 16.601131963062258,
      "grad_norm": 9.375,
      "learning_rate": 4.271662480668995e-06,
      "loss": 0.9454,
      "num_input_tokens_seen": 64663312,
      "step": 111460
    },
    {
      "epoch": 16.60187667560322,
      "grad_norm": 12.1875,
      "learning_rate": 4.269846064564498e-06,
      "loss": 1.0003,
      "num_input_tokens_seen": 64665808,
      "step": 111465
    },
    {
      "epoch": 16.602621388144176,
      "grad_norm": 8.875,
      "learning_rate": 4.268029998674883e-06,
      "loss": 0.9455,
      "num_input_tokens_seen": 64668752,
      "step": 111470
    },
    {
      "epoch": 16.603366100685136,
      "grad_norm": 6.59375,
      "learning_rate": 4.266214283030825e-06,
      "loss": 1.059,
      "num_input_tokens_seen": 64671952,
      "step": 111475
    },
    {
      "epoch": 16.604110813226093,
      "grad_norm": 8.25,
      "learning_rate": 4.2643989176630095e-06,
      "loss": 1.049,
      "num_input_tokens_seen": 64674832,
      "step": 111480
    },
    {
      "epoch": 16.604855525767054,
      "grad_norm": 6.71875,
      "learning_rate": 4.262583902602094e-06,
      "loss": 1.0763,
      "num_input_tokens_seen": 64677616,
      "step": 111485
    },
    {
      "epoch": 16.605600238308014,
      "grad_norm": 8.375,
      "learning_rate": 4.2607692378787535e-06,
      "loss": 1.2661,
      "num_input_tokens_seen": 64680784,
      "step": 111490
    },
    {
      "epoch": 16.60634495084897,
      "grad_norm": 3.78125,
      "learning_rate": 4.258954923523636e-06,
      "loss": 0.7384,
      "num_input_tokens_seen": 64683888,
      "step": 111495
    },
    {
      "epoch": 16.60708966338993,
      "grad_norm": 12.375,
      "learning_rate": 4.25714095956739e-06,
      "loss": 0.888,
      "num_input_tokens_seen": 64686704,
      "step": 111500
    },
    {
      "epoch": 16.607834375930892,
      "grad_norm": 10.625,
      "learning_rate": 4.255327346040672e-06,
      "loss": 0.9615,
      "num_input_tokens_seen": 64690000,
      "step": 111505
    },
    {
      "epoch": 16.60857908847185,
      "grad_norm": 10.375,
      "learning_rate": 4.253514082974108e-06,
      "loss": 1.0129,
      "num_input_tokens_seen": 64692880,
      "step": 111510
    },
    {
      "epoch": 16.60932380101281,
      "grad_norm": 7.21875,
      "learning_rate": 4.251701170398342e-06,
      "loss": 1.0512,
      "num_input_tokens_seen": 64695856,
      "step": 111515
    },
    {
      "epoch": 16.610068513553767,
      "grad_norm": 7.15625,
      "learning_rate": 4.2498886083439995e-06,
      "loss": 1.2568,
      "num_input_tokens_seen": 64698736,
      "step": 111520
    },
    {
      "epoch": 16.610813226094727,
      "grad_norm": 12.6875,
      "learning_rate": 4.2480763968416996e-06,
      "loss": 1.331,
      "num_input_tokens_seen": 64701392,
      "step": 111525
    },
    {
      "epoch": 16.611557938635688,
      "grad_norm": 10.25,
      "learning_rate": 4.246264535922051e-06,
      "loss": 1.2358,
      "num_input_tokens_seen": 64704560,
      "step": 111530
    },
    {
      "epoch": 16.612302651176645,
      "grad_norm": 11.5,
      "learning_rate": 4.244453025615674e-06,
      "loss": 1.0709,
      "num_input_tokens_seen": 64707792,
      "step": 111535
    },
    {
      "epoch": 16.613047363717605,
      "grad_norm": 7.78125,
      "learning_rate": 4.242641865953173e-06,
      "loss": 0.9897,
      "num_input_tokens_seen": 64710544,
      "step": 111540
    },
    {
      "epoch": 16.613792076258566,
      "grad_norm": 10.25,
      "learning_rate": 4.240831056965131e-06,
      "loss": 1.0486,
      "num_input_tokens_seen": 64713136,
      "step": 111545
    },
    {
      "epoch": 16.614536788799523,
      "grad_norm": 5.03125,
      "learning_rate": 4.239020598682155e-06,
      "loss": 1.0799,
      "num_input_tokens_seen": 64715792,
      "step": 111550
    },
    {
      "epoch": 16.615281501340483,
      "grad_norm": 3.90625,
      "learning_rate": 4.237210491134821e-06,
      "loss": 0.8726,
      "num_input_tokens_seen": 64718672,
      "step": 111555
    },
    {
      "epoch": 16.61602621388144,
      "grad_norm": 5.5625,
      "learning_rate": 4.23540073435372e-06,
      "loss": 0.981,
      "num_input_tokens_seen": 64721360,
      "step": 111560
    },
    {
      "epoch": 16.6167709264224,
      "grad_norm": 10.625,
      "learning_rate": 4.2335913283694126e-06,
      "loss": 1.0936,
      "num_input_tokens_seen": 64724208,
      "step": 111565
    },
    {
      "epoch": 16.61751563896336,
      "grad_norm": 6.65625,
      "learning_rate": 4.231782273212481e-06,
      "loss": 0.9282,
      "num_input_tokens_seen": 64727376,
      "step": 111570
    },
    {
      "epoch": 16.61826035150432,
      "grad_norm": 8.125,
      "learning_rate": 4.2299735689134784e-06,
      "loss": 1.2074,
      "num_input_tokens_seen": 64730032,
      "step": 111575
    },
    {
      "epoch": 16.61900506404528,
      "grad_norm": 10.125,
      "learning_rate": 4.228165215502958e-06,
      "loss": 1.0582,
      "num_input_tokens_seen": 64733040,
      "step": 111580
    },
    {
      "epoch": 16.61974977658624,
      "grad_norm": 7.8125,
      "learning_rate": 4.226357213011478e-06,
      "loss": 1.1381,
      "num_input_tokens_seen": 64736240,
      "step": 111585
    },
    {
      "epoch": 16.620494489127196,
      "grad_norm": 12.25,
      "learning_rate": 4.224549561469582e-06,
      "loss": 0.7996,
      "num_input_tokens_seen": 64739216,
      "step": 111590
    },
    {
      "epoch": 16.621239201668157,
      "grad_norm": 12.5,
      "learning_rate": 4.222742260907806e-06,
      "loss": 0.9768,
      "num_input_tokens_seen": 64742032,
      "step": 111595
    },
    {
      "epoch": 16.621983914209114,
      "grad_norm": 8.1875,
      "learning_rate": 4.220935311356675e-06,
      "loss": 1.2681,
      "num_input_tokens_seen": 64745136,
      "step": 111600
    },
    {
      "epoch": 16.622728626750074,
      "grad_norm": 9.0,
      "learning_rate": 4.219128712846729e-06,
      "loss": 0.7215,
      "num_input_tokens_seen": 64748112,
      "step": 111605
    },
    {
      "epoch": 16.623473339291035,
      "grad_norm": 15.1875,
      "learning_rate": 4.217322465408477e-06,
      "loss": 1.1199,
      "num_input_tokens_seen": 64751216,
      "step": 111610
    },
    {
      "epoch": 16.624218051831992,
      "grad_norm": 18.75,
      "learning_rate": 4.2155165690724476e-06,
      "loss": 1.3475,
      "num_input_tokens_seen": 64754480,
      "step": 111615
    },
    {
      "epoch": 16.624962764372953,
      "grad_norm": 9.125,
      "learning_rate": 4.213711023869138e-06,
      "loss": 0.9237,
      "num_input_tokens_seen": 64757264,
      "step": 111620
    },
    {
      "epoch": 16.62570747691391,
      "grad_norm": 7.6875,
      "learning_rate": 4.211905829829049e-06,
      "loss": 1.2867,
      "num_input_tokens_seen": 64760336,
      "step": 111625
    },
    {
      "epoch": 16.62645218945487,
      "grad_norm": 8.5625,
      "learning_rate": 4.21010098698269e-06,
      "loss": 0.8326,
      "num_input_tokens_seen": 64763024,
      "step": 111630
    },
    {
      "epoch": 16.62719690199583,
      "grad_norm": 15.75,
      "learning_rate": 4.208296495360539e-06,
      "loss": 0.9783,
      "num_input_tokens_seen": 64765744,
      "step": 111635
    },
    {
      "epoch": 16.627941614536788,
      "grad_norm": 23.625,
      "learning_rate": 4.206492354993094e-06,
      "loss": 1.3392,
      "num_input_tokens_seen": 64768656,
      "step": 111640
    },
    {
      "epoch": 16.628686327077748,
      "grad_norm": 6.6875,
      "learning_rate": 4.204688565910819e-06,
      "loss": 1.4132,
      "num_input_tokens_seen": 64771600,
      "step": 111645
    },
    {
      "epoch": 16.62943103961871,
      "grad_norm": 9.9375,
      "learning_rate": 4.202885128144202e-06,
      "loss": 0.9144,
      "num_input_tokens_seen": 64774416,
      "step": 111650
    },
    {
      "epoch": 16.630175752159666,
      "grad_norm": 14.3125,
      "learning_rate": 4.201082041723703e-06,
      "loss": 0.6995,
      "num_input_tokens_seen": 64777264,
      "step": 111655
    },
    {
      "epoch": 16.630920464700626,
      "grad_norm": 4.03125,
      "learning_rate": 4.1992793066797845e-06,
      "loss": 0.856,
      "num_input_tokens_seen": 64779952,
      "step": 111660
    },
    {
      "epoch": 16.631665177241583,
      "grad_norm": 16.125,
      "learning_rate": 4.197476923042901e-06,
      "loss": 1.4428,
      "num_input_tokens_seen": 64782704,
      "step": 111665
    },
    {
      "epoch": 16.632409889782544,
      "grad_norm": 10.3125,
      "learning_rate": 4.195674890843495e-06,
      "loss": 1.0093,
      "num_input_tokens_seen": 64786256,
      "step": 111670
    },
    {
      "epoch": 16.633154602323504,
      "grad_norm": 6.90625,
      "learning_rate": 4.193873210112026e-06,
      "loss": 0.9109,
      "num_input_tokens_seen": 64789424,
      "step": 111675
    },
    {
      "epoch": 16.63389931486446,
      "grad_norm": 10.0,
      "learning_rate": 4.192071880878914e-06,
      "loss": 1.0525,
      "num_input_tokens_seen": 64792208,
      "step": 111680
    },
    {
      "epoch": 16.63464402740542,
      "grad_norm": 13.5,
      "learning_rate": 4.1902709031746094e-06,
      "loss": 0.9347,
      "num_input_tokens_seen": 64794992,
      "step": 111685
    },
    {
      "epoch": 16.635388739946382,
      "grad_norm": 10.5,
      "learning_rate": 4.188470277029516e-06,
      "loss": 1.3699,
      "num_input_tokens_seen": 64797872,
      "step": 111690
    },
    {
      "epoch": 16.63613345248734,
      "grad_norm": 5.875,
      "learning_rate": 4.1866700024740745e-06,
      "loss": 1.1164,
      "num_input_tokens_seen": 64800624,
      "step": 111695
    },
    {
      "epoch": 16.6368781650283,
      "grad_norm": 8.875,
      "learning_rate": 4.184870079538692e-06,
      "loss": 1.2812,
      "num_input_tokens_seen": 64803472,
      "step": 111700
    },
    {
      "epoch": 16.637622877569257,
      "grad_norm": 18.375,
      "learning_rate": 4.183070508253764e-06,
      "loss": 1.1785,
      "num_input_tokens_seen": 64806704,
      "step": 111705
    },
    {
      "epoch": 16.638367590110217,
      "grad_norm": 8.5,
      "learning_rate": 4.18127128864971e-06,
      "loss": 1.0343,
      "num_input_tokens_seen": 64809552,
      "step": 111710
    },
    {
      "epoch": 16.639112302651178,
      "grad_norm": 12.0,
      "learning_rate": 4.179472420756911e-06,
      "loss": 1.018,
      "num_input_tokens_seen": 64812304,
      "step": 111715
    },
    {
      "epoch": 16.639857015192135,
      "grad_norm": 13.0,
      "learning_rate": 4.177673904605773e-06,
      "loss": 0.9455,
      "num_input_tokens_seen": 64815472,
      "step": 111720
    },
    {
      "epoch": 16.640601727733095,
      "grad_norm": 12.5,
      "learning_rate": 4.1758757402266675e-06,
      "loss": 0.823,
      "num_input_tokens_seen": 64818032,
      "step": 111725
    },
    {
      "epoch": 16.641346440274056,
      "grad_norm": 4.375,
      "learning_rate": 4.1740779276499805e-06,
      "loss": 1.1088,
      "num_input_tokens_seen": 64820816,
      "step": 111730
    },
    {
      "epoch": 16.642091152815013,
      "grad_norm": 5.90625,
      "learning_rate": 4.172280466906079e-06,
      "loss": 1.0919,
      "num_input_tokens_seen": 64823664,
      "step": 111735
    },
    {
      "epoch": 16.642835865355973,
      "grad_norm": 10.9375,
      "learning_rate": 4.170483358025323e-06,
      "loss": 0.8672,
      "num_input_tokens_seen": 64826608,
      "step": 111740
    },
    {
      "epoch": 16.64358057789693,
      "grad_norm": 12.1875,
      "learning_rate": 4.168686601038091e-06,
      "loss": 1.0876,
      "num_input_tokens_seen": 64829456,
      "step": 111745
    },
    {
      "epoch": 16.64432529043789,
      "grad_norm": 9.25,
      "learning_rate": 4.1668901959747155e-06,
      "loss": 1.2736,
      "num_input_tokens_seen": 64832432,
      "step": 111750
    },
    {
      "epoch": 16.64507000297885,
      "grad_norm": 8.125,
      "learning_rate": 4.165094142865566e-06,
      "loss": 1.2316,
      "num_input_tokens_seen": 64835024,
      "step": 111755
    },
    {
      "epoch": 16.64581471551981,
      "grad_norm": 11.4375,
      "learning_rate": 4.163298441740968e-06,
      "loss": 0.8366,
      "num_input_tokens_seen": 64837968,
      "step": 111760
    },
    {
      "epoch": 16.64655942806077,
      "grad_norm": 11.4375,
      "learning_rate": 4.161503092631272e-06,
      "loss": 0.8904,
      "num_input_tokens_seen": 64840720,
      "step": 111765
    },
    {
      "epoch": 16.647304140601726,
      "grad_norm": 9.6875,
      "learning_rate": 4.159708095566794e-06,
      "loss": 1.2912,
      "num_input_tokens_seen": 64843568,
      "step": 111770
    },
    {
      "epoch": 16.648048853142686,
      "grad_norm": 12.0,
      "learning_rate": 4.157913450577875e-06,
      "loss": 1.2662,
      "num_input_tokens_seen": 64847024,
      "step": 111775
    },
    {
      "epoch": 16.648793565683647,
      "grad_norm": 15.3125,
      "learning_rate": 4.1561191576948235e-06,
      "loss": 1.3995,
      "num_input_tokens_seen": 64849936,
      "step": 111780
    },
    {
      "epoch": 16.649538278224604,
      "grad_norm": 14.75,
      "learning_rate": 4.1543252169479546e-06,
      "loss": 0.8577,
      "num_input_tokens_seen": 64852944,
      "step": 111785
    },
    {
      "epoch": 16.650282990765565,
      "grad_norm": 12.3125,
      "learning_rate": 4.15253162836757e-06,
      "loss": 1.1163,
      "num_input_tokens_seen": 64855856,
      "step": 111790
    },
    {
      "epoch": 16.651027703306525,
      "grad_norm": 11.75,
      "learning_rate": 4.1507383919839795e-06,
      "loss": 0.8531,
      "num_input_tokens_seen": 64858640,
      "step": 111795
    },
    {
      "epoch": 16.651772415847482,
      "grad_norm": 11.5625,
      "learning_rate": 4.148945507827476e-06,
      "loss": 1.1762,
      "num_input_tokens_seen": 64861616,
      "step": 111800
    },
    {
      "epoch": 16.652517128388443,
      "grad_norm": 14.25,
      "learning_rate": 4.147152975928336e-06,
      "loss": 1.3438,
      "num_input_tokens_seen": 64864592,
      "step": 111805
    },
    {
      "epoch": 16.6532618409294,
      "grad_norm": 7.59375,
      "learning_rate": 4.1453607963168604e-06,
      "loss": 0.9878,
      "num_input_tokens_seen": 64867440,
      "step": 111810
    },
    {
      "epoch": 16.65400655347036,
      "grad_norm": 6.96875,
      "learning_rate": 4.1435689690233205e-06,
      "loss": 1.1704,
      "num_input_tokens_seen": 64870256,
      "step": 111815
    },
    {
      "epoch": 16.65475126601132,
      "grad_norm": 5.03125,
      "learning_rate": 4.141777494077978e-06,
      "loss": 0.7965,
      "num_input_tokens_seen": 64873232,
      "step": 111820
    },
    {
      "epoch": 16.655495978552278,
      "grad_norm": 8.4375,
      "learning_rate": 4.139986371511109e-06,
      "loss": 1.0742,
      "num_input_tokens_seen": 64876304,
      "step": 111825
    },
    {
      "epoch": 16.656240691093238,
      "grad_norm": 10.5,
      "learning_rate": 4.138195601352968e-06,
      "loss": 1.0895,
      "num_input_tokens_seen": 64879056,
      "step": 111830
    },
    {
      "epoch": 16.6569854036342,
      "grad_norm": 13.9375,
      "learning_rate": 4.1364051836338125e-06,
      "loss": 1.5585,
      "num_input_tokens_seen": 64882064,
      "step": 111835
    },
    {
      "epoch": 16.657730116175156,
      "grad_norm": 9.8125,
      "learning_rate": 4.134615118383878e-06,
      "loss": 0.6666,
      "num_input_tokens_seen": 64885040,
      "step": 111840
    },
    {
      "epoch": 16.658474828716116,
      "grad_norm": 7.34375,
      "learning_rate": 4.132825405633425e-06,
      "loss": 1.3347,
      "num_input_tokens_seen": 64888208,
      "step": 111845
    },
    {
      "epoch": 16.659219541257073,
      "grad_norm": 8.25,
      "learning_rate": 4.131036045412675e-06,
      "loss": 1.3195,
      "num_input_tokens_seen": 64891088,
      "step": 111850
    },
    {
      "epoch": 16.659964253798034,
      "grad_norm": 14.6875,
      "learning_rate": 4.1292470377518625e-06,
      "loss": 1.1494,
      "num_input_tokens_seen": 64893904,
      "step": 111855
    },
    {
      "epoch": 16.660708966338994,
      "grad_norm": 14.4375,
      "learning_rate": 4.12745838268121e-06,
      "loss": 1.628,
      "num_input_tokens_seen": 64896912,
      "step": 111860
    },
    {
      "epoch": 16.66145367887995,
      "grad_norm": 8.5,
      "learning_rate": 4.125670080230926e-06,
      "loss": 1.0501,
      "num_input_tokens_seen": 64900016,
      "step": 111865
    },
    {
      "epoch": 16.66219839142091,
      "grad_norm": 10.9375,
      "learning_rate": 4.123882130431236e-06,
      "loss": 1.1693,
      "num_input_tokens_seen": 64902896,
      "step": 111870
    },
    {
      "epoch": 16.662943103961872,
      "grad_norm": 13.8125,
      "learning_rate": 4.122094533312337e-06,
      "loss": 1.0052,
      "num_input_tokens_seen": 64905840,
      "step": 111875
    },
    {
      "epoch": 16.66368781650283,
      "grad_norm": 11.6875,
      "learning_rate": 4.120307288904435e-06,
      "loss": 0.9823,
      "num_input_tokens_seen": 64908656,
      "step": 111880
    },
    {
      "epoch": 16.66443252904379,
      "grad_norm": 7.0,
      "learning_rate": 4.118520397237715e-06,
      "loss": 1.2405,
      "num_input_tokens_seen": 64911600,
      "step": 111885
    },
    {
      "epoch": 16.665177241584747,
      "grad_norm": 8.5,
      "learning_rate": 4.1167338583423755e-06,
      "loss": 1.2127,
      "num_input_tokens_seen": 64914384,
      "step": 111890
    },
    {
      "epoch": 16.665921954125707,
      "grad_norm": 17.5,
      "learning_rate": 4.114947672248593e-06,
      "loss": 1.123,
      "num_input_tokens_seen": 64917232,
      "step": 111895
    },
    {
      "epoch": 16.666666666666668,
      "grad_norm": 9.25,
      "learning_rate": 4.113161838986537e-06,
      "loss": 1.0317,
      "num_input_tokens_seen": 64919888,
      "step": 111900
    },
    {
      "epoch": 16.667411379207625,
      "grad_norm": 7.4375,
      "learning_rate": 4.111376358586388e-06,
      "loss": 1.2962,
      "num_input_tokens_seen": 64922800,
      "step": 111905
    },
    {
      "epoch": 16.668156091748585,
      "grad_norm": 10.9375,
      "learning_rate": 4.109591231078303e-06,
      "loss": 1.1402,
      "num_input_tokens_seen": 64925552,
      "step": 111910
    },
    {
      "epoch": 16.668900804289546,
      "grad_norm": 6.125,
      "learning_rate": 4.107806456492444e-06,
      "loss": 1.1037,
      "num_input_tokens_seen": 64928432,
      "step": 111915
    },
    {
      "epoch": 16.669645516830503,
      "grad_norm": 7.375,
      "learning_rate": 4.106022034858961e-06,
      "loss": 0.9364,
      "num_input_tokens_seen": 64931248,
      "step": 111920
    },
    {
      "epoch": 16.670390229371463,
      "grad_norm": 14.875,
      "learning_rate": 4.104237966208002e-06,
      "loss": 1.3046,
      "num_input_tokens_seen": 64934320,
      "step": 111925
    },
    {
      "epoch": 16.67113494191242,
      "grad_norm": 8.3125,
      "learning_rate": 4.102454250569698e-06,
      "loss": 1.1094,
      "num_input_tokens_seen": 64937072,
      "step": 111930
    },
    {
      "epoch": 16.67187965445338,
      "grad_norm": 12.4375,
      "learning_rate": 4.1006708879741975e-06,
      "loss": 1.0573,
      "num_input_tokens_seen": 64939984,
      "step": 111935
    },
    {
      "epoch": 16.67262436699434,
      "grad_norm": 17.125,
      "learning_rate": 4.098887878451621e-06,
      "loss": 0.8672,
      "num_input_tokens_seen": 64942800,
      "step": 111940
    },
    {
      "epoch": 16.6733690795353,
      "grad_norm": 11.25,
      "learning_rate": 4.097105222032083e-06,
      "loss": 0.7307,
      "num_input_tokens_seen": 64945488,
      "step": 111945
    },
    {
      "epoch": 16.67411379207626,
      "grad_norm": 9.6875,
      "learning_rate": 4.095322918745717e-06,
      "loss": 1.2757,
      "num_input_tokens_seen": 64948368,
      "step": 111950
    },
    {
      "epoch": 16.67485850461722,
      "grad_norm": 7.21875,
      "learning_rate": 4.093540968622614e-06,
      "loss": 1.0767,
      "num_input_tokens_seen": 64951216,
      "step": 111955
    },
    {
      "epoch": 16.675603217158177,
      "grad_norm": 7.09375,
      "learning_rate": 4.091759371692896e-06,
      "loss": 0.735,
      "num_input_tokens_seen": 64954352,
      "step": 111960
    },
    {
      "epoch": 16.676347929699137,
      "grad_norm": 12.75,
      "learning_rate": 4.089978127986646e-06,
      "loss": 1.1038,
      "num_input_tokens_seen": 64957232,
      "step": 111965
    },
    {
      "epoch": 16.677092642240094,
      "grad_norm": 10.125,
      "learning_rate": 4.08819723753397e-06,
      "loss": 1.0447,
      "num_input_tokens_seen": 64960240,
      "step": 111970
    },
    {
      "epoch": 16.677837354781055,
      "grad_norm": 12.9375,
      "learning_rate": 4.086416700364948e-06,
      "loss": 1.0001,
      "num_input_tokens_seen": 64963280,
      "step": 111975
    },
    {
      "epoch": 16.678582067322015,
      "grad_norm": 9.25,
      "learning_rate": 4.084636516509654e-06,
      "loss": 1.2723,
      "num_input_tokens_seen": 64966352,
      "step": 111980
    },
    {
      "epoch": 16.679326779862972,
      "grad_norm": 13.5625,
      "learning_rate": 4.0828566859981765e-06,
      "loss": 1.069,
      "num_input_tokens_seen": 64969392,
      "step": 111985
    },
    {
      "epoch": 16.680071492403933,
      "grad_norm": 17.625,
      "learning_rate": 4.081077208860573e-06,
      "loss": 1.5223,
      "num_input_tokens_seen": 64972304,
      "step": 111990
    },
    {
      "epoch": 16.68081620494489,
      "grad_norm": 10.375,
      "learning_rate": 4.079298085126912e-06,
      "loss": 0.9552,
      "num_input_tokens_seen": 64975664,
      "step": 111995
    },
    {
      "epoch": 16.68156091748585,
      "grad_norm": 8.0,
      "learning_rate": 4.0775193148272385e-06,
      "loss": 1.2603,
      "num_input_tokens_seen": 64978640,
      "step": 112000
    },
    {
      "epoch": 16.68230563002681,
      "grad_norm": 7.28125,
      "learning_rate": 4.07574089799162e-06,
      "loss": 0.9014,
      "num_input_tokens_seen": 64981904,
      "step": 112005
    },
    {
      "epoch": 16.683050342567768,
      "grad_norm": 7.40625,
      "learning_rate": 4.073962834650083e-06,
      "loss": 1.0219,
      "num_input_tokens_seen": 64984656,
      "step": 112010
    },
    {
      "epoch": 16.683795055108728,
      "grad_norm": 10.3125,
      "learning_rate": 4.072185124832684e-06,
      "loss": 0.7216,
      "num_input_tokens_seen": 64987248,
      "step": 112015
    },
    {
      "epoch": 16.68453976764969,
      "grad_norm": 7.6875,
      "learning_rate": 4.070407768569448e-06,
      "loss": 1.1174,
      "num_input_tokens_seen": 64990096,
      "step": 112020
    },
    {
      "epoch": 16.685284480190646,
      "grad_norm": 11.3125,
      "learning_rate": 4.068630765890393e-06,
      "loss": 1.103,
      "num_input_tokens_seen": 64992784,
      "step": 112025
    },
    {
      "epoch": 16.686029192731606,
      "grad_norm": 10.0625,
      "learning_rate": 4.0668541168255556e-06,
      "loss": 1.1137,
      "num_input_tokens_seen": 64995536,
      "step": 112030
    },
    {
      "epoch": 16.686773905272563,
      "grad_norm": 6.75,
      "learning_rate": 4.065077821404934e-06,
      "loss": 0.7328,
      "num_input_tokens_seen": 64998448,
      "step": 112035
    },
    {
      "epoch": 16.687518617813524,
      "grad_norm": 11.875,
      "learning_rate": 4.0633018796585514e-06,
      "loss": 1.2205,
      "num_input_tokens_seen": 65001040,
      "step": 112040
    },
    {
      "epoch": 16.688263330354484,
      "grad_norm": 21.5,
      "learning_rate": 4.061526291616399e-06,
      "loss": 1.5586,
      "num_input_tokens_seen": 65003984,
      "step": 112045
    },
    {
      "epoch": 16.68900804289544,
      "grad_norm": 11.0,
      "learning_rate": 4.059751057308486e-06,
      "loss": 1.2446,
      "num_input_tokens_seen": 65007056,
      "step": 112050
    },
    {
      "epoch": 16.689752755436402,
      "grad_norm": 10.1875,
      "learning_rate": 4.057976176764797e-06,
      "loss": 1.271,
      "num_input_tokens_seen": 65009968,
      "step": 112055
    },
    {
      "epoch": 16.690497467977362,
      "grad_norm": 23.75,
      "learning_rate": 4.056201650015315e-06,
      "loss": 1.1583,
      "num_input_tokens_seen": 65012976,
      "step": 112060
    },
    {
      "epoch": 16.69124218051832,
      "grad_norm": 21.125,
      "learning_rate": 4.054427477090019e-06,
      "loss": 1.2486,
      "num_input_tokens_seen": 65016144,
      "step": 112065
    },
    {
      "epoch": 16.69198689305928,
      "grad_norm": 8.8125,
      "learning_rate": 4.0526536580188766e-06,
      "loss": 0.8964,
      "num_input_tokens_seen": 65018960,
      "step": 112070
    },
    {
      "epoch": 16.692731605600237,
      "grad_norm": 16.625,
      "learning_rate": 4.050880192831868e-06,
      "loss": 1.0567,
      "num_input_tokens_seen": 65021872,
      "step": 112075
    },
    {
      "epoch": 16.693476318141197,
      "grad_norm": 9.25,
      "learning_rate": 4.04910708155894e-06,
      "loss": 1.1397,
      "num_input_tokens_seen": 65024656,
      "step": 112080
    },
    {
      "epoch": 16.694221030682158,
      "grad_norm": 7.09375,
      "learning_rate": 4.047334324230059e-06,
      "loss": 0.9111,
      "num_input_tokens_seen": 65027536,
      "step": 112085
    },
    {
      "epoch": 16.694965743223115,
      "grad_norm": 17.125,
      "learning_rate": 4.045561920875165e-06,
      "loss": 0.9544,
      "num_input_tokens_seen": 65030576,
      "step": 112090
    },
    {
      "epoch": 16.695710455764075,
      "grad_norm": 18.75,
      "learning_rate": 4.043789871524212e-06,
      "loss": 1.0871,
      "num_input_tokens_seen": 65033456,
      "step": 112095
    },
    {
      "epoch": 16.696455168305036,
      "grad_norm": 17.0,
      "learning_rate": 4.042018176207127e-06,
      "loss": 1.1791,
      "num_input_tokens_seen": 65036464,
      "step": 112100
    },
    {
      "epoch": 16.697199880845993,
      "grad_norm": 8.6875,
      "learning_rate": 4.04024683495384e-06,
      "loss": 0.8682,
      "num_input_tokens_seen": 65039696,
      "step": 112105
    },
    {
      "epoch": 16.697944593386953,
      "grad_norm": 12.5625,
      "learning_rate": 4.038475847794287e-06,
      "loss": 1.0789,
      "num_input_tokens_seen": 65042224,
      "step": 112110
    },
    {
      "epoch": 16.69868930592791,
      "grad_norm": 15.375,
      "learning_rate": 4.036705214758379e-06,
      "loss": 0.8326,
      "num_input_tokens_seen": 65045104,
      "step": 112115
    },
    {
      "epoch": 16.69943401846887,
      "grad_norm": 9.375,
      "learning_rate": 4.0349349358760255e-06,
      "loss": 1.5566,
      "num_input_tokens_seen": 65047984,
      "step": 112120
    },
    {
      "epoch": 16.70017873100983,
      "grad_norm": 17.5,
      "learning_rate": 4.0331650111771426e-06,
      "loss": 1.3207,
      "num_input_tokens_seen": 65050832,
      "step": 112125
    },
    {
      "epoch": 16.70092344355079,
      "grad_norm": 7.125,
      "learning_rate": 4.031395440691629e-06,
      "loss": 0.8103,
      "num_input_tokens_seen": 65053744,
      "step": 112130
    },
    {
      "epoch": 16.70166815609175,
      "grad_norm": 10.8125,
      "learning_rate": 4.02962622444937e-06,
      "loss": 1.1639,
      "num_input_tokens_seen": 65056688,
      "step": 112135
    },
    {
      "epoch": 16.702412868632706,
      "grad_norm": 13.25,
      "learning_rate": 4.0278573624802695e-06,
      "loss": 1.0683,
      "num_input_tokens_seen": 65059696,
      "step": 112140
    },
    {
      "epoch": 16.703157581173667,
      "grad_norm": 13.125,
      "learning_rate": 4.026088854814205e-06,
      "loss": 0.9757,
      "num_input_tokens_seen": 65062672,
      "step": 112145
    },
    {
      "epoch": 16.703902293714627,
      "grad_norm": 15.8125,
      "learning_rate": 4.024320701481044e-06,
      "loss": 1.1979,
      "num_input_tokens_seen": 65065648,
      "step": 112150
    },
    {
      "epoch": 16.704647006255584,
      "grad_norm": 7.375,
      "learning_rate": 4.0225529025106735e-06,
      "loss": 1.1846,
      "num_input_tokens_seen": 65068592,
      "step": 112155
    },
    {
      "epoch": 16.705391718796545,
      "grad_norm": 4.65625,
      "learning_rate": 4.020785457932946e-06,
      "loss": 0.8777,
      "num_input_tokens_seen": 65071632,
      "step": 112160
    },
    {
      "epoch": 16.706136431337505,
      "grad_norm": 11.625,
      "learning_rate": 4.01901836777773e-06,
      "loss": 1.2205,
      "num_input_tokens_seen": 65074480,
      "step": 112165
    },
    {
      "epoch": 16.706881143878462,
      "grad_norm": 17.25,
      "learning_rate": 4.01725163207487e-06,
      "loss": 0.8748,
      "num_input_tokens_seen": 65077200,
      "step": 112170
    },
    {
      "epoch": 16.707625856419423,
      "grad_norm": 7.59375,
      "learning_rate": 4.015485250854223e-06,
      "loss": 1.0073,
      "num_input_tokens_seen": 65079888,
      "step": 112175
    },
    {
      "epoch": 16.70837056896038,
      "grad_norm": 11.0,
      "learning_rate": 4.013719224145623e-06,
      "loss": 1.0187,
      "num_input_tokens_seen": 65082832,
      "step": 112180
    },
    {
      "epoch": 16.70911528150134,
      "grad_norm": 12.5625,
      "learning_rate": 4.011953551978911e-06,
      "loss": 0.9682,
      "num_input_tokens_seen": 65085520,
      "step": 112185
    },
    {
      "epoch": 16.7098599940423,
      "grad_norm": 15.125,
      "learning_rate": 4.0101882343839105e-06,
      "loss": 0.932,
      "num_input_tokens_seen": 65088624,
      "step": 112190
    },
    {
      "epoch": 16.710604706583258,
      "grad_norm": 7.84375,
      "learning_rate": 4.0084232713904405e-06,
      "loss": 1.0597,
      "num_input_tokens_seen": 65091600,
      "step": 112195
    },
    {
      "epoch": 16.71134941912422,
      "grad_norm": 14.0,
      "learning_rate": 4.006658663028331e-06,
      "loss": 1.1749,
      "num_input_tokens_seen": 65094512,
      "step": 112200
    },
    {
      "epoch": 16.71209413166518,
      "grad_norm": 11.5,
      "learning_rate": 4.004894409327381e-06,
      "loss": 1.1815,
      "num_input_tokens_seen": 65097360,
      "step": 112205
    },
    {
      "epoch": 16.712838844206136,
      "grad_norm": 13.0625,
      "learning_rate": 4.0031305103174076e-06,
      "loss": 1.1927,
      "num_input_tokens_seen": 65100464,
      "step": 112210
    },
    {
      "epoch": 16.713583556747096,
      "grad_norm": 15.0625,
      "learning_rate": 4.001366966028197e-06,
      "loss": 1.1985,
      "num_input_tokens_seen": 65103376,
      "step": 112215
    },
    {
      "epoch": 16.714328269288053,
      "grad_norm": 4.09375,
      "learning_rate": 3.999603776489555e-06,
      "loss": 1.27,
      "num_input_tokens_seen": 65106384,
      "step": 112220
    },
    {
      "epoch": 16.715072981829014,
      "grad_norm": 10.625,
      "learning_rate": 3.997840941731265e-06,
      "loss": 1.0635,
      "num_input_tokens_seen": 65108944,
      "step": 112225
    },
    {
      "epoch": 16.715817694369974,
      "grad_norm": 9.6875,
      "learning_rate": 3.996078461783098e-06,
      "loss": 0.9834,
      "num_input_tokens_seen": 65112048,
      "step": 112230
    },
    {
      "epoch": 16.71656240691093,
      "grad_norm": 6.875,
      "learning_rate": 3.994316336674847e-06,
      "loss": 1.0028,
      "num_input_tokens_seen": 65114640,
      "step": 112235
    },
    {
      "epoch": 16.717307119451892,
      "grad_norm": 6.34375,
      "learning_rate": 3.992554566436263e-06,
      "loss": 0.8399,
      "num_input_tokens_seen": 65117552,
      "step": 112240
    },
    {
      "epoch": 16.718051831992852,
      "grad_norm": 7.5,
      "learning_rate": 3.990793151097128e-06,
      "loss": 1.0222,
      "num_input_tokens_seen": 65120656,
      "step": 112245
    },
    {
      "epoch": 16.71879654453381,
      "grad_norm": 14.125,
      "learning_rate": 3.989032090687189e-06,
      "loss": 1.1287,
      "num_input_tokens_seen": 65123600,
      "step": 112250
    },
    {
      "epoch": 16.71954125707477,
      "grad_norm": 9.5625,
      "learning_rate": 3.987271385236197e-06,
      "loss": 1.1404,
      "num_input_tokens_seen": 65126288,
      "step": 112255
    },
    {
      "epoch": 16.720285969615727,
      "grad_norm": 13.125,
      "learning_rate": 3.9855110347739e-06,
      "loss": 1.0436,
      "num_input_tokens_seen": 65129136,
      "step": 112260
    },
    {
      "epoch": 16.721030682156687,
      "grad_norm": 7.03125,
      "learning_rate": 3.983751039330028e-06,
      "loss": 0.8529,
      "num_input_tokens_seen": 65132112,
      "step": 112265
    },
    {
      "epoch": 16.721775394697648,
      "grad_norm": 13.75,
      "learning_rate": 3.981991398934329e-06,
      "loss": 1.1052,
      "num_input_tokens_seen": 65134832,
      "step": 112270
    },
    {
      "epoch": 16.722520107238605,
      "grad_norm": 8.5625,
      "learning_rate": 3.980232113616519e-06,
      "loss": 1.2252,
      "num_input_tokens_seen": 65137648,
      "step": 112275
    },
    {
      "epoch": 16.723264819779565,
      "grad_norm": 3.875,
      "learning_rate": 3.978473183406328e-06,
      "loss": 1.0571,
      "num_input_tokens_seen": 65140368,
      "step": 112280
    },
    {
      "epoch": 16.724009532320522,
      "grad_norm": 11.1875,
      "learning_rate": 3.9767146083334625e-06,
      "loss": 0.976,
      "num_input_tokens_seen": 65143408,
      "step": 112285
    },
    {
      "epoch": 16.724754244861483,
      "grad_norm": 12.1875,
      "learning_rate": 3.974956388427642e-06,
      "loss": 0.7402,
      "num_input_tokens_seen": 65146256,
      "step": 112290
    },
    {
      "epoch": 16.725498957402444,
      "grad_norm": 19.125,
      "learning_rate": 3.97319852371856e-06,
      "loss": 1.1932,
      "num_input_tokens_seen": 65149392,
      "step": 112295
    },
    {
      "epoch": 16.7262436699434,
      "grad_norm": 5.90625,
      "learning_rate": 3.971441014235921e-06,
      "loss": 1.1514,
      "num_input_tokens_seen": 65152240,
      "step": 112300
    },
    {
      "epoch": 16.72698838248436,
      "grad_norm": 8.5625,
      "learning_rate": 3.969683860009415e-06,
      "loss": 0.8265,
      "num_input_tokens_seen": 65154992,
      "step": 112305
    },
    {
      "epoch": 16.72773309502532,
      "grad_norm": 8.4375,
      "learning_rate": 3.967927061068721e-06,
      "loss": 1.0664,
      "num_input_tokens_seen": 65157840,
      "step": 112310
    },
    {
      "epoch": 16.72847780756628,
      "grad_norm": 17.375,
      "learning_rate": 3.966170617443529e-06,
      "loss": 1.3588,
      "num_input_tokens_seen": 65160528,
      "step": 112315
    },
    {
      "epoch": 16.72922252010724,
      "grad_norm": 9.0625,
      "learning_rate": 3.964414529163507e-06,
      "loss": 0.8576,
      "num_input_tokens_seen": 65163120,
      "step": 112320
    },
    {
      "epoch": 16.7299672326482,
      "grad_norm": 15.6875,
      "learning_rate": 3.96265879625832e-06,
      "loss": 1.3193,
      "num_input_tokens_seen": 65166064,
      "step": 112325
    },
    {
      "epoch": 16.730711945189157,
      "grad_norm": 13.3125,
      "learning_rate": 3.960903418757628e-06,
      "loss": 1.039,
      "num_input_tokens_seen": 65169168,
      "step": 112330
    },
    {
      "epoch": 16.731456657730117,
      "grad_norm": 11.9375,
      "learning_rate": 3.959148396691092e-06,
      "loss": 1.1762,
      "num_input_tokens_seen": 65172304,
      "step": 112335
    },
    {
      "epoch": 16.732201370271074,
      "grad_norm": 11.4375,
      "learning_rate": 3.957393730088363e-06,
      "loss": 1.0018,
      "num_input_tokens_seen": 65175344,
      "step": 112340
    },
    {
      "epoch": 16.732946082812035,
      "grad_norm": 7.03125,
      "learning_rate": 3.9556394189790705e-06,
      "loss": 0.7881,
      "num_input_tokens_seen": 65178288,
      "step": 112345
    },
    {
      "epoch": 16.733690795352995,
      "grad_norm": 6.0625,
      "learning_rate": 3.95388546339287e-06,
      "loss": 1.1553,
      "num_input_tokens_seen": 65181296,
      "step": 112350
    },
    {
      "epoch": 16.734435507893952,
      "grad_norm": 12.5,
      "learning_rate": 3.9521318633593765e-06,
      "loss": 0.9898,
      "num_input_tokens_seen": 65184080,
      "step": 112355
    },
    {
      "epoch": 16.735180220434913,
      "grad_norm": 11.375,
      "learning_rate": 3.950378618908232e-06,
      "loss": 1.1172,
      "num_input_tokens_seen": 65186832,
      "step": 112360
    },
    {
      "epoch": 16.73592493297587,
      "grad_norm": 9.875,
      "learning_rate": 3.948625730069039e-06,
      "loss": 1.0278,
      "num_input_tokens_seen": 65189840,
      "step": 112365
    },
    {
      "epoch": 16.73666964551683,
      "grad_norm": 7.03125,
      "learning_rate": 3.946873196871423e-06,
      "loss": 1.0272,
      "num_input_tokens_seen": 65192880,
      "step": 112370
    },
    {
      "epoch": 16.73741435805779,
      "grad_norm": 9.0,
      "learning_rate": 3.945121019344983e-06,
      "loss": 0.9068,
      "num_input_tokens_seen": 65195728,
      "step": 112375
    },
    {
      "epoch": 16.738159070598748,
      "grad_norm": 9.875,
      "learning_rate": 3.943369197519328e-06,
      "loss": 1.2678,
      "num_input_tokens_seen": 65198512,
      "step": 112380
    },
    {
      "epoch": 16.73890378313971,
      "grad_norm": 13.3125,
      "learning_rate": 3.941617731424052e-06,
      "loss": 1.1724,
      "num_input_tokens_seen": 65201200,
      "step": 112385
    },
    {
      "epoch": 16.73964849568067,
      "grad_norm": 12.625,
      "learning_rate": 3.9398666210887395e-06,
      "loss": 0.7931,
      "num_input_tokens_seen": 65204080,
      "step": 112390
    },
    {
      "epoch": 16.740393208221626,
      "grad_norm": 6.9375,
      "learning_rate": 3.938115866542977e-06,
      "loss": 1.4064,
      "num_input_tokens_seen": 65207216,
      "step": 112395
    },
    {
      "epoch": 16.741137920762586,
      "grad_norm": 6.34375,
      "learning_rate": 3.936365467816333e-06,
      "loss": 0.7813,
      "num_input_tokens_seen": 65209968,
      "step": 112400
    },
    {
      "epoch": 16.741882633303543,
      "grad_norm": 7.375,
      "learning_rate": 3.934615424938395e-06,
      "loss": 1.1013,
      "num_input_tokens_seen": 65213008,
      "step": 112405
    },
    {
      "epoch": 16.742627345844504,
      "grad_norm": 8.75,
      "learning_rate": 3.932865737938712e-06,
      "loss": 0.8943,
      "num_input_tokens_seen": 65215728,
      "step": 112410
    },
    {
      "epoch": 16.743372058385464,
      "grad_norm": 5.5625,
      "learning_rate": 3.9311164068468576e-06,
      "loss": 0.8492,
      "num_input_tokens_seen": 65218928,
      "step": 112415
    },
    {
      "epoch": 16.74411677092642,
      "grad_norm": 8.1875,
      "learning_rate": 3.929367431692377e-06,
      "loss": 0.8102,
      "num_input_tokens_seen": 65221936,
      "step": 112420
    },
    {
      "epoch": 16.744861483467382,
      "grad_norm": 7.15625,
      "learning_rate": 3.927618812504813e-06,
      "loss": 0.8471,
      "num_input_tokens_seen": 65224912,
      "step": 112425
    },
    {
      "epoch": 16.745606196008342,
      "grad_norm": 13.1875,
      "learning_rate": 3.925870549313718e-06,
      "loss": 1.2597,
      "num_input_tokens_seen": 65227920,
      "step": 112430
    },
    {
      "epoch": 16.7463509085493,
      "grad_norm": 4.125,
      "learning_rate": 3.9241226421486145e-06,
      "loss": 0.692,
      "num_input_tokens_seen": 65231152,
      "step": 112435
    },
    {
      "epoch": 16.74709562109026,
      "grad_norm": 7.25,
      "learning_rate": 3.922375091039046e-06,
      "loss": 1.2902,
      "num_input_tokens_seen": 65233968,
      "step": 112440
    },
    {
      "epoch": 16.747840333631217,
      "grad_norm": 5.65625,
      "learning_rate": 3.9206278960145255e-06,
      "loss": 1.1758,
      "num_input_tokens_seen": 65236912,
      "step": 112445
    },
    {
      "epoch": 16.748585046172177,
      "grad_norm": 10.625,
      "learning_rate": 3.918881057104567e-06,
      "loss": 1.0639,
      "num_input_tokens_seen": 65239984,
      "step": 112450
    },
    {
      "epoch": 16.749329758713138,
      "grad_norm": 8.4375,
      "learning_rate": 3.917134574338696e-06,
      "loss": 1.2718,
      "num_input_tokens_seen": 65242992,
      "step": 112455
    },
    {
      "epoch": 16.750074471254095,
      "grad_norm": 15.75,
      "learning_rate": 3.915388447746407e-06,
      "loss": 1.3387,
      "num_input_tokens_seen": 65246000,
      "step": 112460
    },
    {
      "epoch": 16.750819183795056,
      "grad_norm": 9.375,
      "learning_rate": 3.913642677357201e-06,
      "loss": 0.9447,
      "num_input_tokens_seen": 65249008,
      "step": 112465
    },
    {
      "epoch": 16.751563896336016,
      "grad_norm": 11.125,
      "learning_rate": 3.911897263200564e-06,
      "loss": 0.9314,
      "num_input_tokens_seen": 65251824,
      "step": 112470
    },
    {
      "epoch": 16.752308608876973,
      "grad_norm": 3.53125,
      "learning_rate": 3.910152205305998e-06,
      "loss": 0.8297,
      "num_input_tokens_seen": 65254864,
      "step": 112475
    },
    {
      "epoch": 16.753053321417934,
      "grad_norm": 6.40625,
      "learning_rate": 3.908407503702966e-06,
      "loss": 1.0014,
      "num_input_tokens_seen": 65257808,
      "step": 112480
    },
    {
      "epoch": 16.75379803395889,
      "grad_norm": 10.4375,
      "learning_rate": 3.906663158420962e-06,
      "loss": 1.1187,
      "num_input_tokens_seen": 65260560,
      "step": 112485
    },
    {
      "epoch": 16.75454274649985,
      "grad_norm": 14.125,
      "learning_rate": 3.904919169489438e-06,
      "loss": 0.7458,
      "num_input_tokens_seen": 65263440,
      "step": 112490
    },
    {
      "epoch": 16.75528745904081,
      "grad_norm": 6.21875,
      "learning_rate": 3.90317553693787e-06,
      "loss": 1.2886,
      "num_input_tokens_seen": 65266064,
      "step": 112495
    },
    {
      "epoch": 16.75603217158177,
      "grad_norm": 15.3125,
      "learning_rate": 3.90143226079571e-06,
      "loss": 0.9621,
      "num_input_tokens_seen": 65269040,
      "step": 112500
    },
    {
      "epoch": 16.75677688412273,
      "grad_norm": 11.8125,
      "learning_rate": 3.899689341092402e-06,
      "loss": 1.1414,
      "num_input_tokens_seen": 65271920,
      "step": 112505
    },
    {
      "epoch": 16.757521596663686,
      "grad_norm": 9.8125,
      "learning_rate": 3.897946777857406e-06,
      "loss": 1.7631,
      "num_input_tokens_seen": 65274960,
      "step": 112510
    },
    {
      "epoch": 16.758266309204647,
      "grad_norm": 9.75,
      "learning_rate": 3.896204571120149e-06,
      "loss": 0.9013,
      "num_input_tokens_seen": 65277904,
      "step": 112515
    },
    {
      "epoch": 16.759011021745607,
      "grad_norm": 8.125,
      "learning_rate": 3.894462720910067e-06,
      "loss": 0.8627,
      "num_input_tokens_seen": 65281040,
      "step": 112520
    },
    {
      "epoch": 16.759755734286564,
      "grad_norm": 19.5,
      "learning_rate": 3.892721227256582e-06,
      "loss": 1.0272,
      "num_input_tokens_seen": 65283984,
      "step": 112525
    },
    {
      "epoch": 16.760500446827525,
      "grad_norm": 4.9375,
      "learning_rate": 3.890980090189126e-06,
      "loss": 0.8253,
      "num_input_tokens_seen": 65286800,
      "step": 112530
    },
    {
      "epoch": 16.761245159368485,
      "grad_norm": 20.0,
      "learning_rate": 3.8892393097370975e-06,
      "loss": 1.2939,
      "num_input_tokens_seen": 65289712,
      "step": 112535
    },
    {
      "epoch": 16.761989871909442,
      "grad_norm": 22.25,
      "learning_rate": 3.887498885929924e-06,
      "loss": 1.2561,
      "num_input_tokens_seen": 65292400,
      "step": 112540
    },
    {
      "epoch": 16.762734584450403,
      "grad_norm": 16.625,
      "learning_rate": 3.8857588187969975e-06,
      "loss": 1.0088,
      "num_input_tokens_seen": 65295152,
      "step": 112545
    },
    {
      "epoch": 16.76347929699136,
      "grad_norm": 9.6875,
      "learning_rate": 3.884019108367712e-06,
      "loss": 0.8732,
      "num_input_tokens_seen": 65297776,
      "step": 112550
    },
    {
      "epoch": 16.76422400953232,
      "grad_norm": 15.9375,
      "learning_rate": 3.882279754671467e-06,
      "loss": 1.1296,
      "num_input_tokens_seen": 65300240,
      "step": 112555
    },
    {
      "epoch": 16.76496872207328,
      "grad_norm": 7.4375,
      "learning_rate": 3.880540757737636e-06,
      "loss": 0.6954,
      "num_input_tokens_seen": 65303184,
      "step": 112560
    },
    {
      "epoch": 16.765713434614238,
      "grad_norm": 9.3125,
      "learning_rate": 3.878802117595609e-06,
      "loss": 1.0289,
      "num_input_tokens_seen": 65305872,
      "step": 112565
    },
    {
      "epoch": 16.7664581471552,
      "grad_norm": 10.25,
      "learning_rate": 3.877063834274749e-06,
      "loss": 1.1955,
      "num_input_tokens_seen": 65309040,
      "step": 112570
    },
    {
      "epoch": 16.76720285969616,
      "grad_norm": 10.75,
      "learning_rate": 3.8753259078044365e-06,
      "loss": 1.2019,
      "num_input_tokens_seen": 65311856,
      "step": 112575
    },
    {
      "epoch": 16.767947572237116,
      "grad_norm": 12.9375,
      "learning_rate": 3.8735883382140184e-06,
      "loss": 1.1694,
      "num_input_tokens_seen": 65314992,
      "step": 112580
    },
    {
      "epoch": 16.768692284778076,
      "grad_norm": 9.375,
      "learning_rate": 3.871851125532855e-06,
      "loss": 1.1351,
      "num_input_tokens_seen": 65318160,
      "step": 112585
    },
    {
      "epoch": 16.769436997319033,
      "grad_norm": 5.84375,
      "learning_rate": 3.870114269790292e-06,
      "loss": 1.1058,
      "num_input_tokens_seen": 65321200,
      "step": 112590
    },
    {
      "epoch": 16.770181709859994,
      "grad_norm": 9.75,
      "learning_rate": 3.8683777710156685e-06,
      "loss": 0.8224,
      "num_input_tokens_seen": 65324144,
      "step": 112595
    },
    {
      "epoch": 16.770926422400954,
      "grad_norm": 7.4375,
      "learning_rate": 3.866641629238329e-06,
      "loss": 1.0515,
      "num_input_tokens_seen": 65326992,
      "step": 112600
    },
    {
      "epoch": 16.77167113494191,
      "grad_norm": 13.875,
      "learning_rate": 3.864905844487596e-06,
      "loss": 0.9887,
      "num_input_tokens_seen": 65329808,
      "step": 112605
    },
    {
      "epoch": 16.772415847482872,
      "grad_norm": 11.0,
      "learning_rate": 3.8631704167928025e-06,
      "loss": 1.0544,
      "num_input_tokens_seen": 65332624,
      "step": 112610
    },
    {
      "epoch": 16.773160560023832,
      "grad_norm": 9.0,
      "learning_rate": 3.861435346183259e-06,
      "loss": 1.0185,
      "num_input_tokens_seen": 65335632,
      "step": 112615
    },
    {
      "epoch": 16.77390527256479,
      "grad_norm": 13.625,
      "learning_rate": 3.859700632688285e-06,
      "loss": 1.0077,
      "num_input_tokens_seen": 65338640,
      "step": 112620
    },
    {
      "epoch": 16.77464998510575,
      "grad_norm": 12.875,
      "learning_rate": 3.857966276337183e-06,
      "loss": 1.1771,
      "num_input_tokens_seen": 65341296,
      "step": 112625
    },
    {
      "epoch": 16.775394697646707,
      "grad_norm": 12.0625,
      "learning_rate": 3.856232277159247e-06,
      "loss": 1.3632,
      "num_input_tokens_seen": 65344368,
      "step": 112630
    },
    {
      "epoch": 16.776139410187668,
      "grad_norm": 7.28125,
      "learning_rate": 3.8544986351837845e-06,
      "loss": 1.2229,
      "num_input_tokens_seen": 65347184,
      "step": 112635
    },
    {
      "epoch": 16.776884122728628,
      "grad_norm": 11.6875,
      "learning_rate": 3.852765350440069e-06,
      "loss": 0.9844,
      "num_input_tokens_seen": 65349840,
      "step": 112640
    },
    {
      "epoch": 16.777628835269585,
      "grad_norm": 10.8125,
      "learning_rate": 3.8510324229573956e-06,
      "loss": 1.3063,
      "num_input_tokens_seen": 65352528,
      "step": 112645
    },
    {
      "epoch": 16.778373547810546,
      "grad_norm": 10.5,
      "learning_rate": 3.849299852765034e-06,
      "loss": 1.1316,
      "num_input_tokens_seen": 65355504,
      "step": 112650
    },
    {
      "epoch": 16.779118260351503,
      "grad_norm": 13.3125,
      "learning_rate": 3.847567639892255e-06,
      "loss": 1.191,
      "num_input_tokens_seen": 65358256,
      "step": 112655
    },
    {
      "epoch": 16.779862972892463,
      "grad_norm": 19.625,
      "learning_rate": 3.845835784368324e-06,
      "loss": 1.1327,
      "num_input_tokens_seen": 65361232,
      "step": 112660
    },
    {
      "epoch": 16.780607685433424,
      "grad_norm": 5.78125,
      "learning_rate": 3.844104286222492e-06,
      "loss": 1.2188,
      "num_input_tokens_seen": 65364176,
      "step": 112665
    },
    {
      "epoch": 16.78135239797438,
      "grad_norm": 11.8125,
      "learning_rate": 3.8423731454840185e-06,
      "loss": 1.3396,
      "num_input_tokens_seen": 65367184,
      "step": 112670
    },
    {
      "epoch": 16.78209711051534,
      "grad_norm": 5.0,
      "learning_rate": 3.840642362182143e-06,
      "loss": 0.8865,
      "num_input_tokens_seen": 65369904,
      "step": 112675
    },
    {
      "epoch": 16.7828418230563,
      "grad_norm": 11.875,
      "learning_rate": 3.838911936346115e-06,
      "loss": 1.1081,
      "num_input_tokens_seen": 65372688,
      "step": 112680
    },
    {
      "epoch": 16.78358653559726,
      "grad_norm": 7.8125,
      "learning_rate": 3.837181868005158e-06,
      "loss": 0.9584,
      "num_input_tokens_seen": 65375600,
      "step": 112685
    },
    {
      "epoch": 16.78433124813822,
      "grad_norm": 8.125,
      "learning_rate": 3.83545215718851e-06,
      "loss": 1.1569,
      "num_input_tokens_seen": 65378736,
      "step": 112690
    },
    {
      "epoch": 16.785075960679176,
      "grad_norm": 4.5625,
      "learning_rate": 3.83372280392538e-06,
      "loss": 0.7912,
      "num_input_tokens_seen": 65381520,
      "step": 112695
    },
    {
      "epoch": 16.785820673220137,
      "grad_norm": 13.1875,
      "learning_rate": 3.831993808244996e-06,
      "loss": 1.2236,
      "num_input_tokens_seen": 65384432,
      "step": 112700
    },
    {
      "epoch": 16.786565385761097,
      "grad_norm": 9.875,
      "learning_rate": 3.830265170176564e-06,
      "loss": 0.8899,
      "num_input_tokens_seen": 65387184,
      "step": 112705
    },
    {
      "epoch": 16.787310098302054,
      "grad_norm": 13.6875,
      "learning_rate": 3.828536889749282e-06,
      "loss": 1.1461,
      "num_input_tokens_seen": 65390128,
      "step": 112710
    },
    {
      "epoch": 16.788054810843015,
      "grad_norm": 5.0625,
      "learning_rate": 3.826808966992354e-06,
      "loss": 1.0074,
      "num_input_tokens_seen": 65392592,
      "step": 112715
    },
    {
      "epoch": 16.788799523383975,
      "grad_norm": 5.53125,
      "learning_rate": 3.825081401934971e-06,
      "loss": 1.0697,
      "num_input_tokens_seen": 65395472,
      "step": 112720
    },
    {
      "epoch": 16.789544235924932,
      "grad_norm": 8.4375,
      "learning_rate": 3.823354194606316e-06,
      "loss": 1.0177,
      "num_input_tokens_seen": 65398544,
      "step": 112725
    },
    {
      "epoch": 16.790288948465893,
      "grad_norm": 6.125,
      "learning_rate": 3.821627345035561e-06,
      "loss": 1.218,
      "num_input_tokens_seen": 65401328,
      "step": 112730
    },
    {
      "epoch": 16.79103366100685,
      "grad_norm": 11.375,
      "learning_rate": 3.819900853251896e-06,
      "loss": 1.101,
      "num_input_tokens_seen": 65404464,
      "step": 112735
    },
    {
      "epoch": 16.79177837354781,
      "grad_norm": 6.5625,
      "learning_rate": 3.818174719284473e-06,
      "loss": 0.9934,
      "num_input_tokens_seen": 65407632,
      "step": 112740
    },
    {
      "epoch": 16.79252308608877,
      "grad_norm": 7.6875,
      "learning_rate": 3.816448943162465e-06,
      "loss": 0.9164,
      "num_input_tokens_seen": 65410384,
      "step": 112745
    },
    {
      "epoch": 16.793267798629728,
      "grad_norm": 13.9375,
      "learning_rate": 3.814723524915023e-06,
      "loss": 1.3804,
      "num_input_tokens_seen": 65413328,
      "step": 112750
    },
    {
      "epoch": 16.79401251117069,
      "grad_norm": 8.4375,
      "learning_rate": 3.812998464571288e-06,
      "loss": 1.0937,
      "num_input_tokens_seen": 65416336,
      "step": 112755
    },
    {
      "epoch": 16.79475722371165,
      "grad_norm": 7.15625,
      "learning_rate": 3.8112737621604168e-06,
      "loss": 0.7941,
      "num_input_tokens_seen": 65419472,
      "step": 112760
    },
    {
      "epoch": 16.795501936252606,
      "grad_norm": 14.0,
      "learning_rate": 3.8095494177115358e-06,
      "loss": 1.2701,
      "num_input_tokens_seen": 65422128,
      "step": 112765
    },
    {
      "epoch": 16.796246648793566,
      "grad_norm": 18.875,
      "learning_rate": 3.8078254312537836e-06,
      "loss": 1.2668,
      "num_input_tokens_seen": 65425072,
      "step": 112770
    },
    {
      "epoch": 16.796991361334523,
      "grad_norm": 12.375,
      "learning_rate": 3.806101802816284e-06,
      "loss": 1.1188,
      "num_input_tokens_seen": 65427952,
      "step": 112775
    },
    {
      "epoch": 16.797736073875484,
      "grad_norm": 13.9375,
      "learning_rate": 3.804378532428149e-06,
      "loss": 1.0769,
      "num_input_tokens_seen": 65431088,
      "step": 112780
    },
    {
      "epoch": 16.798480786416444,
      "grad_norm": 5.625,
      "learning_rate": 3.8026556201185003e-06,
      "loss": 0.9804,
      "num_input_tokens_seen": 65433840,
      "step": 112785
    },
    {
      "epoch": 16.7992254989574,
      "grad_norm": 5.03125,
      "learning_rate": 3.8009330659164425e-06,
      "loss": 0.8368,
      "num_input_tokens_seen": 65436784,
      "step": 112790
    },
    {
      "epoch": 16.799970211498362,
      "grad_norm": 8.1875,
      "learning_rate": 3.799210869851072e-06,
      "loss": 1.2521,
      "num_input_tokens_seen": 65439568,
      "step": 112795
    },
    {
      "epoch": 16.800714924039323,
      "grad_norm": 17.0,
      "learning_rate": 3.7974890319514816e-06,
      "loss": 1.2371,
      "num_input_tokens_seen": 65442352,
      "step": 112800
    },
    {
      "epoch": 16.80145963658028,
      "grad_norm": 19.875,
      "learning_rate": 3.7957675522467735e-06,
      "loss": 1.3022,
      "num_input_tokens_seen": 65445232,
      "step": 112805
    },
    {
      "epoch": 16.80220434912124,
      "grad_norm": 9.375,
      "learning_rate": 3.7940464307660126e-06,
      "loss": 1.1502,
      "num_input_tokens_seen": 65447824,
      "step": 112810
    },
    {
      "epoch": 16.802949061662197,
      "grad_norm": 14.0,
      "learning_rate": 3.792325667538291e-06,
      "loss": 1.1925,
      "num_input_tokens_seen": 65450800,
      "step": 112815
    },
    {
      "epoch": 16.803693774203158,
      "grad_norm": 12.4375,
      "learning_rate": 3.790605262592667e-06,
      "loss": 1.135,
      "num_input_tokens_seen": 65453712,
      "step": 112820
    },
    {
      "epoch": 16.804438486744118,
      "grad_norm": 8.5,
      "learning_rate": 3.7888852159582154e-06,
      "loss": 0.9056,
      "num_input_tokens_seen": 65456784,
      "step": 112825
    },
    {
      "epoch": 16.805183199285075,
      "grad_norm": 8.4375,
      "learning_rate": 3.7871655276639937e-06,
      "loss": 1.0403,
      "num_input_tokens_seen": 65459696,
      "step": 112830
    },
    {
      "epoch": 16.805927911826036,
      "grad_norm": 15.875,
      "learning_rate": 3.7854461977390426e-06,
      "loss": 0.8123,
      "num_input_tokens_seen": 65462768,
      "step": 112835
    },
    {
      "epoch": 16.806672624366996,
      "grad_norm": 4.8125,
      "learning_rate": 3.7837272262124216e-06,
      "loss": 0.9626,
      "num_input_tokens_seen": 65465392,
      "step": 112840
    },
    {
      "epoch": 16.807417336907953,
      "grad_norm": 5.96875,
      "learning_rate": 3.7820086131131667e-06,
      "loss": 0.9289,
      "num_input_tokens_seen": 65468496,
      "step": 112845
    },
    {
      "epoch": 16.808162049448914,
      "grad_norm": 10.0625,
      "learning_rate": 3.780290358470309e-06,
      "loss": 0.8882,
      "num_input_tokens_seen": 65471536,
      "step": 112850
    },
    {
      "epoch": 16.80890676198987,
      "grad_norm": 16.125,
      "learning_rate": 3.7785724623128766e-06,
      "loss": 0.6838,
      "num_input_tokens_seen": 65474992,
      "step": 112855
    },
    {
      "epoch": 16.80965147453083,
      "grad_norm": 10.25,
      "learning_rate": 3.776854924669898e-06,
      "loss": 1.1286,
      "num_input_tokens_seen": 65478064,
      "step": 112860
    },
    {
      "epoch": 16.81039618707179,
      "grad_norm": 8.9375,
      "learning_rate": 3.775137745570387e-06,
      "loss": 1.0908,
      "num_input_tokens_seen": 65480848,
      "step": 112865
    },
    {
      "epoch": 16.81114089961275,
      "grad_norm": 10.125,
      "learning_rate": 3.7734209250433445e-06,
      "loss": 1.0522,
      "num_input_tokens_seen": 65483632,
      "step": 112870
    },
    {
      "epoch": 16.81188561215371,
      "grad_norm": 9.9375,
      "learning_rate": 3.7717044631177867e-06,
      "loss": 0.7608,
      "num_input_tokens_seen": 65486224,
      "step": 112875
    },
    {
      "epoch": 16.812630324694666,
      "grad_norm": 9.625,
      "learning_rate": 3.7699883598227016e-06,
      "loss": 0.9846,
      "num_input_tokens_seen": 65488880,
      "step": 112880
    },
    {
      "epoch": 16.813375037235627,
      "grad_norm": 19.75,
      "learning_rate": 3.768272615187091e-06,
      "loss": 1.2917,
      "num_input_tokens_seen": 65492048,
      "step": 112885
    },
    {
      "epoch": 16.814119749776587,
      "grad_norm": 9.75,
      "learning_rate": 3.766557229239931e-06,
      "loss": 1.2729,
      "num_input_tokens_seen": 65494960,
      "step": 112890
    },
    {
      "epoch": 16.814864462317544,
      "grad_norm": 12.375,
      "learning_rate": 3.7648422020102105e-06,
      "loss": 1.571,
      "num_input_tokens_seen": 65497904,
      "step": 112895
    },
    {
      "epoch": 16.815609174858505,
      "grad_norm": 6.96875,
      "learning_rate": 3.7631275335268946e-06,
      "loss": 0.9582,
      "num_input_tokens_seen": 65500944,
      "step": 112900
    },
    {
      "epoch": 16.816353887399465,
      "grad_norm": 23.0,
      "learning_rate": 3.7614132238189605e-06,
      "loss": 0.8978,
      "num_input_tokens_seen": 65503696,
      "step": 112905
    },
    {
      "epoch": 16.817098599940422,
      "grad_norm": 6.96875,
      "learning_rate": 3.7596992729153623e-06,
      "loss": 0.9991,
      "num_input_tokens_seen": 65506512,
      "step": 112910
    },
    {
      "epoch": 16.817843312481383,
      "grad_norm": 12.0625,
      "learning_rate": 3.757985680845058e-06,
      "loss": 0.7918,
      "num_input_tokens_seen": 65510672,
      "step": 112915
    },
    {
      "epoch": 16.81858802502234,
      "grad_norm": 15.875,
      "learning_rate": 3.7562724476369963e-06,
      "loss": 0.905,
      "num_input_tokens_seen": 65513616,
      "step": 112920
    },
    {
      "epoch": 16.8193327375633,
      "grad_norm": 7.59375,
      "learning_rate": 3.7545595733201126e-06,
      "loss": 1.0641,
      "num_input_tokens_seen": 65516336,
      "step": 112925
    },
    {
      "epoch": 16.82007745010426,
      "grad_norm": 5.0625,
      "learning_rate": 3.7528470579233614e-06,
      "loss": 0.8668,
      "num_input_tokens_seen": 65519088,
      "step": 112930
    },
    {
      "epoch": 16.820822162645218,
      "grad_norm": 12.9375,
      "learning_rate": 3.751134901475656e-06,
      "loss": 1.2324,
      "num_input_tokens_seen": 65521744,
      "step": 112935
    },
    {
      "epoch": 16.82156687518618,
      "grad_norm": 16.0,
      "learning_rate": 3.7494231040059364e-06,
      "loss": 1.087,
      "num_input_tokens_seen": 65524432,
      "step": 112940
    },
    {
      "epoch": 16.82231158772714,
      "grad_norm": 17.75,
      "learning_rate": 3.7477116655431162e-06,
      "loss": 1.3777,
      "num_input_tokens_seen": 65527440,
      "step": 112945
    },
    {
      "epoch": 16.823056300268096,
      "grad_norm": 7.875,
      "learning_rate": 3.7460005861161024e-06,
      "loss": 0.9293,
      "num_input_tokens_seen": 65530320,
      "step": 112950
    },
    {
      "epoch": 16.823801012809056,
      "grad_norm": 9.5,
      "learning_rate": 3.7442898657538113e-06,
      "loss": 1.3826,
      "num_input_tokens_seen": 65533008,
      "step": 112955
    },
    {
      "epoch": 16.824545725350013,
      "grad_norm": 11.4375,
      "learning_rate": 3.7425795044851357e-06,
      "loss": 0.7753,
      "num_input_tokens_seen": 65535760,
      "step": 112960
    },
    {
      "epoch": 16.825290437890974,
      "grad_norm": 14.25,
      "learning_rate": 3.7408695023389785e-06,
      "loss": 0.7975,
      "num_input_tokens_seen": 65538736,
      "step": 112965
    },
    {
      "epoch": 16.826035150431935,
      "grad_norm": 4.65625,
      "learning_rate": 3.7391598593442184e-06,
      "loss": 1.015,
      "num_input_tokens_seen": 65541776,
      "step": 112970
    },
    {
      "epoch": 16.82677986297289,
      "grad_norm": 14.875,
      "learning_rate": 3.7374505755297494e-06,
      "loss": 1.1447,
      "num_input_tokens_seen": 65544464,
      "step": 112975
    },
    {
      "epoch": 16.827524575513852,
      "grad_norm": 9.125,
      "learning_rate": 3.735741650924443e-06,
      "loss": 1.122,
      "num_input_tokens_seen": 65547600,
      "step": 112980
    },
    {
      "epoch": 16.828269288054813,
      "grad_norm": 9.1875,
      "learning_rate": 3.7340330855571704e-06,
      "loss": 1.1271,
      "num_input_tokens_seen": 65550576,
      "step": 112985
    },
    {
      "epoch": 16.82901400059577,
      "grad_norm": 9.8125,
      "learning_rate": 3.7323248794567942e-06,
      "loss": 1.0554,
      "num_input_tokens_seen": 65553424,
      "step": 112990
    },
    {
      "epoch": 16.82975871313673,
      "grad_norm": 6.90625,
      "learning_rate": 3.7306170326521674e-06,
      "loss": 1.1384,
      "num_input_tokens_seen": 65556464,
      "step": 112995
    },
    {
      "epoch": 16.830503425677687,
      "grad_norm": 7.25,
      "learning_rate": 3.7289095451721546e-06,
      "loss": 1.2115,
      "num_input_tokens_seen": 65559280,
      "step": 113000
    },
    {
      "epoch": 16.831248138218648,
      "grad_norm": 6.65625,
      "learning_rate": 3.7272024170455914e-06,
      "loss": 1.0567,
      "num_input_tokens_seen": 65562096,
      "step": 113005
    },
    {
      "epoch": 16.831992850759608,
      "grad_norm": 15.4375,
      "learning_rate": 3.7254956483013278e-06,
      "loss": 1.1913,
      "num_input_tokens_seen": 65564848,
      "step": 113010
    },
    {
      "epoch": 16.832737563300565,
      "grad_norm": 4.40625,
      "learning_rate": 3.7237892389681866e-06,
      "loss": 0.9451,
      "num_input_tokens_seen": 65567824,
      "step": 113015
    },
    {
      "epoch": 16.833482275841526,
      "grad_norm": 12.9375,
      "learning_rate": 3.7220831890750067e-06,
      "loss": 1.4637,
      "num_input_tokens_seen": 65570512,
      "step": 113020
    },
    {
      "epoch": 16.834226988382483,
      "grad_norm": 10.1875,
      "learning_rate": 3.7203774986506067e-06,
      "loss": 0.7996,
      "num_input_tokens_seen": 65573200,
      "step": 113025
    },
    {
      "epoch": 16.834971700923443,
      "grad_norm": 8.6875,
      "learning_rate": 3.718672167723797e-06,
      "loss": 1.0192,
      "num_input_tokens_seen": 65576240,
      "step": 113030
    },
    {
      "epoch": 16.835716413464404,
      "grad_norm": 7.34375,
      "learning_rate": 3.7169671963233952e-06,
      "loss": 0.7527,
      "num_input_tokens_seen": 65578864,
      "step": 113035
    },
    {
      "epoch": 16.83646112600536,
      "grad_norm": 12.375,
      "learning_rate": 3.7152625844781963e-06,
      "loss": 0.9136,
      "num_input_tokens_seen": 65581872,
      "step": 113040
    },
    {
      "epoch": 16.83720583854632,
      "grad_norm": 8.4375,
      "learning_rate": 3.7135583322170098e-06,
      "loss": 1.0289,
      "num_input_tokens_seen": 65584944,
      "step": 113045
    },
    {
      "epoch": 16.837950551087282,
      "grad_norm": 8.375,
      "learning_rate": 3.7118544395686213e-06,
      "loss": 1.0623,
      "num_input_tokens_seen": 65587504,
      "step": 113050
    },
    {
      "epoch": 16.83869526362824,
      "grad_norm": 14.4375,
      "learning_rate": 3.710150906561813e-06,
      "loss": 1.191,
      "num_input_tokens_seen": 65590512,
      "step": 113055
    },
    {
      "epoch": 16.8394399761692,
      "grad_norm": 6.34375,
      "learning_rate": 3.708447733225362e-06,
      "loss": 1.071,
      "num_input_tokens_seen": 65593424,
      "step": 113060
    },
    {
      "epoch": 16.840184688710156,
      "grad_norm": 8.4375,
      "learning_rate": 3.706744919588054e-06,
      "loss": 1.0267,
      "num_input_tokens_seen": 65596080,
      "step": 113065
    },
    {
      "epoch": 16.840929401251117,
      "grad_norm": 15.6875,
      "learning_rate": 3.7050424656786486e-06,
      "loss": 1.3003,
      "num_input_tokens_seen": 65598992,
      "step": 113070
    },
    {
      "epoch": 16.841674113792077,
      "grad_norm": 8.0,
      "learning_rate": 3.7033403715259014e-06,
      "loss": 0.9217,
      "num_input_tokens_seen": 65601968,
      "step": 113075
    },
    {
      "epoch": 16.842418826333034,
      "grad_norm": 12.75,
      "learning_rate": 3.7016386371585804e-06,
      "loss": 0.8403,
      "num_input_tokens_seen": 65604720,
      "step": 113080
    },
    {
      "epoch": 16.843163538873995,
      "grad_norm": 9.625,
      "learning_rate": 3.6999372626054212e-06,
      "loss": 0.9331,
      "num_input_tokens_seen": 65607440,
      "step": 113085
    },
    {
      "epoch": 16.843908251414955,
      "grad_norm": 8.125,
      "learning_rate": 3.6982362478951786e-06,
      "loss": 1.0688,
      "num_input_tokens_seen": 65610320,
      "step": 113090
    },
    {
      "epoch": 16.844652963955912,
      "grad_norm": 5.40625,
      "learning_rate": 3.6965355930565796e-06,
      "loss": 1.3725,
      "num_input_tokens_seen": 65613584,
      "step": 113095
    },
    {
      "epoch": 16.845397676496873,
      "grad_norm": 10.5625,
      "learning_rate": 3.6948352981183683e-06,
      "loss": 1.1909,
      "num_input_tokens_seen": 65616432,
      "step": 113100
    },
    {
      "epoch": 16.84614238903783,
      "grad_norm": 10.5,
      "learning_rate": 3.693135363109257e-06,
      "loss": 1.2428,
      "num_input_tokens_seen": 65619184,
      "step": 113105
    },
    {
      "epoch": 16.84688710157879,
      "grad_norm": 11.125,
      "learning_rate": 3.6914357880579647e-06,
      "loss": 1.1522,
      "num_input_tokens_seen": 65622000,
      "step": 113110
    },
    {
      "epoch": 16.84763181411975,
      "grad_norm": 9.875,
      "learning_rate": 3.689736572993213e-06,
      "loss": 1.4081,
      "num_input_tokens_seen": 65624816,
      "step": 113115
    },
    {
      "epoch": 16.848376526660708,
      "grad_norm": 10.9375,
      "learning_rate": 3.6880377179437014e-06,
      "loss": 1.3941,
      "num_input_tokens_seen": 65627696,
      "step": 113120
    },
    {
      "epoch": 16.84912123920167,
      "grad_norm": 9.4375,
      "learning_rate": 3.6863392229381342e-06,
      "loss": 1.0242,
      "num_input_tokens_seen": 65630544,
      "step": 113125
    },
    {
      "epoch": 16.84986595174263,
      "grad_norm": 11.875,
      "learning_rate": 3.684641088005197e-06,
      "loss": 1.2438,
      "num_input_tokens_seen": 65633584,
      "step": 113130
    },
    {
      "epoch": 16.850610664283586,
      "grad_norm": 13.0625,
      "learning_rate": 3.6829433131735895e-06,
      "loss": 0.8537,
      "num_input_tokens_seen": 65636592,
      "step": 113135
    },
    {
      "epoch": 16.851355376824547,
      "grad_norm": 6.375,
      "learning_rate": 3.6812458984719857e-06,
      "loss": 1.0369,
      "num_input_tokens_seen": 65639408,
      "step": 113140
    },
    {
      "epoch": 16.852100089365504,
      "grad_norm": 11.625,
      "learning_rate": 3.6795488439290706e-06,
      "loss": 1.1455,
      "num_input_tokens_seen": 65641968,
      "step": 113145
    },
    {
      "epoch": 16.852844801906464,
      "grad_norm": 10.0,
      "learning_rate": 3.6778521495735073e-06,
      "loss": 1.1478,
      "num_input_tokens_seen": 65644720,
      "step": 113150
    },
    {
      "epoch": 16.853589514447425,
      "grad_norm": 5.84375,
      "learning_rate": 3.6761558154339544e-06,
      "loss": 1.1986,
      "num_input_tokens_seen": 65648016,
      "step": 113155
    },
    {
      "epoch": 16.85433422698838,
      "grad_norm": 19.75,
      "learning_rate": 3.6744598415390823e-06,
      "loss": 1.3535,
      "num_input_tokens_seen": 65650800,
      "step": 113160
    },
    {
      "epoch": 16.855078939529342,
      "grad_norm": 5.5625,
      "learning_rate": 3.672764227917533e-06,
      "loss": 1.2087,
      "num_input_tokens_seen": 65653456,
      "step": 113165
    },
    {
      "epoch": 16.8558236520703,
      "grad_norm": 10.0625,
      "learning_rate": 3.6710689745979606e-06,
      "loss": 1.0632,
      "num_input_tokens_seen": 65656368,
      "step": 113170
    },
    {
      "epoch": 16.85656836461126,
      "grad_norm": 7.34375,
      "learning_rate": 3.6693740816090005e-06,
      "loss": 0.9951,
      "num_input_tokens_seen": 65659088,
      "step": 113175
    },
    {
      "epoch": 16.85731307715222,
      "grad_norm": 9.8125,
      "learning_rate": 3.6676795489792857e-06,
      "loss": 1.3782,
      "num_input_tokens_seen": 65661744,
      "step": 113180
    },
    {
      "epoch": 16.858057789693177,
      "grad_norm": 9.4375,
      "learning_rate": 3.665985376737438e-06,
      "loss": 0.7871,
      "num_input_tokens_seen": 65664560,
      "step": 113185
    },
    {
      "epoch": 16.858802502234138,
      "grad_norm": 8.5,
      "learning_rate": 3.6642915649120894e-06,
      "loss": 1.1472,
      "num_input_tokens_seen": 65667376,
      "step": 113190
    },
    {
      "epoch": 16.859547214775098,
      "grad_norm": 13.6875,
      "learning_rate": 3.662598113531851e-06,
      "loss": 1.0349,
      "num_input_tokens_seen": 65670192,
      "step": 113195
    },
    {
      "epoch": 16.860291927316055,
      "grad_norm": 8.75,
      "learning_rate": 3.6609050226253273e-06,
      "loss": 0.9884,
      "num_input_tokens_seen": 65673008,
      "step": 113200
    },
    {
      "epoch": 16.861036639857016,
      "grad_norm": 5.0,
      "learning_rate": 3.6592122922211287e-06,
      "loss": 1.0947,
      "num_input_tokens_seen": 65675920,
      "step": 113205
    },
    {
      "epoch": 16.861781352397973,
      "grad_norm": 11.875,
      "learning_rate": 3.6575199223478436e-06,
      "loss": 1.2727,
      "num_input_tokens_seen": 65678736,
      "step": 113210
    },
    {
      "epoch": 16.862526064938933,
      "grad_norm": 12.3125,
      "learning_rate": 3.655827913034071e-06,
      "loss": 0.7724,
      "num_input_tokens_seen": 65681776,
      "step": 113215
    },
    {
      "epoch": 16.863270777479894,
      "grad_norm": 4.8125,
      "learning_rate": 3.6541362643083887e-06,
      "loss": 0.9425,
      "num_input_tokens_seen": 65684688,
      "step": 113220
    },
    {
      "epoch": 16.86401549002085,
      "grad_norm": 11.25,
      "learning_rate": 3.6524449761993874e-06,
      "loss": 1.3813,
      "num_input_tokens_seen": 65687664,
      "step": 113225
    },
    {
      "epoch": 16.86476020256181,
      "grad_norm": 10.6875,
      "learning_rate": 3.6507540487356274e-06,
      "loss": 1.1991,
      "num_input_tokens_seen": 65690640,
      "step": 113230
    },
    {
      "epoch": 16.865504915102772,
      "grad_norm": 9.6875,
      "learning_rate": 3.6490634819456775e-06,
      "loss": 1.2157,
      "num_input_tokens_seen": 65693680,
      "step": 113235
    },
    {
      "epoch": 16.86624962764373,
      "grad_norm": 10.5,
      "learning_rate": 3.6473732758581013e-06,
      "loss": 0.7425,
      "num_input_tokens_seen": 65696336,
      "step": 113240
    },
    {
      "epoch": 16.86699434018469,
      "grad_norm": 12.3125,
      "learning_rate": 3.6456834305014565e-06,
      "loss": 0.9611,
      "num_input_tokens_seen": 65699088,
      "step": 113245
    },
    {
      "epoch": 16.867739052725646,
      "grad_norm": 12.625,
      "learning_rate": 3.6439939459042836e-06,
      "loss": 1.076,
      "num_input_tokens_seen": 65701840,
      "step": 113250
    },
    {
      "epoch": 16.868483765266607,
      "grad_norm": 6.09375,
      "learning_rate": 3.6423048220951216e-06,
      "loss": 0.8045,
      "num_input_tokens_seen": 65704656,
      "step": 113255
    },
    {
      "epoch": 16.869228477807567,
      "grad_norm": 7.78125,
      "learning_rate": 3.6406160591025196e-06,
      "loss": 0.8576,
      "num_input_tokens_seen": 65707472,
      "step": 113260
    },
    {
      "epoch": 16.869973190348524,
      "grad_norm": 5.21875,
      "learning_rate": 3.638927656954996e-06,
      "loss": 1.095,
      "num_input_tokens_seen": 65710512,
      "step": 113265
    },
    {
      "epoch": 16.870717902889485,
      "grad_norm": 10.375,
      "learning_rate": 3.6372396156810817e-06,
      "loss": 1.1515,
      "num_input_tokens_seen": 65713200,
      "step": 113270
    },
    {
      "epoch": 16.871462615430445,
      "grad_norm": 8.5,
      "learning_rate": 3.6355519353092947e-06,
      "loss": 1.2335,
      "num_input_tokens_seen": 65716304,
      "step": 113275
    },
    {
      "epoch": 16.872207327971402,
      "grad_norm": 11.5,
      "learning_rate": 3.6338646158681377e-06,
      "loss": 1.1673,
      "num_input_tokens_seen": 65719216,
      "step": 113280
    },
    {
      "epoch": 16.872952040512363,
      "grad_norm": 11.0625,
      "learning_rate": 3.6321776573861266e-06,
      "loss": 1.2234,
      "num_input_tokens_seen": 65722128,
      "step": 113285
    },
    {
      "epoch": 16.87369675305332,
      "grad_norm": 7.375,
      "learning_rate": 3.6304910598917526e-06,
      "loss": 1.1375,
      "num_input_tokens_seen": 65725104,
      "step": 113290
    },
    {
      "epoch": 16.87444146559428,
      "grad_norm": 6.125,
      "learning_rate": 3.6288048234135203e-06,
      "loss": 0.9666,
      "num_input_tokens_seen": 65728240,
      "step": 113295
    },
    {
      "epoch": 16.87518617813524,
      "grad_norm": 13.3125,
      "learning_rate": 3.6271189479799017e-06,
      "loss": 1.0803,
      "num_input_tokens_seen": 65731184,
      "step": 113300
    },
    {
      "epoch": 16.875930890676198,
      "grad_norm": 17.875,
      "learning_rate": 3.625433433619391e-06,
      "loss": 0.8206,
      "num_input_tokens_seen": 65734416,
      "step": 113305
    },
    {
      "epoch": 16.87667560321716,
      "grad_norm": 5.0625,
      "learning_rate": 3.623748280360459e-06,
      "loss": 1.1598,
      "num_input_tokens_seen": 65737552,
      "step": 113310
    },
    {
      "epoch": 16.87742031575812,
      "grad_norm": 13.4375,
      "learning_rate": 3.622063488231575e-06,
      "loss": 0.714,
      "num_input_tokens_seen": 65740560,
      "step": 113315
    },
    {
      "epoch": 16.878165028299076,
      "grad_norm": 6.8125,
      "learning_rate": 3.6203790572612002e-06,
      "loss": 0.971,
      "num_input_tokens_seen": 65743664,
      "step": 113320
    },
    {
      "epoch": 16.878909740840037,
      "grad_norm": 8.375,
      "learning_rate": 3.6186949874777858e-06,
      "loss": 0.9424,
      "num_input_tokens_seen": 65746672,
      "step": 113325
    },
    {
      "epoch": 16.879654453380994,
      "grad_norm": 11.1875,
      "learning_rate": 3.617011278909796e-06,
      "loss": 1.2651,
      "num_input_tokens_seen": 65749328,
      "step": 113330
    },
    {
      "epoch": 16.880399165921954,
      "grad_norm": 10.625,
      "learning_rate": 3.6153279315856603e-06,
      "loss": 1.2459,
      "num_input_tokens_seen": 65752240,
      "step": 113335
    },
    {
      "epoch": 16.881143878462915,
      "grad_norm": 7.4375,
      "learning_rate": 3.613644945533831e-06,
      "loss": 1.0821,
      "num_input_tokens_seen": 65755216,
      "step": 113340
    },
    {
      "epoch": 16.88188859100387,
      "grad_norm": 10.375,
      "learning_rate": 3.61196232078273e-06,
      "loss": 1.1573,
      "num_input_tokens_seen": 65758288,
      "step": 113345
    },
    {
      "epoch": 16.882633303544832,
      "grad_norm": 7.59375,
      "learning_rate": 3.610280057360793e-06,
      "loss": 1.1577,
      "num_input_tokens_seen": 65761072,
      "step": 113350
    },
    {
      "epoch": 16.883378016085793,
      "grad_norm": 6.3125,
      "learning_rate": 3.608598155296433e-06,
      "loss": 0.9438,
      "num_input_tokens_seen": 65763952,
      "step": 113355
    },
    {
      "epoch": 16.88412272862675,
      "grad_norm": 6.375,
      "learning_rate": 3.606916614618061e-06,
      "loss": 1.1662,
      "num_input_tokens_seen": 65766960,
      "step": 113360
    },
    {
      "epoch": 16.88486744116771,
      "grad_norm": 9.1875,
      "learning_rate": 3.605235435354096e-06,
      "loss": 0.9298,
      "num_input_tokens_seen": 65769808,
      "step": 113365
    },
    {
      "epoch": 16.885612153708667,
      "grad_norm": 7.125,
      "learning_rate": 3.6035546175329283e-06,
      "loss": 0.8369,
      "num_input_tokens_seen": 65772464,
      "step": 113370
    },
    {
      "epoch": 16.886356866249628,
      "grad_norm": 12.375,
      "learning_rate": 3.6018741611829637e-06,
      "loss": 1.1549,
      "num_input_tokens_seen": 65775568,
      "step": 113375
    },
    {
      "epoch": 16.88710157879059,
      "grad_norm": 9.875,
      "learning_rate": 3.6001940663325876e-06,
      "loss": 1.0413,
      "num_input_tokens_seen": 65778416,
      "step": 113380
    },
    {
      "epoch": 16.887846291331545,
      "grad_norm": 5.3125,
      "learning_rate": 3.598514333010183e-06,
      "loss": 0.8811,
      "num_input_tokens_seen": 65781136,
      "step": 113385
    },
    {
      "epoch": 16.888591003872506,
      "grad_norm": 15.4375,
      "learning_rate": 3.5968349612441277e-06,
      "loss": 1.2264,
      "num_input_tokens_seen": 65784048,
      "step": 113390
    },
    {
      "epoch": 16.889335716413463,
      "grad_norm": 8.375,
      "learning_rate": 3.595155951062787e-06,
      "loss": 1.1715,
      "num_input_tokens_seen": 65787216,
      "step": 113395
    },
    {
      "epoch": 16.890080428954423,
      "grad_norm": 7.03125,
      "learning_rate": 3.5934773024945355e-06,
      "loss": 0.9752,
      "num_input_tokens_seen": 65790288,
      "step": 113400
    },
    {
      "epoch": 16.890825141495384,
      "grad_norm": 11.6875,
      "learning_rate": 3.591799015567726e-06,
      "loss": 1.027,
      "num_input_tokens_seen": 65793104,
      "step": 113405
    },
    {
      "epoch": 16.89156985403634,
      "grad_norm": 5.46875,
      "learning_rate": 3.5901210903107165e-06,
      "loss": 1.0335,
      "num_input_tokens_seen": 65796048,
      "step": 113410
    },
    {
      "epoch": 16.8923145665773,
      "grad_norm": 8.0,
      "learning_rate": 3.5884435267518476e-06,
      "loss": 0.9683,
      "num_input_tokens_seen": 65798864,
      "step": 113415
    },
    {
      "epoch": 16.893059279118262,
      "grad_norm": 10.6875,
      "learning_rate": 3.586766324919466e-06,
      "loss": 1.004,
      "num_input_tokens_seen": 65801488,
      "step": 113420
    },
    {
      "epoch": 16.89380399165922,
      "grad_norm": 11.25,
      "learning_rate": 3.5850894848418997e-06,
      "loss": 1.1138,
      "num_input_tokens_seen": 65804464,
      "step": 113425
    },
    {
      "epoch": 16.89454870420018,
      "grad_norm": 7.125,
      "learning_rate": 3.5834130065474897e-06,
      "loss": 0.9741,
      "num_input_tokens_seen": 65807216,
      "step": 113430
    },
    {
      "epoch": 16.895293416741136,
      "grad_norm": 15.0625,
      "learning_rate": 3.5817368900645466e-06,
      "loss": 1.035,
      "num_input_tokens_seen": 65810096,
      "step": 113435
    },
    {
      "epoch": 16.896038129282097,
      "grad_norm": 13.3125,
      "learning_rate": 3.5800611354213866e-06,
      "loss": 1.0536,
      "num_input_tokens_seen": 65813104,
      "step": 113440
    },
    {
      "epoch": 16.896782841823057,
      "grad_norm": 5.15625,
      "learning_rate": 3.5783857426463286e-06,
      "loss": 0.8853,
      "num_input_tokens_seen": 65816176,
      "step": 113445
    },
    {
      "epoch": 16.897527554364014,
      "grad_norm": 10.625,
      "learning_rate": 3.5767107117676698e-06,
      "loss": 1.2328,
      "num_input_tokens_seen": 65819120,
      "step": 113450
    },
    {
      "epoch": 16.898272266904975,
      "grad_norm": 7.28125,
      "learning_rate": 3.575036042813712e-06,
      "loss": 0.979,
      "num_input_tokens_seen": 65821904,
      "step": 113455
    },
    {
      "epoch": 16.899016979445936,
      "grad_norm": 4.84375,
      "learning_rate": 3.5733617358127384e-06,
      "loss": 1.1503,
      "num_input_tokens_seen": 65824720,
      "step": 113460
    },
    {
      "epoch": 16.899761691986892,
      "grad_norm": 5.0,
      "learning_rate": 3.5716877907930465e-06,
      "loss": 0.7806,
      "num_input_tokens_seen": 65827568,
      "step": 113465
    },
    {
      "epoch": 16.900506404527853,
      "grad_norm": 10.5,
      "learning_rate": 3.5700142077829122e-06,
      "loss": 1.1864,
      "num_input_tokens_seen": 65830768,
      "step": 113470
    },
    {
      "epoch": 16.90125111706881,
      "grad_norm": 6.625,
      "learning_rate": 3.5683409868106004e-06,
      "loss": 1.0274,
      "num_input_tokens_seen": 65833776,
      "step": 113475
    },
    {
      "epoch": 16.90199582960977,
      "grad_norm": 9.5,
      "learning_rate": 3.566668127904391e-06,
      "loss": 1.1503,
      "num_input_tokens_seen": 65836912,
      "step": 113480
    },
    {
      "epoch": 16.90274054215073,
      "grad_norm": 4.65625,
      "learning_rate": 3.5649956310925355e-06,
      "loss": 1.024,
      "num_input_tokens_seen": 65839760,
      "step": 113485
    },
    {
      "epoch": 16.903485254691688,
      "grad_norm": 8.0625,
      "learning_rate": 3.563323496403298e-06,
      "loss": 0.8421,
      "num_input_tokens_seen": 65842448,
      "step": 113490
    },
    {
      "epoch": 16.90422996723265,
      "grad_norm": 12.25,
      "learning_rate": 3.5616517238649172e-06,
      "loss": 1.1537,
      "num_input_tokens_seen": 65845616,
      "step": 113495
    },
    {
      "epoch": 16.90497467977361,
      "grad_norm": 8.6875,
      "learning_rate": 3.5599803135056455e-06,
      "loss": 0.9537,
      "num_input_tokens_seen": 65848496,
      "step": 113500
    },
    {
      "epoch": 16.905719392314566,
      "grad_norm": 11.9375,
      "learning_rate": 3.558309265353718e-06,
      "loss": 1.4373,
      "num_input_tokens_seen": 65851600,
      "step": 113505
    },
    {
      "epoch": 16.906464104855527,
      "grad_norm": 11.125,
      "learning_rate": 3.556638579437363e-06,
      "loss": 0.9117,
      "num_input_tokens_seen": 65854384,
      "step": 113510
    },
    {
      "epoch": 16.907208817396484,
      "grad_norm": 7.375,
      "learning_rate": 3.554968255784799e-06,
      "loss": 1.3437,
      "num_input_tokens_seen": 65857200,
      "step": 113515
    },
    {
      "epoch": 16.907953529937444,
      "grad_norm": 10.125,
      "learning_rate": 3.5532982944242567e-06,
      "loss": 1.0062,
      "num_input_tokens_seen": 65859792,
      "step": 113520
    },
    {
      "epoch": 16.908698242478405,
      "grad_norm": 6.1875,
      "learning_rate": 3.5516286953839406e-06,
      "loss": 0.9552,
      "num_input_tokens_seen": 65862576,
      "step": 113525
    },
    {
      "epoch": 16.90944295501936,
      "grad_norm": 6.6875,
      "learning_rate": 3.549959458692054e-06,
      "loss": 1.1558,
      "num_input_tokens_seen": 65865456,
      "step": 113530
    },
    {
      "epoch": 16.910187667560322,
      "grad_norm": 8.75,
      "learning_rate": 3.5482905843768065e-06,
      "loss": 0.97,
      "num_input_tokens_seen": 65868240,
      "step": 113535
    },
    {
      "epoch": 16.91093238010128,
      "grad_norm": 8.3125,
      "learning_rate": 3.5466220724663824e-06,
      "loss": 0.9941,
      "num_input_tokens_seen": 65871184,
      "step": 113540
    },
    {
      "epoch": 16.91167709264224,
      "grad_norm": 21.25,
      "learning_rate": 3.5449539229889807e-06,
      "loss": 1.2696,
      "num_input_tokens_seen": 65873904,
      "step": 113545
    },
    {
      "epoch": 16.9124218051832,
      "grad_norm": 12.8125,
      "learning_rate": 3.543286135972773e-06,
      "loss": 1.2327,
      "num_input_tokens_seen": 65876880,
      "step": 113550
    },
    {
      "epoch": 16.913166517724157,
      "grad_norm": 16.25,
      "learning_rate": 3.541618711445935e-06,
      "loss": 1.1097,
      "num_input_tokens_seen": 65880208,
      "step": 113555
    },
    {
      "epoch": 16.913911230265118,
      "grad_norm": 8.75,
      "learning_rate": 3.5399516494366457e-06,
      "loss": 0.8654,
      "num_input_tokens_seen": 65882992,
      "step": 113560
    },
    {
      "epoch": 16.91465594280608,
      "grad_norm": 10.25,
      "learning_rate": 3.5382849499730558e-06,
      "loss": 0.9984,
      "num_input_tokens_seen": 65886000,
      "step": 113565
    },
    {
      "epoch": 16.915400655347035,
      "grad_norm": 15.4375,
      "learning_rate": 3.536618613083337e-06,
      "loss": 0.9107,
      "num_input_tokens_seen": 65888560,
      "step": 113570
    },
    {
      "epoch": 16.916145367887996,
      "grad_norm": 7.21875,
      "learning_rate": 3.53495263879563e-06,
      "loss": 1.235,
      "num_input_tokens_seen": 65891568,
      "step": 113575
    },
    {
      "epoch": 16.916890080428953,
      "grad_norm": 6.4375,
      "learning_rate": 3.5332870271380797e-06,
      "loss": 1.2054,
      "num_input_tokens_seen": 65894480,
      "step": 113580
    },
    {
      "epoch": 16.917634792969913,
      "grad_norm": 13.6875,
      "learning_rate": 3.5316217781388243e-06,
      "loss": 1.0921,
      "num_input_tokens_seen": 65897168,
      "step": 113585
    },
    {
      "epoch": 16.918379505510874,
      "grad_norm": 5.75,
      "learning_rate": 3.5299568918260054e-06,
      "loss": 1.011,
      "num_input_tokens_seen": 65900048,
      "step": 113590
    },
    {
      "epoch": 16.91912421805183,
      "grad_norm": 12.125,
      "learning_rate": 3.528292368227745e-06,
      "loss": 1.1537,
      "num_input_tokens_seen": 65902672,
      "step": 113595
    },
    {
      "epoch": 16.91986893059279,
      "grad_norm": 11.3125,
      "learning_rate": 3.526628207372154e-06,
      "loss": 1.0836,
      "num_input_tokens_seen": 65905296,
      "step": 113600
    },
    {
      "epoch": 16.920613643133752,
      "grad_norm": 10.9375,
      "learning_rate": 3.5249644092873624e-06,
      "loss": 1.5746,
      "num_input_tokens_seen": 65908336,
      "step": 113605
    },
    {
      "epoch": 16.92135835567471,
      "grad_norm": 7.0,
      "learning_rate": 3.5233009740014617e-06,
      "loss": 1.0388,
      "num_input_tokens_seen": 65911568,
      "step": 113610
    },
    {
      "epoch": 16.92210306821567,
      "grad_norm": 15.125,
      "learning_rate": 3.521637901542571e-06,
      "loss": 1.4022,
      "num_input_tokens_seen": 65914608,
      "step": 113615
    },
    {
      "epoch": 16.922847780756626,
      "grad_norm": 11.1875,
      "learning_rate": 3.5199751919387735e-06,
      "loss": 1.1278,
      "num_input_tokens_seen": 65917488,
      "step": 113620
    },
    {
      "epoch": 16.923592493297587,
      "grad_norm": 6.75,
      "learning_rate": 3.518312845218169e-06,
      "loss": 1.4749,
      "num_input_tokens_seen": 65920816,
      "step": 113625
    },
    {
      "epoch": 16.924337205838548,
      "grad_norm": 8.9375,
      "learning_rate": 3.516650861408835e-06,
      "loss": 0.9315,
      "num_input_tokens_seen": 65923568,
      "step": 113630
    },
    {
      "epoch": 16.925081918379504,
      "grad_norm": 4.8125,
      "learning_rate": 3.514989240538846e-06,
      "loss": 1.2804,
      "num_input_tokens_seen": 65926224,
      "step": 113635
    },
    {
      "epoch": 16.925826630920465,
      "grad_norm": 7.96875,
      "learning_rate": 3.513327982636283e-06,
      "loss": 1.0633,
      "num_input_tokens_seen": 65929104,
      "step": 113640
    },
    {
      "epoch": 16.926571343461426,
      "grad_norm": 10.4375,
      "learning_rate": 3.5116670877292034e-06,
      "loss": 1.0881,
      "num_input_tokens_seen": 65932016,
      "step": 113645
    },
    {
      "epoch": 16.927316056002383,
      "grad_norm": 12.5625,
      "learning_rate": 3.5100065558456714e-06,
      "loss": 1.4447,
      "num_input_tokens_seen": 65935088,
      "step": 113650
    },
    {
      "epoch": 16.928060768543343,
      "grad_norm": 7.9375,
      "learning_rate": 3.5083463870137306e-06,
      "loss": 1.0073,
      "num_input_tokens_seen": 65938032,
      "step": 113655
    },
    {
      "epoch": 16.9288054810843,
      "grad_norm": 10.75,
      "learning_rate": 3.50668658126144e-06,
      "loss": 1.4156,
      "num_input_tokens_seen": 65940784,
      "step": 113660
    },
    {
      "epoch": 16.92955019362526,
      "grad_norm": 5.9375,
      "learning_rate": 3.5050271386168287e-06,
      "loss": 0.9933,
      "num_input_tokens_seen": 65943440,
      "step": 113665
    },
    {
      "epoch": 16.93029490616622,
      "grad_norm": 8.3125,
      "learning_rate": 3.503368059107942e-06,
      "loss": 1.2386,
      "num_input_tokens_seen": 65946608,
      "step": 113670
    },
    {
      "epoch": 16.931039618707178,
      "grad_norm": 3.9375,
      "learning_rate": 3.5017093427628045e-06,
      "loss": 0.9049,
      "num_input_tokens_seen": 65949488,
      "step": 113675
    },
    {
      "epoch": 16.93178433124814,
      "grad_norm": 6.9375,
      "learning_rate": 3.5000509896094323e-06,
      "loss": 0.8957,
      "num_input_tokens_seen": 65952112,
      "step": 113680
    },
    {
      "epoch": 16.932529043789096,
      "grad_norm": 12.1875,
      "learning_rate": 3.4983929996758535e-06,
      "loss": 1.0485,
      "num_input_tokens_seen": 65955472,
      "step": 113685
    },
    {
      "epoch": 16.933273756330056,
      "grad_norm": 7.6875,
      "learning_rate": 3.496735372990065e-06,
      "loss": 0.7425,
      "num_input_tokens_seen": 65958352,
      "step": 113690
    },
    {
      "epoch": 16.934018468871017,
      "grad_norm": 14.1875,
      "learning_rate": 3.4950781095800828e-06,
      "loss": 1.3928,
      "num_input_tokens_seen": 65961200,
      "step": 113695
    },
    {
      "epoch": 16.934763181411974,
      "grad_norm": 10.75,
      "learning_rate": 3.493421209473896e-06,
      "loss": 1.1974,
      "num_input_tokens_seen": 65964176,
      "step": 113700
    },
    {
      "epoch": 16.935507893952934,
      "grad_norm": 4.34375,
      "learning_rate": 3.4917646726995018e-06,
      "loss": 1.076,
      "num_input_tokens_seen": 65967248,
      "step": 113705
    },
    {
      "epoch": 16.936252606493895,
      "grad_norm": 8.5625,
      "learning_rate": 3.490108499284886e-06,
      "loss": 0.7931,
      "num_input_tokens_seen": 65970288,
      "step": 113710
    },
    {
      "epoch": 16.93699731903485,
      "grad_norm": 18.625,
      "learning_rate": 3.488452689258026e-06,
      "loss": 1.2617,
      "num_input_tokens_seen": 65973264,
      "step": 113715
    },
    {
      "epoch": 16.937742031575812,
      "grad_norm": 16.125,
      "learning_rate": 3.4867972426468915e-06,
      "loss": 1.1528,
      "num_input_tokens_seen": 65976272,
      "step": 113720
    },
    {
      "epoch": 16.93848674411677,
      "grad_norm": 14.75,
      "learning_rate": 3.4851421594794486e-06,
      "loss": 1.3211,
      "num_input_tokens_seen": 65979088,
      "step": 113725
    },
    {
      "epoch": 16.93923145665773,
      "grad_norm": 6.625,
      "learning_rate": 3.483487439783667e-06,
      "loss": 0.7025,
      "num_input_tokens_seen": 65982064,
      "step": 113730
    },
    {
      "epoch": 16.93997616919869,
      "grad_norm": 8.875,
      "learning_rate": 3.4818330835874937e-06,
      "loss": 0.8394,
      "num_input_tokens_seen": 65984944,
      "step": 113735
    },
    {
      "epoch": 16.940720881739647,
      "grad_norm": 8.0625,
      "learning_rate": 3.4801790909188837e-06,
      "loss": 1.0426,
      "num_input_tokens_seen": 65988144,
      "step": 113740
    },
    {
      "epoch": 16.941465594280608,
      "grad_norm": 6.21875,
      "learning_rate": 3.4785254618057707e-06,
      "loss": 1.3626,
      "num_input_tokens_seen": 65991056,
      "step": 113745
    },
    {
      "epoch": 16.94221030682157,
      "grad_norm": 13.4375,
      "learning_rate": 3.4768721962761015e-06,
      "loss": 0.913,
      "num_input_tokens_seen": 65994000,
      "step": 113750
    },
    {
      "epoch": 16.942955019362525,
      "grad_norm": 9.0,
      "learning_rate": 3.4752192943578038e-06,
      "loss": 1.142,
      "num_input_tokens_seen": 65997104,
      "step": 113755
    },
    {
      "epoch": 16.943699731903486,
      "grad_norm": 12.625,
      "learning_rate": 3.4735667560787916e-06,
      "loss": 1.172,
      "num_input_tokens_seen": 65999920,
      "step": 113760
    },
    {
      "epoch": 16.944444444444443,
      "grad_norm": 14.75,
      "learning_rate": 3.471914581466998e-06,
      "loss": 0.7959,
      "num_input_tokens_seen": 66002864,
      "step": 113765
    },
    {
      "epoch": 16.945189156985403,
      "grad_norm": 8.625,
      "learning_rate": 3.4702627705503197e-06,
      "loss": 0.9432,
      "num_input_tokens_seen": 66005520,
      "step": 113770
    },
    {
      "epoch": 16.945933869526364,
      "grad_norm": 13.8125,
      "learning_rate": 3.468611323356677e-06,
      "loss": 1.4217,
      "num_input_tokens_seen": 66008368,
      "step": 113775
    },
    {
      "epoch": 16.94667858206732,
      "grad_norm": 8.5,
      "learning_rate": 3.4669602399139607e-06,
      "loss": 1.0324,
      "num_input_tokens_seen": 66011440,
      "step": 113780
    },
    {
      "epoch": 16.94742329460828,
      "grad_norm": 8.5625,
      "learning_rate": 3.4653095202500677e-06,
      "loss": 1.0188,
      "num_input_tokens_seen": 66014224,
      "step": 113785
    },
    {
      "epoch": 16.948168007149242,
      "grad_norm": 12.875,
      "learning_rate": 3.4636591643928823e-06,
      "loss": 1.392,
      "num_input_tokens_seen": 66017072,
      "step": 113790
    },
    {
      "epoch": 16.9489127196902,
      "grad_norm": 17.125,
      "learning_rate": 3.462009172370284e-06,
      "loss": 1.0718,
      "num_input_tokens_seen": 66020720,
      "step": 113795
    },
    {
      "epoch": 16.94965743223116,
      "grad_norm": 4.71875,
      "learning_rate": 3.4603595442101537e-06,
      "loss": 0.8806,
      "num_input_tokens_seen": 66023824,
      "step": 113800
    },
    {
      "epoch": 16.950402144772116,
      "grad_norm": 15.8125,
      "learning_rate": 3.45871027994035e-06,
      "loss": 1.1944,
      "num_input_tokens_seen": 66026640,
      "step": 113805
    },
    {
      "epoch": 16.951146857313077,
      "grad_norm": 8.25,
      "learning_rate": 3.45706137958875e-06,
      "loss": 1.1996,
      "num_input_tokens_seen": 66029808,
      "step": 113810
    },
    {
      "epoch": 16.951891569854038,
      "grad_norm": 11.625,
      "learning_rate": 3.4554128431831976e-06,
      "loss": 1.2494,
      "num_input_tokens_seen": 66032848,
      "step": 113815
    },
    {
      "epoch": 16.952636282394995,
      "grad_norm": 6.84375,
      "learning_rate": 3.4537646707515527e-06,
      "loss": 0.9152,
      "num_input_tokens_seen": 66035728,
      "step": 113820
    },
    {
      "epoch": 16.953380994935955,
      "grad_norm": 12.375,
      "learning_rate": 3.45211686232165e-06,
      "loss": 1.2356,
      "num_input_tokens_seen": 66038672,
      "step": 113825
    },
    {
      "epoch": 16.954125707476916,
      "grad_norm": 6.5,
      "learning_rate": 3.45046941792134e-06,
      "loss": 1.0293,
      "num_input_tokens_seen": 66041648,
      "step": 113830
    },
    {
      "epoch": 16.954870420017873,
      "grad_norm": 16.25,
      "learning_rate": 3.4488223375784447e-06,
      "loss": 1.0677,
      "num_input_tokens_seen": 66044464,
      "step": 113835
    },
    {
      "epoch": 16.955615132558833,
      "grad_norm": 13.6875,
      "learning_rate": 3.447175621320792e-06,
      "loss": 0.8475,
      "num_input_tokens_seen": 66047280,
      "step": 113840
    },
    {
      "epoch": 16.95635984509979,
      "grad_norm": 12.875,
      "learning_rate": 3.445529269176198e-06,
      "loss": 1.2517,
      "num_input_tokens_seen": 66050032,
      "step": 113845
    },
    {
      "epoch": 16.95710455764075,
      "grad_norm": 10.25,
      "learning_rate": 3.443883281172486e-06,
      "loss": 0.8535,
      "num_input_tokens_seen": 66052848,
      "step": 113850
    },
    {
      "epoch": 16.95784927018171,
      "grad_norm": 11.5,
      "learning_rate": 3.442237657337455e-06,
      "loss": 1.0099,
      "num_input_tokens_seen": 66056016,
      "step": 113855
    },
    {
      "epoch": 16.958593982722668,
      "grad_norm": 2.9375,
      "learning_rate": 3.440592397698905e-06,
      "loss": 0.8518,
      "num_input_tokens_seen": 66059024,
      "step": 113860
    },
    {
      "epoch": 16.95933869526363,
      "grad_norm": 7.28125,
      "learning_rate": 3.4389475022846395e-06,
      "loss": 0.768,
      "num_input_tokens_seen": 66062000,
      "step": 113865
    },
    {
      "epoch": 16.96008340780459,
      "grad_norm": 12.6875,
      "learning_rate": 3.4373029711224356e-06,
      "loss": 1.1639,
      "num_input_tokens_seen": 66064944,
      "step": 113870
    },
    {
      "epoch": 16.960828120345546,
      "grad_norm": 6.40625,
      "learning_rate": 3.435658804240088e-06,
      "loss": 1.4131,
      "num_input_tokens_seen": 66067792,
      "step": 113875
    },
    {
      "epoch": 16.961572832886507,
      "grad_norm": 8.25,
      "learning_rate": 3.4340150016653695e-06,
      "loss": 0.8533,
      "num_input_tokens_seen": 66070864,
      "step": 113880
    },
    {
      "epoch": 16.962317545427464,
      "grad_norm": 4.03125,
      "learning_rate": 3.432371563426043e-06,
      "loss": 0.8508,
      "num_input_tokens_seen": 66073936,
      "step": 113885
    },
    {
      "epoch": 16.963062257968424,
      "grad_norm": 13.5,
      "learning_rate": 3.4307284895498836e-06,
      "loss": 1.2856,
      "num_input_tokens_seen": 66076720,
      "step": 113890
    },
    {
      "epoch": 16.963806970509385,
      "grad_norm": 6.375,
      "learning_rate": 3.429085780064639e-06,
      "loss": 1.3717,
      "num_input_tokens_seen": 66079632,
      "step": 113895
    },
    {
      "epoch": 16.964551683050342,
      "grad_norm": 16.5,
      "learning_rate": 3.427443434998073e-06,
      "loss": 1.4144,
      "num_input_tokens_seen": 66082448,
      "step": 113900
    },
    {
      "epoch": 16.965296395591302,
      "grad_norm": 6.59375,
      "learning_rate": 3.4258014543779222e-06,
      "loss": 1.0319,
      "num_input_tokens_seen": 66086480,
      "step": 113905
    },
    {
      "epoch": 16.96604110813226,
      "grad_norm": 9.1875,
      "learning_rate": 3.4241598382319303e-06,
      "loss": 1.0866,
      "num_input_tokens_seen": 66089392,
      "step": 113910
    },
    {
      "epoch": 16.96678582067322,
      "grad_norm": 14.9375,
      "learning_rate": 3.422518586587831e-06,
      "loss": 1.3083,
      "num_input_tokens_seen": 66092144,
      "step": 113915
    },
    {
      "epoch": 16.96753053321418,
      "grad_norm": 13.0625,
      "learning_rate": 3.4208776994733405e-06,
      "loss": 1.0469,
      "num_input_tokens_seen": 66094896,
      "step": 113920
    },
    {
      "epoch": 16.968275245755137,
      "grad_norm": 11.9375,
      "learning_rate": 3.4192371769161987e-06,
      "loss": 1.0214,
      "num_input_tokens_seen": 66097520,
      "step": 113925
    },
    {
      "epoch": 16.969019958296098,
      "grad_norm": 11.9375,
      "learning_rate": 3.417597018944102e-06,
      "loss": 0.8271,
      "num_input_tokens_seen": 66100240,
      "step": 113930
    },
    {
      "epoch": 16.96976467083706,
      "grad_norm": 9.25,
      "learning_rate": 3.415957225584776e-06,
      "loss": 1.3518,
      "num_input_tokens_seen": 66103280,
      "step": 113935
    },
    {
      "epoch": 16.970509383378015,
      "grad_norm": 6.84375,
      "learning_rate": 3.4143177968659098e-06,
      "loss": 0.9898,
      "num_input_tokens_seen": 66106096,
      "step": 113940
    },
    {
      "epoch": 16.971254095918976,
      "grad_norm": 16.75,
      "learning_rate": 3.412678732815211e-06,
      "loss": 1.3415,
      "num_input_tokens_seen": 66108688,
      "step": 113945
    },
    {
      "epoch": 16.971998808459933,
      "grad_norm": 8.1875,
      "learning_rate": 3.411040033460361e-06,
      "loss": 1.2883,
      "num_input_tokens_seen": 66111344,
      "step": 113950
    },
    {
      "epoch": 16.972743521000893,
      "grad_norm": 6.375,
      "learning_rate": 3.4094016988290512e-06,
      "loss": 1.2492,
      "num_input_tokens_seen": 66114192,
      "step": 113955
    },
    {
      "epoch": 16.973488233541854,
      "grad_norm": 7.59375,
      "learning_rate": 3.407763728948954e-06,
      "loss": 1.2203,
      "num_input_tokens_seen": 66117488,
      "step": 113960
    },
    {
      "epoch": 16.97423294608281,
      "grad_norm": 16.25,
      "learning_rate": 3.4061261238477414e-06,
      "loss": 1.0093,
      "num_input_tokens_seen": 66120400,
      "step": 113965
    },
    {
      "epoch": 16.97497765862377,
      "grad_norm": 11.625,
      "learning_rate": 3.4044888835530835e-06,
      "loss": 1.5727,
      "num_input_tokens_seen": 66123024,
      "step": 113970
    },
    {
      "epoch": 16.975722371164732,
      "grad_norm": 6.03125,
      "learning_rate": 3.4028520080926383e-06,
      "loss": 1.2651,
      "num_input_tokens_seen": 66125840,
      "step": 113975
    },
    {
      "epoch": 16.97646708370569,
      "grad_norm": 14.0625,
      "learning_rate": 3.401215497494059e-06,
      "loss": 1.5471,
      "num_input_tokens_seen": 66128656,
      "step": 113980
    },
    {
      "epoch": 16.97721179624665,
      "grad_norm": 15.3125,
      "learning_rate": 3.3995793517849846e-06,
      "loss": 0.9808,
      "num_input_tokens_seen": 66131504,
      "step": 113985
    },
    {
      "epoch": 16.977956508787607,
      "grad_norm": 13.0,
      "learning_rate": 3.3979435709930703e-06,
      "loss": 1.2895,
      "num_input_tokens_seen": 66134480,
      "step": 113990
    },
    {
      "epoch": 16.978701221328567,
      "grad_norm": 10.0625,
      "learning_rate": 3.3963081551459442e-06,
      "loss": 1.1233,
      "num_input_tokens_seen": 66137136,
      "step": 113995
    },
    {
      "epoch": 16.979445933869528,
      "grad_norm": 6.0,
      "learning_rate": 3.3946731042712286e-06,
      "loss": 1.0436,
      "num_input_tokens_seen": 66140560,
      "step": 114000
    },
    {
      "epoch": 16.980190646410485,
      "grad_norm": 12.1875,
      "learning_rate": 3.3930384183965573e-06,
      "loss": 1.011,
      "num_input_tokens_seen": 66143280,
      "step": 114005
    },
    {
      "epoch": 16.980935358951445,
      "grad_norm": 11.5,
      "learning_rate": 3.3914040975495387e-06,
      "loss": 1.1835,
      "num_input_tokens_seen": 66146256,
      "step": 114010
    },
    {
      "epoch": 16.981680071492406,
      "grad_norm": 11.4375,
      "learning_rate": 3.3897701417577893e-06,
      "loss": 1.1106,
      "num_input_tokens_seen": 66149232,
      "step": 114015
    },
    {
      "epoch": 16.982424784033363,
      "grad_norm": 8.5,
      "learning_rate": 3.388136551048904e-06,
      "loss": 0.9261,
      "num_input_tokens_seen": 66152176,
      "step": 114020
    },
    {
      "epoch": 16.983169496574323,
      "grad_norm": 5.875,
      "learning_rate": 3.386503325450491e-06,
      "loss": 0.9324,
      "num_input_tokens_seen": 66155024,
      "step": 114025
    },
    {
      "epoch": 16.98391420911528,
      "grad_norm": 10.875,
      "learning_rate": 3.3848704649901336e-06,
      "loss": 1.1869,
      "num_input_tokens_seen": 66157936,
      "step": 114030
    },
    {
      "epoch": 16.98465892165624,
      "grad_norm": 10.0625,
      "learning_rate": 3.3832379696954243e-06,
      "loss": 0.9366,
      "num_input_tokens_seen": 66160688,
      "step": 114035
    },
    {
      "epoch": 16.9854036341972,
      "grad_norm": 11.3125,
      "learning_rate": 3.3816058395939434e-06,
      "loss": 0.9628,
      "num_input_tokens_seen": 66163696,
      "step": 114040
    },
    {
      "epoch": 16.986148346738158,
      "grad_norm": 6.8125,
      "learning_rate": 3.3799740747132547e-06,
      "loss": 0.9999,
      "num_input_tokens_seen": 66166928,
      "step": 114045
    },
    {
      "epoch": 16.98689305927912,
      "grad_norm": 6.1875,
      "learning_rate": 3.378342675080934e-06,
      "loss": 0.9524,
      "num_input_tokens_seen": 66169872,
      "step": 114050
    },
    {
      "epoch": 16.987637771820076,
      "grad_norm": 6.84375,
      "learning_rate": 3.376711640724531e-06,
      "loss": 1.2339,
      "num_input_tokens_seen": 66172592,
      "step": 114055
    },
    {
      "epoch": 16.988382484361036,
      "grad_norm": 17.875,
      "learning_rate": 3.375080971671615e-06,
      "loss": 1.1073,
      "num_input_tokens_seen": 66175376,
      "step": 114060
    },
    {
      "epoch": 16.989127196901997,
      "grad_norm": 15.125,
      "learning_rate": 3.3734506679497207e-06,
      "loss": 1.5605,
      "num_input_tokens_seen": 66178992,
      "step": 114065
    },
    {
      "epoch": 16.989871909442954,
      "grad_norm": 8.5625,
      "learning_rate": 3.3718207295864028e-06,
      "loss": 1.0218,
      "num_input_tokens_seen": 66182032,
      "step": 114070
    },
    {
      "epoch": 16.990616621983914,
      "grad_norm": 8.5625,
      "learning_rate": 3.3701911566091925e-06,
      "loss": 0.8616,
      "num_input_tokens_seen": 66184656,
      "step": 114075
    },
    {
      "epoch": 16.991361334524875,
      "grad_norm": 11.0,
      "learning_rate": 3.368561949045615e-06,
      "loss": 0.9806,
      "num_input_tokens_seen": 66187504,
      "step": 114080
    },
    {
      "epoch": 16.992106047065832,
      "grad_norm": 12.25,
      "learning_rate": 3.3669331069232006e-06,
      "loss": 0.9235,
      "num_input_tokens_seen": 66190832,
      "step": 114085
    },
    {
      "epoch": 16.992850759606792,
      "grad_norm": 8.5,
      "learning_rate": 3.3653046302694614e-06,
      "loss": 0.9735,
      "num_input_tokens_seen": 66193968,
      "step": 114090
    },
    {
      "epoch": 16.99359547214775,
      "grad_norm": 13.75,
      "learning_rate": 3.3636765191119165e-06,
      "loss": 0.8405,
      "num_input_tokens_seen": 66196592,
      "step": 114095
    },
    {
      "epoch": 16.99434018468871,
      "grad_norm": 7.875,
      "learning_rate": 3.3620487734780603e-06,
      "loss": 1.3504,
      "num_input_tokens_seen": 66199664,
      "step": 114100
    },
    {
      "epoch": 16.99508489722967,
      "grad_norm": 9.4375,
      "learning_rate": 3.3604213933954048e-06,
      "loss": 1.2871,
      "num_input_tokens_seen": 66202544,
      "step": 114105
    },
    {
      "epoch": 16.995829609770627,
      "grad_norm": 6.15625,
      "learning_rate": 3.358794378891436e-06,
      "loss": 1.1463,
      "num_input_tokens_seen": 66205424,
      "step": 114110
    },
    {
      "epoch": 16.996574322311588,
      "grad_norm": 9.4375,
      "learning_rate": 3.3571677299936403e-06,
      "loss": 1.0392,
      "num_input_tokens_seen": 66208368,
      "step": 114115
    },
    {
      "epoch": 16.99731903485255,
      "grad_norm": 9.5,
      "learning_rate": 3.3555414467295017e-06,
      "loss": 1.0983,
      "num_input_tokens_seen": 66211280,
      "step": 114120
    },
    {
      "epoch": 16.998063747393505,
      "grad_norm": 11.5625,
      "learning_rate": 3.3539155291264833e-06,
      "loss": 1.3716,
      "num_input_tokens_seen": 66214256,
      "step": 114125
    },
    {
      "epoch": 16.998808459934466,
      "grad_norm": 17.625,
      "learning_rate": 3.352289977212067e-06,
      "loss": 1.0438,
      "num_input_tokens_seen": 66216784,
      "step": 114130
    },
    {
      "epoch": 16.999553172475423,
      "grad_norm": 9.125,
      "learning_rate": 3.3506647910137078e-06,
      "loss": 1.0066,
      "num_input_tokens_seen": 66219632,
      "step": 114135
    },
    {
      "epoch": 17.0,
      "eval_loss": 1.0852864980697632,
      "eval_runtime": 43.7758,
      "eval_samples_per_second": 68.166,
      "eval_steps_per_second": 17.041,
      "num_input_tokens_seen": 66220752,
      "step": 114138
    },
    {
      "epoch": 17.000297885016384,
      "grad_norm": 6.625,
      "learning_rate": 3.3490399705588677e-06,
      "loss": 1.0114,
      "num_input_tokens_seen": 66222064,
      "step": 114140
    },
    {
      "epoch": 17.001042597557344,
      "grad_norm": 6.59375,
      "learning_rate": 3.3474155158749854e-06,
      "loss": 0.8845,
      "num_input_tokens_seen": 66225104,
      "step": 114145
    },
    {
      "epoch": 17.0017873100983,
      "grad_norm": 8.1875,
      "learning_rate": 3.345791426989517e-06,
      "loss": 0.8122,
      "num_input_tokens_seen": 66228176,
      "step": 114150
    },
    {
      "epoch": 17.00253202263926,
      "grad_norm": 17.375,
      "learning_rate": 3.3441677039298956e-06,
      "loss": 1.2045,
      "num_input_tokens_seen": 66231216,
      "step": 114155
    },
    {
      "epoch": 17.003276735180222,
      "grad_norm": 9.125,
      "learning_rate": 3.3425443467235443e-06,
      "loss": 0.9207,
      "num_input_tokens_seen": 66233744,
      "step": 114160
    },
    {
      "epoch": 17.00402144772118,
      "grad_norm": 9.4375,
      "learning_rate": 3.3409213553979e-06,
      "loss": 1.0983,
      "num_input_tokens_seen": 66236912,
      "step": 114165
    },
    {
      "epoch": 17.00476616026214,
      "grad_norm": 8.1875,
      "learning_rate": 3.3392987299803753e-06,
      "loss": 0.7239,
      "num_input_tokens_seen": 66239504,
      "step": 114170
    },
    {
      "epoch": 17.005510872803097,
      "grad_norm": 14.125,
      "learning_rate": 3.33767647049838e-06,
      "loss": 0.9169,
      "num_input_tokens_seen": 66242384,
      "step": 114175
    },
    {
      "epoch": 17.006255585344057,
      "grad_norm": 15.5625,
      "learning_rate": 3.3360545769793277e-06,
      "loss": 0.767,
      "num_input_tokens_seen": 66245136,
      "step": 114180
    },
    {
      "epoch": 17.007000297885018,
      "grad_norm": 9.8125,
      "learning_rate": 3.3344330494506166e-06,
      "loss": 1.0441,
      "num_input_tokens_seen": 66247984,
      "step": 114185
    },
    {
      "epoch": 17.007745010425975,
      "grad_norm": 10.25,
      "learning_rate": 3.3328118879396324e-06,
      "loss": 1.0158,
      "num_input_tokens_seen": 66250928,
      "step": 114190
    },
    {
      "epoch": 17.008489722966935,
      "grad_norm": 6.3125,
      "learning_rate": 3.331191092473776e-06,
      "loss": 1.0181,
      "num_input_tokens_seen": 66253712,
      "step": 114195
    },
    {
      "epoch": 17.009234435507896,
      "grad_norm": 4.65625,
      "learning_rate": 3.3295706630804222e-06,
      "loss": 0.865,
      "num_input_tokens_seen": 66256720,
      "step": 114200
    },
    {
      "epoch": 17.009979148048853,
      "grad_norm": 4.84375,
      "learning_rate": 3.3279505997869442e-06,
      "loss": 1.1104,
      "num_input_tokens_seen": 66259504,
      "step": 114205
    },
    {
      "epoch": 17.010723860589813,
      "grad_norm": 9.4375,
      "learning_rate": 3.3263309026207166e-06,
      "loss": 0.9454,
      "num_input_tokens_seen": 66262480,
      "step": 114210
    },
    {
      "epoch": 17.01146857313077,
      "grad_norm": 11.625,
      "learning_rate": 3.3247115716090987e-06,
      "loss": 1.1538,
      "num_input_tokens_seen": 66265168,
      "step": 114215
    },
    {
      "epoch": 17.01221328567173,
      "grad_norm": 9.3125,
      "learning_rate": 3.3230926067794516e-06,
      "loss": 0.9412,
      "num_input_tokens_seen": 66268080,
      "step": 114220
    },
    {
      "epoch": 17.01295799821269,
      "grad_norm": 15.3125,
      "learning_rate": 3.3214740081591173e-06,
      "loss": 1.4625,
      "num_input_tokens_seen": 66271280,
      "step": 114225
    },
    {
      "epoch": 17.01370271075365,
      "grad_norm": 8.375,
      "learning_rate": 3.3198557757754544e-06,
      "loss": 0.8467,
      "num_input_tokens_seen": 66274256,
      "step": 114230
    },
    {
      "epoch": 17.01444742329461,
      "grad_norm": 8.8125,
      "learning_rate": 3.3182379096557916e-06,
      "loss": 1.0575,
      "num_input_tokens_seen": 66277040,
      "step": 114235
    },
    {
      "epoch": 17.015192135835566,
      "grad_norm": 6.6875,
      "learning_rate": 3.3166204098274643e-06,
      "loss": 1.2312,
      "num_input_tokens_seen": 66279952,
      "step": 114240
    },
    {
      "epoch": 17.015936848376526,
      "grad_norm": 7.375,
      "learning_rate": 3.3150032763177962e-06,
      "loss": 1.1781,
      "num_input_tokens_seen": 66282768,
      "step": 114245
    },
    {
      "epoch": 17.016681560917487,
      "grad_norm": 8.9375,
      "learning_rate": 3.3133865091541037e-06,
      "loss": 1.0324,
      "num_input_tokens_seen": 66285616,
      "step": 114250
    },
    {
      "epoch": 17.017426273458444,
      "grad_norm": 17.625,
      "learning_rate": 3.31177010836371e-06,
      "loss": 1.4228,
      "num_input_tokens_seen": 66288624,
      "step": 114255
    },
    {
      "epoch": 17.018170985999404,
      "grad_norm": 18.375,
      "learning_rate": 3.310154073973909e-06,
      "loss": 1.2683,
      "num_input_tokens_seen": 66291376,
      "step": 114260
    },
    {
      "epoch": 17.018915698540365,
      "grad_norm": 10.375,
      "learning_rate": 3.3085384060120185e-06,
      "loss": 1.0289,
      "num_input_tokens_seen": 66294192,
      "step": 114265
    },
    {
      "epoch": 17.019660411081322,
      "grad_norm": 21.875,
      "learning_rate": 3.3069231045053216e-06,
      "loss": 1.2694,
      "num_input_tokens_seen": 66296816,
      "step": 114270
    },
    {
      "epoch": 17.020405123622282,
      "grad_norm": 10.6875,
      "learning_rate": 3.3053081694811137e-06,
      "loss": 1.1786,
      "num_input_tokens_seen": 66299792,
      "step": 114275
    },
    {
      "epoch": 17.02114983616324,
      "grad_norm": 7.8125,
      "learning_rate": 3.303693600966676e-06,
      "loss": 1.313,
      "num_input_tokens_seen": 66302608,
      "step": 114280
    },
    {
      "epoch": 17.0218945487042,
      "grad_norm": 10.5,
      "learning_rate": 3.3020793989892774e-06,
      "loss": 0.871,
      "num_input_tokens_seen": 66305328,
      "step": 114285
    },
    {
      "epoch": 17.02263926124516,
      "grad_norm": 13.25,
      "learning_rate": 3.3004655635761994e-06,
      "loss": 1.0052,
      "num_input_tokens_seen": 66308464,
      "step": 114290
    },
    {
      "epoch": 17.023383973786117,
      "grad_norm": 19.625,
      "learning_rate": 3.298852094754698e-06,
      "loss": 1.2641,
      "num_input_tokens_seen": 66311504,
      "step": 114295
    },
    {
      "epoch": 17.024128686327078,
      "grad_norm": 9.3125,
      "learning_rate": 3.29723899255204e-06,
      "loss": 1.1045,
      "num_input_tokens_seen": 66314512,
      "step": 114300
    },
    {
      "epoch": 17.02487339886804,
      "grad_norm": 4.0,
      "learning_rate": 3.295626256995471e-06,
      "loss": 0.941,
      "num_input_tokens_seen": 66317328,
      "step": 114305
    },
    {
      "epoch": 17.025618111408996,
      "grad_norm": 12.75,
      "learning_rate": 3.294013888112235e-06,
      "loss": 1.4139,
      "num_input_tokens_seen": 66320208,
      "step": 114310
    },
    {
      "epoch": 17.026362823949956,
      "grad_norm": 6.125,
      "learning_rate": 3.2924018859295746e-06,
      "loss": 1.0862,
      "num_input_tokens_seen": 66323088,
      "step": 114315
    },
    {
      "epoch": 17.027107536490913,
      "grad_norm": 10.5625,
      "learning_rate": 3.290790250474718e-06,
      "loss": 1.2643,
      "num_input_tokens_seen": 66325936,
      "step": 114320
    },
    {
      "epoch": 17.027852249031874,
      "grad_norm": 6.90625,
      "learning_rate": 3.2891789817748984e-06,
      "loss": 1.3069,
      "num_input_tokens_seen": 66328752,
      "step": 114325
    },
    {
      "epoch": 17.028596961572834,
      "grad_norm": 8.875,
      "learning_rate": 3.287568079857331e-06,
      "loss": 0.7675,
      "num_input_tokens_seen": 66331696,
      "step": 114330
    },
    {
      "epoch": 17.02934167411379,
      "grad_norm": 4.21875,
      "learning_rate": 3.285957544749238e-06,
      "loss": 1.012,
      "num_input_tokens_seen": 66334608,
      "step": 114335
    },
    {
      "epoch": 17.03008638665475,
      "grad_norm": 16.25,
      "learning_rate": 3.284347376477817e-06,
      "loss": 0.9849,
      "num_input_tokens_seen": 66337456,
      "step": 114340
    },
    {
      "epoch": 17.030831099195712,
      "grad_norm": 6.875,
      "learning_rate": 3.2827375750702825e-06,
      "loss": 1.4717,
      "num_input_tokens_seen": 66340464,
      "step": 114345
    },
    {
      "epoch": 17.03157581173667,
      "grad_norm": 10.5625,
      "learning_rate": 3.2811281405538188e-06,
      "loss": 0.9336,
      "num_input_tokens_seen": 66343120,
      "step": 114350
    },
    {
      "epoch": 17.03232052427763,
      "grad_norm": 5.75,
      "learning_rate": 3.2795190729556254e-06,
      "loss": 1.1111,
      "num_input_tokens_seen": 66346256,
      "step": 114355
    },
    {
      "epoch": 17.033065236818587,
      "grad_norm": 6.15625,
      "learning_rate": 3.2779103723028807e-06,
      "loss": 1.017,
      "num_input_tokens_seen": 66349040,
      "step": 114360
    },
    {
      "epoch": 17.033809949359547,
      "grad_norm": 10.1875,
      "learning_rate": 3.276302038622761e-06,
      "loss": 0.9422,
      "num_input_tokens_seen": 66351824,
      "step": 114365
    },
    {
      "epoch": 17.034554661900508,
      "grad_norm": 7.03125,
      "learning_rate": 3.2746940719424414e-06,
      "loss": 1.0084,
      "num_input_tokens_seen": 66354576,
      "step": 114370
    },
    {
      "epoch": 17.035299374441465,
      "grad_norm": 11.3125,
      "learning_rate": 3.2730864722890886e-06,
      "loss": 1.0528,
      "num_input_tokens_seen": 66357296,
      "step": 114375
    },
    {
      "epoch": 17.036044086982425,
      "grad_norm": 12.25,
      "learning_rate": 3.2714792396898534e-06,
      "loss": 1.1694,
      "num_input_tokens_seen": 66360336,
      "step": 114380
    },
    {
      "epoch": 17.036788799523382,
      "grad_norm": 16.125,
      "learning_rate": 3.2698723741718894e-06,
      "loss": 1.6502,
      "num_input_tokens_seen": 66363216,
      "step": 114385
    },
    {
      "epoch": 17.037533512064343,
      "grad_norm": 10.25,
      "learning_rate": 3.2682658757623526e-06,
      "loss": 1.0766,
      "num_input_tokens_seen": 66366128,
      "step": 114390
    },
    {
      "epoch": 17.038278224605303,
      "grad_norm": 11.625,
      "learning_rate": 3.2666597444883734e-06,
      "loss": 1.4275,
      "num_input_tokens_seen": 66369104,
      "step": 114395
    },
    {
      "epoch": 17.03902293714626,
      "grad_norm": 14.5625,
      "learning_rate": 3.265053980377086e-06,
      "loss": 1.0593,
      "num_input_tokens_seen": 66372240,
      "step": 114400
    },
    {
      "epoch": 17.03976764968722,
      "grad_norm": 10.75,
      "learning_rate": 3.2634485834556276e-06,
      "loss": 1.2046,
      "num_input_tokens_seen": 66375088,
      "step": 114405
    },
    {
      "epoch": 17.04051236222818,
      "grad_norm": 13.625,
      "learning_rate": 3.2618435537511066e-06,
      "loss": 1.1143,
      "num_input_tokens_seen": 66378032,
      "step": 114410
    },
    {
      "epoch": 17.04125707476914,
      "grad_norm": 5.09375,
      "learning_rate": 3.2602388912906482e-06,
      "loss": 1.1935,
      "num_input_tokens_seen": 66381296,
      "step": 114415
    },
    {
      "epoch": 17.0420017873101,
      "grad_norm": 8.25,
      "learning_rate": 3.2586345961013565e-06,
      "loss": 0.6658,
      "num_input_tokens_seen": 66384208,
      "step": 114420
    },
    {
      "epoch": 17.042746499851056,
      "grad_norm": 7.28125,
      "learning_rate": 3.2570306682103396e-06,
      "loss": 1.1645,
      "num_input_tokens_seen": 66387056,
      "step": 114425
    },
    {
      "epoch": 17.043491212392016,
      "grad_norm": 7.5,
      "learning_rate": 3.2554271076446873e-06,
      "loss": 1.1495,
      "num_input_tokens_seen": 66390512,
      "step": 114430
    },
    {
      "epoch": 17.044235924932977,
      "grad_norm": 11.8125,
      "learning_rate": 3.2538239144314974e-06,
      "loss": 0.8108,
      "num_input_tokens_seen": 66393232,
      "step": 114435
    },
    {
      "epoch": 17.044980637473934,
      "grad_norm": 4.0625,
      "learning_rate": 3.252221088597854e-06,
      "loss": 1.1004,
      "num_input_tokens_seen": 66396080,
      "step": 114440
    },
    {
      "epoch": 17.045725350014894,
      "grad_norm": 18.5,
      "learning_rate": 3.250618630170829e-06,
      "loss": 1.2893,
      "num_input_tokens_seen": 66398960,
      "step": 114445
    },
    {
      "epoch": 17.046470062555855,
      "grad_norm": 10.0,
      "learning_rate": 3.2490165391774963e-06,
      "loss": 1.091,
      "num_input_tokens_seen": 66402416,
      "step": 114450
    },
    {
      "epoch": 17.047214775096812,
      "grad_norm": 14.875,
      "learning_rate": 3.2474148156449195e-06,
      "loss": 0.8452,
      "num_input_tokens_seen": 66405072,
      "step": 114455
    },
    {
      "epoch": 17.047959487637772,
      "grad_norm": 9.9375,
      "learning_rate": 3.2458134596001636e-06,
      "loss": 0.9476,
      "num_input_tokens_seen": 66407696,
      "step": 114460
    },
    {
      "epoch": 17.04870420017873,
      "grad_norm": 10.1875,
      "learning_rate": 3.2442124710702764e-06,
      "loss": 1.1918,
      "num_input_tokens_seen": 66410896,
      "step": 114465
    },
    {
      "epoch": 17.04944891271969,
      "grad_norm": 7.6875,
      "learning_rate": 3.242611850082314e-06,
      "loss": 1.0536,
      "num_input_tokens_seen": 66414000,
      "step": 114470
    },
    {
      "epoch": 17.05019362526065,
      "grad_norm": 17.875,
      "learning_rate": 3.2410115966633044e-06,
      "loss": 1.184,
      "num_input_tokens_seen": 66417040,
      "step": 114475
    },
    {
      "epoch": 17.050938337801608,
      "grad_norm": 7.4375,
      "learning_rate": 3.239411710840293e-06,
      "loss": 0.947,
      "num_input_tokens_seen": 66420112,
      "step": 114480
    },
    {
      "epoch": 17.051683050342568,
      "grad_norm": 10.8125,
      "learning_rate": 3.2378121926403077e-06,
      "loss": 0.816,
      "num_input_tokens_seen": 66423056,
      "step": 114485
    },
    {
      "epoch": 17.05242776288353,
      "grad_norm": 8.125,
      "learning_rate": 3.236213042090358e-06,
      "loss": 1.1172,
      "num_input_tokens_seen": 66426160,
      "step": 114490
    },
    {
      "epoch": 17.053172475424486,
      "grad_norm": 7.53125,
      "learning_rate": 3.234614259217478e-06,
      "loss": 0.9439,
      "num_input_tokens_seen": 66428816,
      "step": 114495
    },
    {
      "epoch": 17.053917187965446,
      "grad_norm": 8.6875,
      "learning_rate": 3.2330158440486672e-06,
      "loss": 1.012,
      "num_input_tokens_seen": 66431568,
      "step": 114500
    },
    {
      "epoch": 17.054661900506403,
      "grad_norm": 10.125,
      "learning_rate": 3.231417796610925e-06,
      "loss": 1.2032,
      "num_input_tokens_seen": 66434544,
      "step": 114505
    },
    {
      "epoch": 17.055406613047364,
      "grad_norm": 10.625,
      "learning_rate": 3.229820116931259e-06,
      "loss": 1.1955,
      "num_input_tokens_seen": 66437200,
      "step": 114510
    },
    {
      "epoch": 17.056151325588324,
      "grad_norm": 9.0625,
      "learning_rate": 3.228222805036657e-06,
      "loss": 1.0341,
      "num_input_tokens_seen": 66440080,
      "step": 114515
    },
    {
      "epoch": 17.05689603812928,
      "grad_norm": 12.875,
      "learning_rate": 3.226625860954105e-06,
      "loss": 1.4158,
      "num_input_tokens_seen": 66442960,
      "step": 114520
    },
    {
      "epoch": 17.05764075067024,
      "grad_norm": 18.5,
      "learning_rate": 3.225029284710571e-06,
      "loss": 1.3869,
      "num_input_tokens_seen": 66445776,
      "step": 114525
    },
    {
      "epoch": 17.058385463211202,
      "grad_norm": 16.125,
      "learning_rate": 3.2234330763330432e-06,
      "loss": 1.1986,
      "num_input_tokens_seen": 66448624,
      "step": 114530
    },
    {
      "epoch": 17.05913017575216,
      "grad_norm": 12.1875,
      "learning_rate": 3.221837235848474e-06,
      "loss": 1.3078,
      "num_input_tokens_seen": 66451408,
      "step": 114535
    },
    {
      "epoch": 17.05987488829312,
      "grad_norm": 7.15625,
      "learning_rate": 3.220241763283838e-06,
      "loss": 0.9765,
      "num_input_tokens_seen": 66454000,
      "step": 114540
    },
    {
      "epoch": 17.060619600834077,
      "grad_norm": 10.0,
      "learning_rate": 3.2186466586660746e-06,
      "loss": 0.8381,
      "num_input_tokens_seen": 66456752,
      "step": 114545
    },
    {
      "epoch": 17.061364313375037,
      "grad_norm": 13.3125,
      "learning_rate": 3.2170519220221435e-06,
      "loss": 1.0028,
      "num_input_tokens_seen": 66459536,
      "step": 114550
    },
    {
      "epoch": 17.062109025915998,
      "grad_norm": 14.8125,
      "learning_rate": 3.2154575533789753e-06,
      "loss": 0.969,
      "num_input_tokens_seen": 66462320,
      "step": 114555
    },
    {
      "epoch": 17.062853738456955,
      "grad_norm": 12.375,
      "learning_rate": 3.2138635527635186e-06,
      "loss": 0.9514,
      "num_input_tokens_seen": 66465232,
      "step": 114560
    },
    {
      "epoch": 17.063598450997915,
      "grad_norm": 9.4375,
      "learning_rate": 3.2122699202026927e-06,
      "loss": 1.3401,
      "num_input_tokens_seen": 66468208,
      "step": 114565
    },
    {
      "epoch": 17.064343163538872,
      "grad_norm": 6.3125,
      "learning_rate": 3.2106766557234243e-06,
      "loss": 0.9438,
      "num_input_tokens_seen": 66471056,
      "step": 114570
    },
    {
      "epoch": 17.065087876079833,
      "grad_norm": 13.75,
      "learning_rate": 3.209083759352627e-06,
      "loss": 1.2892,
      "num_input_tokens_seen": 66473872,
      "step": 114575
    },
    {
      "epoch": 17.065832588620793,
      "grad_norm": 11.3125,
      "learning_rate": 3.2074912311172046e-06,
      "loss": 1.0375,
      "num_input_tokens_seen": 66476432,
      "step": 114580
    },
    {
      "epoch": 17.06657730116175,
      "grad_norm": 10.875,
      "learning_rate": 3.2058990710440773e-06,
      "loss": 1.2834,
      "num_input_tokens_seen": 66479248,
      "step": 114585
    },
    {
      "epoch": 17.06732201370271,
      "grad_norm": 5.8125,
      "learning_rate": 3.2043072791601293e-06,
      "loss": 0.8375,
      "num_input_tokens_seen": 66481808,
      "step": 114590
    },
    {
      "epoch": 17.06806672624367,
      "grad_norm": 14.0625,
      "learning_rate": 3.202715855492261e-06,
      "loss": 1.0003,
      "num_input_tokens_seen": 66484752,
      "step": 114595
    },
    {
      "epoch": 17.06881143878463,
      "grad_norm": 13.6875,
      "learning_rate": 3.201124800067357e-06,
      "loss": 0.8954,
      "num_input_tokens_seen": 66487536,
      "step": 114600
    },
    {
      "epoch": 17.06955615132559,
      "grad_norm": 12.6875,
      "learning_rate": 3.1995341129122864e-06,
      "loss": 0.9001,
      "num_input_tokens_seen": 66490352,
      "step": 114605
    },
    {
      "epoch": 17.070300863866546,
      "grad_norm": 11.0,
      "learning_rate": 3.197943794053937e-06,
      "loss": 0.9642,
      "num_input_tokens_seen": 66493136,
      "step": 114610
    },
    {
      "epoch": 17.071045576407506,
      "grad_norm": 5.625,
      "learning_rate": 3.196353843519162e-06,
      "loss": 0.9761,
      "num_input_tokens_seen": 66496080,
      "step": 114615
    },
    {
      "epoch": 17.071790288948467,
      "grad_norm": 11.1875,
      "learning_rate": 3.1947642613348344e-06,
      "loss": 1.2596,
      "num_input_tokens_seen": 66499440,
      "step": 114620
    },
    {
      "epoch": 17.072535001489424,
      "grad_norm": 12.25,
      "learning_rate": 3.193175047527797e-06,
      "loss": 1.0726,
      "num_input_tokens_seen": 66502544,
      "step": 114625
    },
    {
      "epoch": 17.073279714030384,
      "grad_norm": 9.375,
      "learning_rate": 3.1915862021249105e-06,
      "loss": 0.887,
      "num_input_tokens_seen": 66505200,
      "step": 114630
    },
    {
      "epoch": 17.074024426571345,
      "grad_norm": 4.90625,
      "learning_rate": 3.18999772515301e-06,
      "loss": 1.1091,
      "num_input_tokens_seen": 66508048,
      "step": 114635
    },
    {
      "epoch": 17.074769139112302,
      "grad_norm": 7.96875,
      "learning_rate": 3.1884096166389292e-06,
      "loss": 1.2689,
      "num_input_tokens_seen": 66510832,
      "step": 114640
    },
    {
      "epoch": 17.075513851653263,
      "grad_norm": 8.125,
      "learning_rate": 3.1868218766095e-06,
      "loss": 1.2673,
      "num_input_tokens_seen": 66513776,
      "step": 114645
    },
    {
      "epoch": 17.07625856419422,
      "grad_norm": 9.9375,
      "learning_rate": 3.1852345050915415e-06,
      "loss": 1.0969,
      "num_input_tokens_seen": 66516752,
      "step": 114650
    },
    {
      "epoch": 17.07700327673518,
      "grad_norm": 6.5,
      "learning_rate": 3.1836475021118804e-06,
      "loss": 1.3613,
      "num_input_tokens_seen": 66519664,
      "step": 114655
    },
    {
      "epoch": 17.07774798927614,
      "grad_norm": 12.5625,
      "learning_rate": 3.1820608676973144e-06,
      "loss": 1.2435,
      "num_input_tokens_seen": 66522704,
      "step": 114660
    },
    {
      "epoch": 17.078492701817098,
      "grad_norm": 16.5,
      "learning_rate": 3.180474601874661e-06,
      "loss": 1.1146,
      "num_input_tokens_seen": 66525648,
      "step": 114665
    },
    {
      "epoch": 17.079237414358058,
      "grad_norm": 5.34375,
      "learning_rate": 3.1788887046707072e-06,
      "loss": 1.0576,
      "num_input_tokens_seen": 66528656,
      "step": 114670
    },
    {
      "epoch": 17.07998212689902,
      "grad_norm": 8.6875,
      "learning_rate": 3.177303176112256e-06,
      "loss": 1.143,
      "num_input_tokens_seen": 66531344,
      "step": 114675
    },
    {
      "epoch": 17.080726839439976,
      "grad_norm": 15.875,
      "learning_rate": 3.1757180162260897e-06,
      "loss": 0.9126,
      "num_input_tokens_seen": 66534096,
      "step": 114680
    },
    {
      "epoch": 17.081471551980936,
      "grad_norm": 12.25,
      "learning_rate": 3.174133225038978e-06,
      "loss": 1.1151,
      "num_input_tokens_seen": 66537168,
      "step": 114685
    },
    {
      "epoch": 17.082216264521893,
      "grad_norm": 12.375,
      "learning_rate": 3.17254880257771e-06,
      "loss": 1.0902,
      "num_input_tokens_seen": 66539888,
      "step": 114690
    },
    {
      "epoch": 17.082960977062854,
      "grad_norm": 8.6875,
      "learning_rate": 3.1709647488690404e-06,
      "loss": 0.8617,
      "num_input_tokens_seen": 66542704,
      "step": 114695
    },
    {
      "epoch": 17.083705689603814,
      "grad_norm": 7.84375,
      "learning_rate": 3.1693810639397412e-06,
      "loss": 1.1403,
      "num_input_tokens_seen": 66545712,
      "step": 114700
    },
    {
      "epoch": 17.08445040214477,
      "grad_norm": 12.1875,
      "learning_rate": 3.1677977478165588e-06,
      "loss": 1.0433,
      "num_input_tokens_seen": 66548464,
      "step": 114705
    },
    {
      "epoch": 17.08519511468573,
      "grad_norm": 12.25,
      "learning_rate": 3.166214800526246e-06,
      "loss": 0.7761,
      "num_input_tokens_seen": 66551664,
      "step": 114710
    },
    {
      "epoch": 17.085939827226692,
      "grad_norm": 8.75,
      "learning_rate": 3.1646322220955372e-06,
      "loss": 1.1285,
      "num_input_tokens_seen": 66554352,
      "step": 114715
    },
    {
      "epoch": 17.08668453976765,
      "grad_norm": 4.90625,
      "learning_rate": 3.16305001255118e-06,
      "loss": 0.7435,
      "num_input_tokens_seen": 66557328,
      "step": 114720
    },
    {
      "epoch": 17.08742925230861,
      "grad_norm": 11.375,
      "learning_rate": 3.1614681719199015e-06,
      "loss": 1.2793,
      "num_input_tokens_seen": 66560112,
      "step": 114725
    },
    {
      "epoch": 17.088173964849567,
      "grad_norm": 6.3125,
      "learning_rate": 3.1598867002284148e-06,
      "loss": 1.1852,
      "num_input_tokens_seen": 66563184,
      "step": 114730
    },
    {
      "epoch": 17.088918677390527,
      "grad_norm": 16.375,
      "learning_rate": 3.15830559750345e-06,
      "loss": 1.0367,
      "num_input_tokens_seen": 66566032,
      "step": 114735
    },
    {
      "epoch": 17.089663389931488,
      "grad_norm": 8.1875,
      "learning_rate": 3.1567248637717066e-06,
      "loss": 0.7637,
      "num_input_tokens_seen": 66568912,
      "step": 114740
    },
    {
      "epoch": 17.090408102472445,
      "grad_norm": 16.375,
      "learning_rate": 3.1551444990599033e-06,
      "loss": 1.3119,
      "num_input_tokens_seen": 66572048,
      "step": 114745
    },
    {
      "epoch": 17.091152815013405,
      "grad_norm": 7.9375,
      "learning_rate": 3.1535645033947265e-06,
      "loss": 0.9065,
      "num_input_tokens_seen": 66574704,
      "step": 114750
    },
    {
      "epoch": 17.091897527554362,
      "grad_norm": 10.625,
      "learning_rate": 3.15198487680288e-06,
      "loss": 1.0624,
      "num_input_tokens_seen": 66577456,
      "step": 114755
    },
    {
      "epoch": 17.092642240095323,
      "grad_norm": 10.3125,
      "learning_rate": 3.150405619311042e-06,
      "loss": 0.9034,
      "num_input_tokens_seen": 66580688,
      "step": 114760
    },
    {
      "epoch": 17.093386952636283,
      "grad_norm": 6.5625,
      "learning_rate": 3.148826730945889e-06,
      "loss": 1.0689,
      "num_input_tokens_seen": 66583824,
      "step": 114765
    },
    {
      "epoch": 17.09413166517724,
      "grad_norm": 10.6875,
      "learning_rate": 3.147248211734105e-06,
      "loss": 1.1399,
      "num_input_tokens_seen": 66586896,
      "step": 114770
    },
    {
      "epoch": 17.0948763777182,
      "grad_norm": 7.0,
      "learning_rate": 3.145670061702352e-06,
      "loss": 1.0872,
      "num_input_tokens_seen": 66589840,
      "step": 114775
    },
    {
      "epoch": 17.09562109025916,
      "grad_norm": 17.625,
      "learning_rate": 3.144092280877292e-06,
      "loss": 0.8616,
      "num_input_tokens_seen": 66593040,
      "step": 114780
    },
    {
      "epoch": 17.09636580280012,
      "grad_norm": 13.625,
      "learning_rate": 3.1425148692855734e-06,
      "loss": 0.8578,
      "num_input_tokens_seen": 66595824,
      "step": 114785
    },
    {
      "epoch": 17.09711051534108,
      "grad_norm": 12.8125,
      "learning_rate": 3.1409378269538574e-06,
      "loss": 1.1286,
      "num_input_tokens_seen": 66598832,
      "step": 114790
    },
    {
      "epoch": 17.097855227882036,
      "grad_norm": 17.125,
      "learning_rate": 3.1393611539087765e-06,
      "loss": 1.3732,
      "num_input_tokens_seen": 66601936,
      "step": 114795
    },
    {
      "epoch": 17.098599940422996,
      "grad_norm": 11.0,
      "learning_rate": 3.1377848501769724e-06,
      "loss": 1.1355,
      "num_input_tokens_seen": 66605008,
      "step": 114800
    },
    {
      "epoch": 17.099344652963957,
      "grad_norm": 15.0,
      "learning_rate": 3.136208915785077e-06,
      "loss": 1.4984,
      "num_input_tokens_seen": 66608016,
      "step": 114805
    },
    {
      "epoch": 17.100089365504914,
      "grad_norm": 10.375,
      "learning_rate": 3.1346333507597027e-06,
      "loss": 1.2252,
      "num_input_tokens_seen": 66610576,
      "step": 114810
    },
    {
      "epoch": 17.100834078045875,
      "grad_norm": 7.78125,
      "learning_rate": 3.1330581551274827e-06,
      "loss": 0.8609,
      "num_input_tokens_seen": 66613424,
      "step": 114815
    },
    {
      "epoch": 17.101578790586835,
      "grad_norm": 9.875,
      "learning_rate": 3.1314833289150138e-06,
      "loss": 1.1185,
      "num_input_tokens_seen": 66616176,
      "step": 114820
    },
    {
      "epoch": 17.102323503127792,
      "grad_norm": 12.1875,
      "learning_rate": 3.129908872148912e-06,
      "loss": 1.4387,
      "num_input_tokens_seen": 66619216,
      "step": 114825
    },
    {
      "epoch": 17.103068215668753,
      "grad_norm": 9.5,
      "learning_rate": 3.128334784855774e-06,
      "loss": 1.2344,
      "num_input_tokens_seen": 66622224,
      "step": 114830
    },
    {
      "epoch": 17.10381292820971,
      "grad_norm": 13.125,
      "learning_rate": 3.126761067062184e-06,
      "loss": 0.9352,
      "num_input_tokens_seen": 66625168,
      "step": 114835
    },
    {
      "epoch": 17.10455764075067,
      "grad_norm": 7.4375,
      "learning_rate": 3.125187718794742e-06,
      "loss": 1.0539,
      "num_input_tokens_seen": 66628240,
      "step": 114840
    },
    {
      "epoch": 17.10530235329163,
      "grad_norm": 7.96875,
      "learning_rate": 3.1236147400800194e-06,
      "loss": 0.8776,
      "num_input_tokens_seen": 66631280,
      "step": 114845
    },
    {
      "epoch": 17.106047065832588,
      "grad_norm": 8.875,
      "learning_rate": 3.1220421309445913e-06,
      "loss": 1.0385,
      "num_input_tokens_seen": 66634128,
      "step": 114850
    },
    {
      "epoch": 17.106791778373548,
      "grad_norm": 9.0625,
      "learning_rate": 3.1204698914150205e-06,
      "loss": 0.9025,
      "num_input_tokens_seen": 66636848,
      "step": 114855
    },
    {
      "epoch": 17.10753649091451,
      "grad_norm": 8.3125,
      "learning_rate": 3.11889802151788e-06,
      "loss": 1.3493,
      "num_input_tokens_seen": 66640112,
      "step": 114860
    },
    {
      "epoch": 17.108281203455466,
      "grad_norm": 8.9375,
      "learning_rate": 3.117326521279712e-06,
      "loss": 1.2569,
      "num_input_tokens_seen": 66643184,
      "step": 114865
    },
    {
      "epoch": 17.109025915996426,
      "grad_norm": 13.4375,
      "learning_rate": 3.1157553907270766e-06,
      "loss": 1.5766,
      "num_input_tokens_seen": 66645712,
      "step": 114870
    },
    {
      "epoch": 17.109770628537383,
      "grad_norm": 11.875,
      "learning_rate": 3.1141846298865074e-06,
      "loss": 1.3347,
      "num_input_tokens_seen": 66648400,
      "step": 114875
    },
    {
      "epoch": 17.110515341078344,
      "grad_norm": 21.5,
      "learning_rate": 3.11261423878455e-06,
      "loss": 1.2792,
      "num_input_tokens_seen": 66651056,
      "step": 114880
    },
    {
      "epoch": 17.111260053619304,
      "grad_norm": 11.25,
      "learning_rate": 3.111044217447731e-06,
      "loss": 1.2394,
      "num_input_tokens_seen": 66654032,
      "step": 114885
    },
    {
      "epoch": 17.11200476616026,
      "grad_norm": 16.5,
      "learning_rate": 3.1094745659025674e-06,
      "loss": 1.3284,
      "num_input_tokens_seen": 66656784,
      "step": 114890
    },
    {
      "epoch": 17.11274947870122,
      "grad_norm": 5.0,
      "learning_rate": 3.1079052841755857e-06,
      "loss": 0.9886,
      "num_input_tokens_seen": 66659760,
      "step": 114895
    },
    {
      "epoch": 17.113494191242182,
      "grad_norm": 13.9375,
      "learning_rate": 3.1063363722932975e-06,
      "loss": 0.8203,
      "num_input_tokens_seen": 66662736,
      "step": 114900
    },
    {
      "epoch": 17.11423890378314,
      "grad_norm": 9.9375,
      "learning_rate": 3.1047678302822016e-06,
      "loss": 1.021,
      "num_input_tokens_seen": 66665552,
      "step": 114905
    },
    {
      "epoch": 17.1149836163241,
      "grad_norm": 7.03125,
      "learning_rate": 3.1031996581687955e-06,
      "loss": 0.9186,
      "num_input_tokens_seen": 66668432,
      "step": 114910
    },
    {
      "epoch": 17.115728328865057,
      "grad_norm": 2.953125,
      "learning_rate": 3.101631855979581e-06,
      "loss": 1.1589,
      "num_input_tokens_seen": 66671120,
      "step": 114915
    },
    {
      "epoch": 17.116473041406017,
      "grad_norm": 24.0,
      "learning_rate": 3.100064423741042e-06,
      "loss": 1.3484,
      "num_input_tokens_seen": 66674128,
      "step": 114920
    },
    {
      "epoch": 17.117217753946978,
      "grad_norm": 15.625,
      "learning_rate": 3.098497361479649e-06,
      "loss": 1.2084,
      "num_input_tokens_seen": 66676784,
      "step": 114925
    },
    {
      "epoch": 17.117962466487935,
      "grad_norm": 6.21875,
      "learning_rate": 3.0969306692218897e-06,
      "loss": 0.963,
      "num_input_tokens_seen": 66679888,
      "step": 114930
    },
    {
      "epoch": 17.118707179028895,
      "grad_norm": 17.5,
      "learning_rate": 3.0953643469942173e-06,
      "loss": 1.3255,
      "num_input_tokens_seen": 66682928,
      "step": 114935
    },
    {
      "epoch": 17.119451891569852,
      "grad_norm": 8.25,
      "learning_rate": 3.093798394823111e-06,
      "loss": 1.1894,
      "num_input_tokens_seen": 66685648,
      "step": 114940
    },
    {
      "epoch": 17.120196604110813,
      "grad_norm": 15.75,
      "learning_rate": 3.0922328127350076e-06,
      "loss": 1.3067,
      "num_input_tokens_seen": 66688528,
      "step": 114945
    },
    {
      "epoch": 17.120941316651773,
      "grad_norm": 9.5,
      "learning_rate": 3.090667600756372e-06,
      "loss": 1.0914,
      "num_input_tokens_seen": 66691376,
      "step": 114950
    },
    {
      "epoch": 17.12168602919273,
      "grad_norm": 10.75,
      "learning_rate": 3.089102758913634e-06,
      "loss": 0.9592,
      "num_input_tokens_seen": 66694096,
      "step": 114955
    },
    {
      "epoch": 17.12243074173369,
      "grad_norm": 10.5625,
      "learning_rate": 3.087538287233241e-06,
      "loss": 1.25,
      "num_input_tokens_seen": 66697264,
      "step": 114960
    },
    {
      "epoch": 17.12317545427465,
      "grad_norm": 12.125,
      "learning_rate": 3.0859741857416193e-06,
      "loss": 1.0844,
      "num_input_tokens_seen": 66700080,
      "step": 114965
    },
    {
      "epoch": 17.12392016681561,
      "grad_norm": 14.8125,
      "learning_rate": 3.0844104544651893e-06,
      "loss": 1.3822,
      "num_input_tokens_seen": 66702736,
      "step": 114970
    },
    {
      "epoch": 17.12466487935657,
      "grad_norm": 6.5625,
      "learning_rate": 3.082847093430369e-06,
      "loss": 0.8243,
      "num_input_tokens_seen": 66705616,
      "step": 114975
    },
    {
      "epoch": 17.125409591897526,
      "grad_norm": 11.125,
      "learning_rate": 3.0812841026635705e-06,
      "loss": 1.2273,
      "num_input_tokens_seen": 66708688,
      "step": 114980
    },
    {
      "epoch": 17.126154304438487,
      "grad_norm": 9.9375,
      "learning_rate": 3.079721482191203e-06,
      "loss": 0.738,
      "num_input_tokens_seen": 66711536,
      "step": 114985
    },
    {
      "epoch": 17.126899016979447,
      "grad_norm": 11.375,
      "learning_rate": 3.0781592320396568e-06,
      "loss": 1.0459,
      "num_input_tokens_seen": 66714384,
      "step": 114990
    },
    {
      "epoch": 17.127643729520404,
      "grad_norm": 15.1875,
      "learning_rate": 3.076597352235333e-06,
      "loss": 1.2995,
      "num_input_tokens_seen": 66717584,
      "step": 114995
    },
    {
      "epoch": 17.128388442061365,
      "grad_norm": 6.09375,
      "learning_rate": 3.075035842804619e-06,
      "loss": 0.9194,
      "num_input_tokens_seen": 66720400,
      "step": 115000
    },
    {
      "epoch": 17.129133154602325,
      "grad_norm": 6.625,
      "learning_rate": 3.073474703773885e-06,
      "loss": 0.8914,
      "num_input_tokens_seen": 66723216,
      "step": 115005
    },
    {
      "epoch": 17.129877867143282,
      "grad_norm": 14.125,
      "learning_rate": 3.0719139351695125e-06,
      "loss": 1.1992,
      "num_input_tokens_seen": 66726160,
      "step": 115010
    },
    {
      "epoch": 17.130622579684243,
      "grad_norm": 10.375,
      "learning_rate": 3.070353537017867e-06,
      "loss": 1.0899,
      "num_input_tokens_seen": 66729328,
      "step": 115015
    },
    {
      "epoch": 17.1313672922252,
      "grad_norm": 5.375,
      "learning_rate": 3.0687935093453106e-06,
      "loss": 1.2087,
      "num_input_tokens_seen": 66732272,
      "step": 115020
    },
    {
      "epoch": 17.13211200476616,
      "grad_norm": 16.125,
      "learning_rate": 3.0672338521781975e-06,
      "loss": 1.5335,
      "num_input_tokens_seen": 66735024,
      "step": 115025
    },
    {
      "epoch": 17.13285671730712,
      "grad_norm": 10.8125,
      "learning_rate": 3.0656745655428783e-06,
      "loss": 1.1471,
      "num_input_tokens_seen": 66737840,
      "step": 115030
    },
    {
      "epoch": 17.133601429848078,
      "grad_norm": 5.625,
      "learning_rate": 3.0641156494656957e-06,
      "loss": 1.3426,
      "num_input_tokens_seen": 66740496,
      "step": 115035
    },
    {
      "epoch": 17.134346142389038,
      "grad_norm": 11.1875,
      "learning_rate": 3.062557103972985e-06,
      "loss": 1.2353,
      "num_input_tokens_seen": 66743472,
      "step": 115040
    },
    {
      "epoch": 17.13509085493,
      "grad_norm": 17.75,
      "learning_rate": 3.0609989290910775e-06,
      "loss": 1.1087,
      "num_input_tokens_seen": 66746480,
      "step": 115045
    },
    {
      "epoch": 17.135835567470956,
      "grad_norm": 7.5,
      "learning_rate": 3.059441124846288e-06,
      "loss": 1.0758,
      "num_input_tokens_seen": 66749424,
      "step": 115050
    },
    {
      "epoch": 17.136580280011916,
      "grad_norm": 9.5,
      "learning_rate": 3.0578836912649458e-06,
      "loss": 0.8816,
      "num_input_tokens_seen": 66752240,
      "step": 115055
    },
    {
      "epoch": 17.137324992552873,
      "grad_norm": 6.625,
      "learning_rate": 3.0563266283733517e-06,
      "loss": 0.9475,
      "num_input_tokens_seen": 66754768,
      "step": 115060
    },
    {
      "epoch": 17.138069705093834,
      "grad_norm": 9.0,
      "learning_rate": 3.054769936197824e-06,
      "loss": 0.8278,
      "num_input_tokens_seen": 66757872,
      "step": 115065
    },
    {
      "epoch": 17.138814417634794,
      "grad_norm": 5.78125,
      "learning_rate": 3.0532136147646496e-06,
      "loss": 1.2274,
      "num_input_tokens_seen": 66760944,
      "step": 115070
    },
    {
      "epoch": 17.13955913017575,
      "grad_norm": 13.375,
      "learning_rate": 3.05165766410013e-06,
      "loss": 1.5281,
      "num_input_tokens_seen": 66764144,
      "step": 115075
    },
    {
      "epoch": 17.140303842716712,
      "grad_norm": 5.53125,
      "learning_rate": 3.050102084230541e-06,
      "loss": 0.8985,
      "num_input_tokens_seen": 66766928,
      "step": 115080
    },
    {
      "epoch": 17.14104855525767,
      "grad_norm": 14.6875,
      "learning_rate": 3.0485468751821735e-06,
      "loss": 1.1276,
      "num_input_tokens_seen": 66769776,
      "step": 115085
    },
    {
      "epoch": 17.14179326779863,
      "grad_norm": 8.5625,
      "learning_rate": 3.046992036981294e-06,
      "loss": 0.8998,
      "num_input_tokens_seen": 66772592,
      "step": 115090
    },
    {
      "epoch": 17.14253798033959,
      "grad_norm": 16.0,
      "learning_rate": 3.0454375696541694e-06,
      "loss": 1.2476,
      "num_input_tokens_seen": 66775696,
      "step": 115095
    },
    {
      "epoch": 17.143282692880547,
      "grad_norm": 7.65625,
      "learning_rate": 3.0438834732270686e-06,
      "loss": 1.0417,
      "num_input_tokens_seen": 66778576,
      "step": 115100
    },
    {
      "epoch": 17.144027405421507,
      "grad_norm": 4.71875,
      "learning_rate": 3.0423297477262415e-06,
      "loss": 1.1215,
      "num_input_tokens_seen": 66781616,
      "step": 115105
    },
    {
      "epoch": 17.144772117962468,
      "grad_norm": 14.0,
      "learning_rate": 3.0407763931779354e-06,
      "loss": 1.1283,
      "num_input_tokens_seen": 66784496,
      "step": 115110
    },
    {
      "epoch": 17.145516830503425,
      "grad_norm": 9.375,
      "learning_rate": 3.039223409608391e-06,
      "loss": 1.1189,
      "num_input_tokens_seen": 66787408,
      "step": 115115
    },
    {
      "epoch": 17.146261543044385,
      "grad_norm": 7.21875,
      "learning_rate": 3.0376707970438513e-06,
      "loss": 1.1041,
      "num_input_tokens_seen": 66790224,
      "step": 115120
    },
    {
      "epoch": 17.147006255585342,
      "grad_norm": 15.625,
      "learning_rate": 3.036118555510539e-06,
      "loss": 0.7954,
      "num_input_tokens_seen": 66792944,
      "step": 115125
    },
    {
      "epoch": 17.147750968126303,
      "grad_norm": 8.875,
      "learning_rate": 3.0345666850346787e-06,
      "loss": 0.9532,
      "num_input_tokens_seen": 66796016,
      "step": 115130
    },
    {
      "epoch": 17.148495680667263,
      "grad_norm": 10.125,
      "learning_rate": 3.033015185642493e-06,
      "loss": 1.0964,
      "num_input_tokens_seen": 66798576,
      "step": 115135
    },
    {
      "epoch": 17.14924039320822,
      "grad_norm": 4.65625,
      "learning_rate": 3.0314640573601864e-06,
      "loss": 1.0454,
      "num_input_tokens_seen": 66801328,
      "step": 115140
    },
    {
      "epoch": 17.14998510574918,
      "grad_norm": 16.75,
      "learning_rate": 3.029913300213971e-06,
      "loss": 1.0173,
      "num_input_tokens_seen": 66804208,
      "step": 115145
    },
    {
      "epoch": 17.15072981829014,
      "grad_norm": 11.8125,
      "learning_rate": 3.0283629142300347e-06,
      "loss": 1.2699,
      "num_input_tokens_seen": 66806768,
      "step": 115150
    },
    {
      "epoch": 17.1514745308311,
      "grad_norm": 8.75,
      "learning_rate": 3.0268128994345807e-06,
      "loss": 0.8769,
      "num_input_tokens_seen": 66809648,
      "step": 115155
    },
    {
      "epoch": 17.15221924337206,
      "grad_norm": 13.1875,
      "learning_rate": 3.0252632558537913e-06,
      "loss": 1.0816,
      "num_input_tokens_seen": 66812656,
      "step": 115160
    },
    {
      "epoch": 17.152963955913016,
      "grad_norm": 9.5625,
      "learning_rate": 3.0237139835138402e-06,
      "loss": 0.9626,
      "num_input_tokens_seen": 66815440,
      "step": 115165
    },
    {
      "epoch": 17.153708668453977,
      "grad_norm": 8.875,
      "learning_rate": 3.0221650824409114e-06,
      "loss": 1.4329,
      "num_input_tokens_seen": 66818192,
      "step": 115170
    },
    {
      "epoch": 17.154453380994937,
      "grad_norm": 19.375,
      "learning_rate": 3.0206165526611654e-06,
      "loss": 1.1463,
      "num_input_tokens_seen": 66820912,
      "step": 115175
    },
    {
      "epoch": 17.155198093535894,
      "grad_norm": 10.9375,
      "learning_rate": 3.0190683942007637e-06,
      "loss": 0.834,
      "num_input_tokens_seen": 66824016,
      "step": 115180
    },
    {
      "epoch": 17.155942806076855,
      "grad_norm": 10.8125,
      "learning_rate": 3.017520607085858e-06,
      "loss": 1.3165,
      "num_input_tokens_seen": 66826896,
      "step": 115185
    },
    {
      "epoch": 17.156687518617815,
      "grad_norm": 7.03125,
      "learning_rate": 3.0159731913426027e-06,
      "loss": 0.9968,
      "num_input_tokens_seen": 66830064,
      "step": 115190
    },
    {
      "epoch": 17.157432231158772,
      "grad_norm": 9.0625,
      "learning_rate": 3.014426146997132e-06,
      "loss": 0.9642,
      "num_input_tokens_seen": 66833232,
      "step": 115195
    },
    {
      "epoch": 17.158176943699733,
      "grad_norm": 8.0,
      "learning_rate": 3.0128794740755916e-06,
      "loss": 0.9293,
      "num_input_tokens_seen": 66835952,
      "step": 115200
    },
    {
      "epoch": 17.15892165624069,
      "grad_norm": 16.375,
      "learning_rate": 3.0113331726041055e-06,
      "loss": 0.9634,
      "num_input_tokens_seen": 66838800,
      "step": 115205
    },
    {
      "epoch": 17.15966636878165,
      "grad_norm": 10.3125,
      "learning_rate": 3.0097872426087914e-06,
      "loss": 1.3124,
      "num_input_tokens_seen": 66841680,
      "step": 115210
    },
    {
      "epoch": 17.16041108132261,
      "grad_norm": 8.6875,
      "learning_rate": 3.0082416841157783e-06,
      "loss": 1.1577,
      "num_input_tokens_seen": 66844400,
      "step": 115215
    },
    {
      "epoch": 17.161155793863568,
      "grad_norm": 12.125,
      "learning_rate": 3.006696497151165e-06,
      "loss": 1.1717,
      "num_input_tokens_seen": 66847152,
      "step": 115220
    },
    {
      "epoch": 17.16190050640453,
      "grad_norm": 10.25,
      "learning_rate": 3.005151681741067e-06,
      "loss": 0.9442,
      "num_input_tokens_seen": 66850128,
      "step": 115225
    },
    {
      "epoch": 17.16264521894549,
      "grad_norm": 7.5625,
      "learning_rate": 3.0036072379115737e-06,
      "loss": 1.1041,
      "num_input_tokens_seen": 66853040,
      "step": 115230
    },
    {
      "epoch": 17.163389931486446,
      "grad_norm": 6.03125,
      "learning_rate": 3.0020631656887845e-06,
      "loss": 0.8815,
      "num_input_tokens_seen": 66856016,
      "step": 115235
    },
    {
      "epoch": 17.164134644027406,
      "grad_norm": 11.4375,
      "learning_rate": 3.000519465098772e-06,
      "loss": 0.8455,
      "num_input_tokens_seen": 66858896,
      "step": 115240
    },
    {
      "epoch": 17.164879356568363,
      "grad_norm": 8.9375,
      "learning_rate": 2.9989761361676306e-06,
      "loss": 1.1314,
      "num_input_tokens_seen": 66861616,
      "step": 115245
    },
    {
      "epoch": 17.165624069109324,
      "grad_norm": 6.1875,
      "learning_rate": 2.997433178921427e-06,
      "loss": 1.2476,
      "num_input_tokens_seen": 66864816,
      "step": 115250
    },
    {
      "epoch": 17.166368781650284,
      "grad_norm": 13.125,
      "learning_rate": 2.995890593386222e-06,
      "loss": 1.3047,
      "num_input_tokens_seen": 66867952,
      "step": 115255
    },
    {
      "epoch": 17.16711349419124,
      "grad_norm": 9.1875,
      "learning_rate": 2.9943483795880854e-06,
      "loss": 1.4301,
      "num_input_tokens_seen": 66870672,
      "step": 115260
    },
    {
      "epoch": 17.167858206732202,
      "grad_norm": 8.6875,
      "learning_rate": 2.992806537553064e-06,
      "loss": 1.252,
      "num_input_tokens_seen": 66873488,
      "step": 115265
    },
    {
      "epoch": 17.16860291927316,
      "grad_norm": 7.21875,
      "learning_rate": 2.9912650673072113e-06,
      "loss": 1.4343,
      "num_input_tokens_seen": 66876528,
      "step": 115270
    },
    {
      "epoch": 17.16934763181412,
      "grad_norm": 11.6875,
      "learning_rate": 2.989723968876565e-06,
      "loss": 1.2651,
      "num_input_tokens_seen": 66879248,
      "step": 115275
    },
    {
      "epoch": 17.17009234435508,
      "grad_norm": 9.5625,
      "learning_rate": 2.9881832422871654e-06,
      "loss": 1.1499,
      "num_input_tokens_seen": 66882288,
      "step": 115280
    },
    {
      "epoch": 17.170837056896037,
      "grad_norm": 13.4375,
      "learning_rate": 2.986642887565036e-06,
      "loss": 1.0988,
      "num_input_tokens_seen": 66885104,
      "step": 115285
    },
    {
      "epoch": 17.171581769436997,
      "grad_norm": 5.3125,
      "learning_rate": 2.9851029047362008e-06,
      "loss": 1.0764,
      "num_input_tokens_seen": 66888016,
      "step": 115290
    },
    {
      "epoch": 17.172326481977958,
      "grad_norm": 18.375,
      "learning_rate": 2.98356329382668e-06,
      "loss": 1.229,
      "num_input_tokens_seen": 66890800,
      "step": 115295
    },
    {
      "epoch": 17.173071194518915,
      "grad_norm": 7.78125,
      "learning_rate": 2.9820240548624814e-06,
      "loss": 1.3523,
      "num_input_tokens_seen": 66893840,
      "step": 115300
    },
    {
      "epoch": 17.173815907059875,
      "grad_norm": 11.3125,
      "learning_rate": 2.9804851878696054e-06,
      "loss": 1.1015,
      "num_input_tokens_seen": 66896528,
      "step": 115305
    },
    {
      "epoch": 17.174560619600832,
      "grad_norm": 5.71875,
      "learning_rate": 2.9789466928740515e-06,
      "loss": 1.2112,
      "num_input_tokens_seen": 66899728,
      "step": 115310
    },
    {
      "epoch": 17.175305332141793,
      "grad_norm": 18.25,
      "learning_rate": 2.9774085699018158e-06,
      "loss": 1.0289,
      "num_input_tokens_seen": 66902448,
      "step": 115315
    },
    {
      "epoch": 17.176050044682754,
      "grad_norm": 11.375,
      "learning_rate": 2.9758708189788736e-06,
      "loss": 1.2835,
      "num_input_tokens_seen": 66905488,
      "step": 115320
    },
    {
      "epoch": 17.17679475722371,
      "grad_norm": 11.0625,
      "learning_rate": 2.9743334401312133e-06,
      "loss": 0.9838,
      "num_input_tokens_seen": 66908336,
      "step": 115325
    },
    {
      "epoch": 17.17753946976467,
      "grad_norm": 7.5625,
      "learning_rate": 2.9727964333848056e-06,
      "loss": 1.2139,
      "num_input_tokens_seen": 66911312,
      "step": 115330
    },
    {
      "epoch": 17.17828418230563,
      "grad_norm": 8.75,
      "learning_rate": 2.9712597987656105e-06,
      "loss": 1.0116,
      "num_input_tokens_seen": 66914384,
      "step": 115335
    },
    {
      "epoch": 17.17902889484659,
      "grad_norm": 10.3125,
      "learning_rate": 2.9697235362995955e-06,
      "loss": 1.2702,
      "num_input_tokens_seen": 66917232,
      "step": 115340
    },
    {
      "epoch": 17.17977360738755,
      "grad_norm": 10.0625,
      "learning_rate": 2.9681876460127073e-06,
      "loss": 1.164,
      "num_input_tokens_seen": 66920016,
      "step": 115345
    },
    {
      "epoch": 17.180518319928506,
      "grad_norm": 6.53125,
      "learning_rate": 2.9666521279309023e-06,
      "loss": 1.0153,
      "num_input_tokens_seen": 66922928,
      "step": 115350
    },
    {
      "epoch": 17.181263032469467,
      "grad_norm": 7.90625,
      "learning_rate": 2.965116982080107e-06,
      "loss": 0.8991,
      "num_input_tokens_seen": 66925680,
      "step": 115355
    },
    {
      "epoch": 17.182007745010427,
      "grad_norm": 12.1875,
      "learning_rate": 2.9635822084862737e-06,
      "loss": 0.9115,
      "num_input_tokens_seen": 66928432,
      "step": 115360
    },
    {
      "epoch": 17.182752457551384,
      "grad_norm": 10.875,
      "learning_rate": 2.9620478071753223e-06,
      "loss": 1.0768,
      "num_input_tokens_seen": 66931280,
      "step": 115365
    },
    {
      "epoch": 17.183497170092345,
      "grad_norm": 8.4375,
      "learning_rate": 2.9605137781731713e-06,
      "loss": 0.9979,
      "num_input_tokens_seen": 66934128,
      "step": 115370
    },
    {
      "epoch": 17.184241882633305,
      "grad_norm": 9.1875,
      "learning_rate": 2.9589801215057445e-06,
      "loss": 1.4116,
      "num_input_tokens_seen": 66937264,
      "step": 115375
    },
    {
      "epoch": 17.184986595174262,
      "grad_norm": 12.6875,
      "learning_rate": 2.9574468371989378e-06,
      "loss": 1.1454,
      "num_input_tokens_seen": 66940560,
      "step": 115380
    },
    {
      "epoch": 17.185731307715223,
      "grad_norm": 11.25,
      "learning_rate": 2.955913925278672e-06,
      "loss": 0.8729,
      "num_input_tokens_seen": 66943280,
      "step": 115385
    },
    {
      "epoch": 17.18647602025618,
      "grad_norm": 9.6875,
      "learning_rate": 2.95438138577083e-06,
      "loss": 0.7807,
      "num_input_tokens_seen": 66946256,
      "step": 115390
    },
    {
      "epoch": 17.18722073279714,
      "grad_norm": 6.3125,
      "learning_rate": 2.952849218701312e-06,
      "loss": 1.1397,
      "num_input_tokens_seen": 66949136,
      "step": 115395
    },
    {
      "epoch": 17.1879654453381,
      "grad_norm": 9.5625,
      "learning_rate": 2.951317424095995e-06,
      "loss": 1.3122,
      "num_input_tokens_seen": 66951696,
      "step": 115400
    },
    {
      "epoch": 17.188710157879058,
      "grad_norm": 12.6875,
      "learning_rate": 2.9497860019807643e-06,
      "loss": 0.9881,
      "num_input_tokens_seen": 66954512,
      "step": 115405
    },
    {
      "epoch": 17.18945487042002,
      "grad_norm": 15.75,
      "learning_rate": 2.948254952381491e-06,
      "loss": 1.1717,
      "num_input_tokens_seen": 66957552,
      "step": 115410
    },
    {
      "epoch": 17.19019958296098,
      "grad_norm": 8.9375,
      "learning_rate": 2.946724275324031e-06,
      "loss": 1.3079,
      "num_input_tokens_seen": 66960656,
      "step": 115415
    },
    {
      "epoch": 17.190944295501936,
      "grad_norm": 11.0,
      "learning_rate": 2.9451939708342564e-06,
      "loss": 1.2756,
      "num_input_tokens_seen": 66963568,
      "step": 115420
    },
    {
      "epoch": 17.191689008042896,
      "grad_norm": 19.75,
      "learning_rate": 2.9436640389380073e-06,
      "loss": 1.0986,
      "num_input_tokens_seen": 66966416,
      "step": 115425
    },
    {
      "epoch": 17.192433720583853,
      "grad_norm": 16.75,
      "learning_rate": 2.9421344796611435e-06,
      "loss": 1.1576,
      "num_input_tokens_seen": 66969360,
      "step": 115430
    },
    {
      "epoch": 17.193178433124814,
      "grad_norm": 10.4375,
      "learning_rate": 2.9406052930295e-06,
      "loss": 1.0695,
      "num_input_tokens_seen": 66971984,
      "step": 115435
    },
    {
      "epoch": 17.193923145665774,
      "grad_norm": 14.0625,
      "learning_rate": 2.9390764790689085e-06,
      "loss": 0.8508,
      "num_input_tokens_seen": 66974832,
      "step": 115440
    },
    {
      "epoch": 17.19466785820673,
      "grad_norm": 8.9375,
      "learning_rate": 2.9375480378051987e-06,
      "loss": 0.9085,
      "num_input_tokens_seen": 66978192,
      "step": 115445
    },
    {
      "epoch": 17.195412570747692,
      "grad_norm": 6.75,
      "learning_rate": 2.9360199692641864e-06,
      "loss": 0.9692,
      "num_input_tokens_seen": 66981040,
      "step": 115450
    },
    {
      "epoch": 17.19615728328865,
      "grad_norm": 9.3125,
      "learning_rate": 2.9344922734716977e-06,
      "loss": 0.9087,
      "num_input_tokens_seen": 66983920,
      "step": 115455
    },
    {
      "epoch": 17.19690199582961,
      "grad_norm": 21.625,
      "learning_rate": 2.932964950453529e-06,
      "loss": 1.1384,
      "num_input_tokens_seen": 66986928,
      "step": 115460
    },
    {
      "epoch": 17.19764670837057,
      "grad_norm": 4.15625,
      "learning_rate": 2.9314380002354953e-06,
      "loss": 0.8959,
      "num_input_tokens_seen": 66989648,
      "step": 115465
    },
    {
      "epoch": 17.198391420911527,
      "grad_norm": 7.46875,
      "learning_rate": 2.9299114228433816e-06,
      "loss": 1.3092,
      "num_input_tokens_seen": 66992656,
      "step": 115470
    },
    {
      "epoch": 17.199136133452487,
      "grad_norm": 13.8125,
      "learning_rate": 2.9283852183029898e-06,
      "loss": 1.2982,
      "num_input_tokens_seen": 66995600,
      "step": 115475
    },
    {
      "epoch": 17.199880845993448,
      "grad_norm": 5.28125,
      "learning_rate": 2.9268593866400907e-06,
      "loss": 0.8924,
      "num_input_tokens_seen": 66998608,
      "step": 115480
    },
    {
      "epoch": 17.200625558534405,
      "grad_norm": 16.25,
      "learning_rate": 2.9253339278804748e-06,
      "loss": 1.0329,
      "num_input_tokens_seen": 67001424,
      "step": 115485
    },
    {
      "epoch": 17.201370271075366,
      "grad_norm": 21.625,
      "learning_rate": 2.923808842049905e-06,
      "loss": 1.2376,
      "num_input_tokens_seen": 67004208,
      "step": 115490
    },
    {
      "epoch": 17.202114983616323,
      "grad_norm": 12.3125,
      "learning_rate": 2.922284129174141e-06,
      "loss": 1.0704,
      "num_input_tokens_seen": 67006896,
      "step": 115495
    },
    {
      "epoch": 17.202859696157283,
      "grad_norm": 5.1875,
      "learning_rate": 2.920759789278957e-06,
      "loss": 1.3813,
      "num_input_tokens_seen": 67009680,
      "step": 115500
    },
    {
      "epoch": 17.203604408698244,
      "grad_norm": 11.0,
      "learning_rate": 2.919235822390093e-06,
      "loss": 1.2543,
      "num_input_tokens_seen": 67012336,
      "step": 115505
    },
    {
      "epoch": 17.2043491212392,
      "grad_norm": 8.3125,
      "learning_rate": 2.9177122285332982e-06,
      "loss": 1.1566,
      "num_input_tokens_seen": 67015504,
      "step": 115510
    },
    {
      "epoch": 17.20509383378016,
      "grad_norm": 10.5625,
      "learning_rate": 2.9161890077343074e-06,
      "loss": 0.9784,
      "num_input_tokens_seen": 67018256,
      "step": 115515
    },
    {
      "epoch": 17.20583854632112,
      "grad_norm": 9.5625,
      "learning_rate": 2.914666160018864e-06,
      "loss": 1.0032,
      "num_input_tokens_seen": 67021072,
      "step": 115520
    },
    {
      "epoch": 17.20658325886208,
      "grad_norm": 8.125,
      "learning_rate": 2.9131436854126894e-06,
      "loss": 1.576,
      "num_input_tokens_seen": 67024080,
      "step": 115525
    },
    {
      "epoch": 17.20732797140304,
      "grad_norm": 9.0625,
      "learning_rate": 2.9116215839414986e-06,
      "loss": 1.3167,
      "num_input_tokens_seen": 67027120,
      "step": 115530
    },
    {
      "epoch": 17.208072683943996,
      "grad_norm": 7.625,
      "learning_rate": 2.9100998556310153e-06,
      "loss": 1.1678,
      "num_input_tokens_seen": 67029936,
      "step": 115535
    },
    {
      "epoch": 17.208817396484957,
      "grad_norm": 9.625,
      "learning_rate": 2.9085785005069394e-06,
      "loss": 0.9861,
      "num_input_tokens_seen": 67032848,
      "step": 115540
    },
    {
      "epoch": 17.209562109025917,
      "grad_norm": 8.25,
      "learning_rate": 2.907057518594983e-06,
      "loss": 0.9164,
      "num_input_tokens_seen": 67035792,
      "step": 115545
    },
    {
      "epoch": 17.210306821566874,
      "grad_norm": 4.59375,
      "learning_rate": 2.9055369099208306e-06,
      "loss": 0.9201,
      "num_input_tokens_seen": 67038736,
      "step": 115550
    },
    {
      "epoch": 17.211051534107835,
      "grad_norm": 7.1875,
      "learning_rate": 2.904016674510179e-06,
      "loss": 1.0192,
      "num_input_tokens_seen": 67041840,
      "step": 115555
    },
    {
      "epoch": 17.211796246648795,
      "grad_norm": 25.75,
      "learning_rate": 2.9024968123887107e-06,
      "loss": 1.1263,
      "num_input_tokens_seen": 67044528,
      "step": 115560
    },
    {
      "epoch": 17.212540959189752,
      "grad_norm": 18.625,
      "learning_rate": 2.900977323582099e-06,
      "loss": 1.1319,
      "num_input_tokens_seen": 67047152,
      "step": 115565
    },
    {
      "epoch": 17.213285671730713,
      "grad_norm": 30.5,
      "learning_rate": 2.8994582081160155e-06,
      "loss": 1.0027,
      "num_input_tokens_seen": 67049776,
      "step": 115570
    },
    {
      "epoch": 17.21403038427167,
      "grad_norm": 8.75,
      "learning_rate": 2.897939466016117e-06,
      "loss": 0.9196,
      "num_input_tokens_seen": 67053040,
      "step": 115575
    },
    {
      "epoch": 17.21477509681263,
      "grad_norm": 10.875,
      "learning_rate": 2.8964210973080745e-06,
      "loss": 1.1319,
      "num_input_tokens_seen": 67056016,
      "step": 115580
    },
    {
      "epoch": 17.21551980935359,
      "grad_norm": 7.9375,
      "learning_rate": 2.8949031020175264e-06,
      "loss": 0.9904,
      "num_input_tokens_seen": 67058800,
      "step": 115585
    },
    {
      "epoch": 17.216264521894548,
      "grad_norm": 11.5625,
      "learning_rate": 2.89338548017013e-06,
      "loss": 1.0033,
      "num_input_tokens_seen": 67061872,
      "step": 115590
    },
    {
      "epoch": 17.21700923443551,
      "grad_norm": 15.25,
      "learning_rate": 2.8918682317915115e-06,
      "loss": 1.0781,
      "num_input_tokens_seen": 67064848,
      "step": 115595
    },
    {
      "epoch": 17.217753946976465,
      "grad_norm": 19.5,
      "learning_rate": 2.890351356907314e-06,
      "loss": 1.1421,
      "num_input_tokens_seen": 67067472,
      "step": 115600
    },
    {
      "epoch": 17.218498659517426,
      "grad_norm": 7.84375,
      "learning_rate": 2.8888348555431625e-06,
      "loss": 1.0596,
      "num_input_tokens_seen": 67070256,
      "step": 115605
    },
    {
      "epoch": 17.219243372058386,
      "grad_norm": 12.3125,
      "learning_rate": 2.887318727724664e-06,
      "loss": 1.2268,
      "num_input_tokens_seen": 67073136,
      "step": 115610
    },
    {
      "epoch": 17.219988084599343,
      "grad_norm": 17.375,
      "learning_rate": 2.88580297347745e-06,
      "loss": 0.8856,
      "num_input_tokens_seen": 67076240,
      "step": 115615
    },
    {
      "epoch": 17.220732797140304,
      "grad_norm": 22.75,
      "learning_rate": 2.884287592827112e-06,
      "loss": 1.1625,
      "num_input_tokens_seen": 67079216,
      "step": 115620
    },
    {
      "epoch": 17.221477509681264,
      "grad_norm": 9.1875,
      "learning_rate": 2.882772585799262e-06,
      "loss": 0.8472,
      "num_input_tokens_seen": 67082160,
      "step": 115625
    },
    {
      "epoch": 17.22222222222222,
      "grad_norm": 9.625,
      "learning_rate": 2.8812579524194916e-06,
      "loss": 1.2424,
      "num_input_tokens_seen": 67085264,
      "step": 115630
    },
    {
      "epoch": 17.222966934763182,
      "grad_norm": 8.625,
      "learning_rate": 2.879743692713388e-06,
      "loss": 0.8268,
      "num_input_tokens_seen": 67088048,
      "step": 115635
    },
    {
      "epoch": 17.22371164730414,
      "grad_norm": 9.5625,
      "learning_rate": 2.8782298067065256e-06,
      "loss": 0.992,
      "num_input_tokens_seen": 67090864,
      "step": 115640
    },
    {
      "epoch": 17.2244563598451,
      "grad_norm": 11.6875,
      "learning_rate": 2.8767162944244918e-06,
      "loss": 0.7447,
      "num_input_tokens_seen": 67093712,
      "step": 115645
    },
    {
      "epoch": 17.22520107238606,
      "grad_norm": 4.78125,
      "learning_rate": 2.875203155892853e-06,
      "loss": 1.0843,
      "num_input_tokens_seen": 67096400,
      "step": 115650
    },
    {
      "epoch": 17.225945784927017,
      "grad_norm": 5.71875,
      "learning_rate": 2.8736903911371652e-06,
      "loss": 0.9117,
      "num_input_tokens_seen": 67099280,
      "step": 115655
    },
    {
      "epoch": 17.226690497467978,
      "grad_norm": 10.3125,
      "learning_rate": 2.8721780001829956e-06,
      "loss": 0.9714,
      "num_input_tokens_seen": 67102192,
      "step": 115660
    },
    {
      "epoch": 17.227435210008938,
      "grad_norm": 7.6875,
      "learning_rate": 2.870665983055881e-06,
      "loss": 1.1053,
      "num_input_tokens_seen": 67104880,
      "step": 115665
    },
    {
      "epoch": 17.228179922549895,
      "grad_norm": 7.875,
      "learning_rate": 2.8691543397813824e-06,
      "loss": 0.8596,
      "num_input_tokens_seen": 67107600,
      "step": 115670
    },
    {
      "epoch": 17.228924635090856,
      "grad_norm": 7.84375,
      "learning_rate": 2.8676430703850206e-06,
      "loss": 1.1229,
      "num_input_tokens_seen": 67110512,
      "step": 115675
    },
    {
      "epoch": 17.229669347631813,
      "grad_norm": 9.4375,
      "learning_rate": 2.8661321748923416e-06,
      "loss": 1.1064,
      "num_input_tokens_seen": 67113232,
      "step": 115680
    },
    {
      "epoch": 17.230414060172773,
      "grad_norm": 6.1875,
      "learning_rate": 2.8646216533288556e-06,
      "loss": 0.7898,
      "num_input_tokens_seen": 67115664,
      "step": 115685
    },
    {
      "epoch": 17.231158772713734,
      "grad_norm": 9.0,
      "learning_rate": 2.863111505720098e-06,
      "loss": 1.1581,
      "num_input_tokens_seen": 67118320,
      "step": 115690
    },
    {
      "epoch": 17.23190348525469,
      "grad_norm": 5.28125,
      "learning_rate": 2.8616017320915704e-06,
      "loss": 1.2757,
      "num_input_tokens_seen": 67121232,
      "step": 115695
    },
    {
      "epoch": 17.23264819779565,
      "grad_norm": 8.125,
      "learning_rate": 2.8600923324687807e-06,
      "loss": 0.8581,
      "num_input_tokens_seen": 67124176,
      "step": 115700
    },
    {
      "epoch": 17.23339291033661,
      "grad_norm": 7.25,
      "learning_rate": 2.85858330687723e-06,
      "loss": 0.9856,
      "num_input_tokens_seen": 67126864,
      "step": 115705
    },
    {
      "epoch": 17.23413762287757,
      "grad_norm": 15.5,
      "learning_rate": 2.8570746553424065e-06,
      "loss": 1.2449,
      "num_input_tokens_seen": 67129520,
      "step": 115710
    },
    {
      "epoch": 17.23488233541853,
      "grad_norm": 9.5,
      "learning_rate": 2.8555663778898066e-06,
      "loss": 0.9106,
      "num_input_tokens_seen": 67132624,
      "step": 115715
    },
    {
      "epoch": 17.235627047959486,
      "grad_norm": 13.3125,
      "learning_rate": 2.854058474544899e-06,
      "loss": 0.9144,
      "num_input_tokens_seen": 67135952,
      "step": 115720
    },
    {
      "epoch": 17.236371760500447,
      "grad_norm": 19.125,
      "learning_rate": 2.852550945333174e-06,
      "loss": 1.2817,
      "num_input_tokens_seen": 67139376,
      "step": 115725
    },
    {
      "epoch": 17.237116473041407,
      "grad_norm": 7.5625,
      "learning_rate": 2.851043790280089e-06,
      "loss": 1.2553,
      "num_input_tokens_seen": 67142608,
      "step": 115730
    },
    {
      "epoch": 17.237861185582364,
      "grad_norm": 7.84375,
      "learning_rate": 2.849537009411102e-06,
      "loss": 0.7334,
      "num_input_tokens_seen": 67145392,
      "step": 115735
    },
    {
      "epoch": 17.238605898123325,
      "grad_norm": 10.125,
      "learning_rate": 2.8480306027516807e-06,
      "loss": 1.0138,
      "num_input_tokens_seen": 67148112,
      "step": 115740
    },
    {
      "epoch": 17.239350610664285,
      "grad_norm": 12.8125,
      "learning_rate": 2.8465245703272607e-06,
      "loss": 1.0814,
      "num_input_tokens_seen": 67150672,
      "step": 115745
    },
    {
      "epoch": 17.240095323205242,
      "grad_norm": 10.625,
      "learning_rate": 2.8450189121632998e-06,
      "loss": 0.9477,
      "num_input_tokens_seen": 67153296,
      "step": 115750
    },
    {
      "epoch": 17.240840035746203,
      "grad_norm": 6.03125,
      "learning_rate": 2.8435136282852217e-06,
      "loss": 1.053,
      "num_input_tokens_seen": 67156336,
      "step": 115755
    },
    {
      "epoch": 17.24158474828716,
      "grad_norm": 10.3125,
      "learning_rate": 2.842008718718467e-06,
      "loss": 0.972,
      "num_input_tokens_seen": 67159344,
      "step": 115760
    },
    {
      "epoch": 17.24232946082812,
      "grad_norm": 10.9375,
      "learning_rate": 2.840504183488457e-06,
      "loss": 1.2828,
      "num_input_tokens_seen": 67162448,
      "step": 115765
    },
    {
      "epoch": 17.24307417336908,
      "grad_norm": 13.6875,
      "learning_rate": 2.8390000226206025e-06,
      "loss": 0.8157,
      "num_input_tokens_seen": 67165104,
      "step": 115770
    },
    {
      "epoch": 17.243818885910038,
      "grad_norm": 8.875,
      "learning_rate": 2.837496236140322e-06,
      "loss": 1.0143,
      "num_input_tokens_seen": 67167920,
      "step": 115775
    },
    {
      "epoch": 17.244563598451,
      "grad_norm": 25.5,
      "learning_rate": 2.835992824073011e-06,
      "loss": 1.0277,
      "num_input_tokens_seen": 67170544,
      "step": 115780
    },
    {
      "epoch": 17.245308310991955,
      "grad_norm": 9.125,
      "learning_rate": 2.8344897864440805e-06,
      "loss": 1.2852,
      "num_input_tokens_seen": 67173264,
      "step": 115785
    },
    {
      "epoch": 17.246053023532916,
      "grad_norm": 8.375,
      "learning_rate": 2.832987123278913e-06,
      "loss": 0.8839,
      "num_input_tokens_seen": 67176208,
      "step": 115790
    },
    {
      "epoch": 17.246797736073876,
      "grad_norm": 7.9375,
      "learning_rate": 2.8314848346029017e-06,
      "loss": 0.9543,
      "num_input_tokens_seen": 67178960,
      "step": 115795
    },
    {
      "epoch": 17.247542448614833,
      "grad_norm": 7.71875,
      "learning_rate": 2.829982920441421e-06,
      "loss": 1.0462,
      "num_input_tokens_seen": 67181712,
      "step": 115800
    },
    {
      "epoch": 17.248287161155794,
      "grad_norm": 9.875,
      "learning_rate": 2.8284813808198473e-06,
      "loss": 1.1281,
      "num_input_tokens_seen": 67184528,
      "step": 115805
    },
    {
      "epoch": 17.249031873696755,
      "grad_norm": 14.875,
      "learning_rate": 2.82698021576355e-06,
      "loss": 1.2262,
      "num_input_tokens_seen": 67187312,
      "step": 115810
    },
    {
      "epoch": 17.24977658623771,
      "grad_norm": 13.25,
      "learning_rate": 2.825479425297878e-06,
      "loss": 0.9994,
      "num_input_tokens_seen": 67190288,
      "step": 115815
    },
    {
      "epoch": 17.250521298778672,
      "grad_norm": 19.625,
      "learning_rate": 2.823979009448202e-06,
      "loss": 0.9927,
      "num_input_tokens_seen": 67192880,
      "step": 115820
    },
    {
      "epoch": 17.25126601131963,
      "grad_norm": 8.8125,
      "learning_rate": 2.8224789682398556e-06,
      "loss": 1.0547,
      "num_input_tokens_seen": 67195792,
      "step": 115825
    },
    {
      "epoch": 17.25201072386059,
      "grad_norm": 8.0,
      "learning_rate": 2.8209793016981927e-06,
      "loss": 1.3081,
      "num_input_tokens_seen": 67198640,
      "step": 115830
    },
    {
      "epoch": 17.25275543640155,
      "grad_norm": 9.375,
      "learning_rate": 2.8194800098485407e-06,
      "loss": 0.9908,
      "num_input_tokens_seen": 67201584,
      "step": 115835
    },
    {
      "epoch": 17.253500148942507,
      "grad_norm": 7.59375,
      "learning_rate": 2.817981092716232e-06,
      "loss": 0.8382,
      "num_input_tokens_seen": 67204304,
      "step": 115840
    },
    {
      "epoch": 17.254244861483468,
      "grad_norm": 14.1875,
      "learning_rate": 2.8164825503265825e-06,
      "loss": 1.3846,
      "num_input_tokens_seen": 67207248,
      "step": 115845
    },
    {
      "epoch": 17.254989574024428,
      "grad_norm": 11.875,
      "learning_rate": 2.8149843827049186e-06,
      "loss": 0.8768,
      "num_input_tokens_seen": 67209872,
      "step": 115850
    },
    {
      "epoch": 17.255734286565385,
      "grad_norm": 12.4375,
      "learning_rate": 2.813486589876549e-06,
      "loss": 1.0524,
      "num_input_tokens_seen": 67212656,
      "step": 115855
    },
    {
      "epoch": 17.256478999106346,
      "grad_norm": 11.1875,
      "learning_rate": 2.8119891718667664e-06,
      "loss": 1.1832,
      "num_input_tokens_seen": 67215280,
      "step": 115860
    },
    {
      "epoch": 17.257223711647303,
      "grad_norm": 8.375,
      "learning_rate": 2.8104921287008785e-06,
      "loss": 1.1078,
      "num_input_tokens_seen": 67218320,
      "step": 115865
    },
    {
      "epoch": 17.257968424188263,
      "grad_norm": 12.6875,
      "learning_rate": 2.8089954604041734e-06,
      "loss": 0.8619,
      "num_input_tokens_seen": 67221104,
      "step": 115870
    },
    {
      "epoch": 17.258713136729224,
      "grad_norm": 7.65625,
      "learning_rate": 2.807499167001937e-06,
      "loss": 0.9923,
      "num_input_tokens_seen": 67223824,
      "step": 115875
    },
    {
      "epoch": 17.25945784927018,
      "grad_norm": 12.625,
      "learning_rate": 2.8060032485194453e-06,
      "loss": 1.4424,
      "num_input_tokens_seen": 67226576,
      "step": 115880
    },
    {
      "epoch": 17.26020256181114,
      "grad_norm": 10.25,
      "learning_rate": 2.8045077049819733e-06,
      "loss": 1.18,
      "num_input_tokens_seen": 67229168,
      "step": 115885
    },
    {
      "epoch": 17.2609472743521,
      "grad_norm": 11.5,
      "learning_rate": 2.8030125364147868e-06,
      "loss": 1.1467,
      "num_input_tokens_seen": 67232112,
      "step": 115890
    },
    {
      "epoch": 17.26169198689306,
      "grad_norm": 5.125,
      "learning_rate": 2.8015177428431433e-06,
      "loss": 1.0644,
      "num_input_tokens_seen": 67235120,
      "step": 115895
    },
    {
      "epoch": 17.26243669943402,
      "grad_norm": 7.65625,
      "learning_rate": 2.8000233242922973e-06,
      "loss": 0.9075,
      "num_input_tokens_seen": 67238160,
      "step": 115900
    },
    {
      "epoch": 17.263181411974976,
      "grad_norm": 17.0,
      "learning_rate": 2.7985292807874873e-06,
      "loss": 1.4143,
      "num_input_tokens_seen": 67240688,
      "step": 115905
    },
    {
      "epoch": 17.263926124515937,
      "grad_norm": 6.875,
      "learning_rate": 2.797035612353968e-06,
      "loss": 1.0172,
      "num_input_tokens_seen": 67243248,
      "step": 115910
    },
    {
      "epoch": 17.264670837056897,
      "grad_norm": 4.53125,
      "learning_rate": 2.7955423190169585e-06,
      "loss": 0.7892,
      "num_input_tokens_seen": 67246256,
      "step": 115915
    },
    {
      "epoch": 17.265415549597854,
      "grad_norm": 11.25,
      "learning_rate": 2.794049400801699e-06,
      "loss": 0.9257,
      "num_input_tokens_seen": 67249072,
      "step": 115920
    },
    {
      "epoch": 17.266160262138815,
      "grad_norm": 13.0625,
      "learning_rate": 2.792556857733403e-06,
      "loss": 1.3471,
      "num_input_tokens_seen": 67251728,
      "step": 115925
    },
    {
      "epoch": 17.266904974679775,
      "grad_norm": 15.0,
      "learning_rate": 2.7910646898372916e-06,
      "loss": 1.3056,
      "num_input_tokens_seen": 67254480,
      "step": 115930
    },
    {
      "epoch": 17.267649687220732,
      "grad_norm": 11.8125,
      "learning_rate": 2.7895728971385706e-06,
      "loss": 1.1517,
      "num_input_tokens_seen": 67257456,
      "step": 115935
    },
    {
      "epoch": 17.268394399761693,
      "grad_norm": 19.125,
      "learning_rate": 2.7880814796624355e-06,
      "loss": 1.1279,
      "num_input_tokens_seen": 67260304,
      "step": 115940
    },
    {
      "epoch": 17.26913911230265,
      "grad_norm": 7.90625,
      "learning_rate": 2.7865904374340947e-06,
      "loss": 1.1966,
      "num_input_tokens_seen": 67263376,
      "step": 115945
    },
    {
      "epoch": 17.26988382484361,
      "grad_norm": 11.6875,
      "learning_rate": 2.7850997704787244e-06,
      "loss": 0.9735,
      "num_input_tokens_seen": 67266256,
      "step": 115950
    },
    {
      "epoch": 17.27062853738457,
      "grad_norm": 9.75,
      "learning_rate": 2.783609478821525e-06,
      "loss": 0.9463,
      "num_input_tokens_seen": 67269072,
      "step": 115955
    },
    {
      "epoch": 17.271373249925528,
      "grad_norm": 12.25,
      "learning_rate": 2.782119562487662e-06,
      "loss": 1.2272,
      "num_input_tokens_seen": 67271920,
      "step": 115960
    },
    {
      "epoch": 17.27211796246649,
      "grad_norm": 14.5625,
      "learning_rate": 2.7806300215023063e-06,
      "loss": 1.0742,
      "num_input_tokens_seen": 67275088,
      "step": 115965
    },
    {
      "epoch": 17.272862675007445,
      "grad_norm": 11.3125,
      "learning_rate": 2.7791408558906245e-06,
      "loss": 1.2618,
      "num_input_tokens_seen": 67278064,
      "step": 115970
    },
    {
      "epoch": 17.273607387548406,
      "grad_norm": 9.75,
      "learning_rate": 2.777652065677766e-06,
      "loss": 1.197,
      "num_input_tokens_seen": 67280816,
      "step": 115975
    },
    {
      "epoch": 17.274352100089367,
      "grad_norm": 9.0625,
      "learning_rate": 2.7761636508888995e-06,
      "loss": 1.0382,
      "num_input_tokens_seen": 67283792,
      "step": 115980
    },
    {
      "epoch": 17.275096812630323,
      "grad_norm": 10.125,
      "learning_rate": 2.774675611549152e-06,
      "loss": 0.9514,
      "num_input_tokens_seen": 67286672,
      "step": 115985
    },
    {
      "epoch": 17.275841525171284,
      "grad_norm": 7.0,
      "learning_rate": 2.773187947683678e-06,
      "loss": 1.0885,
      "num_input_tokens_seen": 67289552,
      "step": 115990
    },
    {
      "epoch": 17.276586237712245,
      "grad_norm": 12.125,
      "learning_rate": 2.7717006593175997e-06,
      "loss": 1.1,
      "num_input_tokens_seen": 67292336,
      "step": 115995
    },
    {
      "epoch": 17.2773309502532,
      "grad_norm": 7.5625,
      "learning_rate": 2.7702137464760497e-06,
      "loss": 1.3836,
      "num_input_tokens_seen": 67295088,
      "step": 116000
    },
    {
      "epoch": 17.278075662794162,
      "grad_norm": 8.75,
      "learning_rate": 2.768727209184141e-06,
      "loss": 1.1852,
      "num_input_tokens_seen": 67298096,
      "step": 116005
    },
    {
      "epoch": 17.27882037533512,
      "grad_norm": 10.1875,
      "learning_rate": 2.767241047466998e-06,
      "loss": 0.9297,
      "num_input_tokens_seen": 67301200,
      "step": 116010
    },
    {
      "epoch": 17.27956508787608,
      "grad_norm": 11.3125,
      "learning_rate": 2.765755261349717e-06,
      "loss": 0.8624,
      "num_input_tokens_seen": 67304432,
      "step": 116015
    },
    {
      "epoch": 17.28030980041704,
      "grad_norm": 14.5,
      "learning_rate": 2.764269850857401e-06,
      "loss": 1.2478,
      "num_input_tokens_seen": 67307312,
      "step": 116020
    },
    {
      "epoch": 17.281054512957997,
      "grad_norm": 5.75,
      "learning_rate": 2.7627848160151513e-06,
      "loss": 1.2753,
      "num_input_tokens_seen": 67310416,
      "step": 116025
    },
    {
      "epoch": 17.281799225498958,
      "grad_norm": 15.875,
      "learning_rate": 2.7613001568480514e-06,
      "loss": 0.9212,
      "num_input_tokens_seen": 67313328,
      "step": 116030
    },
    {
      "epoch": 17.282543938039918,
      "grad_norm": 8.9375,
      "learning_rate": 2.759815873381183e-06,
      "loss": 0.9744,
      "num_input_tokens_seen": 67316304,
      "step": 116035
    },
    {
      "epoch": 17.283288650580875,
      "grad_norm": 22.125,
      "learning_rate": 2.7583319656396155e-06,
      "loss": 1.1322,
      "num_input_tokens_seen": 67318928,
      "step": 116040
    },
    {
      "epoch": 17.284033363121836,
      "grad_norm": 13.875,
      "learning_rate": 2.756848433648429e-06,
      "loss": 1.1336,
      "num_input_tokens_seen": 67321968,
      "step": 116045
    },
    {
      "epoch": 17.284778075662793,
      "grad_norm": 8.75,
      "learning_rate": 2.755365277432681e-06,
      "loss": 1.2342,
      "num_input_tokens_seen": 67324976,
      "step": 116050
    },
    {
      "epoch": 17.285522788203753,
      "grad_norm": 4.78125,
      "learning_rate": 2.753882497017424e-06,
      "loss": 0.7525,
      "num_input_tokens_seen": 67327856,
      "step": 116055
    },
    {
      "epoch": 17.286267500744714,
      "grad_norm": 4.4375,
      "learning_rate": 2.7524000924277178e-06,
      "loss": 1.0733,
      "num_input_tokens_seen": 67330640,
      "step": 116060
    },
    {
      "epoch": 17.28701221328567,
      "grad_norm": 8.1875,
      "learning_rate": 2.7509180636885927e-06,
      "loss": 1.2924,
      "num_input_tokens_seen": 67333360,
      "step": 116065
    },
    {
      "epoch": 17.28775692582663,
      "grad_norm": 6.71875,
      "learning_rate": 2.7494364108251016e-06,
      "loss": 1.0435,
      "num_input_tokens_seen": 67336368,
      "step": 116070
    },
    {
      "epoch": 17.288501638367592,
      "grad_norm": 14.0,
      "learning_rate": 2.747955133862262e-06,
      "loss": 1.1109,
      "num_input_tokens_seen": 67338992,
      "step": 116075
    },
    {
      "epoch": 17.28924635090855,
      "grad_norm": 12.4375,
      "learning_rate": 2.746474232825111e-06,
      "loss": 1.4621,
      "num_input_tokens_seen": 67341680,
      "step": 116080
    },
    {
      "epoch": 17.28999106344951,
      "grad_norm": 9.6875,
      "learning_rate": 2.744993707738655e-06,
      "loss": 1.0959,
      "num_input_tokens_seen": 67344688,
      "step": 116085
    },
    {
      "epoch": 17.290735775990466,
      "grad_norm": 9.0625,
      "learning_rate": 2.7435135586279165e-06,
      "loss": 1.1189,
      "num_input_tokens_seen": 67347600,
      "step": 116090
    },
    {
      "epoch": 17.291480488531427,
      "grad_norm": 9.0625,
      "learning_rate": 2.7420337855178944e-06,
      "loss": 0.8724,
      "num_input_tokens_seen": 67350544,
      "step": 116095
    },
    {
      "epoch": 17.292225201072387,
      "grad_norm": 17.875,
      "learning_rate": 2.7405543884335887e-06,
      "loss": 1.182,
      "num_input_tokens_seen": 67353680,
      "step": 116100
    },
    {
      "epoch": 17.292969913613344,
      "grad_norm": 6.125,
      "learning_rate": 2.739075367399996e-06,
      "loss": 1.0805,
      "num_input_tokens_seen": 67356432,
      "step": 116105
    },
    {
      "epoch": 17.293714626154305,
      "grad_norm": 15.6875,
      "learning_rate": 2.7375967224420928e-06,
      "loss": 1.0676,
      "num_input_tokens_seen": 67359120,
      "step": 116110
    },
    {
      "epoch": 17.294459338695262,
      "grad_norm": 8.5,
      "learning_rate": 2.736118453584871e-06,
      "loss": 1.0766,
      "num_input_tokens_seen": 67361776,
      "step": 116115
    },
    {
      "epoch": 17.295204051236222,
      "grad_norm": 13.375,
      "learning_rate": 2.7346405608532965e-06,
      "loss": 1.1935,
      "num_input_tokens_seen": 67364560,
      "step": 116120
    },
    {
      "epoch": 17.295948763777183,
      "grad_norm": 7.59375,
      "learning_rate": 2.7331630442723466e-06,
      "loss": 0.9219,
      "num_input_tokens_seen": 67367664,
      "step": 116125
    },
    {
      "epoch": 17.29669347631814,
      "grad_norm": 7.53125,
      "learning_rate": 2.7316859038669736e-06,
      "loss": 1.1594,
      "num_input_tokens_seen": 67370576,
      "step": 116130
    },
    {
      "epoch": 17.2974381888591,
      "grad_norm": 10.0625,
      "learning_rate": 2.7302091396621294e-06,
      "loss": 1.2053,
      "num_input_tokens_seen": 67373488,
      "step": 116135
    },
    {
      "epoch": 17.29818290140006,
      "grad_norm": 10.5625,
      "learning_rate": 2.7287327516827748e-06,
      "loss": 1.4361,
      "num_input_tokens_seen": 67376272,
      "step": 116140
    },
    {
      "epoch": 17.298927613941018,
      "grad_norm": 9.625,
      "learning_rate": 2.7272567399538375e-06,
      "loss": 1.1491,
      "num_input_tokens_seen": 67379504,
      "step": 116145
    },
    {
      "epoch": 17.29967232648198,
      "grad_norm": 4.15625,
      "learning_rate": 2.725781104500269e-06,
      "loss": 0.8438,
      "num_input_tokens_seen": 67382640,
      "step": 116150
    },
    {
      "epoch": 17.300417039022935,
      "grad_norm": 5.875,
      "learning_rate": 2.7243058453469835e-06,
      "loss": 0.7047,
      "num_input_tokens_seen": 67385200,
      "step": 116155
    },
    {
      "epoch": 17.301161751563896,
      "grad_norm": 8.625,
      "learning_rate": 2.722830962518913e-06,
      "loss": 0.9109,
      "num_input_tokens_seen": 67388208,
      "step": 116160
    },
    {
      "epoch": 17.301906464104857,
      "grad_norm": 18.5,
      "learning_rate": 2.7213564560409743e-06,
      "loss": 1.2701,
      "num_input_tokens_seen": 67390768,
      "step": 116165
    },
    {
      "epoch": 17.302651176645814,
      "grad_norm": 7.34375,
      "learning_rate": 2.7198823259380777e-06,
      "loss": 1.0769,
      "num_input_tokens_seen": 67393808,
      "step": 116170
    },
    {
      "epoch": 17.303395889186774,
      "grad_norm": 9.0625,
      "learning_rate": 2.7184085722351205e-06,
      "loss": 1.1799,
      "num_input_tokens_seen": 67396624,
      "step": 116175
    },
    {
      "epoch": 17.304140601727735,
      "grad_norm": 14.9375,
      "learning_rate": 2.7169351949570017e-06,
      "loss": 0.9432,
      "num_input_tokens_seen": 67399696,
      "step": 116180
    },
    {
      "epoch": 17.30488531426869,
      "grad_norm": 15.875,
      "learning_rate": 2.7154621941286206e-06,
      "loss": 1.0509,
      "num_input_tokens_seen": 67402736,
      "step": 116185
    },
    {
      "epoch": 17.305630026809652,
      "grad_norm": 22.125,
      "learning_rate": 2.7139895697748496e-06,
      "loss": 1.227,
      "num_input_tokens_seen": 67405744,
      "step": 116190
    },
    {
      "epoch": 17.30637473935061,
      "grad_norm": 10.75,
      "learning_rate": 2.7125173219205824e-06,
      "loss": 1.133,
      "num_input_tokens_seen": 67408624,
      "step": 116195
    },
    {
      "epoch": 17.30711945189157,
      "grad_norm": 10.9375,
      "learning_rate": 2.711045450590677e-06,
      "loss": 1.2628,
      "num_input_tokens_seen": 67411312,
      "step": 116200
    },
    {
      "epoch": 17.30786416443253,
      "grad_norm": 11.8125,
      "learning_rate": 2.7095739558100074e-06,
      "loss": 0.7761,
      "num_input_tokens_seen": 67414352,
      "step": 116205
    },
    {
      "epoch": 17.308608876973487,
      "grad_norm": 8.3125,
      "learning_rate": 2.708102837603435e-06,
      "loss": 1.2292,
      "num_input_tokens_seen": 67417008,
      "step": 116210
    },
    {
      "epoch": 17.309353589514448,
      "grad_norm": 11.1875,
      "learning_rate": 2.706632095995801e-06,
      "loss": 0.985,
      "num_input_tokens_seen": 67420176,
      "step": 116215
    },
    {
      "epoch": 17.31009830205541,
      "grad_norm": 11.625,
      "learning_rate": 2.7051617310119653e-06,
      "loss": 0.97,
      "num_input_tokens_seen": 67422992,
      "step": 116220
    },
    {
      "epoch": 17.310843014596365,
      "grad_norm": 6.84375,
      "learning_rate": 2.7036917426767615e-06,
      "loss": 1.3275,
      "num_input_tokens_seen": 67425648,
      "step": 116225
    },
    {
      "epoch": 17.311587727137326,
      "grad_norm": 12.9375,
      "learning_rate": 2.702222131015028e-06,
      "loss": 1.1061,
      "num_input_tokens_seen": 67428624,
      "step": 116230
    },
    {
      "epoch": 17.312332439678283,
      "grad_norm": 11.25,
      "learning_rate": 2.700752896051581e-06,
      "loss": 1.2019,
      "num_input_tokens_seen": 67431632,
      "step": 116235
    },
    {
      "epoch": 17.313077152219243,
      "grad_norm": 14.5,
      "learning_rate": 2.699284037811256e-06,
      "loss": 1.1171,
      "num_input_tokens_seen": 67434608,
      "step": 116240
    },
    {
      "epoch": 17.313821864760204,
      "grad_norm": 4.78125,
      "learning_rate": 2.6978155563188583e-06,
      "loss": 0.7664,
      "num_input_tokens_seen": 67437936,
      "step": 116245
    },
    {
      "epoch": 17.31456657730116,
      "grad_norm": 7.1875,
      "learning_rate": 2.6963474515992044e-06,
      "loss": 0.9595,
      "num_input_tokens_seen": 67441136,
      "step": 116250
    },
    {
      "epoch": 17.31531128984212,
      "grad_norm": 14.5625,
      "learning_rate": 2.6948797236770907e-06,
      "loss": 0.8783,
      "num_input_tokens_seen": 67443888,
      "step": 116255
    },
    {
      "epoch": 17.316056002383082,
      "grad_norm": 13.3125,
      "learning_rate": 2.6934123725773088e-06,
      "loss": 0.8295,
      "num_input_tokens_seen": 67446576,
      "step": 116260
    },
    {
      "epoch": 17.31680071492404,
      "grad_norm": 9.0,
      "learning_rate": 2.6919453983246577e-06,
      "loss": 0.9227,
      "num_input_tokens_seen": 67449552,
      "step": 116265
    },
    {
      "epoch": 17.317545427465,
      "grad_norm": 7.5625,
      "learning_rate": 2.690478800943913e-06,
      "loss": 0.8703,
      "num_input_tokens_seen": 67452368,
      "step": 116270
    },
    {
      "epoch": 17.318290140005956,
      "grad_norm": 9.1875,
      "learning_rate": 2.68901258045986e-06,
      "loss": 1.1693,
      "num_input_tokens_seen": 67455216,
      "step": 116275
    },
    {
      "epoch": 17.319034852546917,
      "grad_norm": 8.6875,
      "learning_rate": 2.6875467368972563e-06,
      "loss": 1.0666,
      "num_input_tokens_seen": 67457904,
      "step": 116280
    },
    {
      "epoch": 17.319779565087877,
      "grad_norm": 16.875,
      "learning_rate": 2.6860812702808795e-06,
      "loss": 1.0748,
      "num_input_tokens_seen": 67460976,
      "step": 116285
    },
    {
      "epoch": 17.320524277628834,
      "grad_norm": 6.3125,
      "learning_rate": 2.6846161806354826e-06,
      "loss": 0.8696,
      "num_input_tokens_seen": 67463888,
      "step": 116290
    },
    {
      "epoch": 17.321268990169795,
      "grad_norm": 11.75,
      "learning_rate": 2.6831514679858115e-06,
      "loss": 1.1626,
      "num_input_tokens_seen": 67466864,
      "step": 116295
    },
    {
      "epoch": 17.322013702710752,
      "grad_norm": 3.4375,
      "learning_rate": 2.6816871323566165e-06,
      "loss": 0.9513,
      "num_input_tokens_seen": 67469936,
      "step": 116300
    },
    {
      "epoch": 17.322758415251712,
      "grad_norm": 6.8125,
      "learning_rate": 2.6802231737726307e-06,
      "loss": 0.7347,
      "num_input_tokens_seen": 67472752,
      "step": 116305
    },
    {
      "epoch": 17.323503127792673,
      "grad_norm": 14.4375,
      "learning_rate": 2.6787595922585924e-06,
      "loss": 1.1064,
      "num_input_tokens_seen": 67475856,
      "step": 116310
    },
    {
      "epoch": 17.32424784033363,
      "grad_norm": 6.03125,
      "learning_rate": 2.6772963878392177e-06,
      "loss": 0.9889,
      "num_input_tokens_seen": 67478736,
      "step": 116315
    },
    {
      "epoch": 17.32499255287459,
      "grad_norm": 9.1875,
      "learning_rate": 2.67583356053924e-06,
      "loss": 0.9968,
      "num_input_tokens_seen": 67481584,
      "step": 116320
    },
    {
      "epoch": 17.32573726541555,
      "grad_norm": 10.8125,
      "learning_rate": 2.6743711103833614e-06,
      "loss": 1.3118,
      "num_input_tokens_seen": 67484240,
      "step": 116325
    },
    {
      "epoch": 17.326481977956508,
      "grad_norm": 8.625,
      "learning_rate": 2.6729090373962957e-06,
      "loss": 0.7853,
      "num_input_tokens_seen": 67487216,
      "step": 116330
    },
    {
      "epoch": 17.32722669049747,
      "grad_norm": 17.875,
      "learning_rate": 2.67144734160274e-06,
      "loss": 1.411,
      "num_input_tokens_seen": 67489776,
      "step": 116335
    },
    {
      "epoch": 17.327971403038426,
      "grad_norm": 7.59375,
      "learning_rate": 2.669986023027382e-06,
      "loss": 1.1258,
      "num_input_tokens_seen": 67492592,
      "step": 116340
    },
    {
      "epoch": 17.328716115579386,
      "grad_norm": 9.0625,
      "learning_rate": 2.6685250816949196e-06,
      "loss": 0.9069,
      "num_input_tokens_seen": 67495920,
      "step": 116345
    },
    {
      "epoch": 17.329460828120347,
      "grad_norm": 5.875,
      "learning_rate": 2.6670645176300246e-06,
      "loss": 0.9017,
      "num_input_tokens_seen": 67498864,
      "step": 116350
    },
    {
      "epoch": 17.330205540661304,
      "grad_norm": 13.5,
      "learning_rate": 2.6656043308573826e-06,
      "loss": 0.846,
      "num_input_tokens_seen": 67501584,
      "step": 116355
    },
    {
      "epoch": 17.330950253202264,
      "grad_norm": 9.125,
      "learning_rate": 2.664144521401654e-06,
      "loss": 1.0386,
      "num_input_tokens_seen": 67504624,
      "step": 116360
    },
    {
      "epoch": 17.331694965743225,
      "grad_norm": 8.625,
      "learning_rate": 2.6626850892875033e-06,
      "loss": 1.1057,
      "num_input_tokens_seen": 67507504,
      "step": 116365
    },
    {
      "epoch": 17.33243967828418,
      "grad_norm": 11.25,
      "learning_rate": 2.6612260345395797e-06,
      "loss": 0.8396,
      "num_input_tokens_seen": 67510320,
      "step": 116370
    },
    {
      "epoch": 17.333184390825142,
      "grad_norm": 10.75,
      "learning_rate": 2.6597673571825436e-06,
      "loss": 1.1276,
      "num_input_tokens_seen": 67513008,
      "step": 116375
    },
    {
      "epoch": 17.3339291033661,
      "grad_norm": 14.875,
      "learning_rate": 2.658309057241032e-06,
      "loss": 1.0502,
      "num_input_tokens_seen": 67515760,
      "step": 116380
    },
    {
      "epoch": 17.33467381590706,
      "grad_norm": 17.0,
      "learning_rate": 2.6568511347396795e-06,
      "loss": 1.3814,
      "num_input_tokens_seen": 67518608,
      "step": 116385
    },
    {
      "epoch": 17.33541852844802,
      "grad_norm": 5.53125,
      "learning_rate": 2.6553935897031203e-06,
      "loss": 0.8006,
      "num_input_tokens_seen": 67521424,
      "step": 116390
    },
    {
      "epoch": 17.336163240988977,
      "grad_norm": 13.5,
      "learning_rate": 2.6539364221559725e-06,
      "loss": 1.0635,
      "num_input_tokens_seen": 67524272,
      "step": 116395
    },
    {
      "epoch": 17.336907953529938,
      "grad_norm": 11.3125,
      "learning_rate": 2.652479632122862e-06,
      "loss": 1.0877,
      "num_input_tokens_seen": 67526800,
      "step": 116400
    },
    {
      "epoch": 17.3376526660709,
      "grad_norm": 8.75,
      "learning_rate": 2.65102321962839e-06,
      "loss": 1.1197,
      "num_input_tokens_seen": 67530224,
      "step": 116405
    },
    {
      "epoch": 17.338397378611855,
      "grad_norm": 8.4375,
      "learning_rate": 2.6495671846971716e-06,
      "loss": 0.9676,
      "num_input_tokens_seen": 67533136,
      "step": 116410
    },
    {
      "epoch": 17.339142091152816,
      "grad_norm": 19.625,
      "learning_rate": 2.6481115273538e-06,
      "loss": 1.1962,
      "num_input_tokens_seen": 67536112,
      "step": 116415
    },
    {
      "epoch": 17.339886803693773,
      "grad_norm": 15.0625,
      "learning_rate": 2.6466562476228612e-06,
      "loss": 1.2347,
      "num_input_tokens_seen": 67539024,
      "step": 116420
    },
    {
      "epoch": 17.340631516234733,
      "grad_norm": 10.0625,
      "learning_rate": 2.645201345528953e-06,
      "loss": 1.0653,
      "num_input_tokens_seen": 67542256,
      "step": 116425
    },
    {
      "epoch": 17.341376228775694,
      "grad_norm": 9.375,
      "learning_rate": 2.643746821096646e-06,
      "loss": 1.1155,
      "num_input_tokens_seen": 67545232,
      "step": 116430
    },
    {
      "epoch": 17.34212094131665,
      "grad_norm": 17.625,
      "learning_rate": 2.6422926743505132e-06,
      "loss": 1.0343,
      "num_input_tokens_seen": 67547856,
      "step": 116435
    },
    {
      "epoch": 17.34286565385761,
      "grad_norm": 7.34375,
      "learning_rate": 2.6408389053151185e-06,
      "loss": 0.984,
      "num_input_tokens_seen": 67550832,
      "step": 116440
    },
    {
      "epoch": 17.343610366398572,
      "grad_norm": 11.8125,
      "learning_rate": 2.6393855140150304e-06,
      "loss": 1.1853,
      "num_input_tokens_seen": 67553872,
      "step": 116445
    },
    {
      "epoch": 17.34435507893953,
      "grad_norm": 14.875,
      "learning_rate": 2.6379325004747937e-06,
      "loss": 1.0707,
      "num_input_tokens_seen": 67556912,
      "step": 116450
    },
    {
      "epoch": 17.34509979148049,
      "grad_norm": 8.1875,
      "learning_rate": 2.636479864718966e-06,
      "loss": 0.981,
      "num_input_tokens_seen": 67559760,
      "step": 116455
    },
    {
      "epoch": 17.345844504021446,
      "grad_norm": 8.75,
      "learning_rate": 2.635027606772078e-06,
      "loss": 1.0883,
      "num_input_tokens_seen": 67562640,
      "step": 116460
    },
    {
      "epoch": 17.346589216562407,
      "grad_norm": 19.375,
      "learning_rate": 2.633575726658666e-06,
      "loss": 0.9408,
      "num_input_tokens_seen": 67565616,
      "step": 116465
    },
    {
      "epoch": 17.347333929103367,
      "grad_norm": 5.8125,
      "learning_rate": 2.632124224403262e-06,
      "loss": 1.1831,
      "num_input_tokens_seen": 67568560,
      "step": 116470
    },
    {
      "epoch": 17.348078641644324,
      "grad_norm": 9.5625,
      "learning_rate": 2.6306731000303842e-06,
      "loss": 1.3488,
      "num_input_tokens_seen": 67571440,
      "step": 116475
    },
    {
      "epoch": 17.348823354185285,
      "grad_norm": 8.875,
      "learning_rate": 2.6292223535645507e-06,
      "loss": 0.9127,
      "num_input_tokens_seen": 67574544,
      "step": 116480
    },
    {
      "epoch": 17.349568066726242,
      "grad_norm": 8.25,
      "learning_rate": 2.627771985030264e-06,
      "loss": 0.9102,
      "num_input_tokens_seen": 67577264,
      "step": 116485
    },
    {
      "epoch": 17.350312779267203,
      "grad_norm": 10.1875,
      "learning_rate": 2.6263219944520383e-06,
      "loss": 1.3299,
      "num_input_tokens_seen": 67579952,
      "step": 116490
    },
    {
      "epoch": 17.351057491808163,
      "grad_norm": 12.75,
      "learning_rate": 2.6248723818543625e-06,
      "loss": 1.0089,
      "num_input_tokens_seen": 67582864,
      "step": 116495
    },
    {
      "epoch": 17.35180220434912,
      "grad_norm": 21.375,
      "learning_rate": 2.6234231472617276e-06,
      "loss": 1.5016,
      "num_input_tokens_seen": 67585680,
      "step": 116500
    },
    {
      "epoch": 17.35254691689008,
      "grad_norm": 15.8125,
      "learning_rate": 2.6219742906986143e-06,
      "loss": 1.0053,
      "num_input_tokens_seen": 67588528,
      "step": 116505
    },
    {
      "epoch": 17.35329162943104,
      "grad_norm": 6.25,
      "learning_rate": 2.6205258121894976e-06,
      "loss": 1.137,
      "num_input_tokens_seen": 67591152,
      "step": 116510
    },
    {
      "epoch": 17.354036341971998,
      "grad_norm": 11.5,
      "learning_rate": 2.619077711758858e-06,
      "loss": 0.9653,
      "num_input_tokens_seen": 67594128,
      "step": 116515
    },
    {
      "epoch": 17.35478105451296,
      "grad_norm": 14.0,
      "learning_rate": 2.6176299894311444e-06,
      "loss": 1.269,
      "num_input_tokens_seen": 67597008,
      "step": 116520
    },
    {
      "epoch": 17.355525767053916,
      "grad_norm": 12.375,
      "learning_rate": 2.616182645230833e-06,
      "loss": 1.405,
      "num_input_tokens_seen": 67599792,
      "step": 116525
    },
    {
      "epoch": 17.356270479594876,
      "grad_norm": 4.59375,
      "learning_rate": 2.614735679182359e-06,
      "loss": 1.015,
      "num_input_tokens_seen": 67602832,
      "step": 116530
    },
    {
      "epoch": 17.357015192135837,
      "grad_norm": 7.0625,
      "learning_rate": 2.6132890913101783e-06,
      "loss": 1.0122,
      "num_input_tokens_seen": 67605552,
      "step": 116535
    },
    {
      "epoch": 17.357759904676794,
      "grad_norm": 15.625,
      "learning_rate": 2.6118428816387265e-06,
      "loss": 0.653,
      "num_input_tokens_seen": 67608368,
      "step": 116540
    },
    {
      "epoch": 17.358504617217754,
      "grad_norm": 12.875,
      "learning_rate": 2.610397050192431e-06,
      "loss": 1.3201,
      "num_input_tokens_seen": 67611248,
      "step": 116545
    },
    {
      "epoch": 17.359249329758715,
      "grad_norm": 11.9375,
      "learning_rate": 2.6089515969957263e-06,
      "loss": 1.332,
      "num_input_tokens_seen": 67613872,
      "step": 116550
    },
    {
      "epoch": 17.35999404229967,
      "grad_norm": 10.125,
      "learning_rate": 2.607506522073025e-06,
      "loss": 1.1626,
      "num_input_tokens_seen": 67616752,
      "step": 116555
    },
    {
      "epoch": 17.360738754840632,
      "grad_norm": 9.25,
      "learning_rate": 2.606061825448744e-06,
      "loss": 1.2968,
      "num_input_tokens_seen": 67619536,
      "step": 116560
    },
    {
      "epoch": 17.36148346738159,
      "grad_norm": 13.0625,
      "learning_rate": 2.6046175071472835e-06,
      "loss": 1.1565,
      "num_input_tokens_seen": 67622288,
      "step": 116565
    },
    {
      "epoch": 17.36222817992255,
      "grad_norm": 12.3125,
      "learning_rate": 2.603173567193054e-06,
      "loss": 0.8461,
      "num_input_tokens_seen": 67625136,
      "step": 116570
    },
    {
      "epoch": 17.36297289246351,
      "grad_norm": 3.5,
      "learning_rate": 2.601730005610442e-06,
      "loss": 0.9728,
      "num_input_tokens_seen": 67627824,
      "step": 116575
    },
    {
      "epoch": 17.363717605004467,
      "grad_norm": 12.8125,
      "learning_rate": 2.6002868224238334e-06,
      "loss": 1.2858,
      "num_input_tokens_seen": 67630480,
      "step": 116580
    },
    {
      "epoch": 17.364462317545428,
      "grad_norm": 15.625,
      "learning_rate": 2.5988440176576172e-06,
      "loss": 0.9635,
      "num_input_tokens_seen": 67633328,
      "step": 116585
    },
    {
      "epoch": 17.36520703008639,
      "grad_norm": 7.90625,
      "learning_rate": 2.5974015913361597e-06,
      "loss": 0.7554,
      "num_input_tokens_seen": 67636112,
      "step": 116590
    },
    {
      "epoch": 17.365951742627345,
      "grad_norm": 5.40625,
      "learning_rate": 2.5959595434838363e-06,
      "loss": 1.0264,
      "num_input_tokens_seen": 67638960,
      "step": 116595
    },
    {
      "epoch": 17.366696455168306,
      "grad_norm": 11.9375,
      "learning_rate": 2.594517874125005e-06,
      "loss": 1.1002,
      "num_input_tokens_seen": 67642064,
      "step": 116600
    },
    {
      "epoch": 17.367441167709263,
      "grad_norm": 10.375,
      "learning_rate": 2.5930765832840238e-06,
      "loss": 0.898,
      "num_input_tokens_seen": 67645264,
      "step": 116605
    },
    {
      "epoch": 17.368185880250223,
      "grad_norm": 15.875,
      "learning_rate": 2.5916356709852373e-06,
      "loss": 1.0936,
      "num_input_tokens_seen": 67648272,
      "step": 116610
    },
    {
      "epoch": 17.368930592791184,
      "grad_norm": 11.5625,
      "learning_rate": 2.5901951372529933e-06,
      "loss": 1.0789,
      "num_input_tokens_seen": 67650960,
      "step": 116615
    },
    {
      "epoch": 17.36967530533214,
      "grad_norm": 7.21875,
      "learning_rate": 2.5887549821116297e-06,
      "loss": 1.0445,
      "num_input_tokens_seen": 67653968,
      "step": 116620
    },
    {
      "epoch": 17.3704200178731,
      "grad_norm": 7.875,
      "learning_rate": 2.5873152055854694e-06,
      "loss": 1.1141,
      "num_input_tokens_seen": 67656592,
      "step": 116625
    },
    {
      "epoch": 17.37116473041406,
      "grad_norm": 5.53125,
      "learning_rate": 2.5858758076988425e-06,
      "loss": 1.2673,
      "num_input_tokens_seen": 67659696,
      "step": 116630
    },
    {
      "epoch": 17.37190944295502,
      "grad_norm": 9.3125,
      "learning_rate": 2.5844367884760577e-06,
      "loss": 1.1956,
      "num_input_tokens_seen": 67662672,
      "step": 116635
    },
    {
      "epoch": 17.37265415549598,
      "grad_norm": 10.5625,
      "learning_rate": 2.5829981479414346e-06,
      "loss": 1.134,
      "num_input_tokens_seen": 67665584,
      "step": 116640
    },
    {
      "epoch": 17.373398868036936,
      "grad_norm": 10.4375,
      "learning_rate": 2.58155988611927e-06,
      "loss": 1.2679,
      "num_input_tokens_seen": 67668496,
      "step": 116645
    },
    {
      "epoch": 17.374143580577897,
      "grad_norm": 17.5,
      "learning_rate": 2.580122003033872e-06,
      "loss": 1.3301,
      "num_input_tokens_seen": 67671216,
      "step": 116650
    },
    {
      "epoch": 17.374888293118858,
      "grad_norm": 19.375,
      "learning_rate": 2.578684498709524e-06,
      "loss": 1.3115,
      "num_input_tokens_seen": 67674224,
      "step": 116655
    },
    {
      "epoch": 17.375633005659815,
      "grad_norm": 6.0,
      "learning_rate": 2.5772473731705106e-06,
      "loss": 1.0174,
      "num_input_tokens_seen": 67677456,
      "step": 116660
    },
    {
      "epoch": 17.376377718200775,
      "grad_norm": 19.375,
      "learning_rate": 2.5758106264411193e-06,
      "loss": 1.1284,
      "num_input_tokens_seen": 67680336,
      "step": 116665
    },
    {
      "epoch": 17.377122430741732,
      "grad_norm": 6.125,
      "learning_rate": 2.574374258545609e-06,
      "loss": 1.0064,
      "num_input_tokens_seen": 67683088,
      "step": 116670
    },
    {
      "epoch": 17.377867143282693,
      "grad_norm": 13.1875,
      "learning_rate": 2.5729382695082572e-06,
      "loss": 1.0239,
      "num_input_tokens_seen": 67686224,
      "step": 116675
    },
    {
      "epoch": 17.378611855823653,
      "grad_norm": 8.0,
      "learning_rate": 2.5715026593533172e-06,
      "loss": 0.9765,
      "num_input_tokens_seen": 67689392,
      "step": 116680
    },
    {
      "epoch": 17.37935656836461,
      "grad_norm": 13.0,
      "learning_rate": 2.5700674281050496e-06,
      "loss": 0.9707,
      "num_input_tokens_seen": 67692112,
      "step": 116685
    },
    {
      "epoch": 17.38010128090557,
      "grad_norm": 5.96875,
      "learning_rate": 2.568632575787694e-06,
      "loss": 0.8717,
      "num_input_tokens_seen": 67695120,
      "step": 116690
    },
    {
      "epoch": 17.38084599344653,
      "grad_norm": 11.75,
      "learning_rate": 2.567198102425494e-06,
      "loss": 1.0626,
      "num_input_tokens_seen": 67698096,
      "step": 116695
    },
    {
      "epoch": 17.381590705987488,
      "grad_norm": 8.125,
      "learning_rate": 2.5657640080426815e-06,
      "loss": 0.9284,
      "num_input_tokens_seen": 67700944,
      "step": 116700
    },
    {
      "epoch": 17.38233541852845,
      "grad_norm": 11.25,
      "learning_rate": 2.5643302926634804e-06,
      "loss": 0.987,
      "num_input_tokens_seen": 67704240,
      "step": 116705
    },
    {
      "epoch": 17.383080131069406,
      "grad_norm": 9.3125,
      "learning_rate": 2.562896956312122e-06,
      "loss": 1.0887,
      "num_input_tokens_seen": 67706864,
      "step": 116710
    },
    {
      "epoch": 17.383824843610366,
      "grad_norm": 14.625,
      "learning_rate": 2.5614639990128113e-06,
      "loss": 0.9606,
      "num_input_tokens_seen": 67709744,
      "step": 116715
    },
    {
      "epoch": 17.384569556151327,
      "grad_norm": 8.125,
      "learning_rate": 2.560031420789763e-06,
      "loss": 1.2434,
      "num_input_tokens_seen": 67712560,
      "step": 116720
    },
    {
      "epoch": 17.385314268692284,
      "grad_norm": 6.78125,
      "learning_rate": 2.558599221667174e-06,
      "loss": 1.0679,
      "num_input_tokens_seen": 67715472,
      "step": 116725
    },
    {
      "epoch": 17.386058981233244,
      "grad_norm": 5.84375,
      "learning_rate": 2.557167401669247e-06,
      "loss": 1.2907,
      "num_input_tokens_seen": 67718544,
      "step": 116730
    },
    {
      "epoch": 17.386803693774205,
      "grad_norm": 6.78125,
      "learning_rate": 2.555735960820169e-06,
      "loss": 1.0744,
      "num_input_tokens_seen": 67721520,
      "step": 116735
    },
    {
      "epoch": 17.38754840631516,
      "grad_norm": 5.71875,
      "learning_rate": 2.554304899144111e-06,
      "loss": 1.2556,
      "num_input_tokens_seen": 67724304,
      "step": 116740
    },
    {
      "epoch": 17.388293118856122,
      "grad_norm": 22.0,
      "learning_rate": 2.552874216665269e-06,
      "loss": 1.2439,
      "num_input_tokens_seen": 67727248,
      "step": 116745
    },
    {
      "epoch": 17.38903783139708,
      "grad_norm": 10.9375,
      "learning_rate": 2.5514439134077945e-06,
      "loss": 0.966,
      "num_input_tokens_seen": 67730064,
      "step": 116750
    },
    {
      "epoch": 17.38978254393804,
      "grad_norm": 5.21875,
      "learning_rate": 2.5500139893958663e-06,
      "loss": 0.9471,
      "num_input_tokens_seen": 67732784,
      "step": 116755
    },
    {
      "epoch": 17.390527256479,
      "grad_norm": 14.3125,
      "learning_rate": 2.5485844446536316e-06,
      "loss": 1.4052,
      "num_input_tokens_seen": 67735824,
      "step": 116760
    },
    {
      "epoch": 17.391271969019957,
      "grad_norm": 6.28125,
      "learning_rate": 2.5471552792052463e-06,
      "loss": 1.3383,
      "num_input_tokens_seen": 67738576,
      "step": 116765
    },
    {
      "epoch": 17.392016681560918,
      "grad_norm": 9.0,
      "learning_rate": 2.545726493074849e-06,
      "loss": 1.2058,
      "num_input_tokens_seen": 67741456,
      "step": 116770
    },
    {
      "epoch": 17.39276139410188,
      "grad_norm": 12.625,
      "learning_rate": 2.544298086286584e-06,
      "loss": 1.4108,
      "num_input_tokens_seen": 67744336,
      "step": 116775
    },
    {
      "epoch": 17.393506106642835,
      "grad_norm": 10.375,
      "learning_rate": 2.54287005886458e-06,
      "loss": 1.1972,
      "num_input_tokens_seen": 67746960,
      "step": 116780
    },
    {
      "epoch": 17.394250819183796,
      "grad_norm": 10.5,
      "learning_rate": 2.541442410832959e-06,
      "loss": 1.19,
      "num_input_tokens_seen": 67749808,
      "step": 116785
    },
    {
      "epoch": 17.394995531724753,
      "grad_norm": 6.875,
      "learning_rate": 2.540015142215846e-06,
      "loss": 1.388,
      "num_input_tokens_seen": 67752432,
      "step": 116790
    },
    {
      "epoch": 17.395740244265713,
      "grad_norm": 14.75,
      "learning_rate": 2.5385882530373438e-06,
      "loss": 1.105,
      "num_input_tokens_seen": 67755472,
      "step": 116795
    },
    {
      "epoch": 17.396484956806674,
      "grad_norm": 7.4375,
      "learning_rate": 2.53716174332157e-06,
      "loss": 1.4246,
      "num_input_tokens_seen": 67758160,
      "step": 116800
    },
    {
      "epoch": 17.39722966934763,
      "grad_norm": 8.125,
      "learning_rate": 2.535735613092613e-06,
      "loss": 0.9443,
      "num_input_tokens_seen": 67760880,
      "step": 116805
    },
    {
      "epoch": 17.39797438188859,
      "grad_norm": 10.5625,
      "learning_rate": 2.5343098623745787e-06,
      "loss": 0.9628,
      "num_input_tokens_seen": 67763600,
      "step": 116810
    },
    {
      "epoch": 17.39871909442955,
      "grad_norm": 15.0625,
      "learning_rate": 2.532884491191542e-06,
      "loss": 1.1227,
      "num_input_tokens_seen": 67766448,
      "step": 116815
    },
    {
      "epoch": 17.39946380697051,
      "grad_norm": 11.6875,
      "learning_rate": 2.5314594995675845e-06,
      "loss": 1.127,
      "num_input_tokens_seen": 67769392,
      "step": 116820
    },
    {
      "epoch": 17.40020851951147,
      "grad_norm": 10.4375,
      "learning_rate": 2.530034887526789e-06,
      "loss": 0.7761,
      "num_input_tokens_seen": 67772016,
      "step": 116825
    },
    {
      "epoch": 17.400953232052427,
      "grad_norm": 11.4375,
      "learning_rate": 2.5286106550932164e-06,
      "loss": 0.9915,
      "num_input_tokens_seen": 67774960,
      "step": 116830
    },
    {
      "epoch": 17.401697944593387,
      "grad_norm": 7.6875,
      "learning_rate": 2.527186802290926e-06,
      "loss": 1.0549,
      "num_input_tokens_seen": 67777616,
      "step": 116835
    },
    {
      "epoch": 17.402442657134348,
      "grad_norm": 9.375,
      "learning_rate": 2.525763329143971e-06,
      "loss": 1.1691,
      "num_input_tokens_seen": 67780240,
      "step": 116840
    },
    {
      "epoch": 17.403187369675305,
      "grad_norm": 25.0,
      "learning_rate": 2.5243402356764063e-06,
      "loss": 1.0103,
      "num_input_tokens_seen": 67782896,
      "step": 116845
    },
    {
      "epoch": 17.403932082216265,
      "grad_norm": 17.0,
      "learning_rate": 2.522917521912266e-06,
      "loss": 1.0879,
      "num_input_tokens_seen": 67785712,
      "step": 116850
    },
    {
      "epoch": 17.404676794757222,
      "grad_norm": 10.625,
      "learning_rate": 2.5214951878755944e-06,
      "loss": 1.0605,
      "num_input_tokens_seen": 67788464,
      "step": 116855
    },
    {
      "epoch": 17.405421507298183,
      "grad_norm": 9.5625,
      "learning_rate": 2.520073233590414e-06,
      "loss": 0.8184,
      "num_input_tokens_seen": 67791600,
      "step": 116860
    },
    {
      "epoch": 17.406166219839143,
      "grad_norm": 8.9375,
      "learning_rate": 2.5186516590807453e-06,
      "loss": 0.8521,
      "num_input_tokens_seen": 67794480,
      "step": 116865
    },
    {
      "epoch": 17.4069109323801,
      "grad_norm": 13.375,
      "learning_rate": 2.5172304643706123e-06,
      "loss": 1.3296,
      "num_input_tokens_seen": 67797488,
      "step": 116870
    },
    {
      "epoch": 17.40765564492106,
      "grad_norm": 11.75,
      "learning_rate": 2.515809649484016e-06,
      "loss": 1.0929,
      "num_input_tokens_seen": 67800272,
      "step": 116875
    },
    {
      "epoch": 17.40840035746202,
      "grad_norm": 4.0,
      "learning_rate": 2.5143892144449676e-06,
      "loss": 0.9447,
      "num_input_tokens_seen": 67802992,
      "step": 116880
    },
    {
      "epoch": 17.409145070002978,
      "grad_norm": 6.03125,
      "learning_rate": 2.512969159277459e-06,
      "loss": 0.9189,
      "num_input_tokens_seen": 67805904,
      "step": 116885
    },
    {
      "epoch": 17.40988978254394,
      "grad_norm": 6.46875,
      "learning_rate": 2.511549484005485e-06,
      "loss": 1.1985,
      "num_input_tokens_seen": 67808880,
      "step": 116890
    },
    {
      "epoch": 17.410634495084896,
      "grad_norm": 7.03125,
      "learning_rate": 2.5101301886530177e-06,
      "loss": 1.1216,
      "num_input_tokens_seen": 67811792,
      "step": 116895
    },
    {
      "epoch": 17.411379207625856,
      "grad_norm": 7.1875,
      "learning_rate": 2.508711273244052e-06,
      "loss": 1.3803,
      "num_input_tokens_seen": 67814864,
      "step": 116900
    },
    {
      "epoch": 17.412123920166817,
      "grad_norm": 15.1875,
      "learning_rate": 2.5072927378025467e-06,
      "loss": 1.3125,
      "num_input_tokens_seen": 67818064,
      "step": 116905
    },
    {
      "epoch": 17.412868632707774,
      "grad_norm": 7.53125,
      "learning_rate": 2.505874582352466e-06,
      "loss": 1.2069,
      "num_input_tokens_seen": 67821232,
      "step": 116910
    },
    {
      "epoch": 17.413613345248734,
      "grad_norm": 22.875,
      "learning_rate": 2.5044568069177794e-06,
      "loss": 1.0181,
      "num_input_tokens_seen": 67824336,
      "step": 116915
    },
    {
      "epoch": 17.414358057789695,
      "grad_norm": 16.125,
      "learning_rate": 2.503039411522423e-06,
      "loss": 1.2938,
      "num_input_tokens_seen": 67826864,
      "step": 116920
    },
    {
      "epoch": 17.415102770330652,
      "grad_norm": 11.875,
      "learning_rate": 2.501622396190359e-06,
      "loss": 1.1128,
      "num_input_tokens_seen": 67829520,
      "step": 116925
    },
    {
      "epoch": 17.415847482871612,
      "grad_norm": 16.125,
      "learning_rate": 2.500205760945512e-06,
      "loss": 1.1379,
      "num_input_tokens_seen": 67832176,
      "step": 116930
    },
    {
      "epoch": 17.41659219541257,
      "grad_norm": 12.0,
      "learning_rate": 2.4987895058118244e-06,
      "loss": 0.8979,
      "num_input_tokens_seen": 67834864,
      "step": 116935
    },
    {
      "epoch": 17.41733690795353,
      "grad_norm": 8.5625,
      "learning_rate": 2.4973736308132214e-06,
      "loss": 0.9908,
      "num_input_tokens_seen": 67838032,
      "step": 116940
    },
    {
      "epoch": 17.41808162049449,
      "grad_norm": 4.03125,
      "learning_rate": 2.4959581359736137e-06,
      "loss": 0.7322,
      "num_input_tokens_seen": 67840752,
      "step": 116945
    },
    {
      "epoch": 17.418826333035447,
      "grad_norm": 6.53125,
      "learning_rate": 2.494543021316928e-06,
      "loss": 1.0945,
      "num_input_tokens_seen": 67843824,
      "step": 116950
    },
    {
      "epoch": 17.419571045576408,
      "grad_norm": 6.96875,
      "learning_rate": 2.4931282868670634e-06,
      "loss": 1.0727,
      "num_input_tokens_seen": 67846800,
      "step": 116955
    },
    {
      "epoch": 17.42031575811737,
      "grad_norm": 8.9375,
      "learning_rate": 2.4917139326479213e-06,
      "loss": 1.0341,
      "num_input_tokens_seen": 67849712,
      "step": 116960
    },
    {
      "epoch": 17.421060470658325,
      "grad_norm": 9.875,
      "learning_rate": 2.4902999586833897e-06,
      "loss": 1.2051,
      "num_input_tokens_seen": 67852336,
      "step": 116965
    },
    {
      "epoch": 17.421805183199286,
      "grad_norm": 6.875,
      "learning_rate": 2.48888636499737e-06,
      "loss": 0.8027,
      "num_input_tokens_seen": 67855088,
      "step": 116970
    },
    {
      "epoch": 17.422549895740243,
      "grad_norm": 16.75,
      "learning_rate": 2.4874731516137283e-06,
      "loss": 0.9806,
      "num_input_tokens_seen": 67857872,
      "step": 116975
    },
    {
      "epoch": 17.423294608281203,
      "grad_norm": 7.5,
      "learning_rate": 2.486060318556352e-06,
      "loss": 1.0312,
      "num_input_tokens_seen": 67860816,
      "step": 116980
    },
    {
      "epoch": 17.424039320822164,
      "grad_norm": 5.96875,
      "learning_rate": 2.4846478658491075e-06,
      "loss": 1.0699,
      "num_input_tokens_seen": 67863824,
      "step": 116985
    },
    {
      "epoch": 17.42478403336312,
      "grad_norm": 11.625,
      "learning_rate": 2.483235793515845e-06,
      "loss": 0.9063,
      "num_input_tokens_seen": 67866768,
      "step": 116990
    },
    {
      "epoch": 17.42552874590408,
      "grad_norm": 12.9375,
      "learning_rate": 2.4818241015804376e-06,
      "loss": 1.1025,
      "num_input_tokens_seen": 67869840,
      "step": 116995
    },
    {
      "epoch": 17.42627345844504,
      "grad_norm": 14.75,
      "learning_rate": 2.480412790066719e-06,
      "loss": 1.4264,
      "num_input_tokens_seen": 67872464,
      "step": 117000
    },
    {
      "epoch": 17.427018170986,
      "grad_norm": 5.03125,
      "learning_rate": 2.479001858998542e-06,
      "loss": 0.9864,
      "num_input_tokens_seen": 67875824,
      "step": 117005
    },
    {
      "epoch": 17.42776288352696,
      "grad_norm": 11.8125,
      "learning_rate": 2.4775913083997348e-06,
      "loss": 1.0841,
      "num_input_tokens_seen": 67878832,
      "step": 117010
    },
    {
      "epoch": 17.428507596067917,
      "grad_norm": 14.3125,
      "learning_rate": 2.476181138294137e-06,
      "loss": 0.9295,
      "num_input_tokens_seen": 67881552,
      "step": 117015
    },
    {
      "epoch": 17.429252308608877,
      "grad_norm": 21.5,
      "learning_rate": 2.474771348705568e-06,
      "loss": 1.079,
      "num_input_tokens_seen": 67884304,
      "step": 117020
    },
    {
      "epoch": 17.429997021149838,
      "grad_norm": 9.875,
      "learning_rate": 2.4733619396578422e-06,
      "loss": 1.0916,
      "num_input_tokens_seen": 67887024,
      "step": 117025
    },
    {
      "epoch": 17.430741733690795,
      "grad_norm": 9.625,
      "learning_rate": 2.4719529111747715e-06,
      "loss": 1.2474,
      "num_input_tokens_seen": 67890320,
      "step": 117030
    },
    {
      "epoch": 17.431486446231755,
      "grad_norm": 17.375,
      "learning_rate": 2.470544263280153e-06,
      "loss": 0.9786,
      "num_input_tokens_seen": 67893168,
      "step": 117035
    },
    {
      "epoch": 17.432231158772712,
      "grad_norm": 10.3125,
      "learning_rate": 2.4691359959977985e-06,
      "loss": 1.098,
      "num_input_tokens_seen": 67895856,
      "step": 117040
    },
    {
      "epoch": 17.432975871313673,
      "grad_norm": 5.15625,
      "learning_rate": 2.467728109351486e-06,
      "loss": 1.0257,
      "num_input_tokens_seen": 67898640,
      "step": 117045
    },
    {
      "epoch": 17.433720583854633,
      "grad_norm": 7.0,
      "learning_rate": 2.466320603365013e-06,
      "loss": 1.2866,
      "num_input_tokens_seen": 67901648,
      "step": 117050
    },
    {
      "epoch": 17.43446529639559,
      "grad_norm": 12.9375,
      "learning_rate": 2.4649134780621446e-06,
      "loss": 1.4251,
      "num_input_tokens_seen": 67904336,
      "step": 117055
    },
    {
      "epoch": 17.43521000893655,
      "grad_norm": 16.5,
      "learning_rate": 2.463506733466667e-06,
      "loss": 1.0765,
      "num_input_tokens_seen": 67907312,
      "step": 117060
    },
    {
      "epoch": 17.43595472147751,
      "grad_norm": 11.3125,
      "learning_rate": 2.4621003696023354e-06,
      "loss": 1.1799,
      "num_input_tokens_seen": 67910320,
      "step": 117065
    },
    {
      "epoch": 17.43669943401847,
      "grad_norm": 9.5,
      "learning_rate": 2.4606943864929064e-06,
      "loss": 0.6492,
      "num_input_tokens_seen": 67913040,
      "step": 117070
    },
    {
      "epoch": 17.43744414655943,
      "grad_norm": 15.375,
      "learning_rate": 2.459288784162142e-06,
      "loss": 1.1054,
      "num_input_tokens_seen": 67915984,
      "step": 117075
    },
    {
      "epoch": 17.438188859100386,
      "grad_norm": 6.4375,
      "learning_rate": 2.457883562633781e-06,
      "loss": 1.069,
      "num_input_tokens_seen": 67919120,
      "step": 117080
    },
    {
      "epoch": 17.438933571641346,
      "grad_norm": 14.1875,
      "learning_rate": 2.456478721931571e-06,
      "loss": 1.174,
      "num_input_tokens_seen": 67922000,
      "step": 117085
    },
    {
      "epoch": 17.439678284182307,
      "grad_norm": 10.25,
      "learning_rate": 2.4550742620792404e-06,
      "loss": 0.9172,
      "num_input_tokens_seen": 67924784,
      "step": 117090
    },
    {
      "epoch": 17.440422996723264,
      "grad_norm": 12.5,
      "learning_rate": 2.4536701831005177e-06,
      "loss": 1.0647,
      "num_input_tokens_seen": 67927792,
      "step": 117095
    },
    {
      "epoch": 17.441167709264224,
      "grad_norm": 7.3125,
      "learning_rate": 2.4522664850191223e-06,
      "loss": 0.824,
      "num_input_tokens_seen": 67931056,
      "step": 117100
    },
    {
      "epoch": 17.441912421805185,
      "grad_norm": 5.90625,
      "learning_rate": 2.4508631678587635e-06,
      "loss": 1.0153,
      "num_input_tokens_seen": 67934192,
      "step": 117105
    },
    {
      "epoch": 17.442657134346142,
      "grad_norm": 7.5,
      "learning_rate": 2.4494602316431554e-06,
      "loss": 1.0978,
      "num_input_tokens_seen": 67937520,
      "step": 117110
    },
    {
      "epoch": 17.443401846887102,
      "grad_norm": 9.5,
      "learning_rate": 2.4480576763959956e-06,
      "loss": 1.3733,
      "num_input_tokens_seen": 67940592,
      "step": 117115
    },
    {
      "epoch": 17.44414655942806,
      "grad_norm": 10.875,
      "learning_rate": 2.4466555021409848e-06,
      "loss": 0.8326,
      "num_input_tokens_seen": 67943376,
      "step": 117120
    },
    {
      "epoch": 17.44489127196902,
      "grad_norm": 15.1875,
      "learning_rate": 2.445253708901801e-06,
      "loss": 1.3319,
      "num_input_tokens_seen": 67946160,
      "step": 117125
    },
    {
      "epoch": 17.44563598450998,
      "grad_norm": 10.8125,
      "learning_rate": 2.443852296702137e-06,
      "loss": 0.9058,
      "num_input_tokens_seen": 67948848,
      "step": 117130
    },
    {
      "epoch": 17.446380697050937,
      "grad_norm": 7.46875,
      "learning_rate": 2.4424512655656597e-06,
      "loss": 0.9635,
      "num_input_tokens_seen": 67951536,
      "step": 117135
    },
    {
      "epoch": 17.447125409591898,
      "grad_norm": 7.375,
      "learning_rate": 2.4410506155160467e-06,
      "loss": 1.0131,
      "num_input_tokens_seen": 67954672,
      "step": 117140
    },
    {
      "epoch": 17.447870122132855,
      "grad_norm": 8.375,
      "learning_rate": 2.439650346576955e-06,
      "loss": 1.1814,
      "num_input_tokens_seen": 67957520,
      "step": 117145
    },
    {
      "epoch": 17.448614834673815,
      "grad_norm": 17.5,
      "learning_rate": 2.438250458772037e-06,
      "loss": 1.0324,
      "num_input_tokens_seen": 67960752,
      "step": 117150
    },
    {
      "epoch": 17.449359547214776,
      "grad_norm": 5.0,
      "learning_rate": 2.436850952124953e-06,
      "loss": 0.6679,
      "num_input_tokens_seen": 67963696,
      "step": 117155
    },
    {
      "epoch": 17.450104259755733,
      "grad_norm": 10.625,
      "learning_rate": 2.435451826659338e-06,
      "loss": 0.96,
      "num_input_tokens_seen": 67966512,
      "step": 117160
    },
    {
      "epoch": 17.450848972296694,
      "grad_norm": 11.5,
      "learning_rate": 2.4340530823988292e-06,
      "loss": 1.0785,
      "num_input_tokens_seen": 67969552,
      "step": 117165
    },
    {
      "epoch": 17.451593684837654,
      "grad_norm": 4.0625,
      "learning_rate": 2.4326547193670556e-06,
      "loss": 0.9885,
      "num_input_tokens_seen": 67972400,
      "step": 117170
    },
    {
      "epoch": 17.45233839737861,
      "grad_norm": 7.375,
      "learning_rate": 2.4312567375876503e-06,
      "loss": 1.058,
      "num_input_tokens_seen": 67975440,
      "step": 117175
    },
    {
      "epoch": 17.45308310991957,
      "grad_norm": 13.8125,
      "learning_rate": 2.4298591370842227e-06,
      "loss": 1.3168,
      "num_input_tokens_seen": 67978160,
      "step": 117180
    },
    {
      "epoch": 17.45382782246053,
      "grad_norm": 6.59375,
      "learning_rate": 2.4284619178803814e-06,
      "loss": 1.0812,
      "num_input_tokens_seen": 67981168,
      "step": 117185
    },
    {
      "epoch": 17.45457253500149,
      "grad_norm": 6.15625,
      "learning_rate": 2.427065079999741e-06,
      "loss": 0.9627,
      "num_input_tokens_seen": 67984240,
      "step": 117190
    },
    {
      "epoch": 17.45531724754245,
      "grad_norm": 4.5,
      "learning_rate": 2.4256686234658877e-06,
      "loss": 1.391,
      "num_input_tokens_seen": 67987056,
      "step": 117195
    },
    {
      "epoch": 17.456061960083407,
      "grad_norm": 14.1875,
      "learning_rate": 2.4242725483024252e-06,
      "loss": 1.0909,
      "num_input_tokens_seen": 67990160,
      "step": 117200
    },
    {
      "epoch": 17.456806672624367,
      "grad_norm": 5.15625,
      "learning_rate": 2.4228768545329267e-06,
      "loss": 1.2014,
      "num_input_tokens_seen": 67993488,
      "step": 117205
    },
    {
      "epoch": 17.457551385165328,
      "grad_norm": 7.9375,
      "learning_rate": 2.421481542180984e-06,
      "loss": 1.1163,
      "num_input_tokens_seen": 67996432,
      "step": 117210
    },
    {
      "epoch": 17.458296097706285,
      "grad_norm": 9.125,
      "learning_rate": 2.4200866112701643e-06,
      "loss": 1.1245,
      "num_input_tokens_seen": 67999504,
      "step": 117215
    },
    {
      "epoch": 17.459040810247245,
      "grad_norm": 9.875,
      "learning_rate": 2.4186920618240294e-06,
      "loss": 0.9959,
      "num_input_tokens_seen": 68002288,
      "step": 117220
    },
    {
      "epoch": 17.459785522788202,
      "grad_norm": 9.0,
      "learning_rate": 2.417297893866138e-06,
      "loss": 0.9612,
      "num_input_tokens_seen": 68005552,
      "step": 117225
    },
    {
      "epoch": 17.460530235329163,
      "grad_norm": 21.625,
      "learning_rate": 2.41590410742005e-06,
      "loss": 1.1118,
      "num_input_tokens_seen": 68008304,
      "step": 117230
    },
    {
      "epoch": 17.461274947870123,
      "grad_norm": 9.125,
      "learning_rate": 2.4145107025093095e-06,
      "loss": 0.7677,
      "num_input_tokens_seen": 68011056,
      "step": 117235
    },
    {
      "epoch": 17.46201966041108,
      "grad_norm": 8.0,
      "learning_rate": 2.4131176791574504e-06,
      "loss": 1.3499,
      "num_input_tokens_seen": 68014160,
      "step": 117240
    },
    {
      "epoch": 17.46276437295204,
      "grad_norm": 6.84375,
      "learning_rate": 2.4117250373880184e-06,
      "loss": 1.2452,
      "num_input_tokens_seen": 68017008,
      "step": 117245
    },
    {
      "epoch": 17.463509085493,
      "grad_norm": 15.25,
      "learning_rate": 2.4103327772245275e-06,
      "loss": 1.176,
      "num_input_tokens_seen": 68019824,
      "step": 117250
    },
    {
      "epoch": 17.46425379803396,
      "grad_norm": 10.6875,
      "learning_rate": 2.408940898690512e-06,
      "loss": 1.4557,
      "num_input_tokens_seen": 68022448,
      "step": 117255
    },
    {
      "epoch": 17.46499851057492,
      "grad_norm": 9.75,
      "learning_rate": 2.407549401809478e-06,
      "loss": 1.1858,
      "num_input_tokens_seen": 68025552,
      "step": 117260
    },
    {
      "epoch": 17.465743223115876,
      "grad_norm": 21.25,
      "learning_rate": 2.4061582866049285e-06,
      "loss": 1.1326,
      "num_input_tokens_seen": 68028272,
      "step": 117265
    },
    {
      "epoch": 17.466487935656836,
      "grad_norm": 16.25,
      "learning_rate": 2.4047675531003787e-06,
      "loss": 1.425,
      "num_input_tokens_seen": 68031088,
      "step": 117270
    },
    {
      "epoch": 17.467232648197797,
      "grad_norm": 5.5625,
      "learning_rate": 2.4033772013193123e-06,
      "loss": 1.15,
      "num_input_tokens_seen": 68033968,
      "step": 117275
    },
    {
      "epoch": 17.467977360738754,
      "grad_norm": 10.9375,
      "learning_rate": 2.4019872312852244e-06,
      "loss": 1.2182,
      "num_input_tokens_seen": 68037008,
      "step": 117280
    },
    {
      "epoch": 17.468722073279714,
      "grad_norm": 13.375,
      "learning_rate": 2.400597643021596e-06,
      "loss": 1.3795,
      "num_input_tokens_seen": 68040176,
      "step": 117285
    },
    {
      "epoch": 17.469466785820675,
      "grad_norm": 10.6875,
      "learning_rate": 2.3992084365519004e-06,
      "loss": 1.2262,
      "num_input_tokens_seen": 68042992,
      "step": 117290
    },
    {
      "epoch": 17.470211498361632,
      "grad_norm": 10.125,
      "learning_rate": 2.3978196118996043e-06,
      "loss": 0.8668,
      "num_input_tokens_seen": 68046000,
      "step": 117295
    },
    {
      "epoch": 17.470956210902592,
      "grad_norm": 11.5625,
      "learning_rate": 2.3964311690881786e-06,
      "loss": 1.3463,
      "num_input_tokens_seen": 68048944,
      "step": 117300
    },
    {
      "epoch": 17.47170092344355,
      "grad_norm": 11.125,
      "learning_rate": 2.3950431081410734e-06,
      "loss": 1.2008,
      "num_input_tokens_seen": 68051792,
      "step": 117305
    },
    {
      "epoch": 17.47244563598451,
      "grad_norm": 8.75,
      "learning_rate": 2.393655429081737e-06,
      "loss": 1.0634,
      "num_input_tokens_seen": 68055088,
      "step": 117310
    },
    {
      "epoch": 17.47319034852547,
      "grad_norm": 6.0625,
      "learning_rate": 2.3922681319336197e-06,
      "loss": 0.9498,
      "num_input_tokens_seen": 68057808,
      "step": 117315
    },
    {
      "epoch": 17.473935061066427,
      "grad_norm": 26.5,
      "learning_rate": 2.3908812167201532e-06,
      "loss": 1.3391,
      "num_input_tokens_seen": 68060560,
      "step": 117320
    },
    {
      "epoch": 17.474679773607388,
      "grad_norm": 10.9375,
      "learning_rate": 2.3894946834647713e-06,
      "loss": 0.8824,
      "num_input_tokens_seen": 68063568,
      "step": 117325
    },
    {
      "epoch": 17.47542448614835,
      "grad_norm": 11.875,
      "learning_rate": 2.3881085321908912e-06,
      "loss": 1.7432,
      "num_input_tokens_seen": 68066416,
      "step": 117330
    },
    {
      "epoch": 17.476169198689306,
      "grad_norm": 12.9375,
      "learning_rate": 2.3867227629219417e-06,
      "loss": 0.8597,
      "num_input_tokens_seen": 68069552,
      "step": 117335
    },
    {
      "epoch": 17.476913911230266,
      "grad_norm": 12.3125,
      "learning_rate": 2.385337375681329e-06,
      "loss": 1.2426,
      "num_input_tokens_seen": 68072368,
      "step": 117340
    },
    {
      "epoch": 17.477658623771223,
      "grad_norm": 8.0625,
      "learning_rate": 2.383952370492451e-06,
      "loss": 1.0693,
      "num_input_tokens_seen": 68075280,
      "step": 117345
    },
    {
      "epoch": 17.478403336312184,
      "grad_norm": 11.5,
      "learning_rate": 2.3825677473787168e-06,
      "loss": 0.9263,
      "num_input_tokens_seen": 68078160,
      "step": 117350
    },
    {
      "epoch": 17.479148048853144,
      "grad_norm": 13.3125,
      "learning_rate": 2.381183506363513e-06,
      "loss": 0.8935,
      "num_input_tokens_seen": 68081136,
      "step": 117355
    },
    {
      "epoch": 17.4798927613941,
      "grad_norm": 17.25,
      "learning_rate": 2.3797996474702267e-06,
      "loss": 1.2892,
      "num_input_tokens_seen": 68083920,
      "step": 117360
    },
    {
      "epoch": 17.48063747393506,
      "grad_norm": 6.0625,
      "learning_rate": 2.378416170722228e-06,
      "loss": 0.7609,
      "num_input_tokens_seen": 68087056,
      "step": 117365
    },
    {
      "epoch": 17.48138218647602,
      "grad_norm": 7.65625,
      "learning_rate": 2.3770330761429012e-06,
      "loss": 1.4151,
      "num_input_tokens_seen": 68090000,
      "step": 117370
    },
    {
      "epoch": 17.48212689901698,
      "grad_norm": 12.5,
      "learning_rate": 2.375650363755605e-06,
      "loss": 1.2235,
      "num_input_tokens_seen": 68093168,
      "step": 117375
    },
    {
      "epoch": 17.48287161155794,
      "grad_norm": 15.8125,
      "learning_rate": 2.3742680335837042e-06,
      "loss": 1.392,
      "num_input_tokens_seen": 68096336,
      "step": 117380
    },
    {
      "epoch": 17.483616324098897,
      "grad_norm": 10.75,
      "learning_rate": 2.3728860856505526e-06,
      "loss": 1.354,
      "num_input_tokens_seen": 68099024,
      "step": 117385
    },
    {
      "epoch": 17.484361036639857,
      "grad_norm": 13.5,
      "learning_rate": 2.371504519979484e-06,
      "loss": 1.2575,
      "num_input_tokens_seen": 68101840,
      "step": 117390
    },
    {
      "epoch": 17.485105749180818,
      "grad_norm": 13.0625,
      "learning_rate": 2.3701233365938547e-06,
      "loss": 1.1002,
      "num_input_tokens_seen": 68104400,
      "step": 117395
    },
    {
      "epoch": 17.485850461721775,
      "grad_norm": 9.875,
      "learning_rate": 2.368742535516988e-06,
      "loss": 1.1743,
      "num_input_tokens_seen": 68107088,
      "step": 117400
    },
    {
      "epoch": 17.486595174262735,
      "grad_norm": 4.90625,
      "learning_rate": 2.3673621167722202e-06,
      "loss": 0.8249,
      "num_input_tokens_seen": 68109904,
      "step": 117405
    },
    {
      "epoch": 17.487339886803692,
      "grad_norm": 8.375,
      "learning_rate": 2.3659820803828586e-06,
      "loss": 0.9924,
      "num_input_tokens_seen": 68113008,
      "step": 117410
    },
    {
      "epoch": 17.488084599344653,
      "grad_norm": 9.25,
      "learning_rate": 2.364602426372234e-06,
      "loss": 1.0042,
      "num_input_tokens_seen": 68116208,
      "step": 117415
    },
    {
      "epoch": 17.488829311885613,
      "grad_norm": 13.625,
      "learning_rate": 2.3632231547636443e-06,
      "loss": 1.3284,
      "num_input_tokens_seen": 68119152,
      "step": 117420
    },
    {
      "epoch": 17.48957402442657,
      "grad_norm": 11.375,
      "learning_rate": 2.3618442655803903e-06,
      "loss": 1.0134,
      "num_input_tokens_seen": 68121776,
      "step": 117425
    },
    {
      "epoch": 17.49031873696753,
      "grad_norm": 14.75,
      "learning_rate": 2.36046575884577e-06,
      "loss": 0.9078,
      "num_input_tokens_seen": 68124720,
      "step": 117430
    },
    {
      "epoch": 17.49106344950849,
      "grad_norm": 18.125,
      "learning_rate": 2.359087634583068e-06,
      "loss": 1.7197,
      "num_input_tokens_seen": 68127728,
      "step": 117435
    },
    {
      "epoch": 17.49180816204945,
      "grad_norm": 15.25,
      "learning_rate": 2.357709892815574e-06,
      "loss": 1.1532,
      "num_input_tokens_seen": 68130736,
      "step": 117440
    },
    {
      "epoch": 17.49255287459041,
      "grad_norm": 11.75,
      "learning_rate": 2.356332533566552e-06,
      "loss": 0.9112,
      "num_input_tokens_seen": 68133424,
      "step": 117445
    },
    {
      "epoch": 17.493297587131366,
      "grad_norm": 11.1875,
      "learning_rate": 2.354955556859284e-06,
      "loss": 1.3463,
      "num_input_tokens_seen": 68136528,
      "step": 117450
    },
    {
      "epoch": 17.494042299672326,
      "grad_norm": 10.8125,
      "learning_rate": 2.3535789627170205e-06,
      "loss": 1.1773,
      "num_input_tokens_seen": 68139568,
      "step": 117455
    },
    {
      "epoch": 17.494787012213287,
      "grad_norm": 10.375,
      "learning_rate": 2.3522027511630297e-06,
      "loss": 1.2606,
      "num_input_tokens_seen": 68142224,
      "step": 117460
    },
    {
      "epoch": 17.495531724754244,
      "grad_norm": 10.5,
      "learning_rate": 2.350826922220553e-06,
      "loss": 0.8559,
      "num_input_tokens_seen": 68144816,
      "step": 117465
    },
    {
      "epoch": 17.496276437295204,
      "grad_norm": 8.0,
      "learning_rate": 2.349451475912834e-06,
      "loss": 0.9097,
      "num_input_tokens_seen": 68147728,
      "step": 117470
    },
    {
      "epoch": 17.497021149836165,
      "grad_norm": 15.125,
      "learning_rate": 2.3480764122631143e-06,
      "loss": 1.6025,
      "num_input_tokens_seen": 68150640,
      "step": 117475
    },
    {
      "epoch": 17.497765862377122,
      "grad_norm": 9.4375,
      "learning_rate": 2.3467017312946175e-06,
      "loss": 1.1684,
      "num_input_tokens_seen": 68153584,
      "step": 117480
    },
    {
      "epoch": 17.498510574918082,
      "grad_norm": 10.9375,
      "learning_rate": 2.345327433030575e-06,
      "loss": 1.3133,
      "num_input_tokens_seen": 68156368,
      "step": 117485
    },
    {
      "epoch": 17.49925528745904,
      "grad_norm": 17.375,
      "learning_rate": 2.343953517494202e-06,
      "loss": 1.3511,
      "num_input_tokens_seen": 68159344,
      "step": 117490
    },
    {
      "epoch": 17.5,
      "grad_norm": 7.28125,
      "learning_rate": 2.3425799847087066e-06,
      "loss": 1.1057,
      "num_input_tokens_seen": 68162384,
      "step": 117495
    },
    {
      "epoch": 17.50074471254096,
      "grad_norm": 20.0,
      "learning_rate": 2.341206834697288e-06,
      "loss": 1.1268,
      "num_input_tokens_seen": 68165328,
      "step": 117500
    },
    {
      "epoch": 17.501489425081918,
      "grad_norm": 14.0,
      "learning_rate": 2.339834067483157e-06,
      "loss": 1.1956,
      "num_input_tokens_seen": 68168112,
      "step": 117505
    },
    {
      "epoch": 17.502234137622878,
      "grad_norm": 14.0625,
      "learning_rate": 2.338461683089499e-06,
      "loss": 0.8831,
      "num_input_tokens_seen": 68171184,
      "step": 117510
    },
    {
      "epoch": 17.502978850163835,
      "grad_norm": 10.75,
      "learning_rate": 2.337089681539495e-06,
      "loss": 1.0145,
      "num_input_tokens_seen": 68173968,
      "step": 117515
    },
    {
      "epoch": 17.503723562704796,
      "grad_norm": 6.71875,
      "learning_rate": 2.335718062856329e-06,
      "loss": 1.1953,
      "num_input_tokens_seen": 68177104,
      "step": 117520
    },
    {
      "epoch": 17.504468275245756,
      "grad_norm": 7.875,
      "learning_rate": 2.334346827063169e-06,
      "loss": 1.224,
      "num_input_tokens_seen": 68179952,
      "step": 117525
    },
    {
      "epoch": 17.505212987786713,
      "grad_norm": 13.0625,
      "learning_rate": 2.332975974183185e-06,
      "loss": 1.3887,
      "num_input_tokens_seen": 68182864,
      "step": 117530
    },
    {
      "epoch": 17.505957700327674,
      "grad_norm": 5.375,
      "learning_rate": 2.331605504239534e-06,
      "loss": 0.9068,
      "num_input_tokens_seen": 68185744,
      "step": 117535
    },
    {
      "epoch": 17.506702412868634,
      "grad_norm": 12.6875,
      "learning_rate": 2.330235417255369e-06,
      "loss": 0.8975,
      "num_input_tokens_seen": 68188560,
      "step": 117540
    },
    {
      "epoch": 17.50744712540959,
      "grad_norm": 17.75,
      "learning_rate": 2.3288657132538387e-06,
      "loss": 1.1188,
      "num_input_tokens_seen": 68191504,
      "step": 117545
    },
    {
      "epoch": 17.50819183795055,
      "grad_norm": 4.875,
      "learning_rate": 2.32749639225808e-06,
      "loss": 1.0345,
      "num_input_tokens_seen": 68194640,
      "step": 117550
    },
    {
      "epoch": 17.50893655049151,
      "grad_norm": 10.0625,
      "learning_rate": 2.3261274542912213e-06,
      "loss": 1.1004,
      "num_input_tokens_seen": 68197520,
      "step": 117555
    },
    {
      "epoch": 17.50968126303247,
      "grad_norm": 12.0,
      "learning_rate": 2.3247588993764e-06,
      "loss": 0.9795,
      "num_input_tokens_seen": 68200752,
      "step": 117560
    },
    {
      "epoch": 17.51042597557343,
      "grad_norm": 10.0625,
      "learning_rate": 2.3233907275367283e-06,
      "loss": 0.942,
      "num_input_tokens_seen": 68203472,
      "step": 117565
    },
    {
      "epoch": 17.511170688114387,
      "grad_norm": 9.9375,
      "learning_rate": 2.3220229387953207e-06,
      "loss": 1.1681,
      "num_input_tokens_seen": 68206224,
      "step": 117570
    },
    {
      "epoch": 17.511915400655347,
      "grad_norm": 9.375,
      "learning_rate": 2.3206555331752922e-06,
      "loss": 1.0324,
      "num_input_tokens_seen": 68208976,
      "step": 117575
    },
    {
      "epoch": 17.512660113196308,
      "grad_norm": 16.625,
      "learning_rate": 2.3192885106997327e-06,
      "loss": 0.9982,
      "num_input_tokens_seen": 68211888,
      "step": 117580
    },
    {
      "epoch": 17.513404825737265,
      "grad_norm": 15.625,
      "learning_rate": 2.3179218713917456e-06,
      "loss": 1.1147,
      "num_input_tokens_seen": 68214736,
      "step": 117585
    },
    {
      "epoch": 17.514149538278225,
      "grad_norm": 7.25,
      "learning_rate": 2.3165556152744154e-06,
      "loss": 1.1166,
      "num_input_tokens_seen": 68217584,
      "step": 117590
    },
    {
      "epoch": 17.514894250819182,
      "grad_norm": 8.375,
      "learning_rate": 2.3151897423708214e-06,
      "loss": 1.2262,
      "num_input_tokens_seen": 68220432,
      "step": 117595
    },
    {
      "epoch": 17.515638963360143,
      "grad_norm": 10.5625,
      "learning_rate": 2.3138242527040416e-06,
      "loss": 1.1364,
      "num_input_tokens_seen": 68223344,
      "step": 117600
    },
    {
      "epoch": 17.516383675901103,
      "grad_norm": 8.8125,
      "learning_rate": 2.3124591462971414e-06,
      "loss": 0.9534,
      "num_input_tokens_seen": 68226384,
      "step": 117605
    },
    {
      "epoch": 17.51712838844206,
      "grad_norm": 11.5625,
      "learning_rate": 2.3110944231731907e-06,
      "loss": 1.1911,
      "num_input_tokens_seen": 68229136,
      "step": 117610
    },
    {
      "epoch": 17.51787310098302,
      "grad_norm": 7.625,
      "learning_rate": 2.309730083355238e-06,
      "loss": 0.9836,
      "num_input_tokens_seen": 68232336,
      "step": 117615
    },
    {
      "epoch": 17.51861781352398,
      "grad_norm": 7.8125,
      "learning_rate": 2.3083661268663376e-06,
      "loss": 1.2282,
      "num_input_tokens_seen": 68235152,
      "step": 117620
    },
    {
      "epoch": 17.51936252606494,
      "grad_norm": 6.59375,
      "learning_rate": 2.3070025537295257e-06,
      "loss": 0.9535,
      "num_input_tokens_seen": 68238256,
      "step": 117625
    },
    {
      "epoch": 17.5201072386059,
      "grad_norm": 16.625,
      "learning_rate": 2.3056393639678374e-06,
      "loss": 1.211,
      "num_input_tokens_seen": 68241264,
      "step": 117630
    },
    {
      "epoch": 17.520851951146856,
      "grad_norm": 8.25,
      "learning_rate": 2.3042765576043124e-06,
      "loss": 1.0133,
      "num_input_tokens_seen": 68244144,
      "step": 117635
    },
    {
      "epoch": 17.521596663687816,
      "grad_norm": 11.875,
      "learning_rate": 2.3029141346619624e-06,
      "loss": 0.9054,
      "num_input_tokens_seen": 68247024,
      "step": 117640
    },
    {
      "epoch": 17.522341376228777,
      "grad_norm": 18.25,
      "learning_rate": 2.301552095163814e-06,
      "loss": 1.369,
      "num_input_tokens_seen": 68249520,
      "step": 117645
    },
    {
      "epoch": 17.523086088769734,
      "grad_norm": 8.3125,
      "learning_rate": 2.3001904391328683e-06,
      "loss": 1.268,
      "num_input_tokens_seen": 68252560,
      "step": 117650
    },
    {
      "epoch": 17.523830801310694,
      "grad_norm": 11.4375,
      "learning_rate": 2.2988291665921396e-06,
      "loss": 1.0466,
      "num_input_tokens_seen": 68255568,
      "step": 117655
    },
    {
      "epoch": 17.52457551385165,
      "grad_norm": 7.5625,
      "learning_rate": 2.2974682775646157e-06,
      "loss": 0.9105,
      "num_input_tokens_seen": 68258832,
      "step": 117660
    },
    {
      "epoch": 17.525320226392612,
      "grad_norm": 10.0625,
      "learning_rate": 2.2961077720732943e-06,
      "loss": 1.0881,
      "num_input_tokens_seen": 68262032,
      "step": 117665
    },
    {
      "epoch": 17.526064938933573,
      "grad_norm": 10.1875,
      "learning_rate": 2.2947476501411573e-06,
      "loss": 0.9294,
      "num_input_tokens_seen": 68265232,
      "step": 117670
    },
    {
      "epoch": 17.52680965147453,
      "grad_norm": 12.75,
      "learning_rate": 2.293387911791178e-06,
      "loss": 1.2711,
      "num_input_tokens_seen": 68268048,
      "step": 117675
    },
    {
      "epoch": 17.52755436401549,
      "grad_norm": 9.25,
      "learning_rate": 2.2920285570463386e-06,
      "loss": 0.9911,
      "num_input_tokens_seen": 68270896,
      "step": 117680
    },
    {
      "epoch": 17.52829907655645,
      "grad_norm": 8.6875,
      "learning_rate": 2.2906695859295946e-06,
      "loss": 1.0301,
      "num_input_tokens_seen": 68273936,
      "step": 117685
    },
    {
      "epoch": 17.529043789097408,
      "grad_norm": 9.6875,
      "learning_rate": 2.2893109984639086e-06,
      "loss": 1.201,
      "num_input_tokens_seen": 68276720,
      "step": 117690
    },
    {
      "epoch": 17.529788501638368,
      "grad_norm": 7.96875,
      "learning_rate": 2.287952794672227e-06,
      "loss": 1.2053,
      "num_input_tokens_seen": 68279600,
      "step": 117695
    },
    {
      "epoch": 17.530533214179325,
      "grad_norm": 5.9375,
      "learning_rate": 2.2865949745775024e-06,
      "loss": 1.0115,
      "num_input_tokens_seen": 68282448,
      "step": 117700
    },
    {
      "epoch": 17.531277926720286,
      "grad_norm": 12.625,
      "learning_rate": 2.2852375382026727e-06,
      "loss": 1.0509,
      "num_input_tokens_seen": 68285200,
      "step": 117705
    },
    {
      "epoch": 17.532022639261246,
      "grad_norm": 10.8125,
      "learning_rate": 2.2838804855706606e-06,
      "loss": 1.2679,
      "num_input_tokens_seen": 68288080,
      "step": 117710
    },
    {
      "epoch": 17.532767351802203,
      "grad_norm": 7.875,
      "learning_rate": 2.282523816704407e-06,
      "loss": 1.0799,
      "num_input_tokens_seen": 68290928,
      "step": 117715
    },
    {
      "epoch": 17.533512064343164,
      "grad_norm": 7.53125,
      "learning_rate": 2.2811675316268212e-06,
      "loss": 1.1174,
      "num_input_tokens_seen": 68293584,
      "step": 117720
    },
    {
      "epoch": 17.534256776884124,
      "grad_norm": 16.75,
      "learning_rate": 2.279811630360823e-06,
      "loss": 0.9132,
      "num_input_tokens_seen": 68296688,
      "step": 117725
    },
    {
      "epoch": 17.53500148942508,
      "grad_norm": 10.0,
      "learning_rate": 2.2784561129293086e-06,
      "loss": 0.9248,
      "num_input_tokens_seen": 68299696,
      "step": 117730
    },
    {
      "epoch": 17.53574620196604,
      "grad_norm": 17.375,
      "learning_rate": 2.2771009793551927e-06,
      "loss": 1.2214,
      "num_input_tokens_seen": 68302608,
      "step": 117735
    },
    {
      "epoch": 17.536490914507,
      "grad_norm": 13.125,
      "learning_rate": 2.275746229661355e-06,
      "loss": 1.107,
      "num_input_tokens_seen": 68305584,
      "step": 117740
    },
    {
      "epoch": 17.53723562704796,
      "grad_norm": 12.3125,
      "learning_rate": 2.2743918638706952e-06,
      "loss": 1.5307,
      "num_input_tokens_seen": 68308784,
      "step": 117745
    },
    {
      "epoch": 17.53798033958892,
      "grad_norm": 12.8125,
      "learning_rate": 2.2730378820060847e-06,
      "loss": 0.9954,
      "num_input_tokens_seen": 68311792,
      "step": 117750
    },
    {
      "epoch": 17.538725052129877,
      "grad_norm": 7.5625,
      "learning_rate": 2.271684284090403e-06,
      "loss": 1.2197,
      "num_input_tokens_seen": 68314544,
      "step": 117755
    },
    {
      "epoch": 17.539469764670837,
      "grad_norm": 12.3125,
      "learning_rate": 2.270331070146514e-06,
      "loss": 0.5242,
      "num_input_tokens_seen": 68317360,
      "step": 117760
    },
    {
      "epoch": 17.540214477211798,
      "grad_norm": 12.25,
      "learning_rate": 2.2689782401972783e-06,
      "loss": 0.9117,
      "num_input_tokens_seen": 68320560,
      "step": 117765
    },
    {
      "epoch": 17.540959189752755,
      "grad_norm": 5.9375,
      "learning_rate": 2.2676257942655544e-06,
      "loss": 0.6671,
      "num_input_tokens_seen": 68324656,
      "step": 117770
    },
    {
      "epoch": 17.541703902293715,
      "grad_norm": 10.125,
      "learning_rate": 2.266273732374183e-06,
      "loss": 1.1151,
      "num_input_tokens_seen": 68327504,
      "step": 117775
    },
    {
      "epoch": 17.542448614834672,
      "grad_norm": 8.5,
      "learning_rate": 2.2649220545460176e-06,
      "loss": 1.0285,
      "num_input_tokens_seen": 68330512,
      "step": 117780
    },
    {
      "epoch": 17.543193327375633,
      "grad_norm": 7.28125,
      "learning_rate": 2.2635707608038877e-06,
      "loss": 1.2792,
      "num_input_tokens_seen": 68333392,
      "step": 117785
    },
    {
      "epoch": 17.543938039916593,
      "grad_norm": 7.4375,
      "learning_rate": 2.262219851170616e-06,
      "loss": 0.8382,
      "num_input_tokens_seen": 68336272,
      "step": 117790
    },
    {
      "epoch": 17.54468275245755,
      "grad_norm": 15.4375,
      "learning_rate": 2.260869325669035e-06,
      "loss": 0.9022,
      "num_input_tokens_seen": 68339280,
      "step": 117795
    },
    {
      "epoch": 17.54542746499851,
      "grad_norm": 10.5,
      "learning_rate": 2.259519184321951e-06,
      "loss": 0.576,
      "num_input_tokens_seen": 68342128,
      "step": 117800
    },
    {
      "epoch": 17.54617217753947,
      "grad_norm": 10.4375,
      "learning_rate": 2.2581694271521817e-06,
      "loss": 1.1841,
      "num_input_tokens_seen": 68345392,
      "step": 117805
    },
    {
      "epoch": 17.54691689008043,
      "grad_norm": 5.65625,
      "learning_rate": 2.2568200541825236e-06,
      "loss": 1.136,
      "num_input_tokens_seen": 68348880,
      "step": 117810
    },
    {
      "epoch": 17.54766160262139,
      "grad_norm": 16.0,
      "learning_rate": 2.2554710654357796e-06,
      "loss": 1.2343,
      "num_input_tokens_seen": 68352176,
      "step": 117815
    },
    {
      "epoch": 17.548406315162346,
      "grad_norm": 11.125,
      "learning_rate": 2.254122460934735e-06,
      "loss": 0.819,
      "num_input_tokens_seen": 68355440,
      "step": 117820
    },
    {
      "epoch": 17.549151027703306,
      "grad_norm": 11.5,
      "learning_rate": 2.2527742407021742e-06,
      "loss": 0.9125,
      "num_input_tokens_seen": 68358288,
      "step": 117825
    },
    {
      "epoch": 17.549895740244267,
      "grad_norm": 7.0,
      "learning_rate": 2.251426404760873e-06,
      "loss": 0.6233,
      "num_input_tokens_seen": 68361136,
      "step": 117830
    },
    {
      "epoch": 17.550640452785224,
      "grad_norm": 6.75,
      "learning_rate": 2.250078953133597e-06,
      "loss": 1.3713,
      "num_input_tokens_seen": 68364272,
      "step": 117835
    },
    {
      "epoch": 17.551385165326185,
      "grad_norm": 14.5625,
      "learning_rate": 2.2487318858431193e-06,
      "loss": 0.9078,
      "num_input_tokens_seen": 68366896,
      "step": 117840
    },
    {
      "epoch": 17.552129877867145,
      "grad_norm": 13.75,
      "learning_rate": 2.247385202912189e-06,
      "loss": 0.8191,
      "num_input_tokens_seen": 68369552,
      "step": 117845
    },
    {
      "epoch": 17.552874590408102,
      "grad_norm": 13.0,
      "learning_rate": 2.246038904363565e-06,
      "loss": 1.007,
      "num_input_tokens_seen": 68372304,
      "step": 117850
    },
    {
      "epoch": 17.553619302949063,
      "grad_norm": 21.625,
      "learning_rate": 2.2446929902199847e-06,
      "loss": 1.2433,
      "num_input_tokens_seen": 68375088,
      "step": 117855
    },
    {
      "epoch": 17.55436401549002,
      "grad_norm": 16.875,
      "learning_rate": 2.2433474605041917e-06,
      "loss": 1.0878,
      "num_input_tokens_seen": 68378000,
      "step": 117860
    },
    {
      "epoch": 17.55510872803098,
      "grad_norm": 6.53125,
      "learning_rate": 2.242002315238917e-06,
      "loss": 1.4237,
      "num_input_tokens_seen": 68380880,
      "step": 117865
    },
    {
      "epoch": 17.55585344057194,
      "grad_norm": 11.0,
      "learning_rate": 2.240657554446876e-06,
      "loss": 0.9638,
      "num_input_tokens_seen": 68383664,
      "step": 117870
    },
    {
      "epoch": 17.556598153112898,
      "grad_norm": 15.6875,
      "learning_rate": 2.239313178150798e-06,
      "loss": 1.2526,
      "num_input_tokens_seen": 68386320,
      "step": 117875
    },
    {
      "epoch": 17.557342865653858,
      "grad_norm": 8.25,
      "learning_rate": 2.2379691863733927e-06,
      "loss": 0.9556,
      "num_input_tokens_seen": 68389488,
      "step": 117880
    },
    {
      "epoch": 17.558087578194815,
      "grad_norm": 9.75,
      "learning_rate": 2.236625579137358e-06,
      "loss": 1.1231,
      "num_input_tokens_seen": 68392464,
      "step": 117885
    },
    {
      "epoch": 17.558832290735776,
      "grad_norm": 5.8125,
      "learning_rate": 2.235282356465404e-06,
      "loss": 1.0467,
      "num_input_tokens_seen": 68395280,
      "step": 117890
    },
    {
      "epoch": 17.559577003276736,
      "grad_norm": 6.96875,
      "learning_rate": 2.233939518380218e-06,
      "loss": 1.666,
      "num_input_tokens_seen": 68398256,
      "step": 117895
    },
    {
      "epoch": 17.560321715817693,
      "grad_norm": 8.1875,
      "learning_rate": 2.232597064904479e-06,
      "loss": 1.1741,
      "num_input_tokens_seen": 68400976,
      "step": 117900
    },
    {
      "epoch": 17.561066428358654,
      "grad_norm": 5.15625,
      "learning_rate": 2.2312549960608804e-06,
      "loss": 0.8272,
      "num_input_tokens_seen": 68403824,
      "step": 117905
    },
    {
      "epoch": 17.561811140899614,
      "grad_norm": 10.3125,
      "learning_rate": 2.2299133118720837e-06,
      "loss": 1.3544,
      "num_input_tokens_seen": 68406448,
      "step": 117910
    },
    {
      "epoch": 17.56255585344057,
      "grad_norm": 9.1875,
      "learning_rate": 2.228572012360758e-06,
      "loss": 1.0141,
      "num_input_tokens_seen": 68409168,
      "step": 117915
    },
    {
      "epoch": 17.56330056598153,
      "grad_norm": 12.5,
      "learning_rate": 2.2272310975495673e-06,
      "loss": 1.284,
      "num_input_tokens_seen": 68411984,
      "step": 117920
    },
    {
      "epoch": 17.56404527852249,
      "grad_norm": 9.4375,
      "learning_rate": 2.2258905674611556e-06,
      "loss": 0.8373,
      "num_input_tokens_seen": 68415216,
      "step": 117925
    },
    {
      "epoch": 17.56478999106345,
      "grad_norm": 11.25,
      "learning_rate": 2.224550422118185e-06,
      "loss": 1.2059,
      "num_input_tokens_seen": 68418032,
      "step": 117930
    },
    {
      "epoch": 17.56553470360441,
      "grad_norm": 7.4375,
      "learning_rate": 2.223210661543279e-06,
      "loss": 0.7194,
      "num_input_tokens_seen": 68420752,
      "step": 117935
    },
    {
      "epoch": 17.566279416145367,
      "grad_norm": 9.0625,
      "learning_rate": 2.221871285759086e-06,
      "loss": 1.5002,
      "num_input_tokens_seen": 68423792,
      "step": 117940
    },
    {
      "epoch": 17.567024128686327,
      "grad_norm": 6.03125,
      "learning_rate": 2.2205322947882245e-06,
      "loss": 0.9807,
      "num_input_tokens_seen": 68426992,
      "step": 117945
    },
    {
      "epoch": 17.567768841227288,
      "grad_norm": 7.71875,
      "learning_rate": 2.2191936886533206e-06,
      "loss": 1.2292,
      "num_input_tokens_seen": 68429776,
      "step": 117950
    },
    {
      "epoch": 17.568513553768245,
      "grad_norm": 7.375,
      "learning_rate": 2.2178554673769863e-06,
      "loss": 1.0693,
      "num_input_tokens_seen": 68433072,
      "step": 117955
    },
    {
      "epoch": 17.569258266309205,
      "grad_norm": 9.3125,
      "learning_rate": 2.216517630981821e-06,
      "loss": 0.9704,
      "num_input_tokens_seen": 68435888,
      "step": 117960
    },
    {
      "epoch": 17.570002978850162,
      "grad_norm": 5.9375,
      "learning_rate": 2.215180179490442e-06,
      "loss": 0.9467,
      "num_input_tokens_seen": 68439088,
      "step": 117965
    },
    {
      "epoch": 17.570747691391123,
      "grad_norm": 11.3125,
      "learning_rate": 2.2138431129254318e-06,
      "loss": 1.0675,
      "num_input_tokens_seen": 68441936,
      "step": 117970
    },
    {
      "epoch": 17.571492403932083,
      "grad_norm": 7.53125,
      "learning_rate": 2.2125064313093857e-06,
      "loss": 1.0366,
      "num_input_tokens_seen": 68444688,
      "step": 117975
    },
    {
      "epoch": 17.57223711647304,
      "grad_norm": 7.0,
      "learning_rate": 2.2111701346648806e-06,
      "loss": 0.884,
      "num_input_tokens_seen": 68447536,
      "step": 117980
    },
    {
      "epoch": 17.572981829014,
      "grad_norm": 9.9375,
      "learning_rate": 2.209834223014498e-06,
      "loss": 0.8943,
      "num_input_tokens_seen": 68450448,
      "step": 117985
    },
    {
      "epoch": 17.57372654155496,
      "grad_norm": 22.0,
      "learning_rate": 2.2084986963808036e-06,
      "loss": 1.6552,
      "num_input_tokens_seen": 68453360,
      "step": 117990
    },
    {
      "epoch": 17.57447125409592,
      "grad_norm": 8.5,
      "learning_rate": 2.2071635547863565e-06,
      "loss": 1.4551,
      "num_input_tokens_seen": 68456272,
      "step": 117995
    },
    {
      "epoch": 17.57521596663688,
      "grad_norm": 12.0,
      "learning_rate": 2.2058287982537175e-06,
      "loss": 1.0403,
      "num_input_tokens_seen": 68459184,
      "step": 118000
    },
    {
      "epoch": 17.575960679177836,
      "grad_norm": 8.0,
      "learning_rate": 2.2044944268054315e-06,
      "loss": 0.7637,
      "num_input_tokens_seen": 68461744,
      "step": 118005
    },
    {
      "epoch": 17.576705391718797,
      "grad_norm": 12.625,
      "learning_rate": 2.203160440464047e-06,
      "loss": 1.1822,
      "num_input_tokens_seen": 68464336,
      "step": 118010
    },
    {
      "epoch": 17.577450104259757,
      "grad_norm": 14.4375,
      "learning_rate": 2.2018268392520998e-06,
      "loss": 1.1481,
      "num_input_tokens_seen": 68467504,
      "step": 118015
    },
    {
      "epoch": 17.578194816800714,
      "grad_norm": 9.5625,
      "learning_rate": 2.2004936231921153e-06,
      "loss": 1.0204,
      "num_input_tokens_seen": 68470320,
      "step": 118020
    },
    {
      "epoch": 17.578939529341675,
      "grad_norm": 8.5625,
      "learning_rate": 2.1991607923066176e-06,
      "loss": 1.3414,
      "num_input_tokens_seen": 68472976,
      "step": 118025
    },
    {
      "epoch": 17.57968424188263,
      "grad_norm": 11.125,
      "learning_rate": 2.197828346618122e-06,
      "loss": 0.6864,
      "num_input_tokens_seen": 68475696,
      "step": 118030
    },
    {
      "epoch": 17.580428954423592,
      "grad_norm": 11.625,
      "learning_rate": 2.196496286149144e-06,
      "loss": 0.7957,
      "num_input_tokens_seen": 68478384,
      "step": 118035
    },
    {
      "epoch": 17.581173666964553,
      "grad_norm": 9.3125,
      "learning_rate": 2.195164610922182e-06,
      "loss": 1.2099,
      "num_input_tokens_seen": 68481232,
      "step": 118040
    },
    {
      "epoch": 17.58191837950551,
      "grad_norm": 9.8125,
      "learning_rate": 2.1938333209597373e-06,
      "loss": 1.4522,
      "num_input_tokens_seen": 68484496,
      "step": 118045
    },
    {
      "epoch": 17.58266309204647,
      "grad_norm": 7.3125,
      "learning_rate": 2.1925024162842978e-06,
      "loss": 1.264,
      "num_input_tokens_seen": 68487376,
      "step": 118050
    },
    {
      "epoch": 17.58340780458743,
      "grad_norm": 13.5,
      "learning_rate": 2.1911718969183535e-06,
      "loss": 1.0574,
      "num_input_tokens_seen": 68490544,
      "step": 118055
    },
    {
      "epoch": 17.584152517128388,
      "grad_norm": 20.75,
      "learning_rate": 2.189841762884376e-06,
      "loss": 1.14,
      "num_input_tokens_seen": 68493232,
      "step": 118060
    },
    {
      "epoch": 17.584897229669348,
      "grad_norm": 10.8125,
      "learning_rate": 2.188512014204841e-06,
      "loss": 0.9902,
      "num_input_tokens_seen": 68495728,
      "step": 118065
    },
    {
      "epoch": 17.585641942210305,
      "grad_norm": 13.1875,
      "learning_rate": 2.1871826509022086e-06,
      "loss": 1.2272,
      "num_input_tokens_seen": 68498544,
      "step": 118070
    },
    {
      "epoch": 17.586386654751266,
      "grad_norm": 10.625,
      "learning_rate": 2.1858536729989388e-06,
      "loss": 0.9311,
      "num_input_tokens_seen": 68501456,
      "step": 118075
    },
    {
      "epoch": 17.587131367292226,
      "grad_norm": 8.375,
      "learning_rate": 2.1845250805174854e-06,
      "loss": 0.7815,
      "num_input_tokens_seen": 68504176,
      "step": 118080
    },
    {
      "epoch": 17.587876079833183,
      "grad_norm": 23.625,
      "learning_rate": 2.183196873480295e-06,
      "loss": 0.9197,
      "num_input_tokens_seen": 68507024,
      "step": 118085
    },
    {
      "epoch": 17.588620792374144,
      "grad_norm": 21.75,
      "learning_rate": 2.1818690519098018e-06,
      "loss": 1.1458,
      "num_input_tokens_seen": 68510096,
      "step": 118090
    },
    {
      "epoch": 17.589365504915104,
      "grad_norm": 7.65625,
      "learning_rate": 2.1805416158284355e-06,
      "loss": 0.9319,
      "num_input_tokens_seen": 68513168,
      "step": 118095
    },
    {
      "epoch": 17.59011021745606,
      "grad_norm": 4.1875,
      "learning_rate": 2.1792145652586305e-06,
      "loss": 0.972,
      "num_input_tokens_seen": 68516016,
      "step": 118100
    },
    {
      "epoch": 17.590854929997022,
      "grad_norm": 12.25,
      "learning_rate": 2.177887900222797e-06,
      "loss": 0.9608,
      "num_input_tokens_seen": 68519152,
      "step": 118105
    },
    {
      "epoch": 17.59159964253798,
      "grad_norm": 12.3125,
      "learning_rate": 2.176561620743356e-06,
      "loss": 1.4286,
      "num_input_tokens_seen": 68522192,
      "step": 118110
    },
    {
      "epoch": 17.59234435507894,
      "grad_norm": 6.59375,
      "learning_rate": 2.1752357268427086e-06,
      "loss": 1.0637,
      "num_input_tokens_seen": 68525168,
      "step": 118115
    },
    {
      "epoch": 17.5930890676199,
      "grad_norm": 19.625,
      "learning_rate": 2.173910218543254e-06,
      "loss": 1.2186,
      "num_input_tokens_seen": 68528240,
      "step": 118120
    },
    {
      "epoch": 17.593833780160857,
      "grad_norm": 7.8125,
      "learning_rate": 2.172585095867391e-06,
      "loss": 1.2215,
      "num_input_tokens_seen": 68530896,
      "step": 118125
    },
    {
      "epoch": 17.594578492701817,
      "grad_norm": 9.625,
      "learning_rate": 2.1712603588374956e-06,
      "loss": 0.9575,
      "num_input_tokens_seen": 68533936,
      "step": 118130
    },
    {
      "epoch": 17.595323205242778,
      "grad_norm": 8.875,
      "learning_rate": 2.169936007475959e-06,
      "loss": 0.965,
      "num_input_tokens_seen": 68536848,
      "step": 118135
    },
    {
      "epoch": 17.596067917783735,
      "grad_norm": 4.5,
      "learning_rate": 2.1686120418051457e-06,
      "loss": 0.7263,
      "num_input_tokens_seen": 68539824,
      "step": 118140
    },
    {
      "epoch": 17.596812630324695,
      "grad_norm": 4.65625,
      "learning_rate": 2.167288461847433e-06,
      "loss": 0.9842,
      "num_input_tokens_seen": 68542800,
      "step": 118145
    },
    {
      "epoch": 17.597557342865652,
      "grad_norm": 7.5,
      "learning_rate": 2.165965267625175e-06,
      "loss": 0.9346,
      "num_input_tokens_seen": 68545904,
      "step": 118150
    },
    {
      "epoch": 17.598302055406613,
      "grad_norm": 9.75,
      "learning_rate": 2.164642459160726e-06,
      "loss": 1.2296,
      "num_input_tokens_seen": 68548720,
      "step": 118155
    },
    {
      "epoch": 17.599046767947573,
      "grad_norm": 11.0625,
      "learning_rate": 2.163320036476435e-06,
      "loss": 1.1373,
      "num_input_tokens_seen": 68551696,
      "step": 118160
    },
    {
      "epoch": 17.59979148048853,
      "grad_norm": 14.625,
      "learning_rate": 2.1619979995946365e-06,
      "loss": 1.3112,
      "num_input_tokens_seen": 68554896,
      "step": 118165
    },
    {
      "epoch": 17.60053619302949,
      "grad_norm": 14.875,
      "learning_rate": 2.160676348537674e-06,
      "loss": 1.0887,
      "num_input_tokens_seen": 68557904,
      "step": 118170
    },
    {
      "epoch": 17.601280905570448,
      "grad_norm": 12.4375,
      "learning_rate": 2.159355083327866e-06,
      "loss": 1.0607,
      "num_input_tokens_seen": 68560656,
      "step": 118175
    },
    {
      "epoch": 17.60202561811141,
      "grad_norm": 12.1875,
      "learning_rate": 2.158034203987547e-06,
      "loss": 1.1417,
      "num_input_tokens_seen": 68563408,
      "step": 118180
    },
    {
      "epoch": 17.60277033065237,
      "grad_norm": 10.6875,
      "learning_rate": 2.1567137105390183e-06,
      "loss": 1.3068,
      "num_input_tokens_seen": 68566416,
      "step": 118185
    },
    {
      "epoch": 17.603515043193326,
      "grad_norm": 14.0625,
      "learning_rate": 2.155393603004602e-06,
      "loss": 1.7406,
      "num_input_tokens_seen": 68569232,
      "step": 118190
    },
    {
      "epoch": 17.604259755734287,
      "grad_norm": 9.6875,
      "learning_rate": 2.15407388140659e-06,
      "loss": 1.0557,
      "num_input_tokens_seen": 68571888,
      "step": 118195
    },
    {
      "epoch": 17.605004468275247,
      "grad_norm": 12.0625,
      "learning_rate": 2.1527545457672743e-06,
      "loss": 0.8405,
      "num_input_tokens_seen": 68574960,
      "step": 118200
    },
    {
      "epoch": 17.605749180816204,
      "grad_norm": 13.5625,
      "learning_rate": 2.1514355961089583e-06,
      "loss": 0.9073,
      "num_input_tokens_seen": 68577968,
      "step": 118205
    },
    {
      "epoch": 17.606493893357165,
      "grad_norm": 4.21875,
      "learning_rate": 2.1501170324539134e-06,
      "loss": 0.9514,
      "num_input_tokens_seen": 68580720,
      "step": 118210
    },
    {
      "epoch": 17.60723860589812,
      "grad_norm": 14.5,
      "learning_rate": 2.1487988548244133e-06,
      "loss": 0.9533,
      "num_input_tokens_seen": 68583664,
      "step": 118215
    },
    {
      "epoch": 17.607983318439082,
      "grad_norm": 18.375,
      "learning_rate": 2.1474810632427344e-06,
      "loss": 1.009,
      "num_input_tokens_seen": 68586640,
      "step": 118220
    },
    {
      "epoch": 17.608728030980043,
      "grad_norm": 9.8125,
      "learning_rate": 2.1461636577311373e-06,
      "loss": 1.0896,
      "num_input_tokens_seen": 68589520,
      "step": 118225
    },
    {
      "epoch": 17.609472743521,
      "grad_norm": 12.9375,
      "learning_rate": 2.1448466383118786e-06,
      "loss": 1.1926,
      "num_input_tokens_seen": 68592272,
      "step": 118230
    },
    {
      "epoch": 17.61021745606196,
      "grad_norm": 7.625,
      "learning_rate": 2.143530005007202e-06,
      "loss": 1.0438,
      "num_input_tokens_seen": 68595312,
      "step": 118235
    },
    {
      "epoch": 17.61096216860292,
      "grad_norm": 12.375,
      "learning_rate": 2.142213757839362e-06,
      "loss": 0.5911,
      "num_input_tokens_seen": 68598032,
      "step": 118240
    },
    {
      "epoch": 17.611706881143878,
      "grad_norm": 12.4375,
      "learning_rate": 2.140897896830582e-06,
      "loss": 0.8937,
      "num_input_tokens_seen": 68601232,
      "step": 118245
    },
    {
      "epoch": 17.61245159368484,
      "grad_norm": 9.125,
      "learning_rate": 2.139582422003106e-06,
      "loss": 1.0171,
      "num_input_tokens_seen": 68604272,
      "step": 118250
    },
    {
      "epoch": 17.613196306225795,
      "grad_norm": 8.5625,
      "learning_rate": 2.138267333379146e-06,
      "loss": 0.9762,
      "num_input_tokens_seen": 68607440,
      "step": 118255
    },
    {
      "epoch": 17.613941018766756,
      "grad_norm": 12.75,
      "learning_rate": 2.1369526309809263e-06,
      "loss": 1.0869,
      "num_input_tokens_seen": 68610288,
      "step": 118260
    },
    {
      "epoch": 17.614685731307716,
      "grad_norm": 10.5,
      "learning_rate": 2.1356383148306515e-06,
      "loss": 0.7763,
      "num_input_tokens_seen": 68612816,
      "step": 118265
    },
    {
      "epoch": 17.615430443848673,
      "grad_norm": 9.5,
      "learning_rate": 2.134324384950531e-06,
      "loss": 0.8966,
      "num_input_tokens_seen": 68615632,
      "step": 118270
    },
    {
      "epoch": 17.616175156389634,
      "grad_norm": 10.9375,
      "learning_rate": 2.1330108413627647e-06,
      "loss": 0.9575,
      "num_input_tokens_seen": 68618416,
      "step": 118275
    },
    {
      "epoch": 17.616919868930594,
      "grad_norm": 8.5,
      "learning_rate": 2.131697684089537e-06,
      "loss": 1.1644,
      "num_input_tokens_seen": 68621456,
      "step": 118280
    },
    {
      "epoch": 17.61766458147155,
      "grad_norm": 15.0625,
      "learning_rate": 2.1303849131530357e-06,
      "loss": 1.1241,
      "num_input_tokens_seen": 68624368,
      "step": 118285
    },
    {
      "epoch": 17.618409294012512,
      "grad_norm": 9.875,
      "learning_rate": 2.129072528575432e-06,
      "loss": 1.4931,
      "num_input_tokens_seen": 68627216,
      "step": 118290
    },
    {
      "epoch": 17.61915400655347,
      "grad_norm": 8.875,
      "learning_rate": 2.1277605303789057e-06,
      "loss": 1.1816,
      "num_input_tokens_seen": 68629936,
      "step": 118295
    },
    {
      "epoch": 17.61989871909443,
      "grad_norm": 19.125,
      "learning_rate": 2.126448918585619e-06,
      "loss": 1.3743,
      "num_input_tokens_seen": 68633072,
      "step": 118300
    },
    {
      "epoch": 17.62064343163539,
      "grad_norm": 4.5,
      "learning_rate": 2.12513769321773e-06,
      "loss": 0.9421,
      "num_input_tokens_seen": 68635824,
      "step": 118305
    },
    {
      "epoch": 17.621388144176347,
      "grad_norm": 4.21875,
      "learning_rate": 2.123826854297395e-06,
      "loss": 1.2456,
      "num_input_tokens_seen": 68638672,
      "step": 118310
    },
    {
      "epoch": 17.622132856717307,
      "grad_norm": 8.1875,
      "learning_rate": 2.1225164018467468e-06,
      "loss": 0.8167,
      "num_input_tokens_seen": 68641648,
      "step": 118315
    },
    {
      "epoch": 17.622877569258268,
      "grad_norm": 15.25,
      "learning_rate": 2.1212063358879374e-06,
      "loss": 1.2091,
      "num_input_tokens_seen": 68644304,
      "step": 118320
    },
    {
      "epoch": 17.623622281799225,
      "grad_norm": 12.5625,
      "learning_rate": 2.1198966564430935e-06,
      "loss": 0.8827,
      "num_input_tokens_seen": 68647088,
      "step": 118325
    },
    {
      "epoch": 17.624366994340185,
      "grad_norm": 13.25,
      "learning_rate": 2.1185873635343413e-06,
      "loss": 1.0353,
      "num_input_tokens_seen": 68650000,
      "step": 118330
    },
    {
      "epoch": 17.625111706881142,
      "grad_norm": 12.4375,
      "learning_rate": 2.1172784571837965e-06,
      "loss": 0.9738,
      "num_input_tokens_seen": 68652816,
      "step": 118335
    },
    {
      "epoch": 17.625856419422103,
      "grad_norm": 11.0625,
      "learning_rate": 2.115969937413581e-06,
      "loss": 1.2127,
      "num_input_tokens_seen": 68655632,
      "step": 118340
    },
    {
      "epoch": 17.626601131963064,
      "grad_norm": 5.59375,
      "learning_rate": 2.1146618042457935e-06,
      "loss": 1.0627,
      "num_input_tokens_seen": 68658672,
      "step": 118345
    },
    {
      "epoch": 17.62734584450402,
      "grad_norm": 9.875,
      "learning_rate": 2.113354057702535e-06,
      "loss": 1.3403,
      "num_input_tokens_seen": 68661776,
      "step": 118350
    },
    {
      "epoch": 17.62809055704498,
      "grad_norm": 8.625,
      "learning_rate": 2.1120466978059e-06,
      "loss": 1.0978,
      "num_input_tokens_seen": 68664464,
      "step": 118355
    },
    {
      "epoch": 17.62883526958594,
      "grad_norm": 8.625,
      "learning_rate": 2.1107397245779705e-06,
      "loss": 0.9735,
      "num_input_tokens_seen": 68667472,
      "step": 118360
    },
    {
      "epoch": 17.6295799821269,
      "grad_norm": 12.8125,
      "learning_rate": 2.109433138040834e-06,
      "loss": 1.1679,
      "num_input_tokens_seen": 68670160,
      "step": 118365
    },
    {
      "epoch": 17.63032469466786,
      "grad_norm": 14.125,
      "learning_rate": 2.1081269382165536e-06,
      "loss": 1.2891,
      "num_input_tokens_seen": 68673072,
      "step": 118370
    },
    {
      "epoch": 17.631069407208816,
      "grad_norm": 10.875,
      "learning_rate": 2.1068211251272063e-06,
      "loss": 1.4329,
      "num_input_tokens_seen": 68676016,
      "step": 118375
    },
    {
      "epoch": 17.631814119749777,
      "grad_norm": 7.9375,
      "learning_rate": 2.105515698794844e-06,
      "loss": 1.172,
      "num_input_tokens_seen": 68678896,
      "step": 118380
    },
    {
      "epoch": 17.632558832290737,
      "grad_norm": 8.0625,
      "learning_rate": 2.104210659241532e-06,
      "loss": 1.0379,
      "num_input_tokens_seen": 68681744,
      "step": 118385
    },
    {
      "epoch": 17.633303544831694,
      "grad_norm": 9.4375,
      "learning_rate": 2.102906006489308e-06,
      "loss": 0.8562,
      "num_input_tokens_seen": 68684240,
      "step": 118390
    },
    {
      "epoch": 17.634048257372655,
      "grad_norm": 11.625,
      "learning_rate": 2.1016017405602135e-06,
      "loss": 1.5313,
      "num_input_tokens_seen": 68687184,
      "step": 118395
    },
    {
      "epoch": 17.63479296991361,
      "grad_norm": 7.75,
      "learning_rate": 2.100297861476286e-06,
      "loss": 1.0787,
      "num_input_tokens_seen": 68689680,
      "step": 118400
    },
    {
      "epoch": 17.635537682454572,
      "grad_norm": 5.6875,
      "learning_rate": 2.0989943692595495e-06,
      "loss": 1.0338,
      "num_input_tokens_seen": 68692368,
      "step": 118405
    },
    {
      "epoch": 17.636282394995533,
      "grad_norm": 9.6875,
      "learning_rate": 2.0976912639320336e-06,
      "loss": 0.9832,
      "num_input_tokens_seen": 68695152,
      "step": 118410
    },
    {
      "epoch": 17.63702710753649,
      "grad_norm": 9.8125,
      "learning_rate": 2.096388545515743e-06,
      "loss": 1.3682,
      "num_input_tokens_seen": 68697840,
      "step": 118415
    },
    {
      "epoch": 17.63777182007745,
      "grad_norm": 21.625,
      "learning_rate": 2.095086214032693e-06,
      "loss": 1.2472,
      "num_input_tokens_seen": 68700912,
      "step": 118420
    },
    {
      "epoch": 17.63851653261841,
      "grad_norm": 10.25,
      "learning_rate": 2.0937842695048754e-06,
      "loss": 1.1244,
      "num_input_tokens_seen": 68703664,
      "step": 118425
    },
    {
      "epoch": 17.639261245159368,
      "grad_norm": 7.375,
      "learning_rate": 2.0924827119542965e-06,
      "loss": 1.2212,
      "num_input_tokens_seen": 68706704,
      "step": 118430
    },
    {
      "epoch": 17.64000595770033,
      "grad_norm": 8.625,
      "learning_rate": 2.0911815414029423e-06,
      "loss": 1.2144,
      "num_input_tokens_seen": 68709456,
      "step": 118435
    },
    {
      "epoch": 17.640750670241285,
      "grad_norm": 21.125,
      "learning_rate": 2.089880757872786e-06,
      "loss": 0.8988,
      "num_input_tokens_seen": 68712176,
      "step": 118440
    },
    {
      "epoch": 17.641495382782246,
      "grad_norm": 11.0625,
      "learning_rate": 2.088580361385814e-06,
      "loss": 1.2424,
      "num_input_tokens_seen": 68714992,
      "step": 118445
    },
    {
      "epoch": 17.642240095323206,
      "grad_norm": 7.75,
      "learning_rate": 2.0872803519639856e-06,
      "loss": 0.9208,
      "num_input_tokens_seen": 68717936,
      "step": 118450
    },
    {
      "epoch": 17.642984807864163,
      "grad_norm": 4.8125,
      "learning_rate": 2.085980729629275e-06,
      "loss": 0.9998,
      "num_input_tokens_seen": 68720688,
      "step": 118455
    },
    {
      "epoch": 17.643729520405124,
      "grad_norm": 10.125,
      "learning_rate": 2.084681494403623e-06,
      "loss": 1.0796,
      "num_input_tokens_seen": 68723600,
      "step": 118460
    },
    {
      "epoch": 17.644474232946084,
      "grad_norm": 6.46875,
      "learning_rate": 2.0833826463089957e-06,
      "loss": 0.8063,
      "num_input_tokens_seen": 68726416,
      "step": 118465
    },
    {
      "epoch": 17.64521894548704,
      "grad_norm": 4.4375,
      "learning_rate": 2.0820841853673252e-06,
      "loss": 1.1657,
      "num_input_tokens_seen": 68729072,
      "step": 118470
    },
    {
      "epoch": 17.645963658028002,
      "grad_norm": 13.5625,
      "learning_rate": 2.080786111600544e-06,
      "loss": 1.0446,
      "num_input_tokens_seen": 68731920,
      "step": 118475
    },
    {
      "epoch": 17.64670837056896,
      "grad_norm": 10.5,
      "learning_rate": 2.079488425030593e-06,
      "loss": 0.9789,
      "num_input_tokens_seen": 68734992,
      "step": 118480
    },
    {
      "epoch": 17.64745308310992,
      "grad_norm": 9.375,
      "learning_rate": 2.0781911256793905e-06,
      "loss": 1.2098,
      "num_input_tokens_seen": 68738128,
      "step": 118485
    },
    {
      "epoch": 17.64819779565088,
      "grad_norm": 6.0625,
      "learning_rate": 2.076894213568853e-06,
      "loss": 0.954,
      "num_input_tokens_seen": 68740944,
      "step": 118490
    },
    {
      "epoch": 17.648942508191837,
      "grad_norm": 11.5625,
      "learning_rate": 2.0755976887208815e-06,
      "loss": 1.0913,
      "num_input_tokens_seen": 68743728,
      "step": 118495
    },
    {
      "epoch": 17.649687220732797,
      "grad_norm": 7.46875,
      "learning_rate": 2.0743015511573952e-06,
      "loss": 1.0151,
      "num_input_tokens_seen": 68746928,
      "step": 118500
    },
    {
      "epoch": 17.650431933273758,
      "grad_norm": 12.9375,
      "learning_rate": 2.0730058009002793e-06,
      "loss": 0.9977,
      "num_input_tokens_seen": 68749840,
      "step": 118505
    },
    {
      "epoch": 17.651176645814715,
      "grad_norm": 7.875,
      "learning_rate": 2.0717104379714304e-06,
      "loss": 1.0039,
      "num_input_tokens_seen": 68752784,
      "step": 118510
    },
    {
      "epoch": 17.651921358355676,
      "grad_norm": 17.875,
      "learning_rate": 2.070415462392733e-06,
      "loss": 1.0083,
      "num_input_tokens_seen": 68755792,
      "step": 118515
    },
    {
      "epoch": 17.652666070896633,
      "grad_norm": 15.0,
      "learning_rate": 2.0691208741860562e-06,
      "loss": 1.0707,
      "num_input_tokens_seen": 68759024,
      "step": 118520
    },
    {
      "epoch": 17.653410783437593,
      "grad_norm": 10.0625,
      "learning_rate": 2.06782667337328e-06,
      "loss": 1.2224,
      "num_input_tokens_seen": 68762032,
      "step": 118525
    },
    {
      "epoch": 17.654155495978554,
      "grad_norm": 7.53125,
      "learning_rate": 2.0665328599762613e-06,
      "loss": 1.1703,
      "num_input_tokens_seen": 68764848,
      "step": 118530
    },
    {
      "epoch": 17.65490020851951,
      "grad_norm": 14.0,
      "learning_rate": 2.0652394340168664e-06,
      "loss": 0.8139,
      "num_input_tokens_seen": 68767856,
      "step": 118535
    },
    {
      "epoch": 17.65564492106047,
      "grad_norm": 10.0625,
      "learning_rate": 2.0639463955169414e-06,
      "loss": 0.8729,
      "num_input_tokens_seen": 68770608,
      "step": 118540
    },
    {
      "epoch": 17.656389633601428,
      "grad_norm": 7.875,
      "learning_rate": 2.0626537444983274e-06,
      "loss": 0.9517,
      "num_input_tokens_seen": 68773584,
      "step": 118545
    },
    {
      "epoch": 17.65713434614239,
      "grad_norm": 11.25,
      "learning_rate": 2.061361480982868e-06,
      "loss": 0.929,
      "num_input_tokens_seen": 68776336,
      "step": 118550
    },
    {
      "epoch": 17.65787905868335,
      "grad_norm": 5.25,
      "learning_rate": 2.060069604992393e-06,
      "loss": 1.2152,
      "num_input_tokens_seen": 68779472,
      "step": 118555
    },
    {
      "epoch": 17.658623771224306,
      "grad_norm": 16.25,
      "learning_rate": 2.058778116548729e-06,
      "loss": 0.9975,
      "num_input_tokens_seen": 68782320,
      "step": 118560
    },
    {
      "epoch": 17.659368483765267,
      "grad_norm": 2.984375,
      "learning_rate": 2.057487015673684e-06,
      "loss": 1.1009,
      "num_input_tokens_seen": 68785232,
      "step": 118565
    },
    {
      "epoch": 17.660113196306227,
      "grad_norm": 10.1875,
      "learning_rate": 2.0561963023890853e-06,
      "loss": 1.0429,
      "num_input_tokens_seen": 68787792,
      "step": 118570
    },
    {
      "epoch": 17.660857908847184,
      "grad_norm": 10.3125,
      "learning_rate": 2.0549059767167255e-06,
      "loss": 0.8602,
      "num_input_tokens_seen": 68790864,
      "step": 118575
    },
    {
      "epoch": 17.661602621388145,
      "grad_norm": 10.875,
      "learning_rate": 2.053616038678416e-06,
      "loss": 0.8242,
      "num_input_tokens_seen": 68793744,
      "step": 118580
    },
    {
      "epoch": 17.6623473339291,
      "grad_norm": 6.125,
      "learning_rate": 2.0523264882959357e-06,
      "loss": 0.9892,
      "num_input_tokens_seen": 68796560,
      "step": 118585
    },
    {
      "epoch": 17.663092046470062,
      "grad_norm": 14.1875,
      "learning_rate": 2.051037325591079e-06,
      "loss": 1.2625,
      "num_input_tokens_seen": 68799440,
      "step": 118590
    },
    {
      "epoch": 17.663836759011023,
      "grad_norm": 24.0,
      "learning_rate": 2.0497485505856256e-06,
      "loss": 1.6681,
      "num_input_tokens_seen": 68802256,
      "step": 118595
    },
    {
      "epoch": 17.66458147155198,
      "grad_norm": 12.8125,
      "learning_rate": 2.0484601633013383e-06,
      "loss": 1.1582,
      "num_input_tokens_seen": 68805424,
      "step": 118600
    },
    {
      "epoch": 17.66532618409294,
      "grad_norm": 5.28125,
      "learning_rate": 2.0471721637599944e-06,
      "loss": 1.217,
      "num_input_tokens_seen": 68808304,
      "step": 118605
    },
    {
      "epoch": 17.6660708966339,
      "grad_norm": 12.625,
      "learning_rate": 2.0458845519833487e-06,
      "loss": 1.2221,
      "num_input_tokens_seen": 68811312,
      "step": 118610
    },
    {
      "epoch": 17.666815609174858,
      "grad_norm": 5.28125,
      "learning_rate": 2.044597327993153e-06,
      "loss": 0.9709,
      "num_input_tokens_seen": 68814096,
      "step": 118615
    },
    {
      "epoch": 17.66756032171582,
      "grad_norm": 6.03125,
      "learning_rate": 2.0433104918111514e-06,
      "loss": 0.874,
      "num_input_tokens_seen": 68816816,
      "step": 118620
    },
    {
      "epoch": 17.668305034256775,
      "grad_norm": 10.6875,
      "learning_rate": 2.0420240434590925e-06,
      "loss": 1.0361,
      "num_input_tokens_seen": 68819696,
      "step": 118625
    },
    {
      "epoch": 17.669049746797736,
      "grad_norm": 20.75,
      "learning_rate": 2.0407379829587013e-06,
      "loss": 1.6043,
      "num_input_tokens_seen": 68822576,
      "step": 118630
    },
    {
      "epoch": 17.669794459338696,
      "grad_norm": 10.6875,
      "learning_rate": 2.039452310331705e-06,
      "loss": 0.631,
      "num_input_tokens_seen": 68825296,
      "step": 118635
    },
    {
      "epoch": 17.670539171879653,
      "grad_norm": 12.75,
      "learning_rate": 2.0381670255998297e-06,
      "loss": 0.9866,
      "num_input_tokens_seen": 68828432,
      "step": 118640
    },
    {
      "epoch": 17.671283884420614,
      "grad_norm": 9.4375,
      "learning_rate": 2.0368821287847785e-06,
      "loss": 0.7882,
      "num_input_tokens_seen": 68831088,
      "step": 118645
    },
    {
      "epoch": 17.672028596961574,
      "grad_norm": 6.5625,
      "learning_rate": 2.035597619908272e-06,
      "loss": 1.034,
      "num_input_tokens_seen": 68833904,
      "step": 118650
    },
    {
      "epoch": 17.67277330950253,
      "grad_norm": 14.0,
      "learning_rate": 2.0343134989919995e-06,
      "loss": 1.5523,
      "num_input_tokens_seen": 68836784,
      "step": 118655
    },
    {
      "epoch": 17.673518022043492,
      "grad_norm": 10.9375,
      "learning_rate": 2.033029766057662e-06,
      "loss": 1.2481,
      "num_input_tokens_seen": 68839600,
      "step": 118660
    },
    {
      "epoch": 17.67426273458445,
      "grad_norm": 10.75,
      "learning_rate": 2.03174642112694e-06,
      "loss": 0.9894,
      "num_input_tokens_seen": 68842160,
      "step": 118665
    },
    {
      "epoch": 17.67500744712541,
      "grad_norm": 13.9375,
      "learning_rate": 2.0304634642215215e-06,
      "loss": 1.29,
      "num_input_tokens_seen": 68845488,
      "step": 118670
    },
    {
      "epoch": 17.67575215966637,
      "grad_norm": 8.75,
      "learning_rate": 2.029180895363081e-06,
      "loss": 1.3078,
      "num_input_tokens_seen": 68848560,
      "step": 118675
    },
    {
      "epoch": 17.676496872207327,
      "grad_norm": 19.25,
      "learning_rate": 2.0278987145732786e-06,
      "loss": 0.9413,
      "num_input_tokens_seen": 68851408,
      "step": 118680
    },
    {
      "epoch": 17.677241584748288,
      "grad_norm": 4.40625,
      "learning_rate": 2.0266169218737836e-06,
      "loss": 0.7965,
      "num_input_tokens_seen": 68854000,
      "step": 118685
    },
    {
      "epoch": 17.677986297289245,
      "grad_norm": 8.75,
      "learning_rate": 2.0253355172862394e-06,
      "loss": 1.3505,
      "num_input_tokens_seen": 68856848,
      "step": 118690
    },
    {
      "epoch": 17.678731009830205,
      "grad_norm": 7.96875,
      "learning_rate": 2.0240545008323064e-06,
      "loss": 1.0138,
      "num_input_tokens_seen": 68859920,
      "step": 118695
    },
    {
      "epoch": 17.679475722371166,
      "grad_norm": 9.375,
      "learning_rate": 2.0227738725336176e-06,
      "loss": 0.9541,
      "num_input_tokens_seen": 68863248,
      "step": 118700
    },
    {
      "epoch": 17.680220434912123,
      "grad_norm": 7.375,
      "learning_rate": 2.0214936324118137e-06,
      "loss": 1.3803,
      "num_input_tokens_seen": 68865872,
      "step": 118705
    },
    {
      "epoch": 17.680965147453083,
      "grad_norm": 6.90625,
      "learning_rate": 2.0202137804885196e-06,
      "loss": 0.9753,
      "num_input_tokens_seen": 68868976,
      "step": 118710
    },
    {
      "epoch": 17.681709859994044,
      "grad_norm": 12.25,
      "learning_rate": 2.018934316785359e-06,
      "loss": 1.093,
      "num_input_tokens_seen": 68871952,
      "step": 118715
    },
    {
      "epoch": 17.682454572535,
      "grad_norm": 8.4375,
      "learning_rate": 2.017655241323946e-06,
      "loss": 1.2156,
      "num_input_tokens_seen": 68874832,
      "step": 118720
    },
    {
      "epoch": 17.68319928507596,
      "grad_norm": 13.75,
      "learning_rate": 2.016376554125887e-06,
      "loss": 1.1878,
      "num_input_tokens_seen": 68877648,
      "step": 118725
    },
    {
      "epoch": 17.683943997616918,
      "grad_norm": 8.0625,
      "learning_rate": 2.0150982552127913e-06,
      "loss": 1.2403,
      "num_input_tokens_seen": 68880560,
      "step": 118730
    },
    {
      "epoch": 17.68468871015788,
      "grad_norm": 10.625,
      "learning_rate": 2.0138203446062433e-06,
      "loss": 0.6465,
      "num_input_tokens_seen": 68883280,
      "step": 118735
    },
    {
      "epoch": 17.68543342269884,
      "grad_norm": 12.0625,
      "learning_rate": 2.0125428223278453e-06,
      "loss": 1.3091,
      "num_input_tokens_seen": 68886128,
      "step": 118740
    },
    {
      "epoch": 17.686178135239796,
      "grad_norm": 12.4375,
      "learning_rate": 2.011265688399172e-06,
      "loss": 1.0914,
      "num_input_tokens_seen": 68888944,
      "step": 118745
    },
    {
      "epoch": 17.686922847780757,
      "grad_norm": 12.9375,
      "learning_rate": 2.009988942841798e-06,
      "loss": 0.983,
      "num_input_tokens_seen": 68892112,
      "step": 118750
    },
    {
      "epoch": 17.687667560321717,
      "grad_norm": 11.8125,
      "learning_rate": 2.008712585677297e-06,
      "loss": 0.9704,
      "num_input_tokens_seen": 68894928,
      "step": 118755
    },
    {
      "epoch": 17.688412272862674,
      "grad_norm": 11.3125,
      "learning_rate": 2.007436616927225e-06,
      "loss": 0.8689,
      "num_input_tokens_seen": 68898032,
      "step": 118760
    },
    {
      "epoch": 17.689156985403635,
      "grad_norm": 14.5625,
      "learning_rate": 2.006161036613147e-06,
      "loss": 1.3327,
      "num_input_tokens_seen": 68901136,
      "step": 118765
    },
    {
      "epoch": 17.68990169794459,
      "grad_norm": 13.75,
      "learning_rate": 2.0048858447566045e-06,
      "loss": 1.2169,
      "num_input_tokens_seen": 68903664,
      "step": 118770
    },
    {
      "epoch": 17.690646410485552,
      "grad_norm": 17.0,
      "learning_rate": 2.003611041379147e-06,
      "loss": 1.1119,
      "num_input_tokens_seen": 68906544,
      "step": 118775
    },
    {
      "epoch": 17.691391123026513,
      "grad_norm": 12.375,
      "learning_rate": 2.0023366265023074e-06,
      "loss": 1.6429,
      "num_input_tokens_seen": 68909744,
      "step": 118780
    },
    {
      "epoch": 17.69213583556747,
      "grad_norm": 8.8125,
      "learning_rate": 2.0010626001476184e-06,
      "loss": 1.1903,
      "num_input_tokens_seen": 68912688,
      "step": 118785
    },
    {
      "epoch": 17.69288054810843,
      "grad_norm": 9.8125,
      "learning_rate": 1.999788962336599e-06,
      "loss": 0.7871,
      "num_input_tokens_seen": 68915472,
      "step": 118790
    },
    {
      "epoch": 17.69362526064939,
      "grad_norm": 8.0,
      "learning_rate": 1.9985157130907707e-06,
      "loss": 0.9996,
      "num_input_tokens_seen": 68918416,
      "step": 118795
    },
    {
      "epoch": 17.694369973190348,
      "grad_norm": 10.5,
      "learning_rate": 1.997242852431644e-06,
      "loss": 1.1254,
      "num_input_tokens_seen": 68921296,
      "step": 118800
    },
    {
      "epoch": 17.69511468573131,
      "grad_norm": 7.65625,
      "learning_rate": 1.9959703803807156e-06,
      "loss": 1.0403,
      "num_input_tokens_seen": 68924208,
      "step": 118805
    },
    {
      "epoch": 17.695859398272265,
      "grad_norm": 17.25,
      "learning_rate": 1.994698296959491e-06,
      "loss": 1.0609,
      "num_input_tokens_seen": 68927312,
      "step": 118810
    },
    {
      "epoch": 17.696604110813226,
      "grad_norm": 10.25,
      "learning_rate": 1.9934266021894575e-06,
      "loss": 0.8979,
      "num_input_tokens_seen": 68930384,
      "step": 118815
    },
    {
      "epoch": 17.697348823354186,
      "grad_norm": 14.0,
      "learning_rate": 1.9921552960920994e-06,
      "loss": 1.0958,
      "num_input_tokens_seen": 68933392,
      "step": 118820
    },
    {
      "epoch": 17.698093535895143,
      "grad_norm": 7.15625,
      "learning_rate": 1.990884378688887e-06,
      "loss": 1.1302,
      "num_input_tokens_seen": 68936464,
      "step": 118825
    },
    {
      "epoch": 17.698838248436104,
      "grad_norm": 9.6875,
      "learning_rate": 1.989613850001304e-06,
      "loss": 1.3649,
      "num_input_tokens_seen": 68939248,
      "step": 118830
    },
    {
      "epoch": 17.699582960977065,
      "grad_norm": 6.65625,
      "learning_rate": 1.988343710050808e-06,
      "loss": 1.1451,
      "num_input_tokens_seen": 68942416,
      "step": 118835
    },
    {
      "epoch": 17.70032767351802,
      "grad_norm": 13.8125,
      "learning_rate": 1.987073958858851e-06,
      "loss": 1.1986,
      "num_input_tokens_seen": 68945392,
      "step": 118840
    },
    {
      "epoch": 17.701072386058982,
      "grad_norm": 6.09375,
      "learning_rate": 1.985804596446897e-06,
      "loss": 0.965,
      "num_input_tokens_seen": 68948272,
      "step": 118845
    },
    {
      "epoch": 17.70181709859994,
      "grad_norm": 4.34375,
      "learning_rate": 1.984535622836378e-06,
      "loss": 0.9667,
      "num_input_tokens_seen": 68951024,
      "step": 118850
    },
    {
      "epoch": 17.7025618111409,
      "grad_norm": 9.625,
      "learning_rate": 1.983267038048742e-06,
      "loss": 1.2149,
      "num_input_tokens_seen": 68954096,
      "step": 118855
    },
    {
      "epoch": 17.70330652368186,
      "grad_norm": 5.9375,
      "learning_rate": 1.981998842105412e-06,
      "loss": 1.1637,
      "num_input_tokens_seen": 68957136,
      "step": 118860
    },
    {
      "epoch": 17.704051236222817,
      "grad_norm": 3.9375,
      "learning_rate": 1.980731035027822e-06,
      "loss": 0.7533,
      "num_input_tokens_seen": 68959792,
      "step": 118865
    },
    {
      "epoch": 17.704795948763778,
      "grad_norm": 6.6875,
      "learning_rate": 1.979463616837385e-06,
      "loss": 0.8626,
      "num_input_tokens_seen": 68962768,
      "step": 118870
    },
    {
      "epoch": 17.705540661304738,
      "grad_norm": 8.9375,
      "learning_rate": 1.9781965875555087e-06,
      "loss": 1.5641,
      "num_input_tokens_seen": 68965648,
      "step": 118875
    },
    {
      "epoch": 17.706285373845695,
      "grad_norm": 11.3125,
      "learning_rate": 1.976929947203607e-06,
      "loss": 0.7426,
      "num_input_tokens_seen": 68968656,
      "step": 118880
    },
    {
      "epoch": 17.707030086386656,
      "grad_norm": 6.25,
      "learning_rate": 1.9756636958030733e-06,
      "loss": 0.9829,
      "num_input_tokens_seen": 68971472,
      "step": 118885
    },
    {
      "epoch": 17.707774798927613,
      "grad_norm": 9.0,
      "learning_rate": 1.9743978333753023e-06,
      "loss": 0.9545,
      "num_input_tokens_seen": 68975312,
      "step": 118890
    },
    {
      "epoch": 17.708519511468573,
      "grad_norm": 16.75,
      "learning_rate": 1.9731323599416736e-06,
      "loss": 1.1469,
      "num_input_tokens_seen": 68978256,
      "step": 118895
    },
    {
      "epoch": 17.709264224009534,
      "grad_norm": 8.9375,
      "learning_rate": 1.9718672755235728e-06,
      "loss": 1.039,
      "num_input_tokens_seen": 68981040,
      "step": 118900
    },
    {
      "epoch": 17.71000893655049,
      "grad_norm": 13.75,
      "learning_rate": 1.9706025801423666e-06,
      "loss": 1.4573,
      "num_input_tokens_seen": 68984144,
      "step": 118905
    },
    {
      "epoch": 17.71075364909145,
      "grad_norm": 11.0,
      "learning_rate": 1.969338273819429e-06,
      "loss": 1.1525,
      "num_input_tokens_seen": 68986960,
      "step": 118910
    },
    {
      "epoch": 17.711498361632408,
      "grad_norm": 10.125,
      "learning_rate": 1.9680743565761107e-06,
      "loss": 1.1072,
      "num_input_tokens_seen": 68989584,
      "step": 118915
    },
    {
      "epoch": 17.71224307417337,
      "grad_norm": 13.6875,
      "learning_rate": 1.9668108284337654e-06,
      "loss": 0.9596,
      "num_input_tokens_seen": 68992464,
      "step": 118920
    },
    {
      "epoch": 17.71298778671433,
      "grad_norm": 4.34375,
      "learning_rate": 1.9655476894137465e-06,
      "loss": 0.6768,
      "num_input_tokens_seen": 68995184,
      "step": 118925
    },
    {
      "epoch": 17.713732499255286,
      "grad_norm": 15.5,
      "learning_rate": 1.9642849395373836e-06,
      "loss": 1.0422,
      "num_input_tokens_seen": 68997936,
      "step": 118930
    },
    {
      "epoch": 17.714477211796247,
      "grad_norm": 5.40625,
      "learning_rate": 1.963022578826018e-06,
      "loss": 1.086,
      "num_input_tokens_seen": 69000752,
      "step": 118935
    },
    {
      "epoch": 17.715221924337207,
      "grad_norm": 10.25,
      "learning_rate": 1.961760607300972e-06,
      "loss": 0.7549,
      "num_input_tokens_seen": 69003504,
      "step": 118940
    },
    {
      "epoch": 17.715966636878164,
      "grad_norm": 12.125,
      "learning_rate": 1.960499024983564e-06,
      "loss": 1.3763,
      "num_input_tokens_seen": 69006384,
      "step": 118945
    },
    {
      "epoch": 17.716711349419125,
      "grad_norm": 10.875,
      "learning_rate": 1.9592378318951054e-06,
      "loss": 1.1302,
      "num_input_tokens_seen": 69009232,
      "step": 118950
    },
    {
      "epoch": 17.717456061960082,
      "grad_norm": 9.375,
      "learning_rate": 1.957977028056912e-06,
      "loss": 1.2438,
      "num_input_tokens_seen": 69011856,
      "step": 118955
    },
    {
      "epoch": 17.718200774501042,
      "grad_norm": 6.5,
      "learning_rate": 1.9567166134902752e-06,
      "loss": 0.9397,
      "num_input_tokens_seen": 69014576,
      "step": 118960
    },
    {
      "epoch": 17.718945487042003,
      "grad_norm": 12.0625,
      "learning_rate": 1.955456588216489e-06,
      "loss": 1.1527,
      "num_input_tokens_seen": 69017296,
      "step": 118965
    },
    {
      "epoch": 17.71969019958296,
      "grad_norm": 6.03125,
      "learning_rate": 1.9541969522568456e-06,
      "loss": 1.0155,
      "num_input_tokens_seen": 69020176,
      "step": 118970
    },
    {
      "epoch": 17.72043491212392,
      "grad_norm": 10.875,
      "learning_rate": 1.9529377056326183e-06,
      "loss": 1.1394,
      "num_input_tokens_seen": 69022992,
      "step": 118975
    },
    {
      "epoch": 17.72117962466488,
      "grad_norm": 21.625,
      "learning_rate": 1.951678848365088e-06,
      "loss": 1.0785,
      "num_input_tokens_seen": 69026064,
      "step": 118980
    },
    {
      "epoch": 17.721924337205838,
      "grad_norm": 15.4375,
      "learning_rate": 1.950420380475515e-06,
      "loss": 1.3771,
      "num_input_tokens_seen": 69029072,
      "step": 118985
    },
    {
      "epoch": 17.7226690497468,
      "grad_norm": 15.25,
      "learning_rate": 1.949162301985166e-06,
      "loss": 1.187,
      "num_input_tokens_seen": 69032080,
      "step": 118990
    },
    {
      "epoch": 17.723413762287755,
      "grad_norm": 6.5625,
      "learning_rate": 1.947904612915294e-06,
      "loss": 0.9154,
      "num_input_tokens_seen": 69034928,
      "step": 118995
    },
    {
      "epoch": 17.724158474828716,
      "grad_norm": 7.0,
      "learning_rate": 1.9466473132871392e-06,
      "loss": 1.2389,
      "num_input_tokens_seen": 69037904,
      "step": 119000
    },
    {
      "epoch": 17.724903187369677,
      "grad_norm": 12.4375,
      "learning_rate": 1.945390403121952e-06,
      "loss": 1.0746,
      "num_input_tokens_seen": 69040848,
      "step": 119005
    },
    {
      "epoch": 17.725647899910633,
      "grad_norm": 21.75,
      "learning_rate": 1.944133882440963e-06,
      "loss": 1.1093,
      "num_input_tokens_seen": 69043696,
      "step": 119010
    },
    {
      "epoch": 17.726392612451594,
      "grad_norm": 10.4375,
      "learning_rate": 1.9428777512653957e-06,
      "loss": 1.0319,
      "num_input_tokens_seen": 69046512,
      "step": 119015
    },
    {
      "epoch": 17.727137324992555,
      "grad_norm": 11.375,
      "learning_rate": 1.941622009616473e-06,
      "loss": 0.898,
      "num_input_tokens_seen": 69049232,
      "step": 119020
    },
    {
      "epoch": 17.72788203753351,
      "grad_norm": 10.625,
      "learning_rate": 1.9403666575154163e-06,
      "loss": 1.4454,
      "num_input_tokens_seen": 69052016,
      "step": 119025
    },
    {
      "epoch": 17.728626750074472,
      "grad_norm": 14.4375,
      "learning_rate": 1.9391116949834227e-06,
      "loss": 1.0781,
      "num_input_tokens_seen": 69054864,
      "step": 119030
    },
    {
      "epoch": 17.72937146261543,
      "grad_norm": 9.5,
      "learning_rate": 1.937857122041703e-06,
      "loss": 1.2918,
      "num_input_tokens_seen": 69058000,
      "step": 119035
    },
    {
      "epoch": 17.73011617515639,
      "grad_norm": 10.8125,
      "learning_rate": 1.936602938711449e-06,
      "loss": 1.1502,
      "num_input_tokens_seen": 69060944,
      "step": 119040
    },
    {
      "epoch": 17.73086088769735,
      "grad_norm": 9.75,
      "learning_rate": 1.93534914501384e-06,
      "loss": 1.1804,
      "num_input_tokens_seen": 69063984,
      "step": 119045
    },
    {
      "epoch": 17.731605600238307,
      "grad_norm": 9.875,
      "learning_rate": 1.934095740970074e-06,
      "loss": 1.3169,
      "num_input_tokens_seen": 69066672,
      "step": 119050
    },
    {
      "epoch": 17.732350312779268,
      "grad_norm": 16.25,
      "learning_rate": 1.932842726601311e-06,
      "loss": 1.5564,
      "num_input_tokens_seen": 69069424,
      "step": 119055
    },
    {
      "epoch": 17.733095025320225,
      "grad_norm": 5.03125,
      "learning_rate": 1.931590101928729e-06,
      "loss": 1.4174,
      "num_input_tokens_seen": 69072144,
      "step": 119060
    },
    {
      "epoch": 17.733839737861185,
      "grad_norm": 12.0,
      "learning_rate": 1.9303378669734834e-06,
      "loss": 1.0984,
      "num_input_tokens_seen": 69075088,
      "step": 119065
    },
    {
      "epoch": 17.734584450402146,
      "grad_norm": 10.1875,
      "learning_rate": 1.9290860217567374e-06,
      "loss": 0.9555,
      "num_input_tokens_seen": 69077904,
      "step": 119070
    },
    {
      "epoch": 17.735329162943103,
      "grad_norm": 14.125,
      "learning_rate": 1.9278345662996356e-06,
      "loss": 1.175,
      "num_input_tokens_seen": 69080816,
      "step": 119075
    },
    {
      "epoch": 17.736073875484063,
      "grad_norm": 7.75,
      "learning_rate": 1.926583500623316e-06,
      "loss": 1.0537,
      "num_input_tokens_seen": 69083632,
      "step": 119080
    },
    {
      "epoch": 17.736818588025024,
      "grad_norm": 11.25,
      "learning_rate": 1.925332824748921e-06,
      "loss": 0.955,
      "num_input_tokens_seen": 69086672,
      "step": 119085
    },
    {
      "epoch": 17.73756330056598,
      "grad_norm": 25.5,
      "learning_rate": 1.9240825386975692e-06,
      "loss": 1.3384,
      "num_input_tokens_seen": 69089808,
      "step": 119090
    },
    {
      "epoch": 17.73830801310694,
      "grad_norm": 9.4375,
      "learning_rate": 1.9228326424903966e-06,
      "loss": 1.2499,
      "num_input_tokens_seen": 69092624,
      "step": 119095
    },
    {
      "epoch": 17.7390527256479,
      "grad_norm": 13.75,
      "learning_rate": 1.9215831361485054e-06,
      "loss": 0.7823,
      "num_input_tokens_seen": 69095632,
      "step": 119100
    },
    {
      "epoch": 17.73979743818886,
      "grad_norm": 13.9375,
      "learning_rate": 1.920334019693015e-06,
      "loss": 0.8292,
      "num_input_tokens_seen": 69098256,
      "step": 119105
    },
    {
      "epoch": 17.74054215072982,
      "grad_norm": 16.0,
      "learning_rate": 1.9190852931450204e-06,
      "loss": 1.2267,
      "num_input_tokens_seen": 69101424,
      "step": 119110
    },
    {
      "epoch": 17.741286863270776,
      "grad_norm": 11.5625,
      "learning_rate": 1.917836956525626e-06,
      "loss": 0.948,
      "num_input_tokens_seen": 69104368,
      "step": 119115
    },
    {
      "epoch": 17.742031575811737,
      "grad_norm": 8.0625,
      "learning_rate": 1.916589009855918e-06,
      "loss": 0.9746,
      "num_input_tokens_seen": 69107440,
      "step": 119120
    },
    {
      "epoch": 17.742776288352697,
      "grad_norm": 17.875,
      "learning_rate": 1.9153414531569713e-06,
      "loss": 0.9995,
      "num_input_tokens_seen": 69110256,
      "step": 119125
    },
    {
      "epoch": 17.743521000893654,
      "grad_norm": 6.9375,
      "learning_rate": 1.9140942864498747e-06,
      "loss": 0.8426,
      "num_input_tokens_seen": 69112944,
      "step": 119130
    },
    {
      "epoch": 17.744265713434615,
      "grad_norm": 7.5,
      "learning_rate": 1.912847509755686e-06,
      "loss": 1.0209,
      "num_input_tokens_seen": 69115568,
      "step": 119135
    },
    {
      "epoch": 17.745010425975572,
      "grad_norm": 7.25,
      "learning_rate": 1.911601123095477e-06,
      "loss": 1.1739,
      "num_input_tokens_seen": 69118480,
      "step": 119140
    },
    {
      "epoch": 17.745755138516532,
      "grad_norm": 8.625,
      "learning_rate": 1.910355126490304e-06,
      "loss": 0.9077,
      "num_input_tokens_seen": 69121328,
      "step": 119145
    },
    {
      "epoch": 17.746499851057493,
      "grad_norm": 13.125,
      "learning_rate": 1.909109519961211e-06,
      "loss": 0.9151,
      "num_input_tokens_seen": 69124048,
      "step": 119150
    },
    {
      "epoch": 17.74724456359845,
      "grad_norm": 24.375,
      "learning_rate": 1.907864303529247e-06,
      "loss": 1.2296,
      "num_input_tokens_seen": 69126832,
      "step": 119155
    },
    {
      "epoch": 17.74798927613941,
      "grad_norm": 8.625,
      "learning_rate": 1.9066194772154379e-06,
      "loss": 1.2912,
      "num_input_tokens_seen": 69129840,
      "step": 119160
    },
    {
      "epoch": 17.74873398868037,
      "grad_norm": 6.28125,
      "learning_rate": 1.90537504104083e-06,
      "loss": 1.1166,
      "num_input_tokens_seen": 69132656,
      "step": 119165
    },
    {
      "epoch": 17.749478701221328,
      "grad_norm": 6.0,
      "learning_rate": 1.9041309950264319e-06,
      "loss": 1.226,
      "num_input_tokens_seen": 69135888,
      "step": 119170
    },
    {
      "epoch": 17.75022341376229,
      "grad_norm": 4.875,
      "learning_rate": 1.902887339193271e-06,
      "loss": 1.3773,
      "num_input_tokens_seen": 69138672,
      "step": 119175
    },
    {
      "epoch": 17.750968126303245,
      "grad_norm": 12.0,
      "learning_rate": 1.9016440735623503e-06,
      "loss": 0.8955,
      "num_input_tokens_seen": 69141200,
      "step": 119180
    },
    {
      "epoch": 17.751712838844206,
      "grad_norm": 14.0625,
      "learning_rate": 1.9004011981546804e-06,
      "loss": 1.1606,
      "num_input_tokens_seen": 69144304,
      "step": 119185
    },
    {
      "epoch": 17.752457551385167,
      "grad_norm": 9.1875,
      "learning_rate": 1.8991587129912531e-06,
      "loss": 1.3875,
      "num_input_tokens_seen": 69147376,
      "step": 119190
    },
    {
      "epoch": 17.753202263926124,
      "grad_norm": 10.75,
      "learning_rate": 1.8979166180930625e-06,
      "loss": 1.1409,
      "num_input_tokens_seen": 69150224,
      "step": 119195
    },
    {
      "epoch": 17.753946976467084,
      "grad_norm": 15.3125,
      "learning_rate": 1.896674913481089e-06,
      "loss": 1.1386,
      "num_input_tokens_seen": 69153008,
      "step": 119200
    },
    {
      "epoch": 17.75469168900804,
      "grad_norm": 13.875,
      "learning_rate": 1.8954335991763107e-06,
      "loss": 1.6452,
      "num_input_tokens_seen": 69156176,
      "step": 119205
    },
    {
      "epoch": 17.755436401549,
      "grad_norm": 6.15625,
      "learning_rate": 1.8941926751997018e-06,
      "loss": 0.9841,
      "num_input_tokens_seen": 69158864,
      "step": 119210
    },
    {
      "epoch": 17.756181114089962,
      "grad_norm": 14.1875,
      "learning_rate": 1.8929521415722267e-06,
      "loss": 0.9929,
      "num_input_tokens_seen": 69161776,
      "step": 119215
    },
    {
      "epoch": 17.75692582663092,
      "grad_norm": 5.90625,
      "learning_rate": 1.8917119983148378e-06,
      "loss": 1.1058,
      "num_input_tokens_seen": 69164944,
      "step": 119220
    },
    {
      "epoch": 17.75767053917188,
      "grad_norm": 11.5625,
      "learning_rate": 1.8904722454484825e-06,
      "loss": 1.1001,
      "num_input_tokens_seen": 69167728,
      "step": 119225
    },
    {
      "epoch": 17.75841525171284,
      "grad_norm": 13.3125,
      "learning_rate": 1.8892328829941186e-06,
      "loss": 1.3446,
      "num_input_tokens_seen": 69170576,
      "step": 119230
    },
    {
      "epoch": 17.759159964253797,
      "grad_norm": 9.125,
      "learning_rate": 1.8879939109726713e-06,
      "loss": 1.2861,
      "num_input_tokens_seen": 69173296,
      "step": 119235
    },
    {
      "epoch": 17.759904676794758,
      "grad_norm": 8.5625,
      "learning_rate": 1.8867553294050795e-06,
      "loss": 0.9003,
      "num_input_tokens_seen": 69176304,
      "step": 119240
    },
    {
      "epoch": 17.76064938933572,
      "grad_norm": 6.75,
      "learning_rate": 1.8855171383122677e-06,
      "loss": 0.8223,
      "num_input_tokens_seen": 69179152,
      "step": 119245
    },
    {
      "epoch": 17.761394101876675,
      "grad_norm": 15.125,
      "learning_rate": 1.8842793377151446e-06,
      "loss": 1.5932,
      "num_input_tokens_seen": 69182032,
      "step": 119250
    },
    {
      "epoch": 17.762138814417636,
      "grad_norm": 17.5,
      "learning_rate": 1.8830419276346352e-06,
      "loss": 1.404,
      "num_input_tokens_seen": 69184976,
      "step": 119255
    },
    {
      "epoch": 17.762883526958593,
      "grad_norm": 21.5,
      "learning_rate": 1.8818049080916305e-06,
      "loss": 1.3586,
      "num_input_tokens_seen": 69187856,
      "step": 119260
    },
    {
      "epoch": 17.763628239499553,
      "grad_norm": 12.375,
      "learning_rate": 1.8805682791070422e-06,
      "loss": 1.0827,
      "num_input_tokens_seen": 69190768,
      "step": 119265
    },
    {
      "epoch": 17.764372952040514,
      "grad_norm": 5.28125,
      "learning_rate": 1.8793320407017534e-06,
      "loss": 1.0332,
      "num_input_tokens_seen": 69193776,
      "step": 119270
    },
    {
      "epoch": 17.76511766458147,
      "grad_norm": 10.9375,
      "learning_rate": 1.8780961928966528e-06,
      "loss": 1.3151,
      "num_input_tokens_seen": 69196752,
      "step": 119275
    },
    {
      "epoch": 17.76586237712243,
      "grad_norm": 9.75,
      "learning_rate": 1.8768607357126128e-06,
      "loss": 0.9347,
      "num_input_tokens_seen": 69199824,
      "step": 119280
    },
    {
      "epoch": 17.76660708966339,
      "grad_norm": 11.8125,
      "learning_rate": 1.875625669170511e-06,
      "loss": 1.3368,
      "num_input_tokens_seen": 69202608,
      "step": 119285
    },
    {
      "epoch": 17.76735180220435,
      "grad_norm": 8.8125,
      "learning_rate": 1.874390993291214e-06,
      "loss": 1.1181,
      "num_input_tokens_seen": 69205392,
      "step": 119290
    },
    {
      "epoch": 17.76809651474531,
      "grad_norm": 7.375,
      "learning_rate": 1.8731567080955692e-06,
      "loss": 0.9661,
      "num_input_tokens_seen": 69208240,
      "step": 119295
    },
    {
      "epoch": 17.768841227286266,
      "grad_norm": 10.9375,
      "learning_rate": 1.871922813604443e-06,
      "loss": 0.9818,
      "num_input_tokens_seen": 69211056,
      "step": 119300
    },
    {
      "epoch": 17.769585939827227,
      "grad_norm": 14.0625,
      "learning_rate": 1.870689309838672e-06,
      "loss": 0.8956,
      "num_input_tokens_seen": 69213872,
      "step": 119305
    },
    {
      "epoch": 17.770330652368187,
      "grad_norm": 12.0,
      "learning_rate": 1.8694561968191e-06,
      "loss": 1.0891,
      "num_input_tokens_seen": 69216816,
      "step": 119310
    },
    {
      "epoch": 17.771075364909144,
      "grad_norm": 6.625,
      "learning_rate": 1.8682234745665522e-06,
      "loss": 0.7394,
      "num_input_tokens_seen": 69219824,
      "step": 119315
    },
    {
      "epoch": 17.771820077450105,
      "grad_norm": 18.75,
      "learning_rate": 1.866991143101865e-06,
      "loss": 1.0688,
      "num_input_tokens_seen": 69222992,
      "step": 119320
    },
    {
      "epoch": 17.772564789991062,
      "grad_norm": 12.375,
      "learning_rate": 1.8657592024458491e-06,
      "loss": 0.999,
      "num_input_tokens_seen": 69225840,
      "step": 119325
    },
    {
      "epoch": 17.773309502532022,
      "grad_norm": 6.15625,
      "learning_rate": 1.8645276526193162e-06,
      "loss": 1.0577,
      "num_input_tokens_seen": 69228656,
      "step": 119330
    },
    {
      "epoch": 17.774054215072983,
      "grad_norm": 6.6875,
      "learning_rate": 1.8632964936430768e-06,
      "loss": 0.9075,
      "num_input_tokens_seen": 69231312,
      "step": 119335
    },
    {
      "epoch": 17.77479892761394,
      "grad_norm": 11.4375,
      "learning_rate": 1.8620657255379314e-06,
      "loss": 1.0688,
      "num_input_tokens_seen": 69234576,
      "step": 119340
    },
    {
      "epoch": 17.7755436401549,
      "grad_norm": 12.75,
      "learning_rate": 1.860835348324666e-06,
      "loss": 1.27,
      "num_input_tokens_seen": 69237520,
      "step": 119345
    },
    {
      "epoch": 17.77628835269586,
      "grad_norm": 9.8125,
      "learning_rate": 1.8596053620240667e-06,
      "loss": 0.9972,
      "num_input_tokens_seen": 69240528,
      "step": 119350
    },
    {
      "epoch": 17.777033065236818,
      "grad_norm": 21.625,
      "learning_rate": 1.8583757666569196e-06,
      "loss": 1.0663,
      "num_input_tokens_seen": 69243760,
      "step": 119355
    },
    {
      "epoch": 17.77777777777778,
      "grad_norm": 9.6875,
      "learning_rate": 1.8571465622439943e-06,
      "loss": 0.9718,
      "num_input_tokens_seen": 69246704,
      "step": 119360
    },
    {
      "epoch": 17.778522490318736,
      "grad_norm": 5.71875,
      "learning_rate": 1.8559177488060547e-06,
      "loss": 1.0654,
      "num_input_tokens_seen": 69249584,
      "step": 119365
    },
    {
      "epoch": 17.779267202859696,
      "grad_norm": 16.75,
      "learning_rate": 1.854689326363862e-06,
      "loss": 0.948,
      "num_input_tokens_seen": 69252272,
      "step": 119370
    },
    {
      "epoch": 17.780011915400657,
      "grad_norm": 14.9375,
      "learning_rate": 1.8534612949381691e-06,
      "loss": 1.2896,
      "num_input_tokens_seen": 69254928,
      "step": 119375
    },
    {
      "epoch": 17.780756627941614,
      "grad_norm": 15.25,
      "learning_rate": 1.8522336545497232e-06,
      "loss": 0.9854,
      "num_input_tokens_seen": 69257840,
      "step": 119380
    },
    {
      "epoch": 17.781501340482574,
      "grad_norm": 10.9375,
      "learning_rate": 1.8510064052192604e-06,
      "loss": 1.0209,
      "num_input_tokens_seen": 69260688,
      "step": 119385
    },
    {
      "epoch": 17.782246053023535,
      "grad_norm": 11.4375,
      "learning_rate": 1.8497795469675227e-06,
      "loss": 1.2406,
      "num_input_tokens_seen": 69263504,
      "step": 119390
    },
    {
      "epoch": 17.78299076556449,
      "grad_norm": 12.5625,
      "learning_rate": 1.848553079815224e-06,
      "loss": 1.1649,
      "num_input_tokens_seen": 69266416,
      "step": 119395
    },
    {
      "epoch": 17.783735478105452,
      "grad_norm": 10.875,
      "learning_rate": 1.8473270037830975e-06,
      "loss": 1.0871,
      "num_input_tokens_seen": 69269488,
      "step": 119400
    },
    {
      "epoch": 17.78448019064641,
      "grad_norm": 12.0,
      "learning_rate": 1.8461013188918492e-06,
      "loss": 1.5787,
      "num_input_tokens_seen": 69272272,
      "step": 119405
    },
    {
      "epoch": 17.78522490318737,
      "grad_norm": 9.625,
      "learning_rate": 1.8448760251621844e-06,
      "loss": 1.1315,
      "num_input_tokens_seen": 69275216,
      "step": 119410
    },
    {
      "epoch": 17.78596961572833,
      "grad_norm": 9.6875,
      "learning_rate": 1.843651122614809e-06,
      "loss": 1.2439,
      "num_input_tokens_seen": 69278224,
      "step": 119415
    },
    {
      "epoch": 17.786714328269287,
      "grad_norm": 7.15625,
      "learning_rate": 1.8424266112704064e-06,
      "loss": 1.0435,
      "num_input_tokens_seen": 69281392,
      "step": 119420
    },
    {
      "epoch": 17.787459040810248,
      "grad_norm": 11.625,
      "learning_rate": 1.841202491149674e-06,
      "loss": 1.0533,
      "num_input_tokens_seen": 69284912,
      "step": 119425
    },
    {
      "epoch": 17.788203753351205,
      "grad_norm": 12.3125,
      "learning_rate": 1.839978762273284e-06,
      "loss": 0.8173,
      "num_input_tokens_seen": 69287984,
      "step": 119430
    },
    {
      "epoch": 17.788948465892165,
      "grad_norm": 10.4375,
      "learning_rate": 1.838755424661917e-06,
      "loss": 1.1347,
      "num_input_tokens_seen": 69290736,
      "step": 119435
    },
    {
      "epoch": 17.789693178433126,
      "grad_norm": 10.8125,
      "learning_rate": 1.8375324783362402e-06,
      "loss": 1.0967,
      "num_input_tokens_seen": 69293712,
      "step": 119440
    },
    {
      "epoch": 17.790437890974083,
      "grad_norm": 10.4375,
      "learning_rate": 1.8363099233169034e-06,
      "loss": 0.9787,
      "num_input_tokens_seen": 69296592,
      "step": 119445
    },
    {
      "epoch": 17.791182603515043,
      "grad_norm": 10.4375,
      "learning_rate": 1.8350877596245735e-06,
      "loss": 1.0077,
      "num_input_tokens_seen": 69299664,
      "step": 119450
    },
    {
      "epoch": 17.791927316056004,
      "grad_norm": 7.25,
      "learning_rate": 1.8338659872798896e-06,
      "loss": 0.8389,
      "num_input_tokens_seen": 69302512,
      "step": 119455
    },
    {
      "epoch": 17.79267202859696,
      "grad_norm": 11.75,
      "learning_rate": 1.8326446063034964e-06,
      "loss": 0.8054,
      "num_input_tokens_seen": 69305168,
      "step": 119460
    },
    {
      "epoch": 17.79341674113792,
      "grad_norm": 9.9375,
      "learning_rate": 1.8314236167160243e-06,
      "loss": 0.9361,
      "num_input_tokens_seen": 69308304,
      "step": 119465
    },
    {
      "epoch": 17.79416145367888,
      "grad_norm": 12.75,
      "learning_rate": 1.8302030185381042e-06,
      "loss": 1.0139,
      "num_input_tokens_seen": 69310928,
      "step": 119470
    },
    {
      "epoch": 17.79490616621984,
      "grad_norm": 18.125,
      "learning_rate": 1.8289828117903584e-06,
      "loss": 1.1394,
      "num_input_tokens_seen": 69313584,
      "step": 119475
    },
    {
      "epoch": 17.7956508787608,
      "grad_norm": 11.0,
      "learning_rate": 1.8277629964933958e-06,
      "loss": 1.0749,
      "num_input_tokens_seen": 69316464,
      "step": 119480
    },
    {
      "epoch": 17.796395591301756,
      "grad_norm": 7.75,
      "learning_rate": 1.8265435726678271e-06,
      "loss": 1.4268,
      "num_input_tokens_seen": 69319600,
      "step": 119485
    },
    {
      "epoch": 17.797140303842717,
      "grad_norm": 17.125,
      "learning_rate": 1.8253245403342472e-06,
      "loss": 1.0515,
      "num_input_tokens_seen": 69322512,
      "step": 119490
    },
    {
      "epoch": 17.797885016383677,
      "grad_norm": 15.125,
      "learning_rate": 1.824105899513262e-06,
      "loss": 1.1154,
      "num_input_tokens_seen": 69325392,
      "step": 119495
    },
    {
      "epoch": 17.798629728924634,
      "grad_norm": 15.9375,
      "learning_rate": 1.8228876502254465e-06,
      "loss": 1.5065,
      "num_input_tokens_seen": 69328560,
      "step": 119500
    },
    {
      "epoch": 17.799374441465595,
      "grad_norm": 7.5625,
      "learning_rate": 1.8216697924913928e-06,
      "loss": 1.0544,
      "num_input_tokens_seen": 69331472,
      "step": 119505
    },
    {
      "epoch": 17.800119154006552,
      "grad_norm": 9.5625,
      "learning_rate": 1.8204523263316647e-06,
      "loss": 1.222,
      "num_input_tokens_seen": 69334448,
      "step": 119510
    },
    {
      "epoch": 17.800863866547513,
      "grad_norm": 19.375,
      "learning_rate": 1.8192352517668432e-06,
      "loss": 1.4705,
      "num_input_tokens_seen": 69337456,
      "step": 119515
    },
    {
      "epoch": 17.801608579088473,
      "grad_norm": 9.75,
      "learning_rate": 1.818018568817481e-06,
      "loss": 1.0937,
      "num_input_tokens_seen": 69340272,
      "step": 119520
    },
    {
      "epoch": 17.80235329162943,
      "grad_norm": 22.375,
      "learning_rate": 1.8168022775041288e-06,
      "loss": 1.5509,
      "num_input_tokens_seen": 69342928,
      "step": 119525
    },
    {
      "epoch": 17.80309800417039,
      "grad_norm": 11.375,
      "learning_rate": 1.8155863778473447e-06,
      "loss": 0.9454,
      "num_input_tokens_seen": 69346224,
      "step": 119530
    },
    {
      "epoch": 17.80384271671135,
      "grad_norm": 18.125,
      "learning_rate": 1.8143708698676597e-06,
      "loss": 1.5515,
      "num_input_tokens_seen": 69348912,
      "step": 119535
    },
    {
      "epoch": 17.804587429252308,
      "grad_norm": 7.0625,
      "learning_rate": 1.8131557535856214e-06,
      "loss": 0.9337,
      "num_input_tokens_seen": 69351632,
      "step": 119540
    },
    {
      "epoch": 17.80533214179327,
      "grad_norm": 12.875,
      "learning_rate": 1.8119410290217465e-06,
      "loss": 0.9758,
      "num_input_tokens_seen": 69354704,
      "step": 119545
    },
    {
      "epoch": 17.806076854334226,
      "grad_norm": 11.6875,
      "learning_rate": 1.810726696196563e-06,
      "loss": 1.0255,
      "num_input_tokens_seen": 69357456,
      "step": 119550
    },
    {
      "epoch": 17.806821566875186,
      "grad_norm": 4.40625,
      "learning_rate": 1.8095127551305797e-06,
      "loss": 1.058,
      "num_input_tokens_seen": 69360176,
      "step": 119555
    },
    {
      "epoch": 17.807566279416147,
      "grad_norm": 8.6875,
      "learning_rate": 1.8082992058443132e-06,
      "loss": 1.227,
      "num_input_tokens_seen": 69363376,
      "step": 119560
    },
    {
      "epoch": 17.808310991957104,
      "grad_norm": 8.5,
      "learning_rate": 1.8070860483582585e-06,
      "loss": 0.8674,
      "num_input_tokens_seen": 69366352,
      "step": 119565
    },
    {
      "epoch": 17.809055704498064,
      "grad_norm": 9.625,
      "learning_rate": 1.8058732826929104e-06,
      "loss": 1.0511,
      "num_input_tokens_seen": 69369296,
      "step": 119570
    },
    {
      "epoch": 17.80980041703902,
      "grad_norm": 10.0625,
      "learning_rate": 1.8046609088687633e-06,
      "loss": 0.928,
      "num_input_tokens_seen": 69371984,
      "step": 119575
    },
    {
      "epoch": 17.81054512957998,
      "grad_norm": 10.4375,
      "learning_rate": 1.8034489269062899e-06,
      "loss": 1.4863,
      "num_input_tokens_seen": 69374800,
      "step": 119580
    },
    {
      "epoch": 17.811289842120942,
      "grad_norm": 8.75,
      "learning_rate": 1.8022373368259765e-06,
      "loss": 0.9384,
      "num_input_tokens_seen": 69377808,
      "step": 119585
    },
    {
      "epoch": 17.8120345546619,
      "grad_norm": 4.90625,
      "learning_rate": 1.801026138648282e-06,
      "loss": 1.3549,
      "num_input_tokens_seen": 69380688,
      "step": 119590
    },
    {
      "epoch": 17.81277926720286,
      "grad_norm": 5.65625,
      "learning_rate": 1.7998153323936755e-06,
      "loss": 1.0179,
      "num_input_tokens_seen": 69383760,
      "step": 119595
    },
    {
      "epoch": 17.81352397974382,
      "grad_norm": 12.125,
      "learning_rate": 1.798604918082611e-06,
      "loss": 1.1612,
      "num_input_tokens_seen": 69386512,
      "step": 119600
    },
    {
      "epoch": 17.814268692284777,
      "grad_norm": 8.625,
      "learning_rate": 1.7973948957355352e-06,
      "loss": 1.1367,
      "num_input_tokens_seen": 69389136,
      "step": 119605
    },
    {
      "epoch": 17.815013404825738,
      "grad_norm": 6.53125,
      "learning_rate": 1.796185265372885e-06,
      "loss": 1.09,
      "num_input_tokens_seen": 69391984,
      "step": 119610
    },
    {
      "epoch": 17.815758117366695,
      "grad_norm": 7.8125,
      "learning_rate": 1.7949760270151078e-06,
      "loss": 1.0323,
      "num_input_tokens_seen": 69394928,
      "step": 119615
    },
    {
      "epoch": 17.816502829907655,
      "grad_norm": 10.375,
      "learning_rate": 1.7937671806826262e-06,
      "loss": 1.1544,
      "num_input_tokens_seen": 69397776,
      "step": 119620
    },
    {
      "epoch": 17.817247542448616,
      "grad_norm": 18.5,
      "learning_rate": 1.792558726395857e-06,
      "loss": 1.055,
      "num_input_tokens_seen": 69400496,
      "step": 119625
    },
    {
      "epoch": 17.817992254989573,
      "grad_norm": 6.3125,
      "learning_rate": 1.791350664175223e-06,
      "loss": 0.894,
      "num_input_tokens_seen": 69403440,
      "step": 119630
    },
    {
      "epoch": 17.818736967530533,
      "grad_norm": 12.3125,
      "learning_rate": 1.7901429940411301e-06,
      "loss": 1.2126,
      "num_input_tokens_seen": 69406448,
      "step": 119635
    },
    {
      "epoch": 17.819481680071494,
      "grad_norm": 13.3125,
      "learning_rate": 1.788935716013987e-06,
      "loss": 1.2173,
      "num_input_tokens_seen": 69409232,
      "step": 119640
    },
    {
      "epoch": 17.82022639261245,
      "grad_norm": 7.0,
      "learning_rate": 1.7877288301141826e-06,
      "loss": 0.8669,
      "num_input_tokens_seen": 69412112,
      "step": 119645
    },
    {
      "epoch": 17.82097110515341,
      "grad_norm": 7.875,
      "learning_rate": 1.7865223363621037e-06,
      "loss": 0.8514,
      "num_input_tokens_seen": 69414960,
      "step": 119650
    },
    {
      "epoch": 17.82171581769437,
      "grad_norm": 7.15625,
      "learning_rate": 1.7853162347781394e-06,
      "loss": 1.0167,
      "num_input_tokens_seen": 69417904,
      "step": 119655
    },
    {
      "epoch": 17.82246053023533,
      "grad_norm": 26.875,
      "learning_rate": 1.7841105253826596e-06,
      "loss": 1.1342,
      "num_input_tokens_seen": 69420752,
      "step": 119660
    },
    {
      "epoch": 17.82320524277629,
      "grad_norm": 10.8125,
      "learning_rate": 1.7829052081960423e-06,
      "loss": 0.9232,
      "num_input_tokens_seen": 69423568,
      "step": 119665
    },
    {
      "epoch": 17.823949955317246,
      "grad_norm": 12.375,
      "learning_rate": 1.7817002832386436e-06,
      "loss": 0.9984,
      "num_input_tokens_seen": 69426352,
      "step": 119670
    },
    {
      "epoch": 17.824694667858207,
      "grad_norm": 15.75,
      "learning_rate": 1.7804957505308224e-06,
      "loss": 0.8512,
      "num_input_tokens_seen": 69429424,
      "step": 119675
    },
    {
      "epoch": 17.825439380399168,
      "grad_norm": 9.8125,
      "learning_rate": 1.7792916100929258e-06,
      "loss": 0.7288,
      "num_input_tokens_seen": 69432400,
      "step": 119680
    },
    {
      "epoch": 17.826184092940125,
      "grad_norm": 8.5,
      "learning_rate": 1.7780878619452905e-06,
      "loss": 1.0358,
      "num_input_tokens_seen": 69435024,
      "step": 119685
    },
    {
      "epoch": 17.826928805481085,
      "grad_norm": 15.5625,
      "learning_rate": 1.7768845061082646e-06,
      "loss": 1.276,
      "num_input_tokens_seen": 69437776,
      "step": 119690
    },
    {
      "epoch": 17.827673518022042,
      "grad_norm": 7.25,
      "learning_rate": 1.7756815426021673e-06,
      "loss": 1.0068,
      "num_input_tokens_seen": 69440656,
      "step": 119695
    },
    {
      "epoch": 17.828418230563003,
      "grad_norm": 8.25,
      "learning_rate": 1.7744789714473325e-06,
      "loss": 1.357,
      "num_input_tokens_seen": 69443376,
      "step": 119700
    },
    {
      "epoch": 17.829162943103963,
      "grad_norm": 12.0,
      "learning_rate": 1.7732767926640636e-06,
      "loss": 1.125,
      "num_input_tokens_seen": 69446544,
      "step": 119705
    },
    {
      "epoch": 17.82990765564492,
      "grad_norm": 10.0,
      "learning_rate": 1.7720750062726831e-06,
      "loss": 0.9574,
      "num_input_tokens_seen": 69449360,
      "step": 119710
    },
    {
      "epoch": 17.83065236818588,
      "grad_norm": 11.8125,
      "learning_rate": 1.7708736122934805e-06,
      "loss": 1.3089,
      "num_input_tokens_seen": 69452272,
      "step": 119715
    },
    {
      "epoch": 17.83139708072684,
      "grad_norm": 7.625,
      "learning_rate": 1.7696726107467643e-06,
      "loss": 0.885,
      "num_input_tokens_seen": 69455088,
      "step": 119720
    },
    {
      "epoch": 17.832141793267798,
      "grad_norm": 7.375,
      "learning_rate": 1.768472001652821e-06,
      "loss": 1.0705,
      "num_input_tokens_seen": 69458032,
      "step": 119725
    },
    {
      "epoch": 17.83288650580876,
      "grad_norm": 13.8125,
      "learning_rate": 1.7672717850319264e-06,
      "loss": 0.8421,
      "num_input_tokens_seen": 69460720,
      "step": 119730
    },
    {
      "epoch": 17.833631218349716,
      "grad_norm": 4.25,
      "learning_rate": 1.766071960904367e-06,
      "loss": 1.0838,
      "num_input_tokens_seen": 69463536,
      "step": 119735
    },
    {
      "epoch": 17.834375930890676,
      "grad_norm": 21.5,
      "learning_rate": 1.7648725292904067e-06,
      "loss": 1.0208,
      "num_input_tokens_seen": 69466416,
      "step": 119740
    },
    {
      "epoch": 17.835120643431637,
      "grad_norm": 8.1875,
      "learning_rate": 1.7636734902103102e-06,
      "loss": 1.0011,
      "num_input_tokens_seen": 69469520,
      "step": 119745
    },
    {
      "epoch": 17.835865355972594,
      "grad_norm": 6.65625,
      "learning_rate": 1.7624748436843308e-06,
      "loss": 0.9567,
      "num_input_tokens_seen": 69472464,
      "step": 119750
    },
    {
      "epoch": 17.836610068513554,
      "grad_norm": 6.8125,
      "learning_rate": 1.7612765897327244e-06,
      "loss": 0.7906,
      "num_input_tokens_seen": 69475216,
      "step": 119755
    },
    {
      "epoch": 17.837354781054515,
      "grad_norm": 9.4375,
      "learning_rate": 1.7600787283757303e-06,
      "loss": 1.1117,
      "num_input_tokens_seen": 69477904,
      "step": 119760
    },
    {
      "epoch": 17.83809949359547,
      "grad_norm": 12.125,
      "learning_rate": 1.7588812596335824e-06,
      "loss": 1.1248,
      "num_input_tokens_seen": 69480688,
      "step": 119765
    },
    {
      "epoch": 17.838844206136432,
      "grad_norm": 14.25,
      "learning_rate": 1.7576841835265202e-06,
      "loss": 0.7605,
      "num_input_tokens_seen": 69483696,
      "step": 119770
    },
    {
      "epoch": 17.83958891867739,
      "grad_norm": 12.0625,
      "learning_rate": 1.756487500074755e-06,
      "loss": 1.1331,
      "num_input_tokens_seen": 69486256,
      "step": 119775
    },
    {
      "epoch": 17.84033363121835,
      "grad_norm": 8.8125,
      "learning_rate": 1.7552912092985153e-06,
      "loss": 1.0737,
      "num_input_tokens_seen": 69489072,
      "step": 119780
    },
    {
      "epoch": 17.84107834375931,
      "grad_norm": 14.5625,
      "learning_rate": 1.7540953112180014e-06,
      "loss": 0.9934,
      "num_input_tokens_seen": 69491920,
      "step": 119785
    },
    {
      "epoch": 17.841823056300267,
      "grad_norm": 9.9375,
      "learning_rate": 1.752899805853425e-06,
      "loss": 1.0977,
      "num_input_tokens_seen": 69494992,
      "step": 119790
    },
    {
      "epoch": 17.842567768841228,
      "grad_norm": 11.5,
      "learning_rate": 1.7517046932249758e-06,
      "loss": 1.0805,
      "num_input_tokens_seen": 69497968,
      "step": 119795
    },
    {
      "epoch": 17.843312481382185,
      "grad_norm": 9.375,
      "learning_rate": 1.7505099733528514e-06,
      "loss": 0.8332,
      "num_input_tokens_seen": 69500944,
      "step": 119800
    },
    {
      "epoch": 17.844057193923145,
      "grad_norm": 7.15625,
      "learning_rate": 1.7493156462572296e-06,
      "loss": 1.1458,
      "num_input_tokens_seen": 69503920,
      "step": 119805
    },
    {
      "epoch": 17.844801906464106,
      "grad_norm": 8.6875,
      "learning_rate": 1.7481217119582921e-06,
      "loss": 1.1561,
      "num_input_tokens_seen": 69506864,
      "step": 119810
    },
    {
      "epoch": 17.845546619005063,
      "grad_norm": 10.0625,
      "learning_rate": 1.746928170476203e-06,
      "loss": 1.2657,
      "num_input_tokens_seen": 69509936,
      "step": 119815
    },
    {
      "epoch": 17.846291331546023,
      "grad_norm": 6.5,
      "learning_rate": 1.7457350218311269e-06,
      "loss": 1.2938,
      "num_input_tokens_seen": 69512912,
      "step": 119820
    },
    {
      "epoch": 17.847036044086984,
      "grad_norm": 9.5625,
      "learning_rate": 1.7445422660432254e-06,
      "loss": 1.0616,
      "num_input_tokens_seen": 69515792,
      "step": 119825
    },
    {
      "epoch": 17.84778075662794,
      "grad_norm": 12.6875,
      "learning_rate": 1.7433499031326434e-06,
      "loss": 0.9665,
      "num_input_tokens_seen": 69518704,
      "step": 119830
    },
    {
      "epoch": 17.8485254691689,
      "grad_norm": 17.375,
      "learning_rate": 1.7421579331195314e-06,
      "loss": 1.0092,
      "num_input_tokens_seen": 69521776,
      "step": 119835
    },
    {
      "epoch": 17.84927018170986,
      "grad_norm": 15.875,
      "learning_rate": 1.7409663560240209e-06,
      "loss": 1.3964,
      "num_input_tokens_seen": 69524368,
      "step": 119840
    },
    {
      "epoch": 17.85001489425082,
      "grad_norm": 9.5625,
      "learning_rate": 1.7397751718662452e-06,
      "loss": 1.2129,
      "num_input_tokens_seen": 69527472,
      "step": 119845
    },
    {
      "epoch": 17.85075960679178,
      "grad_norm": 13.625,
      "learning_rate": 1.7385843806663304e-06,
      "loss": 0.8877,
      "num_input_tokens_seen": 69530640,
      "step": 119850
    },
    {
      "epoch": 17.851504319332737,
      "grad_norm": 15.0625,
      "learning_rate": 1.7373939824443853e-06,
      "loss": 1.3361,
      "num_input_tokens_seen": 69533616,
      "step": 119855
    },
    {
      "epoch": 17.852249031873697,
      "grad_norm": 9.6875,
      "learning_rate": 1.7362039772205296e-06,
      "loss": 1.0125,
      "num_input_tokens_seen": 69537040,
      "step": 119860
    },
    {
      "epoch": 17.852993744414658,
      "grad_norm": 8.125,
      "learning_rate": 1.7350143650148587e-06,
      "loss": 1.0169,
      "num_input_tokens_seen": 69540080,
      "step": 119865
    },
    {
      "epoch": 17.853738456955615,
      "grad_norm": 12.0625,
      "learning_rate": 1.7338251458474786e-06,
      "loss": 1.2809,
      "num_input_tokens_seen": 69542896,
      "step": 119870
    },
    {
      "epoch": 17.854483169496575,
      "grad_norm": 7.875,
      "learning_rate": 1.7326363197384788e-06,
      "loss": 1.1127,
      "num_input_tokens_seen": 69546000,
      "step": 119875
    },
    {
      "epoch": 17.855227882037532,
      "grad_norm": 9.125,
      "learning_rate": 1.7314478867079376e-06,
      "loss": 1.0612,
      "num_input_tokens_seen": 69548688,
      "step": 119880
    },
    {
      "epoch": 17.855972594578493,
      "grad_norm": 13.0,
      "learning_rate": 1.7302598467759362e-06,
      "loss": 1.3468,
      "num_input_tokens_seen": 69551664,
      "step": 119885
    },
    {
      "epoch": 17.856717307119453,
      "grad_norm": 10.375,
      "learning_rate": 1.729072199962542e-06,
      "loss": 0.7381,
      "num_input_tokens_seen": 69554640,
      "step": 119890
    },
    {
      "epoch": 17.85746201966041,
      "grad_norm": 14.25,
      "learning_rate": 1.7278849462878223e-06,
      "loss": 1.1541,
      "num_input_tokens_seen": 69557520,
      "step": 119895
    },
    {
      "epoch": 17.85820673220137,
      "grad_norm": 9.125,
      "learning_rate": 1.7266980857718328e-06,
      "loss": 1.0795,
      "num_input_tokens_seen": 69560624,
      "step": 119900
    },
    {
      "epoch": 17.85895144474233,
      "grad_norm": 8.75,
      "learning_rate": 1.7255116184346277e-06,
      "loss": 1.1043,
      "num_input_tokens_seen": 69563728,
      "step": 119905
    },
    {
      "epoch": 17.859696157283288,
      "grad_norm": 14.75,
      "learning_rate": 1.724325544296243e-06,
      "loss": 0.8397,
      "num_input_tokens_seen": 69566704,
      "step": 119910
    },
    {
      "epoch": 17.86044086982425,
      "grad_norm": 8.875,
      "learning_rate": 1.7231398633767272e-06,
      "loss": 0.9151,
      "num_input_tokens_seen": 69569296,
      "step": 119915
    },
    {
      "epoch": 17.861185582365206,
      "grad_norm": 14.625,
      "learning_rate": 1.7219545756961025e-06,
      "loss": 0.7633,
      "num_input_tokens_seen": 69572048,
      "step": 119920
    },
    {
      "epoch": 17.861930294906166,
      "grad_norm": 6.78125,
      "learning_rate": 1.7207696812744007e-06,
      "loss": 0.7984,
      "num_input_tokens_seen": 69574640,
      "step": 119925
    },
    {
      "epoch": 17.862675007447127,
      "grad_norm": 9.1875,
      "learning_rate": 1.719585180131636e-06,
      "loss": 1.262,
      "num_input_tokens_seen": 69577488,
      "step": 119930
    },
    {
      "epoch": 17.863419719988084,
      "grad_norm": 9.125,
      "learning_rate": 1.7184010722878146e-06,
      "loss": 0.9534,
      "num_input_tokens_seen": 69580496,
      "step": 119935
    },
    {
      "epoch": 17.864164432529044,
      "grad_norm": 14.25,
      "learning_rate": 1.7172173577629459e-06,
      "loss": 0.8042,
      "num_input_tokens_seen": 69583568,
      "step": 119940
    },
    {
      "epoch": 17.86490914507,
      "grad_norm": 10.625,
      "learning_rate": 1.7160340365770272e-06,
      "loss": 0.8671,
      "num_input_tokens_seen": 69586480,
      "step": 119945
    },
    {
      "epoch": 17.865653857610962,
      "grad_norm": 6.15625,
      "learning_rate": 1.7148511087500485e-06,
      "loss": 0.9531,
      "num_input_tokens_seen": 69589168,
      "step": 119950
    },
    {
      "epoch": 17.866398570151922,
      "grad_norm": 6.3125,
      "learning_rate": 1.7136685743019909e-06,
      "loss": 1.0109,
      "num_input_tokens_seen": 69592176,
      "step": 119955
    },
    {
      "epoch": 17.86714328269288,
      "grad_norm": 8.3125,
      "learning_rate": 1.7124864332528412e-06,
      "loss": 1.2237,
      "num_input_tokens_seen": 69595152,
      "step": 119960
    },
    {
      "epoch": 17.86788799523384,
      "grad_norm": 7.40625,
      "learning_rate": 1.7113046856225611e-06,
      "loss": 0.7953,
      "num_input_tokens_seen": 69598288,
      "step": 119965
    },
    {
      "epoch": 17.8686327077748,
      "grad_norm": 7.75,
      "learning_rate": 1.7101233314311181e-06,
      "loss": 0.9457,
      "num_input_tokens_seen": 69601168,
      "step": 119970
    },
    {
      "epoch": 17.869377420315757,
      "grad_norm": 6.53125,
      "learning_rate": 1.7089423706984742e-06,
      "loss": 0.9464,
      "num_input_tokens_seen": 69603952,
      "step": 119975
    },
    {
      "epoch": 17.870122132856718,
      "grad_norm": 10.875,
      "learning_rate": 1.7077618034445714e-06,
      "loss": 1.0832,
      "num_input_tokens_seen": 69607152,
      "step": 119980
    },
    {
      "epoch": 17.870866845397675,
      "grad_norm": 10.8125,
      "learning_rate": 1.706581629689366e-06,
      "loss": 1.3896,
      "num_input_tokens_seen": 69610800,
      "step": 119985
    },
    {
      "epoch": 17.871611557938635,
      "grad_norm": 9.0,
      "learning_rate": 1.705401849452784e-06,
      "loss": 0.8448,
      "num_input_tokens_seen": 69613584,
      "step": 119990
    },
    {
      "epoch": 17.872356270479596,
      "grad_norm": 3.4375,
      "learning_rate": 1.7042224627547676e-06,
      "loss": 1.2298,
      "num_input_tokens_seen": 69616368,
      "step": 119995
    },
    {
      "epoch": 17.873100983020553,
      "grad_norm": 12.875,
      "learning_rate": 1.7030434696152342e-06,
      "loss": 1.1178,
      "num_input_tokens_seen": 69619408,
      "step": 120000
    },
    {
      "epoch": 17.873845695561513,
      "grad_norm": 8.75,
      "learning_rate": 1.701864870054104e-06,
      "loss": 0.9545,
      "num_input_tokens_seen": 69622384,
      "step": 120005
    },
    {
      "epoch": 17.874590408102474,
      "grad_norm": 11.625,
      "learning_rate": 1.700686664091286e-06,
      "loss": 1.2776,
      "num_input_tokens_seen": 69625520,
      "step": 120010
    },
    {
      "epoch": 17.87533512064343,
      "grad_norm": 13.3125,
      "learning_rate": 1.6995088517466867e-06,
      "loss": 1.4299,
      "num_input_tokens_seen": 69628528,
      "step": 120015
    },
    {
      "epoch": 17.87607983318439,
      "grad_norm": 10.0625,
      "learning_rate": 1.6983314330402039e-06,
      "loss": 1.662,
      "num_input_tokens_seen": 69631312,
      "step": 120020
    },
    {
      "epoch": 17.87682454572535,
      "grad_norm": 13.125,
      "learning_rate": 1.6971544079917273e-06,
      "loss": 1.2625,
      "num_input_tokens_seen": 69634480,
      "step": 120025
    },
    {
      "epoch": 17.87756925826631,
      "grad_norm": 8.9375,
      "learning_rate": 1.6959777766211437e-06,
      "loss": 1.0887,
      "num_input_tokens_seen": 69637296,
      "step": 120030
    },
    {
      "epoch": 17.87831397080727,
      "grad_norm": 5.40625,
      "learning_rate": 1.6948015389483291e-06,
      "loss": 0.9264,
      "num_input_tokens_seen": 69640080,
      "step": 120035
    },
    {
      "epoch": 17.879058683348227,
      "grad_norm": 10.6875,
      "learning_rate": 1.6936256949931618e-06,
      "loss": 1.4232,
      "num_input_tokens_seen": 69642960,
      "step": 120040
    },
    {
      "epoch": 17.879803395889187,
      "grad_norm": 15.75,
      "learning_rate": 1.692450244775498e-06,
      "loss": 1.5149,
      "num_input_tokens_seen": 69645680,
      "step": 120045
    },
    {
      "epoch": 17.880548108430148,
      "grad_norm": 19.375,
      "learning_rate": 1.6912751883151945e-06,
      "loss": 1.188,
      "num_input_tokens_seen": 69648496,
      "step": 120050
    },
    {
      "epoch": 17.881292820971105,
      "grad_norm": 10.3125,
      "learning_rate": 1.6901005256321128e-06,
      "loss": 1.1838,
      "num_input_tokens_seen": 69651376,
      "step": 120055
    },
    {
      "epoch": 17.882037533512065,
      "grad_norm": 12.3125,
      "learning_rate": 1.6889262567460846e-06,
      "loss": 1.2057,
      "num_input_tokens_seen": 69654352,
      "step": 120060
    },
    {
      "epoch": 17.882782246053022,
      "grad_norm": 12.125,
      "learning_rate": 1.6877523816769603e-06,
      "loss": 0.8623,
      "num_input_tokens_seen": 69657616,
      "step": 120065
    },
    {
      "epoch": 17.883526958593983,
      "grad_norm": 3.84375,
      "learning_rate": 1.6865789004445686e-06,
      "loss": 0.6609,
      "num_input_tokens_seen": 69660336,
      "step": 120070
    },
    {
      "epoch": 17.884271671134943,
      "grad_norm": 14.6875,
      "learning_rate": 1.6854058130687272e-06,
      "loss": 0.8146,
      "num_input_tokens_seen": 69663248,
      "step": 120075
    },
    {
      "epoch": 17.8850163836759,
      "grad_norm": 9.9375,
      "learning_rate": 1.684233119569259e-06,
      "loss": 0.9859,
      "num_input_tokens_seen": 69665840,
      "step": 120080
    },
    {
      "epoch": 17.88576109621686,
      "grad_norm": 14.375,
      "learning_rate": 1.683060819965976e-06,
      "loss": 1.3862,
      "num_input_tokens_seen": 69668720,
      "step": 120085
    },
    {
      "epoch": 17.886505808757818,
      "grad_norm": 7.28125,
      "learning_rate": 1.6818889142786842e-06,
      "loss": 1.0119,
      "num_input_tokens_seen": 69671984,
      "step": 120090
    },
    {
      "epoch": 17.88725052129878,
      "grad_norm": 11.75,
      "learning_rate": 1.6807174025271737e-06,
      "loss": 1.05,
      "num_input_tokens_seen": 69675152,
      "step": 120095
    },
    {
      "epoch": 17.88799523383974,
      "grad_norm": 7.625,
      "learning_rate": 1.6795462847312481e-06,
      "loss": 1.4269,
      "num_input_tokens_seen": 69678032,
      "step": 120100
    },
    {
      "epoch": 17.888739946380696,
      "grad_norm": 15.625,
      "learning_rate": 1.6783755609106804e-06,
      "loss": 1.1442,
      "num_input_tokens_seen": 69680880,
      "step": 120105
    },
    {
      "epoch": 17.889484658921656,
      "grad_norm": 16.75,
      "learning_rate": 1.6772052310852605e-06,
      "loss": 1.1637,
      "num_input_tokens_seen": 69684048,
      "step": 120110
    },
    {
      "epoch": 17.890229371462617,
      "grad_norm": 9.375,
      "learning_rate": 1.6760352952747472e-06,
      "loss": 1.2757,
      "num_input_tokens_seen": 69687600,
      "step": 120115
    },
    {
      "epoch": 17.890974084003574,
      "grad_norm": 9.8125,
      "learning_rate": 1.6748657534989194e-06,
      "loss": 0.841,
      "num_input_tokens_seen": 69690256,
      "step": 120120
    },
    {
      "epoch": 17.891718796544534,
      "grad_norm": 12.375,
      "learning_rate": 1.673696605777525e-06,
      "loss": 0.9192,
      "num_input_tokens_seen": 69692976,
      "step": 120125
    },
    {
      "epoch": 17.89246350908549,
      "grad_norm": 7.28125,
      "learning_rate": 1.6725278521303178e-06,
      "loss": 0.9098,
      "num_input_tokens_seen": 69695792,
      "step": 120130
    },
    {
      "epoch": 17.893208221626452,
      "grad_norm": 6.25,
      "learning_rate": 1.6713594925770459e-06,
      "loss": 0.9665,
      "num_input_tokens_seen": 69698864,
      "step": 120135
    },
    {
      "epoch": 17.893952934167412,
      "grad_norm": 14.6875,
      "learning_rate": 1.6701915271374436e-06,
      "loss": 1.0214,
      "num_input_tokens_seen": 69701456,
      "step": 120140
    },
    {
      "epoch": 17.89469764670837,
      "grad_norm": 13.0,
      "learning_rate": 1.6690239558312476e-06,
      "loss": 1.2336,
      "num_input_tokens_seen": 69704752,
      "step": 120145
    },
    {
      "epoch": 17.89544235924933,
      "grad_norm": 13.0,
      "learning_rate": 1.667856778678173e-06,
      "loss": 1.3174,
      "num_input_tokens_seen": 69708048,
      "step": 120150
    },
    {
      "epoch": 17.89618707179029,
      "grad_norm": 11.6875,
      "learning_rate": 1.6666899956979483e-06,
      "loss": 1.2541,
      "num_input_tokens_seen": 69710800,
      "step": 120155
    },
    {
      "epoch": 17.896931784331247,
      "grad_norm": 18.125,
      "learning_rate": 1.665523606910277e-06,
      "loss": 1.1147,
      "num_input_tokens_seen": 69714000,
      "step": 120160
    },
    {
      "epoch": 17.897676496872208,
      "grad_norm": 12.5625,
      "learning_rate": 1.6643576123348741e-06,
      "loss": 1.0544,
      "num_input_tokens_seen": 69716720,
      "step": 120165
    },
    {
      "epoch": 17.898421209413165,
      "grad_norm": 7.5,
      "learning_rate": 1.6631920119914296e-06,
      "loss": 1.2956,
      "num_input_tokens_seen": 69719440,
      "step": 120170
    },
    {
      "epoch": 17.899165921954125,
      "grad_norm": 8.625,
      "learning_rate": 1.6620268058996357e-06,
      "loss": 0.867,
      "num_input_tokens_seen": 69722288,
      "step": 120175
    },
    {
      "epoch": 17.899910634495086,
      "grad_norm": 21.0,
      "learning_rate": 1.6608619940791826e-06,
      "loss": 1.1347,
      "num_input_tokens_seen": 69725264,
      "step": 120180
    },
    {
      "epoch": 17.900655347036043,
      "grad_norm": 13.875,
      "learning_rate": 1.6596975765497403e-06,
      "loss": 1.0654,
      "num_input_tokens_seen": 69728304,
      "step": 120185
    },
    {
      "epoch": 17.901400059577004,
      "grad_norm": 5.84375,
      "learning_rate": 1.6585335533309903e-06,
      "loss": 1.0183,
      "num_input_tokens_seen": 69730768,
      "step": 120190
    },
    {
      "epoch": 17.902144772117964,
      "grad_norm": 12.5,
      "learning_rate": 1.6573699244425895e-06,
      "loss": 1.3099,
      "num_input_tokens_seen": 69733744,
      "step": 120195
    },
    {
      "epoch": 17.90288948465892,
      "grad_norm": 13.5625,
      "learning_rate": 1.6562066899042023e-06,
      "loss": 1.1152,
      "num_input_tokens_seen": 69736432,
      "step": 120200
    },
    {
      "epoch": 17.90363419719988,
      "grad_norm": 9.625,
      "learning_rate": 1.655043849735477e-06,
      "loss": 1.1393,
      "num_input_tokens_seen": 69739152,
      "step": 120205
    },
    {
      "epoch": 17.90437890974084,
      "grad_norm": 7.65625,
      "learning_rate": 1.653881403956062e-06,
      "loss": 1.265,
      "num_input_tokens_seen": 69742160,
      "step": 120210
    },
    {
      "epoch": 17.9051236222818,
      "grad_norm": 7.0,
      "learning_rate": 1.6527193525855911e-06,
      "loss": 0.8653,
      "num_input_tokens_seen": 69744880,
      "step": 120215
    },
    {
      "epoch": 17.90586833482276,
      "grad_norm": 13.8125,
      "learning_rate": 1.6515576956436906e-06,
      "loss": 1.3317,
      "num_input_tokens_seen": 69747728,
      "step": 120220
    },
    {
      "epoch": 17.906613047363717,
      "grad_norm": 13.8125,
      "learning_rate": 1.6503964331500004e-06,
      "loss": 1.1097,
      "num_input_tokens_seen": 69751024,
      "step": 120225
    },
    {
      "epoch": 17.907357759904677,
      "grad_norm": 12.25,
      "learning_rate": 1.649235565124127e-06,
      "loss": 0.7366,
      "num_input_tokens_seen": 69753968,
      "step": 120230
    },
    {
      "epoch": 17.908102472445638,
      "grad_norm": 6.625,
      "learning_rate": 1.648075091585688e-06,
      "loss": 0.8997,
      "num_input_tokens_seen": 69756784,
      "step": 120235
    },
    {
      "epoch": 17.908847184986595,
      "grad_norm": 16.375,
      "learning_rate": 1.6469150125542843e-06,
      "loss": 1.4278,
      "num_input_tokens_seen": 69759792,
      "step": 120240
    },
    {
      "epoch": 17.909591897527555,
      "grad_norm": 14.0625,
      "learning_rate": 1.6457553280495168e-06,
      "loss": 1.2846,
      "num_input_tokens_seen": 69762448,
      "step": 120245
    },
    {
      "epoch": 17.910336610068512,
      "grad_norm": 13.125,
      "learning_rate": 1.6445960380909814e-06,
      "loss": 0.9154,
      "num_input_tokens_seen": 69765200,
      "step": 120250
    },
    {
      "epoch": 17.911081322609473,
      "grad_norm": 10.0625,
      "learning_rate": 1.6434371426982508e-06,
      "loss": 1.1364,
      "num_input_tokens_seen": 69768272,
      "step": 120255
    },
    {
      "epoch": 17.911826035150433,
      "grad_norm": 13.6875,
      "learning_rate": 1.642278641890918e-06,
      "loss": 1.3145,
      "num_input_tokens_seen": 69770960,
      "step": 120260
    },
    {
      "epoch": 17.91257074769139,
      "grad_norm": 14.125,
      "learning_rate": 1.641120535688548e-06,
      "loss": 1.1098,
      "num_input_tokens_seen": 69774000,
      "step": 120265
    },
    {
      "epoch": 17.91331546023235,
      "grad_norm": 9.0625,
      "learning_rate": 1.6399628241106996e-06,
      "loss": 1.0769,
      "num_input_tokens_seen": 69776912,
      "step": 120270
    },
    {
      "epoch": 17.91406017277331,
      "grad_norm": 11.5625,
      "learning_rate": 1.638805507176941e-06,
      "loss": 1.1033,
      "num_input_tokens_seen": 69779920,
      "step": 120275
    },
    {
      "epoch": 17.91480488531427,
      "grad_norm": 16.75,
      "learning_rate": 1.637648584906823e-06,
      "loss": 1.0354,
      "num_input_tokens_seen": 69783024,
      "step": 120280
    },
    {
      "epoch": 17.91554959785523,
      "grad_norm": 7.875,
      "learning_rate": 1.6364920573198856e-06,
      "loss": 0.9956,
      "num_input_tokens_seen": 69785936,
      "step": 120285
    },
    {
      "epoch": 17.916294310396186,
      "grad_norm": 5.53125,
      "learning_rate": 1.6353359244356658e-06,
      "loss": 1.2136,
      "num_input_tokens_seen": 69788912,
      "step": 120290
    },
    {
      "epoch": 17.917039022937146,
      "grad_norm": 6.09375,
      "learning_rate": 1.634180186273701e-06,
      "loss": 0.7467,
      "num_input_tokens_seen": 69791664,
      "step": 120295
    },
    {
      "epoch": 17.917783735478107,
      "grad_norm": 7.5625,
      "learning_rate": 1.6330248428535117e-06,
      "loss": 1.1644,
      "num_input_tokens_seen": 69794384,
      "step": 120300
    },
    {
      "epoch": 17.918528448019064,
      "grad_norm": 14.875,
      "learning_rate": 1.6318698941946237e-06,
      "loss": 1.1681,
      "num_input_tokens_seen": 69797680,
      "step": 120305
    },
    {
      "epoch": 17.919273160560024,
      "grad_norm": 10.6875,
      "learning_rate": 1.6307153403165382e-06,
      "loss": 0.9693,
      "num_input_tokens_seen": 69800560,
      "step": 120310
    },
    {
      "epoch": 17.92001787310098,
      "grad_norm": 11.3125,
      "learning_rate": 1.6295611812387673e-06,
      "loss": 1.0611,
      "num_input_tokens_seen": 69803504,
      "step": 120315
    },
    {
      "epoch": 17.920762585641942,
      "grad_norm": 7.625,
      "learning_rate": 1.6284074169808067e-06,
      "loss": 1.1216,
      "num_input_tokens_seen": 69806352,
      "step": 120320
    },
    {
      "epoch": 17.921507298182902,
      "grad_norm": 9.1875,
      "learning_rate": 1.6272540475621518e-06,
      "loss": 1.2854,
      "num_input_tokens_seen": 69809424,
      "step": 120325
    },
    {
      "epoch": 17.92225201072386,
      "grad_norm": 11.4375,
      "learning_rate": 1.6261010730022842e-06,
      "loss": 0.9419,
      "num_input_tokens_seen": 69812464,
      "step": 120330
    },
    {
      "epoch": 17.92299672326482,
      "grad_norm": 14.875,
      "learning_rate": 1.6249484933206853e-06,
      "loss": 1.3185,
      "num_input_tokens_seen": 69815216,
      "step": 120335
    },
    {
      "epoch": 17.92374143580578,
      "grad_norm": 11.6875,
      "learning_rate": 1.6237963085368236e-06,
      "loss": 0.7842,
      "num_input_tokens_seen": 69818128,
      "step": 120340
    },
    {
      "epoch": 17.924486148346737,
      "grad_norm": 8.25,
      "learning_rate": 1.6226445186701577e-06,
      "loss": 0.8827,
      "num_input_tokens_seen": 69820880,
      "step": 120345
    },
    {
      "epoch": 17.925230860887698,
      "grad_norm": 31.75,
      "learning_rate": 1.6214931237401588e-06,
      "loss": 1.7692,
      "num_input_tokens_seen": 69824016,
      "step": 120350
    },
    {
      "epoch": 17.925975573428655,
      "grad_norm": 8.8125,
      "learning_rate": 1.6203421237662692e-06,
      "loss": 0.9246,
      "num_input_tokens_seen": 69826864,
      "step": 120355
    },
    {
      "epoch": 17.926720285969616,
      "grad_norm": 7.84375,
      "learning_rate": 1.619191518767943e-06,
      "loss": 1.1104,
      "num_input_tokens_seen": 69829552,
      "step": 120360
    },
    {
      "epoch": 17.927464998510576,
      "grad_norm": 9.0,
      "learning_rate": 1.618041308764609e-06,
      "loss": 1.1495,
      "num_input_tokens_seen": 69832336,
      "step": 120365
    },
    {
      "epoch": 17.928209711051533,
      "grad_norm": 6.40625,
      "learning_rate": 1.6168914937757019e-06,
      "loss": 1.1766,
      "num_input_tokens_seen": 69835120,
      "step": 120370
    },
    {
      "epoch": 17.928954423592494,
      "grad_norm": 8.6875,
      "learning_rate": 1.6157420738206503e-06,
      "loss": 1.1245,
      "num_input_tokens_seen": 69838032,
      "step": 120375
    },
    {
      "epoch": 17.929699136133454,
      "grad_norm": 10.1875,
      "learning_rate": 1.6145930489188666e-06,
      "loss": 1.1467,
      "num_input_tokens_seen": 69841168,
      "step": 120380
    },
    {
      "epoch": 17.93044384867441,
      "grad_norm": 9.375,
      "learning_rate": 1.6134444190897685e-06,
      "loss": 0.7726,
      "num_input_tokens_seen": 69843824,
      "step": 120385
    },
    {
      "epoch": 17.93118856121537,
      "grad_norm": 10.8125,
      "learning_rate": 1.6122961843527546e-06,
      "loss": 0.8634,
      "num_input_tokens_seen": 69846864,
      "step": 120390
    },
    {
      "epoch": 17.93193327375633,
      "grad_norm": 13.1875,
      "learning_rate": 1.6111483447272286e-06,
      "loss": 0.9818,
      "num_input_tokens_seen": 69849552,
      "step": 120395
    },
    {
      "epoch": 17.93267798629729,
      "grad_norm": 6.6875,
      "learning_rate": 1.6100009002325806e-06,
      "loss": 1.2114,
      "num_input_tokens_seen": 69853104,
      "step": 120400
    },
    {
      "epoch": 17.93342269883825,
      "grad_norm": 5.90625,
      "learning_rate": 1.608853850888195e-06,
      "loss": 0.8892,
      "num_input_tokens_seen": 69855824,
      "step": 120405
    },
    {
      "epoch": 17.934167411379207,
      "grad_norm": 14.5,
      "learning_rate": 1.6077071967134511e-06,
      "loss": 1.2984,
      "num_input_tokens_seen": 69858704,
      "step": 120410
    },
    {
      "epoch": 17.934912123920167,
      "grad_norm": 5.875,
      "learning_rate": 1.6065609377277136e-06,
      "loss": 0.9737,
      "num_input_tokens_seen": 69861552,
      "step": 120415
    },
    {
      "epoch": 17.935656836461128,
      "grad_norm": 7.5625,
      "learning_rate": 1.6054150739503587e-06,
      "loss": 1.1324,
      "num_input_tokens_seen": 69864432,
      "step": 120420
    },
    {
      "epoch": 17.936401549002085,
      "grad_norm": 10.875,
      "learning_rate": 1.604269605400735e-06,
      "loss": 0.9342,
      "num_input_tokens_seen": 69867440,
      "step": 120425
    },
    {
      "epoch": 17.937146261543045,
      "grad_norm": 18.375,
      "learning_rate": 1.6031245320982018e-06,
      "loss": 0.9176,
      "num_input_tokens_seen": 69870224,
      "step": 120430
    },
    {
      "epoch": 17.937890974084002,
      "grad_norm": 7.65625,
      "learning_rate": 1.601979854062094e-06,
      "loss": 0.9945,
      "num_input_tokens_seen": 69873104,
      "step": 120435
    },
    {
      "epoch": 17.938635686624963,
      "grad_norm": 7.40625,
      "learning_rate": 1.6008355713117623e-06,
      "loss": 0.8385,
      "num_input_tokens_seen": 69876016,
      "step": 120440
    },
    {
      "epoch": 17.939380399165923,
      "grad_norm": 7.09375,
      "learning_rate": 1.599691683866525e-06,
      "loss": 0.9001,
      "num_input_tokens_seen": 69878800,
      "step": 120445
    },
    {
      "epoch": 17.94012511170688,
      "grad_norm": 8.375,
      "learning_rate": 1.5985481917457217e-06,
      "loss": 1.5264,
      "num_input_tokens_seen": 69881680,
      "step": 120450
    },
    {
      "epoch": 17.94086982424784,
      "grad_norm": 9.8125,
      "learning_rate": 1.5974050949686597e-06,
      "loss": 1.2266,
      "num_input_tokens_seen": 69884400,
      "step": 120455
    },
    {
      "epoch": 17.941614536788798,
      "grad_norm": 13.4375,
      "learning_rate": 1.5962623935546483e-06,
      "loss": 1.3142,
      "num_input_tokens_seen": 69887056,
      "step": 120460
    },
    {
      "epoch": 17.94235924932976,
      "grad_norm": 8.1875,
      "learning_rate": 1.5951200875230055e-06,
      "loss": 0.9434,
      "num_input_tokens_seen": 69889712,
      "step": 120465
    },
    {
      "epoch": 17.94310396187072,
      "grad_norm": 5.53125,
      "learning_rate": 1.5939781768930185e-06,
      "loss": 1.3008,
      "num_input_tokens_seen": 69892496,
      "step": 120470
    },
    {
      "epoch": 17.943848674411676,
      "grad_norm": 6.34375,
      "learning_rate": 1.59283666168398e-06,
      "loss": 0.9515,
      "num_input_tokens_seen": 69895472,
      "step": 120475
    },
    {
      "epoch": 17.944593386952636,
      "grad_norm": 8.125,
      "learning_rate": 1.5916955419151725e-06,
      "loss": 0.9225,
      "num_input_tokens_seen": 69898640,
      "step": 120480
    },
    {
      "epoch": 17.945338099493597,
      "grad_norm": 7.3125,
      "learning_rate": 1.5905548176058826e-06,
      "loss": 1.1023,
      "num_input_tokens_seen": 69901456,
      "step": 120485
    },
    {
      "epoch": 17.946082812034554,
      "grad_norm": 16.875,
      "learning_rate": 1.5894144887753786e-06,
      "loss": 1.0886,
      "num_input_tokens_seen": 69904272,
      "step": 120490
    },
    {
      "epoch": 17.946827524575514,
      "grad_norm": 6.28125,
      "learning_rate": 1.5882745554429174e-06,
      "loss": 1.0251,
      "num_input_tokens_seen": 69907056,
      "step": 120495
    },
    {
      "epoch": 17.94757223711647,
      "grad_norm": 12.625,
      "learning_rate": 1.5871350176277667e-06,
      "loss": 1.5763,
      "num_input_tokens_seen": 69909808,
      "step": 120500
    },
    {
      "epoch": 17.948316949657432,
      "grad_norm": 8.3125,
      "learning_rate": 1.585995875349172e-06,
      "loss": 1.1505,
      "num_input_tokens_seen": 69912752,
      "step": 120505
    },
    {
      "epoch": 17.949061662198392,
      "grad_norm": 16.0,
      "learning_rate": 1.5848571286263825e-06,
      "loss": 0.8929,
      "num_input_tokens_seen": 69915984,
      "step": 120510
    },
    {
      "epoch": 17.94980637473935,
      "grad_norm": 10.0,
      "learning_rate": 1.5837187774786293e-06,
      "loss": 1.3545,
      "num_input_tokens_seen": 69918864,
      "step": 120515
    },
    {
      "epoch": 17.95055108728031,
      "grad_norm": 15.6875,
      "learning_rate": 1.5825808219251532e-06,
      "loss": 1.1743,
      "num_input_tokens_seen": 69921776,
      "step": 120520
    },
    {
      "epoch": 17.95129579982127,
      "grad_norm": 11.875,
      "learning_rate": 1.5814432619851687e-06,
      "loss": 1.168,
      "num_input_tokens_seen": 69924624,
      "step": 120525
    },
    {
      "epoch": 17.952040512362228,
      "grad_norm": 16.75,
      "learning_rate": 1.5803060976779026e-06,
      "loss": 1.1775,
      "num_input_tokens_seen": 69927440,
      "step": 120530
    },
    {
      "epoch": 17.952785224903188,
      "grad_norm": 12.5625,
      "learning_rate": 1.5791693290225646e-06,
      "loss": 1.1127,
      "num_input_tokens_seen": 69929904,
      "step": 120535
    },
    {
      "epoch": 17.953529937444145,
      "grad_norm": 9.0,
      "learning_rate": 1.5780329560383527e-06,
      "loss": 1.0507,
      "num_input_tokens_seen": 69932976,
      "step": 120540
    },
    {
      "epoch": 17.954274649985106,
      "grad_norm": 13.0625,
      "learning_rate": 1.5768969787444716e-06,
      "loss": 1.1799,
      "num_input_tokens_seen": 69935792,
      "step": 120545
    },
    {
      "epoch": 17.955019362526066,
      "grad_norm": 14.125,
      "learning_rate": 1.5757613971601054e-06,
      "loss": 1.3993,
      "num_input_tokens_seen": 69938704,
      "step": 120550
    },
    {
      "epoch": 17.955764075067023,
      "grad_norm": 16.5,
      "learning_rate": 1.5746262113044474e-06,
      "loss": 1.1943,
      "num_input_tokens_seen": 69941456,
      "step": 120555
    },
    {
      "epoch": 17.956508787607984,
      "grad_norm": 7.90625,
      "learning_rate": 1.5734914211966683e-06,
      "loss": 1.2062,
      "num_input_tokens_seen": 69944304,
      "step": 120560
    },
    {
      "epoch": 17.957253500148944,
      "grad_norm": 12.0625,
      "learning_rate": 1.5723570268559445e-06,
      "loss": 1.1222,
      "num_input_tokens_seen": 69947024,
      "step": 120565
    },
    {
      "epoch": 17.9579982126899,
      "grad_norm": 5.40625,
      "learning_rate": 1.5712230283014385e-06,
      "loss": 0.9624,
      "num_input_tokens_seen": 69949968,
      "step": 120570
    },
    {
      "epoch": 17.95874292523086,
      "grad_norm": 10.9375,
      "learning_rate": 1.570089425552304e-06,
      "loss": 0.8303,
      "num_input_tokens_seen": 69952880,
      "step": 120575
    },
    {
      "epoch": 17.95948763777182,
      "grad_norm": 4.90625,
      "learning_rate": 1.5689562186276986e-06,
      "loss": 1.1305,
      "num_input_tokens_seen": 69955728,
      "step": 120580
    },
    {
      "epoch": 17.96023235031278,
      "grad_norm": 9.3125,
      "learning_rate": 1.567823407546759e-06,
      "loss": 1.0005,
      "num_input_tokens_seen": 69958768,
      "step": 120585
    },
    {
      "epoch": 17.96097706285374,
      "grad_norm": 6.03125,
      "learning_rate": 1.5666909923286315e-06,
      "loss": 1.1334,
      "num_input_tokens_seen": 69961520,
      "step": 120590
    },
    {
      "epoch": 17.961721775394697,
      "grad_norm": 12.9375,
      "learning_rate": 1.5655589729924453e-06,
      "loss": 1.191,
      "num_input_tokens_seen": 69964080,
      "step": 120595
    },
    {
      "epoch": 17.962466487935657,
      "grad_norm": 8.8125,
      "learning_rate": 1.564427349557318e-06,
      "loss": 0.9413,
      "num_input_tokens_seen": 69967024,
      "step": 120600
    },
    {
      "epoch": 17.963211200476614,
      "grad_norm": 10.125,
      "learning_rate": 1.5632961220423737e-06,
      "loss": 1.2054,
      "num_input_tokens_seen": 69969712,
      "step": 120605
    },
    {
      "epoch": 17.963955913017575,
      "grad_norm": 8.75,
      "learning_rate": 1.5621652904667244e-06,
      "loss": 0.9477,
      "num_input_tokens_seen": 69972400,
      "step": 120610
    },
    {
      "epoch": 17.964700625558535,
      "grad_norm": 6.8125,
      "learning_rate": 1.5610348548494692e-06,
      "loss": 1.0812,
      "num_input_tokens_seen": 69975344,
      "step": 120615
    },
    {
      "epoch": 17.965445338099492,
      "grad_norm": 4.96875,
      "learning_rate": 1.5599048152097034e-06,
      "loss": 0.9992,
      "num_input_tokens_seen": 69977968,
      "step": 120620
    },
    {
      "epoch": 17.966190050640453,
      "grad_norm": 5.46875,
      "learning_rate": 1.558775171566529e-06,
      "loss": 0.8713,
      "num_input_tokens_seen": 69980720,
      "step": 120625
    },
    {
      "epoch": 17.966934763181413,
      "grad_norm": 9.25,
      "learning_rate": 1.5576459239390162e-06,
      "loss": 1.0728,
      "num_input_tokens_seen": 69983664,
      "step": 120630
    },
    {
      "epoch": 17.96767947572237,
      "grad_norm": 16.125,
      "learning_rate": 1.5565170723462558e-06,
      "loss": 1.0404,
      "num_input_tokens_seen": 69986160,
      "step": 120635
    },
    {
      "epoch": 17.96842418826333,
      "grad_norm": 7.53125,
      "learning_rate": 1.5553886168073073e-06,
      "loss": 0.9027,
      "num_input_tokens_seen": 69989104,
      "step": 120640
    },
    {
      "epoch": 17.969168900804288,
      "grad_norm": 13.75,
      "learning_rate": 1.5542605573412444e-06,
      "loss": 0.9463,
      "num_input_tokens_seen": 69991888,
      "step": 120645
    },
    {
      "epoch": 17.96991361334525,
      "grad_norm": 11.0,
      "learning_rate": 1.5531328939671215e-06,
      "loss": 1.2829,
      "num_input_tokens_seen": 69994800,
      "step": 120650
    },
    {
      "epoch": 17.97065832588621,
      "grad_norm": 7.5625,
      "learning_rate": 1.552005626703984e-06,
      "loss": 0.8456,
      "num_input_tokens_seen": 69997808,
      "step": 120655
    },
    {
      "epoch": 17.971403038427166,
      "grad_norm": 9.625,
      "learning_rate": 1.550878755570881e-06,
      "loss": 1.126,
      "num_input_tokens_seen": 70000560,
      "step": 120660
    },
    {
      "epoch": 17.972147750968126,
      "grad_norm": 8.4375,
      "learning_rate": 1.5497522805868525e-06,
      "loss": 0.8462,
      "num_input_tokens_seen": 70003344,
      "step": 120665
    },
    {
      "epoch": 17.972892463509087,
      "grad_norm": 7.28125,
      "learning_rate": 1.5486262017709225e-06,
      "loss": 1.1499,
      "num_input_tokens_seen": 70006288,
      "step": 120670
    },
    {
      "epoch": 17.973637176050044,
      "grad_norm": 14.3125,
      "learning_rate": 1.5475005191421172e-06,
      "loss": 1.2141,
      "num_input_tokens_seen": 70009136,
      "step": 120675
    },
    {
      "epoch": 17.974381888591004,
      "grad_norm": 10.375,
      "learning_rate": 1.5463752327194548e-06,
      "loss": 1.2146,
      "num_input_tokens_seen": 70011984,
      "step": 120680
    },
    {
      "epoch": 17.97512660113196,
      "grad_norm": 8.75,
      "learning_rate": 1.545250342521945e-06,
      "loss": 1.1529,
      "num_input_tokens_seen": 70014768,
      "step": 120685
    },
    {
      "epoch": 17.975871313672922,
      "grad_norm": 7.53125,
      "learning_rate": 1.544125848568595e-06,
      "loss": 1.079,
      "num_input_tokens_seen": 70017584,
      "step": 120690
    },
    {
      "epoch": 17.976616026213883,
      "grad_norm": 14.9375,
      "learning_rate": 1.5430017508783978e-06,
      "loss": 1.0536,
      "num_input_tokens_seen": 70021232,
      "step": 120695
    },
    {
      "epoch": 17.97736073875484,
      "grad_norm": 9.0,
      "learning_rate": 1.541878049470344e-06,
      "loss": 1.0626,
      "num_input_tokens_seen": 70024144,
      "step": 120700
    },
    {
      "epoch": 17.9781054512958,
      "grad_norm": 10.875,
      "learning_rate": 1.5407547443634206e-06,
      "loss": 1.0289,
      "num_input_tokens_seen": 70027536,
      "step": 120705
    },
    {
      "epoch": 17.97885016383676,
      "grad_norm": 8.5,
      "learning_rate": 1.5396318355765993e-06,
      "loss": 0.8907,
      "num_input_tokens_seen": 70030512,
      "step": 120710
    },
    {
      "epoch": 17.979594876377718,
      "grad_norm": 4.71875,
      "learning_rate": 1.538509323128859e-06,
      "loss": 0.924,
      "num_input_tokens_seen": 70033488,
      "step": 120715
    },
    {
      "epoch": 17.980339588918678,
      "grad_norm": 7.65625,
      "learning_rate": 1.5373872070391536e-06,
      "loss": 1.0422,
      "num_input_tokens_seen": 70036624,
      "step": 120720
    },
    {
      "epoch": 17.981084301459635,
      "grad_norm": 12.8125,
      "learning_rate": 1.5362654873264493e-06,
      "loss": 0.9882,
      "num_input_tokens_seen": 70039440,
      "step": 120725
    },
    {
      "epoch": 17.981829014000596,
      "grad_norm": 5.53125,
      "learning_rate": 1.5351441640096941e-06,
      "loss": 0.9483,
      "num_input_tokens_seen": 70042192,
      "step": 120730
    },
    {
      "epoch": 17.982573726541556,
      "grad_norm": 7.75,
      "learning_rate": 1.5340232371078262e-06,
      "loss": 0.9848,
      "num_input_tokens_seen": 70045264,
      "step": 120735
    },
    {
      "epoch": 17.983318439082513,
      "grad_norm": 7.15625,
      "learning_rate": 1.5329027066397884e-06,
      "loss": 0.8383,
      "num_input_tokens_seen": 70048016,
      "step": 120740
    },
    {
      "epoch": 17.984063151623474,
      "grad_norm": 8.0625,
      "learning_rate": 1.5317825726245045e-06,
      "loss": 0.8583,
      "num_input_tokens_seen": 70050704,
      "step": 120745
    },
    {
      "epoch": 17.984807864164434,
      "grad_norm": 15.6875,
      "learning_rate": 1.5306628350809037e-06,
      "loss": 1.5002,
      "num_input_tokens_seen": 70053424,
      "step": 120750
    },
    {
      "epoch": 17.98555257670539,
      "grad_norm": 14.3125,
      "learning_rate": 1.529543494027899e-06,
      "loss": 0.9813,
      "num_input_tokens_seen": 70056368,
      "step": 120755
    },
    {
      "epoch": 17.98629728924635,
      "grad_norm": 15.125,
      "learning_rate": 1.5284245494844057e-06,
      "loss": 1.1554,
      "num_input_tokens_seen": 70059600,
      "step": 120760
    },
    {
      "epoch": 17.98704200178731,
      "grad_norm": 9.3125,
      "learning_rate": 1.5273060014693224e-06,
      "loss": 1.0696,
      "num_input_tokens_seen": 70062544,
      "step": 120765
    },
    {
      "epoch": 17.98778671432827,
      "grad_norm": 4.8125,
      "learning_rate": 1.526187850001548e-06,
      "loss": 0.8795,
      "num_input_tokens_seen": 70065328,
      "step": 120770
    },
    {
      "epoch": 17.98853142686923,
      "grad_norm": 6.40625,
      "learning_rate": 1.5250700950999758e-06,
      "loss": 1.0958,
      "num_input_tokens_seen": 70068304,
      "step": 120775
    },
    {
      "epoch": 17.989276139410187,
      "grad_norm": 7.40625,
      "learning_rate": 1.5239527367834794e-06,
      "loss": 1.1902,
      "num_input_tokens_seen": 70071184,
      "step": 120780
    },
    {
      "epoch": 17.990020851951147,
      "grad_norm": 11.0625,
      "learning_rate": 1.5228357750709465e-06,
      "loss": 1.2314,
      "num_input_tokens_seen": 70074288,
      "step": 120785
    },
    {
      "epoch": 17.990765564492108,
      "grad_norm": 12.5625,
      "learning_rate": 1.5217192099812372e-06,
      "loss": 1.0936,
      "num_input_tokens_seen": 70077264,
      "step": 120790
    },
    {
      "epoch": 17.991510277033065,
      "grad_norm": 15.4375,
      "learning_rate": 1.5206030415332223e-06,
      "loss": 1.5559,
      "num_input_tokens_seen": 70080720,
      "step": 120795
    },
    {
      "epoch": 17.992254989574025,
      "grad_norm": 5.71875,
      "learning_rate": 1.519487269745759e-06,
      "loss": 1.0066,
      "num_input_tokens_seen": 70083600,
      "step": 120800
    },
    {
      "epoch": 17.992999702114982,
      "grad_norm": 10.8125,
      "learning_rate": 1.5183718946376907e-06,
      "loss": 1.118,
      "num_input_tokens_seen": 70086416,
      "step": 120805
    },
    {
      "epoch": 17.993744414655943,
      "grad_norm": 7.625,
      "learning_rate": 1.5172569162278661e-06,
      "loss": 0.9041,
      "num_input_tokens_seen": 70089584,
      "step": 120810
    },
    {
      "epoch": 17.994489127196903,
      "grad_norm": 11.4375,
      "learning_rate": 1.5161423345351116e-06,
      "loss": 1.1957,
      "num_input_tokens_seen": 70092528,
      "step": 120815
    },
    {
      "epoch": 17.99523383973786,
      "grad_norm": 9.25,
      "learning_rate": 1.515028149578271e-06,
      "loss": 1.1584,
      "num_input_tokens_seen": 70095696,
      "step": 120820
    },
    {
      "epoch": 17.99597855227882,
      "grad_norm": 8.4375,
      "learning_rate": 1.5139143613761565e-06,
      "loss": 1.1773,
      "num_input_tokens_seen": 70098480,
      "step": 120825
    },
    {
      "epoch": 17.996723264819778,
      "grad_norm": 11.375,
      "learning_rate": 1.5128009699475948e-06,
      "loss": 1.2499,
      "num_input_tokens_seen": 70101232,
      "step": 120830
    },
    {
      "epoch": 17.99746797736074,
      "grad_norm": 8.125,
      "learning_rate": 1.5116879753113822e-06,
      "loss": 1.4262,
      "num_input_tokens_seen": 70103888,
      "step": 120835
    },
    {
      "epoch": 17.9982126899017,
      "grad_norm": 10.5625,
      "learning_rate": 1.510575377486334e-06,
      "loss": 1.0535,
      "num_input_tokens_seen": 70107280,
      "step": 120840
    },
    {
      "epoch": 17.998957402442656,
      "grad_norm": 10.75,
      "learning_rate": 1.5094631764912354e-06,
      "loss": 1.1273,
      "num_input_tokens_seen": 70110128,
      "step": 120845
    },
    {
      "epoch": 17.999702114983616,
      "grad_norm": 11.625,
      "learning_rate": 1.5083513723448877e-06,
      "loss": 1.4593,
      "num_input_tokens_seen": 70112976,
      "step": 120850
    },
    {
      "epoch": 18.0,
      "eval_loss": 1.0852864980697632,
      "eval_runtime": 43.6922,
      "eval_samples_per_second": 68.296,
      "eval_steps_per_second": 17.074,
      "num_input_tokens_seen": 70113640,
      "step": 120852
    },
    {
      "epoch": 18.000446827524577,
      "grad_norm": 15.4375,
      "learning_rate": 1.507239965066068e-06,
      "loss": 0.8607,
      "num_input_tokens_seen": 70115656,
      "step": 120855
    },
    {
      "epoch": 18.001191540065534,
      "grad_norm": 13.625,
      "learning_rate": 1.506128954673547e-06,
      "loss": 1.1564,
      "num_input_tokens_seen": 70118568,
      "step": 120860
    },
    {
      "epoch": 18.001936252606495,
      "grad_norm": 11.4375,
      "learning_rate": 1.505018341186104e-06,
      "loss": 1.1533,
      "num_input_tokens_seen": 70121384,
      "step": 120865
    },
    {
      "epoch": 18.00268096514745,
      "grad_norm": 10.0,
      "learning_rate": 1.5039081246224967e-06,
      "loss": 0.8952,
      "num_input_tokens_seen": 70124360,
      "step": 120870
    },
    {
      "epoch": 18.003425677688412,
      "grad_norm": 4.9375,
      "learning_rate": 1.502798305001482e-06,
      "loss": 0.821,
      "num_input_tokens_seen": 70127048,
      "step": 120875
    },
    {
      "epoch": 18.004170390229373,
      "grad_norm": 9.75,
      "learning_rate": 1.5016888823418035e-06,
      "loss": 1.0201,
      "num_input_tokens_seen": 70130120,
      "step": 120880
    },
    {
      "epoch": 18.00491510277033,
      "grad_norm": 9.9375,
      "learning_rate": 1.5005798566622125e-06,
      "loss": 0.6931,
      "num_input_tokens_seen": 70132776,
      "step": 120885
    },
    {
      "epoch": 18.00565981531129,
      "grad_norm": 14.6875,
      "learning_rate": 1.4994712279814415e-06,
      "loss": 0.8951,
      "num_input_tokens_seen": 70135688,
      "step": 120890
    },
    {
      "epoch": 18.00640452785225,
      "grad_norm": 10.6875,
      "learning_rate": 1.4983629963182143e-06,
      "loss": 0.9692,
      "num_input_tokens_seen": 70138376,
      "step": 120895
    },
    {
      "epoch": 18.007149240393208,
      "grad_norm": 5.6875,
      "learning_rate": 1.4972551616912633e-06,
      "loss": 0.7634,
      "num_input_tokens_seen": 70141032,
      "step": 120900
    },
    {
      "epoch": 18.007893952934168,
      "grad_norm": 7.09375,
      "learning_rate": 1.4961477241192956e-06,
      "loss": 0.9942,
      "num_input_tokens_seen": 70143976,
      "step": 120905
    },
    {
      "epoch": 18.008638665475125,
      "grad_norm": 5.3125,
      "learning_rate": 1.4950406836210267e-06,
      "loss": 1.1716,
      "num_input_tokens_seen": 70147080,
      "step": 120910
    },
    {
      "epoch": 18.009383378016086,
      "grad_norm": 6.875,
      "learning_rate": 1.493934040215153e-06,
      "loss": 0.7936,
      "num_input_tokens_seen": 70149992,
      "step": 120915
    },
    {
      "epoch": 18.010128090557046,
      "grad_norm": 17.0,
      "learning_rate": 1.492827793920376e-06,
      "loss": 1.1518,
      "num_input_tokens_seen": 70152680,
      "step": 120920
    },
    {
      "epoch": 18.010872803098003,
      "grad_norm": 5.0,
      "learning_rate": 1.4917219447553838e-06,
      "loss": 1.005,
      "num_input_tokens_seen": 70155464,
      "step": 120925
    },
    {
      "epoch": 18.011617515638964,
      "grad_norm": 12.625,
      "learning_rate": 1.49061649273885e-06,
      "loss": 1.0556,
      "num_input_tokens_seen": 70158504,
      "step": 120930
    },
    {
      "epoch": 18.012362228179924,
      "grad_norm": 9.6875,
      "learning_rate": 1.4895114378894625e-06,
      "loss": 0.9495,
      "num_input_tokens_seen": 70161384,
      "step": 120935
    },
    {
      "epoch": 18.01310694072088,
      "grad_norm": 8.5625,
      "learning_rate": 1.4884067802258845e-06,
      "loss": 1.1986,
      "num_input_tokens_seen": 70164232,
      "step": 120940
    },
    {
      "epoch": 18.013851653261842,
      "grad_norm": 17.25,
      "learning_rate": 1.4873025197667756e-06,
      "loss": 1.2955,
      "num_input_tokens_seen": 70167016,
      "step": 120945
    },
    {
      "epoch": 18.0145963658028,
      "grad_norm": 7.03125,
      "learning_rate": 1.4861986565307935e-06,
      "loss": 0.9537,
      "num_input_tokens_seen": 70169960,
      "step": 120950
    },
    {
      "epoch": 18.01534107834376,
      "grad_norm": 9.0625,
      "learning_rate": 1.4850951905365868e-06,
      "loss": 0.8401,
      "num_input_tokens_seen": 70172872,
      "step": 120955
    },
    {
      "epoch": 18.01608579088472,
      "grad_norm": 8.125,
      "learning_rate": 1.4839921218027935e-06,
      "loss": 1.4396,
      "num_input_tokens_seen": 70175624,
      "step": 120960
    },
    {
      "epoch": 18.016830503425677,
      "grad_norm": 7.53125,
      "learning_rate": 1.4828894503480601e-06,
      "loss": 1.0102,
      "num_input_tokens_seen": 70178312,
      "step": 120965
    },
    {
      "epoch": 18.017575215966637,
      "grad_norm": 8.875,
      "learning_rate": 1.4817871761910047e-06,
      "loss": 0.9322,
      "num_input_tokens_seen": 70181256,
      "step": 120970
    },
    {
      "epoch": 18.018319928507594,
      "grad_norm": 18.5,
      "learning_rate": 1.4806852993502485e-06,
      "loss": 1.5023,
      "num_input_tokens_seen": 70184200,
      "step": 120975
    },
    {
      "epoch": 18.019064641048555,
      "grad_norm": 24.625,
      "learning_rate": 1.4795838198444157e-06,
      "loss": 1.0821,
      "num_input_tokens_seen": 70187112,
      "step": 120980
    },
    {
      "epoch": 18.019809353589515,
      "grad_norm": 11.0,
      "learning_rate": 1.4784827376921052e-06,
      "loss": 0.7191,
      "num_input_tokens_seen": 70189800,
      "step": 120985
    },
    {
      "epoch": 18.020554066130472,
      "grad_norm": 3.609375,
      "learning_rate": 1.4773820529119243e-06,
      "loss": 0.8231,
      "num_input_tokens_seen": 70192584,
      "step": 120990
    },
    {
      "epoch": 18.021298778671433,
      "grad_norm": 5.1875,
      "learning_rate": 1.4762817655224664e-06,
      "loss": 1.0514,
      "num_input_tokens_seen": 70195624,
      "step": 120995
    },
    {
      "epoch": 18.022043491212393,
      "grad_norm": 7.125,
      "learning_rate": 1.4751818755423225e-06,
      "loss": 0.9077,
      "num_input_tokens_seen": 70198504,
      "step": 121000
    },
    {
      "epoch": 18.02278820375335,
      "grad_norm": 8.375,
      "learning_rate": 1.4740823829900664e-06,
      "loss": 0.9563,
      "num_input_tokens_seen": 70201800,
      "step": 121005
    },
    {
      "epoch": 18.02353291629431,
      "grad_norm": 11.25,
      "learning_rate": 1.4729832878842803e-06,
      "loss": 1.0687,
      "num_input_tokens_seen": 70204616,
      "step": 121010
    },
    {
      "epoch": 18.024277628835268,
      "grad_norm": 12.75,
      "learning_rate": 1.4718845902435303e-06,
      "loss": 0.9003,
      "num_input_tokens_seen": 70207336,
      "step": 121015
    },
    {
      "epoch": 18.02502234137623,
      "grad_norm": 8.25,
      "learning_rate": 1.4707862900863734e-06,
      "loss": 1.1153,
      "num_input_tokens_seen": 70210440,
      "step": 121020
    },
    {
      "epoch": 18.02576705391719,
      "grad_norm": 10.6875,
      "learning_rate": 1.4696883874313727e-06,
      "loss": 0.9341,
      "num_input_tokens_seen": 70213320,
      "step": 121025
    },
    {
      "epoch": 18.026511766458146,
      "grad_norm": 7.5625,
      "learning_rate": 1.4685908822970663e-06,
      "loss": 1.0577,
      "num_input_tokens_seen": 70216264,
      "step": 121030
    },
    {
      "epoch": 18.027256478999107,
      "grad_norm": 12.4375,
      "learning_rate": 1.4674937747020057e-06,
      "loss": 1.3402,
      "num_input_tokens_seen": 70219208,
      "step": 121035
    },
    {
      "epoch": 18.028001191540067,
      "grad_norm": 5.59375,
      "learning_rate": 1.4663970646647152e-06,
      "loss": 0.6686,
      "num_input_tokens_seen": 70221992,
      "step": 121040
    },
    {
      "epoch": 18.028745904081024,
      "grad_norm": 7.375,
      "learning_rate": 1.4653007522037325e-06,
      "loss": 0.9475,
      "num_input_tokens_seen": 70224616,
      "step": 121045
    },
    {
      "epoch": 18.029490616621985,
      "grad_norm": 9.1875,
      "learning_rate": 1.4642048373375712e-06,
      "loss": 0.9188,
      "num_input_tokens_seen": 70227432,
      "step": 121050
    },
    {
      "epoch": 18.03023532916294,
      "grad_norm": 5.90625,
      "learning_rate": 1.4631093200847517e-06,
      "loss": 0.8031,
      "num_input_tokens_seen": 70230280,
      "step": 121055
    },
    {
      "epoch": 18.030980041703902,
      "grad_norm": 9.8125,
      "learning_rate": 1.4620142004637766e-06,
      "loss": 0.9998,
      "num_input_tokens_seen": 70233224,
      "step": 121060
    },
    {
      "epoch": 18.031724754244863,
      "grad_norm": 5.46875,
      "learning_rate": 1.4609194784931502e-06,
      "loss": 0.8934,
      "num_input_tokens_seen": 70235944,
      "step": 121065
    },
    {
      "epoch": 18.03246946678582,
      "grad_norm": 6.6875,
      "learning_rate": 1.459825154191366e-06,
      "loss": 1.1346,
      "num_input_tokens_seen": 70238664,
      "step": 121070
    },
    {
      "epoch": 18.03321417932678,
      "grad_norm": 6.375,
      "learning_rate": 1.4587312275769065e-06,
      "loss": 0.8345,
      "num_input_tokens_seen": 70241352,
      "step": 121075
    },
    {
      "epoch": 18.03395889186774,
      "grad_norm": 13.0,
      "learning_rate": 1.4576376986682598e-06,
      "loss": 1.0551,
      "num_input_tokens_seen": 70244456,
      "step": 121080
    },
    {
      "epoch": 18.034703604408698,
      "grad_norm": 16.125,
      "learning_rate": 1.4565445674838919e-06,
      "loss": 1.5168,
      "num_input_tokens_seen": 70247688,
      "step": 121085
    },
    {
      "epoch": 18.035448316949658,
      "grad_norm": 5.15625,
      "learning_rate": 1.455451834042279e-06,
      "loss": 0.9158,
      "num_input_tokens_seen": 70250984,
      "step": 121090
    },
    {
      "epoch": 18.036193029490615,
      "grad_norm": 19.25,
      "learning_rate": 1.4543594983618792e-06,
      "loss": 1.1091,
      "num_input_tokens_seen": 70253640,
      "step": 121095
    },
    {
      "epoch": 18.036937742031576,
      "grad_norm": 7.03125,
      "learning_rate": 1.4532675604611412e-06,
      "loss": 1.3069,
      "num_input_tokens_seen": 70256456,
      "step": 121100
    },
    {
      "epoch": 18.037682454572536,
      "grad_norm": 13.0625,
      "learning_rate": 1.45217602035852e-06,
      "loss": 1.3857,
      "num_input_tokens_seen": 70259048,
      "step": 121105
    },
    {
      "epoch": 18.038427167113493,
      "grad_norm": 7.8125,
      "learning_rate": 1.451084878072448e-06,
      "loss": 1.1219,
      "num_input_tokens_seen": 70262216,
      "step": 121110
    },
    {
      "epoch": 18.039171879654454,
      "grad_norm": 7.3125,
      "learning_rate": 1.4499941336213657e-06,
      "loss": 1.1012,
      "num_input_tokens_seen": 70265256,
      "step": 121115
    },
    {
      "epoch": 18.039916592195414,
      "grad_norm": 9.5,
      "learning_rate": 1.448903787023692e-06,
      "loss": 1.0394,
      "num_input_tokens_seen": 70268328,
      "step": 121120
    },
    {
      "epoch": 18.04066130473637,
      "grad_norm": 6.40625,
      "learning_rate": 1.4478138382978595e-06,
      "loss": 0.8711,
      "num_input_tokens_seen": 70271240,
      "step": 121125
    },
    {
      "epoch": 18.041406017277332,
      "grad_norm": 14.375,
      "learning_rate": 1.4467242874622726e-06,
      "loss": 1.3202,
      "num_input_tokens_seen": 70273960,
      "step": 121130
    },
    {
      "epoch": 18.04215072981829,
      "grad_norm": 12.5,
      "learning_rate": 1.445635134535342e-06,
      "loss": 0.9851,
      "num_input_tokens_seen": 70276744,
      "step": 121135
    },
    {
      "epoch": 18.04289544235925,
      "grad_norm": 6.125,
      "learning_rate": 1.4445463795354664e-06,
      "loss": 0.9846,
      "num_input_tokens_seen": 70279848,
      "step": 121140
    },
    {
      "epoch": 18.04364015490021,
      "grad_norm": 8.5625,
      "learning_rate": 1.443458022481034e-06,
      "loss": 0.956,
      "num_input_tokens_seen": 70282984,
      "step": 121145
    },
    {
      "epoch": 18.044384867441167,
      "grad_norm": 8.375,
      "learning_rate": 1.4423700633904414e-06,
      "loss": 1.2531,
      "num_input_tokens_seen": 70285960,
      "step": 121150
    },
    {
      "epoch": 18.045129579982127,
      "grad_norm": 22.625,
      "learning_rate": 1.4412825022820598e-06,
      "loss": 1.2817,
      "num_input_tokens_seen": 70288744,
      "step": 121155
    },
    {
      "epoch": 18.045874292523084,
      "grad_norm": 8.25,
      "learning_rate": 1.440195339174269e-06,
      "loss": 1.3869,
      "num_input_tokens_seen": 70291528,
      "step": 121160
    },
    {
      "epoch": 18.046619005064045,
      "grad_norm": 6.5,
      "learning_rate": 1.4391085740854293e-06,
      "loss": 0.9686,
      "num_input_tokens_seen": 70294600,
      "step": 121165
    },
    {
      "epoch": 18.047363717605005,
      "grad_norm": 12.9375,
      "learning_rate": 1.4380222070339095e-06,
      "loss": 1.3052,
      "num_input_tokens_seen": 70297512,
      "step": 121170
    },
    {
      "epoch": 18.048108430145962,
      "grad_norm": 16.0,
      "learning_rate": 1.4369362380380558e-06,
      "loss": 1.1373,
      "num_input_tokens_seen": 70300488,
      "step": 121175
    },
    {
      "epoch": 18.048853142686923,
      "grad_norm": 6.6875,
      "learning_rate": 1.435850667116212e-06,
      "loss": 0.8581,
      "num_input_tokens_seen": 70303944,
      "step": 121180
    },
    {
      "epoch": 18.049597855227884,
      "grad_norm": 9.1875,
      "learning_rate": 1.4347654942867245e-06,
      "loss": 1.1492,
      "num_input_tokens_seen": 70307272,
      "step": 121185
    },
    {
      "epoch": 18.05034256776884,
      "grad_norm": 12.3125,
      "learning_rate": 1.4336807195679203e-06,
      "loss": 1.2962,
      "num_input_tokens_seen": 70310280,
      "step": 121190
    },
    {
      "epoch": 18.0510872803098,
      "grad_norm": 8.6875,
      "learning_rate": 1.4325963429781347e-06,
      "loss": 1.0351,
      "num_input_tokens_seen": 70312936,
      "step": 121195
    },
    {
      "epoch": 18.051831992850758,
      "grad_norm": 7.59375,
      "learning_rate": 1.4315123645356782e-06,
      "loss": 0.8956,
      "num_input_tokens_seen": 70315816,
      "step": 121200
    },
    {
      "epoch": 18.05257670539172,
      "grad_norm": 8.4375,
      "learning_rate": 1.4304287842588665e-06,
      "loss": 1.1676,
      "num_input_tokens_seen": 70318888,
      "step": 121205
    },
    {
      "epoch": 18.05332141793268,
      "grad_norm": 11.25,
      "learning_rate": 1.4293456021660018e-06,
      "loss": 1.3932,
      "num_input_tokens_seen": 70321928,
      "step": 121210
    },
    {
      "epoch": 18.054066130473636,
      "grad_norm": 11.0625,
      "learning_rate": 1.4282628182753915e-06,
      "loss": 1.0634,
      "num_input_tokens_seen": 70324744,
      "step": 121215
    },
    {
      "epoch": 18.054810843014597,
      "grad_norm": 6.625,
      "learning_rate": 1.4271804326053239e-06,
      "loss": 1.1065,
      "num_input_tokens_seen": 70327720,
      "step": 121220
    },
    {
      "epoch": 18.055555555555557,
      "grad_norm": 7.9375,
      "learning_rate": 1.4260984451740815e-06,
      "loss": 0.9441,
      "num_input_tokens_seen": 70330568,
      "step": 121225
    },
    {
      "epoch": 18.056300268096514,
      "grad_norm": 16.5,
      "learning_rate": 1.4250168559999499e-06,
      "loss": 1.3445,
      "num_input_tokens_seen": 70333512,
      "step": 121230
    },
    {
      "epoch": 18.057044980637475,
      "grad_norm": 4.21875,
      "learning_rate": 1.423935665101192e-06,
      "loss": 0.7561,
      "num_input_tokens_seen": 70336712,
      "step": 121235
    },
    {
      "epoch": 18.05778969317843,
      "grad_norm": 7.65625,
      "learning_rate": 1.422854872496085e-06,
      "loss": 1.1126,
      "num_input_tokens_seen": 70339464,
      "step": 121240
    },
    {
      "epoch": 18.058534405719392,
      "grad_norm": 7.96875,
      "learning_rate": 1.421774478202878e-06,
      "loss": 0.7878,
      "num_input_tokens_seen": 70342184,
      "step": 121245
    },
    {
      "epoch": 18.059279118260353,
      "grad_norm": 11.1875,
      "learning_rate": 1.4206944822398316e-06,
      "loss": 1.0856,
      "num_input_tokens_seen": 70345032,
      "step": 121250
    },
    {
      "epoch": 18.06002383080131,
      "grad_norm": 13.875,
      "learning_rate": 1.4196148846251867e-06,
      "loss": 1.0883,
      "num_input_tokens_seen": 70348040,
      "step": 121255
    },
    {
      "epoch": 18.06076854334227,
      "grad_norm": 4.90625,
      "learning_rate": 1.4185356853771787e-06,
      "loss": 0.9477,
      "num_input_tokens_seen": 70350984,
      "step": 121260
    },
    {
      "epoch": 18.06151325588323,
      "grad_norm": 7.3125,
      "learning_rate": 1.4174568845140457e-06,
      "loss": 0.9379,
      "num_input_tokens_seen": 70353832,
      "step": 121265
    },
    {
      "epoch": 18.062257968424188,
      "grad_norm": 9.5,
      "learning_rate": 1.416378482054012e-06,
      "loss": 1.1258,
      "num_input_tokens_seen": 70356904,
      "step": 121270
    },
    {
      "epoch": 18.06300268096515,
      "grad_norm": 17.25,
      "learning_rate": 1.4153004780152939e-06,
      "loss": 1.5019,
      "num_input_tokens_seen": 70359592,
      "step": 121275
    },
    {
      "epoch": 18.063747393506105,
      "grad_norm": 18.5,
      "learning_rate": 1.414222872416099e-06,
      "loss": 1.1037,
      "num_input_tokens_seen": 70362632,
      "step": 121280
    },
    {
      "epoch": 18.064492106047066,
      "grad_norm": 9.625,
      "learning_rate": 1.4131456652746428e-06,
      "loss": 0.8883,
      "num_input_tokens_seen": 70365544,
      "step": 121285
    },
    {
      "epoch": 18.065236818588026,
      "grad_norm": 6.9375,
      "learning_rate": 1.4120688566091112e-06,
      "loss": 0.811,
      "num_input_tokens_seen": 70368232,
      "step": 121290
    },
    {
      "epoch": 18.065981531128983,
      "grad_norm": 9.1875,
      "learning_rate": 1.410992446437709e-06,
      "loss": 0.7754,
      "num_input_tokens_seen": 70371080,
      "step": 121295
    },
    {
      "epoch": 18.066726243669944,
      "grad_norm": 7.34375,
      "learning_rate": 1.4099164347786132e-06,
      "loss": 1.1107,
      "num_input_tokens_seen": 70373672,
      "step": 121300
    },
    {
      "epoch": 18.0674709562109,
      "grad_norm": 13.3125,
      "learning_rate": 1.408840821650001e-06,
      "loss": 1.0923,
      "num_input_tokens_seen": 70376264,
      "step": 121305
    },
    {
      "epoch": 18.06821566875186,
      "grad_norm": 8.5,
      "learning_rate": 1.4077656070700496e-06,
      "loss": 0.9861,
      "num_input_tokens_seen": 70379464,
      "step": 121310
    },
    {
      "epoch": 18.068960381292822,
      "grad_norm": 10.4375,
      "learning_rate": 1.4066907910569166e-06,
      "loss": 1.115,
      "num_input_tokens_seen": 70382600,
      "step": 121315
    },
    {
      "epoch": 18.06970509383378,
      "grad_norm": 11.0625,
      "learning_rate": 1.4056163736287682e-06,
      "loss": 1.1068,
      "num_input_tokens_seen": 70385480,
      "step": 121320
    },
    {
      "epoch": 18.07044980637474,
      "grad_norm": 12.9375,
      "learning_rate": 1.4045423548037478e-06,
      "loss": 1.0654,
      "num_input_tokens_seen": 70388616,
      "step": 121325
    },
    {
      "epoch": 18.0711945189157,
      "grad_norm": 10.5,
      "learning_rate": 1.4034687346000052e-06,
      "loss": 0.9024,
      "num_input_tokens_seen": 70391496,
      "step": 121330
    },
    {
      "epoch": 18.071939231456657,
      "grad_norm": 15.8125,
      "learning_rate": 1.4023955130356758e-06,
      "loss": 0.8964,
      "num_input_tokens_seen": 70394440,
      "step": 121335
    },
    {
      "epoch": 18.072683943997617,
      "grad_norm": 6.78125,
      "learning_rate": 1.4013226901288868e-06,
      "loss": 1.2078,
      "num_input_tokens_seen": 70397192,
      "step": 121340
    },
    {
      "epoch": 18.073428656538574,
      "grad_norm": 11.1875,
      "learning_rate": 1.400250265897768e-06,
      "loss": 1.1127,
      "num_input_tokens_seen": 70400264,
      "step": 121345
    },
    {
      "epoch": 18.074173369079535,
      "grad_norm": 13.875,
      "learning_rate": 1.3991782403604353e-06,
      "loss": 0.944,
      "num_input_tokens_seen": 70403144,
      "step": 121350
    },
    {
      "epoch": 18.074918081620496,
      "grad_norm": 7.78125,
      "learning_rate": 1.3981066135349995e-06,
      "loss": 1.19,
      "num_input_tokens_seen": 70406088,
      "step": 121355
    },
    {
      "epoch": 18.075662794161452,
      "grad_norm": 10.375,
      "learning_rate": 1.397035385439563e-06,
      "loss": 1.1804,
      "num_input_tokens_seen": 70409096,
      "step": 121360
    },
    {
      "epoch": 18.076407506702413,
      "grad_norm": 7.90625,
      "learning_rate": 1.3959645560922275e-06,
      "loss": 0.9589,
      "num_input_tokens_seen": 70411848,
      "step": 121365
    },
    {
      "epoch": 18.077152219243374,
      "grad_norm": 10.1875,
      "learning_rate": 1.3948941255110787e-06,
      "loss": 1.3002,
      "num_input_tokens_seen": 70414952,
      "step": 121370
    },
    {
      "epoch": 18.07789693178433,
      "grad_norm": 23.375,
      "learning_rate": 1.393824093714205e-06,
      "loss": 1.1946,
      "num_input_tokens_seen": 70418216,
      "step": 121375
    },
    {
      "epoch": 18.07864164432529,
      "grad_norm": 7.6875,
      "learning_rate": 1.3927544607196807e-06,
      "loss": 0.5541,
      "num_input_tokens_seen": 70420904,
      "step": 121380
    },
    {
      "epoch": 18.079386356866248,
      "grad_norm": 16.125,
      "learning_rate": 1.3916852265455722e-06,
      "loss": 1.4532,
      "num_input_tokens_seen": 70423752,
      "step": 121385
    },
    {
      "epoch": 18.08013106940721,
      "grad_norm": 10.1875,
      "learning_rate": 1.3906163912099506e-06,
      "loss": 1.0518,
      "num_input_tokens_seen": 70426600,
      "step": 121390
    },
    {
      "epoch": 18.08087578194817,
      "grad_norm": 9.8125,
      "learning_rate": 1.3895479547308716e-06,
      "loss": 0.8843,
      "num_input_tokens_seen": 70429416,
      "step": 121395
    },
    {
      "epoch": 18.081620494489126,
      "grad_norm": 16.125,
      "learning_rate": 1.3884799171263841e-06,
      "loss": 1.0291,
      "num_input_tokens_seen": 70432168,
      "step": 121400
    },
    {
      "epoch": 18.082365207030087,
      "grad_norm": 5.46875,
      "learning_rate": 1.3874122784145239e-06,
      "loss": 0.8603,
      "num_input_tokens_seen": 70434920,
      "step": 121405
    },
    {
      "epoch": 18.083109919571047,
      "grad_norm": 20.75,
      "learning_rate": 1.3863450386133402e-06,
      "loss": 1.0868,
      "num_input_tokens_seen": 70438088,
      "step": 121410
    },
    {
      "epoch": 18.083854632112004,
      "grad_norm": 10.0,
      "learning_rate": 1.385278197740858e-06,
      "loss": 0.7319,
      "num_input_tokens_seen": 70440840,
      "step": 121415
    },
    {
      "epoch": 18.084599344652965,
      "grad_norm": 8.5,
      "learning_rate": 1.384211755815093e-06,
      "loss": 0.9898,
      "num_input_tokens_seen": 70443720,
      "step": 121420
    },
    {
      "epoch": 18.08534405719392,
      "grad_norm": 15.1875,
      "learning_rate": 1.3831457128540753e-06,
      "loss": 1.1142,
      "num_input_tokens_seen": 70446728,
      "step": 121425
    },
    {
      "epoch": 18.086088769734882,
      "grad_norm": 11.75,
      "learning_rate": 1.3820800688758018e-06,
      "loss": 0.8816,
      "num_input_tokens_seen": 70449736,
      "step": 121430
    },
    {
      "epoch": 18.086833482275843,
      "grad_norm": 10.0,
      "learning_rate": 1.3810148238982857e-06,
      "loss": 0.8263,
      "num_input_tokens_seen": 70452616,
      "step": 121435
    },
    {
      "epoch": 18.0875781948168,
      "grad_norm": 8.875,
      "learning_rate": 1.3799499779395152e-06,
      "loss": 1.1354,
      "num_input_tokens_seen": 70455592,
      "step": 121440
    },
    {
      "epoch": 18.08832290735776,
      "grad_norm": 11.625,
      "learning_rate": 1.3788855310174876e-06,
      "loss": 1.0257,
      "num_input_tokens_seen": 70458696,
      "step": 121445
    },
    {
      "epoch": 18.08906761989872,
      "grad_norm": 6.875,
      "learning_rate": 1.3778214831501767e-06,
      "loss": 0.9722,
      "num_input_tokens_seen": 70461608,
      "step": 121450
    },
    {
      "epoch": 18.089812332439678,
      "grad_norm": 6.03125,
      "learning_rate": 1.3767578343555688e-06,
      "loss": 0.9786,
      "num_input_tokens_seen": 70464584,
      "step": 121455
    },
    {
      "epoch": 18.09055704498064,
      "grad_norm": 21.5,
      "learning_rate": 1.3756945846516267e-06,
      "loss": 1.3117,
      "num_input_tokens_seen": 70467592,
      "step": 121460
    },
    {
      "epoch": 18.091301757521595,
      "grad_norm": 15.625,
      "learning_rate": 1.3746317340563142e-06,
      "loss": 1.1982,
      "num_input_tokens_seen": 70470568,
      "step": 121465
    },
    {
      "epoch": 18.092046470062556,
      "grad_norm": 11.5625,
      "learning_rate": 1.3735692825875861e-06,
      "loss": 0.9402,
      "num_input_tokens_seen": 70473608,
      "step": 121470
    },
    {
      "epoch": 18.092791182603516,
      "grad_norm": 8.75,
      "learning_rate": 1.3725072302633895e-06,
      "loss": 1.0174,
      "num_input_tokens_seen": 70476552,
      "step": 121475
    },
    {
      "epoch": 18.093535895144473,
      "grad_norm": 10.9375,
      "learning_rate": 1.371445577101671e-06,
      "loss": 1.1763,
      "num_input_tokens_seen": 70479432,
      "step": 121480
    },
    {
      "epoch": 18.094280607685434,
      "grad_norm": 9.9375,
      "learning_rate": 1.3703843231203634e-06,
      "loss": 0.8796,
      "num_input_tokens_seen": 70482344,
      "step": 121485
    },
    {
      "epoch": 18.09502532022639,
      "grad_norm": 8.375,
      "learning_rate": 1.3693234683373997e-06,
      "loss": 1.0182,
      "num_input_tokens_seen": 70485032,
      "step": 121490
    },
    {
      "epoch": 18.09577003276735,
      "grad_norm": 12.6875,
      "learning_rate": 1.3682630127706958e-06,
      "loss": 0.7354,
      "num_input_tokens_seen": 70487944,
      "step": 121495
    },
    {
      "epoch": 18.096514745308312,
      "grad_norm": 8.75,
      "learning_rate": 1.3672029564381711e-06,
      "loss": 0.7545,
      "num_input_tokens_seen": 70491080,
      "step": 121500
    },
    {
      "epoch": 18.09725945784927,
      "grad_norm": 9.6875,
      "learning_rate": 1.3661432993577333e-06,
      "loss": 0.9389,
      "num_input_tokens_seen": 70493896,
      "step": 121505
    },
    {
      "epoch": 18.09800417039023,
      "grad_norm": 8.1875,
      "learning_rate": 1.365084041547282e-06,
      "loss": 1.177,
      "num_input_tokens_seen": 70496520,
      "step": 121510
    },
    {
      "epoch": 18.09874888293119,
      "grad_norm": 13.875,
      "learning_rate": 1.364025183024717e-06,
      "loss": 0.9235,
      "num_input_tokens_seen": 70499592,
      "step": 121515
    },
    {
      "epoch": 18.099493595472147,
      "grad_norm": 10.5,
      "learning_rate": 1.3629667238079208e-06,
      "loss": 1.1996,
      "num_input_tokens_seen": 70502440,
      "step": 121520
    },
    {
      "epoch": 18.100238308013108,
      "grad_norm": 8.75,
      "learning_rate": 1.361908663914782e-06,
      "loss": 0.8378,
      "num_input_tokens_seen": 70505384,
      "step": 121525
    },
    {
      "epoch": 18.100983020554064,
      "grad_norm": 7.15625,
      "learning_rate": 1.3608510033631728e-06,
      "loss": 0.8303,
      "num_input_tokens_seen": 70508104,
      "step": 121530
    },
    {
      "epoch": 18.101727733095025,
      "grad_norm": 9.8125,
      "learning_rate": 1.3597937421709588e-06,
      "loss": 1.3035,
      "num_input_tokens_seen": 70510728,
      "step": 121535
    },
    {
      "epoch": 18.102472445635986,
      "grad_norm": 8.625,
      "learning_rate": 1.358736880356004e-06,
      "loss": 1.0483,
      "num_input_tokens_seen": 70513864,
      "step": 121540
    },
    {
      "epoch": 18.103217158176943,
      "grad_norm": 10.125,
      "learning_rate": 1.3576804179361552e-06,
      "loss": 1.1415,
      "num_input_tokens_seen": 70516520,
      "step": 121545
    },
    {
      "epoch": 18.103961870717903,
      "grad_norm": 6.6875,
      "learning_rate": 1.356624354929273e-06,
      "loss": 1.2211,
      "num_input_tokens_seen": 70519336,
      "step": 121550
    },
    {
      "epoch": 18.104706583258864,
      "grad_norm": 10.875,
      "learning_rate": 1.3555686913531874e-06,
      "loss": 1.0554,
      "num_input_tokens_seen": 70521960,
      "step": 121555
    },
    {
      "epoch": 18.10545129579982,
      "grad_norm": 7.59375,
      "learning_rate": 1.3545134272257426e-06,
      "loss": 1.1388,
      "num_input_tokens_seen": 70525192,
      "step": 121560
    },
    {
      "epoch": 18.10619600834078,
      "grad_norm": 16.875,
      "learning_rate": 1.3534585625647578e-06,
      "loss": 1.3029,
      "num_input_tokens_seen": 70527976,
      "step": 121565
    },
    {
      "epoch": 18.106940720881738,
      "grad_norm": 10.5,
      "learning_rate": 1.3524040973880575e-06,
      "loss": 0.9029,
      "num_input_tokens_seen": 70530888,
      "step": 121570
    },
    {
      "epoch": 18.1076854334227,
      "grad_norm": 5.375,
      "learning_rate": 1.3513500317134582e-06,
      "loss": 1.1183,
      "num_input_tokens_seen": 70533768,
      "step": 121575
    },
    {
      "epoch": 18.10843014596366,
      "grad_norm": 13.625,
      "learning_rate": 1.3502963655587619e-06,
      "loss": 1.0502,
      "num_input_tokens_seen": 70536776,
      "step": 121580
    },
    {
      "epoch": 18.109174858504616,
      "grad_norm": 6.3125,
      "learning_rate": 1.3492430989417742e-06,
      "loss": 0.8346,
      "num_input_tokens_seen": 70539752,
      "step": 121585
    },
    {
      "epoch": 18.109919571045577,
      "grad_norm": 11.125,
      "learning_rate": 1.3481902318802835e-06,
      "loss": 1.1277,
      "num_input_tokens_seen": 70542824,
      "step": 121590
    },
    {
      "epoch": 18.110664283586537,
      "grad_norm": 15.1875,
      "learning_rate": 1.3471377643920841e-06,
      "loss": 1.2686,
      "num_input_tokens_seen": 70545704,
      "step": 121595
    },
    {
      "epoch": 18.111408996127494,
      "grad_norm": 6.0,
      "learning_rate": 1.346085696494953e-06,
      "loss": 1.0289,
      "num_input_tokens_seen": 70548456,
      "step": 121600
    },
    {
      "epoch": 18.112153708668455,
      "grad_norm": 7.375,
      "learning_rate": 1.3450340282066625e-06,
      "loss": 0.8201,
      "num_input_tokens_seen": 70551560,
      "step": 121605
    },
    {
      "epoch": 18.11289842120941,
      "grad_norm": 3.6875,
      "learning_rate": 1.3439827595449761e-06,
      "loss": 0.874,
      "num_input_tokens_seen": 70554152,
      "step": 121610
    },
    {
      "epoch": 18.113643133750372,
      "grad_norm": 13.4375,
      "learning_rate": 1.3429318905276627e-06,
      "loss": 0.9269,
      "num_input_tokens_seen": 70556968,
      "step": 121615
    },
    {
      "epoch": 18.114387846291333,
      "grad_norm": 11.0,
      "learning_rate": 1.3418814211724695e-06,
      "loss": 1.0552,
      "num_input_tokens_seen": 70559976,
      "step": 121620
    },
    {
      "epoch": 18.11513255883229,
      "grad_norm": 10.875,
      "learning_rate": 1.3408313514971432e-06,
      "loss": 1.1388,
      "num_input_tokens_seen": 70562888,
      "step": 121625
    },
    {
      "epoch": 18.11587727137325,
      "grad_norm": 15.5,
      "learning_rate": 1.3397816815194281e-06,
      "loss": 1.1309,
      "num_input_tokens_seen": 70565800,
      "step": 121630
    },
    {
      "epoch": 18.11662198391421,
      "grad_norm": 11.25,
      "learning_rate": 1.3387324112570488e-06,
      "loss": 1.0297,
      "num_input_tokens_seen": 70568776,
      "step": 121635
    },
    {
      "epoch": 18.117366696455168,
      "grad_norm": 5.625,
      "learning_rate": 1.3376835407277437e-06,
      "loss": 0.9041,
      "num_input_tokens_seen": 70571528,
      "step": 121640
    },
    {
      "epoch": 18.11811140899613,
      "grad_norm": 9.625,
      "learning_rate": 1.3366350699492214e-06,
      "loss": 1.4669,
      "num_input_tokens_seen": 70574280,
      "step": 121645
    },
    {
      "epoch": 18.118856121537085,
      "grad_norm": 7.125,
      "learning_rate": 1.3355869989392005e-06,
      "loss": 1.068,
      "num_input_tokens_seen": 70577032,
      "step": 121650
    },
    {
      "epoch": 18.119600834078046,
      "grad_norm": 11.625,
      "learning_rate": 1.3345393277153896e-06,
      "loss": 1.0055,
      "num_input_tokens_seen": 70579752,
      "step": 121655
    },
    {
      "epoch": 18.120345546619006,
      "grad_norm": 17.25,
      "learning_rate": 1.3334920562954822e-06,
      "loss": 1.0652,
      "num_input_tokens_seen": 70582504,
      "step": 121660
    },
    {
      "epoch": 18.121090259159963,
      "grad_norm": 9.125,
      "learning_rate": 1.332445184697173e-06,
      "loss": 1.0693,
      "num_input_tokens_seen": 70585352,
      "step": 121665
    },
    {
      "epoch": 18.121834971700924,
      "grad_norm": 6.78125,
      "learning_rate": 1.3313987129381422e-06,
      "loss": 1.1203,
      "num_input_tokens_seen": 70588232,
      "step": 121670
    },
    {
      "epoch": 18.12257968424188,
      "grad_norm": 11.0625,
      "learning_rate": 1.3303526410360811e-06,
      "loss": 1.0506,
      "num_input_tokens_seen": 70591208,
      "step": 121675
    },
    {
      "epoch": 18.12332439678284,
      "grad_norm": 10.3125,
      "learning_rate": 1.3293069690086506e-06,
      "loss": 1.1769,
      "num_input_tokens_seen": 70594184,
      "step": 121680
    },
    {
      "epoch": 18.124069109323802,
      "grad_norm": 7.125,
      "learning_rate": 1.3282616968735256e-06,
      "loss": 0.9564,
      "num_input_tokens_seen": 70597224,
      "step": 121685
    },
    {
      "epoch": 18.12481382186476,
      "grad_norm": 13.75,
      "learning_rate": 1.3272168246483557e-06,
      "loss": 1.041,
      "num_input_tokens_seen": 70599688,
      "step": 121690
    },
    {
      "epoch": 18.12555853440572,
      "grad_norm": 8.75,
      "learning_rate": 1.3261723523508018e-06,
      "loss": 0.8874,
      "num_input_tokens_seen": 70602984,
      "step": 121695
    },
    {
      "epoch": 18.12630324694668,
      "grad_norm": 8.4375,
      "learning_rate": 1.3251282799985026e-06,
      "loss": 0.7448,
      "num_input_tokens_seen": 70605768,
      "step": 121700
    },
    {
      "epoch": 18.127047959487637,
      "grad_norm": 10.0,
      "learning_rate": 1.3240846076090996e-06,
      "loss": 0.8759,
      "num_input_tokens_seen": 70608968,
      "step": 121705
    },
    {
      "epoch": 18.127792672028598,
      "grad_norm": 6.46875,
      "learning_rate": 1.3230413352002259e-06,
      "loss": 0.696,
      "num_input_tokens_seen": 70611848,
      "step": 121710
    },
    {
      "epoch": 18.128537384569555,
      "grad_norm": 13.9375,
      "learning_rate": 1.3219984627895005e-06,
      "loss": 1.1769,
      "num_input_tokens_seen": 70614440,
      "step": 121715
    },
    {
      "epoch": 18.129282097110515,
      "grad_norm": 10.75,
      "learning_rate": 1.3209559903945483e-06,
      "loss": 1.0194,
      "num_input_tokens_seen": 70617160,
      "step": 121720
    },
    {
      "epoch": 18.130026809651476,
      "grad_norm": 12.0625,
      "learning_rate": 1.3199139180329806e-06,
      "loss": 1.2683,
      "num_input_tokens_seen": 70619816,
      "step": 121725
    },
    {
      "epoch": 18.130771522192433,
      "grad_norm": 7.96875,
      "learning_rate": 1.3188722457223995e-06,
      "loss": 1.1947,
      "num_input_tokens_seen": 70622504,
      "step": 121730
    },
    {
      "epoch": 18.131516234733393,
      "grad_norm": 10.6875,
      "learning_rate": 1.3178309734803968e-06,
      "loss": 1.2891,
      "num_input_tokens_seen": 70625320,
      "step": 121735
    },
    {
      "epoch": 18.132260947274354,
      "grad_norm": 5.1875,
      "learning_rate": 1.3167901013245775e-06,
      "loss": 0.8901,
      "num_input_tokens_seen": 70628488,
      "step": 121740
    },
    {
      "epoch": 18.13300565981531,
      "grad_norm": 12.1875,
      "learning_rate": 1.3157496292725169e-06,
      "loss": 1.038,
      "num_input_tokens_seen": 70631464,
      "step": 121745
    },
    {
      "epoch": 18.13375037235627,
      "grad_norm": 7.4375,
      "learning_rate": 1.3147095573417923e-06,
      "loss": 1.2549,
      "num_input_tokens_seen": 70634184,
      "step": 121750
    },
    {
      "epoch": 18.134495084897228,
      "grad_norm": 6.28125,
      "learning_rate": 1.3136698855499812e-06,
      "loss": 1.1771,
      "num_input_tokens_seen": 70637128,
      "step": 121755
    },
    {
      "epoch": 18.13523979743819,
      "grad_norm": 13.1875,
      "learning_rate": 1.3126306139146394e-06,
      "loss": 1.2385,
      "num_input_tokens_seen": 70640008,
      "step": 121760
    },
    {
      "epoch": 18.13598450997915,
      "grad_norm": 10.3125,
      "learning_rate": 1.3115917424533303e-06,
      "loss": 1.1379,
      "num_input_tokens_seen": 70642920,
      "step": 121765
    },
    {
      "epoch": 18.136729222520106,
      "grad_norm": 15.75,
      "learning_rate": 1.310553271183601e-06,
      "loss": 1.5123,
      "num_input_tokens_seen": 70646216,
      "step": 121770
    },
    {
      "epoch": 18.137473935061067,
      "grad_norm": 19.5,
      "learning_rate": 1.3095152001230015e-06,
      "loss": 0.9962,
      "num_input_tokens_seen": 70649032,
      "step": 121775
    },
    {
      "epoch": 18.138218647602027,
      "grad_norm": 10.6875,
      "learning_rate": 1.308477529289065e-06,
      "loss": 1.2438,
      "num_input_tokens_seen": 70651688,
      "step": 121780
    },
    {
      "epoch": 18.138963360142984,
      "grad_norm": 15.9375,
      "learning_rate": 1.3074402586993162e-06,
      "loss": 1.1157,
      "num_input_tokens_seen": 70654536,
      "step": 121785
    },
    {
      "epoch": 18.139708072683945,
      "grad_norm": 12.4375,
      "learning_rate": 1.3064033883712883e-06,
      "loss": 1.2322,
      "num_input_tokens_seen": 70657384,
      "step": 121790
    },
    {
      "epoch": 18.140452785224902,
      "grad_norm": 16.25,
      "learning_rate": 1.3053669183224954e-06,
      "loss": 1.1323,
      "num_input_tokens_seen": 70660264,
      "step": 121795
    },
    {
      "epoch": 18.141197497765862,
      "grad_norm": 19.25,
      "learning_rate": 1.3043308485704454e-06,
      "loss": 1.136,
      "num_input_tokens_seen": 70663016,
      "step": 121800
    },
    {
      "epoch": 18.141942210306823,
      "grad_norm": 16.375,
      "learning_rate": 1.303295179132638e-06,
      "loss": 1.4652,
      "num_input_tokens_seen": 70665736,
      "step": 121805
    },
    {
      "epoch": 18.14268692284778,
      "grad_norm": 13.8125,
      "learning_rate": 1.3022599100265791e-06,
      "loss": 1.3604,
      "num_input_tokens_seen": 70668840,
      "step": 121810
    },
    {
      "epoch": 18.14343163538874,
      "grad_norm": 8.0,
      "learning_rate": 1.3012250412697519e-06,
      "loss": 1.1688,
      "num_input_tokens_seen": 70671848,
      "step": 121815
    },
    {
      "epoch": 18.1441763479297,
      "grad_norm": 4.09375,
      "learning_rate": 1.300190572879642e-06,
      "loss": 0.7892,
      "num_input_tokens_seen": 70674344,
      "step": 121820
    },
    {
      "epoch": 18.144921060470658,
      "grad_norm": 11.4375,
      "learning_rate": 1.2991565048737248e-06,
      "loss": 1.3718,
      "num_input_tokens_seen": 70677608,
      "step": 121825
    },
    {
      "epoch": 18.14566577301162,
      "grad_norm": 10.5625,
      "learning_rate": 1.2981228372694692e-06,
      "loss": 1.318,
      "num_input_tokens_seen": 70680360,
      "step": 121830
    },
    {
      "epoch": 18.146410485552575,
      "grad_norm": 10.6875,
      "learning_rate": 1.2970895700843394e-06,
      "loss": 1.4621,
      "num_input_tokens_seen": 70683272,
      "step": 121835
    },
    {
      "epoch": 18.147155198093536,
      "grad_norm": 7.71875,
      "learning_rate": 1.296056703335788e-06,
      "loss": 0.9149,
      "num_input_tokens_seen": 70686216,
      "step": 121840
    },
    {
      "epoch": 18.147899910634496,
      "grad_norm": 12.0,
      "learning_rate": 1.2950242370412702e-06,
      "loss": 1.074,
      "num_input_tokens_seen": 70689032,
      "step": 121845
    },
    {
      "epoch": 18.148644623175453,
      "grad_norm": 7.75,
      "learning_rate": 1.2939921712182223e-06,
      "loss": 1.1316,
      "num_input_tokens_seen": 70692072,
      "step": 121850
    },
    {
      "epoch": 18.149389335716414,
      "grad_norm": 8.1875,
      "learning_rate": 1.2929605058840889e-06,
      "loss": 1.0812,
      "num_input_tokens_seen": 70694888,
      "step": 121855
    },
    {
      "epoch": 18.15013404825737,
      "grad_norm": 7.59375,
      "learning_rate": 1.2919292410562889e-06,
      "loss": 1.1075,
      "num_input_tokens_seen": 70697896,
      "step": 121860
    },
    {
      "epoch": 18.15087876079833,
      "grad_norm": 10.9375,
      "learning_rate": 1.2908983767522504e-06,
      "loss": 1.1641,
      "num_input_tokens_seen": 70700680,
      "step": 121865
    },
    {
      "epoch": 18.151623473339292,
      "grad_norm": 11.0,
      "learning_rate": 1.289867912989387e-06,
      "loss": 0.8977,
      "num_input_tokens_seen": 70703304,
      "step": 121870
    },
    {
      "epoch": 18.15236818588025,
      "grad_norm": 14.75,
      "learning_rate": 1.2888378497851045e-06,
      "loss": 1.1933,
      "num_input_tokens_seen": 70706472,
      "step": 121875
    },
    {
      "epoch": 18.15311289842121,
      "grad_norm": 12.75,
      "learning_rate": 1.2878081871568082e-06,
      "loss": 1.1227,
      "num_input_tokens_seen": 70709640,
      "step": 121880
    },
    {
      "epoch": 18.15385761096217,
      "grad_norm": 15.0,
      "learning_rate": 1.2867789251218925e-06,
      "loss": 1.1963,
      "num_input_tokens_seen": 70712360,
      "step": 121885
    },
    {
      "epoch": 18.154602323503127,
      "grad_norm": 13.8125,
      "learning_rate": 1.2857500636977466e-06,
      "loss": 1.4661,
      "num_input_tokens_seen": 70715080,
      "step": 121890
    },
    {
      "epoch": 18.155347036044088,
      "grad_norm": 6.28125,
      "learning_rate": 1.2847216029017506e-06,
      "loss": 1.1329,
      "num_input_tokens_seen": 70717992,
      "step": 121895
    },
    {
      "epoch": 18.156091748585045,
      "grad_norm": 6.375,
      "learning_rate": 1.2836935427512826e-06,
      "loss": 0.8964,
      "num_input_tokens_seen": 70720744,
      "step": 121900
    },
    {
      "epoch": 18.156836461126005,
      "grad_norm": 8.875,
      "learning_rate": 1.2826658832637062e-06,
      "loss": 1.3017,
      "num_input_tokens_seen": 70723560,
      "step": 121905
    },
    {
      "epoch": 18.157581173666966,
      "grad_norm": 6.25,
      "learning_rate": 1.2816386244563827e-06,
      "loss": 0.9871,
      "num_input_tokens_seen": 70726760,
      "step": 121910
    },
    {
      "epoch": 18.158325886207923,
      "grad_norm": 12.3125,
      "learning_rate": 1.2806117663466704e-06,
      "loss": 1.3087,
      "num_input_tokens_seen": 70729736,
      "step": 121915
    },
    {
      "epoch": 18.159070598748883,
      "grad_norm": 4.65625,
      "learning_rate": 1.279585308951914e-06,
      "loss": 0.6913,
      "num_input_tokens_seen": 70732424,
      "step": 121920
    },
    {
      "epoch": 18.159815311289844,
      "grad_norm": 13.25,
      "learning_rate": 1.2785592522894573e-06,
      "loss": 1.3055,
      "num_input_tokens_seen": 70735400,
      "step": 121925
    },
    {
      "epoch": 18.1605600238308,
      "grad_norm": 11.0,
      "learning_rate": 1.2775335963766317e-06,
      "loss": 1.1327,
      "num_input_tokens_seen": 70738568,
      "step": 121930
    },
    {
      "epoch": 18.16130473637176,
      "grad_norm": 7.40625,
      "learning_rate": 1.2765083412307672e-06,
      "loss": 1.1375,
      "num_input_tokens_seen": 70741672,
      "step": 121935
    },
    {
      "epoch": 18.162049448912718,
      "grad_norm": 6.65625,
      "learning_rate": 1.2754834868691834e-06,
      "loss": 0.9959,
      "num_input_tokens_seen": 70744552,
      "step": 121940
    },
    {
      "epoch": 18.16279416145368,
      "grad_norm": 7.9375,
      "learning_rate": 1.2744590333091888e-06,
      "loss": 1.0745,
      "num_input_tokens_seen": 70747368,
      "step": 121945
    },
    {
      "epoch": 18.16353887399464,
      "grad_norm": 7.8125,
      "learning_rate": 1.2734349805680974e-06,
      "loss": 1.1315,
      "num_input_tokens_seen": 70750344,
      "step": 121950
    },
    {
      "epoch": 18.164283586535596,
      "grad_norm": 7.9375,
      "learning_rate": 1.2724113286632061e-06,
      "loss": 1.522,
      "num_input_tokens_seen": 70753448,
      "step": 121955
    },
    {
      "epoch": 18.165028299076557,
      "grad_norm": 4.71875,
      "learning_rate": 1.2713880776118126e-06,
      "loss": 0.6252,
      "num_input_tokens_seen": 70756328,
      "step": 121960
    },
    {
      "epoch": 18.165773011617517,
      "grad_norm": 10.8125,
      "learning_rate": 1.2703652274311973e-06,
      "loss": 1.2211,
      "num_input_tokens_seen": 70759400,
      "step": 121965
    },
    {
      "epoch": 18.166517724158474,
      "grad_norm": 8.1875,
      "learning_rate": 1.2693427781386464e-06,
      "loss": 1.1342,
      "num_input_tokens_seen": 70762216,
      "step": 121970
    },
    {
      "epoch": 18.167262436699435,
      "grad_norm": 7.875,
      "learning_rate": 1.2683207297514293e-06,
      "loss": 1.1908,
      "num_input_tokens_seen": 70765384,
      "step": 121975
    },
    {
      "epoch": 18.168007149240392,
      "grad_norm": 6.09375,
      "learning_rate": 1.267299082286816e-06,
      "loss": 0.8796,
      "num_input_tokens_seen": 70768296,
      "step": 121980
    },
    {
      "epoch": 18.168751861781352,
      "grad_norm": 16.375,
      "learning_rate": 1.2662778357620614e-06,
      "loss": 0.8465,
      "num_input_tokens_seen": 70771176,
      "step": 121985
    },
    {
      "epoch": 18.169496574322313,
      "grad_norm": 6.4375,
      "learning_rate": 1.2652569901944244e-06,
      "loss": 0.8242,
      "num_input_tokens_seen": 70774344,
      "step": 121990
    },
    {
      "epoch": 18.17024128686327,
      "grad_norm": 7.28125,
      "learning_rate": 1.2642365456011467e-06,
      "loss": 1.1083,
      "num_input_tokens_seen": 70777256,
      "step": 121995
    },
    {
      "epoch": 18.17098599940423,
      "grad_norm": 5.84375,
      "learning_rate": 1.2632165019994646e-06,
      "loss": 0.9557,
      "num_input_tokens_seen": 70780136,
      "step": 122000
    },
    {
      "epoch": 18.171730711945187,
      "grad_norm": 16.75,
      "learning_rate": 1.262196859406617e-06,
      "loss": 1.0556,
      "num_input_tokens_seen": 70782952,
      "step": 122005
    },
    {
      "epoch": 18.172475424486148,
      "grad_norm": 13.1875,
      "learning_rate": 1.261177617839826e-06,
      "loss": 1.2398,
      "num_input_tokens_seen": 70785640,
      "step": 122010
    },
    {
      "epoch": 18.17322013702711,
      "grad_norm": 5.28125,
      "learning_rate": 1.2601587773163142e-06,
      "loss": 1.2548,
      "num_input_tokens_seen": 70788456,
      "step": 122015
    },
    {
      "epoch": 18.173964849568065,
      "grad_norm": 13.0625,
      "learning_rate": 1.259140337853293e-06,
      "loss": 0.7371,
      "num_input_tokens_seen": 70791208,
      "step": 122020
    },
    {
      "epoch": 18.174709562109026,
      "grad_norm": 8.375,
      "learning_rate": 1.258122299467962e-06,
      "loss": 1.2751,
      "num_input_tokens_seen": 70794184,
      "step": 122025
    },
    {
      "epoch": 18.175454274649987,
      "grad_norm": 4.21875,
      "learning_rate": 1.2571046621775273e-06,
      "loss": 0.9883,
      "num_input_tokens_seen": 70797320,
      "step": 122030
    },
    {
      "epoch": 18.176198987190944,
      "grad_norm": 7.96875,
      "learning_rate": 1.2560874259991778e-06,
      "loss": 0.9681,
      "num_input_tokens_seen": 70800680,
      "step": 122035
    },
    {
      "epoch": 18.176943699731904,
      "grad_norm": 8.4375,
      "learning_rate": 1.2550705909500998e-06,
      "loss": 1.1685,
      "num_input_tokens_seen": 70803688,
      "step": 122040
    },
    {
      "epoch": 18.17768841227286,
      "grad_norm": 15.8125,
      "learning_rate": 1.2540541570474684e-06,
      "loss": 1.1506,
      "num_input_tokens_seen": 70806568,
      "step": 122045
    },
    {
      "epoch": 18.17843312481382,
      "grad_norm": 7.59375,
      "learning_rate": 1.2530381243084616e-06,
      "loss": 0.9818,
      "num_input_tokens_seen": 70809544,
      "step": 122050
    },
    {
      "epoch": 18.179177837354782,
      "grad_norm": 6.0,
      "learning_rate": 1.2520224927502405e-06,
      "loss": 1.1356,
      "num_input_tokens_seen": 70812232,
      "step": 122055
    },
    {
      "epoch": 18.17992254989574,
      "grad_norm": 6.0625,
      "learning_rate": 1.251007262389961e-06,
      "loss": 0.7635,
      "num_input_tokens_seen": 70815080,
      "step": 122060
    },
    {
      "epoch": 18.1806672624367,
      "grad_norm": 6.4375,
      "learning_rate": 1.2499924332447788e-06,
      "loss": 1.1751,
      "num_input_tokens_seen": 70818088,
      "step": 122065
    },
    {
      "epoch": 18.18141197497766,
      "grad_norm": 6.03125,
      "learning_rate": 1.24897800533183e-06,
      "loss": 1.1467,
      "num_input_tokens_seen": 70821224,
      "step": 122070
    },
    {
      "epoch": 18.182156687518617,
      "grad_norm": 19.375,
      "learning_rate": 1.247963978668265e-06,
      "loss": 1.2218,
      "num_input_tokens_seen": 70824424,
      "step": 122075
    },
    {
      "epoch": 18.182901400059578,
      "grad_norm": 11.875,
      "learning_rate": 1.2469503532712008e-06,
      "loss": 1.0322,
      "num_input_tokens_seen": 70827368,
      "step": 122080
    },
    {
      "epoch": 18.183646112600535,
      "grad_norm": 9.375,
      "learning_rate": 1.2459371291577759e-06,
      "loss": 0.9053,
      "num_input_tokens_seen": 70830120,
      "step": 122085
    },
    {
      "epoch": 18.184390825141495,
      "grad_norm": 10.4375,
      "learning_rate": 1.2449243063450967e-06,
      "loss": 1.1124,
      "num_input_tokens_seen": 70833032,
      "step": 122090
    },
    {
      "epoch": 18.185135537682456,
      "grad_norm": 17.0,
      "learning_rate": 1.2439118848502796e-06,
      "loss": 0.9576,
      "num_input_tokens_seen": 70836008,
      "step": 122095
    },
    {
      "epoch": 18.185880250223413,
      "grad_norm": 9.6875,
      "learning_rate": 1.2428998646904277e-06,
      "loss": 0.998,
      "num_input_tokens_seen": 70838984,
      "step": 122100
    },
    {
      "epoch": 18.186624962764373,
      "grad_norm": 11.3125,
      "learning_rate": 1.241888245882633e-06,
      "loss": 1.4781,
      "num_input_tokens_seen": 70842152,
      "step": 122105
    },
    {
      "epoch": 18.187369675305334,
      "grad_norm": 26.25,
      "learning_rate": 1.240877028443993e-06,
      "loss": 1.2098,
      "num_input_tokens_seen": 70845160,
      "step": 122110
    },
    {
      "epoch": 18.18811438784629,
      "grad_norm": 16.75,
      "learning_rate": 1.2398662123915827e-06,
      "loss": 1.2824,
      "num_input_tokens_seen": 70848360,
      "step": 122115
    },
    {
      "epoch": 18.18885910038725,
      "grad_norm": 17.375,
      "learning_rate": 1.2388557977424915e-06,
      "loss": 1.0415,
      "num_input_tokens_seen": 70851208,
      "step": 122120
    },
    {
      "epoch": 18.18960381292821,
      "grad_norm": 10.9375,
      "learning_rate": 1.2378457845137775e-06,
      "loss": 1.0485,
      "num_input_tokens_seen": 70853992,
      "step": 122125
    },
    {
      "epoch": 18.19034852546917,
      "grad_norm": 14.0625,
      "learning_rate": 1.236836172722511e-06,
      "loss": 0.8782,
      "num_input_tokens_seen": 70856904,
      "step": 122130
    },
    {
      "epoch": 18.19109323801013,
      "grad_norm": 19.75,
      "learning_rate": 1.2358269623857416e-06,
      "loss": 1.2813,
      "num_input_tokens_seen": 70859816,
      "step": 122135
    },
    {
      "epoch": 18.191837950551086,
      "grad_norm": 10.4375,
      "learning_rate": 1.2348181535205283e-06,
      "loss": 0.7859,
      "num_input_tokens_seen": 70862792,
      "step": 122140
    },
    {
      "epoch": 18.192582663092047,
      "grad_norm": 10.125,
      "learning_rate": 1.2338097461439047e-06,
      "loss": 1.2227,
      "num_input_tokens_seen": 70865832,
      "step": 122145
    },
    {
      "epoch": 18.193327375633007,
      "grad_norm": 7.65625,
      "learning_rate": 1.2328017402729098e-06,
      "loss": 0.9506,
      "num_input_tokens_seen": 70869704,
      "step": 122150
    },
    {
      "epoch": 18.194072088173964,
      "grad_norm": 5.71875,
      "learning_rate": 1.2317941359245744e-06,
      "loss": 1.0042,
      "num_input_tokens_seen": 70872872,
      "step": 122155
    },
    {
      "epoch": 18.194816800714925,
      "grad_norm": 10.9375,
      "learning_rate": 1.2307869331159182e-06,
      "loss": 0.9373,
      "num_input_tokens_seen": 70875816,
      "step": 122160
    },
    {
      "epoch": 18.195561513255882,
      "grad_norm": 7.9375,
      "learning_rate": 1.229780131863964e-06,
      "loss": 1.0495,
      "num_input_tokens_seen": 70878568,
      "step": 122165
    },
    {
      "epoch": 18.196306225796842,
      "grad_norm": 15.1875,
      "learning_rate": 1.2287737321857118e-06,
      "loss": 0.932,
      "num_input_tokens_seen": 70881640,
      "step": 122170
    },
    {
      "epoch": 18.197050938337803,
      "grad_norm": 14.3125,
      "learning_rate": 1.2277677340981675e-06,
      "loss": 0.835,
      "num_input_tokens_seen": 70884424,
      "step": 122175
    },
    {
      "epoch": 18.19779565087876,
      "grad_norm": 11.125,
      "learning_rate": 1.2267621376183286e-06,
      "loss": 1.0328,
      "num_input_tokens_seen": 70887112,
      "step": 122180
    },
    {
      "epoch": 18.19854036341972,
      "grad_norm": 4.5,
      "learning_rate": 1.2257569427631789e-06,
      "loss": 1.1171,
      "num_input_tokens_seen": 70889608,
      "step": 122185
    },
    {
      "epoch": 18.199285075960677,
      "grad_norm": 9.3125,
      "learning_rate": 1.2247521495497043e-06,
      "loss": 1.1882,
      "num_input_tokens_seen": 70892488,
      "step": 122190
    },
    {
      "epoch": 18.200029788501638,
      "grad_norm": 13.0625,
      "learning_rate": 1.223747757994878e-06,
      "loss": 1.27,
      "num_input_tokens_seen": 70895272,
      "step": 122195
    },
    {
      "epoch": 18.2007745010426,
      "grad_norm": 15.5,
      "learning_rate": 1.2227437681156695e-06,
      "loss": 1.1682,
      "num_input_tokens_seen": 70898216,
      "step": 122200
    },
    {
      "epoch": 18.201519213583556,
      "grad_norm": 6.59375,
      "learning_rate": 1.2217401799290345e-06,
      "loss": 1.3981,
      "num_input_tokens_seen": 70901160,
      "step": 122205
    },
    {
      "epoch": 18.202263926124516,
      "grad_norm": 15.0625,
      "learning_rate": 1.2207369934519347e-06,
      "loss": 1.3582,
      "num_input_tokens_seen": 70904264,
      "step": 122210
    },
    {
      "epoch": 18.203008638665477,
      "grad_norm": 10.3125,
      "learning_rate": 1.2197342087013146e-06,
      "loss": 1.1035,
      "num_input_tokens_seen": 70907848,
      "step": 122215
    },
    {
      "epoch": 18.203753351206434,
      "grad_norm": 12.5625,
      "learning_rate": 1.2187318256941166e-06,
      "loss": 1.0103,
      "num_input_tokens_seen": 70910664,
      "step": 122220
    },
    {
      "epoch": 18.204498063747394,
      "grad_norm": 12.0,
      "learning_rate": 1.2177298444472741e-06,
      "loss": 0.9636,
      "num_input_tokens_seen": 70913800,
      "step": 122225
    },
    {
      "epoch": 18.20524277628835,
      "grad_norm": 9.1875,
      "learning_rate": 1.2167282649777123e-06,
      "loss": 1.1689,
      "num_input_tokens_seen": 70916840,
      "step": 122230
    },
    {
      "epoch": 18.20598748882931,
      "grad_norm": 7.1875,
      "learning_rate": 1.2157270873023596e-06,
      "loss": 1.2037,
      "num_input_tokens_seen": 70919656,
      "step": 122235
    },
    {
      "epoch": 18.206732201370272,
      "grad_norm": 11.3125,
      "learning_rate": 1.2147263114381191e-06,
      "loss": 1.2577,
      "num_input_tokens_seen": 70922408,
      "step": 122240
    },
    {
      "epoch": 18.20747691391123,
      "grad_norm": 4.1875,
      "learning_rate": 1.213725937401905e-06,
      "loss": 0.8305,
      "num_input_tokens_seen": 70925096,
      "step": 122245
    },
    {
      "epoch": 18.20822162645219,
      "grad_norm": 5.40625,
      "learning_rate": 1.2127259652106149e-06,
      "loss": 1.134,
      "num_input_tokens_seen": 70927656,
      "step": 122250
    },
    {
      "epoch": 18.20896633899315,
      "grad_norm": 9.625,
      "learning_rate": 1.2117263948811464e-06,
      "loss": 1.0584,
      "num_input_tokens_seen": 70930792,
      "step": 122255
    },
    {
      "epoch": 18.209711051534107,
      "grad_norm": 15.75,
      "learning_rate": 1.2107272264303831e-06,
      "loss": 1.051,
      "num_input_tokens_seen": 70933736,
      "step": 122260
    },
    {
      "epoch": 18.210455764075068,
      "grad_norm": 7.53125,
      "learning_rate": 1.2097284598752034e-06,
      "loss": 1.0477,
      "num_input_tokens_seen": 70936616,
      "step": 122265
    },
    {
      "epoch": 18.211200476616025,
      "grad_norm": 5.34375,
      "learning_rate": 1.2087300952324826e-06,
      "loss": 0.9573,
      "num_input_tokens_seen": 70939432,
      "step": 122270
    },
    {
      "epoch": 18.211945189156985,
      "grad_norm": 12.75,
      "learning_rate": 1.2077321325190849e-06,
      "loss": 0.938,
      "num_input_tokens_seen": 70942568,
      "step": 122275
    },
    {
      "epoch": 18.212689901697946,
      "grad_norm": 12.5,
      "learning_rate": 1.206734571751872e-06,
      "loss": 1.1506,
      "num_input_tokens_seen": 70945352,
      "step": 122280
    },
    {
      "epoch": 18.213434614238903,
      "grad_norm": 10.0,
      "learning_rate": 1.2057374129476968e-06,
      "loss": 1.0916,
      "num_input_tokens_seen": 70947976,
      "step": 122285
    },
    {
      "epoch": 18.214179326779863,
      "grad_norm": 29.0,
      "learning_rate": 1.2047406561234042e-06,
      "loss": 1.4463,
      "num_input_tokens_seen": 70950664,
      "step": 122290
    },
    {
      "epoch": 18.214924039320824,
      "grad_norm": 6.34375,
      "learning_rate": 1.203744301295831e-06,
      "loss": 0.857,
      "num_input_tokens_seen": 70953704,
      "step": 122295
    },
    {
      "epoch": 18.21566875186178,
      "grad_norm": 9.8125,
      "learning_rate": 1.2027483484818165e-06,
      "loss": 1.0817,
      "num_input_tokens_seen": 70956584,
      "step": 122300
    },
    {
      "epoch": 18.21641346440274,
      "grad_norm": 9.6875,
      "learning_rate": 1.2017527976981828e-06,
      "loss": 1.0074,
      "num_input_tokens_seen": 70959464,
      "step": 122305
    },
    {
      "epoch": 18.2171581769437,
      "grad_norm": 9.25,
      "learning_rate": 1.200757648961745e-06,
      "loss": 1.2106,
      "num_input_tokens_seen": 70962536,
      "step": 122310
    },
    {
      "epoch": 18.21790288948466,
      "grad_norm": 5.96875,
      "learning_rate": 1.1997629022893198e-06,
      "loss": 0.9951,
      "num_input_tokens_seen": 70965128,
      "step": 122315
    },
    {
      "epoch": 18.21864760202562,
      "grad_norm": 11.5,
      "learning_rate": 1.1987685576977131e-06,
      "loss": 0.7767,
      "num_input_tokens_seen": 70967880,
      "step": 122320
    },
    {
      "epoch": 18.219392314566576,
      "grad_norm": 10.0625,
      "learning_rate": 1.19777461520372e-06,
      "loss": 0.9832,
      "num_input_tokens_seen": 70970792,
      "step": 122325
    },
    {
      "epoch": 18.220137027107537,
      "grad_norm": 17.375,
      "learning_rate": 1.19678107482413e-06,
      "loss": 1.1389,
      "num_input_tokens_seen": 70973704,
      "step": 122330
    },
    {
      "epoch": 18.220881739648497,
      "grad_norm": 10.4375,
      "learning_rate": 1.1957879365757346e-06,
      "loss": 0.6451,
      "num_input_tokens_seen": 70976584,
      "step": 122335
    },
    {
      "epoch": 18.221626452189454,
      "grad_norm": 9.8125,
      "learning_rate": 1.1947952004753044e-06,
      "loss": 1.1286,
      "num_input_tokens_seen": 70979496,
      "step": 122340
    },
    {
      "epoch": 18.222371164730415,
      "grad_norm": 10.875,
      "learning_rate": 1.1938028665396173e-06,
      "loss": 1.0521,
      "num_input_tokens_seen": 70982408,
      "step": 122345
    },
    {
      "epoch": 18.223115877271372,
      "grad_norm": 8.75,
      "learning_rate": 1.1928109347854377e-06,
      "loss": 1.1413,
      "num_input_tokens_seen": 70985192,
      "step": 122350
    },
    {
      "epoch": 18.223860589812332,
      "grad_norm": 15.5,
      "learning_rate": 1.1918194052295162e-06,
      "loss": 1.076,
      "num_input_tokens_seen": 70988168,
      "step": 122355
    },
    {
      "epoch": 18.224605302353293,
      "grad_norm": 7.40625,
      "learning_rate": 1.1908282778886115e-06,
      "loss": 1.1456,
      "num_input_tokens_seen": 70991080,
      "step": 122360
    },
    {
      "epoch": 18.22535001489425,
      "grad_norm": 27.0,
      "learning_rate": 1.1898375527794603e-06,
      "loss": 1.4379,
      "num_input_tokens_seen": 70994024,
      "step": 122365
    },
    {
      "epoch": 18.22609472743521,
      "grad_norm": 12.5625,
      "learning_rate": 1.1888472299188102e-06,
      "loss": 1.1217,
      "num_input_tokens_seen": 70997128,
      "step": 122370
    },
    {
      "epoch": 18.226839439976168,
      "grad_norm": 6.375,
      "learning_rate": 1.1878573093233814e-06,
      "loss": 0.9364,
      "num_input_tokens_seen": 71000104,
      "step": 122375
    },
    {
      "epoch": 18.227584152517128,
      "grad_norm": 11.0625,
      "learning_rate": 1.1868677910099018e-06,
      "loss": 1.3109,
      "num_input_tokens_seen": 71002856,
      "step": 122380
    },
    {
      "epoch": 18.22832886505809,
      "grad_norm": 6.34375,
      "learning_rate": 1.1858786749950919e-06,
      "loss": 1.075,
      "num_input_tokens_seen": 71005960,
      "step": 122385
    },
    {
      "epoch": 18.229073577599046,
      "grad_norm": 6.78125,
      "learning_rate": 1.184889961295657e-06,
      "loss": 1.188,
      "num_input_tokens_seen": 71008744,
      "step": 122390
    },
    {
      "epoch": 18.229818290140006,
      "grad_norm": 10.3125,
      "learning_rate": 1.1839016499283013e-06,
      "loss": 1.1243,
      "num_input_tokens_seen": 71011688,
      "step": 122395
    },
    {
      "epoch": 18.230563002680967,
      "grad_norm": 10.5625,
      "learning_rate": 1.1829137409097191e-06,
      "loss": 1.1507,
      "num_input_tokens_seen": 71014824,
      "step": 122400
    },
    {
      "epoch": 18.231307715221924,
      "grad_norm": 7.5,
      "learning_rate": 1.1819262342566056e-06,
      "loss": 1.0002,
      "num_input_tokens_seen": 71017992,
      "step": 122405
    },
    {
      "epoch": 18.232052427762884,
      "grad_norm": 7.25,
      "learning_rate": 1.1809391299856365e-06,
      "loss": 1.2741,
      "num_input_tokens_seen": 71020936,
      "step": 122410
    },
    {
      "epoch": 18.23279714030384,
      "grad_norm": 9.125,
      "learning_rate": 1.1799524281134983e-06,
      "loss": 0.8976,
      "num_input_tokens_seen": 71023656,
      "step": 122415
    },
    {
      "epoch": 18.2335418528448,
      "grad_norm": 10.1875,
      "learning_rate": 1.1789661286568472e-06,
      "loss": 1.0061,
      "num_input_tokens_seen": 71026376,
      "step": 122420
    },
    {
      "epoch": 18.234286565385762,
      "grad_norm": 5.8125,
      "learning_rate": 1.1779802316323585e-06,
      "loss": 1.1091,
      "num_input_tokens_seen": 71029032,
      "step": 122425
    },
    {
      "epoch": 18.23503127792672,
      "grad_norm": 12.25,
      "learning_rate": 1.176994737056683e-06,
      "loss": 1.2761,
      "num_input_tokens_seen": 71032072,
      "step": 122430
    },
    {
      "epoch": 18.23577599046768,
      "grad_norm": 13.0,
      "learning_rate": 1.176009644946463e-06,
      "loss": 1.0393,
      "num_input_tokens_seen": 71034728,
      "step": 122435
    },
    {
      "epoch": 18.23652070300864,
      "grad_norm": 11.125,
      "learning_rate": 1.1750249553183518e-06,
      "loss": 0.7671,
      "num_input_tokens_seen": 71037480,
      "step": 122440
    },
    {
      "epoch": 18.237265415549597,
      "grad_norm": 15.5625,
      "learning_rate": 1.1740406681889748e-06,
      "loss": 0.9791,
      "num_input_tokens_seen": 71040456,
      "step": 122445
    },
    {
      "epoch": 18.238010128090558,
      "grad_norm": 8.5,
      "learning_rate": 1.173056783574969e-06,
      "loss": 1.0723,
      "num_input_tokens_seen": 71043400,
      "step": 122450
    },
    {
      "epoch": 18.238754840631515,
      "grad_norm": 8.5,
      "learning_rate": 1.1720733014929514e-06,
      "loss": 1.0307,
      "num_input_tokens_seen": 71046376,
      "step": 122455
    },
    {
      "epoch": 18.239499553172475,
      "grad_norm": 20.75,
      "learning_rate": 1.1710902219595366e-06,
      "loss": 0.8504,
      "num_input_tokens_seen": 71049512,
      "step": 122460
    },
    {
      "epoch": 18.240244265713436,
      "grad_norm": 9.75,
      "learning_rate": 1.1701075449913363e-06,
      "loss": 0.8351,
      "num_input_tokens_seen": 71052904,
      "step": 122465
    },
    {
      "epoch": 18.240988978254393,
      "grad_norm": 10.0625,
      "learning_rate": 1.1691252706049456e-06,
      "loss": 1.3708,
      "num_input_tokens_seen": 71055880,
      "step": 122470
    },
    {
      "epoch": 18.241733690795353,
      "grad_norm": 9.5625,
      "learning_rate": 1.168143398816965e-06,
      "loss": 1.0297,
      "num_input_tokens_seen": 71058920,
      "step": 122475
    },
    {
      "epoch": 18.242478403336314,
      "grad_norm": 13.1875,
      "learning_rate": 1.1671619296439785e-06,
      "loss": 1.2891,
      "num_input_tokens_seen": 71061928,
      "step": 122480
    },
    {
      "epoch": 18.24322311587727,
      "grad_norm": 10.625,
      "learning_rate": 1.16618086310257e-06,
      "loss": 1.1456,
      "num_input_tokens_seen": 71064872,
      "step": 122485
    },
    {
      "epoch": 18.24396782841823,
      "grad_norm": 12.9375,
      "learning_rate": 1.1652001992093097e-06,
      "loss": 1.2771,
      "num_input_tokens_seen": 71067592,
      "step": 122490
    },
    {
      "epoch": 18.24471254095919,
      "grad_norm": 10.5,
      "learning_rate": 1.1642199379807706e-06,
      "loss": 1.2081,
      "num_input_tokens_seen": 71070280,
      "step": 122495
    },
    {
      "epoch": 18.24545725350015,
      "grad_norm": 15.1875,
      "learning_rate": 1.1632400794335084e-06,
      "loss": 0.9841,
      "num_input_tokens_seen": 71073320,
      "step": 122500
    },
    {
      "epoch": 18.24620196604111,
      "grad_norm": 8.75,
      "learning_rate": 1.16226062358408e-06,
      "loss": 0.748,
      "num_input_tokens_seen": 71076008,
      "step": 122505
    },
    {
      "epoch": 18.246946678582066,
      "grad_norm": 10.0625,
      "learning_rate": 1.1612815704490298e-06,
      "loss": 1.0612,
      "num_input_tokens_seen": 71078920,
      "step": 122510
    },
    {
      "epoch": 18.247691391123027,
      "grad_norm": 7.96875,
      "learning_rate": 1.1603029200448978e-06,
      "loss": 1.1372,
      "num_input_tokens_seen": 71081640,
      "step": 122515
    },
    {
      "epoch": 18.248436103663984,
      "grad_norm": 16.0,
      "learning_rate": 1.1593246723882206e-06,
      "loss": 1.0414,
      "num_input_tokens_seen": 71084520,
      "step": 122520
    },
    {
      "epoch": 18.249180816204944,
      "grad_norm": 9.375,
      "learning_rate": 1.158346827495524e-06,
      "loss": 1.0342,
      "num_input_tokens_seen": 71087304,
      "step": 122525
    },
    {
      "epoch": 18.249925528745905,
      "grad_norm": 14.1875,
      "learning_rate": 1.1573693853833224e-06,
      "loss": 0.7841,
      "num_input_tokens_seen": 71090248,
      "step": 122530
    },
    {
      "epoch": 18.250670241286862,
      "grad_norm": 8.625,
      "learning_rate": 1.156392346068133e-06,
      "loss": 1.2417,
      "num_input_tokens_seen": 71093128,
      "step": 122535
    },
    {
      "epoch": 18.251414953827823,
      "grad_norm": 11.125,
      "learning_rate": 1.1554157095664625e-06,
      "loss": 1.3507,
      "num_input_tokens_seen": 71097192,
      "step": 122540
    },
    {
      "epoch": 18.252159666368783,
      "grad_norm": 11.5,
      "learning_rate": 1.1544394758948112e-06,
      "loss": 0.8129,
      "num_input_tokens_seen": 71100072,
      "step": 122545
    },
    {
      "epoch": 18.25290437890974,
      "grad_norm": 13.0,
      "learning_rate": 1.1534636450696634e-06,
      "loss": 1.0575,
      "num_input_tokens_seen": 71102664,
      "step": 122550
    },
    {
      "epoch": 18.2536490914507,
      "grad_norm": 18.5,
      "learning_rate": 1.1524882171075168e-06,
      "loss": 0.8304,
      "num_input_tokens_seen": 71105256,
      "step": 122555
    },
    {
      "epoch": 18.254393803991658,
      "grad_norm": 9.3125,
      "learning_rate": 1.151513192024839e-06,
      "loss": 0.9391,
      "num_input_tokens_seen": 71108328,
      "step": 122560
    },
    {
      "epoch": 18.255138516532618,
      "grad_norm": 12.875,
      "learning_rate": 1.150538569838111e-06,
      "loss": 1.3172,
      "num_input_tokens_seen": 71111112,
      "step": 122565
    },
    {
      "epoch": 18.25588322907358,
      "grad_norm": 7.78125,
      "learning_rate": 1.1495643505637922e-06,
      "loss": 0.9095,
      "num_input_tokens_seen": 71114184,
      "step": 122570
    },
    {
      "epoch": 18.256627941614536,
      "grad_norm": 15.625,
      "learning_rate": 1.148590534218344e-06,
      "loss": 1.2082,
      "num_input_tokens_seen": 71117000,
      "step": 122575
    },
    {
      "epoch": 18.257372654155496,
      "grad_norm": 11.6875,
      "learning_rate": 1.1476171208182146e-06,
      "loss": 1.1801,
      "num_input_tokens_seen": 71120200,
      "step": 122580
    },
    {
      "epoch": 18.258117366696457,
      "grad_norm": 6.1875,
      "learning_rate": 1.1466441103798575e-06,
      "loss": 1.0251,
      "num_input_tokens_seen": 71123144,
      "step": 122585
    },
    {
      "epoch": 18.258862079237414,
      "grad_norm": 8.1875,
      "learning_rate": 1.1456715029197012e-06,
      "loss": 1.1279,
      "num_input_tokens_seen": 71125960,
      "step": 122590
    },
    {
      "epoch": 18.259606791778374,
      "grad_norm": 8.375,
      "learning_rate": 1.1446992984541827e-06,
      "loss": 1.0791,
      "num_input_tokens_seen": 71128840,
      "step": 122595
    },
    {
      "epoch": 18.26035150431933,
      "grad_norm": 10.625,
      "learning_rate": 1.143727496999722e-06,
      "loss": 0.8631,
      "num_input_tokens_seen": 71131624,
      "step": 122600
    },
    {
      "epoch": 18.26109621686029,
      "grad_norm": 12.3125,
      "learning_rate": 1.1427560985727392e-06,
      "loss": 0.8747,
      "num_input_tokens_seen": 71134600,
      "step": 122605
    },
    {
      "epoch": 18.261840929401252,
      "grad_norm": 7.1875,
      "learning_rate": 1.1417851031896438e-06,
      "loss": 1.0278,
      "num_input_tokens_seen": 71137352,
      "step": 122610
    },
    {
      "epoch": 18.26258564194221,
      "grad_norm": 14.25,
      "learning_rate": 1.140814510866839e-06,
      "loss": 1.0105,
      "num_input_tokens_seen": 71140200,
      "step": 122615
    },
    {
      "epoch": 18.26333035448317,
      "grad_norm": 7.5,
      "learning_rate": 1.1398443216207282e-06,
      "loss": 0.8002,
      "num_input_tokens_seen": 71143080,
      "step": 122620
    },
    {
      "epoch": 18.26407506702413,
      "grad_norm": 8.125,
      "learning_rate": 1.138874535467696e-06,
      "loss": 0.996,
      "num_input_tokens_seen": 71145928,
      "step": 122625
    },
    {
      "epoch": 18.264819779565087,
      "grad_norm": 18.625,
      "learning_rate": 1.1379051524241236e-06,
      "loss": 1.4273,
      "num_input_tokens_seen": 71148744,
      "step": 122630
    },
    {
      "epoch": 18.265564492106048,
      "grad_norm": 8.4375,
      "learning_rate": 1.1369361725063948e-06,
      "loss": 1.0602,
      "num_input_tokens_seen": 71151656,
      "step": 122635
    },
    {
      "epoch": 18.266309204647005,
      "grad_norm": 13.5625,
      "learning_rate": 1.1359675957308745e-06,
      "loss": 1.1025,
      "num_input_tokens_seen": 71154696,
      "step": 122640
    },
    {
      "epoch": 18.267053917187965,
      "grad_norm": 11.75,
      "learning_rate": 1.1349994221139276e-06,
      "loss": 1.2069,
      "num_input_tokens_seen": 71157512,
      "step": 122645
    },
    {
      "epoch": 18.267798629728926,
      "grad_norm": 17.375,
      "learning_rate": 1.134031651671913e-06,
      "loss": 1.286,
      "num_input_tokens_seen": 71160424,
      "step": 122650
    },
    {
      "epoch": 18.268543342269883,
      "grad_norm": 7.28125,
      "learning_rate": 1.133064284421176e-06,
      "loss": 1.251,
      "num_input_tokens_seen": 71163336,
      "step": 122655
    },
    {
      "epoch": 18.269288054810843,
      "grad_norm": 12.75,
      "learning_rate": 1.132097320378056e-06,
      "loss": 1.3443,
      "num_input_tokens_seen": 71166408,
      "step": 122660
    },
    {
      "epoch": 18.270032767351804,
      "grad_norm": 12.625,
      "learning_rate": 1.1311307595588987e-06,
      "loss": 1.0498,
      "num_input_tokens_seen": 71169224,
      "step": 122665
    },
    {
      "epoch": 18.27077747989276,
      "grad_norm": 13.5,
      "learning_rate": 1.130164601980027e-06,
      "loss": 1.0155,
      "num_input_tokens_seen": 71172136,
      "step": 122670
    },
    {
      "epoch": 18.27152219243372,
      "grad_norm": 13.125,
      "learning_rate": 1.1291988476577614e-06,
      "loss": 1.2658,
      "num_input_tokens_seen": 71174952,
      "step": 122675
    },
    {
      "epoch": 18.27226690497468,
      "grad_norm": 6.0,
      "learning_rate": 1.1282334966084246e-06,
      "loss": 0.9216,
      "num_input_tokens_seen": 71177992,
      "step": 122680
    },
    {
      "epoch": 18.27301161751564,
      "grad_norm": 8.4375,
      "learning_rate": 1.1272685488483148e-06,
      "loss": 1.0046,
      "num_input_tokens_seen": 71180808,
      "step": 122685
    },
    {
      "epoch": 18.2737563300566,
      "grad_norm": 12.6875,
      "learning_rate": 1.126304004393744e-06,
      "loss": 1.2669,
      "num_input_tokens_seen": 71183976,
      "step": 122690
    },
    {
      "epoch": 18.274501042597556,
      "grad_norm": 11.875,
      "learning_rate": 1.125339863261002e-06,
      "loss": 1.2016,
      "num_input_tokens_seen": 71187528,
      "step": 122695
    },
    {
      "epoch": 18.275245755138517,
      "grad_norm": 14.5625,
      "learning_rate": 1.1243761254663781e-06,
      "loss": 0.9364,
      "num_input_tokens_seen": 71190984,
      "step": 122700
    },
    {
      "epoch": 18.275990467679474,
      "grad_norm": 8.5625,
      "learning_rate": 1.1234127910261543e-06,
      "loss": 0.8883,
      "num_input_tokens_seen": 71194056,
      "step": 122705
    },
    {
      "epoch": 18.276735180220435,
      "grad_norm": 6.90625,
      "learning_rate": 1.1224498599566009e-06,
      "loss": 1.025,
      "num_input_tokens_seen": 71196872,
      "step": 122710
    },
    {
      "epoch": 18.277479892761395,
      "grad_norm": 6.84375,
      "learning_rate": 1.1214873322739933e-06,
      "loss": 1.0436,
      "num_input_tokens_seen": 71199720,
      "step": 122715
    },
    {
      "epoch": 18.278224605302352,
      "grad_norm": 10.25,
      "learning_rate": 1.1205252079945882e-06,
      "loss": 0.8376,
      "num_input_tokens_seen": 71202536,
      "step": 122720
    },
    {
      "epoch": 18.278969317843313,
      "grad_norm": 7.875,
      "learning_rate": 1.1195634871346395e-06,
      "loss": 0.9742,
      "num_input_tokens_seen": 71205512,
      "step": 122725
    },
    {
      "epoch": 18.279714030384273,
      "grad_norm": 11.5,
      "learning_rate": 1.1186021697103893e-06,
      "loss": 1.0658,
      "num_input_tokens_seen": 71208456,
      "step": 122730
    },
    {
      "epoch": 18.28045874292523,
      "grad_norm": 8.25,
      "learning_rate": 1.1176412557380888e-06,
      "loss": 0.6361,
      "num_input_tokens_seen": 71211432,
      "step": 122735
    },
    {
      "epoch": 18.28120345546619,
      "grad_norm": 9.5,
      "learning_rate": 1.116680745233961e-06,
      "loss": 0.8782,
      "num_input_tokens_seen": 71214248,
      "step": 122740
    },
    {
      "epoch": 18.281948168007148,
      "grad_norm": 11.1875,
      "learning_rate": 1.1157206382142433e-06,
      "loss": 0.7255,
      "num_input_tokens_seen": 71217352,
      "step": 122745
    },
    {
      "epoch": 18.282692880548108,
      "grad_norm": 13.5,
      "learning_rate": 1.1147609346951526e-06,
      "loss": 1.0077,
      "num_input_tokens_seen": 71220008,
      "step": 122750
    },
    {
      "epoch": 18.28343759308907,
      "grad_norm": 5.25,
      "learning_rate": 1.113801634692893e-06,
      "loss": 1.409,
      "num_input_tokens_seen": 71222792,
      "step": 122755
    },
    {
      "epoch": 18.284182305630026,
      "grad_norm": 15.125,
      "learning_rate": 1.1128427382236823e-06,
      "loss": 1.229,
      "num_input_tokens_seen": 71225960,
      "step": 122760
    },
    {
      "epoch": 18.284927018170986,
      "grad_norm": 9.375,
      "learning_rate": 1.1118842453037126e-06,
      "loss": 1.0308,
      "num_input_tokens_seen": 71229448,
      "step": 122765
    },
    {
      "epoch": 18.285671730711947,
      "grad_norm": 5.71875,
      "learning_rate": 1.1109261559491823e-06,
      "loss": 1.44,
      "num_input_tokens_seen": 71232584,
      "step": 122770
    },
    {
      "epoch": 18.286416443252904,
      "grad_norm": 10.0625,
      "learning_rate": 1.109968470176273e-06,
      "loss": 1.0215,
      "num_input_tokens_seen": 71235112,
      "step": 122775
    },
    {
      "epoch": 18.287161155793864,
      "grad_norm": 5.875,
      "learning_rate": 1.109011188001169e-06,
      "loss": 1.1217,
      "num_input_tokens_seen": 71237928,
      "step": 122780
    },
    {
      "epoch": 18.28790586833482,
      "grad_norm": 16.75,
      "learning_rate": 1.1080543094400374e-06,
      "loss": 0.8276,
      "num_input_tokens_seen": 71240968,
      "step": 122785
    },
    {
      "epoch": 18.28865058087578,
      "grad_norm": 11.5625,
      "learning_rate": 1.1070978345090494e-06,
      "loss": 1.0852,
      "num_input_tokens_seen": 71243880,
      "step": 122790
    },
    {
      "epoch": 18.289395293416742,
      "grad_norm": 11.5625,
      "learning_rate": 1.1061417632243554e-06,
      "loss": 1.4574,
      "num_input_tokens_seen": 71246792,
      "step": 122795
    },
    {
      "epoch": 18.2901400059577,
      "grad_norm": 9.0625,
      "learning_rate": 1.105186095602112e-06,
      "loss": 1.089,
      "num_input_tokens_seen": 71249480,
      "step": 122800
    },
    {
      "epoch": 18.29088471849866,
      "grad_norm": 11.125,
      "learning_rate": 1.1042308316584649e-06,
      "loss": 0.8433,
      "num_input_tokens_seen": 71252072,
      "step": 122805
    },
    {
      "epoch": 18.29162943103962,
      "grad_norm": 8.75,
      "learning_rate": 1.1032759714095481e-06,
      "loss": 1.1064,
      "num_input_tokens_seen": 71255048,
      "step": 122810
    },
    {
      "epoch": 18.292374143580577,
      "grad_norm": 17.0,
      "learning_rate": 1.1023215148714988e-06,
      "loss": 1.0798,
      "num_input_tokens_seen": 71257928,
      "step": 122815
    },
    {
      "epoch": 18.293118856121538,
      "grad_norm": 11.4375,
      "learning_rate": 1.1013674620604376e-06,
      "loss": 1.3862,
      "num_input_tokens_seen": 71260872,
      "step": 122820
    },
    {
      "epoch": 18.293863568662495,
      "grad_norm": 6.625,
      "learning_rate": 1.1004138129924874e-06,
      "loss": 0.9854,
      "num_input_tokens_seen": 71263464,
      "step": 122825
    },
    {
      "epoch": 18.294608281203455,
      "grad_norm": 11.625,
      "learning_rate": 1.0994605676837521e-06,
      "loss": 1.2571,
      "num_input_tokens_seen": 71266152,
      "step": 122830
    },
    {
      "epoch": 18.295352993744416,
      "grad_norm": 9.5,
      "learning_rate": 1.0985077261503384e-06,
      "loss": 0.9265,
      "num_input_tokens_seen": 71269064,
      "step": 122835
    },
    {
      "epoch": 18.296097706285373,
      "grad_norm": 14.0,
      "learning_rate": 1.0975552884083473e-06,
      "loss": 0.9226,
      "num_input_tokens_seen": 71271656,
      "step": 122840
    },
    {
      "epoch": 18.296842418826333,
      "grad_norm": 10.75,
      "learning_rate": 1.096603254473863e-06,
      "loss": 1.1574,
      "num_input_tokens_seen": 71274536,
      "step": 122845
    },
    {
      "epoch": 18.297587131367294,
      "grad_norm": 7.9375,
      "learning_rate": 1.0956516243629754e-06,
      "loss": 0.9653,
      "num_input_tokens_seen": 71277128,
      "step": 122850
    },
    {
      "epoch": 18.29833184390825,
      "grad_norm": 12.4375,
      "learning_rate": 1.094700398091758e-06,
      "loss": 1.3292,
      "num_input_tokens_seen": 71280168,
      "step": 122855
    },
    {
      "epoch": 18.29907655644921,
      "grad_norm": 8.3125,
      "learning_rate": 1.093749575676281e-06,
      "loss": 1.015,
      "num_input_tokens_seen": 71283112,
      "step": 122860
    },
    {
      "epoch": 18.29982126899017,
      "grad_norm": 4.1875,
      "learning_rate": 1.092799157132604e-06,
      "loss": 0.899,
      "num_input_tokens_seen": 71285992,
      "step": 122865
    },
    {
      "epoch": 18.30056598153113,
      "grad_norm": 21.125,
      "learning_rate": 1.091849142476792e-06,
      "loss": 1.4824,
      "num_input_tokens_seen": 71288808,
      "step": 122870
    },
    {
      "epoch": 18.30131069407209,
      "grad_norm": 14.5,
      "learning_rate": 1.0908995317248898e-06,
      "loss": 1.1352,
      "num_input_tokens_seen": 71291656,
      "step": 122875
    },
    {
      "epoch": 18.302055406613047,
      "grad_norm": 15.875,
      "learning_rate": 1.0899503248929355e-06,
      "loss": 1.2926,
      "num_input_tokens_seen": 71294600,
      "step": 122880
    },
    {
      "epoch": 18.302800119154007,
      "grad_norm": 6.96875,
      "learning_rate": 1.0890015219969713e-06,
      "loss": 0.7626,
      "num_input_tokens_seen": 71297544,
      "step": 122885
    },
    {
      "epoch": 18.303544831694964,
      "grad_norm": 9.3125,
      "learning_rate": 1.0880531230530233e-06,
      "loss": 0.9875,
      "num_input_tokens_seen": 71300520,
      "step": 122890
    },
    {
      "epoch": 18.304289544235925,
      "grad_norm": 6.40625,
      "learning_rate": 1.0871051280771178e-06,
      "loss": 0.901,
      "num_input_tokens_seen": 71303496,
      "step": 122895
    },
    {
      "epoch": 18.305034256776885,
      "grad_norm": 13.3125,
      "learning_rate": 1.0861575370852612e-06,
      "loss": 1.467,
      "num_input_tokens_seen": 71306504,
      "step": 122900
    },
    {
      "epoch": 18.305778969317842,
      "grad_norm": 12.375,
      "learning_rate": 1.085210350093474e-06,
      "loss": 1.2888,
      "num_input_tokens_seen": 71309384,
      "step": 122905
    },
    {
      "epoch": 18.306523681858803,
      "grad_norm": 5.59375,
      "learning_rate": 1.084263567117752e-06,
      "loss": 0.9383,
      "num_input_tokens_seen": 71312392,
      "step": 122910
    },
    {
      "epoch": 18.307268394399763,
      "grad_norm": 9.3125,
      "learning_rate": 1.083317188174085e-06,
      "loss": 1.2668,
      "num_input_tokens_seen": 71315304,
      "step": 122915
    },
    {
      "epoch": 18.30801310694072,
      "grad_norm": 16.875,
      "learning_rate": 1.0823712132784713e-06,
      "loss": 0.9521,
      "num_input_tokens_seen": 71318184,
      "step": 122920
    },
    {
      "epoch": 18.30875781948168,
      "grad_norm": 5.21875,
      "learning_rate": 1.0814256424468872e-06,
      "loss": 1.0713,
      "num_input_tokens_seen": 71320904,
      "step": 122925
    },
    {
      "epoch": 18.309502532022638,
      "grad_norm": 11.3125,
      "learning_rate": 1.0804804756953057e-06,
      "loss": 1.0129,
      "num_input_tokens_seen": 71323784,
      "step": 122930
    },
    {
      "epoch": 18.310247244563598,
      "grad_norm": 17.5,
      "learning_rate": 1.079535713039695e-06,
      "loss": 1.5193,
      "num_input_tokens_seen": 71326728,
      "step": 122935
    },
    {
      "epoch": 18.31099195710456,
      "grad_norm": 10.625,
      "learning_rate": 1.078591354496017e-06,
      "loss": 1.1195,
      "num_input_tokens_seen": 71329800,
      "step": 122940
    },
    {
      "epoch": 18.311736669645516,
      "grad_norm": 14.8125,
      "learning_rate": 1.0776474000802255e-06,
      "loss": 1.0311,
      "num_input_tokens_seen": 71332904,
      "step": 122945
    },
    {
      "epoch": 18.312481382186476,
      "grad_norm": 4.96875,
      "learning_rate": 1.0767038498082694e-06,
      "loss": 0.8671,
      "num_input_tokens_seen": 71335592,
      "step": 122950
    },
    {
      "epoch": 18.313226094727437,
      "grad_norm": 10.625,
      "learning_rate": 1.0757607036960853e-06,
      "loss": 1.0825,
      "num_input_tokens_seen": 71338856,
      "step": 122955
    },
    {
      "epoch": 18.313970807268394,
      "grad_norm": 9.3125,
      "learning_rate": 1.0748179617596082e-06,
      "loss": 1.1582,
      "num_input_tokens_seen": 71342120,
      "step": 122960
    },
    {
      "epoch": 18.314715519809354,
      "grad_norm": 7.28125,
      "learning_rate": 1.0738756240147668e-06,
      "loss": 0.9428,
      "num_input_tokens_seen": 71344744,
      "step": 122965
    },
    {
      "epoch": 18.31546023235031,
      "grad_norm": 17.5,
      "learning_rate": 1.0729336904774762e-06,
      "loss": 1.0645,
      "num_input_tokens_seen": 71347720,
      "step": 122970
    },
    {
      "epoch": 18.316204944891272,
      "grad_norm": 7.09375,
      "learning_rate": 1.071992161163654e-06,
      "loss": 1.2728,
      "num_input_tokens_seen": 71350664,
      "step": 122975
    },
    {
      "epoch": 18.316949657432232,
      "grad_norm": 13.75,
      "learning_rate": 1.0710510360892072e-06,
      "loss": 1.0949,
      "num_input_tokens_seen": 71353320,
      "step": 122980
    },
    {
      "epoch": 18.31769436997319,
      "grad_norm": 7.25,
      "learning_rate": 1.0701103152700343e-06,
      "loss": 0.9598,
      "num_input_tokens_seen": 71356296,
      "step": 122985
    },
    {
      "epoch": 18.31843908251415,
      "grad_norm": 7.625,
      "learning_rate": 1.0691699987220194e-06,
      "loss": 1.2036,
      "num_input_tokens_seen": 71358728,
      "step": 122990
    },
    {
      "epoch": 18.31918379505511,
      "grad_norm": 11.1875,
      "learning_rate": 1.068230086461061e-06,
      "loss": 1.1384,
      "num_input_tokens_seen": 71361736,
      "step": 122995
    },
    {
      "epoch": 18.319928507596067,
      "grad_norm": 4.875,
      "learning_rate": 1.067290578503033e-06,
      "loss": 0.8945,
      "num_input_tokens_seen": 71364584,
      "step": 123000
    },
    {
      "epoch": 18.320673220137028,
      "grad_norm": 5.96875,
      "learning_rate": 1.0663514748637998e-06,
      "loss": 1.3686,
      "num_input_tokens_seen": 71367560,
      "step": 123005
    },
    {
      "epoch": 18.321417932677985,
      "grad_norm": 15.3125,
      "learning_rate": 1.0654127755592381e-06,
      "loss": 1.0446,
      "num_input_tokens_seen": 71370472,
      "step": 123010
    },
    {
      "epoch": 18.322162645218945,
      "grad_norm": 10.375,
      "learning_rate": 1.0644744806051988e-06,
      "loss": 0.991,
      "num_input_tokens_seen": 71373576,
      "step": 123015
    },
    {
      "epoch": 18.322907357759906,
      "grad_norm": 12.6875,
      "learning_rate": 1.0635365900175414e-06,
      "loss": 1.2678,
      "num_input_tokens_seen": 71376264,
      "step": 123020
    },
    {
      "epoch": 18.323652070300863,
      "grad_norm": 28.375,
      "learning_rate": 1.062599103812098e-06,
      "loss": 1.228,
      "num_input_tokens_seen": 71379592,
      "step": 123025
    },
    {
      "epoch": 18.324396782841823,
      "grad_norm": 9.625,
      "learning_rate": 1.0616620220047197e-06,
      "loss": 1.3072,
      "num_input_tokens_seen": 71382280,
      "step": 123030
    },
    {
      "epoch": 18.32514149538278,
      "grad_norm": 11.625,
      "learning_rate": 1.0607253446112324e-06,
      "loss": 1.1023,
      "num_input_tokens_seen": 71385160,
      "step": 123035
    },
    {
      "epoch": 18.32588620792374,
      "grad_norm": 12.3125,
      "learning_rate": 1.0597890716474545e-06,
      "loss": 0.9657,
      "num_input_tokens_seen": 71387880,
      "step": 123040
    },
    {
      "epoch": 18.3266309204647,
      "grad_norm": 11.0625,
      "learning_rate": 1.058853203129212e-06,
      "loss": 1.241,
      "num_input_tokens_seen": 71390952,
      "step": 123045
    },
    {
      "epoch": 18.32737563300566,
      "grad_norm": 12.625,
      "learning_rate": 1.0579177390723116e-06,
      "loss": 0.9085,
      "num_input_tokens_seen": 71393960,
      "step": 123050
    },
    {
      "epoch": 18.32812034554662,
      "grad_norm": 8.4375,
      "learning_rate": 1.0569826794925602e-06,
      "loss": 1.0519,
      "num_input_tokens_seen": 71397000,
      "step": 123055
    },
    {
      "epoch": 18.32886505808758,
      "grad_norm": 14.9375,
      "learning_rate": 1.0560480244057452e-06,
      "loss": 1.1712,
      "num_input_tokens_seen": 71399944,
      "step": 123060
    },
    {
      "epoch": 18.329609770628537,
      "grad_norm": 11.375,
      "learning_rate": 1.0551137738276678e-06,
      "loss": 1.0579,
      "num_input_tokens_seen": 71402856,
      "step": 123065
    },
    {
      "epoch": 18.330354483169497,
      "grad_norm": 12.625,
      "learning_rate": 1.0541799277741071e-06,
      "loss": 1.1795,
      "num_input_tokens_seen": 71405640,
      "step": 123070
    },
    {
      "epoch": 18.331099195710454,
      "grad_norm": 6.78125,
      "learning_rate": 1.0532464862608366e-06,
      "loss": 1.2022,
      "num_input_tokens_seen": 71408424,
      "step": 123075
    },
    {
      "epoch": 18.331843908251415,
      "grad_norm": 8.4375,
      "learning_rate": 1.0523134493036296e-06,
      "loss": 0.8142,
      "num_input_tokens_seen": 71411048,
      "step": 123080
    },
    {
      "epoch": 18.332588620792375,
      "grad_norm": 13.0,
      "learning_rate": 1.051380816918243e-06,
      "loss": 1.2625,
      "num_input_tokens_seen": 71414280,
      "step": 123085
    },
    {
      "epoch": 18.333333333333332,
      "grad_norm": 13.625,
      "learning_rate": 1.0504485891204452e-06,
      "loss": 1.018,
      "num_input_tokens_seen": 71417448,
      "step": 123090
    },
    {
      "epoch": 18.334078045874293,
      "grad_norm": 13.0625,
      "learning_rate": 1.0495167659259703e-06,
      "loss": 1.0752,
      "num_input_tokens_seen": 71420488,
      "step": 123095
    },
    {
      "epoch": 18.334822758415253,
      "grad_norm": 6.03125,
      "learning_rate": 1.0485853473505724e-06,
      "loss": 1.0529,
      "num_input_tokens_seen": 71423144,
      "step": 123100
    },
    {
      "epoch": 18.33556747095621,
      "grad_norm": 14.75,
      "learning_rate": 1.0476543334099781e-06,
      "loss": 1.012,
      "num_input_tokens_seen": 71425768,
      "step": 123105
    },
    {
      "epoch": 18.33631218349717,
      "grad_norm": 15.25,
      "learning_rate": 1.0467237241199218e-06,
      "loss": 1.2866,
      "num_input_tokens_seen": 71428552,
      "step": 123110
    },
    {
      "epoch": 18.337056896038128,
      "grad_norm": 9.125,
      "learning_rate": 1.0457935194961245e-06,
      "loss": 0.8662,
      "num_input_tokens_seen": 71431208,
      "step": 123115
    },
    {
      "epoch": 18.33780160857909,
      "grad_norm": 14.125,
      "learning_rate": 1.044863719554298e-06,
      "loss": 1.0829,
      "num_input_tokens_seen": 71434184,
      "step": 123120
    },
    {
      "epoch": 18.33854632112005,
      "grad_norm": 18.875,
      "learning_rate": 1.0439343243101558e-06,
      "loss": 1.0122,
      "num_input_tokens_seen": 71437160,
      "step": 123125
    },
    {
      "epoch": 18.339291033661006,
      "grad_norm": 9.375,
      "learning_rate": 1.04300533377939e-06,
      "loss": 0.9948,
      "num_input_tokens_seen": 71439976,
      "step": 123130
    },
    {
      "epoch": 18.340035746201966,
      "grad_norm": 5.59375,
      "learning_rate": 1.0420767479777022e-06,
      "loss": 1.097,
      "num_input_tokens_seen": 71443336,
      "step": 123135
    },
    {
      "epoch": 18.340780458742927,
      "grad_norm": 6.21875,
      "learning_rate": 1.0411485669207772e-06,
      "loss": 1.2963,
      "num_input_tokens_seen": 71446280,
      "step": 123140
    },
    {
      "epoch": 18.341525171283884,
      "grad_norm": 8.125,
      "learning_rate": 1.0402207906242966e-06,
      "loss": 1.0729,
      "num_input_tokens_seen": 71449096,
      "step": 123145
    },
    {
      "epoch": 18.342269883824844,
      "grad_norm": 8.625,
      "learning_rate": 1.0392934191039372e-06,
      "loss": 0.8343,
      "num_input_tokens_seen": 71451944,
      "step": 123150
    },
    {
      "epoch": 18.3430145963658,
      "grad_norm": 14.0,
      "learning_rate": 1.0383664523753584e-06,
      "loss": 1.1311,
      "num_input_tokens_seen": 71454888,
      "step": 123155
    },
    {
      "epoch": 18.343759308906762,
      "grad_norm": 8.375,
      "learning_rate": 1.0374398904542283e-06,
      "loss": 0.8936,
      "num_input_tokens_seen": 71457608,
      "step": 123160
    },
    {
      "epoch": 18.344504021447722,
      "grad_norm": 6.0,
      "learning_rate": 1.0365137333561925e-06,
      "loss": 0.947,
      "num_input_tokens_seen": 71460392,
      "step": 123165
    },
    {
      "epoch": 18.34524873398868,
      "grad_norm": 13.0625,
      "learning_rate": 1.0355879810969054e-06,
      "loss": 0.8329,
      "num_input_tokens_seen": 71463528,
      "step": 123170
    },
    {
      "epoch": 18.34599344652964,
      "grad_norm": 22.25,
      "learning_rate": 1.0346626336920019e-06,
      "loss": 1.3389,
      "num_input_tokens_seen": 71466664,
      "step": 123175
    },
    {
      "epoch": 18.3467381590706,
      "grad_norm": 14.375,
      "learning_rate": 1.0337376911571161e-06,
      "loss": 1.1655,
      "num_input_tokens_seen": 71469704,
      "step": 123180
    },
    {
      "epoch": 18.347482871611557,
      "grad_norm": 8.8125,
      "learning_rate": 1.032813153507875e-06,
      "loss": 1.0873,
      "num_input_tokens_seen": 71472456,
      "step": 123185
    },
    {
      "epoch": 18.348227584152518,
      "grad_norm": 7.8125,
      "learning_rate": 1.0318890207598963e-06,
      "loss": 1.2093,
      "num_input_tokens_seen": 71475368,
      "step": 123190
    },
    {
      "epoch": 18.348972296693475,
      "grad_norm": 6.53125,
      "learning_rate": 1.0309652929287926e-06,
      "loss": 1.2873,
      "num_input_tokens_seen": 71478184,
      "step": 123195
    },
    {
      "epoch": 18.349717009234435,
      "grad_norm": 14.0625,
      "learning_rate": 1.0300419700301684e-06,
      "loss": 1.1425,
      "num_input_tokens_seen": 71481128,
      "step": 123200
    },
    {
      "epoch": 18.350461721775396,
      "grad_norm": 6.28125,
      "learning_rate": 1.0291190520796246e-06,
      "loss": 1.1883,
      "num_input_tokens_seen": 71484360,
      "step": 123205
    },
    {
      "epoch": 18.351206434316353,
      "grad_norm": 8.25,
      "learning_rate": 1.028196539092746e-06,
      "loss": 1.3745,
      "num_input_tokens_seen": 71487112,
      "step": 123210
    },
    {
      "epoch": 18.351951146857314,
      "grad_norm": 9.1875,
      "learning_rate": 1.027274431085129e-06,
      "loss": 0.8669,
      "num_input_tokens_seen": 71490184,
      "step": 123215
    },
    {
      "epoch": 18.35269585939827,
      "grad_norm": 12.1875,
      "learning_rate": 1.0263527280723411e-06,
      "loss": 0.7486,
      "num_input_tokens_seen": 71493384,
      "step": 123220
    },
    {
      "epoch": 18.35344057193923,
      "grad_norm": 15.6875,
      "learning_rate": 1.025431430069962e-06,
      "loss": 1.1232,
      "num_input_tokens_seen": 71496200,
      "step": 123225
    },
    {
      "epoch": 18.35418528448019,
      "grad_norm": 10.0625,
      "learning_rate": 1.0245105370935536e-06,
      "loss": 1.186,
      "num_input_tokens_seen": 71499144,
      "step": 123230
    },
    {
      "epoch": 18.35492999702115,
      "grad_norm": 9.8125,
      "learning_rate": 1.0235900491586652e-06,
      "loss": 1.062,
      "num_input_tokens_seen": 71501960,
      "step": 123235
    },
    {
      "epoch": 18.35567470956211,
      "grad_norm": 13.5,
      "learning_rate": 1.022669966280862e-06,
      "loss": 0.921,
      "num_input_tokens_seen": 71504680,
      "step": 123240
    },
    {
      "epoch": 18.35641942210307,
      "grad_norm": 15.4375,
      "learning_rate": 1.021750288475673e-06,
      "loss": 0.9781,
      "num_input_tokens_seen": 71507752,
      "step": 123245
    },
    {
      "epoch": 18.357164134644027,
      "grad_norm": 11.125,
      "learning_rate": 1.0208310157586497e-06,
      "loss": 1.1776,
      "num_input_tokens_seen": 71510568,
      "step": 123250
    },
    {
      "epoch": 18.357908847184987,
      "grad_norm": 8.375,
      "learning_rate": 1.0199121481453106e-06,
      "loss": 1.5537,
      "num_input_tokens_seen": 71513736,
      "step": 123255
    },
    {
      "epoch": 18.358653559725944,
      "grad_norm": 12.125,
      "learning_rate": 1.0189936856511873e-06,
      "loss": 1.075,
      "num_input_tokens_seen": 71516744,
      "step": 123260
    },
    {
      "epoch": 18.359398272266905,
      "grad_norm": 10.5625,
      "learning_rate": 1.018075628291787e-06,
      "loss": 1.2717,
      "num_input_tokens_seen": 71519464,
      "step": 123265
    },
    {
      "epoch": 18.360142984807865,
      "grad_norm": 7.09375,
      "learning_rate": 1.0171579760826279e-06,
      "loss": 0.9966,
      "num_input_tokens_seen": 71522504,
      "step": 123270
    },
    {
      "epoch": 18.360887697348822,
      "grad_norm": 12.9375,
      "learning_rate": 1.0162407290392112e-06,
      "loss": 1.1241,
      "num_input_tokens_seen": 71525064,
      "step": 123275
    },
    {
      "epoch": 18.361632409889783,
      "grad_norm": 7.75,
      "learning_rate": 1.0153238871770277e-06,
      "loss": 1.1197,
      "num_input_tokens_seen": 71527656,
      "step": 123280
    },
    {
      "epoch": 18.362377122430743,
      "grad_norm": 13.0625,
      "learning_rate": 1.014407450511573e-06,
      "loss": 1.0157,
      "num_input_tokens_seen": 71530664,
      "step": 123285
    },
    {
      "epoch": 18.3631218349717,
      "grad_norm": 11.0625,
      "learning_rate": 1.013491419058324e-06,
      "loss": 0.8281,
      "num_input_tokens_seen": 71533800,
      "step": 123290
    },
    {
      "epoch": 18.36386654751266,
      "grad_norm": 7.15625,
      "learning_rate": 1.0125757928327623e-06,
      "loss": 1.2398,
      "num_input_tokens_seen": 71537160,
      "step": 123295
    },
    {
      "epoch": 18.364611260053618,
      "grad_norm": 14.0625,
      "learning_rate": 1.011660571850348e-06,
      "loss": 1.265,
      "num_input_tokens_seen": 71540072,
      "step": 123300
    },
    {
      "epoch": 18.36535597259458,
      "grad_norm": 13.25,
      "learning_rate": 1.010745756126552e-06,
      "loss": 1.4112,
      "num_input_tokens_seen": 71542728,
      "step": 123305
    },
    {
      "epoch": 18.36610068513554,
      "grad_norm": 10.25,
      "learning_rate": 1.0098313456768233e-06,
      "loss": 1.0794,
      "num_input_tokens_seen": 71545576,
      "step": 123310
    },
    {
      "epoch": 18.366845397676496,
      "grad_norm": 9.4375,
      "learning_rate": 1.008917340516613e-06,
      "loss": 1.4244,
      "num_input_tokens_seen": 71548872,
      "step": 123315
    },
    {
      "epoch": 18.367590110217456,
      "grad_norm": 18.375,
      "learning_rate": 1.008003740661359e-06,
      "loss": 1.2909,
      "num_input_tokens_seen": 71551848,
      "step": 123320
    },
    {
      "epoch": 18.368334822758417,
      "grad_norm": 13.1875,
      "learning_rate": 1.007090546126499e-06,
      "loss": 1.4026,
      "num_input_tokens_seen": 71554824,
      "step": 123325
    },
    {
      "epoch": 18.369079535299374,
      "grad_norm": 13.1875,
      "learning_rate": 1.0061777569274593e-06,
      "loss": 1.1234,
      "num_input_tokens_seen": 71557544,
      "step": 123330
    },
    {
      "epoch": 18.369824247840334,
      "grad_norm": 8.5625,
      "learning_rate": 1.0052653730796558e-06,
      "loss": 0.934,
      "num_input_tokens_seen": 71560296,
      "step": 123335
    },
    {
      "epoch": 18.37056896038129,
      "grad_norm": 6.625,
      "learning_rate": 1.004353394598509e-06,
      "loss": 0.9777,
      "num_input_tokens_seen": 71563176,
      "step": 123340
    },
    {
      "epoch": 18.371313672922252,
      "grad_norm": 9.75,
      "learning_rate": 1.0034418214994235e-06,
      "loss": 0.859,
      "num_input_tokens_seen": 71566088,
      "step": 123345
    },
    {
      "epoch": 18.372058385463212,
      "grad_norm": 4.625,
      "learning_rate": 1.0025306537978007e-06,
      "loss": 0.8046,
      "num_input_tokens_seen": 71569128,
      "step": 123350
    },
    {
      "epoch": 18.37280309800417,
      "grad_norm": 11.5625,
      "learning_rate": 1.001619891509034e-06,
      "loss": 1.4026,
      "num_input_tokens_seen": 71571976,
      "step": 123355
    },
    {
      "epoch": 18.37354781054513,
      "grad_norm": 15.0,
      "learning_rate": 1.000709534648503e-06,
      "loss": 0.9172,
      "num_input_tokens_seen": 71575176,
      "step": 123360
    },
    {
      "epoch": 18.37429252308609,
      "grad_norm": 11.6875,
      "learning_rate": 9.997995832315977e-07,
      "loss": 1.3317,
      "num_input_tokens_seen": 71577832,
      "step": 123365
    },
    {
      "epoch": 18.375037235627047,
      "grad_norm": 11.1875,
      "learning_rate": 9.988900372736808e-07,
      "loss": 0.9369,
      "num_input_tokens_seen": 71580488,
      "step": 123370
    },
    {
      "epoch": 18.375781948168008,
      "grad_norm": 12.0,
      "learning_rate": 9.979808967901267e-07,
      "loss": 1.1582,
      "num_input_tokens_seen": 71583144,
      "step": 123375
    },
    {
      "epoch": 18.376526660708965,
      "grad_norm": 8.4375,
      "learning_rate": 9.97072161796292e-07,
      "loss": 1.5535,
      "num_input_tokens_seen": 71586376,
      "step": 123380
    },
    {
      "epoch": 18.377271373249926,
      "grad_norm": 11.375,
      "learning_rate": 9.961638323075284e-07,
      "loss": 0.7736,
      "num_input_tokens_seen": 71589224,
      "step": 123385
    },
    {
      "epoch": 18.378016085790886,
      "grad_norm": 16.375,
      "learning_rate": 9.952559083391765e-07,
      "loss": 1.2755,
      "num_input_tokens_seen": 71592168,
      "step": 123390
    },
    {
      "epoch": 18.378760798331843,
      "grad_norm": 7.875,
      "learning_rate": 9.943483899065798e-07,
      "loss": 1.2742,
      "num_input_tokens_seen": 71595240,
      "step": 123395
    },
    {
      "epoch": 18.379505510872804,
      "grad_norm": 19.625,
      "learning_rate": 9.9344127702507e-07,
      "loss": 1.0999,
      "num_input_tokens_seen": 71598088,
      "step": 123400
    },
    {
      "epoch": 18.38025022341376,
      "grad_norm": 7.34375,
      "learning_rate": 9.925345697099686e-07,
      "loss": 0.9101,
      "num_input_tokens_seen": 71600808,
      "step": 123405
    },
    {
      "epoch": 18.38099493595472,
      "grad_norm": 9.1875,
      "learning_rate": 9.916282679765965e-07,
      "loss": 0.8919,
      "num_input_tokens_seen": 71603816,
      "step": 123410
    },
    {
      "epoch": 18.38173964849568,
      "grad_norm": 19.125,
      "learning_rate": 9.907223718402608e-07,
      "loss": 1.4159,
      "num_input_tokens_seen": 71606696,
      "step": 123415
    },
    {
      "epoch": 18.38248436103664,
      "grad_norm": 6.6875,
      "learning_rate": 9.898168813162744e-07,
      "loss": 1.2456,
      "num_input_tokens_seen": 71609416,
      "step": 123420
    },
    {
      "epoch": 18.3832290735776,
      "grad_norm": 16.25,
      "learning_rate": 9.889117964199252e-07,
      "loss": 1.4785,
      "num_input_tokens_seen": 71612136,
      "step": 123425
    },
    {
      "epoch": 18.38397378611856,
      "grad_norm": 15.4375,
      "learning_rate": 9.880071171665089e-07,
      "loss": 0.9932,
      "num_input_tokens_seen": 71614888,
      "step": 123430
    },
    {
      "epoch": 18.384718498659517,
      "grad_norm": 19.125,
      "learning_rate": 9.871028435713081e-07,
      "loss": 1.0261,
      "num_input_tokens_seen": 71617736,
      "step": 123435
    },
    {
      "epoch": 18.385463211200477,
      "grad_norm": 6.09375,
      "learning_rate": 9.861989756495965e-07,
      "loss": 1.057,
      "num_input_tokens_seen": 71620776,
      "step": 123440
    },
    {
      "epoch": 18.386207923741434,
      "grad_norm": 17.125,
      "learning_rate": 9.852955134166481e-07,
      "loss": 1.2339,
      "num_input_tokens_seen": 71623560,
      "step": 123445
    },
    {
      "epoch": 18.386952636282395,
      "grad_norm": 7.125,
      "learning_rate": 9.843924568877282e-07,
      "loss": 1.1449,
      "num_input_tokens_seen": 71626376,
      "step": 123450
    },
    {
      "epoch": 18.387697348823355,
      "grad_norm": 8.3125,
      "learning_rate": 9.834898060780861e-07,
      "loss": 0.8263,
      "num_input_tokens_seen": 71629128,
      "step": 123455
    },
    {
      "epoch": 18.388442061364312,
      "grad_norm": 12.0,
      "learning_rate": 9.825875610029733e-07,
      "loss": 1.4324,
      "num_input_tokens_seen": 71631816,
      "step": 123460
    },
    {
      "epoch": 18.389186773905273,
      "grad_norm": 9.5625,
      "learning_rate": 9.81685721677636e-07,
      "loss": 1.2656,
      "num_input_tokens_seen": 71635240,
      "step": 123465
    },
    {
      "epoch": 18.389931486446233,
      "grad_norm": 14.5,
      "learning_rate": 9.807842881173034e-07,
      "loss": 1.1069,
      "num_input_tokens_seen": 71637992,
      "step": 123470
    },
    {
      "epoch": 18.39067619898719,
      "grad_norm": 9.0,
      "learning_rate": 9.79883260337211e-07,
      "loss": 0.7391,
      "num_input_tokens_seen": 71640808,
      "step": 123475
    },
    {
      "epoch": 18.39142091152815,
      "grad_norm": 8.8125,
      "learning_rate": 9.789826383525796e-07,
      "loss": 1.3154,
      "num_input_tokens_seen": 71644104,
      "step": 123480
    },
    {
      "epoch": 18.392165624069108,
      "grad_norm": 7.03125,
      "learning_rate": 9.780824221786195e-07,
      "loss": 0.947,
      "num_input_tokens_seen": 71647336,
      "step": 123485
    },
    {
      "epoch": 18.39291033661007,
      "grad_norm": 17.25,
      "learning_rate": 9.771826118305432e-07,
      "loss": 1.0919,
      "num_input_tokens_seen": 71650120,
      "step": 123490
    },
    {
      "epoch": 18.39365504915103,
      "grad_norm": 8.4375,
      "learning_rate": 9.762832073235501e-07,
      "loss": 1.2115,
      "num_input_tokens_seen": 71653000,
      "step": 123495
    },
    {
      "epoch": 18.394399761691986,
      "grad_norm": 23.625,
      "learning_rate": 9.75384208672836e-07,
      "loss": 1.6033,
      "num_input_tokens_seen": 71656360,
      "step": 123500
    },
    {
      "epoch": 18.395144474232946,
      "grad_norm": 10.125,
      "learning_rate": 9.744856158935888e-07,
      "loss": 1.0213,
      "num_input_tokens_seen": 71659208,
      "step": 123505
    },
    {
      "epoch": 18.395889186773907,
      "grad_norm": 6.28125,
      "learning_rate": 9.735874290009884e-07,
      "loss": 0.8908,
      "num_input_tokens_seen": 71661960,
      "step": 123510
    },
    {
      "epoch": 18.396633899314864,
      "grad_norm": 13.375,
      "learning_rate": 9.72689648010211e-07,
      "loss": 1.189,
      "num_input_tokens_seen": 71664840,
      "step": 123515
    },
    {
      "epoch": 18.397378611855824,
      "grad_norm": 6.25,
      "learning_rate": 9.717922729364198e-07,
      "loss": 1.0232,
      "num_input_tokens_seen": 71667848,
      "step": 123520
    },
    {
      "epoch": 18.39812332439678,
      "grad_norm": 6.25,
      "learning_rate": 9.708953037947804e-07,
      "loss": 1.1768,
      "num_input_tokens_seen": 71670568,
      "step": 123525
    },
    {
      "epoch": 18.398868036937742,
      "grad_norm": 17.75,
      "learning_rate": 9.699987406004364e-07,
      "loss": 1.3527,
      "num_input_tokens_seen": 71673384,
      "step": 123530
    },
    {
      "epoch": 18.399612749478703,
      "grad_norm": 11.1875,
      "learning_rate": 9.691025833685446e-07,
      "loss": 1.2668,
      "num_input_tokens_seen": 71676328,
      "step": 123535
    },
    {
      "epoch": 18.40035746201966,
      "grad_norm": 17.875,
      "learning_rate": 9.68206832114238e-07,
      "loss": 1.4243,
      "num_input_tokens_seen": 71679144,
      "step": 123540
    },
    {
      "epoch": 18.40110217456062,
      "grad_norm": 5.8125,
      "learning_rate": 9.673114868526568e-07,
      "loss": 1.1101,
      "num_input_tokens_seen": 71681960,
      "step": 123545
    },
    {
      "epoch": 18.401846887101577,
      "grad_norm": 10.25,
      "learning_rate": 9.664165475989168e-07,
      "loss": 1.2507,
      "num_input_tokens_seen": 71684552,
      "step": 123550
    },
    {
      "epoch": 18.402591599642538,
      "grad_norm": 21.625,
      "learning_rate": 9.655220143681476e-07,
      "loss": 0.8964,
      "num_input_tokens_seen": 71687464,
      "step": 123555
    },
    {
      "epoch": 18.403336312183498,
      "grad_norm": 8.25,
      "learning_rate": 9.646278871754539e-07,
      "loss": 1.2938,
      "num_input_tokens_seen": 71690248,
      "step": 123560
    },
    {
      "epoch": 18.404081024724455,
      "grad_norm": 6.8125,
      "learning_rate": 9.637341660359428e-07,
      "loss": 1.144,
      "num_input_tokens_seen": 71693288,
      "step": 123565
    },
    {
      "epoch": 18.404825737265416,
      "grad_norm": 9.25,
      "learning_rate": 9.628408509647164e-07,
      "loss": 0.9225,
      "num_input_tokens_seen": 71696040,
      "step": 123570
    },
    {
      "epoch": 18.405570449806376,
      "grad_norm": 13.6875,
      "learning_rate": 9.619479419768596e-07,
      "loss": 1.202,
      "num_input_tokens_seen": 71699112,
      "step": 123575
    },
    {
      "epoch": 18.406315162347333,
      "grad_norm": 7.25,
      "learning_rate": 9.610554390874632e-07,
      "loss": 1.5078,
      "num_input_tokens_seen": 71701992,
      "step": 123580
    },
    {
      "epoch": 18.407059874888294,
      "grad_norm": 9.25,
      "learning_rate": 9.60163342311604e-07,
      "loss": 0.9417,
      "num_input_tokens_seen": 71705448,
      "step": 123585
    },
    {
      "epoch": 18.40780458742925,
      "grad_norm": 18.625,
      "learning_rate": 9.592716516643536e-07,
      "loss": 1.2602,
      "num_input_tokens_seen": 71708968,
      "step": 123590
    },
    {
      "epoch": 18.40854929997021,
      "grad_norm": 7.25,
      "learning_rate": 9.583803671607743e-07,
      "loss": 1.335,
      "num_input_tokens_seen": 71712168,
      "step": 123595
    },
    {
      "epoch": 18.40929401251117,
      "grad_norm": 7.6875,
      "learning_rate": 9.574894888159186e-07,
      "loss": 1.2005,
      "num_input_tokens_seen": 71715048,
      "step": 123600
    },
    {
      "epoch": 18.41003872505213,
      "grad_norm": 16.875,
      "learning_rate": 9.565990166448463e-07,
      "loss": 0.8475,
      "num_input_tokens_seen": 71718088,
      "step": 123605
    },
    {
      "epoch": 18.41078343759309,
      "grad_norm": 6.28125,
      "learning_rate": 9.557089506625954e-07,
      "loss": 0.9899,
      "num_input_tokens_seen": 71720904,
      "step": 123610
    },
    {
      "epoch": 18.41152815013405,
      "grad_norm": 9.5625,
      "learning_rate": 9.54819290884207e-07,
      "loss": 0.748,
      "num_input_tokens_seen": 71723592,
      "step": 123615
    },
    {
      "epoch": 18.412272862675007,
      "grad_norm": 6.03125,
      "learning_rate": 9.539300373247045e-07,
      "loss": 0.8959,
      "num_input_tokens_seen": 71726856,
      "step": 123620
    },
    {
      "epoch": 18.413017575215967,
      "grad_norm": 9.75,
      "learning_rate": 9.530411899991182e-07,
      "loss": 0.9869,
      "num_input_tokens_seen": 71729672,
      "step": 123625
    },
    {
      "epoch": 18.413762287756924,
      "grad_norm": 10.0,
      "learning_rate": 9.521527489224552e-07,
      "loss": 0.8729,
      "num_input_tokens_seen": 71732712,
      "step": 123630
    },
    {
      "epoch": 18.414507000297885,
      "grad_norm": 10.375,
      "learning_rate": 9.512647141097369e-07,
      "loss": 0.8888,
      "num_input_tokens_seen": 71735688,
      "step": 123635
    },
    {
      "epoch": 18.415251712838845,
      "grad_norm": 8.4375,
      "learning_rate": 9.503770855759569e-07,
      "loss": 1.1246,
      "num_input_tokens_seen": 71738504,
      "step": 123640
    },
    {
      "epoch": 18.415996425379802,
      "grad_norm": 10.0,
      "learning_rate": 9.494898633361144e-07,
      "loss": 1.3501,
      "num_input_tokens_seen": 71741320,
      "step": 123645
    },
    {
      "epoch": 18.416741137920763,
      "grad_norm": 5.25,
      "learning_rate": 9.486030474051944e-07,
      "loss": 0.7624,
      "num_input_tokens_seen": 71744104,
      "step": 123650
    },
    {
      "epoch": 18.417485850461723,
      "grad_norm": 11.25,
      "learning_rate": 9.477166377981822e-07,
      "loss": 1.127,
      "num_input_tokens_seen": 71746984,
      "step": 123655
    },
    {
      "epoch": 18.41823056300268,
      "grad_norm": 10.3125,
      "learning_rate": 9.468306345300548e-07,
      "loss": 1.1982,
      "num_input_tokens_seen": 71749832,
      "step": 123660
    },
    {
      "epoch": 18.41897527554364,
      "grad_norm": 10.4375,
      "learning_rate": 9.459450376157697e-07,
      "loss": 0.9997,
      "num_input_tokens_seen": 71753032,
      "step": 123665
    },
    {
      "epoch": 18.419719988084598,
      "grad_norm": 14.5,
      "learning_rate": 9.450598470703037e-07,
      "loss": 0.9915,
      "num_input_tokens_seen": 71756456,
      "step": 123670
    },
    {
      "epoch": 18.42046470062556,
      "grad_norm": 8.8125,
      "learning_rate": 9.441750629086004e-07,
      "loss": 1.3382,
      "num_input_tokens_seen": 71759400,
      "step": 123675
    },
    {
      "epoch": 18.42120941316652,
      "grad_norm": 18.0,
      "learning_rate": 9.432906851456064e-07,
      "loss": 1.4065,
      "num_input_tokens_seen": 71762536,
      "step": 123680
    },
    {
      "epoch": 18.421954125707476,
      "grad_norm": 12.4375,
      "learning_rate": 9.424067137962705e-07,
      "loss": 1.0736,
      "num_input_tokens_seen": 71765288,
      "step": 123685
    },
    {
      "epoch": 18.422698838248436,
      "grad_norm": 21.875,
      "learning_rate": 9.4152314887552e-07,
      "loss": 1.3414,
      "num_input_tokens_seen": 71767848,
      "step": 123690
    },
    {
      "epoch": 18.423443550789397,
      "grad_norm": 6.84375,
      "learning_rate": 9.406399903982844e-07,
      "loss": 1.2433,
      "num_input_tokens_seen": 71770952,
      "step": 123695
    },
    {
      "epoch": 18.424188263330354,
      "grad_norm": 9.4375,
      "learning_rate": 9.397572383794823e-07,
      "loss": 1.0326,
      "num_input_tokens_seen": 71773768,
      "step": 123700
    },
    {
      "epoch": 18.424932975871315,
      "grad_norm": 10.75,
      "learning_rate": 9.388748928340296e-07,
      "loss": 0.9982,
      "num_input_tokens_seen": 71776616,
      "step": 123705
    },
    {
      "epoch": 18.42567768841227,
      "grad_norm": 8.3125,
      "learning_rate": 9.379929537768339e-07,
      "loss": 1.0112,
      "num_input_tokens_seen": 71779560,
      "step": 123710
    },
    {
      "epoch": 18.426422400953232,
      "grad_norm": 11.375,
      "learning_rate": 9.371114212227889e-07,
      "loss": 1.2115,
      "num_input_tokens_seen": 71782344,
      "step": 123715
    },
    {
      "epoch": 18.427167113494193,
      "grad_norm": 10.9375,
      "learning_rate": 9.362302951867907e-07,
      "loss": 1.2643,
      "num_input_tokens_seen": 71784968,
      "step": 123720
    },
    {
      "epoch": 18.42791182603515,
      "grad_norm": 15.3125,
      "learning_rate": 9.353495756837222e-07,
      "loss": 1.2446,
      "num_input_tokens_seen": 71788008,
      "step": 123725
    },
    {
      "epoch": 18.42865653857611,
      "grad_norm": 6.875,
      "learning_rate": 9.344692627284657e-07,
      "loss": 0.9358,
      "num_input_tokens_seen": 71790728,
      "step": 123730
    },
    {
      "epoch": 18.42940125111707,
      "grad_norm": 10.5,
      "learning_rate": 9.335893563358899e-07,
      "loss": 1.1908,
      "num_input_tokens_seen": 71793800,
      "step": 123735
    },
    {
      "epoch": 18.430145963658028,
      "grad_norm": 11.0625,
      "learning_rate": 9.327098565208636e-07,
      "loss": 1.1597,
      "num_input_tokens_seen": 71796808,
      "step": 123740
    },
    {
      "epoch": 18.430890676198988,
      "grad_norm": 12.375,
      "learning_rate": 9.318307632982415e-07,
      "loss": 1.0451,
      "num_input_tokens_seen": 71800008,
      "step": 123745
    },
    {
      "epoch": 18.431635388739945,
      "grad_norm": 8.875,
      "learning_rate": 9.309520766828811e-07,
      "loss": 1.0611,
      "num_input_tokens_seen": 71802792,
      "step": 123750
    },
    {
      "epoch": 18.432380101280906,
      "grad_norm": 11.8125,
      "learning_rate": 9.300737966896206e-07,
      "loss": 0.987,
      "num_input_tokens_seen": 71805576,
      "step": 123755
    },
    {
      "epoch": 18.433124813821866,
      "grad_norm": 11.25,
      "learning_rate": 9.291959233332981e-07,
      "loss": 0.9792,
      "num_input_tokens_seen": 71808488,
      "step": 123760
    },
    {
      "epoch": 18.433869526362823,
      "grad_norm": 13.25,
      "learning_rate": 9.28318456628749e-07,
      "loss": 1.2627,
      "num_input_tokens_seen": 71811208,
      "step": 123765
    },
    {
      "epoch": 18.434614238903784,
      "grad_norm": 6.5625,
      "learning_rate": 9.274413965907919e-07,
      "loss": 0.9593,
      "num_input_tokens_seen": 71814152,
      "step": 123770
    },
    {
      "epoch": 18.43535895144474,
      "grad_norm": 6.5,
      "learning_rate": 9.265647432342455e-07,
      "loss": 0.9696,
      "num_input_tokens_seen": 71817032,
      "step": 123775
    },
    {
      "epoch": 18.4361036639857,
      "grad_norm": 10.3125,
      "learning_rate": 9.256884965739232e-07,
      "loss": 1.0059,
      "num_input_tokens_seen": 71819816,
      "step": 123780
    },
    {
      "epoch": 18.43684837652666,
      "grad_norm": 10.5,
      "learning_rate": 9.248126566246267e-07,
      "loss": 1.1379,
      "num_input_tokens_seen": 71822952,
      "step": 123785
    },
    {
      "epoch": 18.43759308906762,
      "grad_norm": 10.4375,
      "learning_rate": 9.239372234011473e-07,
      "loss": 0.892,
      "num_input_tokens_seen": 71825704,
      "step": 123790
    },
    {
      "epoch": 18.43833780160858,
      "grad_norm": 10.5625,
      "learning_rate": 9.230621969182812e-07,
      "loss": 1.3887,
      "num_input_tokens_seen": 71828520,
      "step": 123795
    },
    {
      "epoch": 18.43908251414954,
      "grad_norm": 6.875,
      "learning_rate": 9.221875771908084e-07,
      "loss": 0.7387,
      "num_input_tokens_seen": 71831208,
      "step": 123800
    },
    {
      "epoch": 18.439827226690497,
      "grad_norm": 9.625,
      "learning_rate": 9.213133642335031e-07,
      "loss": 1.2175,
      "num_input_tokens_seen": 71833960,
      "step": 123805
    },
    {
      "epoch": 18.440571939231457,
      "grad_norm": 11.6875,
      "learning_rate": 9.204395580611397e-07,
      "loss": 1.6948,
      "num_input_tokens_seen": 71836936,
      "step": 123810
    },
    {
      "epoch": 18.441316651772414,
      "grad_norm": 8.25,
      "learning_rate": 9.195661586884729e-07,
      "loss": 1.0095,
      "num_input_tokens_seen": 71840072,
      "step": 123815
    },
    {
      "epoch": 18.442061364313375,
      "grad_norm": 7.21875,
      "learning_rate": 9.186931661302634e-07,
      "loss": 1.102,
      "num_input_tokens_seen": 71843112,
      "step": 123820
    },
    {
      "epoch": 18.442806076854335,
      "grad_norm": 6.78125,
      "learning_rate": 9.178205804012546e-07,
      "loss": 0.9164,
      "num_input_tokens_seen": 71845896,
      "step": 123825
    },
    {
      "epoch": 18.443550789395292,
      "grad_norm": 12.4375,
      "learning_rate": 9.16948401516196e-07,
      "loss": 1.3901,
      "num_input_tokens_seen": 71848648,
      "step": 123830
    },
    {
      "epoch": 18.444295501936253,
      "grad_norm": 12.5625,
      "learning_rate": 9.160766294898148e-07,
      "loss": 0.9844,
      "num_input_tokens_seen": 71851560,
      "step": 123835
    },
    {
      "epoch": 18.445040214477213,
      "grad_norm": 9.875,
      "learning_rate": 9.152052643368408e-07,
      "loss": 0.9596,
      "num_input_tokens_seen": 71854280,
      "step": 123840
    },
    {
      "epoch": 18.44578492701817,
      "grad_norm": 9.0625,
      "learning_rate": 9.143343060719956e-07,
      "loss": 0.9542,
      "num_input_tokens_seen": 71857288,
      "step": 123845
    },
    {
      "epoch": 18.44652963955913,
      "grad_norm": 13.25,
      "learning_rate": 9.13463754709995e-07,
      "loss": 1.3606,
      "num_input_tokens_seen": 71859976,
      "step": 123850
    },
    {
      "epoch": 18.447274352100088,
      "grad_norm": 4.9375,
      "learning_rate": 9.125936102655414e-07,
      "loss": 1.1331,
      "num_input_tokens_seen": 71862920,
      "step": 123855
    },
    {
      "epoch": 18.44801906464105,
      "grad_norm": 11.25,
      "learning_rate": 9.117238727533367e-07,
      "loss": 1.3998,
      "num_input_tokens_seen": 71865608,
      "step": 123860
    },
    {
      "epoch": 18.44876377718201,
      "grad_norm": 6.875,
      "learning_rate": 9.108545421880776e-07,
      "loss": 1.1059,
      "num_input_tokens_seen": 71868616,
      "step": 123865
    },
    {
      "epoch": 18.449508489722966,
      "grad_norm": 17.375,
      "learning_rate": 9.09985618584444e-07,
      "loss": 1.2855,
      "num_input_tokens_seen": 71871720,
      "step": 123870
    },
    {
      "epoch": 18.450253202263927,
      "grad_norm": 5.96875,
      "learning_rate": 9.091171019571215e-07,
      "loss": 0.9841,
      "num_input_tokens_seen": 71874504,
      "step": 123875
    },
    {
      "epoch": 18.450997914804887,
      "grad_norm": 10.5,
      "learning_rate": 9.082489923207815e-07,
      "loss": 1.1246,
      "num_input_tokens_seen": 71877288,
      "step": 123880
    },
    {
      "epoch": 18.451742627345844,
      "grad_norm": 9.6875,
      "learning_rate": 9.073812896900874e-07,
      "loss": 1.11,
      "num_input_tokens_seen": 71880040,
      "step": 123885
    },
    {
      "epoch": 18.452487339886805,
      "grad_norm": 7.15625,
      "learning_rate": 9.065139940797024e-07,
      "loss": 0.9613,
      "num_input_tokens_seen": 71883016,
      "step": 123890
    },
    {
      "epoch": 18.45323205242776,
      "grad_norm": 8.9375,
      "learning_rate": 9.056471055042732e-07,
      "loss": 1.3554,
      "num_input_tokens_seen": 71885928,
      "step": 123895
    },
    {
      "epoch": 18.453976764968722,
      "grad_norm": 8.75,
      "learning_rate": 9.04780623978449e-07,
      "loss": 1.3064,
      "num_input_tokens_seen": 71888680,
      "step": 123900
    },
    {
      "epoch": 18.454721477509683,
      "grad_norm": 7.90625,
      "learning_rate": 9.039145495168655e-07,
      "loss": 1.317,
      "num_input_tokens_seen": 71891656,
      "step": 123905
    },
    {
      "epoch": 18.45546619005064,
      "grad_norm": 5.75,
      "learning_rate": 9.030488821341554e-07,
      "loss": 0.9457,
      "num_input_tokens_seen": 71894632,
      "step": 123910
    },
    {
      "epoch": 18.4562109025916,
      "grad_norm": 11.375,
      "learning_rate": 9.021836218449459e-07,
      "loss": 0.9657,
      "num_input_tokens_seen": 71897800,
      "step": 123915
    },
    {
      "epoch": 18.456955615132557,
      "grad_norm": 9.0,
      "learning_rate": 9.01318768663853e-07,
      "loss": 0.9075,
      "num_input_tokens_seen": 71900808,
      "step": 123920
    },
    {
      "epoch": 18.457700327673518,
      "grad_norm": 5.875,
      "learning_rate": 9.004543226054846e-07,
      "loss": 1.3502,
      "num_input_tokens_seen": 71903720,
      "step": 123925
    },
    {
      "epoch": 18.458445040214478,
      "grad_norm": 13.4375,
      "learning_rate": 8.995902836844455e-07,
      "loss": 0.9659,
      "num_input_tokens_seen": 71906760,
      "step": 123930
    },
    {
      "epoch": 18.459189752755435,
      "grad_norm": 4.84375,
      "learning_rate": 8.987266519153353e-07,
      "loss": 1.1741,
      "num_input_tokens_seen": 71909704,
      "step": 123935
    },
    {
      "epoch": 18.459934465296396,
      "grad_norm": 15.0625,
      "learning_rate": 8.978634273127424e-07,
      "loss": 0.9547,
      "num_input_tokens_seen": 71912520,
      "step": 123940
    },
    {
      "epoch": 18.460679177837356,
      "grad_norm": 12.5625,
      "learning_rate": 8.97000609891252e-07,
      "loss": 0.8063,
      "num_input_tokens_seen": 71915368,
      "step": 123945
    },
    {
      "epoch": 18.461423890378313,
      "grad_norm": 9.5625,
      "learning_rate": 8.961381996654361e-07,
      "loss": 0.8543,
      "num_input_tokens_seen": 71918152,
      "step": 123950
    },
    {
      "epoch": 18.462168602919274,
      "grad_norm": 15.875,
      "learning_rate": 8.952761966498691e-07,
      "loss": 1.1531,
      "num_input_tokens_seen": 71921192,
      "step": 123955
    },
    {
      "epoch": 18.46291331546023,
      "grad_norm": 20.25,
      "learning_rate": 8.944146008591143e-07,
      "loss": 1.0938,
      "num_input_tokens_seen": 71923816,
      "step": 123960
    },
    {
      "epoch": 18.46365802800119,
      "grad_norm": 14.875,
      "learning_rate": 8.93553412307721e-07,
      "loss": 1.3871,
      "num_input_tokens_seen": 71926984,
      "step": 123965
    },
    {
      "epoch": 18.464402740542152,
      "grad_norm": 6.75,
      "learning_rate": 8.926926310102445e-07,
      "loss": 0.9857,
      "num_input_tokens_seen": 71929800,
      "step": 123970
    },
    {
      "epoch": 18.46514745308311,
      "grad_norm": 13.3125,
      "learning_rate": 8.918322569812259e-07,
      "loss": 1.4388,
      "num_input_tokens_seen": 71932616,
      "step": 123975
    },
    {
      "epoch": 18.46589216562407,
      "grad_norm": 8.5625,
      "learning_rate": 8.909722902351924e-07,
      "loss": 1.1727,
      "num_input_tokens_seen": 71935624,
      "step": 123980
    },
    {
      "epoch": 18.46663687816503,
      "grad_norm": 15.4375,
      "learning_rate": 8.901127307866852e-07,
      "loss": 1.4881,
      "num_input_tokens_seen": 71938600,
      "step": 123985
    },
    {
      "epoch": 18.467381590705987,
      "grad_norm": 15.25,
      "learning_rate": 8.892535786502176e-07,
      "loss": 1.4487,
      "num_input_tokens_seen": 71941640,
      "step": 123990
    },
    {
      "epoch": 18.468126303246947,
      "grad_norm": 4.71875,
      "learning_rate": 8.883948338403058e-07,
      "loss": 0.9387,
      "num_input_tokens_seen": 71944488,
      "step": 123995
    },
    {
      "epoch": 18.468871015787904,
      "grad_norm": 14.375,
      "learning_rate": 8.87536496371455e-07,
      "loss": 1.2398,
      "num_input_tokens_seen": 71947592,
      "step": 124000
    },
    {
      "epoch": 18.469615728328865,
      "grad_norm": 5.9375,
      "learning_rate": 8.866785662581728e-07,
      "loss": 0.9513,
      "num_input_tokens_seen": 71950472,
      "step": 124005
    },
    {
      "epoch": 18.470360440869825,
      "grad_norm": 12.0,
      "learning_rate": 8.858210435149422e-07,
      "loss": 0.7508,
      "num_input_tokens_seen": 71953704,
      "step": 124010
    },
    {
      "epoch": 18.471105153410782,
      "grad_norm": 7.53125,
      "learning_rate": 8.849639281562628e-07,
      "loss": 1.2485,
      "num_input_tokens_seen": 71956712,
      "step": 124015
    },
    {
      "epoch": 18.471849865951743,
      "grad_norm": 11.75,
      "learning_rate": 8.841072201966033e-07,
      "loss": 1.2009,
      "num_input_tokens_seen": 71959624,
      "step": 124020
    },
    {
      "epoch": 18.472594578492703,
      "grad_norm": 9.0625,
      "learning_rate": 8.83250919650444e-07,
      "loss": 0.9634,
      "num_input_tokens_seen": 71962664,
      "step": 124025
    },
    {
      "epoch": 18.47333929103366,
      "grad_norm": 15.25,
      "learning_rate": 8.823950265322484e-07,
      "loss": 1.001,
      "num_input_tokens_seen": 71965640,
      "step": 124030
    },
    {
      "epoch": 18.47408400357462,
      "grad_norm": 12.0,
      "learning_rate": 8.815395408564797e-07,
      "loss": 0.759,
      "num_input_tokens_seen": 71968392,
      "step": 124035
    },
    {
      "epoch": 18.474828716115578,
      "grad_norm": 6.25,
      "learning_rate": 8.806844626375848e-07,
      "loss": 1.3351,
      "num_input_tokens_seen": 71971304,
      "step": 124040
    },
    {
      "epoch": 18.47557342865654,
      "grad_norm": 9.8125,
      "learning_rate": 8.798297918900162e-07,
      "loss": 0.9178,
      "num_input_tokens_seen": 71974312,
      "step": 124045
    },
    {
      "epoch": 18.4763181411975,
      "grad_norm": 8.125,
      "learning_rate": 8.789755286282065e-07,
      "loss": 1.0343,
      "num_input_tokens_seen": 71977480,
      "step": 124050
    },
    {
      "epoch": 18.477062853738456,
      "grad_norm": 9.5625,
      "learning_rate": 8.781216728665859e-07,
      "loss": 1.0035,
      "num_input_tokens_seen": 71980136,
      "step": 124055
    },
    {
      "epoch": 18.477807566279417,
      "grad_norm": 16.75,
      "learning_rate": 8.772682246195873e-07,
      "loss": 1.3871,
      "num_input_tokens_seen": 71983048,
      "step": 124060
    },
    {
      "epoch": 18.478552278820374,
      "grad_norm": 12.9375,
      "learning_rate": 8.764151839016216e-07,
      "loss": 1.2298,
      "num_input_tokens_seen": 71985800,
      "step": 124065
    },
    {
      "epoch": 18.479296991361334,
      "grad_norm": 19.875,
      "learning_rate": 8.755625507271076e-07,
      "loss": 1.4032,
      "num_input_tokens_seen": 71988680,
      "step": 124070
    },
    {
      "epoch": 18.480041703902295,
      "grad_norm": 15.125,
      "learning_rate": 8.747103251104394e-07,
      "loss": 1.024,
      "num_input_tokens_seen": 71991464,
      "step": 124075
    },
    {
      "epoch": 18.48078641644325,
      "grad_norm": 11.0625,
      "learning_rate": 8.738585070660249e-07,
      "loss": 1.2148,
      "num_input_tokens_seen": 71994312,
      "step": 124080
    },
    {
      "epoch": 18.481531128984212,
      "grad_norm": 10.0,
      "learning_rate": 8.730070966082499e-07,
      "loss": 1.0564,
      "num_input_tokens_seen": 71997192,
      "step": 124085
    },
    {
      "epoch": 18.482275841525173,
      "grad_norm": 10.5,
      "learning_rate": 8.721560937514972e-07,
      "loss": 1.6068,
      "num_input_tokens_seen": 71999976,
      "step": 124090
    },
    {
      "epoch": 18.48302055406613,
      "grad_norm": 9.625,
      "learning_rate": 8.71305498510147e-07,
      "loss": 0.9093,
      "num_input_tokens_seen": 72002728,
      "step": 124095
    },
    {
      "epoch": 18.48376526660709,
      "grad_norm": 5.40625,
      "learning_rate": 8.70455310898563e-07,
      "loss": 0.958,
      "num_input_tokens_seen": 72005672,
      "step": 124100
    },
    {
      "epoch": 18.484509979148047,
      "grad_norm": 11.1875,
      "learning_rate": 8.696055309311169e-07,
      "loss": 1.1966,
      "num_input_tokens_seen": 72008744,
      "step": 124105
    },
    {
      "epoch": 18.485254691689008,
      "grad_norm": 7.4375,
      "learning_rate": 8.687561586221582e-07,
      "loss": 0.7981,
      "num_input_tokens_seen": 72011464,
      "step": 124110
    },
    {
      "epoch": 18.48599940422997,
      "grad_norm": 7.125,
      "learning_rate": 8.679071939860394e-07,
      "loss": 1.1946,
      "num_input_tokens_seen": 72014280,
      "step": 124115
    },
    {
      "epoch": 18.486744116770925,
      "grad_norm": 11.1875,
      "learning_rate": 8.67058637037102e-07,
      "loss": 1.0157,
      "num_input_tokens_seen": 72017192,
      "step": 124120
    },
    {
      "epoch": 18.487488829311886,
      "grad_norm": 4.46875,
      "learning_rate": 8.662104877896788e-07,
      "loss": 1.0624,
      "num_input_tokens_seen": 72020040,
      "step": 124125
    },
    {
      "epoch": 18.488233541852846,
      "grad_norm": 8.1875,
      "learning_rate": 8.653627462581027e-07,
      "loss": 1.1291,
      "num_input_tokens_seen": 72022856,
      "step": 124130
    },
    {
      "epoch": 18.488978254393803,
      "grad_norm": 10.625,
      "learning_rate": 8.645154124566929e-07,
      "loss": 1.1958,
      "num_input_tokens_seen": 72025672,
      "step": 124135
    },
    {
      "epoch": 18.489722966934764,
      "grad_norm": 14.125,
      "learning_rate": 8.636684863997657e-07,
      "loss": 1.2322,
      "num_input_tokens_seen": 72028584,
      "step": 124140
    },
    {
      "epoch": 18.49046767947572,
      "grad_norm": 9.8125,
      "learning_rate": 8.628219681016264e-07,
      "loss": 0.865,
      "num_input_tokens_seen": 72031432,
      "step": 124145
    },
    {
      "epoch": 18.49121239201668,
      "grad_norm": 14.1875,
      "learning_rate": 8.619758575765801e-07,
      "loss": 1.153,
      "num_input_tokens_seen": 72034088,
      "step": 124150
    },
    {
      "epoch": 18.491957104557642,
      "grad_norm": 6.875,
      "learning_rate": 8.611301548389155e-07,
      "loss": 1.2114,
      "num_input_tokens_seen": 72037000,
      "step": 124155
    },
    {
      "epoch": 18.4927018170986,
      "grad_norm": 14.375,
      "learning_rate": 8.602848599029267e-07,
      "loss": 0.8149,
      "num_input_tokens_seen": 72040040,
      "step": 124160
    },
    {
      "epoch": 18.49344652963956,
      "grad_norm": 10.6875,
      "learning_rate": 8.59439972782894e-07,
      "loss": 1.0625,
      "num_input_tokens_seen": 72042792,
      "step": 124165
    },
    {
      "epoch": 18.49419124218052,
      "grad_norm": 20.375,
      "learning_rate": 8.585954934930806e-07,
      "loss": 1.2481,
      "num_input_tokens_seen": 72045672,
      "step": 124170
    },
    {
      "epoch": 18.494935954721477,
      "grad_norm": 9.0,
      "learning_rate": 8.577514220477644e-07,
      "loss": 1.2511,
      "num_input_tokens_seen": 72048584,
      "step": 124175
    },
    {
      "epoch": 18.495680667262437,
      "grad_norm": 8.875,
      "learning_rate": 8.569077584612006e-07,
      "loss": 1.4718,
      "num_input_tokens_seen": 72051656,
      "step": 124180
    },
    {
      "epoch": 18.496425379803394,
      "grad_norm": 10.875,
      "learning_rate": 8.560645027476416e-07,
      "loss": 1.2488,
      "num_input_tokens_seen": 72054824,
      "step": 124185
    },
    {
      "epoch": 18.497170092344355,
      "grad_norm": 10.8125,
      "learning_rate": 8.552216549213316e-07,
      "loss": 1.1722,
      "num_input_tokens_seen": 72057576,
      "step": 124190
    },
    {
      "epoch": 18.497914804885315,
      "grad_norm": 7.875,
      "learning_rate": 8.543792149965174e-07,
      "loss": 1.2184,
      "num_input_tokens_seen": 72060328,
      "step": 124195
    },
    {
      "epoch": 18.498659517426272,
      "grad_norm": 12.4375,
      "learning_rate": 8.535371829874239e-07,
      "loss": 1.2269,
      "num_input_tokens_seen": 72063432,
      "step": 124200
    },
    {
      "epoch": 18.499404229967233,
      "grad_norm": 8.1875,
      "learning_rate": 8.52695558908273e-07,
      "loss": 1.0495,
      "num_input_tokens_seen": 72066440,
      "step": 124205
    },
    {
      "epoch": 18.500148942508194,
      "grad_norm": 13.25,
      "learning_rate": 8.51854342773295e-07,
      "loss": 1.1246,
      "num_input_tokens_seen": 72069480,
      "step": 124210
    },
    {
      "epoch": 18.50089365504915,
      "grad_norm": 16.75,
      "learning_rate": 8.510135345966897e-07,
      "loss": 0.892,
      "num_input_tokens_seen": 72072264,
      "step": 124215
    },
    {
      "epoch": 18.50163836759011,
      "grad_norm": 9.75,
      "learning_rate": 8.501731343926706e-07,
      "loss": 1.3657,
      "num_input_tokens_seen": 72075688,
      "step": 124220
    },
    {
      "epoch": 18.502383080131068,
      "grad_norm": 15.4375,
      "learning_rate": 8.493331421754291e-07,
      "loss": 0.9808,
      "num_input_tokens_seen": 72079080,
      "step": 124225
    },
    {
      "epoch": 18.50312779267203,
      "grad_norm": 8.25,
      "learning_rate": 8.484935579591596e-07,
      "loss": 0.9725,
      "num_input_tokens_seen": 72082344,
      "step": 124230
    },
    {
      "epoch": 18.50387250521299,
      "grad_norm": 7.21875,
      "learning_rate": 8.476543817580451e-07,
      "loss": 0.911,
      "num_input_tokens_seen": 72085480,
      "step": 124235
    },
    {
      "epoch": 18.504617217753946,
      "grad_norm": 10.0625,
      "learning_rate": 8.468156135862631e-07,
      "loss": 1.0966,
      "num_input_tokens_seen": 72088488,
      "step": 124240
    },
    {
      "epoch": 18.505361930294907,
      "grad_norm": 13.75,
      "learning_rate": 8.45977253457983e-07,
      "loss": 1.3689,
      "num_input_tokens_seen": 72091720,
      "step": 124245
    },
    {
      "epoch": 18.506106642835867,
      "grad_norm": 15.5625,
      "learning_rate": 8.451393013873682e-07,
      "loss": 1.4833,
      "num_input_tokens_seen": 72094536,
      "step": 124250
    },
    {
      "epoch": 18.506851355376824,
      "grad_norm": 18.25,
      "learning_rate": 8.443017573885769e-07,
      "loss": 1.2861,
      "num_input_tokens_seen": 72097384,
      "step": 124255
    },
    {
      "epoch": 18.507596067917785,
      "grad_norm": 9.8125,
      "learning_rate": 8.434646214757536e-07,
      "loss": 0.9688,
      "num_input_tokens_seen": 72100008,
      "step": 124260
    },
    {
      "epoch": 18.50834078045874,
      "grad_norm": 7.90625,
      "learning_rate": 8.42627893663045e-07,
      "loss": 1.0938,
      "num_input_tokens_seen": 72102664,
      "step": 124265
    },
    {
      "epoch": 18.509085492999702,
      "grad_norm": 13.125,
      "learning_rate": 8.417915739645815e-07,
      "loss": 1.2486,
      "num_input_tokens_seen": 72105320,
      "step": 124270
    },
    {
      "epoch": 18.509830205540663,
      "grad_norm": 9.375,
      "learning_rate": 8.40955662394502e-07,
      "loss": 1.3152,
      "num_input_tokens_seen": 72108168,
      "step": 124275
    },
    {
      "epoch": 18.51057491808162,
      "grad_norm": 8.0,
      "learning_rate": 8.401201589669227e-07,
      "loss": 1.0204,
      "num_input_tokens_seen": 72111432,
      "step": 124280
    },
    {
      "epoch": 18.51131963062258,
      "grad_norm": 11.125,
      "learning_rate": 8.392850636959521e-07,
      "loss": 0.9908,
      "num_input_tokens_seen": 72114248,
      "step": 124285
    },
    {
      "epoch": 18.512064343163537,
      "grad_norm": 10.0,
      "learning_rate": 8.384503765957091e-07,
      "loss": 1.091,
      "num_input_tokens_seen": 72117256,
      "step": 124290
    },
    {
      "epoch": 18.512809055704498,
      "grad_norm": 10.6875,
      "learning_rate": 8.376160976802882e-07,
      "loss": 1.0826,
      "num_input_tokens_seen": 72119944,
      "step": 124295
    },
    {
      "epoch": 18.51355376824546,
      "grad_norm": 9.8125,
      "learning_rate": 8.367822269637892e-07,
      "loss": 1.3494,
      "num_input_tokens_seen": 72122888,
      "step": 124300
    },
    {
      "epoch": 18.514298480786415,
      "grad_norm": 14.0,
      "learning_rate": 8.359487644602954e-07,
      "loss": 1.2954,
      "num_input_tokens_seen": 72125704,
      "step": 124305
    },
    {
      "epoch": 18.515043193327376,
      "grad_norm": 13.6875,
      "learning_rate": 8.351157101838842e-07,
      "loss": 1.2281,
      "num_input_tokens_seen": 72128552,
      "step": 124310
    },
    {
      "epoch": 18.515787905868336,
      "grad_norm": 15.375,
      "learning_rate": 8.34283064148636e-07,
      "loss": 0.7587,
      "num_input_tokens_seen": 72131496,
      "step": 124315
    },
    {
      "epoch": 18.516532618409293,
      "grad_norm": 14.6875,
      "learning_rate": 8.334508263686147e-07,
      "loss": 1.1218,
      "num_input_tokens_seen": 72134440,
      "step": 124320
    },
    {
      "epoch": 18.517277330950254,
      "grad_norm": 15.8125,
      "learning_rate": 8.326189968578785e-07,
      "loss": 1.2564,
      "num_input_tokens_seen": 72137320,
      "step": 124325
    },
    {
      "epoch": 18.51802204349121,
      "grad_norm": 19.125,
      "learning_rate": 8.317875756304827e-07,
      "loss": 1.0914,
      "num_input_tokens_seen": 72140136,
      "step": 124330
    },
    {
      "epoch": 18.51876675603217,
      "grad_norm": 11.0625,
      "learning_rate": 8.309565627004717e-07,
      "loss": 1.1975,
      "num_input_tokens_seen": 72143048,
      "step": 124335
    },
    {
      "epoch": 18.519511468573132,
      "grad_norm": 16.125,
      "learning_rate": 8.301259580818843e-07,
      "loss": 1.2211,
      "num_input_tokens_seen": 72145704,
      "step": 124340
    },
    {
      "epoch": 18.52025618111409,
      "grad_norm": 13.375,
      "learning_rate": 8.292957617887537e-07,
      "loss": 1.0037,
      "num_input_tokens_seen": 72148456,
      "step": 124345
    },
    {
      "epoch": 18.52100089365505,
      "grad_norm": 17.875,
      "learning_rate": 8.284659738351047e-07,
      "loss": 0.967,
      "num_input_tokens_seen": 72151272,
      "step": 124350
    },
    {
      "epoch": 18.52174560619601,
      "grad_norm": 11.875,
      "learning_rate": 8.276365942349595e-07,
      "loss": 1.1312,
      "num_input_tokens_seen": 72154056,
      "step": 124355
    },
    {
      "epoch": 18.522490318736967,
      "grad_norm": 9.0,
      "learning_rate": 8.268076230023264e-07,
      "loss": 1.0417,
      "num_input_tokens_seen": 72156968,
      "step": 124360
    },
    {
      "epoch": 18.523235031277927,
      "grad_norm": 9.6875,
      "learning_rate": 8.259790601512052e-07,
      "loss": 1.0685,
      "num_input_tokens_seen": 72159688,
      "step": 124365
    },
    {
      "epoch": 18.523979743818884,
      "grad_norm": 14.5,
      "learning_rate": 8.251509056956042e-07,
      "loss": 1.3074,
      "num_input_tokens_seen": 72162632,
      "step": 124370
    },
    {
      "epoch": 18.524724456359845,
      "grad_norm": 11.0,
      "learning_rate": 8.243231596495066e-07,
      "loss": 1.5776,
      "num_input_tokens_seen": 72165480,
      "step": 124375
    },
    {
      "epoch": 18.525469168900806,
      "grad_norm": 14.875,
      "learning_rate": 8.234958220268985e-07,
      "loss": 1.0809,
      "num_input_tokens_seen": 72168456,
      "step": 124380
    },
    {
      "epoch": 18.526213881441763,
      "grad_norm": 7.03125,
      "learning_rate": 8.22668892841752e-07,
      "loss": 1.2362,
      "num_input_tokens_seen": 72171496,
      "step": 124385
    },
    {
      "epoch": 18.526958593982723,
      "grad_norm": 14.375,
      "learning_rate": 8.218423721080476e-07,
      "loss": 1.4444,
      "num_input_tokens_seen": 72174408,
      "step": 124390
    },
    {
      "epoch": 18.527703306523684,
      "grad_norm": 6.40625,
      "learning_rate": 8.21016259839738e-07,
      "loss": 0.9059,
      "num_input_tokens_seen": 72177256,
      "step": 124395
    },
    {
      "epoch": 18.52844801906464,
      "grad_norm": 5.71875,
      "learning_rate": 8.201905560507872e-07,
      "loss": 0.8281,
      "num_input_tokens_seen": 72180296,
      "step": 124400
    },
    {
      "epoch": 18.5291927316056,
      "grad_norm": 8.8125,
      "learning_rate": 8.193652607551422e-07,
      "loss": 0.8356,
      "num_input_tokens_seen": 72183144,
      "step": 124405
    },
    {
      "epoch": 18.529937444146558,
      "grad_norm": 20.375,
      "learning_rate": 8.185403739667419e-07,
      "loss": 1.2084,
      "num_input_tokens_seen": 72185864,
      "step": 124410
    },
    {
      "epoch": 18.53068215668752,
      "grad_norm": 12.0,
      "learning_rate": 8.177158956995279e-07,
      "loss": 1.4867,
      "num_input_tokens_seen": 72188616,
      "step": 124415
    },
    {
      "epoch": 18.53142686922848,
      "grad_norm": 17.125,
      "learning_rate": 8.168918259674224e-07,
      "loss": 1.1891,
      "num_input_tokens_seen": 72191304,
      "step": 124420
    },
    {
      "epoch": 18.532171581769436,
      "grad_norm": 7.875,
      "learning_rate": 8.16068164784356e-07,
      "loss": 1.1452,
      "num_input_tokens_seen": 72194120,
      "step": 124425
    },
    {
      "epoch": 18.532916294310397,
      "grad_norm": 10.875,
      "learning_rate": 8.152449121642342e-07,
      "loss": 0.9852,
      "num_input_tokens_seen": 72197128,
      "step": 124430
    },
    {
      "epoch": 18.533661006851354,
      "grad_norm": 8.1875,
      "learning_rate": 8.144220681209708e-07,
      "loss": 1.0896,
      "num_input_tokens_seen": 72200008,
      "step": 124435
    },
    {
      "epoch": 18.534405719392314,
      "grad_norm": 9.0625,
      "learning_rate": 8.135996326684686e-07,
      "loss": 1.0798,
      "num_input_tokens_seen": 72202728,
      "step": 124440
    },
    {
      "epoch": 18.535150431933275,
      "grad_norm": 14.125,
      "learning_rate": 8.127776058206166e-07,
      "loss": 1.5239,
      "num_input_tokens_seen": 72205704,
      "step": 124445
    },
    {
      "epoch": 18.53589514447423,
      "grad_norm": 12.875,
      "learning_rate": 8.119559875913036e-07,
      "loss": 1.5451,
      "num_input_tokens_seen": 72208520,
      "step": 124450
    },
    {
      "epoch": 18.536639857015192,
      "grad_norm": 12.9375,
      "learning_rate": 8.111347779944101e-07,
      "loss": 1.2161,
      "num_input_tokens_seen": 72211208,
      "step": 124455
    },
    {
      "epoch": 18.537384569556153,
      "grad_norm": 6.71875,
      "learning_rate": 8.103139770438112e-07,
      "loss": 1.1897,
      "num_input_tokens_seen": 72213928,
      "step": 124460
    },
    {
      "epoch": 18.53812928209711,
      "grad_norm": 16.75,
      "learning_rate": 8.09493584753368e-07,
      "loss": 0.8909,
      "num_input_tokens_seen": 72216648,
      "step": 124465
    },
    {
      "epoch": 18.53887399463807,
      "grad_norm": 6.46875,
      "learning_rate": 8.0867360113695e-07,
      "loss": 1.0116,
      "num_input_tokens_seen": 72219240,
      "step": 124470
    },
    {
      "epoch": 18.539618707179027,
      "grad_norm": 15.9375,
      "learning_rate": 8.078540262084017e-07,
      "loss": 1.7177,
      "num_input_tokens_seen": 72222184,
      "step": 124475
    },
    {
      "epoch": 18.540363419719988,
      "grad_norm": 13.25,
      "learning_rate": 8.07034859981573e-07,
      "loss": 0.9589,
      "num_input_tokens_seen": 72225032,
      "step": 124480
    },
    {
      "epoch": 18.54110813226095,
      "grad_norm": 7.875,
      "learning_rate": 8.062161024703029e-07,
      "loss": 0.8945,
      "num_input_tokens_seen": 72227784,
      "step": 124485
    },
    {
      "epoch": 18.541852844801905,
      "grad_norm": 15.5,
      "learning_rate": 8.053977536884194e-07,
      "loss": 1.3569,
      "num_input_tokens_seen": 72230664,
      "step": 124490
    },
    {
      "epoch": 18.542597557342866,
      "grad_norm": 12.25,
      "learning_rate": 8.045798136497529e-07,
      "loss": 1.0162,
      "num_input_tokens_seen": 72233512,
      "step": 124495
    },
    {
      "epoch": 18.543342269883826,
      "grad_norm": 16.875,
      "learning_rate": 8.037622823681174e-07,
      "loss": 1.2313,
      "num_input_tokens_seen": 72236712,
      "step": 124500
    },
    {
      "epoch": 18.544086982424783,
      "grad_norm": 13.0,
      "learning_rate": 8.029451598573267e-07,
      "loss": 1.282,
      "num_input_tokens_seen": 72239528,
      "step": 124505
    },
    {
      "epoch": 18.544831694965744,
      "grad_norm": 11.0,
      "learning_rate": 8.021284461311867e-07,
      "loss": 0.9473,
      "num_input_tokens_seen": 72242344,
      "step": 124510
    },
    {
      "epoch": 18.5455764075067,
      "grad_norm": 5.6875,
      "learning_rate": 8.013121412034919e-07,
      "loss": 1.1856,
      "num_input_tokens_seen": 72245288,
      "step": 124515
    },
    {
      "epoch": 18.54632112004766,
      "grad_norm": 15.0,
      "learning_rate": 8.004962450880338e-07,
      "loss": 1.3867,
      "num_input_tokens_seen": 72248360,
      "step": 124520
    },
    {
      "epoch": 18.547065832588622,
      "grad_norm": 7.3125,
      "learning_rate": 7.99680757798596e-07,
      "loss": 1.1476,
      "num_input_tokens_seen": 72251048,
      "step": 124525
    },
    {
      "epoch": 18.54781054512958,
      "grad_norm": 10.9375,
      "learning_rate": 7.988656793489563e-07,
      "loss": 1.0796,
      "num_input_tokens_seen": 72254088,
      "step": 124530
    },
    {
      "epoch": 18.54855525767054,
      "grad_norm": 11.75,
      "learning_rate": 7.980510097528815e-07,
      "loss": 1.3481,
      "num_input_tokens_seen": 72257352,
      "step": 124535
    },
    {
      "epoch": 18.5492999702115,
      "grad_norm": 6.03125,
      "learning_rate": 7.972367490241412e-07,
      "loss": 0.7567,
      "num_input_tokens_seen": 72260456,
      "step": 124540
    },
    {
      "epoch": 18.550044682752457,
      "grad_norm": 14.375,
      "learning_rate": 7.964228971764826e-07,
      "loss": 1.0841,
      "num_input_tokens_seen": 72263240,
      "step": 124545
    },
    {
      "epoch": 18.550789395293418,
      "grad_norm": 19.375,
      "learning_rate": 7.956094542236642e-07,
      "loss": 1.1636,
      "num_input_tokens_seen": 72266120,
      "step": 124550
    },
    {
      "epoch": 18.551534107834375,
      "grad_norm": 8.8125,
      "learning_rate": 7.947964201794223e-07,
      "loss": 1.0774,
      "num_input_tokens_seen": 72269160,
      "step": 124555
    },
    {
      "epoch": 18.552278820375335,
      "grad_norm": 10.25,
      "learning_rate": 7.93983795057493e-07,
      "loss": 1.1357,
      "num_input_tokens_seen": 72271848,
      "step": 124560
    },
    {
      "epoch": 18.553023532916296,
      "grad_norm": 11.0625,
      "learning_rate": 7.931715788716071e-07,
      "loss": 1.5374,
      "num_input_tokens_seen": 72274696,
      "step": 124565
    },
    {
      "epoch": 18.553768245457253,
      "grad_norm": 13.25,
      "learning_rate": 7.923597716354841e-07,
      "loss": 1.3255,
      "num_input_tokens_seen": 72277512,
      "step": 124570
    },
    {
      "epoch": 18.554512957998213,
      "grad_norm": 9.0,
      "learning_rate": 7.915483733628382e-07,
      "loss": 0.7362,
      "num_input_tokens_seen": 72280552,
      "step": 124575
    },
    {
      "epoch": 18.55525767053917,
      "grad_norm": 9.9375,
      "learning_rate": 7.907373840673804e-07,
      "loss": 0.8714,
      "num_input_tokens_seen": 72283464,
      "step": 124580
    },
    {
      "epoch": 18.55600238308013,
      "grad_norm": 12.1875,
      "learning_rate": 7.899268037628082e-07,
      "loss": 0.892,
      "num_input_tokens_seen": 72286600,
      "step": 124585
    },
    {
      "epoch": 18.55674709562109,
      "grad_norm": 13.5,
      "learning_rate": 7.891166324628163e-07,
      "loss": 1.3248,
      "num_input_tokens_seen": 72289448,
      "step": 124590
    },
    {
      "epoch": 18.557491808162048,
      "grad_norm": 13.6875,
      "learning_rate": 7.883068701810936e-07,
      "loss": 1.0022,
      "num_input_tokens_seen": 72292328,
      "step": 124595
    },
    {
      "epoch": 18.55823652070301,
      "grad_norm": 14.0,
      "learning_rate": 7.874975169313181e-07,
      "loss": 1.3207,
      "num_input_tokens_seen": 72295784,
      "step": 124600
    },
    {
      "epoch": 18.55898123324397,
      "grad_norm": 7.65625,
      "learning_rate": 7.866885727271594e-07,
      "loss": 0.9962,
      "num_input_tokens_seen": 72298984,
      "step": 124605
    },
    {
      "epoch": 18.559725945784926,
      "grad_norm": 9.6875,
      "learning_rate": 7.858800375822928e-07,
      "loss": 1.0565,
      "num_input_tokens_seen": 72301896,
      "step": 124610
    },
    {
      "epoch": 18.560470658325887,
      "grad_norm": 16.0,
      "learning_rate": 7.850719115103683e-07,
      "loss": 1.2846,
      "num_input_tokens_seen": 72304968,
      "step": 124615
    },
    {
      "epoch": 18.561215370866844,
      "grad_norm": 11.625,
      "learning_rate": 7.842641945250473e-07,
      "loss": 1.0701,
      "num_input_tokens_seen": 72307720,
      "step": 124620
    },
    {
      "epoch": 18.561960083407804,
      "grad_norm": 10.0625,
      "learning_rate": 7.834568866399688e-07,
      "loss": 1.2498,
      "num_input_tokens_seen": 72310376,
      "step": 124625
    },
    {
      "epoch": 18.562704795948765,
      "grad_norm": 8.25,
      "learning_rate": 7.826499878687749e-07,
      "loss": 1.2347,
      "num_input_tokens_seen": 72313256,
      "step": 124630
    },
    {
      "epoch": 18.56344950848972,
      "grad_norm": 21.375,
      "learning_rate": 7.81843498225096e-07,
      "loss": 0.9893,
      "num_input_tokens_seen": 72316232,
      "step": 124635
    },
    {
      "epoch": 18.564194221030682,
      "grad_norm": 7.28125,
      "learning_rate": 7.810374177225549e-07,
      "loss": 0.9233,
      "num_input_tokens_seen": 72319336,
      "step": 124640
    },
    {
      "epoch": 18.564938933571643,
      "grad_norm": 5.375,
      "learning_rate": 7.802317463747738e-07,
      "loss": 0.8529,
      "num_input_tokens_seen": 72322312,
      "step": 124645
    },
    {
      "epoch": 18.5656836461126,
      "grad_norm": 14.625,
      "learning_rate": 7.794264841953613e-07,
      "loss": 1.5807,
      "num_input_tokens_seen": 72325032,
      "step": 124650
    },
    {
      "epoch": 18.56642835865356,
      "grad_norm": 11.5,
      "learning_rate": 7.786216311979233e-07,
      "loss": 0.9461,
      "num_input_tokens_seen": 72327912,
      "step": 124655
    },
    {
      "epoch": 18.567173071194517,
      "grad_norm": 11.0625,
      "learning_rate": 7.778171873960516e-07,
      "loss": 1.2378,
      "num_input_tokens_seen": 72330792,
      "step": 124660
    },
    {
      "epoch": 18.567917783735478,
      "grad_norm": 5.71875,
      "learning_rate": 7.770131528033409e-07,
      "loss": 0.7536,
      "num_input_tokens_seen": 72333736,
      "step": 124665
    },
    {
      "epoch": 18.56866249627644,
      "grad_norm": 8.4375,
      "learning_rate": 7.762095274333747e-07,
      "loss": 0.7424,
      "num_input_tokens_seen": 72336360,
      "step": 124670
    },
    {
      "epoch": 18.569407208817395,
      "grad_norm": 8.4375,
      "learning_rate": 7.754063112997284e-07,
      "loss": 0.9887,
      "num_input_tokens_seen": 72339112,
      "step": 124675
    },
    {
      "epoch": 18.570151921358356,
      "grad_norm": 10.3125,
      "learning_rate": 7.746035044159688e-07,
      "loss": 1.1299,
      "num_input_tokens_seen": 72342184,
      "step": 124680
    },
    {
      "epoch": 18.570896633899316,
      "grad_norm": 10.5625,
      "learning_rate": 7.738011067956658e-07,
      "loss": 0.9739,
      "num_input_tokens_seen": 72345032,
      "step": 124685
    },
    {
      "epoch": 18.571641346440273,
      "grad_norm": 17.25,
      "learning_rate": 7.729991184523722e-07,
      "loss": 1.4788,
      "num_input_tokens_seen": 72347880,
      "step": 124690
    },
    {
      "epoch": 18.572386058981234,
      "grad_norm": 10.0625,
      "learning_rate": 7.7219753939963e-07,
      "loss": 1.0212,
      "num_input_tokens_seen": 72350920,
      "step": 124695
    },
    {
      "epoch": 18.57313077152219,
      "grad_norm": 8.9375,
      "learning_rate": 7.713963696509896e-07,
      "loss": 1.285,
      "num_input_tokens_seen": 72353960,
      "step": 124700
    },
    {
      "epoch": 18.57387548406315,
      "grad_norm": 8.5625,
      "learning_rate": 7.705956092199818e-07,
      "loss": 1.1797,
      "num_input_tokens_seen": 72357288,
      "step": 124705
    },
    {
      "epoch": 18.574620196604112,
      "grad_norm": 16.75,
      "learning_rate": 7.697952581201373e-07,
      "loss": 1.2215,
      "num_input_tokens_seen": 72360072,
      "step": 124710
    },
    {
      "epoch": 18.57536490914507,
      "grad_norm": 8.5,
      "learning_rate": 7.689953163649704e-07,
      "loss": 1.3718,
      "num_input_tokens_seen": 72362824,
      "step": 124715
    },
    {
      "epoch": 18.57610962168603,
      "grad_norm": 12.5625,
      "learning_rate": 7.681957839680065e-07,
      "loss": 1.2891,
      "num_input_tokens_seen": 72365608,
      "step": 124720
    },
    {
      "epoch": 18.57685433422699,
      "grad_norm": 10.9375,
      "learning_rate": 7.67396660942743e-07,
      "loss": 1.1819,
      "num_input_tokens_seen": 72368616,
      "step": 124725
    },
    {
      "epoch": 18.577599046767947,
      "grad_norm": 10.5,
      "learning_rate": 7.66597947302683e-07,
      "loss": 1.1951,
      "num_input_tokens_seen": 72371592,
      "step": 124730
    },
    {
      "epoch": 18.578343759308908,
      "grad_norm": 6.46875,
      "learning_rate": 7.65799643061324e-07,
      "loss": 1.3507,
      "num_input_tokens_seen": 72374216,
      "step": 124735
    },
    {
      "epoch": 18.579088471849865,
      "grad_norm": 13.375,
      "learning_rate": 7.65001748232147e-07,
      "loss": 1.2096,
      "num_input_tokens_seen": 72377064,
      "step": 124740
    },
    {
      "epoch": 18.579833184390825,
      "grad_norm": 8.1875,
      "learning_rate": 7.642042628286355e-07,
      "loss": 1.0953,
      "num_input_tokens_seen": 72380008,
      "step": 124745
    },
    {
      "epoch": 18.580577896931786,
      "grad_norm": 6.40625,
      "learning_rate": 7.634071868642595e-07,
      "loss": 0.9492,
      "num_input_tokens_seen": 72382920,
      "step": 124750
    },
    {
      "epoch": 18.581322609472743,
      "grad_norm": 11.625,
      "learning_rate": 7.626105203524886e-07,
      "loss": 1.0969,
      "num_input_tokens_seen": 72385704,
      "step": 124755
    },
    {
      "epoch": 18.582067322013703,
      "grad_norm": 16.375,
      "learning_rate": 7.61814263306776e-07,
      "loss": 1.3577,
      "num_input_tokens_seen": 72388200,
      "step": 124760
    },
    {
      "epoch": 18.582812034554664,
      "grad_norm": 12.4375,
      "learning_rate": 7.610184157405803e-07,
      "loss": 1.2113,
      "num_input_tokens_seen": 72391240,
      "step": 124765
    },
    {
      "epoch": 18.58355674709562,
      "grad_norm": 8.0625,
      "learning_rate": 7.602229776673409e-07,
      "loss": 1.007,
      "num_input_tokens_seen": 72394248,
      "step": 124770
    },
    {
      "epoch": 18.58430145963658,
      "grad_norm": 14.9375,
      "learning_rate": 7.594279491004997e-07,
      "loss": 0.8398,
      "num_input_tokens_seen": 72397224,
      "step": 124775
    },
    {
      "epoch": 18.585046172177538,
      "grad_norm": 11.125,
      "learning_rate": 7.586333300534876e-07,
      "loss": 1.0531,
      "num_input_tokens_seen": 72400360,
      "step": 124780
    },
    {
      "epoch": 18.5857908847185,
      "grad_norm": 15.9375,
      "learning_rate": 7.578391205397218e-07,
      "loss": 1.0744,
      "num_input_tokens_seen": 72403304,
      "step": 124785
    },
    {
      "epoch": 18.58653559725946,
      "grad_norm": 6.90625,
      "learning_rate": 7.570453205726303e-07,
      "loss": 1.2481,
      "num_input_tokens_seen": 72406056,
      "step": 124790
    },
    {
      "epoch": 18.587280309800416,
      "grad_norm": 13.0625,
      "learning_rate": 7.562519301656162e-07,
      "loss": 1.1827,
      "num_input_tokens_seen": 72408840,
      "step": 124795
    },
    {
      "epoch": 18.588025022341377,
      "grad_norm": 6.0,
      "learning_rate": 7.554589493320885e-07,
      "loss": 1.1133,
      "num_input_tokens_seen": 72411880,
      "step": 124800
    },
    {
      "epoch": 18.588769734882334,
      "grad_norm": 15.75,
      "learning_rate": 7.54666378085439e-07,
      "loss": 1.3041,
      "num_input_tokens_seen": 72414856,
      "step": 124805
    },
    {
      "epoch": 18.589514447423294,
      "grad_norm": 9.5625,
      "learning_rate": 7.538742164390572e-07,
      "loss": 0.7748,
      "num_input_tokens_seen": 72417896,
      "step": 124810
    },
    {
      "epoch": 18.590259159964255,
      "grad_norm": 10.875,
      "learning_rate": 7.530824644063295e-07,
      "loss": 1.3348,
      "num_input_tokens_seen": 72420840,
      "step": 124815
    },
    {
      "epoch": 18.591003872505212,
      "grad_norm": 16.125,
      "learning_rate": 7.522911220006285e-07,
      "loss": 1.4508,
      "num_input_tokens_seen": 72423528,
      "step": 124820
    },
    {
      "epoch": 18.591748585046172,
      "grad_norm": 7.8125,
      "learning_rate": 7.515001892353268e-07,
      "loss": 1.022,
      "num_input_tokens_seen": 72426568,
      "step": 124825
    },
    {
      "epoch": 18.592493297587133,
      "grad_norm": 12.5,
      "learning_rate": 7.507096661237834e-07,
      "loss": 1.044,
      "num_input_tokens_seen": 72429288,
      "step": 124830
    },
    {
      "epoch": 18.59323801012809,
      "grad_norm": 17.375,
      "learning_rate": 7.499195526793567e-07,
      "loss": 1.2512,
      "num_input_tokens_seen": 72432072,
      "step": 124835
    },
    {
      "epoch": 18.59398272266905,
      "grad_norm": 11.3125,
      "learning_rate": 7.491298489153919e-07,
      "loss": 1.2037,
      "num_input_tokens_seen": 72435176,
      "step": 124840
    },
    {
      "epoch": 18.594727435210007,
      "grad_norm": 10.8125,
      "learning_rate": 7.483405548452283e-07,
      "loss": 1.4784,
      "num_input_tokens_seen": 72438088,
      "step": 124845
    },
    {
      "epoch": 18.595472147750968,
      "grad_norm": 6.125,
      "learning_rate": 7.47551670482205e-07,
      "loss": 0.9287,
      "num_input_tokens_seen": 72440808,
      "step": 124850
    },
    {
      "epoch": 18.59621686029193,
      "grad_norm": 8.3125,
      "learning_rate": 7.467631958396448e-07,
      "loss": 1.0046,
      "num_input_tokens_seen": 72443880,
      "step": 124855
    },
    {
      "epoch": 18.596961572832885,
      "grad_norm": 16.375,
      "learning_rate": 7.459751309308733e-07,
      "loss": 1.0826,
      "num_input_tokens_seen": 72446792,
      "step": 124860
    },
    {
      "epoch": 18.597706285373846,
      "grad_norm": 10.3125,
      "learning_rate": 7.451874757691991e-07,
      "loss": 1.0739,
      "num_input_tokens_seen": 72449736,
      "step": 124865
    },
    {
      "epoch": 18.598450997914806,
      "grad_norm": 7.6875,
      "learning_rate": 7.444002303679309e-07,
      "loss": 1.0356,
      "num_input_tokens_seen": 72452584,
      "step": 124870
    },
    {
      "epoch": 18.599195710455763,
      "grad_norm": 5.3125,
      "learning_rate": 7.436133947403695e-07,
      "loss": 0.8076,
      "num_input_tokens_seen": 72455592,
      "step": 124875
    },
    {
      "epoch": 18.599940422996724,
      "grad_norm": 11.4375,
      "learning_rate": 7.428269688998068e-07,
      "loss": 1.0326,
      "num_input_tokens_seen": 72458440,
      "step": 124880
    },
    {
      "epoch": 18.60068513553768,
      "grad_norm": 7.96875,
      "learning_rate": 7.420409528595296e-07,
      "loss": 1.2155,
      "num_input_tokens_seen": 72461224,
      "step": 124885
    },
    {
      "epoch": 18.60142984807864,
      "grad_norm": 15.3125,
      "learning_rate": 7.412553466328131e-07,
      "loss": 1.0272,
      "num_input_tokens_seen": 72464232,
      "step": 124890
    },
    {
      "epoch": 18.602174560619602,
      "grad_norm": 8.9375,
      "learning_rate": 7.404701502329331e-07,
      "loss": 0.9516,
      "num_input_tokens_seen": 72466792,
      "step": 124895
    },
    {
      "epoch": 18.60291927316056,
      "grad_norm": 7.28125,
      "learning_rate": 7.396853636731537e-07,
      "loss": 0.892,
      "num_input_tokens_seen": 72469512,
      "step": 124900
    },
    {
      "epoch": 18.60366398570152,
      "grad_norm": 6.46875,
      "learning_rate": 7.389009869667341e-07,
      "loss": 1.2979,
      "num_input_tokens_seen": 72472744,
      "step": 124905
    },
    {
      "epoch": 18.60440869824248,
      "grad_norm": 7.21875,
      "learning_rate": 7.381170201269244e-07,
      "loss": 1.0011,
      "num_input_tokens_seen": 72475592,
      "step": 124910
    },
    {
      "epoch": 18.605153410783437,
      "grad_norm": 7.375,
      "learning_rate": 7.373334631669698e-07,
      "loss": 1.2721,
      "num_input_tokens_seen": 72478568,
      "step": 124915
    },
    {
      "epoch": 18.605898123324398,
      "grad_norm": 10.3125,
      "learning_rate": 7.365503161001013e-07,
      "loss": 1.4812,
      "num_input_tokens_seen": 72481384,
      "step": 124920
    },
    {
      "epoch": 18.606642835865355,
      "grad_norm": 5.25,
      "learning_rate": 7.357675789395613e-07,
      "loss": 0.9841,
      "num_input_tokens_seen": 72484104,
      "step": 124925
    },
    {
      "epoch": 18.607387548406315,
      "grad_norm": 6.6875,
      "learning_rate": 7.349852516985639e-07,
      "loss": 1.4235,
      "num_input_tokens_seen": 72486920,
      "step": 124930
    },
    {
      "epoch": 18.608132260947276,
      "grad_norm": 11.6875,
      "learning_rate": 7.342033343903293e-07,
      "loss": 1.0441,
      "num_input_tokens_seen": 72489896,
      "step": 124935
    },
    {
      "epoch": 18.608876973488233,
      "grad_norm": 7.84375,
      "learning_rate": 7.33421827028069e-07,
      "loss": 1.1037,
      "num_input_tokens_seen": 72492936,
      "step": 124940
    },
    {
      "epoch": 18.609621686029193,
      "grad_norm": 4.25,
      "learning_rate": 7.326407296249782e-07,
      "loss": 1.1178,
      "num_input_tokens_seen": 72496104,
      "step": 124945
    },
    {
      "epoch": 18.61036639857015,
      "grad_norm": 13.875,
      "learning_rate": 7.318600421942628e-07,
      "loss": 1.1224,
      "num_input_tokens_seen": 72498952,
      "step": 124950
    },
    {
      "epoch": 18.61111111111111,
      "grad_norm": 7.4375,
      "learning_rate": 7.310797647491041e-07,
      "loss": 1.1094,
      "num_input_tokens_seen": 72501864,
      "step": 124955
    },
    {
      "epoch": 18.61185582365207,
      "grad_norm": 9.5625,
      "learning_rate": 7.302998973026887e-07,
      "loss": 1.3646,
      "num_input_tokens_seen": 72504840,
      "step": 124960
    },
    {
      "epoch": 18.61260053619303,
      "grad_norm": 6.28125,
      "learning_rate": 7.295204398681893e-07,
      "loss": 1.0514,
      "num_input_tokens_seen": 72507816,
      "step": 124965
    },
    {
      "epoch": 18.61334524873399,
      "grad_norm": 8.6875,
      "learning_rate": 7.287413924587733e-07,
      "loss": 1.2174,
      "num_input_tokens_seen": 72510600,
      "step": 124970
    },
    {
      "epoch": 18.61408996127495,
      "grad_norm": 14.5625,
      "learning_rate": 7.279627550876051e-07,
      "loss": 1.0375,
      "num_input_tokens_seen": 72513512,
      "step": 124975
    },
    {
      "epoch": 18.614834673815906,
      "grad_norm": 11.5,
      "learning_rate": 7.27184527767838e-07,
      "loss": 0.9669,
      "num_input_tokens_seen": 72516296,
      "step": 124980
    },
    {
      "epoch": 18.615579386356867,
      "grad_norm": 3.765625,
      "learning_rate": 7.264067105126199e-07,
      "loss": 0.9709,
      "num_input_tokens_seen": 72519048,
      "step": 124985
    },
    {
      "epoch": 18.616324098897824,
      "grad_norm": 11.8125,
      "learning_rate": 7.256293033350847e-07,
      "loss": 0.9615,
      "num_input_tokens_seen": 72522088,
      "step": 124990
    },
    {
      "epoch": 18.617068811438784,
      "grad_norm": 11.625,
      "learning_rate": 7.248523062483748e-07,
      "loss": 1.1873,
      "num_input_tokens_seen": 72525096,
      "step": 124995
    },
    {
      "epoch": 18.617813523979745,
      "grad_norm": 14.3125,
      "learning_rate": 7.2407571926561e-07,
      "loss": 1.4504,
      "num_input_tokens_seen": 72528008,
      "step": 125000
    },
    {
      "epoch": 18.618558236520702,
      "grad_norm": 9.625,
      "learning_rate": 7.232995423999162e-07,
      "loss": 0.949,
      "num_input_tokens_seen": 72531048,
      "step": 125005
    },
    {
      "epoch": 18.619302949061662,
      "grad_norm": 5.34375,
      "learning_rate": 7.225237756644021e-07,
      "loss": 1.0375,
      "num_input_tokens_seen": 72533800,
      "step": 125010
    },
    {
      "epoch": 18.620047661602623,
      "grad_norm": 6.4375,
      "learning_rate": 7.217484190721712e-07,
      "loss": 1.1089,
      "num_input_tokens_seen": 72536552,
      "step": 125015
    },
    {
      "epoch": 18.62079237414358,
      "grad_norm": 21.625,
      "learning_rate": 7.209734726363299e-07,
      "loss": 1.3356,
      "num_input_tokens_seen": 72539112,
      "step": 125020
    },
    {
      "epoch": 18.62153708668454,
      "grad_norm": 7.96875,
      "learning_rate": 7.201989363699618e-07,
      "loss": 1.2762,
      "num_input_tokens_seen": 72542312,
      "step": 125025
    },
    {
      "epoch": 18.622281799225497,
      "grad_norm": 5.90625,
      "learning_rate": 7.194248102861594e-07,
      "loss": 1.1356,
      "num_input_tokens_seen": 72545256,
      "step": 125030
    },
    {
      "epoch": 18.623026511766458,
      "grad_norm": 10.9375,
      "learning_rate": 7.186510943979957e-07,
      "loss": 1.0628,
      "num_input_tokens_seen": 72548296,
      "step": 125035
    },
    {
      "epoch": 18.62377122430742,
      "grad_norm": 8.4375,
      "learning_rate": 7.178777887185434e-07,
      "loss": 1.0132,
      "num_input_tokens_seen": 72551080,
      "step": 125040
    },
    {
      "epoch": 18.624515936848375,
      "grad_norm": 13.4375,
      "learning_rate": 7.17104893260867e-07,
      "loss": 1.1811,
      "num_input_tokens_seen": 72554120,
      "step": 125045
    },
    {
      "epoch": 18.625260649389336,
      "grad_norm": 6.0625,
      "learning_rate": 7.16332408038023e-07,
      "loss": 0.8675,
      "num_input_tokens_seen": 72556744,
      "step": 125050
    },
    {
      "epoch": 18.626005361930297,
      "grad_norm": 16.875,
      "learning_rate": 7.155603330630617e-07,
      "loss": 1.0923,
      "num_input_tokens_seen": 72559624,
      "step": 125055
    },
    {
      "epoch": 18.626750074471254,
      "grad_norm": 8.0625,
      "learning_rate": 7.147886683490256e-07,
      "loss": 1.1183,
      "num_input_tokens_seen": 72562440,
      "step": 125060
    },
    {
      "epoch": 18.627494787012214,
      "grad_norm": 9.8125,
      "learning_rate": 7.140174139089545e-07,
      "loss": 0.7144,
      "num_input_tokens_seen": 72564968,
      "step": 125065
    },
    {
      "epoch": 18.62823949955317,
      "grad_norm": 11.5,
      "learning_rate": 7.132465697558737e-07,
      "loss": 0.9954,
      "num_input_tokens_seen": 72567592,
      "step": 125070
    },
    {
      "epoch": 18.62898421209413,
      "grad_norm": 15.3125,
      "learning_rate": 7.124761359028121e-07,
      "loss": 1.0783,
      "num_input_tokens_seen": 72570728,
      "step": 125075
    },
    {
      "epoch": 18.629728924635092,
      "grad_norm": 5.40625,
      "learning_rate": 7.117061123627783e-07,
      "loss": 1.0191,
      "num_input_tokens_seen": 72573672,
      "step": 125080
    },
    {
      "epoch": 18.63047363717605,
      "grad_norm": 11.0625,
      "learning_rate": 7.109364991487872e-07,
      "loss": 1.3059,
      "num_input_tokens_seen": 72576392,
      "step": 125085
    },
    {
      "epoch": 18.63121834971701,
      "grad_norm": 11.6875,
      "learning_rate": 7.101672962738365e-07,
      "loss": 1.2219,
      "num_input_tokens_seen": 72579464,
      "step": 125090
    },
    {
      "epoch": 18.631963062257967,
      "grad_norm": 13.1875,
      "learning_rate": 7.093985037509188e-07,
      "loss": 1.2656,
      "num_input_tokens_seen": 72582312,
      "step": 125095
    },
    {
      "epoch": 18.632707774798927,
      "grad_norm": 13.375,
      "learning_rate": 7.086301215930291e-07,
      "loss": 1.1087,
      "num_input_tokens_seen": 72585000,
      "step": 125100
    },
    {
      "epoch": 18.633452487339888,
      "grad_norm": 11.0625,
      "learning_rate": 7.078621498131461e-07,
      "loss": 0.86,
      "num_input_tokens_seen": 72587848,
      "step": 125105
    },
    {
      "epoch": 18.634197199880845,
      "grad_norm": 5.75,
      "learning_rate": 7.070945884242397e-07,
      "loss": 0.8814,
      "num_input_tokens_seen": 72590760,
      "step": 125110
    },
    {
      "epoch": 18.634941912421805,
      "grad_norm": 8.25,
      "learning_rate": 7.063274374392803e-07,
      "loss": 0.8433,
      "num_input_tokens_seen": 72593512,
      "step": 125115
    },
    {
      "epoch": 18.635686624962766,
      "grad_norm": 13.0625,
      "learning_rate": 7.055606968712297e-07,
      "loss": 1.1221,
      "num_input_tokens_seen": 72596712,
      "step": 125120
    },
    {
      "epoch": 18.636431337503723,
      "grad_norm": 10.8125,
      "learning_rate": 7.047943667330386e-07,
      "loss": 1.2,
      "num_input_tokens_seen": 72599528,
      "step": 125125
    },
    {
      "epoch": 18.637176050044683,
      "grad_norm": 8.1875,
      "learning_rate": 7.040284470376523e-07,
      "loss": 1.3087,
      "num_input_tokens_seen": 72602376,
      "step": 125130
    },
    {
      "epoch": 18.63792076258564,
      "grad_norm": 13.125,
      "learning_rate": 7.032629377980133e-07,
      "loss": 1.0447,
      "num_input_tokens_seen": 72605224,
      "step": 125135
    },
    {
      "epoch": 18.6386654751266,
      "grad_norm": 5.5625,
      "learning_rate": 7.024978390270526e-07,
      "loss": 0.6331,
      "num_input_tokens_seen": 72607976,
      "step": 125140
    },
    {
      "epoch": 18.63941018766756,
      "grad_norm": 9.125,
      "learning_rate": 7.017331507376962e-07,
      "loss": 0.9908,
      "num_input_tokens_seen": 72610824,
      "step": 125145
    },
    {
      "epoch": 18.64015490020852,
      "grad_norm": 8.5625,
      "learning_rate": 7.009688729428615e-07,
      "loss": 0.9035,
      "num_input_tokens_seen": 72613640,
      "step": 125150
    },
    {
      "epoch": 18.64089961274948,
      "grad_norm": 7.09375,
      "learning_rate": 7.002050056554632e-07,
      "loss": 0.875,
      "num_input_tokens_seen": 72616808,
      "step": 125155
    },
    {
      "epoch": 18.64164432529044,
      "grad_norm": 11.0625,
      "learning_rate": 6.994415488884021e-07,
      "loss": 1.1331,
      "num_input_tokens_seen": 72619592,
      "step": 125160
    },
    {
      "epoch": 18.642389037831396,
      "grad_norm": 7.4375,
      "learning_rate": 6.986785026545789e-07,
      "loss": 0.9128,
      "num_input_tokens_seen": 72622472,
      "step": 125165
    },
    {
      "epoch": 18.643133750372357,
      "grad_norm": 15.4375,
      "learning_rate": 6.979158669668862e-07,
      "loss": 1.0685,
      "num_input_tokens_seen": 72625352,
      "step": 125170
    },
    {
      "epoch": 18.643878462913314,
      "grad_norm": 14.875,
      "learning_rate": 6.971536418382052e-07,
      "loss": 0.9839,
      "num_input_tokens_seen": 72628232,
      "step": 125175
    },
    {
      "epoch": 18.644623175454274,
      "grad_norm": 12.375,
      "learning_rate": 6.963918272814119e-07,
      "loss": 0.9218,
      "num_input_tokens_seen": 72631432,
      "step": 125180
    },
    {
      "epoch": 18.645367887995235,
      "grad_norm": 11.5,
      "learning_rate": 6.956304233093736e-07,
      "loss": 0.9429,
      "num_input_tokens_seen": 72634280,
      "step": 125185
    },
    {
      "epoch": 18.646112600536192,
      "grad_norm": 16.125,
      "learning_rate": 6.948694299349634e-07,
      "loss": 1.0963,
      "num_input_tokens_seen": 72637160,
      "step": 125190
    },
    {
      "epoch": 18.646857313077152,
      "grad_norm": 8.9375,
      "learning_rate": 6.941088471710266e-07,
      "loss": 1.0387,
      "num_input_tokens_seen": 72640136,
      "step": 125195
    },
    {
      "epoch": 18.647602025618113,
      "grad_norm": 14.625,
      "learning_rate": 6.933486750304197e-07,
      "loss": 0.9933,
      "num_input_tokens_seen": 72642984,
      "step": 125200
    },
    {
      "epoch": 18.64834673815907,
      "grad_norm": 11.0,
      "learning_rate": 6.92588913525985e-07,
      "loss": 1.2385,
      "num_input_tokens_seen": 72645928,
      "step": 125205
    },
    {
      "epoch": 18.64909145070003,
      "grad_norm": 6.21875,
      "learning_rate": 6.918295626705512e-07,
      "loss": 1.0375,
      "num_input_tokens_seen": 72648872,
      "step": 125210
    },
    {
      "epoch": 18.649836163240987,
      "grad_norm": 9.9375,
      "learning_rate": 6.910706224769553e-07,
      "loss": 0.7651,
      "num_input_tokens_seen": 72651560,
      "step": 125215
    },
    {
      "epoch": 18.650580875781948,
      "grad_norm": 10.0625,
      "learning_rate": 6.903120929580092e-07,
      "loss": 1.4613,
      "num_input_tokens_seen": 72654440,
      "step": 125220
    },
    {
      "epoch": 18.65132558832291,
      "grad_norm": 7.125,
      "learning_rate": 6.895539741265389e-07,
      "loss": 1.0638,
      "num_input_tokens_seen": 72657064,
      "step": 125225
    },
    {
      "epoch": 18.652070300863866,
      "grad_norm": 8.6875,
      "learning_rate": 6.887962659953423e-07,
      "loss": 1.2073,
      "num_input_tokens_seen": 72660136,
      "step": 125230
    },
    {
      "epoch": 18.652815013404826,
      "grad_norm": 4.8125,
      "learning_rate": 6.88038968577226e-07,
      "loss": 0.9484,
      "num_input_tokens_seen": 72662824,
      "step": 125235
    },
    {
      "epoch": 18.653559725945787,
      "grad_norm": 13.8125,
      "learning_rate": 6.872820818849823e-07,
      "loss": 1.3988,
      "num_input_tokens_seen": 72665896,
      "step": 125240
    },
    {
      "epoch": 18.654304438486744,
      "grad_norm": 8.25,
      "learning_rate": 6.865256059313985e-07,
      "loss": 1.0275,
      "num_input_tokens_seen": 72668680,
      "step": 125245
    },
    {
      "epoch": 18.655049151027704,
      "grad_norm": 8.625,
      "learning_rate": 6.857695407292503e-07,
      "loss": 1.0307,
      "num_input_tokens_seen": 72671336,
      "step": 125250
    },
    {
      "epoch": 18.65579386356866,
      "grad_norm": 12.25,
      "learning_rate": 6.850138862913136e-07,
      "loss": 0.9831,
      "num_input_tokens_seen": 72674088,
      "step": 125255
    },
    {
      "epoch": 18.65653857610962,
      "grad_norm": 9.3125,
      "learning_rate": 6.842586426303588e-07,
      "loss": 1.182,
      "num_input_tokens_seen": 72677128,
      "step": 125260
    },
    {
      "epoch": 18.657283288650582,
      "grad_norm": 16.75,
      "learning_rate": 6.835038097591367e-07,
      "loss": 1.1465,
      "num_input_tokens_seen": 72680008,
      "step": 125265
    },
    {
      "epoch": 18.65802800119154,
      "grad_norm": 26.375,
      "learning_rate": 6.82749387690404e-07,
      "loss": 1.2782,
      "num_input_tokens_seen": 72682696,
      "step": 125270
    },
    {
      "epoch": 18.6587727137325,
      "grad_norm": 11.0625,
      "learning_rate": 6.819953764369058e-07,
      "loss": 0.8795,
      "num_input_tokens_seen": 72685672,
      "step": 125275
    },
    {
      "epoch": 18.65951742627346,
      "grad_norm": 11.375,
      "learning_rate": 6.812417760113821e-07,
      "loss": 1.0904,
      "num_input_tokens_seen": 72688776,
      "step": 125280
    },
    {
      "epoch": 18.660262138814417,
      "grad_norm": 10.875,
      "learning_rate": 6.804885864265587e-07,
      "loss": 1.1232,
      "num_input_tokens_seen": 72691848,
      "step": 125285
    },
    {
      "epoch": 18.661006851355378,
      "grad_norm": 10.9375,
      "learning_rate": 6.79735807695167e-07,
      "loss": 1.1402,
      "num_input_tokens_seen": 72694760,
      "step": 125290
    },
    {
      "epoch": 18.661751563896335,
      "grad_norm": 7.09375,
      "learning_rate": 6.789834398299194e-07,
      "loss": 1.0335,
      "num_input_tokens_seen": 72697384,
      "step": 125295
    },
    {
      "epoch": 18.662496276437295,
      "grad_norm": 5.875,
      "learning_rate": 6.782314828435249e-07,
      "loss": 1.2609,
      "num_input_tokens_seen": 72700328,
      "step": 125300
    },
    {
      "epoch": 18.663240988978256,
      "grad_norm": 14.375,
      "learning_rate": 6.774799367486956e-07,
      "loss": 1.1722,
      "num_input_tokens_seen": 72703240,
      "step": 125305
    },
    {
      "epoch": 18.663985701519213,
      "grad_norm": 17.75,
      "learning_rate": 6.767288015581186e-07,
      "loss": 1.1738,
      "num_input_tokens_seen": 72706440,
      "step": 125310
    },
    {
      "epoch": 18.664730414060173,
      "grad_norm": 13.6875,
      "learning_rate": 6.759780772844892e-07,
      "loss": 1.2758,
      "num_input_tokens_seen": 72709256,
      "step": 125315
    },
    {
      "epoch": 18.66547512660113,
      "grad_norm": 5.8125,
      "learning_rate": 6.752277639404863e-07,
      "loss": 1.3297,
      "num_input_tokens_seen": 72711976,
      "step": 125320
    },
    {
      "epoch": 18.66621983914209,
      "grad_norm": 13.5625,
      "learning_rate": 6.744778615387914e-07,
      "loss": 0.8404,
      "num_input_tokens_seen": 72714696,
      "step": 125325
    },
    {
      "epoch": 18.66696455168305,
      "grad_norm": 6.75,
      "learning_rate": 6.737283700920666e-07,
      "loss": 0.7647,
      "num_input_tokens_seen": 72717608,
      "step": 125330
    },
    {
      "epoch": 18.66770926422401,
      "grad_norm": 8.5,
      "learning_rate": 6.729792896129767e-07,
      "loss": 1.4322,
      "num_input_tokens_seen": 72720360,
      "step": 125335
    },
    {
      "epoch": 18.66845397676497,
      "grad_norm": 12.25,
      "learning_rate": 6.722306201141781e-07,
      "loss": 1.0827,
      "num_input_tokens_seen": 72723080,
      "step": 125340
    },
    {
      "epoch": 18.66919868930593,
      "grad_norm": 13.3125,
      "learning_rate": 6.714823616083165e-07,
      "loss": 1.2474,
      "num_input_tokens_seen": 72725928,
      "step": 125345
    },
    {
      "epoch": 18.669943401846886,
      "grad_norm": 12.875,
      "learning_rate": 6.707345141080345e-07,
      "loss": 1.2794,
      "num_input_tokens_seen": 72728648,
      "step": 125350
    },
    {
      "epoch": 18.670688114387847,
      "grad_norm": 10.6875,
      "learning_rate": 6.699870776259637e-07,
      "loss": 0.9567,
      "num_input_tokens_seen": 72731656,
      "step": 125355
    },
    {
      "epoch": 18.671432826928804,
      "grad_norm": 7.75,
      "learning_rate": 6.692400521747355e-07,
      "loss": 1.1174,
      "num_input_tokens_seen": 72734632,
      "step": 125360
    },
    {
      "epoch": 18.672177539469764,
      "grad_norm": 13.1875,
      "learning_rate": 6.684934377669705e-07,
      "loss": 0.7444,
      "num_input_tokens_seen": 72737480,
      "step": 125365
    },
    {
      "epoch": 18.672922252010725,
      "grad_norm": 12.6875,
      "learning_rate": 6.67747234415278e-07,
      "loss": 0.7331,
      "num_input_tokens_seen": 72740584,
      "step": 125370
    },
    {
      "epoch": 18.673666964551682,
      "grad_norm": 5.21875,
      "learning_rate": 6.670014421322618e-07,
      "loss": 1.201,
      "num_input_tokens_seen": 72743592,
      "step": 125375
    },
    {
      "epoch": 18.674411677092642,
      "grad_norm": 9.5,
      "learning_rate": 6.662560609305285e-07,
      "loss": 1.393,
      "num_input_tokens_seen": 72746408,
      "step": 125380
    },
    {
      "epoch": 18.675156389633603,
      "grad_norm": 10.5625,
      "learning_rate": 6.655110908226681e-07,
      "loss": 1.3664,
      "num_input_tokens_seen": 72749192,
      "step": 125385
    },
    {
      "epoch": 18.67590110217456,
      "grad_norm": 4.625,
      "learning_rate": 6.647665318212621e-07,
      "loss": 0.7526,
      "num_input_tokens_seen": 72751944,
      "step": 125390
    },
    {
      "epoch": 18.67664581471552,
      "grad_norm": 10.1875,
      "learning_rate": 6.640223839388948e-07,
      "loss": 0.9977,
      "num_input_tokens_seen": 72754824,
      "step": 125395
    },
    {
      "epoch": 18.677390527256478,
      "grad_norm": 8.4375,
      "learning_rate": 6.632786471881342e-07,
      "loss": 1.0384,
      "num_input_tokens_seen": 72757736,
      "step": 125400
    },
    {
      "epoch": 18.678135239797438,
      "grad_norm": 9.375,
      "learning_rate": 6.625353215815478e-07,
      "loss": 1.5764,
      "num_input_tokens_seen": 72760520,
      "step": 125405
    },
    {
      "epoch": 18.6788799523384,
      "grad_norm": 13.125,
      "learning_rate": 6.617924071316894e-07,
      "loss": 1.3697,
      "num_input_tokens_seen": 72763432,
      "step": 125410
    },
    {
      "epoch": 18.679624664879356,
      "grad_norm": 7.875,
      "learning_rate": 6.610499038511131e-07,
      "loss": 0.9835,
      "num_input_tokens_seen": 72766344,
      "step": 125415
    },
    {
      "epoch": 18.680369377420316,
      "grad_norm": 13.25,
      "learning_rate": 6.603078117523615e-07,
      "loss": 0.8846,
      "num_input_tokens_seen": 72769096,
      "step": 125420
    },
    {
      "epoch": 18.681114089961277,
      "grad_norm": 13.5,
      "learning_rate": 6.595661308479717e-07,
      "loss": 0.9748,
      "num_input_tokens_seen": 72771784,
      "step": 125425
    },
    {
      "epoch": 18.681858802502234,
      "grad_norm": 7.375,
      "learning_rate": 6.588248611504755e-07,
      "loss": 0.8554,
      "num_input_tokens_seen": 72774632,
      "step": 125430
    },
    {
      "epoch": 18.682603515043194,
      "grad_norm": 15.0625,
      "learning_rate": 6.580840026723934e-07,
      "loss": 1.4647,
      "num_input_tokens_seen": 72777480,
      "step": 125435
    },
    {
      "epoch": 18.68334822758415,
      "grad_norm": 13.5,
      "learning_rate": 6.573435554262403e-07,
      "loss": 1.2871,
      "num_input_tokens_seen": 72780456,
      "step": 125440
    },
    {
      "epoch": 18.68409294012511,
      "grad_norm": 15.125,
      "learning_rate": 6.566035194245257e-07,
      "loss": 1.2268,
      "num_input_tokens_seen": 72783592,
      "step": 125445
    },
    {
      "epoch": 18.684837652666072,
      "grad_norm": 10.25,
      "learning_rate": 6.558638946797563e-07,
      "loss": 1.0666,
      "num_input_tokens_seen": 72786376,
      "step": 125450
    },
    {
      "epoch": 18.68558236520703,
      "grad_norm": 10.0625,
      "learning_rate": 6.551246812044248e-07,
      "loss": 1.3219,
      "num_input_tokens_seen": 72789512,
      "step": 125455
    },
    {
      "epoch": 18.68632707774799,
      "grad_norm": 13.3125,
      "learning_rate": 6.543858790110158e-07,
      "loss": 1.0163,
      "num_input_tokens_seen": 72792456,
      "step": 125460
    },
    {
      "epoch": 18.687071790288947,
      "grad_norm": 11.8125,
      "learning_rate": 6.536474881120164e-07,
      "loss": 1.0886,
      "num_input_tokens_seen": 72795368,
      "step": 125465
    },
    {
      "epoch": 18.687816502829907,
      "grad_norm": 9.125,
      "learning_rate": 6.529095085198944e-07,
      "loss": 1.3533,
      "num_input_tokens_seen": 72798408,
      "step": 125470
    },
    {
      "epoch": 18.688561215370868,
      "grad_norm": 7.53125,
      "learning_rate": 6.521719402471233e-07,
      "loss": 0.7341,
      "num_input_tokens_seen": 72801128,
      "step": 125475
    },
    {
      "epoch": 18.689305927911825,
      "grad_norm": 12.0,
      "learning_rate": 6.514347833061596e-07,
      "loss": 0.9924,
      "num_input_tokens_seen": 72803944,
      "step": 125480
    },
    {
      "epoch": 18.690050640452785,
      "grad_norm": 7.9375,
      "learning_rate": 6.506980377094601e-07,
      "loss": 0.8606,
      "num_input_tokens_seen": 72806760,
      "step": 125485
    },
    {
      "epoch": 18.690795352993746,
      "grad_norm": 6.6875,
      "learning_rate": 6.499617034694705e-07,
      "loss": 1.4972,
      "num_input_tokens_seen": 72809544,
      "step": 125490
    },
    {
      "epoch": 18.691540065534703,
      "grad_norm": 9.375,
      "learning_rate": 6.492257805986279e-07,
      "loss": 1.0241,
      "num_input_tokens_seen": 72812392,
      "step": 125495
    },
    {
      "epoch": 18.692284778075663,
      "grad_norm": 5.71875,
      "learning_rate": 6.48490269109367e-07,
      "loss": 0.9933,
      "num_input_tokens_seen": 72815112,
      "step": 125500
    },
    {
      "epoch": 18.69302949061662,
      "grad_norm": 11.9375,
      "learning_rate": 6.477551690141165e-07,
      "loss": 0.8502,
      "num_input_tokens_seen": 72818056,
      "step": 125505
    },
    {
      "epoch": 18.69377420315758,
      "grad_norm": 7.0,
      "learning_rate": 6.470204803252888e-07,
      "loss": 1.067,
      "num_input_tokens_seen": 72820968,
      "step": 125510
    },
    {
      "epoch": 18.69451891569854,
      "grad_norm": 7.375,
      "learning_rate": 6.462862030552991e-07,
      "loss": 1.044,
      "num_input_tokens_seen": 72823816,
      "step": 125515
    },
    {
      "epoch": 18.6952636282395,
      "grad_norm": 8.5625,
      "learning_rate": 6.455523372165512e-07,
      "loss": 1.0955,
      "num_input_tokens_seen": 72826696,
      "step": 125520
    },
    {
      "epoch": 18.69600834078046,
      "grad_norm": 11.5625,
      "learning_rate": 6.448188828214435e-07,
      "loss": 1.3787,
      "num_input_tokens_seen": 72829544,
      "step": 125525
    },
    {
      "epoch": 18.69675305332142,
      "grad_norm": 10.0625,
      "learning_rate": 6.44085839882369e-07,
      "loss": 1.5111,
      "num_input_tokens_seen": 72832360,
      "step": 125530
    },
    {
      "epoch": 18.697497765862376,
      "grad_norm": 6.03125,
      "learning_rate": 6.433532084117122e-07,
      "loss": 0.7523,
      "num_input_tokens_seen": 72835176,
      "step": 125535
    },
    {
      "epoch": 18.698242478403337,
      "grad_norm": 7.875,
      "learning_rate": 6.426209884218437e-07,
      "loss": 1.0097,
      "num_input_tokens_seen": 72838088,
      "step": 125540
    },
    {
      "epoch": 18.698987190944294,
      "grad_norm": 8.4375,
      "learning_rate": 6.418891799251397e-07,
      "loss": 0.9991,
      "num_input_tokens_seen": 72840872,
      "step": 125545
    },
    {
      "epoch": 18.699731903485254,
      "grad_norm": 10.5625,
      "learning_rate": 6.411577829339599e-07,
      "loss": 0.9283,
      "num_input_tokens_seen": 72843656,
      "step": 125550
    },
    {
      "epoch": 18.700476616026215,
      "grad_norm": 19.125,
      "learning_rate": 6.404267974606637e-07,
      "loss": 0.9539,
      "num_input_tokens_seen": 72846664,
      "step": 125555
    },
    {
      "epoch": 18.701221328567172,
      "grad_norm": 19.0,
      "learning_rate": 6.396962235175968e-07,
      "loss": 0.936,
      "num_input_tokens_seen": 72849512,
      "step": 125560
    },
    {
      "epoch": 18.701966041108133,
      "grad_norm": 6.71875,
      "learning_rate": 6.38966061117105e-07,
      "loss": 0.9303,
      "num_input_tokens_seen": 72852488,
      "step": 125565
    },
    {
      "epoch": 18.702710753649093,
      "grad_norm": 16.375,
      "learning_rate": 6.382363102715255e-07,
      "loss": 0.9143,
      "num_input_tokens_seen": 72855080,
      "step": 125570
    },
    {
      "epoch": 18.70345546619005,
      "grad_norm": 13.4375,
      "learning_rate": 6.375069709931792e-07,
      "loss": 1.1132,
      "num_input_tokens_seen": 72857960,
      "step": 125575
    },
    {
      "epoch": 18.70420017873101,
      "grad_norm": 12.5,
      "learning_rate": 6.367780432943948e-07,
      "loss": 1.0506,
      "num_input_tokens_seen": 72861000,
      "step": 125580
    },
    {
      "epoch": 18.704944891271968,
      "grad_norm": 6.3125,
      "learning_rate": 6.360495271874794e-07,
      "loss": 0.872,
      "num_input_tokens_seen": 72864104,
      "step": 125585
    },
    {
      "epoch": 18.705689603812928,
      "grad_norm": 11.875,
      "learning_rate": 6.353214226847482e-07,
      "loss": 1.1092,
      "num_input_tokens_seen": 72866888,
      "step": 125590
    },
    {
      "epoch": 18.70643431635389,
      "grad_norm": 14.1875,
      "learning_rate": 6.345937297984966e-07,
      "loss": 1.1526,
      "num_input_tokens_seen": 72869704,
      "step": 125595
    },
    {
      "epoch": 18.707179028894846,
      "grad_norm": 9.0,
      "learning_rate": 6.338664485410206e-07,
      "loss": 1.2008,
      "num_input_tokens_seen": 72872520,
      "step": 125600
    },
    {
      "epoch": 18.707923741435806,
      "grad_norm": 11.125,
      "learning_rate": 6.331395789246048e-07,
      "loss": 1.1765,
      "num_input_tokens_seen": 72875464,
      "step": 125605
    },
    {
      "epoch": 18.708668453976763,
      "grad_norm": 7.875,
      "learning_rate": 6.324131209615336e-07,
      "loss": 1.0237,
      "num_input_tokens_seen": 72878248,
      "step": 125610
    },
    {
      "epoch": 18.709413166517724,
      "grad_norm": 22.75,
      "learning_rate": 6.316870746640751e-07,
      "loss": 1.2788,
      "num_input_tokens_seen": 72880968,
      "step": 125615
    },
    {
      "epoch": 18.710157879058684,
      "grad_norm": 25.125,
      "learning_rate": 6.309614400444946e-07,
      "loss": 1.1043,
      "num_input_tokens_seen": 72883624,
      "step": 125620
    },
    {
      "epoch": 18.71090259159964,
      "grad_norm": 14.625,
      "learning_rate": 6.302362171150572e-07,
      "loss": 1.1891,
      "num_input_tokens_seen": 72886472,
      "step": 125625
    },
    {
      "epoch": 18.7116473041406,
      "grad_norm": 8.8125,
      "learning_rate": 6.295114058880059e-07,
      "loss": 0.8468,
      "num_input_tokens_seen": 72889192,
      "step": 125630
    },
    {
      "epoch": 18.712392016681562,
      "grad_norm": 5.96875,
      "learning_rate": 6.287870063755946e-07,
      "loss": 0.8653,
      "num_input_tokens_seen": 72892072,
      "step": 125635
    },
    {
      "epoch": 18.71313672922252,
      "grad_norm": 10.5625,
      "learning_rate": 6.280630185900555e-07,
      "loss": 1.2518,
      "num_input_tokens_seen": 72895080,
      "step": 125640
    },
    {
      "epoch": 18.71388144176348,
      "grad_norm": 15.5,
      "learning_rate": 6.273394425436202e-07,
      "loss": 1.2337,
      "num_input_tokens_seen": 72897928,
      "step": 125645
    },
    {
      "epoch": 18.714626154304437,
      "grad_norm": 8.0625,
      "learning_rate": 6.26616278248518e-07,
      "loss": 0.9546,
      "num_input_tokens_seen": 72900904,
      "step": 125650
    },
    {
      "epoch": 18.715370866845397,
      "grad_norm": 12.5,
      "learning_rate": 6.258935257169557e-07,
      "loss": 1.0303,
      "num_input_tokens_seen": 72903528,
      "step": 125655
    },
    {
      "epoch": 18.716115579386358,
      "grad_norm": 8.75,
      "learning_rate": 6.251711849611513e-07,
      "loss": 1.1929,
      "num_input_tokens_seen": 72906536,
      "step": 125660
    },
    {
      "epoch": 18.716860291927315,
      "grad_norm": 10.625,
      "learning_rate": 6.244492559933063e-07,
      "loss": 0.9596,
      "num_input_tokens_seen": 72909576,
      "step": 125665
    },
    {
      "epoch": 18.717605004468275,
      "grad_norm": 5.53125,
      "learning_rate": 6.237277388256191e-07,
      "loss": 0.9546,
      "num_input_tokens_seen": 72912456,
      "step": 125670
    },
    {
      "epoch": 18.718349717009236,
      "grad_norm": 14.0,
      "learning_rate": 6.230066334702744e-07,
      "loss": 1.1881,
      "num_input_tokens_seen": 72915144,
      "step": 125675
    },
    {
      "epoch": 18.719094429550193,
      "grad_norm": 16.25,
      "learning_rate": 6.22285939939457e-07,
      "loss": 1.3331,
      "num_input_tokens_seen": 72917832,
      "step": 125680
    },
    {
      "epoch": 18.719839142091153,
      "grad_norm": 14.25,
      "learning_rate": 6.215656582453433e-07,
      "loss": 1.2633,
      "num_input_tokens_seen": 72920808,
      "step": 125685
    },
    {
      "epoch": 18.72058385463211,
      "grad_norm": 5.9375,
      "learning_rate": 6.208457884001012e-07,
      "loss": 1.1987,
      "num_input_tokens_seen": 72923720,
      "step": 125690
    },
    {
      "epoch": 18.72132856717307,
      "grad_norm": 11.4375,
      "learning_rate": 6.201263304158905e-07,
      "loss": 1.401,
      "num_input_tokens_seen": 72926600,
      "step": 125695
    },
    {
      "epoch": 18.72207327971403,
      "grad_norm": 14.0,
      "learning_rate": 6.194072843048681e-07,
      "loss": 1.4337,
      "num_input_tokens_seen": 72929480,
      "step": 125700
    },
    {
      "epoch": 18.72281799225499,
      "grad_norm": 16.125,
      "learning_rate": 6.1868865007918e-07,
      "loss": 0.8026,
      "num_input_tokens_seen": 72932584,
      "step": 125705
    },
    {
      "epoch": 18.72356270479595,
      "grad_norm": 9.0625,
      "learning_rate": 6.179704277509662e-07,
      "loss": 0.9235,
      "num_input_tokens_seen": 72935368,
      "step": 125710
    },
    {
      "epoch": 18.72430741733691,
      "grad_norm": 11.75,
      "learning_rate": 6.172526173323617e-07,
      "loss": 0.8469,
      "num_input_tokens_seen": 72938312,
      "step": 125715
    },
    {
      "epoch": 18.725052129877866,
      "grad_norm": 12.25,
      "learning_rate": 6.165352188354928e-07,
      "loss": 1.3295,
      "num_input_tokens_seen": 72941512,
      "step": 125720
    },
    {
      "epoch": 18.725796842418827,
      "grad_norm": 9.5625,
      "learning_rate": 6.158182322724804e-07,
      "loss": 1.0598,
      "num_input_tokens_seen": 72944264,
      "step": 125725
    },
    {
      "epoch": 18.726541554959784,
      "grad_norm": 7.21875,
      "learning_rate": 6.151016576554341e-07,
      "loss": 1.055,
      "num_input_tokens_seen": 72947464,
      "step": 125730
    },
    {
      "epoch": 18.727286267500745,
      "grad_norm": 12.125,
      "learning_rate": 6.143854949964611e-07,
      "loss": 1.082,
      "num_input_tokens_seen": 72950248,
      "step": 125735
    },
    {
      "epoch": 18.728030980041705,
      "grad_norm": 7.21875,
      "learning_rate": 6.136697443076628e-07,
      "loss": 1.1424,
      "num_input_tokens_seen": 72953096,
      "step": 125740
    },
    {
      "epoch": 18.728775692582662,
      "grad_norm": 15.1875,
      "learning_rate": 6.129544056011266e-07,
      "loss": 1.1134,
      "num_input_tokens_seen": 72955784,
      "step": 125745
    },
    {
      "epoch": 18.729520405123623,
      "grad_norm": 14.1875,
      "learning_rate": 6.122394788889402e-07,
      "loss": 1.1548,
      "num_input_tokens_seen": 72958600,
      "step": 125750
    },
    {
      "epoch": 18.730265117664583,
      "grad_norm": 11.6875,
      "learning_rate": 6.115249641831828e-07,
      "loss": 1.0801,
      "num_input_tokens_seen": 72961544,
      "step": 125755
    },
    {
      "epoch": 18.73100983020554,
      "grad_norm": 11.0,
      "learning_rate": 6.108108614959224e-07,
      "loss": 1.1359,
      "num_input_tokens_seen": 72964552,
      "step": 125760
    },
    {
      "epoch": 18.7317545427465,
      "grad_norm": 9.9375,
      "learning_rate": 6.100971708392272e-07,
      "loss": 1.0053,
      "num_input_tokens_seen": 72967688,
      "step": 125765
    },
    {
      "epoch": 18.732499255287458,
      "grad_norm": 21.5,
      "learning_rate": 6.093838922251488e-07,
      "loss": 1.1077,
      "num_input_tokens_seen": 72970568,
      "step": 125770
    },
    {
      "epoch": 18.733243967828418,
      "grad_norm": 16.125,
      "learning_rate": 6.086710256657413e-07,
      "loss": 1.1659,
      "num_input_tokens_seen": 72973448,
      "step": 125775
    },
    {
      "epoch": 18.73398868036938,
      "grad_norm": 15.75,
      "learning_rate": 6.079585711730451e-07,
      "loss": 0.8202,
      "num_input_tokens_seen": 72976680,
      "step": 125780
    },
    {
      "epoch": 18.734733392910336,
      "grad_norm": 8.5,
      "learning_rate": 6.072465287591005e-07,
      "loss": 1.1062,
      "num_input_tokens_seen": 72979592,
      "step": 125785
    },
    {
      "epoch": 18.735478105451296,
      "grad_norm": 7.8125,
      "learning_rate": 6.065348984359314e-07,
      "loss": 1.0144,
      "num_input_tokens_seen": 72982632,
      "step": 125790
    },
    {
      "epoch": 18.736222817992257,
      "grad_norm": 6.125,
      "learning_rate": 6.058236802155643e-07,
      "loss": 0.8726,
      "num_input_tokens_seen": 72985512,
      "step": 125795
    },
    {
      "epoch": 18.736967530533214,
      "grad_norm": 9.875,
      "learning_rate": 6.051128741100115e-07,
      "loss": 1.1405,
      "num_input_tokens_seen": 72988136,
      "step": 125800
    },
    {
      "epoch": 18.737712243074174,
      "grad_norm": 8.5,
      "learning_rate": 6.044024801312831e-07,
      "loss": 0.9347,
      "num_input_tokens_seen": 72991048,
      "step": 125805
    },
    {
      "epoch": 18.73845695561513,
      "grad_norm": 12.1875,
      "learning_rate": 6.036924982913805e-07,
      "loss": 1.1069,
      "num_input_tokens_seen": 72993960,
      "step": 125810
    },
    {
      "epoch": 18.739201668156092,
      "grad_norm": 13.6875,
      "learning_rate": 6.029829286022998e-07,
      "loss": 1.3119,
      "num_input_tokens_seen": 72996840,
      "step": 125815
    },
    {
      "epoch": 18.739946380697052,
      "grad_norm": 9.75,
      "learning_rate": 6.022737710760256e-07,
      "loss": 1.0045,
      "num_input_tokens_seen": 73000040,
      "step": 125820
    },
    {
      "epoch": 18.74069109323801,
      "grad_norm": 15.5,
      "learning_rate": 6.015650257245348e-07,
      "loss": 1.2646,
      "num_input_tokens_seen": 73002856,
      "step": 125825
    },
    {
      "epoch": 18.74143580577897,
      "grad_norm": 8.1875,
      "learning_rate": 6.008566925598119e-07,
      "loss": 0.8706,
      "num_input_tokens_seen": 73005896,
      "step": 125830
    },
    {
      "epoch": 18.742180518319927,
      "grad_norm": 8.4375,
      "learning_rate": 6.001487715938142e-07,
      "loss": 0.9917,
      "num_input_tokens_seen": 73008744,
      "step": 125835
    },
    {
      "epoch": 18.742925230860887,
      "grad_norm": 8.3125,
      "learning_rate": 5.994412628385043e-07,
      "loss": 1.3433,
      "num_input_tokens_seen": 73011624,
      "step": 125840
    },
    {
      "epoch": 18.743669943401848,
      "grad_norm": 6.0,
      "learning_rate": 5.987341663058338e-07,
      "loss": 1.2275,
      "num_input_tokens_seen": 73014440,
      "step": 125845
    },
    {
      "epoch": 18.744414655942805,
      "grad_norm": 9.0,
      "learning_rate": 5.980274820077514e-07,
      "loss": 1.2518,
      "num_input_tokens_seen": 73017064,
      "step": 125850
    },
    {
      "epoch": 18.745159368483765,
      "grad_norm": 26.25,
      "learning_rate": 5.97321209956192e-07,
      "loss": 1.13,
      "num_input_tokens_seen": 73019784,
      "step": 125855
    },
    {
      "epoch": 18.745904081024726,
      "grad_norm": 12.375,
      "learning_rate": 5.966153501630877e-07,
      "loss": 1.3409,
      "num_input_tokens_seen": 73022504,
      "step": 125860
    },
    {
      "epoch": 18.746648793565683,
      "grad_norm": 4.4375,
      "learning_rate": 5.95909902640368e-07,
      "loss": 1.2164,
      "num_input_tokens_seen": 73025064,
      "step": 125865
    },
    {
      "epoch": 18.747393506106643,
      "grad_norm": 6.53125,
      "learning_rate": 5.952048673999427e-07,
      "loss": 0.8447,
      "num_input_tokens_seen": 73027944,
      "step": 125870
    },
    {
      "epoch": 18.7481382186476,
      "grad_norm": 15.75,
      "learning_rate": 5.945002444537329e-07,
      "loss": 1.2989,
      "num_input_tokens_seen": 73030920,
      "step": 125875
    },
    {
      "epoch": 18.74888293118856,
      "grad_norm": 7.15625,
      "learning_rate": 5.937960338136317e-07,
      "loss": 1.0389,
      "num_input_tokens_seen": 73034088,
      "step": 125880
    },
    {
      "epoch": 18.74962764372952,
      "grad_norm": 7.78125,
      "learning_rate": 5.930922354915436e-07,
      "loss": 1.0244,
      "num_input_tokens_seen": 73037000,
      "step": 125885
    },
    {
      "epoch": 18.75037235627048,
      "grad_norm": 15.5,
      "learning_rate": 5.923888494993562e-07,
      "loss": 1.1546,
      "num_input_tokens_seen": 73039944,
      "step": 125890
    },
    {
      "epoch": 18.75111706881144,
      "grad_norm": 10.75,
      "learning_rate": 5.916858758489519e-07,
      "loss": 1.2721,
      "num_input_tokens_seen": 73042664,
      "step": 125895
    },
    {
      "epoch": 18.7518617813524,
      "grad_norm": 8.0,
      "learning_rate": 5.90983314552207e-07,
      "loss": 0.8495,
      "num_input_tokens_seen": 73045736,
      "step": 125900
    },
    {
      "epoch": 18.752606493893357,
      "grad_norm": 10.8125,
      "learning_rate": 5.902811656209927e-07,
      "loss": 1.1745,
      "num_input_tokens_seen": 73049224,
      "step": 125905
    },
    {
      "epoch": 18.753351206434317,
      "grad_norm": 8.625,
      "learning_rate": 5.895794290671691e-07,
      "loss": 1.3204,
      "num_input_tokens_seen": 73052008,
      "step": 125910
    },
    {
      "epoch": 18.754095918975274,
      "grad_norm": 8.625,
      "learning_rate": 5.888781049025877e-07,
      "loss": 1.2305,
      "num_input_tokens_seen": 73054728,
      "step": 125915
    },
    {
      "epoch": 18.754840631516235,
      "grad_norm": 12.1875,
      "learning_rate": 5.881771931391028e-07,
      "loss": 0.9835,
      "num_input_tokens_seen": 73057224,
      "step": 125920
    },
    {
      "epoch": 18.755585344057195,
      "grad_norm": 7.625,
      "learning_rate": 5.874766937885523e-07,
      "loss": 1.1179,
      "num_input_tokens_seen": 73060232,
      "step": 125925
    },
    {
      "epoch": 18.756330056598152,
      "grad_norm": 18.5,
      "learning_rate": 5.867766068627739e-07,
      "loss": 0.9405,
      "num_input_tokens_seen": 73062952,
      "step": 125930
    },
    {
      "epoch": 18.757074769139113,
      "grad_norm": 10.25,
      "learning_rate": 5.860769323735887e-07,
      "loss": 0.8172,
      "num_input_tokens_seen": 73065640,
      "step": 125935
    },
    {
      "epoch": 18.757819481680073,
      "grad_norm": 10.6875,
      "learning_rate": 5.853776703328207e-07,
      "loss": 1.2122,
      "num_input_tokens_seen": 73068616,
      "step": 125940
    },
    {
      "epoch": 18.75856419422103,
      "grad_norm": 9.75,
      "learning_rate": 5.846788207522852e-07,
      "loss": 1.0596,
      "num_input_tokens_seen": 73071624,
      "step": 125945
    },
    {
      "epoch": 18.75930890676199,
      "grad_norm": 9.9375,
      "learning_rate": 5.83980383643784e-07,
      "loss": 1.124,
      "num_input_tokens_seen": 73074632,
      "step": 125950
    },
    {
      "epoch": 18.760053619302948,
      "grad_norm": 14.0625,
      "learning_rate": 5.832823590191216e-07,
      "loss": 1.169,
      "num_input_tokens_seen": 73077672,
      "step": 125955
    },
    {
      "epoch": 18.760798331843908,
      "grad_norm": 10.5625,
      "learning_rate": 5.825847468900858e-07,
      "loss": 1.0469,
      "num_input_tokens_seen": 73080520,
      "step": 125960
    },
    {
      "epoch": 18.76154304438487,
      "grad_norm": 10.5,
      "learning_rate": 5.81887547268467e-07,
      "loss": 0.9783,
      "num_input_tokens_seen": 73083240,
      "step": 125965
    },
    {
      "epoch": 18.762287756925826,
      "grad_norm": 11.875,
      "learning_rate": 5.811907601660393e-07,
      "loss": 1.1696,
      "num_input_tokens_seen": 73085992,
      "step": 125970
    },
    {
      "epoch": 18.763032469466786,
      "grad_norm": 16.25,
      "learning_rate": 5.804943855945738e-07,
      "loss": 1.3146,
      "num_input_tokens_seen": 73088744,
      "step": 125975
    },
    {
      "epoch": 18.763777182007743,
      "grad_norm": 11.125,
      "learning_rate": 5.797984235658388e-07,
      "loss": 1.0515,
      "num_input_tokens_seen": 73091304,
      "step": 125980
    },
    {
      "epoch": 18.764521894548704,
      "grad_norm": 17.125,
      "learning_rate": 5.791028740915888e-07,
      "loss": 1.3822,
      "num_input_tokens_seen": 73094216,
      "step": 125985
    },
    {
      "epoch": 18.765266607089664,
      "grad_norm": 13.3125,
      "learning_rate": 5.784077371835756e-07,
      "loss": 1.016,
      "num_input_tokens_seen": 73097288,
      "step": 125990
    },
    {
      "epoch": 18.76601131963062,
      "grad_norm": 10.5625,
      "learning_rate": 5.777130128535396e-07,
      "loss": 1.0836,
      "num_input_tokens_seen": 73100264,
      "step": 125995
    },
    {
      "epoch": 18.766756032171582,
      "grad_norm": 9.125,
      "learning_rate": 5.770187011132244e-07,
      "loss": 1.1862,
      "num_input_tokens_seen": 73103144,
      "step": 126000
    },
    {
      "epoch": 18.767500744712542,
      "grad_norm": 6.375,
      "learning_rate": 5.763248019743539e-07,
      "loss": 0.799,
      "num_input_tokens_seen": 73105896,
      "step": 126005
    },
    {
      "epoch": 18.7682454572535,
      "grad_norm": 9.6875,
      "learning_rate": 5.756313154486547e-07,
      "loss": 1.1359,
      "num_input_tokens_seen": 73108968,
      "step": 126010
    },
    {
      "epoch": 18.76899016979446,
      "grad_norm": 8.6875,
      "learning_rate": 5.7493824154784e-07,
      "loss": 0.9637,
      "num_input_tokens_seen": 73111784,
      "step": 126015
    },
    {
      "epoch": 18.769734882335417,
      "grad_norm": 11.875,
      "learning_rate": 5.742455802836166e-07,
      "loss": 1.2705,
      "num_input_tokens_seen": 73114536,
      "step": 126020
    },
    {
      "epoch": 18.770479594876377,
      "grad_norm": 13.3125,
      "learning_rate": 5.735533316676922e-07,
      "loss": 0.9362,
      "num_input_tokens_seen": 73117480,
      "step": 126025
    },
    {
      "epoch": 18.771224307417338,
      "grad_norm": 7.59375,
      "learning_rate": 5.728614957117573e-07,
      "loss": 0.9978,
      "num_input_tokens_seen": 73120712,
      "step": 126030
    },
    {
      "epoch": 18.771969019958295,
      "grad_norm": 9.0625,
      "learning_rate": 5.721700724274997e-07,
      "loss": 0.6275,
      "num_input_tokens_seen": 73123976,
      "step": 126035
    },
    {
      "epoch": 18.772713732499255,
      "grad_norm": 9.75,
      "learning_rate": 5.714790618266019e-07,
      "loss": 0.9709,
      "num_input_tokens_seen": 73127048,
      "step": 126040
    },
    {
      "epoch": 18.773458445040216,
      "grad_norm": 5.75,
      "learning_rate": 5.707884639207406e-07,
      "loss": 0.7226,
      "num_input_tokens_seen": 73129800,
      "step": 126045
    },
    {
      "epoch": 18.774203157581173,
      "grad_norm": 9.4375,
      "learning_rate": 5.700982787215759e-07,
      "loss": 1.1209,
      "num_input_tokens_seen": 73132744,
      "step": 126050
    },
    {
      "epoch": 18.774947870122134,
      "grad_norm": 8.4375,
      "learning_rate": 5.694085062407705e-07,
      "loss": 0.8139,
      "num_input_tokens_seen": 73136072,
      "step": 126055
    },
    {
      "epoch": 18.77569258266309,
      "grad_norm": 7.40625,
      "learning_rate": 5.687191464899821e-07,
      "loss": 0.8969,
      "num_input_tokens_seen": 73139080,
      "step": 126060
    },
    {
      "epoch": 18.77643729520405,
      "grad_norm": 12.875,
      "learning_rate": 5.680301994808485e-07,
      "loss": 0.901,
      "num_input_tokens_seen": 73141704,
      "step": 126065
    },
    {
      "epoch": 18.77718200774501,
      "grad_norm": 11.75,
      "learning_rate": 5.673416652250158e-07,
      "loss": 0.9512,
      "num_input_tokens_seen": 73144904,
      "step": 126070
    },
    {
      "epoch": 18.77792672028597,
      "grad_norm": 9.8125,
      "learning_rate": 5.666535437341108e-07,
      "loss": 1.1491,
      "num_input_tokens_seen": 73147656,
      "step": 126075
    },
    {
      "epoch": 18.77867143282693,
      "grad_norm": 16.0,
      "learning_rate": 5.659658350197661e-07,
      "loss": 1.2636,
      "num_input_tokens_seen": 73150472,
      "step": 126080
    },
    {
      "epoch": 18.77941614536789,
      "grad_norm": 15.6875,
      "learning_rate": 5.652785390935889e-07,
      "loss": 0.9622,
      "num_input_tokens_seen": 73153192,
      "step": 126085
    },
    {
      "epoch": 18.780160857908847,
      "grad_norm": 10.3125,
      "learning_rate": 5.645916559672004e-07,
      "loss": 0.9917,
      "num_input_tokens_seen": 73156168,
      "step": 126090
    },
    {
      "epoch": 18.780905570449807,
      "grad_norm": 13.3125,
      "learning_rate": 5.639051856522026e-07,
      "loss": 1.1494,
      "num_input_tokens_seen": 73158792,
      "step": 126095
    },
    {
      "epoch": 18.781650282990764,
      "grad_norm": 8.625,
      "learning_rate": 5.63219128160189e-07,
      "loss": 1.0278,
      "num_input_tokens_seen": 73161448,
      "step": 126100
    },
    {
      "epoch": 18.782394995531725,
      "grad_norm": 10.0,
      "learning_rate": 5.625334835027502e-07,
      "loss": 0.9682,
      "num_input_tokens_seen": 73164168,
      "step": 126105
    },
    {
      "epoch": 18.783139708072685,
      "grad_norm": 5.0,
      "learning_rate": 5.618482516914714e-07,
      "loss": 1.0419,
      "num_input_tokens_seen": 73167176,
      "step": 126110
    },
    {
      "epoch": 18.783884420613642,
      "grad_norm": 13.25,
      "learning_rate": 5.611634327379295e-07,
      "loss": 1.0433,
      "num_input_tokens_seen": 73170024,
      "step": 126115
    },
    {
      "epoch": 18.784629133154603,
      "grad_norm": 8.8125,
      "learning_rate": 5.60479026653693e-07,
      "loss": 0.9193,
      "num_input_tokens_seen": 73172648,
      "step": 126120
    },
    {
      "epoch": 18.785373845695563,
      "grad_norm": 10.125,
      "learning_rate": 5.59795033450325e-07,
      "loss": 1.1554,
      "num_input_tokens_seen": 73175560,
      "step": 126125
    },
    {
      "epoch": 18.78611855823652,
      "grad_norm": 10.25,
      "learning_rate": 5.591114531393771e-07,
      "loss": 1.17,
      "num_input_tokens_seen": 73178536,
      "step": 126130
    },
    {
      "epoch": 18.78686327077748,
      "grad_norm": 6.1875,
      "learning_rate": 5.584282857324014e-07,
      "loss": 1.2041,
      "num_input_tokens_seen": 73181544,
      "step": 126135
    },
    {
      "epoch": 18.787607983318438,
      "grad_norm": 13.5625,
      "learning_rate": 5.577455312409413e-07,
      "loss": 1.058,
      "num_input_tokens_seen": 73184360,
      "step": 126140
    },
    {
      "epoch": 18.7883526958594,
      "grad_norm": 8.8125,
      "learning_rate": 5.570631896765239e-07,
      "loss": 1.0204,
      "num_input_tokens_seen": 73187048,
      "step": 126145
    },
    {
      "epoch": 18.78909740840036,
      "grad_norm": 8.25,
      "learning_rate": 5.563812610506841e-07,
      "loss": 1.1638,
      "num_input_tokens_seen": 73189928,
      "step": 126150
    },
    {
      "epoch": 18.789842120941316,
      "grad_norm": 8.5625,
      "learning_rate": 5.556997453749379e-07,
      "loss": 0.9225,
      "num_input_tokens_seen": 73192744,
      "step": 126155
    },
    {
      "epoch": 18.790586833482276,
      "grad_norm": 10.6875,
      "learning_rate": 5.550186426608039e-07,
      "loss": 1.3928,
      "num_input_tokens_seen": 73195944,
      "step": 126160
    },
    {
      "epoch": 18.791331546023237,
      "grad_norm": 7.5,
      "learning_rate": 5.543379529197839e-07,
      "loss": 1.3683,
      "num_input_tokens_seen": 73198888,
      "step": 126165
    },
    {
      "epoch": 18.792076258564194,
      "grad_norm": 7.9375,
      "learning_rate": 5.536576761633772e-07,
      "loss": 0.8939,
      "num_input_tokens_seen": 73201928,
      "step": 126170
    },
    {
      "epoch": 18.792820971105154,
      "grad_norm": 7.59375,
      "learning_rate": 5.529778124030799e-07,
      "loss": 0.8268,
      "num_input_tokens_seen": 73204904,
      "step": 126175
    },
    {
      "epoch": 18.79356568364611,
      "grad_norm": 11.875,
      "learning_rate": 5.522983616503746e-07,
      "loss": 1.075,
      "num_input_tokens_seen": 73207784,
      "step": 126180
    },
    {
      "epoch": 18.794310396187072,
      "grad_norm": 19.75,
      "learning_rate": 5.51619323916741e-07,
      "loss": 0.9776,
      "num_input_tokens_seen": 73210568,
      "step": 126185
    },
    {
      "epoch": 18.795055108728032,
      "grad_norm": 8.375,
      "learning_rate": 5.509406992136479e-07,
      "loss": 0.9027,
      "num_input_tokens_seen": 73213320,
      "step": 126190
    },
    {
      "epoch": 18.79579982126899,
      "grad_norm": 15.1875,
      "learning_rate": 5.502624875525664e-07,
      "loss": 1.3344,
      "num_input_tokens_seen": 73215944,
      "step": 126195
    },
    {
      "epoch": 18.79654453380995,
      "grad_norm": 10.3125,
      "learning_rate": 5.495846889449485e-07,
      "loss": 0.7399,
      "num_input_tokens_seen": 73219016,
      "step": 126200
    },
    {
      "epoch": 18.797289246350907,
      "grad_norm": 8.625,
      "learning_rate": 5.48907303402249e-07,
      "loss": 1.1508,
      "num_input_tokens_seen": 73222216,
      "step": 126205
    },
    {
      "epoch": 18.798033958891867,
      "grad_norm": 13.125,
      "learning_rate": 5.48230330935906e-07,
      "loss": 1.3564,
      "num_input_tokens_seen": 73225032,
      "step": 126210
    },
    {
      "epoch": 18.798778671432828,
      "grad_norm": 6.96875,
      "learning_rate": 5.475537715573631e-07,
      "loss": 1.1394,
      "num_input_tokens_seen": 73227848,
      "step": 126215
    },
    {
      "epoch": 18.799523383973785,
      "grad_norm": 11.75,
      "learning_rate": 5.468776252780472e-07,
      "loss": 1.0113,
      "num_input_tokens_seen": 73230760,
      "step": 126220
    },
    {
      "epoch": 18.800268096514746,
      "grad_norm": 6.65625,
      "learning_rate": 5.46201892109377e-07,
      "loss": 1.0092,
      "num_input_tokens_seen": 73233768,
      "step": 126225
    },
    {
      "epoch": 18.801012809055706,
      "grad_norm": 16.125,
      "learning_rate": 5.455265720627767e-07,
      "loss": 1.1505,
      "num_input_tokens_seen": 73236552,
      "step": 126230
    },
    {
      "epoch": 18.801757521596663,
      "grad_norm": 9.0625,
      "learning_rate": 5.448516651496482e-07,
      "loss": 1.0007,
      "num_input_tokens_seen": 73239528,
      "step": 126235
    },
    {
      "epoch": 18.802502234137624,
      "grad_norm": 6.03125,
      "learning_rate": 5.441771713813992e-07,
      "loss": 0.9389,
      "num_input_tokens_seen": 73242216,
      "step": 126240
    },
    {
      "epoch": 18.80324694667858,
      "grad_norm": 6.0,
      "learning_rate": 5.435030907694149e-07,
      "loss": 0.906,
      "num_input_tokens_seen": 73245192,
      "step": 126245
    },
    {
      "epoch": 18.80399165921954,
      "grad_norm": 11.5625,
      "learning_rate": 5.428294233250947e-07,
      "loss": 0.7919,
      "num_input_tokens_seen": 73248104,
      "step": 126250
    },
    {
      "epoch": 18.8047363717605,
      "grad_norm": 12.8125,
      "learning_rate": 5.421561690598126e-07,
      "loss": 0.9853,
      "num_input_tokens_seen": 73250952,
      "step": 126255
    },
    {
      "epoch": 18.80548108430146,
      "grad_norm": 16.125,
      "learning_rate": 5.414833279849429e-07,
      "loss": 1.3869,
      "num_input_tokens_seen": 73253992,
      "step": 126260
    },
    {
      "epoch": 18.80622579684242,
      "grad_norm": 8.25,
      "learning_rate": 5.408109001118544e-07,
      "loss": 0.8451,
      "num_input_tokens_seen": 73256488,
      "step": 126265
    },
    {
      "epoch": 18.80697050938338,
      "grad_norm": 17.375,
      "learning_rate": 5.401388854519046e-07,
      "loss": 1.31,
      "num_input_tokens_seen": 73259208,
      "step": 126270
    },
    {
      "epoch": 18.807715221924337,
      "grad_norm": 8.375,
      "learning_rate": 5.394672840164511e-07,
      "loss": 0.817,
      "num_input_tokens_seen": 73261992,
      "step": 126275
    },
    {
      "epoch": 18.808459934465297,
      "grad_norm": 13.8125,
      "learning_rate": 5.387960958168375e-07,
      "loss": 1.3805,
      "num_input_tokens_seen": 73265096,
      "step": 126280
    },
    {
      "epoch": 18.809204647006254,
      "grad_norm": 15.375,
      "learning_rate": 5.381253208644021e-07,
      "loss": 1.1147,
      "num_input_tokens_seen": 73267880,
      "step": 126285
    },
    {
      "epoch": 18.809949359547215,
      "grad_norm": 5.6875,
      "learning_rate": 5.374549591704747e-07,
      "loss": 0.805,
      "num_input_tokens_seen": 73270760,
      "step": 126290
    },
    {
      "epoch": 18.810694072088175,
      "grad_norm": 5.84375,
      "learning_rate": 5.367850107463879e-07,
      "loss": 0.8895,
      "num_input_tokens_seen": 73273896,
      "step": 126295
    },
    {
      "epoch": 18.811438784629132,
      "grad_norm": 14.1875,
      "learning_rate": 5.36115475603452e-07,
      "loss": 1.1842,
      "num_input_tokens_seen": 73277128,
      "step": 126300
    },
    {
      "epoch": 18.812183497170093,
      "grad_norm": 15.0625,
      "learning_rate": 5.354463537529831e-07,
      "loss": 1.1465,
      "num_input_tokens_seen": 73279976,
      "step": 126305
    },
    {
      "epoch": 18.812928209711053,
      "grad_norm": 8.3125,
      "learning_rate": 5.347776452062831e-07,
      "loss": 1.0732,
      "num_input_tokens_seen": 73282856,
      "step": 126310
    },
    {
      "epoch": 18.81367292225201,
      "grad_norm": 7.09375,
      "learning_rate": 5.341093499746485e-07,
      "loss": 1.1208,
      "num_input_tokens_seen": 73286024,
      "step": 126315
    },
    {
      "epoch": 18.81441763479297,
      "grad_norm": 14.0625,
      "learning_rate": 5.334414680693705e-07,
      "loss": 1.0039,
      "num_input_tokens_seen": 73288936,
      "step": 126320
    },
    {
      "epoch": 18.815162347333928,
      "grad_norm": 9.5,
      "learning_rate": 5.327739995017316e-07,
      "loss": 1.2571,
      "num_input_tokens_seen": 73291912,
      "step": 126325
    },
    {
      "epoch": 18.81590705987489,
      "grad_norm": 12.125,
      "learning_rate": 5.32106944283009e-07,
      "loss": 1.1548,
      "num_input_tokens_seen": 73295048,
      "step": 126330
    },
    {
      "epoch": 18.81665177241585,
      "grad_norm": 19.75,
      "learning_rate": 5.31440302424474e-07,
      "loss": 1.2214,
      "num_input_tokens_seen": 73297736,
      "step": 126335
    },
    {
      "epoch": 18.817396484956806,
      "grad_norm": 10.375,
      "learning_rate": 5.307740739373818e-07,
      "loss": 0.9993,
      "num_input_tokens_seen": 73300648,
      "step": 126340
    },
    {
      "epoch": 18.818141197497766,
      "grad_norm": 9.625,
      "learning_rate": 5.301082588329953e-07,
      "loss": 1.0407,
      "num_input_tokens_seen": 73304040,
      "step": 126345
    },
    {
      "epoch": 18.818885910038723,
      "grad_norm": 15.8125,
      "learning_rate": 5.294428571225585e-07,
      "loss": 1.2932,
      "num_input_tokens_seen": 73306792,
      "step": 126350
    },
    {
      "epoch": 18.819630622579684,
      "grad_norm": 7.84375,
      "learning_rate": 5.287778688173151e-07,
      "loss": 1.2758,
      "num_input_tokens_seen": 73309672,
      "step": 126355
    },
    {
      "epoch": 18.820375335120644,
      "grad_norm": 7.28125,
      "learning_rate": 5.281132939284977e-07,
      "loss": 1.2317,
      "num_input_tokens_seen": 73312808,
      "step": 126360
    },
    {
      "epoch": 18.8211200476616,
      "grad_norm": 12.75,
      "learning_rate": 5.274491324673309e-07,
      "loss": 0.998,
      "num_input_tokens_seen": 73315624,
      "step": 126365
    },
    {
      "epoch": 18.821864760202562,
      "grad_norm": 9.5,
      "learning_rate": 5.267853844450416e-07,
      "loss": 1.0856,
      "num_input_tokens_seen": 73318632,
      "step": 126370
    },
    {
      "epoch": 18.822609472743522,
      "grad_norm": 11.5,
      "learning_rate": 5.261220498728403e-07,
      "loss": 1.5658,
      "num_input_tokens_seen": 73321480,
      "step": 126375
    },
    {
      "epoch": 18.82335418528448,
      "grad_norm": 9.875,
      "learning_rate": 5.254591287619348e-07,
      "loss": 1.1337,
      "num_input_tokens_seen": 73324232,
      "step": 126380
    },
    {
      "epoch": 18.82409889782544,
      "grad_norm": 10.375,
      "learning_rate": 5.247966211235161e-07,
      "loss": 1.0806,
      "num_input_tokens_seen": 73327432,
      "step": 126385
    },
    {
      "epoch": 18.824843610366397,
      "grad_norm": 4.78125,
      "learning_rate": 5.241345269687864e-07,
      "loss": 0.765,
      "num_input_tokens_seen": 73330408,
      "step": 126390
    },
    {
      "epoch": 18.825588322907358,
      "grad_norm": 7.09375,
      "learning_rate": 5.234728463089284e-07,
      "loss": 1.0802,
      "num_input_tokens_seen": 73333064,
      "step": 126395
    },
    {
      "epoch": 18.826333035448318,
      "grad_norm": 9.125,
      "learning_rate": 5.228115791551191e-07,
      "loss": 0.9054,
      "num_input_tokens_seen": 73336104,
      "step": 126400
    },
    {
      "epoch": 18.827077747989275,
      "grad_norm": 7.21875,
      "learning_rate": 5.221507255185304e-07,
      "loss": 1.123,
      "num_input_tokens_seen": 73338824,
      "step": 126405
    },
    {
      "epoch": 18.827822460530236,
      "grad_norm": 16.25,
      "learning_rate": 5.214902854103282e-07,
      "loss": 1.3133,
      "num_input_tokens_seen": 73341864,
      "step": 126410
    },
    {
      "epoch": 18.828567173071196,
      "grad_norm": 10.375,
      "learning_rate": 5.208302588416647e-07,
      "loss": 0.8975,
      "num_input_tokens_seen": 73344648,
      "step": 126415
    },
    {
      "epoch": 18.829311885612153,
      "grad_norm": 2.96875,
      "learning_rate": 5.201706458236977e-07,
      "loss": 1.1106,
      "num_input_tokens_seen": 73347848,
      "step": 126420
    },
    {
      "epoch": 18.830056598153114,
      "grad_norm": 21.75,
      "learning_rate": 5.195114463675682e-07,
      "loss": 1.371,
      "num_input_tokens_seen": 73350632,
      "step": 126425
    },
    {
      "epoch": 18.83080131069407,
      "grad_norm": 12.125,
      "learning_rate": 5.188526604844118e-07,
      "loss": 1.038,
      "num_input_tokens_seen": 73353384,
      "step": 126430
    },
    {
      "epoch": 18.83154602323503,
      "grad_norm": 5.21875,
      "learning_rate": 5.181942881853585e-07,
      "loss": 1.1421,
      "num_input_tokens_seen": 73356456,
      "step": 126435
    },
    {
      "epoch": 18.83229073577599,
      "grad_norm": 14.375,
      "learning_rate": 5.1753632948153e-07,
      "loss": 1.1068,
      "num_input_tokens_seen": 73359208,
      "step": 126440
    },
    {
      "epoch": 18.83303544831695,
      "grad_norm": 7.375,
      "learning_rate": 5.168787843840423e-07,
      "loss": 1.044,
      "num_input_tokens_seen": 73362024,
      "step": 126445
    },
    {
      "epoch": 18.83378016085791,
      "grad_norm": 9.9375,
      "learning_rate": 5.162216529040004e-07,
      "loss": 1.3752,
      "num_input_tokens_seen": 73364936,
      "step": 126450
    },
    {
      "epoch": 18.83452487339887,
      "grad_norm": 16.375,
      "learning_rate": 5.155649350525149e-07,
      "loss": 1.4716,
      "num_input_tokens_seen": 73367688,
      "step": 126455
    },
    {
      "epoch": 18.835269585939827,
      "grad_norm": 23.375,
      "learning_rate": 5.149086308406742e-07,
      "loss": 1.4266,
      "num_input_tokens_seen": 73370536,
      "step": 126460
    },
    {
      "epoch": 18.836014298480787,
      "grad_norm": 8.0,
      "learning_rate": 5.142527402795638e-07,
      "loss": 0.8875,
      "num_input_tokens_seen": 73373160,
      "step": 126465
    },
    {
      "epoch": 18.836759011021744,
      "grad_norm": 10.125,
      "learning_rate": 5.135972633802694e-07,
      "loss": 0.8537,
      "num_input_tokens_seen": 73376104,
      "step": 126470
    },
    {
      "epoch": 18.837503723562705,
      "grad_norm": 9.0625,
      "learning_rate": 5.129422001538597e-07,
      "loss": 1.0501,
      "num_input_tokens_seen": 73379048,
      "step": 126475
    },
    {
      "epoch": 18.838248436103665,
      "grad_norm": 15.4375,
      "learning_rate": 5.122875506114067e-07,
      "loss": 1.0447,
      "num_input_tokens_seen": 73381736,
      "step": 126480
    },
    {
      "epoch": 18.838993148644622,
      "grad_norm": 10.0625,
      "learning_rate": 5.116333147639651e-07,
      "loss": 0.7393,
      "num_input_tokens_seen": 73384616,
      "step": 126485
    },
    {
      "epoch": 18.839737861185583,
      "grad_norm": 6.40625,
      "learning_rate": 5.109794926225903e-07,
      "loss": 1.1905,
      "num_input_tokens_seen": 73387496,
      "step": 126490
    },
    {
      "epoch": 18.84048257372654,
      "grad_norm": 8.6875,
      "learning_rate": 5.103260841983287e-07,
      "loss": 0.8983,
      "num_input_tokens_seen": 73390344,
      "step": 126495
    },
    {
      "epoch": 18.8412272862675,
      "grad_norm": 11.0,
      "learning_rate": 5.096730895022189e-07,
      "loss": 1.5321,
      "num_input_tokens_seen": 73393160,
      "step": 126500
    },
    {
      "epoch": 18.84197199880846,
      "grad_norm": 8.4375,
      "learning_rate": 5.090205085452909e-07,
      "loss": 1.1161,
      "num_input_tokens_seen": 73395816,
      "step": 126505
    },
    {
      "epoch": 18.842716711349418,
      "grad_norm": 15.5625,
      "learning_rate": 5.083683413385665e-07,
      "loss": 1.0561,
      "num_input_tokens_seen": 73398952,
      "step": 126510
    },
    {
      "epoch": 18.84346142389038,
      "grad_norm": 10.3125,
      "learning_rate": 5.077165878930701e-07,
      "loss": 0.7868,
      "num_input_tokens_seen": 73401960,
      "step": 126515
    },
    {
      "epoch": 18.84420613643134,
      "grad_norm": 11.375,
      "learning_rate": 5.070652482198069e-07,
      "loss": 0.9191,
      "num_input_tokens_seen": 73404648,
      "step": 126520
    },
    {
      "epoch": 18.844950848972296,
      "grad_norm": 10.3125,
      "learning_rate": 5.064143223297845e-07,
      "loss": 1.1771,
      "num_input_tokens_seen": 73407272,
      "step": 126525
    },
    {
      "epoch": 18.845695561513256,
      "grad_norm": 7.125,
      "learning_rate": 5.057638102339945e-07,
      "loss": 0.7115,
      "num_input_tokens_seen": 73410088,
      "step": 126530
    },
    {
      "epoch": 18.846440274054213,
      "grad_norm": 7.28125,
      "learning_rate": 5.051137119434362e-07,
      "loss": 1.3377,
      "num_input_tokens_seen": 73412872,
      "step": 126535
    },
    {
      "epoch": 18.847184986595174,
      "grad_norm": 13.875,
      "learning_rate": 5.044640274690815e-07,
      "loss": 1.0919,
      "num_input_tokens_seen": 73415816,
      "step": 126540
    },
    {
      "epoch": 18.847929699136134,
      "grad_norm": 6.03125,
      "learning_rate": 5.038147568219131e-07,
      "loss": 0.8152,
      "num_input_tokens_seen": 73418600,
      "step": 126545
    },
    {
      "epoch": 18.84867441167709,
      "grad_norm": 13.3125,
      "learning_rate": 5.031659000128974e-07,
      "loss": 1.1737,
      "num_input_tokens_seen": 73421448,
      "step": 126550
    },
    {
      "epoch": 18.849419124218052,
      "grad_norm": 20.0,
      "learning_rate": 5.02517457052995e-07,
      "loss": 1.2989,
      "num_input_tokens_seen": 73424296,
      "step": 126555
    },
    {
      "epoch": 18.850163836759013,
      "grad_norm": 9.4375,
      "learning_rate": 5.018694279531638e-07,
      "loss": 0.9705,
      "num_input_tokens_seen": 73427528,
      "step": 126560
    },
    {
      "epoch": 18.85090854929997,
      "grad_norm": 16.5,
      "learning_rate": 5.012218127243478e-07,
      "loss": 1.2374,
      "num_input_tokens_seen": 73430504,
      "step": 126565
    },
    {
      "epoch": 18.85165326184093,
      "grad_norm": 7.625,
      "learning_rate": 5.005746113774912e-07,
      "loss": 1.302,
      "num_input_tokens_seen": 73433384,
      "step": 126570
    },
    {
      "epoch": 18.852397974381887,
      "grad_norm": 5.4375,
      "learning_rate": 4.999278239235267e-07,
      "loss": 1.0317,
      "num_input_tokens_seen": 73436264,
      "step": 126575
    },
    {
      "epoch": 18.853142686922848,
      "grad_norm": 10.9375,
      "learning_rate": 4.992814503733817e-07,
      "loss": 1.226,
      "num_input_tokens_seen": 73439048,
      "step": 126580
    },
    {
      "epoch": 18.853887399463808,
      "grad_norm": 18.375,
      "learning_rate": 4.986354907379726e-07,
      "loss": 1.0756,
      "num_input_tokens_seen": 73441672,
      "step": 126585
    },
    {
      "epoch": 18.854632112004765,
      "grad_norm": 11.3125,
      "learning_rate": 4.979899450282155e-07,
      "loss": 0.6584,
      "num_input_tokens_seen": 73444648,
      "step": 126590
    },
    {
      "epoch": 18.855376824545726,
      "grad_norm": 12.4375,
      "learning_rate": 4.973448132550157e-07,
      "loss": 1.2304,
      "num_input_tokens_seen": 73447720,
      "step": 126595
    },
    {
      "epoch": 18.856121537086686,
      "grad_norm": 12.3125,
      "learning_rate": 4.967000954292728e-07,
      "loss": 0.9861,
      "num_input_tokens_seen": 73450920,
      "step": 126600
    },
    {
      "epoch": 18.856866249627643,
      "grad_norm": 8.625,
      "learning_rate": 4.96055791561878e-07,
      "loss": 0.8847,
      "num_input_tokens_seen": 73453608,
      "step": 126605
    },
    {
      "epoch": 18.857610962168604,
      "grad_norm": 10.0625,
      "learning_rate": 4.954119016637115e-07,
      "loss": 1.0323,
      "num_input_tokens_seen": 73456968,
      "step": 126610
    },
    {
      "epoch": 18.85835567470956,
      "grad_norm": 16.5,
      "learning_rate": 4.94768425745662e-07,
      "loss": 1.0482,
      "num_input_tokens_seen": 73459848,
      "step": 126615
    },
    {
      "epoch": 18.85910038725052,
      "grad_norm": 15.1875,
      "learning_rate": 4.9412536381859e-07,
      "loss": 0.9145,
      "num_input_tokens_seen": 73462600,
      "step": 126620
    },
    {
      "epoch": 18.85984509979148,
      "grad_norm": 8.8125,
      "learning_rate": 4.934827158933647e-07,
      "loss": 1.0676,
      "num_input_tokens_seen": 73465576,
      "step": 126625
    },
    {
      "epoch": 18.86058981233244,
      "grad_norm": 11.375,
      "learning_rate": 4.928404819808413e-07,
      "loss": 1.3479,
      "num_input_tokens_seen": 73468232,
      "step": 126630
    },
    {
      "epoch": 18.8613345248734,
      "grad_norm": 9.6875,
      "learning_rate": 4.921986620918723e-07,
      "loss": 1.3388,
      "num_input_tokens_seen": 73471208,
      "step": 126635
    },
    {
      "epoch": 18.86207923741436,
      "grad_norm": 16.625,
      "learning_rate": 4.915572562372961e-07,
      "loss": 1.283,
      "num_input_tokens_seen": 73474248,
      "step": 126640
    },
    {
      "epoch": 18.862823949955317,
      "grad_norm": 10.8125,
      "learning_rate": 4.909162644279486e-07,
      "loss": 0.8715,
      "num_input_tokens_seen": 73476840,
      "step": 126645
    },
    {
      "epoch": 18.863568662496277,
      "grad_norm": 15.8125,
      "learning_rate": 4.902756866746627e-07,
      "loss": 1.1386,
      "num_input_tokens_seen": 73479528,
      "step": 126650
    },
    {
      "epoch": 18.864313375037234,
      "grad_norm": 9.8125,
      "learning_rate": 4.896355229882576e-07,
      "loss": 0.9167,
      "num_input_tokens_seen": 73482472,
      "step": 126655
    },
    {
      "epoch": 18.865058087578195,
      "grad_norm": 4.65625,
      "learning_rate": 4.889957733795525e-07,
      "loss": 0.9898,
      "num_input_tokens_seen": 73485672,
      "step": 126660
    },
    {
      "epoch": 18.865802800119155,
      "grad_norm": 12.375,
      "learning_rate": 4.883564378593497e-07,
      "loss": 0.7913,
      "num_input_tokens_seen": 73488584,
      "step": 126665
    },
    {
      "epoch": 18.866547512660112,
      "grad_norm": 9.0625,
      "learning_rate": 4.877175164384518e-07,
      "loss": 1.0055,
      "num_input_tokens_seen": 73491720,
      "step": 126670
    },
    {
      "epoch": 18.867292225201073,
      "grad_norm": 6.28125,
      "learning_rate": 4.870790091276555e-07,
      "loss": 1.0135,
      "num_input_tokens_seen": 73494664,
      "step": 126675
    },
    {
      "epoch": 18.868036937742033,
      "grad_norm": 15.375,
      "learning_rate": 4.864409159377415e-07,
      "loss": 1.065,
      "num_input_tokens_seen": 73497640,
      "step": 126680
    },
    {
      "epoch": 18.86878165028299,
      "grad_norm": 9.5,
      "learning_rate": 4.858032368794979e-07,
      "loss": 1.1764,
      "num_input_tokens_seen": 73500264,
      "step": 126685
    },
    {
      "epoch": 18.86952636282395,
      "grad_norm": 9.125,
      "learning_rate": 4.851659719636915e-07,
      "loss": 0.8875,
      "num_input_tokens_seen": 73503144,
      "step": 126690
    },
    {
      "epoch": 18.870271075364908,
      "grad_norm": 12.625,
      "learning_rate": 4.845291212010883e-07,
      "loss": 1.3584,
      "num_input_tokens_seen": 73506088,
      "step": 126695
    },
    {
      "epoch": 18.87101578790587,
      "grad_norm": 11.625,
      "learning_rate": 4.838926846024522e-07,
      "loss": 1.169,
      "num_input_tokens_seen": 73508936,
      "step": 126700
    },
    {
      "epoch": 18.87176050044683,
      "grad_norm": 13.625,
      "learning_rate": 4.832566621785329e-07,
      "loss": 1.1542,
      "num_input_tokens_seen": 73511688,
      "step": 126705
    },
    {
      "epoch": 18.872505212987786,
      "grad_norm": 7.09375,
      "learning_rate": 4.826210539400744e-07,
      "loss": 0.8144,
      "num_input_tokens_seen": 73514632,
      "step": 126710
    },
    {
      "epoch": 18.873249925528746,
      "grad_norm": 8.75,
      "learning_rate": 4.819858598978127e-07,
      "loss": 0.8097,
      "num_input_tokens_seen": 73517608,
      "step": 126715
    },
    {
      "epoch": 18.873994638069703,
      "grad_norm": 8.0625,
      "learning_rate": 4.81351080062481e-07,
      "loss": 0.8946,
      "num_input_tokens_seen": 73520456,
      "step": 126720
    },
    {
      "epoch": 18.874739350610664,
      "grad_norm": 12.0625,
      "learning_rate": 4.807167144448039e-07,
      "loss": 1.2184,
      "num_input_tokens_seen": 73523400,
      "step": 126725
    },
    {
      "epoch": 18.875484063151625,
      "grad_norm": 8.375,
      "learning_rate": 4.800827630554977e-07,
      "loss": 1.2909,
      "num_input_tokens_seen": 73526088,
      "step": 126730
    },
    {
      "epoch": 18.87622877569258,
      "grad_norm": 8.875,
      "learning_rate": 4.794492259052708e-07,
      "loss": 1.0095,
      "num_input_tokens_seen": 73529096,
      "step": 126735
    },
    {
      "epoch": 18.876973488233542,
      "grad_norm": 5.875,
      "learning_rate": 4.788161030048282e-07,
      "loss": 0.848,
      "num_input_tokens_seen": 73531912,
      "step": 126740
    },
    {
      "epoch": 18.877718200774503,
      "grad_norm": 9.3125,
      "learning_rate": 4.781833943648672e-07,
      "loss": 0.8277,
      "num_input_tokens_seen": 73534536,
      "step": 126745
    },
    {
      "epoch": 18.87846291331546,
      "grad_norm": 10.75,
      "learning_rate": 4.775510999960736e-07,
      "loss": 0.9134,
      "num_input_tokens_seen": 73537480,
      "step": 126750
    },
    {
      "epoch": 18.87920762585642,
      "grad_norm": 7.84375,
      "learning_rate": 4.769192199091305e-07,
      "loss": 0.8435,
      "num_input_tokens_seen": 73540520,
      "step": 126755
    },
    {
      "epoch": 18.879952338397377,
      "grad_norm": 7.75,
      "learning_rate": 4.7628775411471536e-07,
      "loss": 1.0723,
      "num_input_tokens_seen": 73543304,
      "step": 126760
    },
    {
      "epoch": 18.880697050938338,
      "grad_norm": 7.34375,
      "learning_rate": 4.7565670262349207e-07,
      "loss": 0.8826,
      "num_input_tokens_seen": 73546120,
      "step": 126765
    },
    {
      "epoch": 18.881441763479298,
      "grad_norm": 9.0625,
      "learning_rate": 4.750260654461214e-07,
      "loss": 1.3328,
      "num_input_tokens_seen": 73548968,
      "step": 126770
    },
    {
      "epoch": 18.882186476020255,
      "grad_norm": 6.5625,
      "learning_rate": 4.743958425932615e-07,
      "loss": 1.0497,
      "num_input_tokens_seen": 73551848,
      "step": 126775
    },
    {
      "epoch": 18.882931188561216,
      "grad_norm": 19.0,
      "learning_rate": 4.737660340755595e-07,
      "loss": 1.4094,
      "num_input_tokens_seen": 73555080,
      "step": 126780
    },
    {
      "epoch": 18.883675901102176,
      "grad_norm": 8.3125,
      "learning_rate": 4.731366399036485e-07,
      "loss": 1.255,
      "num_input_tokens_seen": 73557896,
      "step": 126785
    },
    {
      "epoch": 18.884420613643133,
      "grad_norm": 6.75,
      "learning_rate": 4.7250766008816726e-07,
      "loss": 1.0719,
      "num_input_tokens_seen": 73561064,
      "step": 126790
    },
    {
      "epoch": 18.885165326184094,
      "grad_norm": 9.125,
      "learning_rate": 4.7187909463974054e-07,
      "loss": 1.0562,
      "num_input_tokens_seen": 73563976,
      "step": 126795
    },
    {
      "epoch": 18.88591003872505,
      "grad_norm": 13.3125,
      "learning_rate": 4.712509435689877e-07,
      "loss": 1.1181,
      "num_input_tokens_seen": 73566984,
      "step": 126800
    },
    {
      "epoch": 18.88665475126601,
      "grad_norm": 9.6875,
      "learning_rate": 4.706232068865196e-07,
      "loss": 1.0156,
      "num_input_tokens_seen": 73570280,
      "step": 126805
    },
    {
      "epoch": 18.88739946380697,
      "grad_norm": 22.0,
      "learning_rate": 4.6999588460294177e-07,
      "loss": 1.232,
      "num_input_tokens_seen": 73573192,
      "step": 126810
    },
    {
      "epoch": 18.88814417634793,
      "grad_norm": 12.5,
      "learning_rate": 4.6936897672885117e-07,
      "loss": 1.3738,
      "num_input_tokens_seen": 73576008,
      "step": 126815
    },
    {
      "epoch": 18.88888888888889,
      "grad_norm": 12.875,
      "learning_rate": 4.6874248327484494e-07,
      "loss": 0.8744,
      "num_input_tokens_seen": 73579240,
      "step": 126820
    },
    {
      "epoch": 18.88963360142985,
      "grad_norm": 6.8125,
      "learning_rate": 4.681164042514979e-07,
      "loss": 1.2418,
      "num_input_tokens_seen": 73582184,
      "step": 126825
    },
    {
      "epoch": 18.890378313970807,
      "grad_norm": 6.59375,
      "learning_rate": 4.674907396693934e-07,
      "loss": 0.6815,
      "num_input_tokens_seen": 73585224,
      "step": 126830
    },
    {
      "epoch": 18.891123026511767,
      "grad_norm": 5.4375,
      "learning_rate": 4.668654895390978e-07,
      "loss": 1.1142,
      "num_input_tokens_seen": 73588168,
      "step": 126835
    },
    {
      "epoch": 18.891867739052724,
      "grad_norm": 12.8125,
      "learning_rate": 4.66240653871175e-07,
      "loss": 1.2151,
      "num_input_tokens_seen": 73590920,
      "step": 126840
    },
    {
      "epoch": 18.892612451593685,
      "grad_norm": 10.875,
      "learning_rate": 4.6561623267618037e-07,
      "loss": 1.2224,
      "num_input_tokens_seen": 73593576,
      "step": 126845
    },
    {
      "epoch": 18.893357164134645,
      "grad_norm": 15.5,
      "learning_rate": 4.6499222596466386e-07,
      "loss": 1.1759,
      "num_input_tokens_seen": 73596552,
      "step": 126850
    },
    {
      "epoch": 18.894101876675602,
      "grad_norm": 9.125,
      "learning_rate": 4.6436863374716976e-07,
      "loss": 0.8347,
      "num_input_tokens_seen": 73599464,
      "step": 126855
    },
    {
      "epoch": 18.894846589216563,
      "grad_norm": 9.0625,
      "learning_rate": 4.6374545603423134e-07,
      "loss": 0.7667,
      "num_input_tokens_seen": 73602312,
      "step": 126860
    },
    {
      "epoch": 18.89559130175752,
      "grad_norm": 10.25,
      "learning_rate": 4.6312269283637357e-07,
      "loss": 1.0648,
      "num_input_tokens_seen": 73605128,
      "step": 126865
    },
    {
      "epoch": 18.89633601429848,
      "grad_norm": 11.125,
      "learning_rate": 4.6250034416411845e-07,
      "loss": 0.8545,
      "num_input_tokens_seen": 73607976,
      "step": 126870
    },
    {
      "epoch": 18.89708072683944,
      "grad_norm": 7.1875,
      "learning_rate": 4.618784100279827e-07,
      "loss": 0.7313,
      "num_input_tokens_seen": 73611016,
      "step": 126875
    },
    {
      "epoch": 18.897825439380398,
      "grad_norm": 19.5,
      "learning_rate": 4.6125689043847453e-07,
      "loss": 1.5202,
      "num_input_tokens_seen": 73613704,
      "step": 126880
    },
    {
      "epoch": 18.89857015192136,
      "grad_norm": 13.625,
      "learning_rate": 4.606357854060855e-07,
      "loss": 1.0529,
      "num_input_tokens_seen": 73616552,
      "step": 126885
    },
    {
      "epoch": 18.89931486446232,
      "grad_norm": 14.875,
      "learning_rate": 4.6001509494131846e-07,
      "loss": 1.2375,
      "num_input_tokens_seen": 73619176,
      "step": 126890
    },
    {
      "epoch": 18.900059577003276,
      "grad_norm": 15.0,
      "learning_rate": 4.5939481905465655e-07,
      "loss": 1.1418,
      "num_input_tokens_seen": 73622120,
      "step": 126895
    },
    {
      "epoch": 18.900804289544237,
      "grad_norm": 8.625,
      "learning_rate": 4.5877495775657476e-07,
      "loss": 1.3564,
      "num_input_tokens_seen": 73625064,
      "step": 126900
    },
    {
      "epoch": 18.901549002085194,
      "grad_norm": 11.875,
      "learning_rate": 4.5815551105754804e-07,
      "loss": 1.3001,
      "num_input_tokens_seen": 73627944,
      "step": 126905
    },
    {
      "epoch": 18.902293714626154,
      "grad_norm": 8.4375,
      "learning_rate": 4.575364789680375e-07,
      "loss": 0.8726,
      "num_input_tokens_seen": 73630760,
      "step": 126910
    },
    {
      "epoch": 18.903038427167115,
      "grad_norm": 9.125,
      "learning_rate": 4.5691786149850977e-07,
      "loss": 1.0468,
      "num_input_tokens_seen": 73633480,
      "step": 126915
    },
    {
      "epoch": 18.90378313970807,
      "grad_norm": 8.875,
      "learning_rate": 4.562996586594037e-07,
      "loss": 0.9548,
      "num_input_tokens_seen": 73636584,
      "step": 126920
    },
    {
      "epoch": 18.904527852249032,
      "grad_norm": 10.125,
      "learning_rate": 4.5568187046117484e-07,
      "loss": 0.8994,
      "num_input_tokens_seen": 73639272,
      "step": 126925
    },
    {
      "epoch": 18.905272564789993,
      "grad_norm": 9.9375,
      "learning_rate": 4.550644969142537e-07,
      "loss": 0.8564,
      "num_input_tokens_seen": 73642408,
      "step": 126930
    },
    {
      "epoch": 18.90601727733095,
      "grad_norm": 13.1875,
      "learning_rate": 4.544475380290708e-07,
      "loss": 1.1624,
      "num_input_tokens_seen": 73645640,
      "step": 126935
    },
    {
      "epoch": 18.90676198987191,
      "grad_norm": 13.125,
      "learning_rate": 4.538309938160512e-07,
      "loss": 1.0756,
      "num_input_tokens_seen": 73648424,
      "step": 126940
    },
    {
      "epoch": 18.907506702412867,
      "grad_norm": 6.90625,
      "learning_rate": 4.532148642856088e-07,
      "loss": 1.0273,
      "num_input_tokens_seen": 73651432,
      "step": 126945
    },
    {
      "epoch": 18.908251414953828,
      "grad_norm": 6.59375,
      "learning_rate": 4.5259914944815184e-07,
      "loss": 1.184,
      "num_input_tokens_seen": 73654216,
      "step": 126950
    },
    {
      "epoch": 18.908996127494788,
      "grad_norm": 9.5625,
      "learning_rate": 4.519838493140832e-07,
      "loss": 1.1808,
      "num_input_tokens_seen": 73657064,
      "step": 126955
    },
    {
      "epoch": 18.909740840035745,
      "grad_norm": 18.125,
      "learning_rate": 4.513689638938001e-07,
      "loss": 1.2196,
      "num_input_tokens_seen": 73660008,
      "step": 126960
    },
    {
      "epoch": 18.910485552576706,
      "grad_norm": 14.875,
      "learning_rate": 4.507544931976887e-07,
      "loss": 1.7526,
      "num_input_tokens_seen": 73662856,
      "step": 126965
    },
    {
      "epoch": 18.911230265117666,
      "grad_norm": 11.5,
      "learning_rate": 4.501404372361295e-07,
      "loss": 1.0055,
      "num_input_tokens_seen": 73665672,
      "step": 126970
    },
    {
      "epoch": 18.911974977658623,
      "grad_norm": 8.375,
      "learning_rate": 4.495267960194921e-07,
      "loss": 0.9241,
      "num_input_tokens_seen": 73668584,
      "step": 126975
    },
    {
      "epoch": 18.912719690199584,
      "grad_norm": 12.25,
      "learning_rate": 4.4891356955815145e-07,
      "loss": 1.1458,
      "num_input_tokens_seen": 73671464,
      "step": 126980
    },
    {
      "epoch": 18.91346440274054,
      "grad_norm": 7.875,
      "learning_rate": 4.483007578624632e-07,
      "loss": 0.9824,
      "num_input_tokens_seen": 73674088,
      "step": 126985
    },
    {
      "epoch": 18.9142091152815,
      "grad_norm": 8.0625,
      "learning_rate": 4.476883609427773e-07,
      "loss": 0.8518,
      "num_input_tokens_seen": 73677000,
      "step": 126990
    },
    {
      "epoch": 18.914953827822462,
      "grad_norm": 10.0625,
      "learning_rate": 4.4707637880944675e-07,
      "loss": 0.8795,
      "num_input_tokens_seen": 73679784,
      "step": 126995
    },
    {
      "epoch": 18.91569854036342,
      "grad_norm": 16.25,
      "learning_rate": 4.4646481147280206e-07,
      "loss": 1.0123,
      "num_input_tokens_seen": 73682600,
      "step": 127000
    },
    {
      "epoch": 18.91644325290438,
      "grad_norm": 8.1875,
      "learning_rate": 4.458536589431822e-07,
      "loss": 1.1759,
      "num_input_tokens_seen": 73685288,
      "step": 127005
    },
    {
      "epoch": 18.917187965445336,
      "grad_norm": 10.4375,
      "learning_rate": 4.4524292123090673e-07,
      "loss": 0.8748,
      "num_input_tokens_seen": 73688008,
      "step": 127010
    },
    {
      "epoch": 18.917932677986297,
      "grad_norm": 12.0625,
      "learning_rate": 4.4463259834630066e-07,
      "loss": 1.4792,
      "num_input_tokens_seen": 73691048,
      "step": 127015
    },
    {
      "epoch": 18.918677390527257,
      "grad_norm": 9.6875,
      "learning_rate": 4.440226902996669e-07,
      "loss": 1.0481,
      "num_input_tokens_seen": 73693864,
      "step": 127020
    },
    {
      "epoch": 18.919422103068214,
      "grad_norm": 4.78125,
      "learning_rate": 4.4341319710131115e-07,
      "loss": 1.0956,
      "num_input_tokens_seen": 73696488,
      "step": 127025
    },
    {
      "epoch": 18.920166815609175,
      "grad_norm": 12.625,
      "learning_rate": 4.428041187615306e-07,
      "loss": 1.0831,
      "num_input_tokens_seen": 73699112,
      "step": 127030
    },
    {
      "epoch": 18.920911528150135,
      "grad_norm": 10.375,
      "learning_rate": 4.421954552906199e-07,
      "loss": 1.0918,
      "num_input_tokens_seen": 73701960,
      "step": 127035
    },
    {
      "epoch": 18.921656240691092,
      "grad_norm": 13.9375,
      "learning_rate": 4.415872066988541e-07,
      "loss": 0.9268,
      "num_input_tokens_seen": 73704872,
      "step": 127040
    },
    {
      "epoch": 18.922400953232053,
      "grad_norm": 14.1875,
      "learning_rate": 4.4097937299651115e-07,
      "loss": 0.8689,
      "num_input_tokens_seen": 73707944,
      "step": 127045
    },
    {
      "epoch": 18.92314566577301,
      "grad_norm": 16.75,
      "learning_rate": 4.4037195419386336e-07,
      "loss": 0.9812,
      "num_input_tokens_seen": 73711016,
      "step": 127050
    },
    {
      "epoch": 18.92389037831397,
      "grad_norm": 10.625,
      "learning_rate": 4.3976495030116915e-07,
      "loss": 1.4101,
      "num_input_tokens_seen": 73713672,
      "step": 127055
    },
    {
      "epoch": 18.92463509085493,
      "grad_norm": 8.5,
      "learning_rate": 4.3915836132868426e-07,
      "loss": 1.0909,
      "num_input_tokens_seen": 73716488,
      "step": 127060
    },
    {
      "epoch": 18.925379803395888,
      "grad_norm": 12.0,
      "learning_rate": 4.3855218728665883e-07,
      "loss": 1.2543,
      "num_input_tokens_seen": 73719112,
      "step": 127065
    },
    {
      "epoch": 18.92612451593685,
      "grad_norm": 9.3125,
      "learning_rate": 4.3794642818532905e-07,
      "loss": 0.8866,
      "num_input_tokens_seen": 73721992,
      "step": 127070
    },
    {
      "epoch": 18.92686922847781,
      "grad_norm": 9.625,
      "learning_rate": 4.3734108403493125e-07,
      "loss": 1.248,
      "num_input_tokens_seen": 73724840,
      "step": 127075
    },
    {
      "epoch": 18.927613941018766,
      "grad_norm": 6.3125,
      "learning_rate": 4.3673615484568776e-07,
      "loss": 1.0943,
      "num_input_tokens_seen": 73727688,
      "step": 127080
    },
    {
      "epoch": 18.928358653559727,
      "grad_norm": 12.125,
      "learning_rate": 4.3613164062782653e-07,
      "loss": 1.2297,
      "num_input_tokens_seen": 73730280,
      "step": 127085
    },
    {
      "epoch": 18.929103366100684,
      "grad_norm": 9.3125,
      "learning_rate": 4.3552754139155327e-07,
      "loss": 1.0562,
      "num_input_tokens_seen": 73733160,
      "step": 127090
    },
    {
      "epoch": 18.929848078641644,
      "grad_norm": 11.125,
      "learning_rate": 4.3492385714707927e-07,
      "loss": 1.0171,
      "num_input_tokens_seen": 73735880,
      "step": 127095
    },
    {
      "epoch": 18.930592791182605,
      "grad_norm": 9.0625,
      "learning_rate": 4.343205879045964e-07,
      "loss": 0.8824,
      "num_input_tokens_seen": 73738888,
      "step": 127100
    },
    {
      "epoch": 18.93133750372356,
      "grad_norm": 13.0,
      "learning_rate": 4.3371773367429924e-07,
      "loss": 1.3755,
      "num_input_tokens_seen": 73741544,
      "step": 127105
    },
    {
      "epoch": 18.932082216264522,
      "grad_norm": 17.625,
      "learning_rate": 4.331152944663769e-07,
      "loss": 1.1823,
      "num_input_tokens_seen": 73744296,
      "step": 127110
    },
    {
      "epoch": 18.932826928805483,
      "grad_norm": 10.3125,
      "learning_rate": 4.3251327029099897e-07,
      "loss": 1.1245,
      "num_input_tokens_seen": 73747112,
      "step": 127115
    },
    {
      "epoch": 18.93357164134644,
      "grad_norm": 8.375,
      "learning_rate": 4.319116611583407e-07,
      "loss": 0.8799,
      "num_input_tokens_seen": 73750120,
      "step": 127120
    },
    {
      "epoch": 18.9343163538874,
      "grad_norm": 11.125,
      "learning_rate": 4.3131046707856613e-07,
      "loss": 1.3903,
      "num_input_tokens_seen": 73753096,
      "step": 127125
    },
    {
      "epoch": 18.935061066428357,
      "grad_norm": 9.5,
      "learning_rate": 4.307096880618311e-07,
      "loss": 0.9407,
      "num_input_tokens_seen": 73756008,
      "step": 127130
    },
    {
      "epoch": 18.935805778969318,
      "grad_norm": 8.8125,
      "learning_rate": 4.30109324118283e-07,
      "loss": 1.079,
      "num_input_tokens_seen": 73758856,
      "step": 127135
    },
    {
      "epoch": 18.93655049151028,
      "grad_norm": 9.875,
      "learning_rate": 4.295093752580664e-07,
      "loss": 1.1513,
      "num_input_tokens_seen": 73762056,
      "step": 127140
    },
    {
      "epoch": 18.937295204051235,
      "grad_norm": 10.5,
      "learning_rate": 4.289098414913206e-07,
      "loss": 0.9255,
      "num_input_tokens_seen": 73765128,
      "step": 127145
    },
    {
      "epoch": 18.938039916592196,
      "grad_norm": 13.3125,
      "learning_rate": 4.283107228281652e-07,
      "loss": 1.2089,
      "num_input_tokens_seen": 73768008,
      "step": 127150
    },
    {
      "epoch": 18.938784629133156,
      "grad_norm": 7.625,
      "learning_rate": 4.277120192787282e-07,
      "loss": 0.8691,
      "num_input_tokens_seen": 73770760,
      "step": 127155
    },
    {
      "epoch": 18.939529341674113,
      "grad_norm": 10.75,
      "learning_rate": 4.271137308531237e-07,
      "loss": 0.9202,
      "num_input_tokens_seen": 73773992,
      "step": 127160
    },
    {
      "epoch": 18.940274054215074,
      "grad_norm": 14.75,
      "learning_rate": 4.265158575614575e-07,
      "loss": 0.9406,
      "num_input_tokens_seen": 73777128,
      "step": 127165
    },
    {
      "epoch": 18.94101876675603,
      "grad_norm": 17.0,
      "learning_rate": 4.259183994138299e-07,
      "loss": 1.2882,
      "num_input_tokens_seen": 73780456,
      "step": 127170
    },
    {
      "epoch": 18.94176347929699,
      "grad_norm": 6.3125,
      "learning_rate": 4.2532135642033565e-07,
      "loss": 1.0677,
      "num_input_tokens_seen": 73783592,
      "step": 127175
    },
    {
      "epoch": 18.942508191837952,
      "grad_norm": 6.15625,
      "learning_rate": 4.2472472859105827e-07,
      "loss": 0.9463,
      "num_input_tokens_seen": 73786696,
      "step": 127180
    },
    {
      "epoch": 18.94325290437891,
      "grad_norm": 8.25,
      "learning_rate": 4.241285159360814e-07,
      "loss": 0.9367,
      "num_input_tokens_seen": 73789640,
      "step": 127185
    },
    {
      "epoch": 18.94399761691987,
      "grad_norm": 7.34375,
      "learning_rate": 4.235327184654747e-07,
      "loss": 0.7683,
      "num_input_tokens_seen": 73792456,
      "step": 127190
    },
    {
      "epoch": 18.94474232946083,
      "grad_norm": 12.0625,
      "learning_rate": 4.229373361893024e-07,
      "loss": 1.1489,
      "num_input_tokens_seen": 73795304,
      "step": 127195
    },
    {
      "epoch": 18.945487042001787,
      "grad_norm": 9.875,
      "learning_rate": 4.223423691176287e-07,
      "loss": 1.3137,
      "num_input_tokens_seen": 73798376,
      "step": 127200
    },
    {
      "epoch": 18.946231754542747,
      "grad_norm": 9.125,
      "learning_rate": 4.2174781726049826e-07,
      "loss": 1.0501,
      "num_input_tokens_seen": 73801032,
      "step": 127205
    },
    {
      "epoch": 18.946976467083704,
      "grad_norm": 6.46875,
      "learning_rate": 4.2115368062796147e-07,
      "loss": 0.917,
      "num_input_tokens_seen": 73803912,
      "step": 127210
    },
    {
      "epoch": 18.947721179624665,
      "grad_norm": 7.28125,
      "learning_rate": 4.205599592300491e-07,
      "loss": 0.8363,
      "num_input_tokens_seen": 73806696,
      "step": 127215
    },
    {
      "epoch": 18.948465892165625,
      "grad_norm": 4.21875,
      "learning_rate": 4.199666530767948e-07,
      "loss": 1.09,
      "num_input_tokens_seen": 73809736,
      "step": 127220
    },
    {
      "epoch": 18.949210604706582,
      "grad_norm": 6.4375,
      "learning_rate": 4.19373762178224e-07,
      "loss": 1.1935,
      "num_input_tokens_seen": 73812904,
      "step": 127225
    },
    {
      "epoch": 18.949955317247543,
      "grad_norm": 11.0625,
      "learning_rate": 4.187812865443508e-07,
      "loss": 1.5097,
      "num_input_tokens_seen": 73815624,
      "step": 127230
    },
    {
      "epoch": 18.9507000297885,
      "grad_norm": 10.9375,
      "learning_rate": 4.1818922618518386e-07,
      "loss": 1.3962,
      "num_input_tokens_seen": 73818568,
      "step": 127235
    },
    {
      "epoch": 18.95144474232946,
      "grad_norm": 18.0,
      "learning_rate": 4.175975811107263e-07,
      "loss": 1.2313,
      "num_input_tokens_seen": 73821352,
      "step": 127240
    },
    {
      "epoch": 18.95218945487042,
      "grad_norm": 15.8125,
      "learning_rate": 4.17006351330973e-07,
      "loss": 1.3383,
      "num_input_tokens_seen": 73824232,
      "step": 127245
    },
    {
      "epoch": 18.952934167411378,
      "grad_norm": 5.9375,
      "learning_rate": 4.164155368559103e-07,
      "loss": 1.0883,
      "num_input_tokens_seen": 73827080,
      "step": 127250
    },
    {
      "epoch": 18.95367887995234,
      "grad_norm": 14.3125,
      "learning_rate": 4.1582513769552467e-07,
      "loss": 1.0529,
      "num_input_tokens_seen": 73829960,
      "step": 127255
    },
    {
      "epoch": 18.9544235924933,
      "grad_norm": 11.5,
      "learning_rate": 4.1523515385978317e-07,
      "loss": 0.9127,
      "num_input_tokens_seen": 73832776,
      "step": 127260
    },
    {
      "epoch": 18.955168305034256,
      "grad_norm": 10.6875,
      "learning_rate": 4.1464558535866117e-07,
      "loss": 1.0329,
      "num_input_tokens_seen": 73835528,
      "step": 127265
    },
    {
      "epoch": 18.955913017575217,
      "grad_norm": 6.46875,
      "learning_rate": 4.140564322021145e-07,
      "loss": 0.9927,
      "num_input_tokens_seen": 73838472,
      "step": 127270
    },
    {
      "epoch": 18.956657730116174,
      "grad_norm": 11.25,
      "learning_rate": 4.1346769440009094e-07,
      "loss": 1.0502,
      "num_input_tokens_seen": 73841352,
      "step": 127275
    },
    {
      "epoch": 18.957402442657134,
      "grad_norm": 8.625,
      "learning_rate": 4.12879371962549e-07,
      "loss": 0.9012,
      "num_input_tokens_seen": 73844392,
      "step": 127280
    },
    {
      "epoch": 18.958147155198095,
      "grad_norm": 12.1875,
      "learning_rate": 4.1229146489941416e-07,
      "loss": 1.3328,
      "num_input_tokens_seen": 73847432,
      "step": 127285
    },
    {
      "epoch": 18.95889186773905,
      "grad_norm": 13.125,
      "learning_rate": 4.1170397322063125e-07,
      "loss": 0.9535,
      "num_input_tokens_seen": 73850408,
      "step": 127290
    },
    {
      "epoch": 18.959636580280012,
      "grad_norm": 16.625,
      "learning_rate": 4.111168969361173e-07,
      "loss": 1.66,
      "num_input_tokens_seen": 73853480,
      "step": 127295
    },
    {
      "epoch": 18.960381292820973,
      "grad_norm": 11.625,
      "learning_rate": 4.1053023605579223e-07,
      "loss": 1.3511,
      "num_input_tokens_seen": 73856456,
      "step": 127300
    },
    {
      "epoch": 18.96112600536193,
      "grad_norm": 4.65625,
      "learning_rate": 4.0994399058956743e-07,
      "loss": 0.9233,
      "num_input_tokens_seen": 73859112,
      "step": 127305
    },
    {
      "epoch": 18.96187071790289,
      "grad_norm": 10.75,
      "learning_rate": 4.0935816054734343e-07,
      "loss": 0.9331,
      "num_input_tokens_seen": 73862088,
      "step": 127310
    },
    {
      "epoch": 18.962615430443847,
      "grad_norm": 10.3125,
      "learning_rate": 4.0877274593902335e-07,
      "loss": 1.1079,
      "num_input_tokens_seen": 73864936,
      "step": 127315
    },
    {
      "epoch": 18.963360142984808,
      "grad_norm": 6.28125,
      "learning_rate": 4.0818774677449377e-07,
      "loss": 1.0735,
      "num_input_tokens_seen": 73868104,
      "step": 127320
    },
    {
      "epoch": 18.96410485552577,
      "grad_norm": 6.8125,
      "learning_rate": 4.0760316306363844e-07,
      "loss": 0.9375,
      "num_input_tokens_seen": 73871080,
      "step": 127325
    },
    {
      "epoch": 18.964849568066725,
      "grad_norm": 11.4375,
      "learning_rate": 4.0701899481633277e-07,
      "loss": 0.8419,
      "num_input_tokens_seen": 73873832,
      "step": 127330
    },
    {
      "epoch": 18.965594280607686,
      "grad_norm": 9.125,
      "learning_rate": 4.0643524204244665e-07,
      "loss": 1.2139,
      "num_input_tokens_seen": 73876520,
      "step": 127335
    },
    {
      "epoch": 18.966338993148646,
      "grad_norm": 12.6875,
      "learning_rate": 4.0585190475184166e-07,
      "loss": 1.2485,
      "num_input_tokens_seen": 73879496,
      "step": 127340
    },
    {
      "epoch": 18.967083705689603,
      "grad_norm": 14.625,
      "learning_rate": 4.05268982954371e-07,
      "loss": 1.2278,
      "num_input_tokens_seen": 73882248,
      "step": 127345
    },
    {
      "epoch": 18.967828418230564,
      "grad_norm": 5.84375,
      "learning_rate": 4.0468647665988513e-07,
      "loss": 1.2446,
      "num_input_tokens_seen": 73885000,
      "step": 127350
    },
    {
      "epoch": 18.96857313077152,
      "grad_norm": 11.0625,
      "learning_rate": 4.041043858782234e-07,
      "loss": 1.2224,
      "num_input_tokens_seen": 73888040,
      "step": 127355
    },
    {
      "epoch": 18.96931784331248,
      "grad_norm": 8.875,
      "learning_rate": 4.0352271061921966e-07,
      "loss": 0.9657,
      "num_input_tokens_seen": 73890856,
      "step": 127360
    },
    {
      "epoch": 18.970062555853442,
      "grad_norm": 6.59375,
      "learning_rate": 4.0294145089270205e-07,
      "loss": 0.8724,
      "num_input_tokens_seen": 73893896,
      "step": 127365
    },
    {
      "epoch": 18.9708072683944,
      "grad_norm": 12.125,
      "learning_rate": 4.0236060670848783e-07,
      "loss": 0.9459,
      "num_input_tokens_seen": 73896616,
      "step": 127370
    },
    {
      "epoch": 18.97155198093536,
      "grad_norm": 8.6875,
      "learning_rate": 4.0178017807639136e-07,
      "loss": 1.1767,
      "num_input_tokens_seen": 73899656,
      "step": 127375
    },
    {
      "epoch": 18.972296693476316,
      "grad_norm": 9.9375,
      "learning_rate": 4.012001650062186e-07,
      "loss": 1.2047,
      "num_input_tokens_seen": 73902696,
      "step": 127380
    },
    {
      "epoch": 18.973041406017277,
      "grad_norm": 12.6875,
      "learning_rate": 4.0062056750776734e-07,
      "loss": 1.1359,
      "num_input_tokens_seen": 73905480,
      "step": 127385
    },
    {
      "epoch": 18.973786118558237,
      "grad_norm": 6.65625,
      "learning_rate": 4.000413855908297e-07,
      "loss": 1.0738,
      "num_input_tokens_seen": 73908264,
      "step": 127390
    },
    {
      "epoch": 18.974530831099194,
      "grad_norm": 13.875,
      "learning_rate": 3.9946261926519233e-07,
      "loss": 0.8455,
      "num_input_tokens_seen": 73911272,
      "step": 127395
    },
    {
      "epoch": 18.975275543640155,
      "grad_norm": 5.6875,
      "learning_rate": 3.9888426854063075e-07,
      "loss": 0.7812,
      "num_input_tokens_seen": 73914376,
      "step": 127400
    },
    {
      "epoch": 18.976020256181116,
      "grad_norm": 8.3125,
      "learning_rate": 3.9830633342691494e-07,
      "loss": 1.0309,
      "num_input_tokens_seen": 73917064,
      "step": 127405
    },
    {
      "epoch": 18.976764968722073,
      "grad_norm": 13.5,
      "learning_rate": 3.9772881393380923e-07,
      "loss": 1.0558,
      "num_input_tokens_seen": 73920232,
      "step": 127410
    },
    {
      "epoch": 18.977509681263033,
      "grad_norm": 6.53125,
      "learning_rate": 3.9715171007107256e-07,
      "loss": 1.0092,
      "num_input_tokens_seen": 73922952,
      "step": 127415
    },
    {
      "epoch": 18.97825439380399,
      "grad_norm": 11.5,
      "learning_rate": 3.9657502184844983e-07,
      "loss": 1.0112,
      "num_input_tokens_seen": 73925768,
      "step": 127420
    },
    {
      "epoch": 18.97899910634495,
      "grad_norm": 9.1875,
      "learning_rate": 3.959987492756889e-07,
      "loss": 1.2365,
      "num_input_tokens_seen": 73928616,
      "step": 127425
    },
    {
      "epoch": 18.97974381888591,
      "grad_norm": 18.0,
      "learning_rate": 3.9542289236252363e-07,
      "loss": 0.9234,
      "num_input_tokens_seen": 73931496,
      "step": 127430
    },
    {
      "epoch": 18.980488531426868,
      "grad_norm": 5.84375,
      "learning_rate": 3.948474511186767e-07,
      "loss": 0.7535,
      "num_input_tokens_seen": 73934504,
      "step": 127435
    },
    {
      "epoch": 18.98123324396783,
      "grad_norm": 9.5,
      "learning_rate": 3.9427242555387935e-07,
      "loss": 1.0923,
      "num_input_tokens_seen": 73937704,
      "step": 127440
    },
    {
      "epoch": 18.98197795650879,
      "grad_norm": 12.8125,
      "learning_rate": 3.936978156778376e-07,
      "loss": 1.0235,
      "num_input_tokens_seen": 73940680,
      "step": 127445
    },
    {
      "epoch": 18.982722669049746,
      "grad_norm": 9.0,
      "learning_rate": 3.9312362150026594e-07,
      "loss": 1.3152,
      "num_input_tokens_seen": 73943592,
      "step": 127450
    },
    {
      "epoch": 18.983467381590707,
      "grad_norm": 13.8125,
      "learning_rate": 3.925498430308594e-07,
      "loss": 0.991,
      "num_input_tokens_seen": 73946184,
      "step": 127455
    },
    {
      "epoch": 18.984212094131664,
      "grad_norm": 14.125,
      "learning_rate": 3.91976480279313e-07,
      "loss": 1.274,
      "num_input_tokens_seen": 73948840,
      "step": 127460
    },
    {
      "epoch": 18.984956806672624,
      "grad_norm": 8.5625,
      "learning_rate": 3.914035332553162e-07,
      "loss": 1.202,
      "num_input_tokens_seen": 73951528,
      "step": 127465
    },
    {
      "epoch": 18.985701519213585,
      "grad_norm": 18.875,
      "learning_rate": 3.9083100196854183e-07,
      "loss": 1.2245,
      "num_input_tokens_seen": 73954312,
      "step": 127470
    },
    {
      "epoch": 18.98644623175454,
      "grad_norm": 4.8125,
      "learning_rate": 3.9025888642866827e-07,
      "loss": 0.7387,
      "num_input_tokens_seen": 73957000,
      "step": 127475
    },
    {
      "epoch": 18.987190944295502,
      "grad_norm": 10.0625,
      "learning_rate": 3.896871866453572e-07,
      "loss": 0.8731,
      "num_input_tokens_seen": 73960104,
      "step": 127480
    },
    {
      "epoch": 18.987935656836463,
      "grad_norm": 13.125,
      "learning_rate": 3.891159026282704e-07,
      "loss": 1.2147,
      "num_input_tokens_seen": 73962632,
      "step": 127485
    },
    {
      "epoch": 18.98868036937742,
      "grad_norm": 8.8125,
      "learning_rate": 3.885450343870556e-07,
      "loss": 0.9115,
      "num_input_tokens_seen": 73965704,
      "step": 127490
    },
    {
      "epoch": 18.98942508191838,
      "grad_norm": 13.875,
      "learning_rate": 3.8797458193135793e-07,
      "loss": 1.1719,
      "num_input_tokens_seen": 73968776,
      "step": 127495
    },
    {
      "epoch": 18.990169794459337,
      "grad_norm": 9.4375,
      "learning_rate": 3.8740454527081693e-07,
      "loss": 1.1106,
      "num_input_tokens_seen": 73971624,
      "step": 127500
    },
    {
      "epoch": 18.990914507000298,
      "grad_norm": 9.4375,
      "learning_rate": 3.8683492441506097e-07,
      "loss": 1.0655,
      "num_input_tokens_seen": 73974504,
      "step": 127505
    },
    {
      "epoch": 18.99165921954126,
      "grad_norm": 6.09375,
      "learning_rate": 3.862657193737129e-07,
      "loss": 1.2482,
      "num_input_tokens_seen": 73977736,
      "step": 127510
    },
    {
      "epoch": 18.992403932082215,
      "grad_norm": 13.1875,
      "learning_rate": 3.856969301563873e-07,
      "loss": 1.422,
      "num_input_tokens_seen": 73980648,
      "step": 127515
    },
    {
      "epoch": 18.993148644623176,
      "grad_norm": 9.0625,
      "learning_rate": 3.8512855677269586e-07,
      "loss": 1.2156,
      "num_input_tokens_seen": 73983560,
      "step": 127520
    },
    {
      "epoch": 18.993893357164133,
      "grad_norm": 7.0625,
      "learning_rate": 3.845605992322393e-07,
      "loss": 1.0428,
      "num_input_tokens_seen": 73986376,
      "step": 127525
    },
    {
      "epoch": 18.994638069705093,
      "grad_norm": 6.03125,
      "learning_rate": 3.8399305754461546e-07,
      "loss": 1.3758,
      "num_input_tokens_seen": 73989288,
      "step": 127530
    },
    {
      "epoch": 18.995382782246054,
      "grad_norm": 12.25,
      "learning_rate": 3.834259317194083e-07,
      "loss": 1.4264,
      "num_input_tokens_seen": 73992104,
      "step": 127535
    },
    {
      "epoch": 18.99612749478701,
      "grad_norm": 11.875,
      "learning_rate": 3.828592217662047e-07,
      "loss": 0.9608,
      "num_input_tokens_seen": 73995080,
      "step": 127540
    },
    {
      "epoch": 18.99687220732797,
      "grad_norm": 12.125,
      "learning_rate": 3.8229292769457193e-07,
      "loss": 0.949,
      "num_input_tokens_seen": 73997800,
      "step": 127545
    },
    {
      "epoch": 18.997616919868932,
      "grad_norm": 5.84375,
      "learning_rate": 3.8172704951408013e-07,
      "loss": 0.9978,
      "num_input_tokens_seen": 74000520,
      "step": 127550
    },
    {
      "epoch": 18.99836163240989,
      "grad_norm": 7.28125,
      "learning_rate": 3.811615872342883e-07,
      "loss": 1.034,
      "num_input_tokens_seen": 74003368,
      "step": 127555
    },
    {
      "epoch": 18.99910634495085,
      "grad_norm": 11.125,
      "learning_rate": 3.805965408647527e-07,
      "loss": 1.1023,
      "num_input_tokens_seen": 74006088,
      "step": 127560
    },
    {
      "epoch": 18.999851057491806,
      "grad_norm": 7.21875,
      "learning_rate": 3.8003191041501575e-07,
      "loss": 1.1357,
      "num_input_tokens_seen": 74009000,
      "step": 127565
    },
    {
      "epoch": 19.0,
      "eval_loss": 1.0852864980697632,
      "eval_runtime": 43.8331,
      "eval_samples_per_second": 68.076,
      "eval_steps_per_second": 17.019,
      "num_input_tokens_seen": 74009160,
      "step": 127566
    },
    {
      "epoch": 19.000595770032767,
      "grad_norm": 6.3125,
      "learning_rate": 3.794676958946142e-07,
      "loss": 0.8754,
      "num_input_tokens_seen": 74011400,
      "step": 127570
    },
    {
      "epoch": 19.001340482573728,
      "grad_norm": 4.5625,
      "learning_rate": 3.7890389731308486e-07,
      "loss": 0.9985,
      "num_input_tokens_seen": 74014120,
      "step": 127575
    },
    {
      "epoch": 19.002085195114685,
      "grad_norm": 14.5,
      "learning_rate": 3.783405146799479e-07,
      "loss": 1.2186,
      "num_input_tokens_seen": 74016744,
      "step": 127580
    },
    {
      "epoch": 19.002829907655645,
      "grad_norm": 14.625,
      "learning_rate": 3.777775480047263e-07,
      "loss": 1.015,
      "num_input_tokens_seen": 74019464,
      "step": 127585
    },
    {
      "epoch": 19.003574620196606,
      "grad_norm": 12.75,
      "learning_rate": 3.772149972969291e-07,
      "loss": 1.3159,
      "num_input_tokens_seen": 74022568,
      "step": 127590
    },
    {
      "epoch": 19.004319332737563,
      "grad_norm": 14.75,
      "learning_rate": 3.76652862566057e-07,
      "loss": 1.151,
      "num_input_tokens_seen": 74025320,
      "step": 127595
    },
    {
      "epoch": 19.005064045278523,
      "grad_norm": 12.9375,
      "learning_rate": 3.7609114382160803e-07,
      "loss": 1.0829,
      "num_input_tokens_seen": 74028168,
      "step": 127600
    },
    {
      "epoch": 19.00580875781948,
      "grad_norm": 16.25,
      "learning_rate": 3.7552984107307177e-07,
      "loss": 1.2765,
      "num_input_tokens_seen": 74031048,
      "step": 127605
    },
    {
      "epoch": 19.00655347036044,
      "grad_norm": 10.3125,
      "learning_rate": 3.7496895432993505e-07,
      "loss": 1.0286,
      "num_input_tokens_seen": 74033736,
      "step": 127610
    },
    {
      "epoch": 19.0072981829014,
      "grad_norm": 6.5,
      "learning_rate": 3.7440848360166813e-07,
      "loss": 1.1338,
      "num_input_tokens_seen": 74036360,
      "step": 127615
    },
    {
      "epoch": 19.008042895442358,
      "grad_norm": 5.9375,
      "learning_rate": 3.738484288977412e-07,
      "loss": 0.9894,
      "num_input_tokens_seen": 74038952,
      "step": 127620
    },
    {
      "epoch": 19.00878760798332,
      "grad_norm": 8.1875,
      "learning_rate": 3.7328879022761886e-07,
      "loss": 0.8845,
      "num_input_tokens_seen": 74041576,
      "step": 127625
    },
    {
      "epoch": 19.00953232052428,
      "grad_norm": 10.3125,
      "learning_rate": 3.7272956760075197e-07,
      "loss": 1.3113,
      "num_input_tokens_seen": 74044488,
      "step": 127630
    },
    {
      "epoch": 19.010277033065236,
      "grad_norm": 9.5625,
      "learning_rate": 3.7217076102658845e-07,
      "loss": 1.0428,
      "num_input_tokens_seen": 74047176,
      "step": 127635
    },
    {
      "epoch": 19.011021745606197,
      "grad_norm": 6.21875,
      "learning_rate": 3.7161237051456796e-07,
      "loss": 0.9334,
      "num_input_tokens_seen": 74050024,
      "step": 127640
    },
    {
      "epoch": 19.011766458147154,
      "grad_norm": 10.4375,
      "learning_rate": 3.710543960741275e-07,
      "loss": 1.1026,
      "num_input_tokens_seen": 74052904,
      "step": 127645
    },
    {
      "epoch": 19.012511170688114,
      "grad_norm": 12.9375,
      "learning_rate": 3.7049683771468723e-07,
      "loss": 1.2722,
      "num_input_tokens_seen": 74055592,
      "step": 127650
    },
    {
      "epoch": 19.013255883229075,
      "grad_norm": 9.25,
      "learning_rate": 3.6993969544567575e-07,
      "loss": 1.367,
      "num_input_tokens_seen": 74058760,
      "step": 127655
    },
    {
      "epoch": 19.01400059577003,
      "grad_norm": 20.75,
      "learning_rate": 3.693829692764966e-07,
      "loss": 1.1738,
      "num_input_tokens_seen": 74061864,
      "step": 127660
    },
    {
      "epoch": 19.014745308310992,
      "grad_norm": 7.25,
      "learning_rate": 3.68826659216559e-07,
      "loss": 0.982,
      "num_input_tokens_seen": 74064712,
      "step": 127665
    },
    {
      "epoch": 19.015490020851953,
      "grad_norm": 7.3125,
      "learning_rate": 3.682707652752637e-07,
      "loss": 1.0848,
      "num_input_tokens_seen": 74067496,
      "step": 127670
    },
    {
      "epoch": 19.01623473339291,
      "grad_norm": 11.75,
      "learning_rate": 3.677152874619949e-07,
      "loss": 0.9662,
      "num_input_tokens_seen": 74070376,
      "step": 127675
    },
    {
      "epoch": 19.01697944593387,
      "grad_norm": 8.4375,
      "learning_rate": 3.671602257861451e-07,
      "loss": 0.9057,
      "num_input_tokens_seen": 74073384,
      "step": 127680
    },
    {
      "epoch": 19.017724158474827,
      "grad_norm": 10.5625,
      "learning_rate": 3.666055802570845e-07,
      "loss": 1.105,
      "num_input_tokens_seen": 74076424,
      "step": 127685
    },
    {
      "epoch": 19.018468871015788,
      "grad_norm": 7.21875,
      "learning_rate": 3.6605135088418895e-07,
      "loss": 0.7468,
      "num_input_tokens_seen": 74078952,
      "step": 127690
    },
    {
      "epoch": 19.01921358355675,
      "grad_norm": 7.3125,
      "learning_rate": 3.654975376768205e-07,
      "loss": 1.2042,
      "num_input_tokens_seen": 74082312,
      "step": 127695
    },
    {
      "epoch": 19.019958296097705,
      "grad_norm": 12.9375,
      "learning_rate": 3.649441406443327e-07,
      "loss": 1.0547,
      "num_input_tokens_seen": 74085384,
      "step": 127700
    },
    {
      "epoch": 19.020703008638666,
      "grad_norm": 12.875,
      "learning_rate": 3.643911597960736e-07,
      "loss": 0.9567,
      "num_input_tokens_seen": 74088200,
      "step": 127705
    },
    {
      "epoch": 19.021447721179623,
      "grad_norm": 18.0,
      "learning_rate": 3.6383859514138864e-07,
      "loss": 0.9923,
      "num_input_tokens_seen": 74090888,
      "step": 127710
    },
    {
      "epoch": 19.022192433720583,
      "grad_norm": 9.0,
      "learning_rate": 3.6328644668961187e-07,
      "loss": 1.0809,
      "num_input_tokens_seen": 74093736,
      "step": 127715
    },
    {
      "epoch": 19.022937146261544,
      "grad_norm": 9.75,
      "learning_rate": 3.6273471445006923e-07,
      "loss": 1.2453,
      "num_input_tokens_seen": 74096680,
      "step": 127720
    },
    {
      "epoch": 19.0236818588025,
      "grad_norm": 9.5625,
      "learning_rate": 3.621833984320838e-07,
      "loss": 0.8017,
      "num_input_tokens_seen": 74099720,
      "step": 127725
    },
    {
      "epoch": 19.02442657134346,
      "grad_norm": 12.3125,
      "learning_rate": 3.616324986449676e-07,
      "loss": 0.9352,
      "num_input_tokens_seen": 74102664,
      "step": 127730
    },
    {
      "epoch": 19.025171283884422,
      "grad_norm": 17.125,
      "learning_rate": 3.6108201509803263e-07,
      "loss": 1.5625,
      "num_input_tokens_seen": 74105640,
      "step": 127735
    },
    {
      "epoch": 19.02591599642538,
      "grad_norm": 12.75,
      "learning_rate": 3.605319478005714e-07,
      "loss": 0.9982,
      "num_input_tokens_seen": 74108424,
      "step": 127740
    },
    {
      "epoch": 19.02666070896634,
      "grad_norm": 8.1875,
      "learning_rate": 3.599822967618849e-07,
      "loss": 1.0705,
      "num_input_tokens_seen": 74111400,
      "step": 127745
    },
    {
      "epoch": 19.027405421507297,
      "grad_norm": 6.25,
      "learning_rate": 3.594330619912517e-07,
      "loss": 1.2236,
      "num_input_tokens_seen": 74114344,
      "step": 127750
    },
    {
      "epoch": 19.028150134048257,
      "grad_norm": 6.0625,
      "learning_rate": 3.5888424349795615e-07,
      "loss": 0.7191,
      "num_input_tokens_seen": 74117288,
      "step": 127755
    },
    {
      "epoch": 19.028894846589218,
      "grad_norm": 16.5,
      "learning_rate": 3.5833584129126574e-07,
      "loss": 1.08,
      "num_input_tokens_seen": 74120168,
      "step": 127760
    },
    {
      "epoch": 19.029639559130175,
      "grad_norm": 9.1875,
      "learning_rate": 3.5778785538044255e-07,
      "loss": 1.1745,
      "num_input_tokens_seen": 74123368,
      "step": 127765
    },
    {
      "epoch": 19.030384271671135,
      "grad_norm": 13.75,
      "learning_rate": 3.572402857747542e-07,
      "loss": 1.1853,
      "num_input_tokens_seen": 74126248,
      "step": 127770
    },
    {
      "epoch": 19.031128984212096,
      "grad_norm": 5.46875,
      "learning_rate": 3.566931324834405e-07,
      "loss": 1.0921,
      "num_input_tokens_seen": 74129320,
      "step": 127775
    },
    {
      "epoch": 19.031873696753053,
      "grad_norm": 12.9375,
      "learning_rate": 3.5614639551575235e-07,
      "loss": 1.191,
      "num_input_tokens_seen": 74132200,
      "step": 127780
    },
    {
      "epoch": 19.032618409294013,
      "grad_norm": 7.59375,
      "learning_rate": 3.5560007488092404e-07,
      "loss": 1.1373,
      "num_input_tokens_seen": 74134952,
      "step": 127785
    },
    {
      "epoch": 19.03336312183497,
      "grad_norm": 11.8125,
      "learning_rate": 3.5505417058818437e-07,
      "loss": 0.8947,
      "num_input_tokens_seen": 74137576,
      "step": 127790
    },
    {
      "epoch": 19.03410783437593,
      "grad_norm": 13.3125,
      "learning_rate": 3.545086826467592e-07,
      "loss": 1.1323,
      "num_input_tokens_seen": 74140360,
      "step": 127795
    },
    {
      "epoch": 19.03485254691689,
      "grad_norm": 6.21875,
      "learning_rate": 3.5396361106585787e-07,
      "loss": 1.1538,
      "num_input_tokens_seen": 74143592,
      "step": 127800
    },
    {
      "epoch": 19.035597259457848,
      "grad_norm": 8.4375,
      "learning_rate": 3.534189558546924e-07,
      "loss": 0.9241,
      "num_input_tokens_seen": 74146600,
      "step": 127805
    },
    {
      "epoch": 19.03634197199881,
      "grad_norm": 13.0625,
      "learning_rate": 3.5287471702246386e-07,
      "loss": 1.5661,
      "num_input_tokens_seen": 74149512,
      "step": 127810
    },
    {
      "epoch": 19.03708668453977,
      "grad_norm": 9.1875,
      "learning_rate": 3.5233089457837045e-07,
      "loss": 1.0345,
      "num_input_tokens_seen": 74152680,
      "step": 127815
    },
    {
      "epoch": 19.037831397080726,
      "grad_norm": 9.625,
      "learning_rate": 3.517874885315936e-07,
      "loss": 0.938,
      "num_input_tokens_seen": 74155528,
      "step": 127820
    },
    {
      "epoch": 19.038576109621687,
      "grad_norm": 8.5,
      "learning_rate": 3.5124449889131495e-07,
      "loss": 1.5288,
      "num_input_tokens_seen": 74158472,
      "step": 127825
    },
    {
      "epoch": 19.039320822162644,
      "grad_norm": 9.25,
      "learning_rate": 3.5070192566671046e-07,
      "loss": 1.3659,
      "num_input_tokens_seen": 74161352,
      "step": 127830
    },
    {
      "epoch": 19.040065534703604,
      "grad_norm": 7.5,
      "learning_rate": 3.5015976886694226e-07,
      "loss": 0.9027,
      "num_input_tokens_seen": 74164456,
      "step": 127835
    },
    {
      "epoch": 19.040810247244565,
      "grad_norm": 14.125,
      "learning_rate": 3.496180285011724e-07,
      "loss": 0.9675,
      "num_input_tokens_seen": 74167464,
      "step": 127840
    },
    {
      "epoch": 19.041554959785522,
      "grad_norm": 13.125,
      "learning_rate": 3.49076704578552e-07,
      "loss": 1.065,
      "num_input_tokens_seen": 74170408,
      "step": 127845
    },
    {
      "epoch": 19.042299672326482,
      "grad_norm": 13.4375,
      "learning_rate": 3.4853579710822923e-07,
      "loss": 0.8331,
      "num_input_tokens_seen": 74173160,
      "step": 127850
    },
    {
      "epoch": 19.043044384867443,
      "grad_norm": 8.9375,
      "learning_rate": 3.4799530609933575e-07,
      "loss": 1.2917,
      "num_input_tokens_seen": 74176136,
      "step": 127855
    },
    {
      "epoch": 19.0437890974084,
      "grad_norm": 12.4375,
      "learning_rate": 3.474552315610086e-07,
      "loss": 0.9025,
      "num_input_tokens_seen": 74178856,
      "step": 127860
    },
    {
      "epoch": 19.04453380994936,
      "grad_norm": 13.9375,
      "learning_rate": 3.4691557350236827e-07,
      "loss": 1.3287,
      "num_input_tokens_seen": 74181960,
      "step": 127865
    },
    {
      "epoch": 19.045278522490317,
      "grad_norm": 6.03125,
      "learning_rate": 3.4637633193253525e-07,
      "loss": 0.9739,
      "num_input_tokens_seen": 74185000,
      "step": 127870
    },
    {
      "epoch": 19.046023235031278,
      "grad_norm": 10.625,
      "learning_rate": 3.458375068606162e-07,
      "loss": 1.268,
      "num_input_tokens_seen": 74188008,
      "step": 127875
    },
    {
      "epoch": 19.04676794757224,
      "grad_norm": 7.28125,
      "learning_rate": 3.4529909829571494e-07,
      "loss": 0.8919,
      "num_input_tokens_seen": 74190856,
      "step": 127880
    },
    {
      "epoch": 19.047512660113195,
      "grad_norm": 15.3125,
      "learning_rate": 3.447611062469269e-07,
      "loss": 1.1365,
      "num_input_tokens_seen": 74193480,
      "step": 127885
    },
    {
      "epoch": 19.048257372654156,
      "grad_norm": 12.625,
      "learning_rate": 3.442235307233449e-07,
      "loss": 1.1194,
      "num_input_tokens_seen": 74196328,
      "step": 127890
    },
    {
      "epoch": 19.049002085195113,
      "grad_norm": 12.625,
      "learning_rate": 3.4368637173404494e-07,
      "loss": 1.2408,
      "num_input_tokens_seen": 74199048,
      "step": 127895
    },
    {
      "epoch": 19.049746797736073,
      "grad_norm": 12.5625,
      "learning_rate": 3.4314962928810315e-07,
      "loss": 1.0918,
      "num_input_tokens_seen": 74201960,
      "step": 127900
    },
    {
      "epoch": 19.050491510277034,
      "grad_norm": 12.9375,
      "learning_rate": 3.426133033945872e-07,
      "loss": 1.0539,
      "num_input_tokens_seen": 74204680,
      "step": 127905
    },
    {
      "epoch": 19.05123622281799,
      "grad_norm": 13.4375,
      "learning_rate": 3.420773940625621e-07,
      "loss": 1.014,
      "num_input_tokens_seen": 74207656,
      "step": 127910
    },
    {
      "epoch": 19.05198093535895,
      "grad_norm": 8.5,
      "learning_rate": 3.415419013010762e-07,
      "loss": 0.992,
      "num_input_tokens_seen": 74210568,
      "step": 127915
    },
    {
      "epoch": 19.052725647899912,
      "grad_norm": 23.25,
      "learning_rate": 3.410068251191806e-07,
      "loss": 1.2449,
      "num_input_tokens_seen": 74213512,
      "step": 127920
    },
    {
      "epoch": 19.05347036044087,
      "grad_norm": 7.53125,
      "learning_rate": 3.4047216552590687e-07,
      "loss": 1.1702,
      "num_input_tokens_seen": 74216488,
      "step": 127925
    },
    {
      "epoch": 19.05421507298183,
      "grad_norm": 11.25,
      "learning_rate": 3.399379225302979e-07,
      "loss": 1.1337,
      "num_input_tokens_seen": 74219432,
      "step": 127930
    },
    {
      "epoch": 19.054959785522787,
      "grad_norm": 4.65625,
      "learning_rate": 3.3940409614137135e-07,
      "loss": 1.1665,
      "num_input_tokens_seen": 74221960,
      "step": 127935
    },
    {
      "epoch": 19.055704498063747,
      "grad_norm": 11.875,
      "learning_rate": 3.3887068636815346e-07,
      "loss": 0.9304,
      "num_input_tokens_seen": 74224776,
      "step": 127940
    },
    {
      "epoch": 19.056449210604708,
      "grad_norm": 21.375,
      "learning_rate": 3.3833769321964527e-07,
      "loss": 1.34,
      "num_input_tokens_seen": 74227912,
      "step": 127945
    },
    {
      "epoch": 19.057193923145665,
      "grad_norm": 15.5625,
      "learning_rate": 3.378051167048618e-07,
      "loss": 0.9828,
      "num_input_tokens_seen": 74230664,
      "step": 127950
    },
    {
      "epoch": 19.057938635686625,
      "grad_norm": 14.4375,
      "learning_rate": 3.3727295683279314e-07,
      "loss": 0.9177,
      "num_input_tokens_seen": 74233576,
      "step": 127955
    },
    {
      "epoch": 19.058683348227586,
      "grad_norm": 17.125,
      "learning_rate": 3.367412136124321e-07,
      "loss": 1.0685,
      "num_input_tokens_seen": 74236296,
      "step": 127960
    },
    {
      "epoch": 19.059428060768543,
      "grad_norm": 8.375,
      "learning_rate": 3.3620988705276023e-07,
      "loss": 1.1819,
      "num_input_tokens_seen": 74239016,
      "step": 127965
    },
    {
      "epoch": 19.060172773309503,
      "grad_norm": 7.46875,
      "learning_rate": 3.3567897716275663e-07,
      "loss": 1.1334,
      "num_input_tokens_seen": 74241992,
      "step": 127970
    },
    {
      "epoch": 19.06091748585046,
      "grad_norm": 9.625,
      "learning_rate": 3.351484839513891e-07,
      "loss": 0.8178,
      "num_input_tokens_seen": 74244584,
      "step": 127975
    },
    {
      "epoch": 19.06166219839142,
      "grad_norm": 4.625,
      "learning_rate": 3.3461840742761707e-07,
      "loss": 0.8264,
      "num_input_tokens_seen": 74247464,
      "step": 127980
    },
    {
      "epoch": 19.06240691093238,
      "grad_norm": 22.125,
      "learning_rate": 3.340887476004001e-07,
      "loss": 0.8897,
      "num_input_tokens_seen": 74250632,
      "step": 127985
    },
    {
      "epoch": 19.06315162347334,
      "grad_norm": 11.4375,
      "learning_rate": 3.3355950447868657e-07,
      "loss": 1.0714,
      "num_input_tokens_seen": 74253704,
      "step": 127990
    },
    {
      "epoch": 19.0638963360143,
      "grad_norm": 15.5,
      "learning_rate": 3.3303067807141095e-07,
      "loss": 1.5768,
      "num_input_tokens_seen": 74257000,
      "step": 127995
    },
    {
      "epoch": 19.06464104855526,
      "grad_norm": 11.875,
      "learning_rate": 3.325022683875162e-07,
      "loss": 0.9745,
      "num_input_tokens_seen": 74259720,
      "step": 128000
    },
    {
      "epoch": 19.065385761096216,
      "grad_norm": 8.3125,
      "learning_rate": 3.319742754359201e-07,
      "loss": 1.0451,
      "num_input_tokens_seen": 74262600,
      "step": 128005
    },
    {
      "epoch": 19.066130473637177,
      "grad_norm": 6.96875,
      "learning_rate": 3.314466992255516e-07,
      "loss": 1.0384,
      "num_input_tokens_seen": 74265480,
      "step": 128010
    },
    {
      "epoch": 19.066875186178134,
      "grad_norm": 5.90625,
      "learning_rate": 3.309195397653148e-07,
      "loss": 1.0265,
      "num_input_tokens_seen": 74268648,
      "step": 128015
    },
    {
      "epoch": 19.067619898719094,
      "grad_norm": 10.875,
      "learning_rate": 3.3039279706412465e-07,
      "loss": 0.9996,
      "num_input_tokens_seen": 74271656,
      "step": 128020
    },
    {
      "epoch": 19.068364611260055,
      "grad_norm": 9.25,
      "learning_rate": 3.2986647113087134e-07,
      "loss": 1.1444,
      "num_input_tokens_seen": 74274536,
      "step": 128025
    },
    {
      "epoch": 19.069109323801012,
      "grad_norm": 18.875,
      "learning_rate": 3.293405619744533e-07,
      "loss": 1.379,
      "num_input_tokens_seen": 74277256,
      "step": 128030
    },
    {
      "epoch": 19.069854036341972,
      "grad_norm": 10.3125,
      "learning_rate": 3.288150696037523e-07,
      "loss": 1.319,
      "num_input_tokens_seen": 74280040,
      "step": 128035
    },
    {
      "epoch": 19.070598748882933,
      "grad_norm": 20.125,
      "learning_rate": 3.282899940276418e-07,
      "loss": 1.044,
      "num_input_tokens_seen": 74282952,
      "step": 128040
    },
    {
      "epoch": 19.07134346142389,
      "grad_norm": 9.0625,
      "learning_rate": 3.2776533525500085e-07,
      "loss": 0.6879,
      "num_input_tokens_seen": 74285800,
      "step": 128045
    },
    {
      "epoch": 19.07208817396485,
      "grad_norm": 11.5625,
      "learning_rate": 3.272410932946862e-07,
      "loss": 0.9971,
      "num_input_tokens_seen": 74288616,
      "step": 128050
    },
    {
      "epoch": 19.072832886505807,
      "grad_norm": 8.25,
      "learning_rate": 3.267172681555575e-07,
      "loss": 0.9868,
      "num_input_tokens_seen": 74291720,
      "step": 128055
    },
    {
      "epoch": 19.073577599046768,
      "grad_norm": 16.5,
      "learning_rate": 3.261938598464631e-07,
      "loss": 1.1145,
      "num_input_tokens_seen": 74294664,
      "step": 128060
    },
    {
      "epoch": 19.07432231158773,
      "grad_norm": 6.125,
      "learning_rate": 3.256708683762488e-07,
      "loss": 1.0362,
      "num_input_tokens_seen": 74297480,
      "step": 128065
    },
    {
      "epoch": 19.075067024128685,
      "grad_norm": 7.9375,
      "learning_rate": 3.2514829375374643e-07,
      "loss": 1.0385,
      "num_input_tokens_seen": 74300328,
      "step": 128070
    },
    {
      "epoch": 19.075811736669646,
      "grad_norm": 12.875,
      "learning_rate": 3.24626135987785e-07,
      "loss": 1.0344,
      "num_input_tokens_seen": 74303048,
      "step": 128075
    },
    {
      "epoch": 19.076556449210603,
      "grad_norm": 10.0,
      "learning_rate": 3.2410439508718527e-07,
      "loss": 0.9373,
      "num_input_tokens_seen": 74305736,
      "step": 128080
    },
    {
      "epoch": 19.077301161751564,
      "grad_norm": 10.9375,
      "learning_rate": 3.2358307106076234e-07,
      "loss": 1.4233,
      "num_input_tokens_seen": 74308488,
      "step": 128085
    },
    {
      "epoch": 19.078045874292524,
      "grad_norm": 13.4375,
      "learning_rate": 3.2306216391732593e-07,
      "loss": 1.0928,
      "num_input_tokens_seen": 74311144,
      "step": 128090
    },
    {
      "epoch": 19.07879058683348,
      "grad_norm": 9.1875,
      "learning_rate": 3.225416736656689e-07,
      "loss": 0.8755,
      "num_input_tokens_seen": 74313864,
      "step": 128095
    },
    {
      "epoch": 19.07953529937444,
      "grad_norm": 12.5,
      "learning_rate": 3.220216003145926e-07,
      "loss": 1.3626,
      "num_input_tokens_seen": 74316904,
      "step": 128100
    },
    {
      "epoch": 19.080280011915402,
      "grad_norm": 11.25,
      "learning_rate": 3.215019438728789e-07,
      "loss": 0.7529,
      "num_input_tokens_seen": 74319464,
      "step": 128105
    },
    {
      "epoch": 19.08102472445636,
      "grad_norm": 11.9375,
      "learning_rate": 3.209827043493097e-07,
      "loss": 0.9434,
      "num_input_tokens_seen": 74322216,
      "step": 128110
    },
    {
      "epoch": 19.08176943699732,
      "grad_norm": 9.9375,
      "learning_rate": 3.204638817526528e-07,
      "loss": 1.0296,
      "num_input_tokens_seen": 74325064,
      "step": 128115
    },
    {
      "epoch": 19.082514149538277,
      "grad_norm": 16.75,
      "learning_rate": 3.1994547609167644e-07,
      "loss": 1.2033,
      "num_input_tokens_seen": 74327880,
      "step": 128120
    },
    {
      "epoch": 19.083258862079237,
      "grad_norm": 18.25,
      "learning_rate": 3.194274873751374e-07,
      "loss": 1.1212,
      "num_input_tokens_seen": 74330888,
      "step": 128125
    },
    {
      "epoch": 19.084003574620198,
      "grad_norm": 8.6875,
      "learning_rate": 3.189099156117842e-07,
      "loss": 0.9382,
      "num_input_tokens_seen": 74333704,
      "step": 128130
    },
    {
      "epoch": 19.084748287161155,
      "grad_norm": 16.25,
      "learning_rate": 3.1839276081036816e-07,
      "loss": 1.1211,
      "num_input_tokens_seen": 74336424,
      "step": 128135
    },
    {
      "epoch": 19.085492999702115,
      "grad_norm": 6.28125,
      "learning_rate": 3.1787602297961574e-07,
      "loss": 1.2891,
      "num_input_tokens_seen": 74339272,
      "step": 128140
    },
    {
      "epoch": 19.086237712243076,
      "grad_norm": 15.375,
      "learning_rate": 3.1735970212826705e-07,
      "loss": 1.0831,
      "num_input_tokens_seen": 74342280,
      "step": 128145
    },
    {
      "epoch": 19.086982424784033,
      "grad_norm": 21.375,
      "learning_rate": 3.168437982650374e-07,
      "loss": 1.3555,
      "num_input_tokens_seen": 74345224,
      "step": 128150
    },
    {
      "epoch": 19.087727137324993,
      "grad_norm": 15.375,
      "learning_rate": 3.1632831139864763e-07,
      "loss": 1.031,
      "num_input_tokens_seen": 74348040,
      "step": 128155
    },
    {
      "epoch": 19.08847184986595,
      "grad_norm": 6.90625,
      "learning_rate": 3.158132415378018e-07,
      "loss": 0.8463,
      "num_input_tokens_seen": 74350760,
      "step": 128160
    },
    {
      "epoch": 19.08921656240691,
      "grad_norm": 12.4375,
      "learning_rate": 3.152985886912013e-07,
      "loss": 1.2683,
      "num_input_tokens_seen": 74353672,
      "step": 128165
    },
    {
      "epoch": 19.08996127494787,
      "grad_norm": 8.4375,
      "learning_rate": 3.1478435286754483e-07,
      "loss": 1.2349,
      "num_input_tokens_seen": 74356424,
      "step": 128170
    },
    {
      "epoch": 19.09070598748883,
      "grad_norm": 5.84375,
      "learning_rate": 3.14270534075517e-07,
      "loss": 0.9216,
      "num_input_tokens_seen": 74359304,
      "step": 128175
    },
    {
      "epoch": 19.09145070002979,
      "grad_norm": 13.75,
      "learning_rate": 3.137571323237998e-07,
      "loss": 1.2335,
      "num_input_tokens_seen": 74362120,
      "step": 128180
    },
    {
      "epoch": 19.09219541257075,
      "grad_norm": 20.625,
      "learning_rate": 3.13244147621064e-07,
      "loss": 1.5193,
      "num_input_tokens_seen": 74364872,
      "step": 128185
    },
    {
      "epoch": 19.092940125111706,
      "grad_norm": 12.9375,
      "learning_rate": 3.1273157997598056e-07,
      "loss": 1.4521,
      "num_input_tokens_seen": 74367624,
      "step": 128190
    },
    {
      "epoch": 19.093684837652667,
      "grad_norm": 7.28125,
      "learning_rate": 3.122194293972064e-07,
      "loss": 1.1856,
      "num_input_tokens_seen": 74370440,
      "step": 128195
    },
    {
      "epoch": 19.094429550193624,
      "grad_norm": 25.25,
      "learning_rate": 3.117076958933901e-07,
      "loss": 1.6586,
      "num_input_tokens_seen": 74373320,
      "step": 128200
    },
    {
      "epoch": 19.095174262734584,
      "grad_norm": 13.4375,
      "learning_rate": 3.111963794731831e-07,
      "loss": 1.2278,
      "num_input_tokens_seen": 74376520,
      "step": 128205
    },
    {
      "epoch": 19.095918975275545,
      "grad_norm": 10.9375,
      "learning_rate": 3.106854801452175e-07,
      "loss": 1.0883,
      "num_input_tokens_seen": 74379272,
      "step": 128210
    },
    {
      "epoch": 19.096663687816502,
      "grad_norm": 13.1875,
      "learning_rate": 3.1017499791813067e-07,
      "loss": 1.0562,
      "num_input_tokens_seen": 74382088,
      "step": 128215
    },
    {
      "epoch": 19.097408400357462,
      "grad_norm": 10.75,
      "learning_rate": 3.096649328005435e-07,
      "loss": 1.4047,
      "num_input_tokens_seen": 74384968,
      "step": 128220
    },
    {
      "epoch": 19.098153112898423,
      "grad_norm": 10.0625,
      "learning_rate": 3.091552848010715e-07,
      "loss": 1.2821,
      "num_input_tokens_seen": 74388072,
      "step": 128225
    },
    {
      "epoch": 19.09889782543938,
      "grad_norm": 16.25,
      "learning_rate": 3.08646053928327e-07,
      "loss": 1.3449,
      "num_input_tokens_seen": 74390760,
      "step": 128230
    },
    {
      "epoch": 19.09964253798034,
      "grad_norm": 11.8125,
      "learning_rate": 3.081372401909116e-07,
      "loss": 1.2168,
      "num_input_tokens_seen": 74393448,
      "step": 128235
    },
    {
      "epoch": 19.100387250521297,
      "grad_norm": 4.3125,
      "learning_rate": 3.076288435974239e-07,
      "loss": 0.9249,
      "num_input_tokens_seen": 74396296,
      "step": 128240
    },
    {
      "epoch": 19.101131963062258,
      "grad_norm": 17.875,
      "learning_rate": 3.071208641564488e-07,
      "loss": 0.8253,
      "num_input_tokens_seen": 74399432,
      "step": 128245
    },
    {
      "epoch": 19.10187667560322,
      "grad_norm": 7.0625,
      "learning_rate": 3.06613301876571e-07,
      "loss": 0.8799,
      "num_input_tokens_seen": 74402184,
      "step": 128250
    },
    {
      "epoch": 19.102621388144176,
      "grad_norm": 9.3125,
      "learning_rate": 3.0610615676636144e-07,
      "loss": 1.1236,
      "num_input_tokens_seen": 74404904,
      "step": 128255
    },
    {
      "epoch": 19.103366100685136,
      "grad_norm": 10.5,
      "learning_rate": 3.0559942883439387e-07,
      "loss": 1.1058,
      "num_input_tokens_seen": 74407848,
      "step": 128260
    },
    {
      "epoch": 19.104110813226093,
      "grad_norm": 8.6875,
      "learning_rate": 3.0509311808922526e-07,
      "loss": 1.0218,
      "num_input_tokens_seen": 74410664,
      "step": 128265
    },
    {
      "epoch": 19.104855525767054,
      "grad_norm": 7.03125,
      "learning_rate": 3.045872245394099e-07,
      "loss": 0.9225,
      "num_input_tokens_seen": 74413544,
      "step": 128270
    },
    {
      "epoch": 19.105600238308014,
      "grad_norm": 9.8125,
      "learning_rate": 3.0408174819349377e-07,
      "loss": 1.2053,
      "num_input_tokens_seen": 74416872,
      "step": 128275
    },
    {
      "epoch": 19.10634495084897,
      "grad_norm": 9.125,
      "learning_rate": 3.035766890600145e-07,
      "loss": 1.1143,
      "num_input_tokens_seen": 74419720,
      "step": 128280
    },
    {
      "epoch": 19.10708966338993,
      "grad_norm": 6.40625,
      "learning_rate": 3.030720471475096e-07,
      "loss": 1.4046,
      "num_input_tokens_seen": 74422664,
      "step": 128285
    },
    {
      "epoch": 19.107834375930892,
      "grad_norm": 6.5625,
      "learning_rate": 3.025678224645001e-07,
      "loss": 1.0815,
      "num_input_tokens_seen": 74425544,
      "step": 128290
    },
    {
      "epoch": 19.10857908847185,
      "grad_norm": 10.5,
      "learning_rate": 3.02064015019507e-07,
      "loss": 1.125,
      "num_input_tokens_seen": 74427880,
      "step": 128295
    },
    {
      "epoch": 19.10932380101281,
      "grad_norm": 11.6875,
      "learning_rate": 3.015606248210401e-07,
      "loss": 0.8888,
      "num_input_tokens_seen": 74430568,
      "step": 128300
    },
    {
      "epoch": 19.110068513553767,
      "grad_norm": 8.9375,
      "learning_rate": 3.010576518776037e-07,
      "loss": 1.2295,
      "num_input_tokens_seen": 74433416,
      "step": 128305
    },
    {
      "epoch": 19.110813226094727,
      "grad_norm": 4.46875,
      "learning_rate": 3.005550961976938e-07,
      "loss": 1.1549,
      "num_input_tokens_seen": 74436136,
      "step": 128310
    },
    {
      "epoch": 19.111557938635688,
      "grad_norm": 13.625,
      "learning_rate": 3.0005295778980647e-07,
      "loss": 1.2926,
      "num_input_tokens_seen": 74438824,
      "step": 128315
    },
    {
      "epoch": 19.112302651176645,
      "grad_norm": 5.5,
      "learning_rate": 2.9955123666241814e-07,
      "loss": 0.9639,
      "num_input_tokens_seen": 74441896,
      "step": 128320
    },
    {
      "epoch": 19.113047363717605,
      "grad_norm": 10.8125,
      "learning_rate": 2.990499328240054e-07,
      "loss": 0.8246,
      "num_input_tokens_seen": 74444840,
      "step": 128325
    },
    {
      "epoch": 19.113792076258566,
      "grad_norm": 6.375,
      "learning_rate": 2.9854904628304206e-07,
      "loss": 1.1099,
      "num_input_tokens_seen": 74447720,
      "step": 128330
    },
    {
      "epoch": 19.114536788799523,
      "grad_norm": 10.0,
      "learning_rate": 2.980485770479824e-07,
      "loss": 1.0479,
      "num_input_tokens_seen": 74450536,
      "step": 128335
    },
    {
      "epoch": 19.115281501340483,
      "grad_norm": 12.0,
      "learning_rate": 2.975485251272919e-07,
      "loss": 1.0116,
      "num_input_tokens_seen": 74453512,
      "step": 128340
    },
    {
      "epoch": 19.11602621388144,
      "grad_norm": 9.625,
      "learning_rate": 2.970488905294083e-07,
      "loss": 1.1529,
      "num_input_tokens_seen": 74456296,
      "step": 128345
    },
    {
      "epoch": 19.1167709264224,
      "grad_norm": 15.1875,
      "learning_rate": 2.965496732627804e-07,
      "loss": 0.9474,
      "num_input_tokens_seen": 74459464,
      "step": 128350
    },
    {
      "epoch": 19.11751563896336,
      "grad_norm": 16.0,
      "learning_rate": 2.960508733358375e-07,
      "loss": 1.4723,
      "num_input_tokens_seen": 74462376,
      "step": 128355
    },
    {
      "epoch": 19.11826035150432,
      "grad_norm": 6.96875,
      "learning_rate": 2.955524907570062e-07,
      "loss": 0.976,
      "num_input_tokens_seen": 74465352,
      "step": 128360
    },
    {
      "epoch": 19.11900506404528,
      "grad_norm": 14.5,
      "learning_rate": 2.950545255347076e-07,
      "loss": 1.5513,
      "num_input_tokens_seen": 74468136,
      "step": 128365
    },
    {
      "epoch": 19.11974977658624,
      "grad_norm": 8.5625,
      "learning_rate": 2.9455697767735155e-07,
      "loss": 0.749,
      "num_input_tokens_seen": 74470856,
      "step": 128370
    },
    {
      "epoch": 19.120494489127196,
      "grad_norm": 6.40625,
      "learning_rate": 2.9405984719334814e-07,
      "loss": 0.879,
      "num_input_tokens_seen": 74473960,
      "step": 128375
    },
    {
      "epoch": 19.121239201668157,
      "grad_norm": 14.0625,
      "learning_rate": 2.935631340910933e-07,
      "loss": 1.0448,
      "num_input_tokens_seen": 74476872,
      "step": 128380
    },
    {
      "epoch": 19.121983914209114,
      "grad_norm": 15.8125,
      "learning_rate": 2.930668383789775e-07,
      "loss": 1.0999,
      "num_input_tokens_seen": 74479688,
      "step": 128385
    },
    {
      "epoch": 19.122728626750074,
      "grad_norm": 17.875,
      "learning_rate": 2.925709600653859e-07,
      "loss": 0.9805,
      "num_input_tokens_seen": 74482536,
      "step": 128390
    },
    {
      "epoch": 19.123473339291035,
      "grad_norm": 9.6875,
      "learning_rate": 2.9207549915870045e-07,
      "loss": 1.1566,
      "num_input_tokens_seen": 74485384,
      "step": 128395
    },
    {
      "epoch": 19.124218051831992,
      "grad_norm": 13.4375,
      "learning_rate": 2.915804556672841e-07,
      "loss": 0.8381,
      "num_input_tokens_seen": 74488392,
      "step": 128400
    },
    {
      "epoch": 19.124962764372953,
      "grad_norm": 7.84375,
      "learning_rate": 2.9108582959950504e-07,
      "loss": 1.2154,
      "num_input_tokens_seen": 74491144,
      "step": 128405
    },
    {
      "epoch": 19.12570747691391,
      "grad_norm": 9.8125,
      "learning_rate": 2.9059162096371773e-07,
      "loss": 1.1394,
      "num_input_tokens_seen": 74494056,
      "step": 128410
    },
    {
      "epoch": 19.12645218945487,
      "grad_norm": 5.40625,
      "learning_rate": 2.9009782976827106e-07,
      "loss": 1.2863,
      "num_input_tokens_seen": 74496968,
      "step": 128415
    },
    {
      "epoch": 19.12719690199583,
      "grad_norm": 8.75,
      "learning_rate": 2.896044560215083e-07,
      "loss": 0.8418,
      "num_input_tokens_seen": 74499688,
      "step": 128420
    },
    {
      "epoch": 19.127941614536788,
      "grad_norm": 8.3125,
      "learning_rate": 2.891114997317618e-07,
      "loss": 0.981,
      "num_input_tokens_seen": 74502536,
      "step": 128425
    },
    {
      "epoch": 19.128686327077748,
      "grad_norm": 10.4375,
      "learning_rate": 2.8861896090736365e-07,
      "loss": 1.4255,
      "num_input_tokens_seen": 74505640,
      "step": 128430
    },
    {
      "epoch": 19.12943103961871,
      "grad_norm": 6.59375,
      "learning_rate": 2.881268395566322e-07,
      "loss": 1.1717,
      "num_input_tokens_seen": 74508520,
      "step": 128435
    },
    {
      "epoch": 19.130175752159666,
      "grad_norm": 10.75,
      "learning_rate": 2.8763513568788036e-07,
      "loss": 1.1604,
      "num_input_tokens_seen": 74511624,
      "step": 128440
    },
    {
      "epoch": 19.130920464700626,
      "grad_norm": 14.5625,
      "learning_rate": 2.871438493094153e-07,
      "loss": 0.9072,
      "num_input_tokens_seen": 74514632,
      "step": 128445
    },
    {
      "epoch": 19.131665177241583,
      "grad_norm": 12.25,
      "learning_rate": 2.866529804295387e-07,
      "loss": 1.1095,
      "num_input_tokens_seen": 74517960,
      "step": 128450
    },
    {
      "epoch": 19.132409889782544,
      "grad_norm": 12.8125,
      "learning_rate": 2.8616252905654393e-07,
      "loss": 1.0815,
      "num_input_tokens_seen": 74520776,
      "step": 128455
    },
    {
      "epoch": 19.133154602323504,
      "grad_norm": 11.6875,
      "learning_rate": 2.856724951987161e-07,
      "loss": 1.2705,
      "num_input_tokens_seen": 74523720,
      "step": 128460
    },
    {
      "epoch": 19.13389931486446,
      "grad_norm": 11.4375,
      "learning_rate": 2.851828788643318e-07,
      "loss": 1.1007,
      "num_input_tokens_seen": 74526696,
      "step": 128465
    },
    {
      "epoch": 19.13464402740542,
      "grad_norm": 4.65625,
      "learning_rate": 2.846936800616623e-07,
      "loss": 1.0071,
      "num_input_tokens_seen": 74529576,
      "step": 128470
    },
    {
      "epoch": 19.135388739946382,
      "grad_norm": 17.375,
      "learning_rate": 2.8420489879897595e-07,
      "loss": 0.9565,
      "num_input_tokens_seen": 74532648,
      "step": 128475
    },
    {
      "epoch": 19.13613345248734,
      "grad_norm": 8.9375,
      "learning_rate": 2.8371653508452725e-07,
      "loss": 0.9364,
      "num_input_tokens_seen": 74535656,
      "step": 128480
    },
    {
      "epoch": 19.1368781650283,
      "grad_norm": 8.3125,
      "learning_rate": 2.832285889265651e-07,
      "loss": 0.9919,
      "num_input_tokens_seen": 74538600,
      "step": 128485
    },
    {
      "epoch": 19.137622877569257,
      "grad_norm": 11.0,
      "learning_rate": 2.827410603333386e-07,
      "loss": 0.8516,
      "num_input_tokens_seen": 74541384,
      "step": 128490
    },
    {
      "epoch": 19.138367590110217,
      "grad_norm": 9.375,
      "learning_rate": 2.8225394931307715e-07,
      "loss": 1.0975,
      "num_input_tokens_seen": 74544136,
      "step": 128495
    },
    {
      "epoch": 19.139112302651178,
      "grad_norm": 3.96875,
      "learning_rate": 2.817672558740131e-07,
      "loss": 1.0753,
      "num_input_tokens_seen": 74547048,
      "step": 128500
    },
    {
      "epoch": 19.139857015192135,
      "grad_norm": 12.0,
      "learning_rate": 2.812809800243704e-07,
      "loss": 1.3858,
      "num_input_tokens_seen": 74549928,
      "step": 128505
    },
    {
      "epoch": 19.140601727733095,
      "grad_norm": 8.0,
      "learning_rate": 2.80795121772362e-07,
      "loss": 1.2937,
      "num_input_tokens_seen": 74553064,
      "step": 128510
    },
    {
      "epoch": 19.141346440274056,
      "grad_norm": 8.375,
      "learning_rate": 2.803096811261979e-07,
      "loss": 0.9374,
      "num_input_tokens_seen": 74556072,
      "step": 128515
    },
    {
      "epoch": 19.142091152815013,
      "grad_norm": 7.9375,
      "learning_rate": 2.7982465809407443e-07,
      "loss": 1.0851,
      "num_input_tokens_seen": 74559016,
      "step": 128520
    },
    {
      "epoch": 19.142835865355973,
      "grad_norm": 7.1875,
      "learning_rate": 2.793400526841933e-07,
      "loss": 0.7963,
      "num_input_tokens_seen": 74561672,
      "step": 128525
    },
    {
      "epoch": 19.14358057789693,
      "grad_norm": 11.6875,
      "learning_rate": 2.7885586490473127e-07,
      "loss": 1.0214,
      "num_input_tokens_seen": 74564712,
      "step": 128530
    },
    {
      "epoch": 19.14432529043789,
      "grad_norm": 5.375,
      "learning_rate": 2.7837209476387903e-07,
      "loss": 0.9774,
      "num_input_tokens_seen": 74567624,
      "step": 128535
    },
    {
      "epoch": 19.14507000297885,
      "grad_norm": 13.3125,
      "learning_rate": 2.7788874226980233e-07,
      "loss": 0.98,
      "num_input_tokens_seen": 74570344,
      "step": 128540
    },
    {
      "epoch": 19.14581471551981,
      "grad_norm": 7.25,
      "learning_rate": 2.774058074306696e-07,
      "loss": 1.0523,
      "num_input_tokens_seen": 74573160,
      "step": 128545
    },
    {
      "epoch": 19.14655942806077,
      "grad_norm": 15.3125,
      "learning_rate": 2.7692329025463816e-07,
      "loss": 1.2169,
      "num_input_tokens_seen": 74576072,
      "step": 128550
    },
    {
      "epoch": 19.14730414060173,
      "grad_norm": 10.5,
      "learning_rate": 2.7644119074986263e-07,
      "loss": 1.0228,
      "num_input_tokens_seen": 74578760,
      "step": 128555
    },
    {
      "epoch": 19.148048853142686,
      "grad_norm": 15.0,
      "learning_rate": 2.7595950892448374e-07,
      "loss": 1.0818,
      "num_input_tokens_seen": 74581768,
      "step": 128560
    },
    {
      "epoch": 19.148793565683647,
      "grad_norm": 8.875,
      "learning_rate": 2.754782447866394e-07,
      "loss": 1.0668,
      "num_input_tokens_seen": 74584712,
      "step": 128565
    },
    {
      "epoch": 19.149538278224604,
      "grad_norm": 10.5625,
      "learning_rate": 2.7499739834446204e-07,
      "loss": 1.0335,
      "num_input_tokens_seen": 74587720,
      "step": 128570
    },
    {
      "epoch": 19.150282990765565,
      "grad_norm": 7.21875,
      "learning_rate": 2.745169696060729e-07,
      "loss": 0.93,
      "num_input_tokens_seen": 74590632,
      "step": 128575
    },
    {
      "epoch": 19.151027703306525,
      "grad_norm": 14.125,
      "learning_rate": 2.7403695857959046e-07,
      "loss": 1.4398,
      "num_input_tokens_seen": 74593640,
      "step": 128580
    },
    {
      "epoch": 19.151772415847482,
      "grad_norm": 11.125,
      "learning_rate": 2.735573652731249e-07,
      "loss": 1.0929,
      "num_input_tokens_seen": 74596712,
      "step": 128585
    },
    {
      "epoch": 19.152517128388443,
      "grad_norm": 17.0,
      "learning_rate": 2.730781896947754e-07,
      "loss": 1.4989,
      "num_input_tokens_seen": 74599944,
      "step": 128590
    },
    {
      "epoch": 19.1532618409294,
      "grad_norm": 7.375,
      "learning_rate": 2.7259943185263813e-07,
      "loss": 1.0242,
      "num_input_tokens_seen": 74603080,
      "step": 128595
    },
    {
      "epoch": 19.15400655347036,
      "grad_norm": 17.25,
      "learning_rate": 2.7212109175480114e-07,
      "loss": 1.2895,
      "num_input_tokens_seen": 74606088,
      "step": 128600
    },
    {
      "epoch": 19.15475126601132,
      "grad_norm": 9.6875,
      "learning_rate": 2.7164316940934966e-07,
      "loss": 1.2375,
      "num_input_tokens_seen": 74609096,
      "step": 128605
    },
    {
      "epoch": 19.155495978552278,
      "grad_norm": 10.9375,
      "learning_rate": 2.7116566482434936e-07,
      "loss": 1.4349,
      "num_input_tokens_seen": 74612072,
      "step": 128610
    },
    {
      "epoch": 19.156240691093238,
      "grad_norm": 9.875,
      "learning_rate": 2.706885780078744e-07,
      "loss": 0.9308,
      "num_input_tokens_seen": 74614952,
      "step": 128615
    },
    {
      "epoch": 19.1569854036342,
      "grad_norm": 6.8125,
      "learning_rate": 2.7021190896798223e-07,
      "loss": 0.9088,
      "num_input_tokens_seen": 74617960,
      "step": 128620
    },
    {
      "epoch": 19.157730116175156,
      "grad_norm": 21.875,
      "learning_rate": 2.6973565771272746e-07,
      "loss": 1.2284,
      "num_input_tokens_seen": 74620872,
      "step": 128625
    },
    {
      "epoch": 19.158474828716116,
      "grad_norm": 9.3125,
      "learning_rate": 2.6925982425015097e-07,
      "loss": 0.9255,
      "num_input_tokens_seen": 74623848,
      "step": 128630
    },
    {
      "epoch": 19.159219541257073,
      "grad_norm": 7.34375,
      "learning_rate": 2.6878440858829626e-07,
      "loss": 1.0175,
      "num_input_tokens_seen": 74626856,
      "step": 128635
    },
    {
      "epoch": 19.159964253798034,
      "grad_norm": 12.875,
      "learning_rate": 2.68309410735193e-07,
      "loss": 0.909,
      "num_input_tokens_seen": 74629352,
      "step": 128640
    },
    {
      "epoch": 19.160708966338994,
      "grad_norm": 6.5,
      "learning_rate": 2.678348306988626e-07,
      "loss": 1.0148,
      "num_input_tokens_seen": 74632072,
      "step": 128645
    },
    {
      "epoch": 19.16145367887995,
      "grad_norm": 5.375,
      "learning_rate": 2.67360668487332e-07,
      "loss": 1.1369,
      "num_input_tokens_seen": 74634760,
      "step": 128650
    },
    {
      "epoch": 19.16219839142091,
      "grad_norm": 9.125,
      "learning_rate": 2.6688692410860025e-07,
      "loss": 1.1063,
      "num_input_tokens_seen": 74637672,
      "step": 128655
    },
    {
      "epoch": 19.162943103961872,
      "grad_norm": 11.125,
      "learning_rate": 2.664135975706805e-07,
      "loss": 1.2394,
      "num_input_tokens_seen": 74640680,
      "step": 128660
    },
    {
      "epoch": 19.16368781650283,
      "grad_norm": 8.125,
      "learning_rate": 2.659406888815608e-07,
      "loss": 1.0382,
      "num_input_tokens_seen": 74643464,
      "step": 128665
    },
    {
      "epoch": 19.16443252904379,
      "grad_norm": 7.5625,
      "learning_rate": 2.6546819804923737e-07,
      "loss": 0.9175,
      "num_input_tokens_seen": 74646280,
      "step": 128670
    },
    {
      "epoch": 19.165177241584747,
      "grad_norm": 14.5,
      "learning_rate": 2.6499612508169016e-07,
      "loss": 1.2679,
      "num_input_tokens_seen": 74649224,
      "step": 128675
    },
    {
      "epoch": 19.165921954125707,
      "grad_norm": 8.5625,
      "learning_rate": 2.645244699868932e-07,
      "loss": 0.9814,
      "num_input_tokens_seen": 74651912,
      "step": 128680
    },
    {
      "epoch": 19.166666666666668,
      "grad_norm": 16.375,
      "learning_rate": 2.6405323277281514e-07,
      "loss": 1.098,
      "num_input_tokens_seen": 74654824,
      "step": 128685
    },
    {
      "epoch": 19.167411379207625,
      "grad_norm": 5.28125,
      "learning_rate": 2.6358241344741906e-07,
      "loss": 1.2134,
      "num_input_tokens_seen": 74657544,
      "step": 128690
    },
    {
      "epoch": 19.168156091748585,
      "grad_norm": 9.5,
      "learning_rate": 2.6311201201865423e-07,
      "loss": 1.141,
      "num_input_tokens_seen": 74660392,
      "step": 128695
    },
    {
      "epoch": 19.168900804289546,
      "grad_norm": 9.4375,
      "learning_rate": 2.626420284944725e-07,
      "loss": 0.9387,
      "num_input_tokens_seen": 74663368,
      "step": 128700
    },
    {
      "epoch": 19.169645516830503,
      "grad_norm": 8.875,
      "learning_rate": 2.6217246288281205e-07,
      "loss": 0.9393,
      "num_input_tokens_seen": 74666312,
      "step": 128705
    },
    {
      "epoch": 19.170390229371463,
      "grad_norm": 7.96875,
      "learning_rate": 2.6170331519160264e-07,
      "loss": 1.075,
      "num_input_tokens_seen": 74669256,
      "step": 128710
    },
    {
      "epoch": 19.17113494191242,
      "grad_norm": 8.3125,
      "learning_rate": 2.61234585428774e-07,
      "loss": 0.9249,
      "num_input_tokens_seen": 74672168,
      "step": 128715
    },
    {
      "epoch": 19.17187965445338,
      "grad_norm": 4.40625,
      "learning_rate": 2.607662736022448e-07,
      "loss": 1.3448,
      "num_input_tokens_seen": 74675176,
      "step": 128720
    },
    {
      "epoch": 19.17262436699434,
      "grad_norm": 5.96875,
      "learning_rate": 2.6029837971992545e-07,
      "loss": 0.8812,
      "num_input_tokens_seen": 74678216,
      "step": 128725
    },
    {
      "epoch": 19.1733690795353,
      "grad_norm": 7.59375,
      "learning_rate": 2.5983090378972064e-07,
      "loss": 0.9289,
      "num_input_tokens_seen": 74681256,
      "step": 128730
    },
    {
      "epoch": 19.17411379207626,
      "grad_norm": 13.875,
      "learning_rate": 2.5936384581952686e-07,
      "loss": 1.101,
      "num_input_tokens_seen": 74684136,
      "step": 128735
    },
    {
      "epoch": 19.17485850461722,
      "grad_norm": 15.5,
      "learning_rate": 2.5889720581723506e-07,
      "loss": 1.3625,
      "num_input_tokens_seen": 74687496,
      "step": 128740
    },
    {
      "epoch": 19.175603217158177,
      "grad_norm": 13.3125,
      "learning_rate": 2.584309837907306e-07,
      "loss": 1.235,
      "num_input_tokens_seen": 74690312,
      "step": 128745
    },
    {
      "epoch": 19.176347929699137,
      "grad_norm": 21.375,
      "learning_rate": 2.5796517974789045e-07,
      "loss": 1.8519,
      "num_input_tokens_seen": 74693256,
      "step": 128750
    },
    {
      "epoch": 19.177092642240094,
      "grad_norm": 18.875,
      "learning_rate": 2.5749979369657783e-07,
      "loss": 1.0197,
      "num_input_tokens_seen": 74696392,
      "step": 128755
    },
    {
      "epoch": 19.177837354781055,
      "grad_norm": 6.78125,
      "learning_rate": 2.570348256446614e-07,
      "loss": 0.8236,
      "num_input_tokens_seen": 74699432,
      "step": 128760
    },
    {
      "epoch": 19.178582067322015,
      "grad_norm": 7.59375,
      "learning_rate": 2.5657027559999327e-07,
      "loss": 1.3683,
      "num_input_tokens_seen": 74702376,
      "step": 128765
    },
    {
      "epoch": 19.179326779862972,
      "grad_norm": 8.75,
      "learning_rate": 2.561061435704226e-07,
      "loss": 0.8965,
      "num_input_tokens_seen": 74705576,
      "step": 128770
    },
    {
      "epoch": 19.180071492403933,
      "grad_norm": 7.78125,
      "learning_rate": 2.556424295637905e-07,
      "loss": 0.9058,
      "num_input_tokens_seen": 74708552,
      "step": 128775
    },
    {
      "epoch": 19.18081620494489,
      "grad_norm": 6.40625,
      "learning_rate": 2.5517913358792945e-07,
      "loss": 1.0254,
      "num_input_tokens_seen": 74711336,
      "step": 128780
    },
    {
      "epoch": 19.18156091748585,
      "grad_norm": 9.25,
      "learning_rate": 2.547162556506694e-07,
      "loss": 1.2081,
      "num_input_tokens_seen": 74714280,
      "step": 128785
    },
    {
      "epoch": 19.18230563002681,
      "grad_norm": 9.375,
      "learning_rate": 2.5425379575982343e-07,
      "loss": 0.8862,
      "num_input_tokens_seen": 74717224,
      "step": 128790
    },
    {
      "epoch": 19.183050342567768,
      "grad_norm": 6.75,
      "learning_rate": 2.537917539232132e-07,
      "loss": 1.0572,
      "num_input_tokens_seen": 74719880,
      "step": 128795
    },
    {
      "epoch": 19.183795055108728,
      "grad_norm": 10.5,
      "learning_rate": 2.5333013014864073e-07,
      "loss": 1.0813,
      "num_input_tokens_seen": 74722760,
      "step": 128800
    },
    {
      "epoch": 19.18453976764969,
      "grad_norm": 12.625,
      "learning_rate": 2.528689244439025e-07,
      "loss": 1.2122,
      "num_input_tokens_seen": 74725512,
      "step": 128805
    },
    {
      "epoch": 19.185284480190646,
      "grad_norm": 5.0,
      "learning_rate": 2.524081368167924e-07,
      "loss": 1.1095,
      "num_input_tokens_seen": 74728232,
      "step": 128810
    },
    {
      "epoch": 19.186029192731606,
      "grad_norm": 18.5,
      "learning_rate": 2.5194776727509584e-07,
      "loss": 1.0785,
      "num_input_tokens_seen": 74730952,
      "step": 128815
    },
    {
      "epoch": 19.186773905272563,
      "grad_norm": 13.5625,
      "learning_rate": 2.5148781582658986e-07,
      "loss": 1.2272,
      "num_input_tokens_seen": 74733800,
      "step": 128820
    },
    {
      "epoch": 19.187518617813524,
      "grad_norm": 13.0,
      "learning_rate": 2.5102828247904055e-07,
      "loss": 1.3959,
      "num_input_tokens_seen": 74736712,
      "step": 128825
    },
    {
      "epoch": 19.188263330354484,
      "grad_norm": 13.25,
      "learning_rate": 2.5056916724021663e-07,
      "loss": 1.3309,
      "num_input_tokens_seen": 74739464,
      "step": 128830
    },
    {
      "epoch": 19.18900804289544,
      "grad_norm": 14.9375,
      "learning_rate": 2.5011047011787026e-07,
      "loss": 1.3051,
      "num_input_tokens_seen": 74742088,
      "step": 128835
    },
    {
      "epoch": 19.189752755436402,
      "grad_norm": 20.5,
      "learning_rate": 2.4965219111975635e-07,
      "loss": 1.5081,
      "num_input_tokens_seen": 74745064,
      "step": 128840
    },
    {
      "epoch": 19.190497467977362,
      "grad_norm": 6.78125,
      "learning_rate": 2.491943302536104e-07,
      "loss": 0.6096,
      "num_input_tokens_seen": 74747848,
      "step": 128845
    },
    {
      "epoch": 19.19124218051832,
      "grad_norm": 5.5625,
      "learning_rate": 2.487368875271706e-07,
      "loss": 1.1258,
      "num_input_tokens_seen": 74750536,
      "step": 128850
    },
    {
      "epoch": 19.19198689305928,
      "grad_norm": 9.375,
      "learning_rate": 2.4827986294816696e-07,
      "loss": 1.5498,
      "num_input_tokens_seen": 74753320,
      "step": 128855
    },
    {
      "epoch": 19.192731605600237,
      "grad_norm": 23.75,
      "learning_rate": 2.478232565243183e-07,
      "loss": 1.3378,
      "num_input_tokens_seen": 74755880,
      "step": 128860
    },
    {
      "epoch": 19.193476318141197,
      "grad_norm": 16.5,
      "learning_rate": 2.4736706826333775e-07,
      "loss": 0.9966,
      "num_input_tokens_seen": 74758568,
      "step": 128865
    },
    {
      "epoch": 19.194221030682158,
      "grad_norm": 10.5,
      "learning_rate": 2.4691129817293324e-07,
      "loss": 1.2077,
      "num_input_tokens_seen": 74761480,
      "step": 128870
    },
    {
      "epoch": 19.194965743223115,
      "grad_norm": 7.40625,
      "learning_rate": 2.4645594626080405e-07,
      "loss": 0.947,
      "num_input_tokens_seen": 74764168,
      "step": 128875
    },
    {
      "epoch": 19.195710455764075,
      "grad_norm": 9.3125,
      "learning_rate": 2.460010125346468e-07,
      "loss": 1.1007,
      "num_input_tokens_seen": 74766952,
      "step": 128880
    },
    {
      "epoch": 19.196455168305036,
      "grad_norm": 19.0,
      "learning_rate": 2.455464970021415e-07,
      "loss": 1.192,
      "num_input_tokens_seen": 74769800,
      "step": 128885
    },
    {
      "epoch": 19.197199880845993,
      "grad_norm": 5.96875,
      "learning_rate": 2.450923996709681e-07,
      "loss": 1.168,
      "num_input_tokens_seen": 74772392,
      "step": 128890
    },
    {
      "epoch": 19.197944593386953,
      "grad_norm": 9.5625,
      "learning_rate": 2.446387205487982e-07,
      "loss": 0.8318,
      "num_input_tokens_seen": 74775272,
      "step": 128895
    },
    {
      "epoch": 19.19868930592791,
      "grad_norm": 7.40625,
      "learning_rate": 2.44185459643298e-07,
      "loss": 0.8405,
      "num_input_tokens_seen": 74778088,
      "step": 128900
    },
    {
      "epoch": 19.19943401846887,
      "grad_norm": 8.625,
      "learning_rate": 2.4373261696212237e-07,
      "loss": 1.0622,
      "num_input_tokens_seen": 74780904,
      "step": 128905
    },
    {
      "epoch": 19.20017873100983,
      "grad_norm": 21.625,
      "learning_rate": 2.4328019251292355e-07,
      "loss": 1.0719,
      "num_input_tokens_seen": 74784168,
      "step": 128910
    },
    {
      "epoch": 19.20092344355079,
      "grad_norm": 10.6875,
      "learning_rate": 2.4282818630334547e-07,
      "loss": 1.4003,
      "num_input_tokens_seen": 74787176,
      "step": 128915
    },
    {
      "epoch": 19.20166815609175,
      "grad_norm": 7.625,
      "learning_rate": 2.4237659834102364e-07,
      "loss": 1.0091,
      "num_input_tokens_seen": 74790184,
      "step": 128920
    },
    {
      "epoch": 19.202412868632706,
      "grad_norm": 13.375,
      "learning_rate": 2.4192542863358534e-07,
      "loss": 1.3366,
      "num_input_tokens_seen": 74792936,
      "step": 128925
    },
    {
      "epoch": 19.203157581173667,
      "grad_norm": 17.625,
      "learning_rate": 2.4147467718865227e-07,
      "loss": 1.4409,
      "num_input_tokens_seen": 74795976,
      "step": 128930
    },
    {
      "epoch": 19.203902293714627,
      "grad_norm": 12.3125,
      "learning_rate": 2.410243440138432e-07,
      "loss": 0.8599,
      "num_input_tokens_seen": 74798760,
      "step": 128935
    },
    {
      "epoch": 19.204647006255584,
      "grad_norm": 6.09375,
      "learning_rate": 2.405744291167633e-07,
      "loss": 0.8591,
      "num_input_tokens_seen": 74801800,
      "step": 128940
    },
    {
      "epoch": 19.205391718796545,
      "grad_norm": 10.5,
      "learning_rate": 2.4012493250501476e-07,
      "loss": 1.3285,
      "num_input_tokens_seen": 74804712,
      "step": 128945
    },
    {
      "epoch": 19.206136431337505,
      "grad_norm": 5.21875,
      "learning_rate": 2.3967585418619153e-07,
      "loss": 0.9733,
      "num_input_tokens_seen": 74807720,
      "step": 128950
    },
    {
      "epoch": 19.206881143878462,
      "grad_norm": 8.125,
      "learning_rate": 2.392271941678792e-07,
      "loss": 0.947,
      "num_input_tokens_seen": 74810504,
      "step": 128955
    },
    {
      "epoch": 19.207625856419423,
      "grad_norm": 13.4375,
      "learning_rate": 2.387789524576578e-07,
      "loss": 1.4279,
      "num_input_tokens_seen": 74813256,
      "step": 128960
    },
    {
      "epoch": 19.20837056896038,
      "grad_norm": 9.0,
      "learning_rate": 2.383311290630963e-07,
      "loss": 1.0334,
      "num_input_tokens_seen": 74816104,
      "step": 128965
    },
    {
      "epoch": 19.20911528150134,
      "grad_norm": 12.75,
      "learning_rate": 2.3788372399176638e-07,
      "loss": 1.423,
      "num_input_tokens_seen": 74819400,
      "step": 128970
    },
    {
      "epoch": 19.2098599940423,
      "grad_norm": 8.875,
      "learning_rate": 2.3743673725122318e-07,
      "loss": 0.8315,
      "num_input_tokens_seen": 74822152,
      "step": 128975
    },
    {
      "epoch": 19.210604706583258,
      "grad_norm": 8.5,
      "learning_rate": 2.3699016884901893e-07,
      "loss": 0.7853,
      "num_input_tokens_seen": 74824968,
      "step": 128980
    },
    {
      "epoch": 19.21134941912422,
      "grad_norm": 14.3125,
      "learning_rate": 2.365440187926976e-07,
      "loss": 1.0299,
      "num_input_tokens_seen": 74828136,
      "step": 128985
    },
    {
      "epoch": 19.21209413166518,
      "grad_norm": 7.0625,
      "learning_rate": 2.3609828708979765e-07,
      "loss": 1.0389,
      "num_input_tokens_seen": 74830952,
      "step": 128990
    },
    {
      "epoch": 19.212838844206136,
      "grad_norm": 14.625,
      "learning_rate": 2.3565297374784635e-07,
      "loss": 1.4037,
      "num_input_tokens_seen": 74833928,
      "step": 128995
    },
    {
      "epoch": 19.213583556747096,
      "grad_norm": 15.5,
      "learning_rate": 2.35208078774371e-07,
      "loss": 0.9276,
      "num_input_tokens_seen": 74836968,
      "step": 129000
    },
    {
      "epoch": 19.214328269288053,
      "grad_norm": 12.875,
      "learning_rate": 2.3476360217688508e-07,
      "loss": 1.2896,
      "num_input_tokens_seen": 74840296,
      "step": 129005
    },
    {
      "epoch": 19.215072981829014,
      "grad_norm": 8.625,
      "learning_rate": 2.3431954396289645e-07,
      "loss": 0.9994,
      "num_input_tokens_seen": 74843048,
      "step": 129010
    },
    {
      "epoch": 19.215817694369974,
      "grad_norm": 10.1875,
      "learning_rate": 2.3387590413991022e-07,
      "loss": 1.2982,
      "num_input_tokens_seen": 74846120,
      "step": 129015
    },
    {
      "epoch": 19.21656240691093,
      "grad_norm": 14.6875,
      "learning_rate": 2.3343268271541764e-07,
      "loss": 0.8957,
      "num_input_tokens_seen": 74849064,
      "step": 129020
    },
    {
      "epoch": 19.217307119451892,
      "grad_norm": 13.25,
      "learning_rate": 2.329898796969099e-07,
      "loss": 1.1941,
      "num_input_tokens_seen": 74851816,
      "step": 129025
    },
    {
      "epoch": 19.218051831992852,
      "grad_norm": 13.125,
      "learning_rate": 2.3254749509186434e-07,
      "loss": 1.1465,
      "num_input_tokens_seen": 74854696,
      "step": 129030
    },
    {
      "epoch": 19.21879654453381,
      "grad_norm": 18.125,
      "learning_rate": 2.321055289077584e-07,
      "loss": 1.0758,
      "num_input_tokens_seen": 74857576,
      "step": 129035
    },
    {
      "epoch": 19.21954125707477,
      "grad_norm": 13.875,
      "learning_rate": 2.3166398115205545e-07,
      "loss": 1.0269,
      "num_input_tokens_seen": 74860392,
      "step": 129040
    },
    {
      "epoch": 19.220285969615727,
      "grad_norm": 14.1875,
      "learning_rate": 2.3122285183221627e-07,
      "loss": 1.0175,
      "num_input_tokens_seen": 74863432,
      "step": 129045
    },
    {
      "epoch": 19.221030682156687,
      "grad_norm": 8.4375,
      "learning_rate": 2.3078214095569318e-07,
      "loss": 0.8169,
      "num_input_tokens_seen": 74866472,
      "step": 129050
    },
    {
      "epoch": 19.221775394697648,
      "grad_norm": 17.875,
      "learning_rate": 2.3034184852993025e-07,
      "loss": 1.166,
      "num_input_tokens_seen": 74869608,
      "step": 129055
    },
    {
      "epoch": 19.222520107238605,
      "grad_norm": 9.875,
      "learning_rate": 2.2990197456236873e-07,
      "loss": 0.987,
      "num_input_tokens_seen": 74872264,
      "step": 129060
    },
    {
      "epoch": 19.223264819779565,
      "grad_norm": 4.6875,
      "learning_rate": 2.2946251906043604e-07,
      "loss": 0.9722,
      "num_input_tokens_seen": 74874856,
      "step": 129065
    },
    {
      "epoch": 19.224009532320526,
      "grad_norm": 17.25,
      "learning_rate": 2.2902348203155955e-07,
      "loss": 1.0704,
      "num_input_tokens_seen": 74877768,
      "step": 129070
    },
    {
      "epoch": 19.224754244861483,
      "grad_norm": 15.9375,
      "learning_rate": 2.2858486348315555e-07,
      "loss": 1.0236,
      "num_input_tokens_seen": 74880840,
      "step": 129075
    },
    {
      "epoch": 19.225498957402444,
      "grad_norm": 10.3125,
      "learning_rate": 2.281466634226348e-07,
      "loss": 0.9017,
      "num_input_tokens_seen": 74883848,
      "step": 129080
    },
    {
      "epoch": 19.2262436699434,
      "grad_norm": 11.75,
      "learning_rate": 2.277088818573969e-07,
      "loss": 0.8467,
      "num_input_tokens_seen": 74886568,
      "step": 129085
    },
    {
      "epoch": 19.22698838248436,
      "grad_norm": 10.6875,
      "learning_rate": 2.2727151879484155e-07,
      "loss": 1.074,
      "num_input_tokens_seen": 74889320,
      "step": 129090
    },
    {
      "epoch": 19.22773309502532,
      "grad_norm": 9.4375,
      "learning_rate": 2.2683457424235722e-07,
      "loss": 1.072,
      "num_input_tokens_seen": 74892104,
      "step": 129095
    },
    {
      "epoch": 19.22847780756628,
      "grad_norm": 7.9375,
      "learning_rate": 2.2639804820732135e-07,
      "loss": 1.0168,
      "num_input_tokens_seen": 74895208,
      "step": 129100
    },
    {
      "epoch": 19.22922252010724,
      "grad_norm": 10.0625,
      "learning_rate": 2.259619406971142e-07,
      "loss": 1.3277,
      "num_input_tokens_seen": 74898280,
      "step": 129105
    },
    {
      "epoch": 19.229967232648196,
      "grad_norm": 6.375,
      "learning_rate": 2.2552625171909925e-07,
      "loss": 0.8561,
      "num_input_tokens_seen": 74901160,
      "step": 129110
    },
    {
      "epoch": 19.230711945189157,
      "grad_norm": 12.8125,
      "learning_rate": 2.250909812806401e-07,
      "loss": 1.0414,
      "num_input_tokens_seen": 74904008,
      "step": 129115
    },
    {
      "epoch": 19.231456657730117,
      "grad_norm": 12.6875,
      "learning_rate": 2.246561293890892e-07,
      "loss": 1.2192,
      "num_input_tokens_seen": 74906856,
      "step": 129120
    },
    {
      "epoch": 19.232201370271074,
      "grad_norm": 13.375,
      "learning_rate": 2.2422169605178788e-07,
      "loss": 1.4761,
      "num_input_tokens_seen": 74909576,
      "step": 129125
    },
    {
      "epoch": 19.232946082812035,
      "grad_norm": 9.1875,
      "learning_rate": 2.2378768127608584e-07,
      "loss": 0.9353,
      "num_input_tokens_seen": 74912776,
      "step": 129130
    },
    {
      "epoch": 19.233690795352995,
      "grad_norm": 5.53125,
      "learning_rate": 2.23354085069305e-07,
      "loss": 0.9206,
      "num_input_tokens_seen": 74915624,
      "step": 129135
    },
    {
      "epoch": 19.234435507893952,
      "grad_norm": 14.1875,
      "learning_rate": 2.2292090743877836e-07,
      "loss": 0.9345,
      "num_input_tokens_seen": 74918376,
      "step": 129140
    },
    {
      "epoch": 19.235180220434913,
      "grad_norm": 8.0625,
      "learning_rate": 2.2248814839181953e-07,
      "loss": 0.8867,
      "num_input_tokens_seen": 74921320,
      "step": 129145
    },
    {
      "epoch": 19.23592493297587,
      "grad_norm": 4.375,
      "learning_rate": 2.2205580793573932e-07,
      "loss": 0.8291,
      "num_input_tokens_seen": 74924488,
      "step": 129150
    },
    {
      "epoch": 19.23666964551683,
      "grad_norm": 11.8125,
      "learning_rate": 2.2162388607784578e-07,
      "loss": 0.9887,
      "num_input_tokens_seen": 74927208,
      "step": 129155
    },
    {
      "epoch": 19.23741435805779,
      "grad_norm": 12.3125,
      "learning_rate": 2.2119238282543032e-07,
      "loss": 0.9283,
      "num_input_tokens_seen": 74930056,
      "step": 129160
    },
    {
      "epoch": 19.238159070598748,
      "grad_norm": 6.46875,
      "learning_rate": 2.2076129818578706e-07,
      "loss": 0.8484,
      "num_input_tokens_seen": 74933128,
      "step": 129165
    },
    {
      "epoch": 19.23890378313971,
      "grad_norm": 10.125,
      "learning_rate": 2.203306321661963e-07,
      "loss": 1.2915,
      "num_input_tokens_seen": 74936072,
      "step": 129170
    },
    {
      "epoch": 19.23964849568067,
      "grad_norm": 12.6875,
      "learning_rate": 2.1990038477393559e-07,
      "loss": 1.1197,
      "num_input_tokens_seen": 74938856,
      "step": 129175
    },
    {
      "epoch": 19.240393208221626,
      "grad_norm": 17.0,
      "learning_rate": 2.1947055601627132e-07,
      "loss": 1.0622,
      "num_input_tokens_seen": 74941736,
      "step": 129180
    },
    {
      "epoch": 19.241137920762586,
      "grad_norm": 4.4375,
      "learning_rate": 2.190411459004671e-07,
      "loss": 1.1428,
      "num_input_tokens_seen": 74944552,
      "step": 129185
    },
    {
      "epoch": 19.241882633303543,
      "grad_norm": 7.40625,
      "learning_rate": 2.1861215443377547e-07,
      "loss": 1.2276,
      "num_input_tokens_seen": 74947336,
      "step": 129190
    },
    {
      "epoch": 19.242627345844504,
      "grad_norm": 10.5625,
      "learning_rate": 2.1818358162344622e-07,
      "loss": 0.9718,
      "num_input_tokens_seen": 74950088,
      "step": 129195
    },
    {
      "epoch": 19.243372058385464,
      "grad_norm": 9.5,
      "learning_rate": 2.1775542747671795e-07,
      "loss": 1.0795,
      "num_input_tokens_seen": 74953064,
      "step": 129200
    },
    {
      "epoch": 19.24411677092642,
      "grad_norm": 12.9375,
      "learning_rate": 2.173276920008238e-07,
      "loss": 0.7032,
      "num_input_tokens_seen": 74955880,
      "step": 129205
    },
    {
      "epoch": 19.244861483467382,
      "grad_norm": 6.46875,
      "learning_rate": 2.1690037520299134e-07,
      "loss": 0.8909,
      "num_input_tokens_seen": 74959048,
      "step": 129210
    },
    {
      "epoch": 19.245606196008342,
      "grad_norm": 12.9375,
      "learning_rate": 2.1647347709043696e-07,
      "loss": 1.4617,
      "num_input_tokens_seen": 74961768,
      "step": 129215
    },
    {
      "epoch": 19.2463509085493,
      "grad_norm": 5.125,
      "learning_rate": 2.160469976703744e-07,
      "loss": 0.9921,
      "num_input_tokens_seen": 74964712,
      "step": 129220
    },
    {
      "epoch": 19.24709562109026,
      "grad_norm": 7.6875,
      "learning_rate": 2.1562093695000897e-07,
      "loss": 0.9189,
      "num_input_tokens_seen": 74967400,
      "step": 129225
    },
    {
      "epoch": 19.247840333631217,
      "grad_norm": 6.75,
      "learning_rate": 2.1519529493654045e-07,
      "loss": 1.2754,
      "num_input_tokens_seen": 74970312,
      "step": 129230
    },
    {
      "epoch": 19.248585046172177,
      "grad_norm": 19.5,
      "learning_rate": 2.14770071637152e-07,
      "loss": 1.1739,
      "num_input_tokens_seen": 74973192,
      "step": 129235
    },
    {
      "epoch": 19.249329758713138,
      "grad_norm": 14.125,
      "learning_rate": 2.143452670590379e-07,
      "loss": 1.4402,
      "num_input_tokens_seen": 74976104,
      "step": 129240
    },
    {
      "epoch": 19.250074471254095,
      "grad_norm": 14.3125,
      "learning_rate": 2.1392088120936737e-07,
      "loss": 0.7627,
      "num_input_tokens_seen": 74979176,
      "step": 129245
    },
    {
      "epoch": 19.250819183795056,
      "grad_norm": 6.75,
      "learning_rate": 2.1349691409530968e-07,
      "loss": 0.9758,
      "num_input_tokens_seen": 74982120,
      "step": 129250
    },
    {
      "epoch": 19.251563896336016,
      "grad_norm": 9.75,
      "learning_rate": 2.1307336572403415e-07,
      "loss": 1.0012,
      "num_input_tokens_seen": 74984808,
      "step": 129255
    },
    {
      "epoch": 19.252308608876973,
      "grad_norm": 19.375,
      "learning_rate": 2.1265023610268776e-07,
      "loss": 0.8638,
      "num_input_tokens_seen": 74987752,
      "step": 129260
    },
    {
      "epoch": 19.253053321417934,
      "grad_norm": 9.9375,
      "learning_rate": 2.1222752523842594e-07,
      "loss": 1.4032,
      "num_input_tokens_seen": 74990312,
      "step": 129265
    },
    {
      "epoch": 19.25379803395889,
      "grad_norm": 12.75,
      "learning_rate": 2.1180523313838462e-07,
      "loss": 1.1437,
      "num_input_tokens_seen": 74993352,
      "step": 129270
    },
    {
      "epoch": 19.25454274649985,
      "grad_norm": 30.875,
      "learning_rate": 2.1138335980970258e-07,
      "loss": 1.4363,
      "num_input_tokens_seen": 74996296,
      "step": 129275
    },
    {
      "epoch": 19.25528745904081,
      "grad_norm": 6.5,
      "learning_rate": 2.1096190525950464e-07,
      "loss": 1.0152,
      "num_input_tokens_seen": 74999016,
      "step": 129280
    },
    {
      "epoch": 19.25603217158177,
      "grad_norm": 5.21875,
      "learning_rate": 2.1054086949491013e-07,
      "loss": 0.9591,
      "num_input_tokens_seen": 75001960,
      "step": 129285
    },
    {
      "epoch": 19.25677688412273,
      "grad_norm": 6.375,
      "learning_rate": 2.101202525230328e-07,
      "loss": 0.725,
      "num_input_tokens_seen": 75004872,
      "step": 129290
    },
    {
      "epoch": 19.257521596663686,
      "grad_norm": 10.1875,
      "learning_rate": 2.0970005435097807e-07,
      "loss": 1.3121,
      "num_input_tokens_seen": 75007560,
      "step": 129295
    },
    {
      "epoch": 19.258266309204647,
      "grad_norm": 15.875,
      "learning_rate": 2.0928027498584579e-07,
      "loss": 1.4672,
      "num_input_tokens_seen": 75010760,
      "step": 129300
    },
    {
      "epoch": 19.259011021745607,
      "grad_norm": 7.53125,
      "learning_rate": 2.0886091443472477e-07,
      "loss": 1.2565,
      "num_input_tokens_seen": 75013768,
      "step": 129305
    },
    {
      "epoch": 19.259755734286564,
      "grad_norm": 11.9375,
      "learning_rate": 2.084419727047038e-07,
      "loss": 1.1579,
      "num_input_tokens_seen": 75016488,
      "step": 129310
    },
    {
      "epoch": 19.260500446827525,
      "grad_norm": 7.09375,
      "learning_rate": 2.0802344980285771e-07,
      "loss": 1.0067,
      "num_input_tokens_seen": 75019528,
      "step": 129315
    },
    {
      "epoch": 19.261245159368485,
      "grad_norm": 10.875,
      "learning_rate": 2.0760534573626144e-07,
      "loss": 1.1665,
      "num_input_tokens_seen": 75022344,
      "step": 129320
    },
    {
      "epoch": 19.261989871909442,
      "grad_norm": 11.125,
      "learning_rate": 2.0718766051197048e-07,
      "loss": 0.9232,
      "num_input_tokens_seen": 75025384,
      "step": 129325
    },
    {
      "epoch": 19.262734584450403,
      "grad_norm": 6.125,
      "learning_rate": 2.0677039413704857e-07,
      "loss": 1.1238,
      "num_input_tokens_seen": 75028520,
      "step": 129330
    },
    {
      "epoch": 19.26347929699136,
      "grad_norm": 12.25,
      "learning_rate": 2.063535466185429e-07,
      "loss": 1.1997,
      "num_input_tokens_seen": 75031688,
      "step": 129335
    },
    {
      "epoch": 19.26422400953232,
      "grad_norm": 6.46875,
      "learning_rate": 2.0593711796349225e-07,
      "loss": 0.8677,
      "num_input_tokens_seen": 75034568,
      "step": 129340
    },
    {
      "epoch": 19.26496872207328,
      "grad_norm": 9.375,
      "learning_rate": 2.0552110817893544e-07,
      "loss": 1.1738,
      "num_input_tokens_seen": 75037544,
      "step": 129345
    },
    {
      "epoch": 19.265713434614238,
      "grad_norm": 5.34375,
      "learning_rate": 2.051055172719002e-07,
      "loss": 1.0883,
      "num_input_tokens_seen": 75040456,
      "step": 129350
    },
    {
      "epoch": 19.2664581471552,
      "grad_norm": 9.1875,
      "learning_rate": 2.0469034524940588e-07,
      "loss": 0.9749,
      "num_input_tokens_seen": 75043240,
      "step": 129355
    },
    {
      "epoch": 19.26720285969616,
      "grad_norm": 5.5625,
      "learning_rate": 2.0427559211846915e-07,
      "loss": 0.9127,
      "num_input_tokens_seen": 75046728,
      "step": 129360
    },
    {
      "epoch": 19.267947572237116,
      "grad_norm": 10.8125,
      "learning_rate": 2.0386125788609266e-07,
      "loss": 1.221,
      "num_input_tokens_seen": 75049480,
      "step": 129365
    },
    {
      "epoch": 19.268692284778076,
      "grad_norm": 10.375,
      "learning_rate": 2.034473425592792e-07,
      "loss": 1.2204,
      "num_input_tokens_seen": 75052360,
      "step": 129370
    },
    {
      "epoch": 19.269436997319033,
      "grad_norm": 8.0625,
      "learning_rate": 2.0303384614502042e-07,
      "loss": 0.7868,
      "num_input_tokens_seen": 75055304,
      "step": 129375
    },
    {
      "epoch": 19.270181709859994,
      "grad_norm": 10.0,
      "learning_rate": 2.0262076865030232e-07,
      "loss": 1.2986,
      "num_input_tokens_seen": 75058248,
      "step": 129380
    },
    {
      "epoch": 19.270926422400954,
      "grad_norm": 10.6875,
      "learning_rate": 2.022081100821055e-07,
      "loss": 1.1056,
      "num_input_tokens_seen": 75061480,
      "step": 129385
    },
    {
      "epoch": 19.27167113494191,
      "grad_norm": 6.34375,
      "learning_rate": 2.0179587044739655e-07,
      "loss": 0.9198,
      "num_input_tokens_seen": 75064136,
      "step": 129390
    },
    {
      "epoch": 19.272415847482872,
      "grad_norm": 15.3125,
      "learning_rate": 2.0138404975314495e-07,
      "loss": 1.1952,
      "num_input_tokens_seen": 75066952,
      "step": 129395
    },
    {
      "epoch": 19.273160560023832,
      "grad_norm": 10.8125,
      "learning_rate": 2.0097264800630344e-07,
      "loss": 1.1784,
      "num_input_tokens_seen": 75070120,
      "step": 129400
    },
    {
      "epoch": 19.27390527256479,
      "grad_norm": 9.5625,
      "learning_rate": 2.0056166521382759e-07,
      "loss": 1.2449,
      "num_input_tokens_seen": 75073000,
      "step": 129405
    },
    {
      "epoch": 19.27464998510575,
      "grad_norm": 5.28125,
      "learning_rate": 2.0015110138265624e-07,
      "loss": 0.7817,
      "num_input_tokens_seen": 75075784,
      "step": 129410
    },
    {
      "epoch": 19.275394697646707,
      "grad_norm": 11.3125,
      "learning_rate": 1.997409565197228e-07,
      "loss": 1.3576,
      "num_input_tokens_seen": 75078248,
      "step": 129415
    },
    {
      "epoch": 19.276139410187668,
      "grad_norm": 7.375,
      "learning_rate": 1.9933123063196335e-07,
      "loss": 1.1265,
      "num_input_tokens_seen": 75081128,
      "step": 129420
    },
    {
      "epoch": 19.276884122728628,
      "grad_norm": 8.5,
      "learning_rate": 1.9892192372629737e-07,
      "loss": 1.2267,
      "num_input_tokens_seen": 75084200,
      "step": 129425
    },
    {
      "epoch": 19.277628835269585,
      "grad_norm": 14.9375,
      "learning_rate": 1.9851303580963599e-07,
      "loss": 1.0609,
      "num_input_tokens_seen": 75086824,
      "step": 129430
    },
    {
      "epoch": 19.278373547810546,
      "grad_norm": 12.6875,
      "learning_rate": 1.9810456688889313e-07,
      "loss": 1.13,
      "num_input_tokens_seen": 75089544,
      "step": 129435
    },
    {
      "epoch": 19.279118260351503,
      "grad_norm": 9.0,
      "learning_rate": 1.9769651697096326e-07,
      "loss": 1.153,
      "num_input_tokens_seen": 75092808,
      "step": 129440
    },
    {
      "epoch": 19.279862972892463,
      "grad_norm": 5.0625,
      "learning_rate": 1.9728888606274365e-07,
      "loss": 0.8623,
      "num_input_tokens_seen": 75095688,
      "step": 129445
    },
    {
      "epoch": 19.280607685433424,
      "grad_norm": 6.5,
      "learning_rate": 1.9688167417112047e-07,
      "loss": 1.374,
      "num_input_tokens_seen": 75098504,
      "step": 129450
    },
    {
      "epoch": 19.28135239797438,
      "grad_norm": 6.53125,
      "learning_rate": 1.9647488130297154e-07,
      "loss": 0.9235,
      "num_input_tokens_seen": 75101704,
      "step": 129455
    },
    {
      "epoch": 19.28209711051534,
      "grad_norm": 20.5,
      "learning_rate": 1.960685074651719e-07,
      "loss": 1.1726,
      "num_input_tokens_seen": 75104616,
      "step": 129460
    },
    {
      "epoch": 19.2828418230563,
      "grad_norm": 13.9375,
      "learning_rate": 1.9566255266458278e-07,
      "loss": 1.0215,
      "num_input_tokens_seen": 75107560,
      "step": 129465
    },
    {
      "epoch": 19.28358653559726,
      "grad_norm": 8.4375,
      "learning_rate": 1.9525701690806807e-07,
      "loss": 0.852,
      "num_input_tokens_seen": 75110376,
      "step": 129470
    },
    {
      "epoch": 19.28433124813822,
      "grad_norm": 6.34375,
      "learning_rate": 1.948519002024751e-07,
      "loss": 1.26,
      "num_input_tokens_seen": 75113352,
      "step": 129475
    },
    {
      "epoch": 19.285075960679176,
      "grad_norm": 6.65625,
      "learning_rate": 1.9444720255464844e-07,
      "loss": 1.0382,
      "num_input_tokens_seen": 75116264,
      "step": 129480
    },
    {
      "epoch": 19.285820673220137,
      "grad_norm": 10.5625,
      "learning_rate": 1.940429239714242e-07,
      "loss": 1.7104,
      "num_input_tokens_seen": 75119176,
      "step": 129485
    },
    {
      "epoch": 19.286565385761097,
      "grad_norm": 13.0,
      "learning_rate": 1.936390644596303e-07,
      "loss": 1.1291,
      "num_input_tokens_seen": 75122248,
      "step": 129490
    },
    {
      "epoch": 19.287310098302054,
      "grad_norm": 15.875,
      "learning_rate": 1.932356240260974e-07,
      "loss": 1.2669,
      "num_input_tokens_seen": 75125000,
      "step": 129495
    },
    {
      "epoch": 19.288054810843015,
      "grad_norm": 8.9375,
      "learning_rate": 1.9283260267763115e-07,
      "loss": 1.3279,
      "num_input_tokens_seen": 75127720,
      "step": 129500
    },
    {
      "epoch": 19.288799523383975,
      "grad_norm": 9.875,
      "learning_rate": 1.924300004210483e-07,
      "loss": 1.327,
      "num_input_tokens_seen": 75130440,
      "step": 129505
    },
    {
      "epoch": 19.289544235924932,
      "grad_norm": 8.375,
      "learning_rate": 1.9202781726314622e-07,
      "loss": 1.3167,
      "num_input_tokens_seen": 75133256,
      "step": 129510
    },
    {
      "epoch": 19.290288948465893,
      "grad_norm": 8.8125,
      "learning_rate": 1.9162605321072224e-07,
      "loss": 1.0591,
      "num_input_tokens_seen": 75135912,
      "step": 129515
    },
    {
      "epoch": 19.29103366100685,
      "grad_norm": 9.8125,
      "learning_rate": 1.9122470827055984e-07,
      "loss": 0.9716,
      "num_input_tokens_seen": 75138952,
      "step": 129520
    },
    {
      "epoch": 19.29177837354781,
      "grad_norm": 8.375,
      "learning_rate": 1.9082378244944242e-07,
      "loss": 1.3464,
      "num_input_tokens_seen": 75141768,
      "step": 129525
    },
    {
      "epoch": 19.29252308608877,
      "grad_norm": 6.96875,
      "learning_rate": 1.9042327575414242e-07,
      "loss": 1.0892,
      "num_input_tokens_seen": 75144520,
      "step": 129530
    },
    {
      "epoch": 19.293267798629728,
      "grad_norm": 11.1875,
      "learning_rate": 1.9002318819142661e-07,
      "loss": 0.8832,
      "num_input_tokens_seen": 75147240,
      "step": 129535
    },
    {
      "epoch": 19.29401251117069,
      "grad_norm": 7.875,
      "learning_rate": 1.8962351976805348e-07,
      "loss": 1.0375,
      "num_input_tokens_seen": 75150312,
      "step": 129540
    },
    {
      "epoch": 19.29475722371165,
      "grad_norm": 20.75,
      "learning_rate": 1.892242704907732e-07,
      "loss": 1.0714,
      "num_input_tokens_seen": 75153128,
      "step": 129545
    },
    {
      "epoch": 19.295501936252606,
      "grad_norm": 12.1875,
      "learning_rate": 1.888254403663331e-07,
      "loss": 0.9849,
      "num_input_tokens_seen": 75156200,
      "step": 129550
    },
    {
      "epoch": 19.296246648793566,
      "grad_norm": 6.0625,
      "learning_rate": 1.884270294014695e-07,
      "loss": 1.0462,
      "num_input_tokens_seen": 75159208,
      "step": 129555
    },
    {
      "epoch": 19.296991361334523,
      "grad_norm": 12.3125,
      "learning_rate": 1.880290376029159e-07,
      "loss": 0.9371,
      "num_input_tokens_seen": 75161928,
      "step": 129560
    },
    {
      "epoch": 19.297736073875484,
      "grad_norm": 10.875,
      "learning_rate": 1.8763146497739194e-07,
      "loss": 0.8969,
      "num_input_tokens_seen": 75164776,
      "step": 129565
    },
    {
      "epoch": 19.298480786416444,
      "grad_norm": 11.4375,
      "learning_rate": 1.872343115316144e-07,
      "loss": 1.36,
      "num_input_tokens_seen": 75167688,
      "step": 129570
    },
    {
      "epoch": 19.2992254989574,
      "grad_norm": 12.3125,
      "learning_rate": 1.8683757727229745e-07,
      "loss": 1.2593,
      "num_input_tokens_seen": 75170120,
      "step": 129575
    },
    {
      "epoch": 19.299970211498362,
      "grad_norm": 7.59375,
      "learning_rate": 1.864412622061412e-07,
      "loss": 1.3468,
      "num_input_tokens_seen": 75173096,
      "step": 129580
    },
    {
      "epoch": 19.300714924039323,
      "grad_norm": 7.96875,
      "learning_rate": 1.8604536633984037e-07,
      "loss": 1.304,
      "num_input_tokens_seen": 75175976,
      "step": 129585
    },
    {
      "epoch": 19.30145963658028,
      "grad_norm": 16.75,
      "learning_rate": 1.8564988968008124e-07,
      "loss": 1.1685,
      "num_input_tokens_seen": 75178952,
      "step": 129590
    },
    {
      "epoch": 19.30220434912124,
      "grad_norm": 5.375,
      "learning_rate": 1.8525483223354734e-07,
      "loss": 0.9887,
      "num_input_tokens_seen": 75181832,
      "step": 129595
    },
    {
      "epoch": 19.302949061662197,
      "grad_norm": 3.8125,
      "learning_rate": 1.848601940069139e-07,
      "loss": 0.7753,
      "num_input_tokens_seen": 75184680,
      "step": 129600
    },
    {
      "epoch": 19.303693774203158,
      "grad_norm": 5.3125,
      "learning_rate": 1.8446597500684503e-07,
      "loss": 0.7361,
      "num_input_tokens_seen": 75187496,
      "step": 129605
    },
    {
      "epoch": 19.304438486744118,
      "grad_norm": 10.5625,
      "learning_rate": 1.8407217524000486e-07,
      "loss": 0.9046,
      "num_input_tokens_seen": 75190216,
      "step": 129610
    },
    {
      "epoch": 19.305183199285075,
      "grad_norm": 6.65625,
      "learning_rate": 1.8367879471304084e-07,
      "loss": 1.2105,
      "num_input_tokens_seen": 75193128,
      "step": 129615
    },
    {
      "epoch": 19.305927911826036,
      "grad_norm": 14.25,
      "learning_rate": 1.832858334326032e-07,
      "loss": 1.0165,
      "num_input_tokens_seen": 75196232,
      "step": 129620
    },
    {
      "epoch": 19.306672624366993,
      "grad_norm": 7.46875,
      "learning_rate": 1.828932914053255e-07,
      "loss": 1.1604,
      "num_input_tokens_seen": 75198952,
      "step": 129625
    },
    {
      "epoch": 19.307417336907953,
      "grad_norm": 6.6875,
      "learning_rate": 1.8250116863784694e-07,
      "loss": 0.937,
      "num_input_tokens_seen": 75201768,
      "step": 129630
    },
    {
      "epoch": 19.308162049448914,
      "grad_norm": 4.5625,
      "learning_rate": 1.8210946513678439e-07,
      "loss": 1.0071,
      "num_input_tokens_seen": 75204456,
      "step": 129635
    },
    {
      "epoch": 19.30890676198987,
      "grad_norm": 11.75,
      "learning_rate": 1.8171818090876037e-07,
      "loss": 0.8772,
      "num_input_tokens_seen": 75207272,
      "step": 129640
    },
    {
      "epoch": 19.30965147453083,
      "grad_norm": 12.875,
      "learning_rate": 1.8132731596038345e-07,
      "loss": 1.0615,
      "num_input_tokens_seen": 75210088,
      "step": 129645
    },
    {
      "epoch": 19.31039618707179,
      "grad_norm": 5.125,
      "learning_rate": 1.8093687029825666e-07,
      "loss": 0.7919,
      "num_input_tokens_seen": 75212744,
      "step": 129650
    },
    {
      "epoch": 19.31114089961275,
      "grad_norm": 8.1875,
      "learning_rate": 1.8054684392897758e-07,
      "loss": 0.8238,
      "num_input_tokens_seen": 75215752,
      "step": 129655
    },
    {
      "epoch": 19.31188561215371,
      "grad_norm": 14.125,
      "learning_rate": 1.8015723685913255e-07,
      "loss": 1.1177,
      "num_input_tokens_seen": 75218728,
      "step": 129660
    },
    {
      "epoch": 19.312630324694666,
      "grad_norm": 9.25,
      "learning_rate": 1.797680490953052e-07,
      "loss": 1.1647,
      "num_input_tokens_seen": 75221384,
      "step": 129665
    },
    {
      "epoch": 19.313375037235627,
      "grad_norm": 8.375,
      "learning_rate": 1.7937928064407085e-07,
      "loss": 1.1075,
      "num_input_tokens_seen": 75224232,
      "step": 129670
    },
    {
      "epoch": 19.314119749776587,
      "grad_norm": 4.9375,
      "learning_rate": 1.7899093151199643e-07,
      "loss": 0.9638,
      "num_input_tokens_seen": 75227400,
      "step": 129675
    },
    {
      "epoch": 19.314864462317544,
      "grad_norm": 5.5,
      "learning_rate": 1.7860300170564613e-07,
      "loss": 0.8943,
      "num_input_tokens_seen": 75230504,
      "step": 129680
    },
    {
      "epoch": 19.315609174858505,
      "grad_norm": 10.0625,
      "learning_rate": 1.7821549123156755e-07,
      "loss": 1.5218,
      "num_input_tokens_seen": 75233192,
      "step": 129685
    },
    {
      "epoch": 19.316353887399465,
      "grad_norm": 9.0,
      "learning_rate": 1.7782840009631375e-07,
      "loss": 0.8304,
      "num_input_tokens_seen": 75236168,
      "step": 129690
    },
    {
      "epoch": 19.317098599940422,
      "grad_norm": 7.59375,
      "learning_rate": 1.7744172830641835e-07,
      "loss": 1.1367,
      "num_input_tokens_seen": 75238728,
      "step": 129695
    },
    {
      "epoch": 19.317843312481383,
      "grad_norm": 9.0625,
      "learning_rate": 1.7705547586841785e-07,
      "loss": 1.2536,
      "num_input_tokens_seen": 75241832,
      "step": 129700
    },
    {
      "epoch": 19.31858802502234,
      "grad_norm": 9.75,
      "learning_rate": 1.7666964278883202e-07,
      "loss": 0.8613,
      "num_input_tokens_seen": 75244520,
      "step": 129705
    },
    {
      "epoch": 19.3193327375633,
      "grad_norm": 12.0625,
      "learning_rate": 1.7628422907418894e-07,
      "loss": 0.889,
      "num_input_tokens_seen": 75247368,
      "step": 129710
    },
    {
      "epoch": 19.32007745010426,
      "grad_norm": 13.5,
      "learning_rate": 1.7589923473098902e-07,
      "loss": 0.8801,
      "num_input_tokens_seen": 75250152,
      "step": 129715
    },
    {
      "epoch": 19.320822162645218,
      "grad_norm": 14.3125,
      "learning_rate": 1.7551465976574643e-07,
      "loss": 0.9061,
      "num_input_tokens_seen": 75253000,
      "step": 129720
    },
    {
      "epoch": 19.32156687518618,
      "grad_norm": 5.15625,
      "learning_rate": 1.7513050418495047e-07,
      "loss": 0.6975,
      "num_input_tokens_seen": 75255944,
      "step": 129725
    },
    {
      "epoch": 19.32231158772714,
      "grad_norm": 17.0,
      "learning_rate": 1.7474676799509314e-07,
      "loss": 1.177,
      "num_input_tokens_seen": 75258792,
      "step": 129730
    },
    {
      "epoch": 19.323056300268096,
      "grad_norm": 15.9375,
      "learning_rate": 1.7436345120266095e-07,
      "loss": 1.093,
      "num_input_tokens_seen": 75261512,
      "step": 129735
    },
    {
      "epoch": 19.323801012809056,
      "grad_norm": 12.5,
      "learning_rate": 1.739805538141237e-07,
      "loss": 1.2852,
      "num_input_tokens_seen": 75264392,
      "step": 129740
    },
    {
      "epoch": 19.324545725350013,
      "grad_norm": 7.0625,
      "learning_rate": 1.73598075835954e-07,
      "loss": 0.9079,
      "num_input_tokens_seen": 75267176,
      "step": 129745
    },
    {
      "epoch": 19.325290437890974,
      "grad_norm": 10.0625,
      "learning_rate": 1.7321601727461334e-07,
      "loss": 0.9835,
      "num_input_tokens_seen": 75270056,
      "step": 129750
    },
    {
      "epoch": 19.326035150431935,
      "grad_norm": 4.3125,
      "learning_rate": 1.7283437813655489e-07,
      "loss": 0.997,
      "num_input_tokens_seen": 75272776,
      "step": 129755
    },
    {
      "epoch": 19.32677986297289,
      "grad_norm": 7.34375,
      "learning_rate": 1.7245315842822352e-07,
      "loss": 0.8216,
      "num_input_tokens_seen": 75275656,
      "step": 129760
    },
    {
      "epoch": 19.327524575513852,
      "grad_norm": 17.0,
      "learning_rate": 1.720723581560668e-07,
      "loss": 1.2871,
      "num_input_tokens_seen": 75278984,
      "step": 129765
    },
    {
      "epoch": 19.328269288054813,
      "grad_norm": 10.0625,
      "learning_rate": 1.716919773265102e-07,
      "loss": 1.0191,
      "num_input_tokens_seen": 75282312,
      "step": 129770
    },
    {
      "epoch": 19.32901400059577,
      "grad_norm": 14.0625,
      "learning_rate": 1.7131201594598468e-07,
      "loss": 0.9731,
      "num_input_tokens_seen": 75285256,
      "step": 129775
    },
    {
      "epoch": 19.32975871313673,
      "grad_norm": 9.75,
      "learning_rate": 1.709324740209073e-07,
      "loss": 1.2115,
      "num_input_tokens_seen": 75288008,
      "step": 129780
    },
    {
      "epoch": 19.330503425677687,
      "grad_norm": 5.40625,
      "learning_rate": 1.7055335155769238e-07,
      "loss": 0.8786,
      "num_input_tokens_seen": 75290888,
      "step": 129785
    },
    {
      "epoch": 19.331248138218648,
      "grad_norm": 12.5625,
      "learning_rate": 1.7017464856274033e-07,
      "loss": 1.0853,
      "num_input_tokens_seen": 75293736,
      "step": 129790
    },
    {
      "epoch": 19.331992850759608,
      "grad_norm": 6.03125,
      "learning_rate": 1.6979636504245445e-07,
      "loss": 0.7231,
      "num_input_tokens_seen": 75296680,
      "step": 129795
    },
    {
      "epoch": 19.332737563300565,
      "grad_norm": 7.5625,
      "learning_rate": 1.6941850100322122e-07,
      "loss": 1.1383,
      "num_input_tokens_seen": 75299720,
      "step": 129800
    },
    {
      "epoch": 19.333482275841526,
      "grad_norm": 12.5625,
      "learning_rate": 1.6904105645142444e-07,
      "loss": 1.0588,
      "num_input_tokens_seen": 75302472,
      "step": 129805
    },
    {
      "epoch": 19.334226988382483,
      "grad_norm": 17.75,
      "learning_rate": 1.686640313934451e-07,
      "loss": 1.0251,
      "num_input_tokens_seen": 75305512,
      "step": 129810
    },
    {
      "epoch": 19.334971700923443,
      "grad_norm": 11.75,
      "learning_rate": 1.6828742583564762e-07,
      "loss": 0.8624,
      "num_input_tokens_seen": 75308296,
      "step": 129815
    },
    {
      "epoch": 19.335716413464404,
      "grad_norm": 12.25,
      "learning_rate": 1.6791123978439626e-07,
      "loss": 1.0748,
      "num_input_tokens_seen": 75311272,
      "step": 129820
    },
    {
      "epoch": 19.33646112600536,
      "grad_norm": 14.375,
      "learning_rate": 1.6753547324604713e-07,
      "loss": 1.2133,
      "num_input_tokens_seen": 75314120,
      "step": 129825
    },
    {
      "epoch": 19.33720583854632,
      "grad_norm": 6.9375,
      "learning_rate": 1.671601262269451e-07,
      "loss": 0.9273,
      "num_input_tokens_seen": 75317160,
      "step": 129830
    },
    {
      "epoch": 19.337950551087282,
      "grad_norm": 4.09375,
      "learning_rate": 1.6678519873343789e-07,
      "loss": 0.9804,
      "num_input_tokens_seen": 75320104,
      "step": 129835
    },
    {
      "epoch": 19.33869526362824,
      "grad_norm": 19.25,
      "learning_rate": 1.66410690771851e-07,
      "loss": 1.5358,
      "num_input_tokens_seen": 75323048,
      "step": 129840
    },
    {
      "epoch": 19.3394399761692,
      "grad_norm": 21.125,
      "learning_rate": 1.6603660234851825e-07,
      "loss": 1.0509,
      "num_input_tokens_seen": 75325512,
      "step": 129845
    },
    {
      "epoch": 19.340184688710156,
      "grad_norm": 9.4375,
      "learning_rate": 1.656629334697568e-07,
      "loss": 0.9706,
      "num_input_tokens_seen": 75328392,
      "step": 129850
    },
    {
      "epoch": 19.340929401251117,
      "grad_norm": 7.0,
      "learning_rate": 1.6528968414188107e-07,
      "loss": 0.7925,
      "num_input_tokens_seen": 75331272,
      "step": 129855
    },
    {
      "epoch": 19.341674113792077,
      "grad_norm": 7.4375,
      "learning_rate": 1.6491685437119154e-07,
      "loss": 1.0503,
      "num_input_tokens_seen": 75333896,
      "step": 129860
    },
    {
      "epoch": 19.342418826333034,
      "grad_norm": 7.53125,
      "learning_rate": 1.6454444416399428e-07,
      "loss": 1.073,
      "num_input_tokens_seen": 75336872,
      "step": 129865
    },
    {
      "epoch": 19.343163538873995,
      "grad_norm": 8.875,
      "learning_rate": 1.6417245352657317e-07,
      "loss": 0.943,
      "num_input_tokens_seen": 75339560,
      "step": 129870
    },
    {
      "epoch": 19.343908251414955,
      "grad_norm": 14.0,
      "learning_rate": 1.638008824652204e-07,
      "loss": 1.2924,
      "num_input_tokens_seen": 75342440,
      "step": 129875
    },
    {
      "epoch": 19.344652963955912,
      "grad_norm": 12.4375,
      "learning_rate": 1.6342973098620872e-07,
      "loss": 1.1724,
      "num_input_tokens_seen": 75345288,
      "step": 129880
    },
    {
      "epoch": 19.345397676496873,
      "grad_norm": 9.5625,
      "learning_rate": 1.6305899909580814e-07,
      "loss": 1.3932,
      "num_input_tokens_seen": 75348552,
      "step": 129885
    },
    {
      "epoch": 19.34614238903783,
      "grad_norm": 13.0,
      "learning_rate": 1.6268868680028026e-07,
      "loss": 1.1583,
      "num_input_tokens_seen": 75351304,
      "step": 129890
    },
    {
      "epoch": 19.34688710157879,
      "grad_norm": 13.125,
      "learning_rate": 1.623187941058868e-07,
      "loss": 0.9463,
      "num_input_tokens_seen": 75354024,
      "step": 129895
    },
    {
      "epoch": 19.34763181411975,
      "grad_norm": 7.375,
      "learning_rate": 1.6194932101886995e-07,
      "loss": 1.1898,
      "num_input_tokens_seen": 75357192,
      "step": 129900
    },
    {
      "epoch": 19.348376526660708,
      "grad_norm": 12.4375,
      "learning_rate": 1.615802675454775e-07,
      "loss": 1.5194,
      "num_input_tokens_seen": 75360872,
      "step": 129905
    },
    {
      "epoch": 19.34912123920167,
      "grad_norm": 11.5,
      "learning_rate": 1.6121163369194335e-07,
      "loss": 1.0195,
      "num_input_tokens_seen": 75363816,
      "step": 129910
    },
    {
      "epoch": 19.34986595174263,
      "grad_norm": 8.5,
      "learning_rate": 1.6084341946449033e-07,
      "loss": 1.1047,
      "num_input_tokens_seen": 75366696,
      "step": 129915
    },
    {
      "epoch": 19.350610664283586,
      "grad_norm": 9.375,
      "learning_rate": 1.6047562486934398e-07,
      "loss": 1.157,
      "num_input_tokens_seen": 75369576,
      "step": 129920
    },
    {
      "epoch": 19.351355376824547,
      "grad_norm": 14.625,
      "learning_rate": 1.60108249912716e-07,
      "loss": 0.8512,
      "num_input_tokens_seen": 75372712,
      "step": 129925
    },
    {
      "epoch": 19.352100089365504,
      "grad_norm": 11.5625,
      "learning_rate": 1.5974129460081255e-07,
      "loss": 1.1146,
      "num_input_tokens_seen": 75375656,
      "step": 129930
    },
    {
      "epoch": 19.352844801906464,
      "grad_norm": 4.15625,
      "learning_rate": 1.5937475893983423e-07,
      "loss": 0.9113,
      "num_input_tokens_seen": 75378600,
      "step": 129935
    },
    {
      "epoch": 19.353589514447425,
      "grad_norm": 6.375,
      "learning_rate": 1.5900864293597328e-07,
      "loss": 1.1651,
      "num_input_tokens_seen": 75381384,
      "step": 129940
    },
    {
      "epoch": 19.35433422698838,
      "grad_norm": 12.5625,
      "learning_rate": 1.5864294659541367e-07,
      "loss": 1.0865,
      "num_input_tokens_seen": 75384104,
      "step": 129945
    },
    {
      "epoch": 19.355078939529342,
      "grad_norm": 9.125,
      "learning_rate": 1.5827766992433378e-07,
      "loss": 1.0826,
      "num_input_tokens_seen": 75386792,
      "step": 129950
    },
    {
      "epoch": 19.3558236520703,
      "grad_norm": 8.6875,
      "learning_rate": 1.5791281292890093e-07,
      "loss": 1.2238,
      "num_input_tokens_seen": 75389864,
      "step": 129955
    },
    {
      "epoch": 19.35656836461126,
      "grad_norm": 10.75,
      "learning_rate": 1.575483756152879e-07,
      "loss": 1.0039,
      "num_input_tokens_seen": 75392904,
      "step": 129960
    },
    {
      "epoch": 19.35731307715222,
      "grad_norm": 10.75,
      "learning_rate": 1.5718435798964538e-07,
      "loss": 0.9983,
      "num_input_tokens_seen": 75395784,
      "step": 129965
    },
    {
      "epoch": 19.358057789693177,
      "grad_norm": 11.25,
      "learning_rate": 1.5682076005812118e-07,
      "loss": 1.4929,
      "num_input_tokens_seen": 75398824,
      "step": 129970
    },
    {
      "epoch": 19.358802502234138,
      "grad_norm": 3.921875,
      "learning_rate": 1.564575818268632e-07,
      "loss": 1.2427,
      "num_input_tokens_seen": 75401480,
      "step": 129975
    },
    {
      "epoch": 19.359547214775098,
      "grad_norm": 6.21875,
      "learning_rate": 1.5609482330200265e-07,
      "loss": 1.0239,
      "num_input_tokens_seen": 75404456,
      "step": 129980
    },
    {
      "epoch": 19.360291927316055,
      "grad_norm": 20.375,
      "learning_rate": 1.5573248448967072e-07,
      "loss": 1.1562,
      "num_input_tokens_seen": 75407336,
      "step": 129985
    },
    {
      "epoch": 19.361036639857016,
      "grad_norm": 10.9375,
      "learning_rate": 1.5537056539598748e-07,
      "loss": 1.1154,
      "num_input_tokens_seen": 75410440,
      "step": 129990
    },
    {
      "epoch": 19.361781352397973,
      "grad_norm": 8.5,
      "learning_rate": 1.5500906602706756e-07,
      "loss": 1.1279,
      "num_input_tokens_seen": 75413384,
      "step": 129995
    },
    {
      "epoch": 19.362526064938933,
      "grad_norm": 13.1875,
      "learning_rate": 1.546479863890199e-07,
      "loss": 0.9534,
      "num_input_tokens_seen": 75416296,
      "step": 130000
    },
    {
      "epoch": 19.363270777479894,
      "grad_norm": 9.0,
      "learning_rate": 1.542873264879424e-07,
      "loss": 1.1623,
      "num_input_tokens_seen": 75419112,
      "step": 130005
    },
    {
      "epoch": 19.36401549002085,
      "grad_norm": 19.125,
      "learning_rate": 1.5392708632992748e-07,
      "loss": 1.537,
      "num_input_tokens_seen": 75422280,
      "step": 130010
    },
    {
      "epoch": 19.36476020256181,
      "grad_norm": 5.6875,
      "learning_rate": 1.5356726592106185e-07,
      "loss": 0.7823,
      "num_input_tokens_seen": 75425544,
      "step": 130015
    },
    {
      "epoch": 19.365504915102772,
      "grad_norm": 9.1875,
      "learning_rate": 1.5320786526742682e-07,
      "loss": 1.1575,
      "num_input_tokens_seen": 75428424,
      "step": 130020
    },
    {
      "epoch": 19.36624962764373,
      "grad_norm": 11.8125,
      "learning_rate": 1.5284888437508972e-07,
      "loss": 0.9651,
      "num_input_tokens_seen": 75431304,
      "step": 130025
    },
    {
      "epoch": 19.36699434018469,
      "grad_norm": 14.4375,
      "learning_rate": 1.5249032325011514e-07,
      "loss": 1.1703,
      "num_input_tokens_seen": 75434184,
      "step": 130030
    },
    {
      "epoch": 19.367739052725646,
      "grad_norm": 9.5,
      "learning_rate": 1.5213218189856492e-07,
      "loss": 1.0237,
      "num_input_tokens_seen": 75436936,
      "step": 130035
    },
    {
      "epoch": 19.368483765266607,
      "grad_norm": 12.625,
      "learning_rate": 1.5177446032648702e-07,
      "loss": 1.0834,
      "num_input_tokens_seen": 75440168,
      "step": 130040
    },
    {
      "epoch": 19.369228477807567,
      "grad_norm": 9.875,
      "learning_rate": 1.5141715853992654e-07,
      "loss": 0.8382,
      "num_input_tokens_seen": 75442760,
      "step": 130045
    },
    {
      "epoch": 19.369973190348524,
      "grad_norm": 11.6875,
      "learning_rate": 1.510602765449176e-07,
      "loss": 0.8792,
      "num_input_tokens_seen": 75446024,
      "step": 130050
    },
    {
      "epoch": 19.370717902889485,
      "grad_norm": 17.5,
      "learning_rate": 1.507038143474887e-07,
      "loss": 0.8962,
      "num_input_tokens_seen": 75448712,
      "step": 130055
    },
    {
      "epoch": 19.371462615430445,
      "grad_norm": 13.625,
      "learning_rate": 1.5034777195366278e-07,
      "loss": 1.0745,
      "num_input_tokens_seen": 75451464,
      "step": 130060
    },
    {
      "epoch": 19.372207327971402,
      "grad_norm": 4.75,
      "learning_rate": 1.4999214936945726e-07,
      "loss": 1.1614,
      "num_input_tokens_seen": 75454504,
      "step": 130065
    },
    {
      "epoch": 19.372952040512363,
      "grad_norm": 12.5,
      "learning_rate": 1.496369466008757e-07,
      "loss": 1.2799,
      "num_input_tokens_seen": 75457640,
      "step": 130070
    },
    {
      "epoch": 19.37369675305332,
      "grad_norm": 12.375,
      "learning_rate": 1.4928216365392157e-07,
      "loss": 1.0524,
      "num_input_tokens_seen": 75460584,
      "step": 130075
    },
    {
      "epoch": 19.37444146559428,
      "grad_norm": 9.75,
      "learning_rate": 1.489278005345901e-07,
      "loss": 0.9335,
      "num_input_tokens_seen": 75463240,
      "step": 130080
    },
    {
      "epoch": 19.37518617813524,
      "grad_norm": 14.0625,
      "learning_rate": 1.485738572488654e-07,
      "loss": 0.9332,
      "num_input_tokens_seen": 75466376,
      "step": 130085
    },
    {
      "epoch": 19.375930890676198,
      "grad_norm": 7.84375,
      "learning_rate": 1.4822033380272603e-07,
      "loss": 0.8975,
      "num_input_tokens_seen": 75469160,
      "step": 130090
    },
    {
      "epoch": 19.37667560321716,
      "grad_norm": 9.3125,
      "learning_rate": 1.47867230202145e-07,
      "loss": 1.0552,
      "num_input_tokens_seen": 75471720,
      "step": 130095
    },
    {
      "epoch": 19.37742031575812,
      "grad_norm": 9.625,
      "learning_rate": 1.4751454645309248e-07,
      "loss": 1.2805,
      "num_input_tokens_seen": 75475048,
      "step": 130100
    },
    {
      "epoch": 19.378165028299076,
      "grad_norm": 12.0,
      "learning_rate": 1.471622825615193e-07,
      "loss": 1.0633,
      "num_input_tokens_seen": 75478216,
      "step": 130105
    },
    {
      "epoch": 19.378909740840037,
      "grad_norm": 10.1875,
      "learning_rate": 1.4681043853338184e-07,
      "loss": 1.0232,
      "num_input_tokens_seen": 75481064,
      "step": 130110
    },
    {
      "epoch": 19.379654453380994,
      "grad_norm": 10.8125,
      "learning_rate": 1.4645901437461972e-07,
      "loss": 1.1434,
      "num_input_tokens_seen": 75483752,
      "step": 130115
    },
    {
      "epoch": 19.380399165921954,
      "grad_norm": 12.375,
      "learning_rate": 1.4610801009117548e-07,
      "loss": 1.035,
      "num_input_tokens_seen": 75486472,
      "step": 130120
    },
    {
      "epoch": 19.381143878462915,
      "grad_norm": 20.125,
      "learning_rate": 1.4575742568897488e-07,
      "loss": 1.2459,
      "num_input_tokens_seen": 75489320,
      "step": 130125
    },
    {
      "epoch": 19.38188859100387,
      "grad_norm": 16.875,
      "learning_rate": 1.45407261173941e-07,
      "loss": 1.0934,
      "num_input_tokens_seen": 75492392,
      "step": 130130
    },
    {
      "epoch": 19.382633303544832,
      "grad_norm": 12.625,
      "learning_rate": 1.4505751655199405e-07,
      "loss": 0.9188,
      "num_input_tokens_seen": 75495080,
      "step": 130135
    },
    {
      "epoch": 19.38337801608579,
      "grad_norm": 11.8125,
      "learning_rate": 1.4470819182903493e-07,
      "loss": 0.741,
      "num_input_tokens_seen": 75498024,
      "step": 130140
    },
    {
      "epoch": 19.38412272862675,
      "grad_norm": 9.25,
      "learning_rate": 1.443592870109728e-07,
      "loss": 1.5909,
      "num_input_tokens_seen": 75500872,
      "step": 130145
    },
    {
      "epoch": 19.38486744116771,
      "grad_norm": 9.3125,
      "learning_rate": 1.4401080210369454e-07,
      "loss": 0.9929,
      "num_input_tokens_seen": 75503656,
      "step": 130150
    },
    {
      "epoch": 19.385612153708667,
      "grad_norm": 11.8125,
      "learning_rate": 1.4366273711309275e-07,
      "loss": 0.8928,
      "num_input_tokens_seen": 75506600,
      "step": 130155
    },
    {
      "epoch": 19.386356866249628,
      "grad_norm": 9.125,
      "learning_rate": 1.43315092045046e-07,
      "loss": 1.5702,
      "num_input_tokens_seen": 75509704,
      "step": 130160
    },
    {
      "epoch": 19.38710157879059,
      "grad_norm": 10.3125,
      "learning_rate": 1.429678669054274e-07,
      "loss": 1.3474,
      "num_input_tokens_seen": 75512872,
      "step": 130165
    },
    {
      "epoch": 19.387846291331545,
      "grad_norm": 9.25,
      "learning_rate": 1.4262106170010447e-07,
      "loss": 1.1728,
      "num_input_tokens_seen": 75516328,
      "step": 130170
    },
    {
      "epoch": 19.388591003872506,
      "grad_norm": 9.75,
      "learning_rate": 1.4227467643493364e-07,
      "loss": 1.0225,
      "num_input_tokens_seen": 75519304,
      "step": 130175
    },
    {
      "epoch": 19.389335716413463,
      "grad_norm": 17.75,
      "learning_rate": 1.4192871111576856e-07,
      "loss": 1.0446,
      "num_input_tokens_seen": 75522120,
      "step": 130180
    },
    {
      "epoch": 19.390080428954423,
      "grad_norm": 11.375,
      "learning_rate": 1.4158316574845175e-07,
      "loss": 0.7369,
      "num_input_tokens_seen": 75524968,
      "step": 130185
    },
    {
      "epoch": 19.390825141495384,
      "grad_norm": 12.6875,
      "learning_rate": 1.4123804033882305e-07,
      "loss": 1.3411,
      "num_input_tokens_seen": 75527976,
      "step": 130190
    },
    {
      "epoch": 19.39156985403634,
      "grad_norm": 4.375,
      "learning_rate": 1.4089333489271384e-07,
      "loss": 1.0559,
      "num_input_tokens_seen": 75530984,
      "step": 130195
    },
    {
      "epoch": 19.3923145665773,
      "grad_norm": 9.8125,
      "learning_rate": 1.405490494159445e-07,
      "loss": 1.0207,
      "num_input_tokens_seen": 75534056,
      "step": 130200
    },
    {
      "epoch": 19.393059279118262,
      "grad_norm": 5.5,
      "learning_rate": 1.4020518391433258e-07,
      "loss": 0.742,
      "num_input_tokens_seen": 75536936,
      "step": 130205
    },
    {
      "epoch": 19.39380399165922,
      "grad_norm": 9.4375,
      "learning_rate": 1.398617383936901e-07,
      "loss": 0.8853,
      "num_input_tokens_seen": 75540072,
      "step": 130210
    },
    {
      "epoch": 19.39454870420018,
      "grad_norm": 14.375,
      "learning_rate": 1.395187128598152e-07,
      "loss": 1.055,
      "num_input_tokens_seen": 75542888,
      "step": 130215
    },
    {
      "epoch": 19.395293416741136,
      "grad_norm": 4.8125,
      "learning_rate": 1.3917610731850328e-07,
      "loss": 0.9259,
      "num_input_tokens_seen": 75545768,
      "step": 130220
    },
    {
      "epoch": 19.396038129282097,
      "grad_norm": 9.5625,
      "learning_rate": 1.3883392177554688e-07,
      "loss": 1.0828,
      "num_input_tokens_seen": 75548424,
      "step": 130225
    },
    {
      "epoch": 19.396782841823057,
      "grad_norm": 9.5,
      "learning_rate": 1.3849215623672197e-07,
      "loss": 1.0647,
      "num_input_tokens_seen": 75551656,
      "step": 130230
    },
    {
      "epoch": 19.397527554364014,
      "grad_norm": 12.4375,
      "learning_rate": 1.3815081070780167e-07,
      "loss": 1.2692,
      "num_input_tokens_seen": 75554536,
      "step": 130235
    },
    {
      "epoch": 19.398272266904975,
      "grad_norm": 13.625,
      "learning_rate": 1.378098851945564e-07,
      "loss": 0.7676,
      "num_input_tokens_seen": 75557288,
      "step": 130240
    },
    {
      "epoch": 19.399016979445936,
      "grad_norm": 10.6875,
      "learning_rate": 1.3746937970274543e-07,
      "loss": 1.0412,
      "num_input_tokens_seen": 75560424,
      "step": 130245
    },
    {
      "epoch": 19.399761691986892,
      "grad_norm": 20.0,
      "learning_rate": 1.3712929423812247e-07,
      "loss": 0.8981,
      "num_input_tokens_seen": 75563304,
      "step": 130250
    },
    {
      "epoch": 19.400506404527853,
      "grad_norm": 9.5625,
      "learning_rate": 1.3678962880642465e-07,
      "loss": 1.4079,
      "num_input_tokens_seen": 75566056,
      "step": 130255
    },
    {
      "epoch": 19.40125111706881,
      "grad_norm": 7.65625,
      "learning_rate": 1.3645038341340011e-07,
      "loss": 0.9952,
      "num_input_tokens_seen": 75568968,
      "step": 130260
    },
    {
      "epoch": 19.40199582960977,
      "grad_norm": 13.1875,
      "learning_rate": 1.361115580647748e-07,
      "loss": 1.279,
      "num_input_tokens_seen": 75571976,
      "step": 130265
    },
    {
      "epoch": 19.40274054215073,
      "grad_norm": 12.5,
      "learning_rate": 1.357731527662748e-07,
      "loss": 1.2271,
      "num_input_tokens_seen": 75574792,
      "step": 130270
    },
    {
      "epoch": 19.403485254691688,
      "grad_norm": 13.4375,
      "learning_rate": 1.3543516752361763e-07,
      "loss": 1.1865,
      "num_input_tokens_seen": 75578024,
      "step": 130275
    },
    {
      "epoch": 19.40422996723265,
      "grad_norm": 17.0,
      "learning_rate": 1.3509760234251267e-07,
      "loss": 1.4413,
      "num_input_tokens_seen": 75580936,
      "step": 130280
    },
    {
      "epoch": 19.40497467977361,
      "grad_norm": 13.375,
      "learning_rate": 1.3476045722865815e-07,
      "loss": 1.6315,
      "num_input_tokens_seen": 75583752,
      "step": 130285
    },
    {
      "epoch": 19.405719392314566,
      "grad_norm": 19.75,
      "learning_rate": 1.3442373218775784e-07,
      "loss": 1.3446,
      "num_input_tokens_seen": 75586696,
      "step": 130290
    },
    {
      "epoch": 19.406464104855527,
      "grad_norm": 12.4375,
      "learning_rate": 1.340874272254933e-07,
      "loss": 1.0728,
      "num_input_tokens_seen": 75589736,
      "step": 130295
    },
    {
      "epoch": 19.407208817396484,
      "grad_norm": 8.75,
      "learning_rate": 1.3375154234755162e-07,
      "loss": 0.9209,
      "num_input_tokens_seen": 75592680,
      "step": 130300
    },
    {
      "epoch": 19.407953529937444,
      "grad_norm": 5.84375,
      "learning_rate": 1.3341607755960327e-07,
      "loss": 0.8486,
      "num_input_tokens_seen": 75595464,
      "step": 130305
    },
    {
      "epoch": 19.408698242478405,
      "grad_norm": 18.625,
      "learning_rate": 1.3308103286731598e-07,
      "loss": 1.3966,
      "num_input_tokens_seen": 75598344,
      "step": 130310
    },
    {
      "epoch": 19.40944295501936,
      "grad_norm": 16.125,
      "learning_rate": 1.3274640827635187e-07,
      "loss": 1.0413,
      "num_input_tokens_seen": 75601288,
      "step": 130315
    },
    {
      "epoch": 19.410187667560322,
      "grad_norm": 14.125,
      "learning_rate": 1.3241220379236473e-07,
      "loss": 1.2621,
      "num_input_tokens_seen": 75603912,
      "step": 130320
    },
    {
      "epoch": 19.41093238010128,
      "grad_norm": 13.0625,
      "learning_rate": 1.320784194209973e-07,
      "loss": 0.8807,
      "num_input_tokens_seen": 75606824,
      "step": 130325
    },
    {
      "epoch": 19.41167709264224,
      "grad_norm": 11.25,
      "learning_rate": 1.3174505516789226e-07,
      "loss": 1.1462,
      "num_input_tokens_seen": 75609512,
      "step": 130330
    },
    {
      "epoch": 19.4124218051832,
      "grad_norm": 10.4375,
      "learning_rate": 1.3141211103867845e-07,
      "loss": 1.2528,
      "num_input_tokens_seen": 75612520,
      "step": 130335
    },
    {
      "epoch": 19.413166517724157,
      "grad_norm": 4.25,
      "learning_rate": 1.3107958703898193e-07,
      "loss": 0.9075,
      "num_input_tokens_seen": 75615688,
      "step": 130340
    },
    {
      "epoch": 19.413911230265118,
      "grad_norm": 4.875,
      "learning_rate": 1.3074748317442042e-07,
      "loss": 1.2442,
      "num_input_tokens_seen": 75618504,
      "step": 130345
    },
    {
      "epoch": 19.41465594280608,
      "grad_norm": 14.5625,
      "learning_rate": 1.3041579945060335e-07,
      "loss": 1.2394,
      "num_input_tokens_seen": 75621384,
      "step": 130350
    },
    {
      "epoch": 19.415400655347035,
      "grad_norm": 19.375,
      "learning_rate": 1.3008453587313453e-07,
      "loss": 1.2478,
      "num_input_tokens_seen": 75624328,
      "step": 130355
    },
    {
      "epoch": 19.416145367887996,
      "grad_norm": 23.5,
      "learning_rate": 1.2975369244761226e-07,
      "loss": 1.2021,
      "num_input_tokens_seen": 75627496,
      "step": 130360
    },
    {
      "epoch": 19.416890080428953,
      "grad_norm": 11.625,
      "learning_rate": 1.2942326917962377e-07,
      "loss": 1.1702,
      "num_input_tokens_seen": 75630440,
      "step": 130365
    },
    {
      "epoch": 19.417634792969913,
      "grad_norm": 18.625,
      "learning_rate": 1.290932660747507e-07,
      "loss": 1.0337,
      "num_input_tokens_seen": 75633192,
      "step": 130370
    },
    {
      "epoch": 19.418379505510874,
      "grad_norm": 7.53125,
      "learning_rate": 1.287636831385719e-07,
      "loss": 1.4816,
      "num_input_tokens_seen": 75636072,
      "step": 130375
    },
    {
      "epoch": 19.41912421805183,
      "grad_norm": 13.5,
      "learning_rate": 1.2843452037664962e-07,
      "loss": 1.2417,
      "num_input_tokens_seen": 75638760,
      "step": 130380
    },
    {
      "epoch": 19.41986893059279,
      "grad_norm": 14.1875,
      "learning_rate": 1.281057777945488e-07,
      "loss": 0.9866,
      "num_input_tokens_seen": 75641832,
      "step": 130385
    },
    {
      "epoch": 19.420613643133752,
      "grad_norm": 6.875,
      "learning_rate": 1.2777745539782337e-07,
      "loss": 1.3101,
      "num_input_tokens_seen": 75644776,
      "step": 130390
    },
    {
      "epoch": 19.42135835567471,
      "grad_norm": 5.59375,
      "learning_rate": 1.274495531920189e-07,
      "loss": 1.1991,
      "num_input_tokens_seen": 75647976,
      "step": 130395
    },
    {
      "epoch": 19.42210306821567,
      "grad_norm": 11.0625,
      "learning_rate": 1.2712207118267262e-07,
      "loss": 1.2954,
      "num_input_tokens_seen": 75650856,
      "step": 130400
    },
    {
      "epoch": 19.422847780756626,
      "grad_norm": 14.4375,
      "learning_rate": 1.2679500937532173e-07,
      "loss": 1.3448,
      "num_input_tokens_seen": 75653736,
      "step": 130405
    },
    {
      "epoch": 19.423592493297587,
      "grad_norm": 11.0,
      "learning_rate": 1.2646836777548688e-07,
      "loss": 1.0356,
      "num_input_tokens_seen": 75656840,
      "step": 130410
    },
    {
      "epoch": 19.424337205838548,
      "grad_norm": 4.46875,
      "learning_rate": 1.2614214638869137e-07,
      "loss": 0.7607,
      "num_input_tokens_seen": 75659848,
      "step": 130415
    },
    {
      "epoch": 19.425081918379504,
      "grad_norm": 15.0,
      "learning_rate": 1.2581634522044194e-07,
      "loss": 1.2158,
      "num_input_tokens_seen": 75662792,
      "step": 130420
    },
    {
      "epoch": 19.425826630920465,
      "grad_norm": 7.5625,
      "learning_rate": 1.254909642762453e-07,
      "loss": 0.8384,
      "num_input_tokens_seen": 75665704,
      "step": 130425
    },
    {
      "epoch": 19.426571343461426,
      "grad_norm": 6.625,
      "learning_rate": 1.2516600356159701e-07,
      "loss": 1.1166,
      "num_input_tokens_seen": 75668456,
      "step": 130430
    },
    {
      "epoch": 19.427316056002383,
      "grad_norm": 6.53125,
      "learning_rate": 1.248414630819872e-07,
      "loss": 1.043,
      "num_input_tokens_seen": 75671144,
      "step": 130435
    },
    {
      "epoch": 19.428060768543343,
      "grad_norm": 21.5,
      "learning_rate": 1.2451734284289752e-07,
      "loss": 1.1067,
      "num_input_tokens_seen": 75673768,
      "step": 130440
    },
    {
      "epoch": 19.4288054810843,
      "grad_norm": 10.375,
      "learning_rate": 1.2419364284980696e-07,
      "loss": 0.958,
      "num_input_tokens_seen": 75676552,
      "step": 130445
    },
    {
      "epoch": 19.42955019362526,
      "grad_norm": 13.4375,
      "learning_rate": 1.2387036310818334e-07,
      "loss": 1.1148,
      "num_input_tokens_seen": 75679432,
      "step": 130450
    },
    {
      "epoch": 19.43029490616622,
      "grad_norm": 10.125,
      "learning_rate": 1.2354750362348344e-07,
      "loss": 0.8306,
      "num_input_tokens_seen": 75682152,
      "step": 130455
    },
    {
      "epoch": 19.431039618707178,
      "grad_norm": 9.5,
      "learning_rate": 1.2322506440116676e-07,
      "loss": 1.3863,
      "num_input_tokens_seen": 75685096,
      "step": 130460
    },
    {
      "epoch": 19.43178433124814,
      "grad_norm": 15.5,
      "learning_rate": 1.2290304544668174e-07,
      "loss": 0.9335,
      "num_input_tokens_seen": 75687880,
      "step": 130465
    },
    {
      "epoch": 19.432529043789096,
      "grad_norm": 16.375,
      "learning_rate": 1.2258144676546291e-07,
      "loss": 1.1959,
      "num_input_tokens_seen": 75690888,
      "step": 130470
    },
    {
      "epoch": 19.433273756330056,
      "grad_norm": 11.8125,
      "learning_rate": 1.2226026836294756e-07,
      "loss": 1.4108,
      "num_input_tokens_seen": 75693864,
      "step": 130475
    },
    {
      "epoch": 19.434018468871017,
      "grad_norm": 13.3125,
      "learning_rate": 1.2193951024455918e-07,
      "loss": 1.2735,
      "num_input_tokens_seen": 75696552,
      "step": 130480
    },
    {
      "epoch": 19.434763181411974,
      "grad_norm": 10.125,
      "learning_rate": 1.216191724157184e-07,
      "loss": 0.7756,
      "num_input_tokens_seen": 75699592,
      "step": 130485
    },
    {
      "epoch": 19.435507893952934,
      "grad_norm": 11.0,
      "learning_rate": 1.212992548818348e-07,
      "loss": 1.2257,
      "num_input_tokens_seen": 75702408,
      "step": 130490
    },
    {
      "epoch": 19.436252606493895,
      "grad_norm": 8.5,
      "learning_rate": 1.2097975764831516e-07,
      "loss": 1.0826,
      "num_input_tokens_seen": 75705448,
      "step": 130495
    },
    {
      "epoch": 19.43699731903485,
      "grad_norm": 11.3125,
      "learning_rate": 1.206606807205579e-07,
      "loss": 1.0752,
      "num_input_tokens_seen": 75708168,
      "step": 130500
    },
    {
      "epoch": 19.437742031575812,
      "grad_norm": 9.0,
      "learning_rate": 1.2034202410395324e-07,
      "loss": 1.1518,
      "num_input_tokens_seen": 75710984,
      "step": 130505
    },
    {
      "epoch": 19.43848674411677,
      "grad_norm": 9.25,
      "learning_rate": 1.200237878038829e-07,
      "loss": 1.1783,
      "num_input_tokens_seen": 75714024,
      "step": 130510
    },
    {
      "epoch": 19.43923145665773,
      "grad_norm": 13.375,
      "learning_rate": 1.197059718257204e-07,
      "loss": 1.0778,
      "num_input_tokens_seen": 75717192,
      "step": 130515
    },
    {
      "epoch": 19.43997616919869,
      "grad_norm": 8.5,
      "learning_rate": 1.19388576174842e-07,
      "loss": 1.3266,
      "num_input_tokens_seen": 75720200,
      "step": 130520
    },
    {
      "epoch": 19.440720881739647,
      "grad_norm": 18.375,
      "learning_rate": 1.1907160085660451e-07,
      "loss": 1.2204,
      "num_input_tokens_seen": 75723176,
      "step": 130525
    },
    {
      "epoch": 19.441465594280608,
      "grad_norm": 16.125,
      "learning_rate": 1.1875504587636477e-07,
      "loss": 1.1067,
      "num_input_tokens_seen": 75726056,
      "step": 130530
    },
    {
      "epoch": 19.44221030682157,
      "grad_norm": 8.5625,
      "learning_rate": 1.1843891123947126e-07,
      "loss": 1.0497,
      "num_input_tokens_seen": 75728872,
      "step": 130535
    },
    {
      "epoch": 19.442955019362525,
      "grad_norm": 6.0625,
      "learning_rate": 1.1812319695126416e-07,
      "loss": 1.14,
      "num_input_tokens_seen": 75731944,
      "step": 130540
    },
    {
      "epoch": 19.443699731903486,
      "grad_norm": 7.46875,
      "learning_rate": 1.1780790301707533e-07,
      "loss": 1.1083,
      "num_input_tokens_seen": 75734824,
      "step": 130545
    },
    {
      "epoch": 19.444444444444443,
      "grad_norm": 11.0625,
      "learning_rate": 1.1749302944223384e-07,
      "loss": 1.0673,
      "num_input_tokens_seen": 75737960,
      "step": 130550
    },
    {
      "epoch": 19.445189156985403,
      "grad_norm": 11.5625,
      "learning_rate": 1.1717857623205764e-07,
      "loss": 0.8013,
      "num_input_tokens_seen": 75740648,
      "step": 130555
    },
    {
      "epoch": 19.445933869526364,
      "grad_norm": 20.625,
      "learning_rate": 1.1686454339185915e-07,
      "loss": 1.076,
      "num_input_tokens_seen": 75743720,
      "step": 130560
    },
    {
      "epoch": 19.44667858206732,
      "grad_norm": 7.1875,
      "learning_rate": 1.1655093092694525e-07,
      "loss": 1.1659,
      "num_input_tokens_seen": 75746760,
      "step": 130565
    },
    {
      "epoch": 19.44742329460828,
      "grad_norm": 7.21875,
      "learning_rate": 1.1623773884261169e-07,
      "loss": 1.0852,
      "num_input_tokens_seen": 75749768,
      "step": 130570
    },
    {
      "epoch": 19.448168007149242,
      "grad_norm": 10.0,
      "learning_rate": 1.1592496714415147e-07,
      "loss": 1.1628,
      "num_input_tokens_seen": 75752552,
      "step": 130575
    },
    {
      "epoch": 19.4489127196902,
      "grad_norm": 10.875,
      "learning_rate": 1.1561261583684924e-07,
      "loss": 1.1789,
      "num_input_tokens_seen": 75755240,
      "step": 130580
    },
    {
      "epoch": 19.44965743223116,
      "grad_norm": 14.5625,
      "learning_rate": 1.1530068492597856e-07,
      "loss": 1.3059,
      "num_input_tokens_seen": 75758344,
      "step": 130585
    },
    {
      "epoch": 19.450402144772116,
      "grad_norm": 11.0625,
      "learning_rate": 1.1498917441681023e-07,
      "loss": 1.0097,
      "num_input_tokens_seen": 75761736,
      "step": 130590
    },
    {
      "epoch": 19.451146857313077,
      "grad_norm": 14.5,
      "learning_rate": 1.1467808431460947e-07,
      "loss": 1.1184,
      "num_input_tokens_seen": 75764328,
      "step": 130595
    },
    {
      "epoch": 19.451891569854038,
      "grad_norm": 16.0,
      "learning_rate": 1.143674146246304e-07,
      "loss": 1.1294,
      "num_input_tokens_seen": 75767208,
      "step": 130600
    },
    {
      "epoch": 19.452636282394995,
      "grad_norm": 15.6875,
      "learning_rate": 1.1405716535212163e-07,
      "loss": 1.073,
      "num_input_tokens_seen": 75770152,
      "step": 130605
    },
    {
      "epoch": 19.453380994935955,
      "grad_norm": 13.9375,
      "learning_rate": 1.1374733650232338e-07,
      "loss": 1.4042,
      "num_input_tokens_seen": 75772904,
      "step": 130610
    },
    {
      "epoch": 19.454125707476916,
      "grad_norm": 8.4375,
      "learning_rate": 1.1343792808047038e-07,
      "loss": 1.0686,
      "num_input_tokens_seen": 75775880,
      "step": 130615
    },
    {
      "epoch": 19.454870420017873,
      "grad_norm": 12.1875,
      "learning_rate": 1.1312894009179176e-07,
      "loss": 1.0683,
      "num_input_tokens_seen": 75778856,
      "step": 130620
    },
    {
      "epoch": 19.455615132558833,
      "grad_norm": 7.96875,
      "learning_rate": 1.1282037254150279e-07,
      "loss": 0.9855,
      "num_input_tokens_seen": 75781832,
      "step": 130625
    },
    {
      "epoch": 19.45635984509979,
      "grad_norm": 11.6875,
      "learning_rate": 1.1251222543482154e-07,
      "loss": 1.0236,
      "num_input_tokens_seen": 75784936,
      "step": 130630
    },
    {
      "epoch": 19.45710455764075,
      "grad_norm": 7.71875,
      "learning_rate": 1.1220449877694938e-07,
      "loss": 1.0185,
      "num_input_tokens_seen": 75788200,
      "step": 130635
    },
    {
      "epoch": 19.45784927018171,
      "grad_norm": 8.5625,
      "learning_rate": 1.1189719257309051e-07,
      "loss": 1.0308,
      "num_input_tokens_seen": 75791176,
      "step": 130640
    },
    {
      "epoch": 19.458593982722668,
      "grad_norm": 9.6875,
      "learning_rate": 1.1159030682843242e-07,
      "loss": 1.1517,
      "num_input_tokens_seen": 75793832,
      "step": 130645
    },
    {
      "epoch": 19.45933869526363,
      "grad_norm": 6.8125,
      "learning_rate": 1.1128384154815984e-07,
      "loss": 0.8878,
      "num_input_tokens_seen": 75797064,
      "step": 130650
    },
    {
      "epoch": 19.46008340780459,
      "grad_norm": 8.1875,
      "learning_rate": 1.1097779673745201e-07,
      "loss": 0.8773,
      "num_input_tokens_seen": 75799592,
      "step": 130655
    },
    {
      "epoch": 19.460828120345546,
      "grad_norm": 9.3125,
      "learning_rate": 1.1067217240147698e-07,
      "loss": 0.9995,
      "num_input_tokens_seen": 75802600,
      "step": 130660
    },
    {
      "epoch": 19.461572832886507,
      "grad_norm": 10.375,
      "learning_rate": 1.1036696854540007e-07,
      "loss": 1.1696,
      "num_input_tokens_seen": 75805512,
      "step": 130665
    },
    {
      "epoch": 19.462317545427464,
      "grad_norm": 10.125,
      "learning_rate": 1.100621851743755e-07,
      "loss": 1.1887,
      "num_input_tokens_seen": 75808264,
      "step": 130670
    },
    {
      "epoch": 19.463062257968424,
      "grad_norm": 7.53125,
      "learning_rate": 1.0975782229355469e-07,
      "loss": 1.1965,
      "num_input_tokens_seen": 75811432,
      "step": 130675
    },
    {
      "epoch": 19.463806970509385,
      "grad_norm": 7.21875,
      "learning_rate": 1.0945387990807798e-07,
      "loss": 1.2921,
      "num_input_tokens_seen": 75814216,
      "step": 130680
    },
    {
      "epoch": 19.464551683050342,
      "grad_norm": 10.4375,
      "learning_rate": 1.0915035802308016e-07,
      "loss": 1.2081,
      "num_input_tokens_seen": 75817672,
      "step": 130685
    },
    {
      "epoch": 19.465296395591302,
      "grad_norm": 22.25,
      "learning_rate": 1.0884725664368766e-07,
      "loss": 1.3756,
      "num_input_tokens_seen": 75820840,
      "step": 130690
    },
    {
      "epoch": 19.46604110813226,
      "grad_norm": 7.5625,
      "learning_rate": 1.0854457577502419e-07,
      "loss": 0.9045,
      "num_input_tokens_seen": 75823720,
      "step": 130695
    },
    {
      "epoch": 19.46678582067322,
      "grad_norm": 10.0,
      "learning_rate": 1.0824231542220232e-07,
      "loss": 1.3382,
      "num_input_tokens_seen": 75826664,
      "step": 130700
    },
    {
      "epoch": 19.46753053321418,
      "grad_norm": 6.90625,
      "learning_rate": 1.0794047559032627e-07,
      "loss": 1.0435,
      "num_input_tokens_seen": 75829416,
      "step": 130705
    },
    {
      "epoch": 19.468275245755137,
      "grad_norm": 6.59375,
      "learning_rate": 1.0763905628449478e-07,
      "loss": 1.189,
      "num_input_tokens_seen": 75832520,
      "step": 130710
    },
    {
      "epoch": 19.469019958296098,
      "grad_norm": 16.625,
      "learning_rate": 1.0733805750980653e-07,
      "loss": 1.4153,
      "num_input_tokens_seen": 75835336,
      "step": 130715
    },
    {
      "epoch": 19.46976467083706,
      "grad_norm": 7.25,
      "learning_rate": 1.07037479271338e-07,
      "loss": 1.0391,
      "num_input_tokens_seen": 75838152,
      "step": 130720
    },
    {
      "epoch": 19.470509383378015,
      "grad_norm": 10.4375,
      "learning_rate": 1.0673732157417404e-07,
      "loss": 0.9844,
      "num_input_tokens_seen": 75841096,
      "step": 130725
    },
    {
      "epoch": 19.471254095918976,
      "grad_norm": 7.25,
      "learning_rate": 1.0643758442338004e-07,
      "loss": 1.0749,
      "num_input_tokens_seen": 75843976,
      "step": 130730
    },
    {
      "epoch": 19.471998808459933,
      "grad_norm": 12.5625,
      "learning_rate": 1.0613826782402414e-07,
      "loss": 0.8901,
      "num_input_tokens_seen": 75846952,
      "step": 130735
    },
    {
      "epoch": 19.472743521000893,
      "grad_norm": 5.5,
      "learning_rate": 1.0583937178116066e-07,
      "loss": 0.8397,
      "num_input_tokens_seen": 75850152,
      "step": 130740
    },
    {
      "epoch": 19.473488233541854,
      "grad_norm": 9.4375,
      "learning_rate": 1.055408962998411e-07,
      "loss": 1.3303,
      "num_input_tokens_seen": 75852872,
      "step": 130745
    },
    {
      "epoch": 19.47423294608281,
      "grad_norm": 21.75,
      "learning_rate": 1.0524284138510588e-07,
      "loss": 1.0075,
      "num_input_tokens_seen": 75855912,
      "step": 130750
    },
    {
      "epoch": 19.47497765862377,
      "grad_norm": 11.375,
      "learning_rate": 1.0494520704198985e-07,
      "loss": 1.073,
      "num_input_tokens_seen": 75858568,
      "step": 130755
    },
    {
      "epoch": 19.475722371164732,
      "grad_norm": 12.1875,
      "learning_rate": 1.0464799327552232e-07,
      "loss": 1.3657,
      "num_input_tokens_seen": 75861576,
      "step": 130760
    },
    {
      "epoch": 19.47646708370569,
      "grad_norm": 14.375,
      "learning_rate": 1.043512000907243e-07,
      "loss": 1.2503,
      "num_input_tokens_seen": 75864744,
      "step": 130765
    },
    {
      "epoch": 19.47721179624665,
      "grad_norm": 8.8125,
      "learning_rate": 1.040548274926112e-07,
      "loss": 1.0427,
      "num_input_tokens_seen": 75867784,
      "step": 130770
    },
    {
      "epoch": 19.477956508787607,
      "grad_norm": 8.75,
      "learning_rate": 1.0375887548618735e-07,
      "loss": 0.7871,
      "num_input_tokens_seen": 75870888,
      "step": 130775
    },
    {
      "epoch": 19.478701221328567,
      "grad_norm": 8.125,
      "learning_rate": 1.034633440764543e-07,
      "loss": 1.0801,
      "num_input_tokens_seen": 75873928,
      "step": 130780
    },
    {
      "epoch": 19.479445933869528,
      "grad_norm": 14.625,
      "learning_rate": 1.031682332684053e-07,
      "loss": 1.0825,
      "num_input_tokens_seen": 75877096,
      "step": 130785
    },
    {
      "epoch": 19.480190646410485,
      "grad_norm": 12.25,
      "learning_rate": 1.0287354306702524e-07,
      "loss": 0.8669,
      "num_input_tokens_seen": 75880104,
      "step": 130790
    },
    {
      "epoch": 19.480935358951445,
      "grad_norm": 12.1875,
      "learning_rate": 1.0257927347729068e-07,
      "loss": 0.8128,
      "num_input_tokens_seen": 75882760,
      "step": 130795
    },
    {
      "epoch": 19.481680071492406,
      "grad_norm": 12.5625,
      "learning_rate": 1.0228542450417545e-07,
      "loss": 1.2669,
      "num_input_tokens_seen": 75885608,
      "step": 130800
    },
    {
      "epoch": 19.482424784033363,
      "grad_norm": 10.1875,
      "learning_rate": 1.0199199615264499e-07,
      "loss": 0.9527,
      "num_input_tokens_seen": 75888424,
      "step": 130805
    },
    {
      "epoch": 19.483169496574323,
      "grad_norm": 6.71875,
      "learning_rate": 1.0169898842765091e-07,
      "loss": 0.835,
      "num_input_tokens_seen": 75891272,
      "step": 130810
    },
    {
      "epoch": 19.48391420911528,
      "grad_norm": 8.3125,
      "learning_rate": 1.0140640133415036e-07,
      "loss": 1.1723,
      "num_input_tokens_seen": 75894120,
      "step": 130815
    },
    {
      "epoch": 19.48465892165624,
      "grad_norm": 13.125,
      "learning_rate": 1.0111423487708105e-07,
      "loss": 0.9231,
      "num_input_tokens_seen": 75897256,
      "step": 130820
    },
    {
      "epoch": 19.4854036341972,
      "grad_norm": 11.375,
      "learning_rate": 1.008224890613807e-07,
      "loss": 1.0857,
      "num_input_tokens_seen": 75899912,
      "step": 130825
    },
    {
      "epoch": 19.486148346738158,
      "grad_norm": 7.125,
      "learning_rate": 1.0053116389197592e-07,
      "loss": 0.8993,
      "num_input_tokens_seen": 75902792,
      "step": 130830
    },
    {
      "epoch": 19.48689305927912,
      "grad_norm": 8.125,
      "learning_rate": 1.0024025937379333e-07,
      "loss": 0.8,
      "num_input_tokens_seen": 75905704,
      "step": 130835
    },
    {
      "epoch": 19.487637771820076,
      "grad_norm": 9.4375,
      "learning_rate": 9.994977551174289e-08,
      "loss": 1.2763,
      "num_input_tokens_seen": 75908744,
      "step": 130840
    },
    {
      "epoch": 19.488382484361036,
      "grad_norm": 10.1875,
      "learning_rate": 9.965971231073456e-08,
      "loss": 1.1777,
      "num_input_tokens_seen": 75911784,
      "step": 130845
    },
    {
      "epoch": 19.489127196901997,
      "grad_norm": 10.0,
      "learning_rate": 9.937006977566998e-08,
      "loss": 1.1181,
      "num_input_tokens_seen": 75914760,
      "step": 130850
    },
    {
      "epoch": 19.489871909442954,
      "grad_norm": 12.25,
      "learning_rate": 9.90808479114369e-08,
      "loss": 1.1006,
      "num_input_tokens_seen": 75917544,
      "step": 130855
    },
    {
      "epoch": 19.490616621983914,
      "grad_norm": 4.46875,
      "learning_rate": 9.879204672292586e-08,
      "loss": 0.8808,
      "num_input_tokens_seen": 75920616,
      "step": 130860
    },
    {
      "epoch": 19.491361334524875,
      "grad_norm": 9.375,
      "learning_rate": 9.850366621501628e-08,
      "loss": 0.9668,
      "num_input_tokens_seen": 75923560,
      "step": 130865
    },
    {
      "epoch": 19.492106047065832,
      "grad_norm": 10.3125,
      "learning_rate": 9.82157063925765e-08,
      "loss": 0.9018,
      "num_input_tokens_seen": 75926376,
      "step": 130870
    },
    {
      "epoch": 19.492850759606792,
      "grad_norm": 10.375,
      "learning_rate": 9.792816726047482e-08,
      "loss": 1.3036,
      "num_input_tokens_seen": 75929064,
      "step": 130875
    },
    {
      "epoch": 19.49359547214775,
      "grad_norm": 8.8125,
      "learning_rate": 9.764104882356572e-08,
      "loss": 1.0349,
      "num_input_tokens_seen": 75932040,
      "step": 130880
    },
    {
      "epoch": 19.49434018468871,
      "grad_norm": 12.25,
      "learning_rate": 9.735435108670088e-08,
      "loss": 0.8503,
      "num_input_tokens_seen": 75934888,
      "step": 130885
    },
    {
      "epoch": 19.49508489722967,
      "grad_norm": 9.5,
      "learning_rate": 9.70680740547264e-08,
      "loss": 1.0202,
      "num_input_tokens_seen": 75937640,
      "step": 130890
    },
    {
      "epoch": 19.495829609770627,
      "grad_norm": 12.3125,
      "learning_rate": 9.67822177324773e-08,
      "loss": 1.1604,
      "num_input_tokens_seen": 75940584,
      "step": 130895
    },
    {
      "epoch": 19.496574322311588,
      "grad_norm": 12.25,
      "learning_rate": 9.64967821247803e-08,
      "loss": 1.0903,
      "num_input_tokens_seen": 75944008,
      "step": 130900
    },
    {
      "epoch": 19.49731903485255,
      "grad_norm": 10.1875,
      "learning_rate": 9.621176723645931e-08,
      "loss": 1.0561,
      "num_input_tokens_seen": 75946920,
      "step": 130905
    },
    {
      "epoch": 19.498063747393505,
      "grad_norm": 8.75,
      "learning_rate": 9.59271730723299e-08,
      "loss": 1.0728,
      "num_input_tokens_seen": 75949704,
      "step": 130910
    },
    {
      "epoch": 19.498808459934466,
      "grad_norm": 12.375,
      "learning_rate": 9.564299963719936e-08,
      "loss": 0.9214,
      "num_input_tokens_seen": 75952392,
      "step": 130915
    },
    {
      "epoch": 19.499553172475423,
      "grad_norm": 16.75,
      "learning_rate": 9.53592469358694e-08,
      "loss": 1.1579,
      "num_input_tokens_seen": 75955240,
      "step": 130920
    },
    {
      "epoch": 19.500297885016384,
      "grad_norm": 9.125,
      "learning_rate": 9.507591497313063e-08,
      "loss": 1.1313,
      "num_input_tokens_seen": 75958536,
      "step": 130925
    },
    {
      "epoch": 19.501042597557344,
      "grad_norm": 6.59375,
      "learning_rate": 9.479300375377365e-08,
      "loss": 0.9528,
      "num_input_tokens_seen": 75961544,
      "step": 130930
    },
    {
      "epoch": 19.5017873100983,
      "grad_norm": 10.6875,
      "learning_rate": 9.451051328257799e-08,
      "loss": 0.9184,
      "num_input_tokens_seen": 75964392,
      "step": 130935
    },
    {
      "epoch": 19.50253202263926,
      "grad_norm": 13.75,
      "learning_rate": 9.422844356431481e-08,
      "loss": 1.1601,
      "num_input_tokens_seen": 75967400,
      "step": 130940
    },
    {
      "epoch": 19.503276735180222,
      "grad_norm": 16.125,
      "learning_rate": 9.3946794603747e-08,
      "loss": 1.3577,
      "num_input_tokens_seen": 75970312,
      "step": 130945
    },
    {
      "epoch": 19.50402144772118,
      "grad_norm": 12.0,
      "learning_rate": 9.366556640563462e-08,
      "loss": 0.9999,
      "num_input_tokens_seen": 75973320,
      "step": 130950
    },
    {
      "epoch": 19.50476616026214,
      "grad_norm": 5.65625,
      "learning_rate": 9.338475897472942e-08,
      "loss": 1.0592,
      "num_input_tokens_seen": 75976296,
      "step": 130955
    },
    {
      "epoch": 19.505510872803097,
      "grad_norm": 11.0,
      "learning_rate": 9.310437231577207e-08,
      "loss": 1.231,
      "num_input_tokens_seen": 75979272,
      "step": 130960
    },
    {
      "epoch": 19.506255585344057,
      "grad_norm": 7.6875,
      "learning_rate": 9.282440643350598e-08,
      "loss": 1.073,
      "num_input_tokens_seen": 75982024,
      "step": 130965
    },
    {
      "epoch": 19.507000297885018,
      "grad_norm": 6.40625,
      "learning_rate": 9.254486133265517e-08,
      "loss": 0.9481,
      "num_input_tokens_seen": 75984648,
      "step": 130970
    },
    {
      "epoch": 19.507745010425975,
      "grad_norm": 6.15625,
      "learning_rate": 9.226573701794361e-08,
      "loss": 1.0019,
      "num_input_tokens_seen": 75987688,
      "step": 130975
    },
    {
      "epoch": 19.508489722966935,
      "grad_norm": 9.9375,
      "learning_rate": 9.198703349408977e-08,
      "loss": 1.1403,
      "num_input_tokens_seen": 75990248,
      "step": 130980
    },
    {
      "epoch": 19.509234435507892,
      "grad_norm": 6.40625,
      "learning_rate": 9.170875076579821e-08,
      "loss": 0.9292,
      "num_input_tokens_seen": 75992904,
      "step": 130985
    },
    {
      "epoch": 19.509979148048853,
      "grad_norm": 12.5,
      "learning_rate": 9.143088883777073e-08,
      "loss": 1.0731,
      "num_input_tokens_seen": 75995880,
      "step": 130990
    },
    {
      "epoch": 19.510723860589813,
      "grad_norm": 10.625,
      "learning_rate": 9.115344771470357e-08,
      "loss": 0.8173,
      "num_input_tokens_seen": 75998600,
      "step": 130995
    },
    {
      "epoch": 19.51146857313077,
      "grad_norm": 18.625,
      "learning_rate": 9.087642740128188e-08,
      "loss": 1.4347,
      "num_input_tokens_seen": 76001448,
      "step": 131000
    },
    {
      "epoch": 19.51221328567173,
      "grad_norm": 9.8125,
      "learning_rate": 9.059982790218801e-08,
      "loss": 1.2673,
      "num_input_tokens_seen": 76004488,
      "step": 131005
    },
    {
      "epoch": 19.51295799821269,
      "grad_norm": 8.625,
      "learning_rate": 9.032364922209047e-08,
      "loss": 1.0867,
      "num_input_tokens_seen": 76007304,
      "step": 131010
    },
    {
      "epoch": 19.51370271075365,
      "grad_norm": 10.25,
      "learning_rate": 9.00478913656605e-08,
      "loss": 1.1294,
      "num_input_tokens_seen": 76010344,
      "step": 131015
    },
    {
      "epoch": 19.51444742329461,
      "grad_norm": 14.125,
      "learning_rate": 8.977255433755272e-08,
      "loss": 1.3309,
      "num_input_tokens_seen": 76013128,
      "step": 131020
    },
    {
      "epoch": 19.515192135835566,
      "grad_norm": 11.375,
      "learning_rate": 8.949763814242173e-08,
      "loss": 1.6026,
      "num_input_tokens_seen": 76015784,
      "step": 131025
    },
    {
      "epoch": 19.515936848376526,
      "grad_norm": 15.0,
      "learning_rate": 8.922314278490829e-08,
      "loss": 1.05,
      "num_input_tokens_seen": 76019048,
      "step": 131030
    },
    {
      "epoch": 19.516681560917487,
      "grad_norm": 8.4375,
      "learning_rate": 8.89490682696531e-08,
      "loss": 0.994,
      "num_input_tokens_seen": 76021864,
      "step": 131035
    },
    {
      "epoch": 19.517426273458444,
      "grad_norm": 17.375,
      "learning_rate": 8.867541460128304e-08,
      "loss": 0.9681,
      "num_input_tokens_seen": 76025160,
      "step": 131040
    },
    {
      "epoch": 19.518170985999404,
      "grad_norm": 10.25,
      "learning_rate": 8.840218178442494e-08,
      "loss": 0.9669,
      "num_input_tokens_seen": 76027912,
      "step": 131045
    },
    {
      "epoch": 19.518915698540365,
      "grad_norm": 7.3125,
      "learning_rate": 8.81293698236918e-08,
      "loss": 0.768,
      "num_input_tokens_seen": 76030952,
      "step": 131050
    },
    {
      "epoch": 19.519660411081322,
      "grad_norm": 9.0625,
      "learning_rate": 8.785697872369381e-08,
      "loss": 1.0978,
      "num_input_tokens_seen": 76033896,
      "step": 131055
    },
    {
      "epoch": 19.520405123622282,
      "grad_norm": 20.75,
      "learning_rate": 8.758500848903283e-08,
      "loss": 1.3147,
      "num_input_tokens_seen": 76036744,
      "step": 131060
    },
    {
      "epoch": 19.52114983616324,
      "grad_norm": 12.3125,
      "learning_rate": 8.731345912430245e-08,
      "loss": 1.0641,
      "num_input_tokens_seen": 76040040,
      "step": 131065
    },
    {
      "epoch": 19.5218945487042,
      "grad_norm": 11.4375,
      "learning_rate": 8.704233063409339e-08,
      "loss": 0.8771,
      "num_input_tokens_seen": 76043080,
      "step": 131070
    },
    {
      "epoch": 19.52263926124516,
      "grad_norm": 12.9375,
      "learning_rate": 8.677162302298258e-08,
      "loss": 1.1814,
      "num_input_tokens_seen": 76045960,
      "step": 131075
    },
    {
      "epoch": 19.523383973786117,
      "grad_norm": 15.3125,
      "learning_rate": 8.650133629554413e-08,
      "loss": 1.2449,
      "num_input_tokens_seen": 76048808,
      "step": 131080
    },
    {
      "epoch": 19.524128686327078,
      "grad_norm": 6.3125,
      "learning_rate": 8.623147045634383e-08,
      "loss": 0.8585,
      "num_input_tokens_seen": 76051720,
      "step": 131085
    },
    {
      "epoch": 19.52487339886804,
      "grad_norm": 11.25,
      "learning_rate": 8.596202550994193e-08,
      "loss": 1.0018,
      "num_input_tokens_seen": 76054696,
      "step": 131090
    },
    {
      "epoch": 19.525618111408996,
      "grad_norm": 12.1875,
      "learning_rate": 8.569300146089032e-08,
      "loss": 0.9905,
      "num_input_tokens_seen": 76057672,
      "step": 131095
    },
    {
      "epoch": 19.526362823949956,
      "grad_norm": 10.625,
      "learning_rate": 8.542439831373539e-08,
      "loss": 1.1484,
      "num_input_tokens_seen": 76060520,
      "step": 131100
    },
    {
      "epoch": 19.527107536490913,
      "grad_norm": 9.25,
      "learning_rate": 8.515621607301239e-08,
      "loss": 1.0664,
      "num_input_tokens_seen": 76063368,
      "step": 131105
    },
    {
      "epoch": 19.527852249031874,
      "grad_norm": 11.4375,
      "learning_rate": 8.488845474325102e-08,
      "loss": 1.0595,
      "num_input_tokens_seen": 76066056,
      "step": 131110
    },
    {
      "epoch": 19.528596961572834,
      "grad_norm": 12.25,
      "learning_rate": 8.462111432897823e-08,
      "loss": 1.146,
      "num_input_tokens_seen": 76069000,
      "step": 131115
    },
    {
      "epoch": 19.52934167411379,
      "grad_norm": 12.4375,
      "learning_rate": 8.435419483470707e-08,
      "loss": 1.3283,
      "num_input_tokens_seen": 76072264,
      "step": 131120
    },
    {
      "epoch": 19.53008638665475,
      "grad_norm": 7.90625,
      "learning_rate": 8.408769626495061e-08,
      "loss": 1.0541,
      "num_input_tokens_seen": 76075400,
      "step": 131125
    },
    {
      "epoch": 19.530831099195712,
      "grad_norm": 8.0,
      "learning_rate": 8.382161862420801e-08,
      "loss": 1.4632,
      "num_input_tokens_seen": 76078184,
      "step": 131130
    },
    {
      "epoch": 19.53157581173667,
      "grad_norm": 7.65625,
      "learning_rate": 8.355596191697845e-08,
      "loss": 1.1423,
      "num_input_tokens_seen": 76081000,
      "step": 131135
    },
    {
      "epoch": 19.53232052427763,
      "grad_norm": 9.0,
      "learning_rate": 8.329072614774446e-08,
      "loss": 1.3641,
      "num_input_tokens_seen": 76083816,
      "step": 131140
    },
    {
      "epoch": 19.533065236818587,
      "grad_norm": 8.6875,
      "learning_rate": 8.302591132098857e-08,
      "loss": 0.9289,
      "num_input_tokens_seen": 76086664,
      "step": 131145
    },
    {
      "epoch": 19.533809949359547,
      "grad_norm": 5.09375,
      "learning_rate": 8.276151744118777e-08,
      "loss": 0.9198,
      "num_input_tokens_seen": 76089352,
      "step": 131150
    },
    {
      "epoch": 19.534554661900508,
      "grad_norm": 7.71875,
      "learning_rate": 8.249754451280512e-08,
      "loss": 0.9217,
      "num_input_tokens_seen": 76092072,
      "step": 131155
    },
    {
      "epoch": 19.535299374441465,
      "grad_norm": 4.375,
      "learning_rate": 8.223399254030095e-08,
      "loss": 0.8647,
      "num_input_tokens_seen": 76095048,
      "step": 131160
    },
    {
      "epoch": 19.536044086982425,
      "grad_norm": 9.875,
      "learning_rate": 8.197086152812728e-08,
      "loss": 1.0336,
      "num_input_tokens_seen": 76097896,
      "step": 131165
    },
    {
      "epoch": 19.536788799523386,
      "grad_norm": 14.25,
      "learning_rate": 8.17081514807333e-08,
      "loss": 0.9018,
      "num_input_tokens_seen": 76100744,
      "step": 131170
    },
    {
      "epoch": 19.537533512064343,
      "grad_norm": 8.8125,
      "learning_rate": 8.144586240255159e-08,
      "loss": 1.1378,
      "num_input_tokens_seen": 76103560,
      "step": 131175
    },
    {
      "epoch": 19.538278224605303,
      "grad_norm": 22.625,
      "learning_rate": 8.118399429801749e-08,
      "loss": 1.6344,
      "num_input_tokens_seen": 76106696,
      "step": 131180
    },
    {
      "epoch": 19.53902293714626,
      "grad_norm": 8.375,
      "learning_rate": 8.092254717155246e-08,
      "loss": 0.9397,
      "num_input_tokens_seen": 76109672,
      "step": 131185
    },
    {
      "epoch": 19.53976764968722,
      "grad_norm": 11.125,
      "learning_rate": 8.066152102757518e-08,
      "loss": 0.9664,
      "num_input_tokens_seen": 76112392,
      "step": 131190
    },
    {
      "epoch": 19.54051236222818,
      "grad_norm": 14.375,
      "learning_rate": 8.040091587049325e-08,
      "loss": 1.3541,
      "num_input_tokens_seen": 76115528,
      "step": 131195
    },
    {
      "epoch": 19.54125707476914,
      "grad_norm": 6.1875,
      "learning_rate": 8.014073170471149e-08,
      "loss": 1.1701,
      "num_input_tokens_seen": 76118568,
      "step": 131200
    },
    {
      "epoch": 19.5420017873101,
      "grad_norm": 5.34375,
      "learning_rate": 7.988096853462634e-08,
      "loss": 0.9012,
      "num_input_tokens_seen": 76121768,
      "step": 131205
    },
    {
      "epoch": 19.542746499851056,
      "grad_norm": 8.75,
      "learning_rate": 7.962162636462323e-08,
      "loss": 1.0361,
      "num_input_tokens_seen": 76124456,
      "step": 131210
    },
    {
      "epoch": 19.543491212392016,
      "grad_norm": 14.0625,
      "learning_rate": 7.936270519908473e-08,
      "loss": 1.2032,
      "num_input_tokens_seen": 76127240,
      "step": 131215
    },
    {
      "epoch": 19.544235924932977,
      "grad_norm": 13.75,
      "learning_rate": 7.910420504238514e-08,
      "loss": 1.0293,
      "num_input_tokens_seen": 76130056,
      "step": 131220
    },
    {
      "epoch": 19.544980637473934,
      "grad_norm": 11.6875,
      "learning_rate": 7.88461258988904e-08,
      "loss": 1.0876,
      "num_input_tokens_seen": 76132776,
      "step": 131225
    },
    {
      "epoch": 19.545725350014894,
      "grad_norm": 6.34375,
      "learning_rate": 7.858846777296369e-08,
      "loss": 1.1275,
      "num_input_tokens_seen": 76135752,
      "step": 131230
    },
    {
      "epoch": 19.546470062555855,
      "grad_norm": 15.3125,
      "learning_rate": 7.833123066895432e-08,
      "loss": 1.1858,
      "num_input_tokens_seen": 76138728,
      "step": 131235
    },
    {
      "epoch": 19.547214775096812,
      "grad_norm": 6.0,
      "learning_rate": 7.807441459121156e-08,
      "loss": 1.1666,
      "num_input_tokens_seen": 76141800,
      "step": 131240
    },
    {
      "epoch": 19.547959487637772,
      "grad_norm": 11.1875,
      "learning_rate": 7.781801954406809e-08,
      "loss": 1.1864,
      "num_input_tokens_seen": 76144584,
      "step": 131245
    },
    {
      "epoch": 19.54870420017873,
      "grad_norm": 5.84375,
      "learning_rate": 7.756204553186208e-08,
      "loss": 0.8704,
      "num_input_tokens_seen": 76147560,
      "step": 131250
    },
    {
      "epoch": 19.54944891271969,
      "grad_norm": 8.3125,
      "learning_rate": 7.730649255891509e-08,
      "loss": 1.0109,
      "num_input_tokens_seen": 76150472,
      "step": 131255
    },
    {
      "epoch": 19.55019362526065,
      "grad_norm": 8.1875,
      "learning_rate": 7.705136062954587e-08,
      "loss": 0.9745,
      "num_input_tokens_seen": 76153448,
      "step": 131260
    },
    {
      "epoch": 19.550938337801608,
      "grad_norm": 7.8125,
      "learning_rate": 7.679664974806212e-08,
      "loss": 1.0301,
      "num_input_tokens_seen": 76156296,
      "step": 131265
    },
    {
      "epoch": 19.551683050342568,
      "grad_norm": 9.3125,
      "learning_rate": 7.654235991876867e-08,
      "loss": 1.0615,
      "num_input_tokens_seen": 76159176,
      "step": 131270
    },
    {
      "epoch": 19.55242776288353,
      "grad_norm": 8.5625,
      "learning_rate": 7.628849114596214e-08,
      "loss": 1.0047,
      "num_input_tokens_seen": 76162120,
      "step": 131275
    },
    {
      "epoch": 19.553172475424486,
      "grad_norm": 7.34375,
      "learning_rate": 7.603504343392798e-08,
      "loss": 0.9585,
      "num_input_tokens_seen": 76164712,
      "step": 131280
    },
    {
      "epoch": 19.553917187965446,
      "grad_norm": 7.65625,
      "learning_rate": 7.578201678694885e-08,
      "loss": 1.1631,
      "num_input_tokens_seen": 76167560,
      "step": 131285
    },
    {
      "epoch": 19.554661900506403,
      "grad_norm": 32.75,
      "learning_rate": 7.55294112093019e-08,
      "loss": 1.3681,
      "num_input_tokens_seen": 76171048,
      "step": 131290
    },
    {
      "epoch": 19.555406613047364,
      "grad_norm": 12.5,
      "learning_rate": 7.527722670525594e-08,
      "loss": 1.3859,
      "num_input_tokens_seen": 76173992,
      "step": 131295
    },
    {
      "epoch": 19.556151325588324,
      "grad_norm": 9.3125,
      "learning_rate": 7.50254632790659e-08,
      "loss": 1.1645,
      "num_input_tokens_seen": 76177160,
      "step": 131300
    },
    {
      "epoch": 19.55689603812928,
      "grad_norm": 9.6875,
      "learning_rate": 7.477412093498947e-08,
      "loss": 1.1225,
      "num_input_tokens_seen": 76180104,
      "step": 131305
    },
    {
      "epoch": 19.55764075067024,
      "grad_norm": 6.78125,
      "learning_rate": 7.452319967727328e-08,
      "loss": 0.9578,
      "num_input_tokens_seen": 76183304,
      "step": 131310
    },
    {
      "epoch": 19.558385463211202,
      "grad_norm": 12.875,
      "learning_rate": 7.427269951015004e-08,
      "loss": 0.9147,
      "num_input_tokens_seen": 76186312,
      "step": 131315
    },
    {
      "epoch": 19.55913017575216,
      "grad_norm": 15.75,
      "learning_rate": 7.402262043785801e-08,
      "loss": 1.4987,
      "num_input_tokens_seen": 76189128,
      "step": 131320
    },
    {
      "epoch": 19.55987488829312,
      "grad_norm": 5.65625,
      "learning_rate": 7.377296246462162e-08,
      "loss": 1.0835,
      "num_input_tokens_seen": 76192264,
      "step": 131325
    },
    {
      "epoch": 19.560619600834077,
      "grad_norm": 3.828125,
      "learning_rate": 7.352372559465693e-08,
      "loss": 0.9788,
      "num_input_tokens_seen": 76195208,
      "step": 131330
    },
    {
      "epoch": 19.561364313375037,
      "grad_norm": 12.1875,
      "learning_rate": 7.327490983217444e-08,
      "loss": 0.9363,
      "num_input_tokens_seen": 76198024,
      "step": 131335
    },
    {
      "epoch": 19.562109025915998,
      "grad_norm": 10.4375,
      "learning_rate": 7.302651518137638e-08,
      "loss": 1.1246,
      "num_input_tokens_seen": 76200808,
      "step": 131340
    },
    {
      "epoch": 19.562853738456955,
      "grad_norm": 16.0,
      "learning_rate": 7.277854164646214e-08,
      "loss": 1.2643,
      "num_input_tokens_seen": 76203560,
      "step": 131345
    },
    {
      "epoch": 19.563598450997915,
      "grad_norm": 4.90625,
      "learning_rate": 7.253098923162005e-08,
      "loss": 1.2128,
      "num_input_tokens_seen": 76206472,
      "step": 131350
    },
    {
      "epoch": 19.564343163538872,
      "grad_norm": 12.1875,
      "learning_rate": 7.22838579410301e-08,
      "loss": 0.8462,
      "num_input_tokens_seen": 76209416,
      "step": 131355
    },
    {
      "epoch": 19.565087876079833,
      "grad_norm": 10.4375,
      "learning_rate": 7.20371477788695e-08,
      "loss": 1.3565,
      "num_input_tokens_seen": 76212200,
      "step": 131360
    },
    {
      "epoch": 19.565832588620793,
      "grad_norm": 13.1875,
      "learning_rate": 7.179085874930713e-08,
      "loss": 1.435,
      "num_input_tokens_seen": 76215080,
      "step": 131365
    },
    {
      "epoch": 19.56657730116175,
      "grad_norm": 9.25,
      "learning_rate": 7.154499085650079e-08,
      "loss": 0.9011,
      "num_input_tokens_seen": 76217960,
      "step": 131370
    },
    {
      "epoch": 19.56732201370271,
      "grad_norm": 8.5625,
      "learning_rate": 7.129954410460548e-08,
      "loss": 1.0368,
      "num_input_tokens_seen": 76221192,
      "step": 131375
    },
    {
      "epoch": 19.56806672624367,
      "grad_norm": 9.5625,
      "learning_rate": 7.105451849777067e-08,
      "loss": 0.9712,
      "num_input_tokens_seen": 76224200,
      "step": 131380
    },
    {
      "epoch": 19.56881143878463,
      "grad_norm": 4.25,
      "learning_rate": 7.080991404012915e-08,
      "loss": 0.9409,
      "num_input_tokens_seen": 76227336,
      "step": 131385
    },
    {
      "epoch": 19.56955615132559,
      "grad_norm": 14.25,
      "learning_rate": 7.056573073581929e-08,
      "loss": 1.2745,
      "num_input_tokens_seen": 76230184,
      "step": 131390
    },
    {
      "epoch": 19.570300863866546,
      "grad_norm": 5.1875,
      "learning_rate": 7.032196858896279e-08,
      "loss": 0.7447,
      "num_input_tokens_seen": 76233096,
      "step": 131395
    },
    {
      "epoch": 19.571045576407506,
      "grad_norm": 11.1875,
      "learning_rate": 7.007862760368133e-08,
      "loss": 1.029,
      "num_input_tokens_seen": 76236104,
      "step": 131400
    },
    {
      "epoch": 19.571790288948467,
      "grad_norm": 21.0,
      "learning_rate": 6.983570778408277e-08,
      "loss": 0.9166,
      "num_input_tokens_seen": 76238984,
      "step": 131405
    },
    {
      "epoch": 19.572535001489424,
      "grad_norm": 10.4375,
      "learning_rate": 6.959320913427492e-08,
      "loss": 1.296,
      "num_input_tokens_seen": 76242184,
      "step": 131410
    },
    {
      "epoch": 19.573279714030384,
      "grad_norm": 8.875,
      "learning_rate": 6.935113165834616e-08,
      "loss": 1.1895,
      "num_input_tokens_seen": 76245288,
      "step": 131415
    },
    {
      "epoch": 19.574024426571345,
      "grad_norm": 10.375,
      "learning_rate": 6.910947536039603e-08,
      "loss": 0.7279,
      "num_input_tokens_seen": 76248072,
      "step": 131420
    },
    {
      "epoch": 19.574769139112302,
      "grad_norm": 21.0,
      "learning_rate": 6.886824024450178e-08,
      "loss": 1.4143,
      "num_input_tokens_seen": 76251304,
      "step": 131425
    },
    {
      "epoch": 19.575513851653263,
      "grad_norm": 6.5625,
      "learning_rate": 6.862742631473795e-08,
      "loss": 1.022,
      "num_input_tokens_seen": 76254312,
      "step": 131430
    },
    {
      "epoch": 19.57625856419422,
      "grad_norm": 8.8125,
      "learning_rate": 6.838703357517628e-08,
      "loss": 1.1964,
      "num_input_tokens_seen": 76257224,
      "step": 131435
    },
    {
      "epoch": 19.57700327673518,
      "grad_norm": 9.0625,
      "learning_rate": 6.814706202987465e-08,
      "loss": 1.1393,
      "num_input_tokens_seen": 76259912,
      "step": 131440
    },
    {
      "epoch": 19.57774798927614,
      "grad_norm": 7.78125,
      "learning_rate": 6.79075116828909e-08,
      "loss": 1.1953,
      "num_input_tokens_seen": 76263048,
      "step": 131445
    },
    {
      "epoch": 19.578492701817098,
      "grad_norm": 10.9375,
      "learning_rate": 6.766838253826902e-08,
      "loss": 1.1495,
      "num_input_tokens_seen": 76266088,
      "step": 131450
    },
    {
      "epoch": 19.579237414358058,
      "grad_norm": 12.9375,
      "learning_rate": 6.742967460005023e-08,
      "loss": 0.8903,
      "num_input_tokens_seen": 76268744,
      "step": 131455
    },
    {
      "epoch": 19.57998212689902,
      "grad_norm": 11.5625,
      "learning_rate": 6.719138787226464e-08,
      "loss": 1.3059,
      "num_input_tokens_seen": 76271752,
      "step": 131460
    },
    {
      "epoch": 19.580726839439976,
      "grad_norm": 14.1875,
      "learning_rate": 6.695352235894237e-08,
      "loss": 1.4563,
      "num_input_tokens_seen": 76274472,
      "step": 131465
    },
    {
      "epoch": 19.581471551980936,
      "grad_norm": 14.6875,
      "learning_rate": 6.671607806409963e-08,
      "loss": 0.8926,
      "num_input_tokens_seen": 76277128,
      "step": 131470
    },
    {
      "epoch": 19.582216264521893,
      "grad_norm": 15.25,
      "learning_rate": 6.647905499174712e-08,
      "loss": 1.0198,
      "num_input_tokens_seen": 76280136,
      "step": 131475
    },
    {
      "epoch": 19.582960977062854,
      "grad_norm": 12.0,
      "learning_rate": 6.624245314588994e-08,
      "loss": 1.0031,
      "num_input_tokens_seen": 76283272,
      "step": 131480
    },
    {
      "epoch": 19.583705689603814,
      "grad_norm": 14.625,
      "learning_rate": 6.600627253052216e-08,
      "loss": 0.7314,
      "num_input_tokens_seen": 76286024,
      "step": 131485
    },
    {
      "epoch": 19.58445040214477,
      "grad_norm": 10.375,
      "learning_rate": 6.577051314964055e-08,
      "loss": 1.2343,
      "num_input_tokens_seen": 76289000,
      "step": 131490
    },
    {
      "epoch": 19.58519511468573,
      "grad_norm": 10.25,
      "learning_rate": 6.55351750072225e-08,
      "loss": 1.2282,
      "num_input_tokens_seen": 76291720,
      "step": 131495
    },
    {
      "epoch": 19.58593982722669,
      "grad_norm": 6.15625,
      "learning_rate": 6.530025810724539e-08,
      "loss": 0.8601,
      "num_input_tokens_seen": 76294504,
      "step": 131500
    },
    {
      "epoch": 19.58668453976765,
      "grad_norm": 6.6875,
      "learning_rate": 6.506576245367824e-08,
      "loss": 0.9406,
      "num_input_tokens_seen": 76297160,
      "step": 131505
    },
    {
      "epoch": 19.58742925230861,
      "grad_norm": 13.1875,
      "learning_rate": 6.483168805047901e-08,
      "loss": 0.9699,
      "num_input_tokens_seen": 76300136,
      "step": 131510
    },
    {
      "epoch": 19.588173964849567,
      "grad_norm": 6.78125,
      "learning_rate": 6.459803490160843e-08,
      "loss": 0.7916,
      "num_input_tokens_seen": 76303080,
      "step": 131515
    },
    {
      "epoch": 19.588918677390527,
      "grad_norm": 9.25,
      "learning_rate": 6.436480301101055e-08,
      "loss": 1.1067,
      "num_input_tokens_seen": 76305736,
      "step": 131520
    },
    {
      "epoch": 19.589663389931488,
      "grad_norm": 7.625,
      "learning_rate": 6.413199238262668e-08,
      "loss": 1.1111,
      "num_input_tokens_seen": 76308616,
      "step": 131525
    },
    {
      "epoch": 19.590408102472445,
      "grad_norm": 7.0625,
      "learning_rate": 6.389960302038978e-08,
      "loss": 1.1125,
      "num_input_tokens_seen": 76311336,
      "step": 131530
    },
    {
      "epoch": 19.591152815013405,
      "grad_norm": 8.25,
      "learning_rate": 6.366763492822448e-08,
      "loss": 1.0604,
      "num_input_tokens_seen": 76313992,
      "step": 131535
    },
    {
      "epoch": 19.591897527554362,
      "grad_norm": 7.5,
      "learning_rate": 6.343608811004986e-08,
      "loss": 1.0869,
      "num_input_tokens_seen": 76317064,
      "step": 131540
    },
    {
      "epoch": 19.592642240095323,
      "grad_norm": 9.1875,
      "learning_rate": 6.320496256977671e-08,
      "loss": 1.2683,
      "num_input_tokens_seen": 76320200,
      "step": 131545
    },
    {
      "epoch": 19.593386952636283,
      "grad_norm": 10.5625,
      "learning_rate": 6.297425831131299e-08,
      "loss": 1.0385,
      "num_input_tokens_seen": 76323208,
      "step": 131550
    },
    {
      "epoch": 19.59413166517724,
      "grad_norm": 14.0625,
      "learning_rate": 6.274397533855281e-08,
      "loss": 1.1486,
      "num_input_tokens_seen": 76326152,
      "step": 131555
    },
    {
      "epoch": 19.5948763777182,
      "grad_norm": 6.71875,
      "learning_rate": 6.251411365539029e-08,
      "loss": 1.3093,
      "num_input_tokens_seen": 76329032,
      "step": 131560
    },
    {
      "epoch": 19.59562109025916,
      "grad_norm": 5.9375,
      "learning_rate": 6.228467326570286e-08,
      "loss": 1.4016,
      "num_input_tokens_seen": 76331912,
      "step": 131565
    },
    {
      "epoch": 19.59636580280012,
      "grad_norm": 5.6875,
      "learning_rate": 6.205565417337356e-08,
      "loss": 0.9686,
      "num_input_tokens_seen": 76334696,
      "step": 131570
    },
    {
      "epoch": 19.59711051534108,
      "grad_norm": 8.375,
      "learning_rate": 6.182705638226872e-08,
      "loss": 0.9195,
      "num_input_tokens_seen": 76337544,
      "step": 131575
    },
    {
      "epoch": 19.597855227882036,
      "grad_norm": 10.5625,
      "learning_rate": 6.159887989624635e-08,
      "loss": 1.072,
      "num_input_tokens_seen": 76340296,
      "step": 131580
    },
    {
      "epoch": 19.598599940422996,
      "grad_norm": 10.75,
      "learning_rate": 6.137112471916729e-08,
      "loss": 1.1647,
      "num_input_tokens_seen": 76342984,
      "step": 131585
    },
    {
      "epoch": 19.599344652963957,
      "grad_norm": 10.125,
      "learning_rate": 6.114379085487565e-08,
      "loss": 0.9843,
      "num_input_tokens_seen": 76345672,
      "step": 131590
    },
    {
      "epoch": 19.600089365504914,
      "grad_norm": 7.09375,
      "learning_rate": 6.091687830721282e-08,
      "loss": 0.9443,
      "num_input_tokens_seen": 76348296,
      "step": 131595
    },
    {
      "epoch": 19.600834078045875,
      "grad_norm": 13.875,
      "learning_rate": 6.069038708001462e-08,
      "loss": 1.1237,
      "num_input_tokens_seen": 76351528,
      "step": 131600
    },
    {
      "epoch": 19.601578790586835,
      "grad_norm": 9.5,
      "learning_rate": 6.046431717710299e-08,
      "loss": 1.1155,
      "num_input_tokens_seen": 76354376,
      "step": 131605
    },
    {
      "epoch": 19.602323503127792,
      "grad_norm": 9.375,
      "learning_rate": 6.023866860229988e-08,
      "loss": 0.8935,
      "num_input_tokens_seen": 76357320,
      "step": 131610
    },
    {
      "epoch": 19.603068215668753,
      "grad_norm": 9.8125,
      "learning_rate": 6.001344135941611e-08,
      "loss": 1.2915,
      "num_input_tokens_seen": 76360072,
      "step": 131615
    },
    {
      "epoch": 19.60381292820971,
      "grad_norm": 8.9375,
      "learning_rate": 5.9788635452257e-08,
      "loss": 1.2042,
      "num_input_tokens_seen": 76362888,
      "step": 131620
    },
    {
      "epoch": 19.60455764075067,
      "grad_norm": 11.75,
      "learning_rate": 5.9564250884622255e-08,
      "loss": 1.2349,
      "num_input_tokens_seen": 76365992,
      "step": 131625
    },
    {
      "epoch": 19.60530235329163,
      "grad_norm": 16.125,
      "learning_rate": 5.934028766030053e-08,
      "loss": 1.2079,
      "num_input_tokens_seen": 76368904,
      "step": 131630
    },
    {
      "epoch": 19.606047065832588,
      "grad_norm": 12.375,
      "learning_rate": 5.911674578307491e-08,
      "loss": 1.2446,
      "num_input_tokens_seen": 76372008,
      "step": 131635
    },
    {
      "epoch": 19.606791778373548,
      "grad_norm": 6.53125,
      "learning_rate": 5.88936252567257e-08,
      "loss": 1.1083,
      "num_input_tokens_seen": 76374600,
      "step": 131640
    },
    {
      "epoch": 19.60753649091451,
      "grad_norm": 9.875,
      "learning_rate": 5.8670926085016564e-08,
      "loss": 1.2089,
      "num_input_tokens_seen": 76377384,
      "step": 131645
    },
    {
      "epoch": 19.608281203455466,
      "grad_norm": 5.625,
      "learning_rate": 5.8448648271713925e-08,
      "loss": 0.9274,
      "num_input_tokens_seen": 76380424,
      "step": 131650
    },
    {
      "epoch": 19.609025915996426,
      "grad_norm": 8.375,
      "learning_rate": 5.822679182057311e-08,
      "loss": 0.9481,
      "num_input_tokens_seen": 76383112,
      "step": 131655
    },
    {
      "epoch": 19.609770628537383,
      "grad_norm": 11.125,
      "learning_rate": 5.8005356735341135e-08,
      "loss": 1.0023,
      "num_input_tokens_seen": 76385864,
      "step": 131660
    },
    {
      "epoch": 19.610515341078344,
      "grad_norm": 13.25,
      "learning_rate": 5.7784343019759436e-08,
      "loss": 0.9285,
      "num_input_tokens_seen": 76388584,
      "step": 131665
    },
    {
      "epoch": 19.611260053619304,
      "grad_norm": 10.375,
      "learning_rate": 5.756375067755837e-08,
      "loss": 0.9301,
      "num_input_tokens_seen": 76391528,
      "step": 131670
    },
    {
      "epoch": 19.61200476616026,
      "grad_norm": 8.0625,
      "learning_rate": 5.7343579712468286e-08,
      "loss": 0.9817,
      "num_input_tokens_seen": 76394280,
      "step": 131675
    },
    {
      "epoch": 19.61274947870122,
      "grad_norm": 18.625,
      "learning_rate": 5.712383012820843e-08,
      "loss": 0.9637,
      "num_input_tokens_seen": 76397736,
      "step": 131680
    },
    {
      "epoch": 19.613494191242182,
      "grad_norm": 12.3125,
      "learning_rate": 5.6904501928489726e-08,
      "loss": 1.2549,
      "num_input_tokens_seen": 76401032,
      "step": 131685
    },
    {
      "epoch": 19.61423890378314,
      "grad_norm": 8.1875,
      "learning_rate": 5.668559511702032e-08,
      "loss": 1.1975,
      "num_input_tokens_seen": 76403528,
      "step": 131690
    },
    {
      "epoch": 19.6149836163241,
      "grad_norm": 9.1875,
      "learning_rate": 5.646710969749447e-08,
      "loss": 0.8915,
      "num_input_tokens_seen": 76406344,
      "step": 131695
    },
    {
      "epoch": 19.615728328865057,
      "grad_norm": 5.5,
      "learning_rate": 5.6249045673606446e-08,
      "loss": 0.8317,
      "num_input_tokens_seen": 76409544,
      "step": 131700
    },
    {
      "epoch": 19.616473041406017,
      "grad_norm": 10.5,
      "learning_rate": 5.603140304903942e-08,
      "loss": 1.5256,
      "num_input_tokens_seen": 76412424,
      "step": 131705
    },
    {
      "epoch": 19.617217753946978,
      "grad_norm": 6.34375,
      "learning_rate": 5.581418182746823e-08,
      "loss": 0.9287,
      "num_input_tokens_seen": 76415272,
      "step": 131710
    },
    {
      "epoch": 19.617962466487935,
      "grad_norm": 6.5625,
      "learning_rate": 5.55973820125677e-08,
      "loss": 1.0442,
      "num_input_tokens_seen": 76418056,
      "step": 131715
    },
    {
      "epoch": 19.618707179028895,
      "grad_norm": 4.9375,
      "learning_rate": 5.538100360799325e-08,
      "loss": 1.0673,
      "num_input_tokens_seen": 76421256,
      "step": 131720
    },
    {
      "epoch": 19.619451891569852,
      "grad_norm": 11.8125,
      "learning_rate": 5.516504661740585e-08,
      "loss": 1.2763,
      "num_input_tokens_seen": 76423912,
      "step": 131725
    },
    {
      "epoch": 19.620196604110813,
      "grad_norm": 8.875,
      "learning_rate": 5.494951104445256e-08,
      "loss": 1.1099,
      "num_input_tokens_seen": 76426760,
      "step": 131730
    },
    {
      "epoch": 19.620941316651773,
      "grad_norm": 7.875,
      "learning_rate": 5.473439689277493e-08,
      "loss": 1.3217,
      "num_input_tokens_seen": 76429736,
      "step": 131735
    },
    {
      "epoch": 19.62168602919273,
      "grad_norm": 10.9375,
      "learning_rate": 5.451970416600338e-08,
      "loss": 1.1829,
      "num_input_tokens_seen": 76432712,
      "step": 131740
    },
    {
      "epoch": 19.62243074173369,
      "grad_norm": 8.125,
      "learning_rate": 5.430543286777112e-08,
      "loss": 1.1139,
      "num_input_tokens_seen": 76435752,
      "step": 131745
    },
    {
      "epoch": 19.62317545427465,
      "grad_norm": 12.75,
      "learning_rate": 5.4091583001691923e-08,
      "loss": 1.3193,
      "num_input_tokens_seen": 76438664,
      "step": 131750
    },
    {
      "epoch": 19.62392016681561,
      "grad_norm": 16.375,
      "learning_rate": 5.387815457138512e-08,
      "loss": 1.0638,
      "num_input_tokens_seen": 76441544,
      "step": 131755
    },
    {
      "epoch": 19.62466487935657,
      "grad_norm": 8.9375,
      "learning_rate": 5.3665147580450604e-08,
      "loss": 0.9477,
      "num_input_tokens_seen": 76444360,
      "step": 131760
    },
    {
      "epoch": 19.625409591897526,
      "grad_norm": 8.25,
      "learning_rate": 5.3452562032488275e-08,
      "loss": 0.7574,
      "num_input_tokens_seen": 76447144,
      "step": 131765
    },
    {
      "epoch": 19.626154304438487,
      "grad_norm": 8.0,
      "learning_rate": 5.324039793109248e-08,
      "loss": 0.8938,
      "num_input_tokens_seen": 76449928,
      "step": 131770
    },
    {
      "epoch": 19.626899016979447,
      "grad_norm": 19.875,
      "learning_rate": 5.302865527984369e-08,
      "loss": 1.3058,
      "num_input_tokens_seen": 76452744,
      "step": 131775
    },
    {
      "epoch": 19.627643729520404,
      "grad_norm": 8.5,
      "learning_rate": 5.281733408232237e-08,
      "loss": 0.9867,
      "num_input_tokens_seen": 76455528,
      "step": 131780
    },
    {
      "epoch": 19.628388442061365,
      "grad_norm": 10.3125,
      "learning_rate": 5.2606434342095115e-08,
      "loss": 1.2007,
      "num_input_tokens_seen": 76458216,
      "step": 131785
    },
    {
      "epoch": 19.629133154602325,
      "grad_norm": 8.625,
      "learning_rate": 5.2395956062728515e-08,
      "loss": 0.975,
      "num_input_tokens_seen": 76461128,
      "step": 131790
    },
    {
      "epoch": 19.629877867143282,
      "grad_norm": 8.0625,
      "learning_rate": 5.218589924777528e-08,
      "loss": 0.9365,
      "num_input_tokens_seen": 76464360,
      "step": 131795
    },
    {
      "epoch": 19.630622579684243,
      "grad_norm": 8.25,
      "learning_rate": 5.1976263900788136e-08,
      "loss": 0.7871,
      "num_input_tokens_seen": 76467304,
      "step": 131800
    },
    {
      "epoch": 19.6313672922252,
      "grad_norm": 8.75,
      "learning_rate": 5.176705002530313e-08,
      "loss": 0.9699,
      "num_input_tokens_seen": 76470216,
      "step": 131805
    },
    {
      "epoch": 19.63211200476616,
      "grad_norm": 14.4375,
      "learning_rate": 5.155825762485911e-08,
      "loss": 1.097,
      "num_input_tokens_seen": 76473096,
      "step": 131810
    },
    {
      "epoch": 19.63285671730712,
      "grad_norm": 6.59375,
      "learning_rate": 5.134988670298102e-08,
      "loss": 0.9431,
      "num_input_tokens_seen": 76475784,
      "step": 131815
    },
    {
      "epoch": 19.633601429848078,
      "grad_norm": 10.125,
      "learning_rate": 5.1141937263188276e-08,
      "loss": 1.1261,
      "num_input_tokens_seen": 76478696,
      "step": 131820
    },
    {
      "epoch": 19.634346142389038,
      "grad_norm": 8.25,
      "learning_rate": 5.093440930899751e-08,
      "loss": 1.0398,
      "num_input_tokens_seen": 76481448,
      "step": 131825
    },
    {
      "epoch": 19.63509085493,
      "grad_norm": 6.78125,
      "learning_rate": 5.072730284391425e-08,
      "loss": 1.2858,
      "num_input_tokens_seen": 76484328,
      "step": 131830
    },
    {
      "epoch": 19.635835567470956,
      "grad_norm": 12.625,
      "learning_rate": 5.0520617871432916e-08,
      "loss": 0.983,
      "num_input_tokens_seen": 76487208,
      "step": 131835
    },
    {
      "epoch": 19.636580280011916,
      "grad_norm": 8.625,
      "learning_rate": 5.0314354395050724e-08,
      "loss": 1.4399,
      "num_input_tokens_seen": 76490088,
      "step": 131840
    },
    {
      "epoch": 19.637324992552873,
      "grad_norm": 12.375,
      "learning_rate": 5.010851241824821e-08,
      "loss": 0.9627,
      "num_input_tokens_seen": 76493064,
      "step": 131845
    },
    {
      "epoch": 19.638069705093834,
      "grad_norm": 7.21875,
      "learning_rate": 4.990309194450593e-08,
      "loss": 0.7966,
      "num_input_tokens_seen": 76496008,
      "step": 131850
    },
    {
      "epoch": 19.638814417634794,
      "grad_norm": 16.375,
      "learning_rate": 4.9698092977290556e-08,
      "loss": 0.9414,
      "num_input_tokens_seen": 76498664,
      "step": 131855
    },
    {
      "epoch": 19.63955913017575,
      "grad_norm": 19.75,
      "learning_rate": 4.9493515520068754e-08,
      "loss": 1.2686,
      "num_input_tokens_seen": 76501384,
      "step": 131860
    },
    {
      "epoch": 19.640303842716712,
      "grad_norm": 5.125,
      "learning_rate": 4.92893595762961e-08,
      "loss": 0.7023,
      "num_input_tokens_seen": 76504168,
      "step": 131865
    },
    {
      "epoch": 19.64104855525767,
      "grad_norm": 10.875,
      "learning_rate": 4.908562514941983e-08,
      "loss": 1.0911,
      "num_input_tokens_seen": 76506888,
      "step": 131870
    },
    {
      "epoch": 19.64179326779863,
      "grad_norm": 7.0625,
      "learning_rate": 4.888231224288442e-08,
      "loss": 1.2987,
      "num_input_tokens_seen": 76510056,
      "step": 131875
    },
    {
      "epoch": 19.64253798033959,
      "grad_norm": 7.03125,
      "learning_rate": 4.867942086012045e-08,
      "loss": 1.1948,
      "num_input_tokens_seen": 76512840,
      "step": 131880
    },
    {
      "epoch": 19.643282692880547,
      "grad_norm": 7.5,
      "learning_rate": 4.847695100456129e-08,
      "loss": 1.1827,
      "num_input_tokens_seen": 76515880,
      "step": 131885
    },
    {
      "epoch": 19.644027405421507,
      "grad_norm": 5.34375,
      "learning_rate": 4.8274902679623644e-08,
      "loss": 1.0253,
      "num_input_tokens_seen": 76518952,
      "step": 131890
    },
    {
      "epoch": 19.644772117962468,
      "grad_norm": 8.625,
      "learning_rate": 4.807327588871868e-08,
      "loss": 0.9053,
      "num_input_tokens_seen": 76522312,
      "step": 131895
    },
    {
      "epoch": 19.645516830503425,
      "grad_norm": 12.375,
      "learning_rate": 4.7872070635260333e-08,
      "loss": 1.0192,
      "num_input_tokens_seen": 76524936,
      "step": 131900
    },
    {
      "epoch": 19.646261543044385,
      "grad_norm": 8.8125,
      "learning_rate": 4.7671286922640335e-08,
      "loss": 1.0763,
      "num_input_tokens_seen": 76527656,
      "step": 131905
    },
    {
      "epoch": 19.647006255585342,
      "grad_norm": 17.625,
      "learning_rate": 4.7470924754253184e-08,
      "loss": 0.9247,
      "num_input_tokens_seen": 76530600,
      "step": 131910
    },
    {
      "epoch": 19.647750968126303,
      "grad_norm": 7.75,
      "learning_rate": 4.727098413348785e-08,
      "loss": 1.0946,
      "num_input_tokens_seen": 76533320,
      "step": 131915
    },
    {
      "epoch": 19.648495680667263,
      "grad_norm": 12.625,
      "learning_rate": 4.707146506371385e-08,
      "loss": 1.0454,
      "num_input_tokens_seen": 76536296,
      "step": 131920
    },
    {
      "epoch": 19.64924039320822,
      "grad_norm": 12.5625,
      "learning_rate": 4.6872367548309036e-08,
      "loss": 0.9819,
      "num_input_tokens_seen": 76539304,
      "step": 131925
    },
    {
      "epoch": 19.64998510574918,
      "grad_norm": 4.5625,
      "learning_rate": 4.6673691590634614e-08,
      "loss": 1.0034,
      "num_input_tokens_seen": 76542056,
      "step": 131930
    },
    {
      "epoch": 19.65072981829014,
      "grad_norm": 12.6875,
      "learning_rate": 4.6475437194046237e-08,
      "loss": 1.4592,
      "num_input_tokens_seen": 76545032,
      "step": 131935
    },
    {
      "epoch": 19.6514745308311,
      "grad_norm": 7.46875,
      "learning_rate": 4.627760436189121e-08,
      "loss": 0.9011,
      "num_input_tokens_seen": 76547816,
      "step": 131940
    },
    {
      "epoch": 19.65221924337206,
      "grad_norm": 16.875,
      "learning_rate": 4.6080193097516877e-08,
      "loss": 1.5204,
      "num_input_tokens_seen": 76550920,
      "step": 131945
    },
    {
      "epoch": 19.652963955913016,
      "grad_norm": 9.5625,
      "learning_rate": 4.588320340425667e-08,
      "loss": 1.227,
      "num_input_tokens_seen": 76553544,
      "step": 131950
    },
    {
      "epoch": 19.653708668453977,
      "grad_norm": 17.75,
      "learning_rate": 4.5686635285432934e-08,
      "loss": 1.6015,
      "num_input_tokens_seen": 76556360,
      "step": 131955
    },
    {
      "epoch": 19.654453380994937,
      "grad_norm": 6.84375,
      "learning_rate": 4.5490488744376335e-08,
      "loss": 1.2072,
      "num_input_tokens_seen": 76559112,
      "step": 131960
    },
    {
      "epoch": 19.655198093535894,
      "grad_norm": 10.5625,
      "learning_rate": 4.529476378439257e-08,
      "loss": 1.5347,
      "num_input_tokens_seen": 76561864,
      "step": 131965
    },
    {
      "epoch": 19.655942806076855,
      "grad_norm": 3.078125,
      "learning_rate": 4.50994604087901e-08,
      "loss": 0.787,
      "num_input_tokens_seen": 76564648,
      "step": 131970
    },
    {
      "epoch": 19.656687518617815,
      "grad_norm": 10.125,
      "learning_rate": 4.490457862087183e-08,
      "loss": 1.2225,
      "num_input_tokens_seen": 76567592,
      "step": 131975
    },
    {
      "epoch": 19.657432231158772,
      "grad_norm": 11.5625,
      "learning_rate": 4.471011842392403e-08,
      "loss": 1.318,
      "num_input_tokens_seen": 76570472,
      "step": 131980
    },
    {
      "epoch": 19.658176943699733,
      "grad_norm": 15.75,
      "learning_rate": 4.451607982123851e-08,
      "loss": 1.1415,
      "num_input_tokens_seen": 76573192,
      "step": 131985
    },
    {
      "epoch": 19.65892165624069,
      "grad_norm": 6.5,
      "learning_rate": 4.432246281609042e-08,
      "loss": 0.89,
      "num_input_tokens_seen": 76576008,
      "step": 131990
    },
    {
      "epoch": 19.65966636878165,
      "grad_norm": 12.5,
      "learning_rate": 4.4129267411749386e-08,
      "loss": 1.2207,
      "num_input_tokens_seen": 76578856,
      "step": 131995
    },
    {
      "epoch": 19.66041108132261,
      "grad_norm": 14.3125,
      "learning_rate": 4.393649361147944e-08,
      "loss": 0.9516,
      "num_input_tokens_seen": 76581928,
      "step": 132000
    },
    {
      "epoch": 19.661155793863568,
      "grad_norm": 14.75,
      "learning_rate": 4.374414141853911e-08,
      "loss": 0.8381,
      "num_input_tokens_seen": 76585000,
      "step": 132005
    },
    {
      "epoch": 19.66190050640453,
      "grad_norm": 15.9375,
      "learning_rate": 4.355221083617578e-08,
      "loss": 1.4718,
      "num_input_tokens_seen": 76587816,
      "step": 132010
    },
    {
      "epoch": 19.662645218945485,
      "grad_norm": 7.46875,
      "learning_rate": 4.336070186763685e-08,
      "loss": 1.0814,
      "num_input_tokens_seen": 76590728,
      "step": 132015
    },
    {
      "epoch": 19.663389931486446,
      "grad_norm": 7.84375,
      "learning_rate": 4.316961451615031e-08,
      "loss": 1.2366,
      "num_input_tokens_seen": 76593640,
      "step": 132020
    },
    {
      "epoch": 19.664134644027406,
      "grad_norm": 8.0625,
      "learning_rate": 4.297894878494968e-08,
      "loss": 1.0612,
      "num_input_tokens_seen": 76596488,
      "step": 132025
    },
    {
      "epoch": 19.664879356568363,
      "grad_norm": 11.3125,
      "learning_rate": 4.27887046772546e-08,
      "loss": 0.8223,
      "num_input_tokens_seen": 76599336,
      "step": 132030
    },
    {
      "epoch": 19.665624069109324,
      "grad_norm": 13.0625,
      "learning_rate": 4.2598882196279165e-08,
      "loss": 0.8685,
      "num_input_tokens_seen": 76602216,
      "step": 132035
    },
    {
      "epoch": 19.666368781650284,
      "grad_norm": 6.90625,
      "learning_rate": 4.240948134522915e-08,
      "loss": 0.8212,
      "num_input_tokens_seen": 76605384,
      "step": 132040
    },
    {
      "epoch": 19.66711349419124,
      "grad_norm": 6.71875,
      "learning_rate": 4.2220502127304775e-08,
      "loss": 1.0794,
      "num_input_tokens_seen": 76608392,
      "step": 132045
    },
    {
      "epoch": 19.667858206732202,
      "grad_norm": 12.6875,
      "learning_rate": 4.2031944545700696e-08,
      "loss": 1.2416,
      "num_input_tokens_seen": 76611496,
      "step": 132050
    },
    {
      "epoch": 19.66860291927316,
      "grad_norm": 9.25,
      "learning_rate": 4.184380860360049e-08,
      "loss": 1.2066,
      "num_input_tokens_seen": 76614504,
      "step": 132055
    },
    {
      "epoch": 19.66934763181412,
      "grad_norm": 9.9375,
      "learning_rate": 4.165609430418216e-08,
      "loss": 1.1045,
      "num_input_tokens_seen": 76617352,
      "step": 132060
    },
    {
      "epoch": 19.67009234435508,
      "grad_norm": 11.1875,
      "learning_rate": 4.1468801650618175e-08,
      "loss": 1.2446,
      "num_input_tokens_seen": 76620264,
      "step": 132065
    },
    {
      "epoch": 19.670837056896037,
      "grad_norm": 9.0625,
      "learning_rate": 4.128193064606989e-08,
      "loss": 0.8211,
      "num_input_tokens_seen": 76623400,
      "step": 132070
    },
    {
      "epoch": 19.671581769436997,
      "grad_norm": 4.96875,
      "learning_rate": 4.1095481293698665e-08,
      "loss": 0.7581,
      "num_input_tokens_seen": 76626184,
      "step": 132075
    },
    {
      "epoch": 19.672326481977958,
      "grad_norm": 12.0625,
      "learning_rate": 4.0909453596651995e-08,
      "loss": 0.971,
      "num_input_tokens_seen": 76629000,
      "step": 132080
    },
    {
      "epoch": 19.673071194518915,
      "grad_norm": 8.3125,
      "learning_rate": 4.0723847558071795e-08,
      "loss": 0.9943,
      "num_input_tokens_seen": 76632040,
      "step": 132085
    },
    {
      "epoch": 19.673815907059875,
      "grad_norm": 7.9375,
      "learning_rate": 4.0538663181097224e-08,
      "loss": 0.956,
      "num_input_tokens_seen": 76635176,
      "step": 132090
    },
    {
      "epoch": 19.674560619600832,
      "grad_norm": 4.0625,
      "learning_rate": 4.035390046885079e-08,
      "loss": 1.2023,
      "num_input_tokens_seen": 76638152,
      "step": 132095
    },
    {
      "epoch": 19.675305332141793,
      "grad_norm": 8.0,
      "learning_rate": 4.016955942446055e-08,
      "loss": 1.3229,
      "num_input_tokens_seen": 76641096,
      "step": 132100
    },
    {
      "epoch": 19.676050044682754,
      "grad_norm": 7.75,
      "learning_rate": 3.9985640051035114e-08,
      "loss": 1.1312,
      "num_input_tokens_seen": 76644040,
      "step": 132105
    },
    {
      "epoch": 19.67679475722371,
      "grad_norm": 10.4375,
      "learning_rate": 3.980214235168589e-08,
      "loss": 1.1111,
      "num_input_tokens_seen": 76647400,
      "step": 132110
    },
    {
      "epoch": 19.67753946976467,
      "grad_norm": 14.4375,
      "learning_rate": 3.96190663295104e-08,
      "loss": 1.3549,
      "num_input_tokens_seen": 76650440,
      "step": 132115
    },
    {
      "epoch": 19.67828418230563,
      "grad_norm": 11.9375,
      "learning_rate": 3.943641198760062e-08,
      "loss": 1.1196,
      "num_input_tokens_seen": 76653384,
      "step": 132120
    },
    {
      "epoch": 19.67902889484659,
      "grad_norm": 7.125,
      "learning_rate": 3.925417932904574e-08,
      "loss": 0.9882,
      "num_input_tokens_seen": 76656424,
      "step": 132125
    },
    {
      "epoch": 19.67977360738755,
      "grad_norm": 11.25,
      "learning_rate": 3.907236835692385e-08,
      "loss": 1.3064,
      "num_input_tokens_seen": 76659016,
      "step": 132130
    },
    {
      "epoch": 19.680518319928506,
      "grad_norm": 16.625,
      "learning_rate": 3.8890979074301946e-08,
      "loss": 1.2221,
      "num_input_tokens_seen": 76661832,
      "step": 132135
    },
    {
      "epoch": 19.681263032469467,
      "grad_norm": 7.46875,
      "learning_rate": 3.8710011484249795e-08,
      "loss": 1.268,
      "num_input_tokens_seen": 76665032,
      "step": 132140
    },
    {
      "epoch": 19.682007745010427,
      "grad_norm": 14.3125,
      "learning_rate": 3.8529465589820514e-08,
      "loss": 1.0918,
      "num_input_tokens_seen": 76667656,
      "step": 132145
    },
    {
      "epoch": 19.682752457551384,
      "grad_norm": 16.5,
      "learning_rate": 3.8349341394067204e-08,
      "loss": 0.8815,
      "num_input_tokens_seen": 76670344,
      "step": 132150
    },
    {
      "epoch": 19.683497170092345,
      "grad_norm": 9.75,
      "learning_rate": 3.816963890003189e-08,
      "loss": 0.8228,
      "num_input_tokens_seen": 76673096,
      "step": 132155
    },
    {
      "epoch": 19.684241882633305,
      "grad_norm": 10.0625,
      "learning_rate": 3.799035811075102e-08,
      "loss": 1.4145,
      "num_input_tokens_seen": 76676104,
      "step": 132160
    },
    {
      "epoch": 19.684986595174262,
      "grad_norm": 7.46875,
      "learning_rate": 3.7811499029252737e-08,
      "loss": 0.9406,
      "num_input_tokens_seen": 76678760,
      "step": 132165
    },
    {
      "epoch": 19.685731307715223,
      "grad_norm": 8.75,
      "learning_rate": 3.763306165855962e-08,
      "loss": 1.1248,
      "num_input_tokens_seen": 76681768,
      "step": 132170
    },
    {
      "epoch": 19.68647602025618,
      "grad_norm": 8.8125,
      "learning_rate": 3.745504600168315e-08,
      "loss": 0.7296,
      "num_input_tokens_seen": 76684520,
      "step": 132175
    },
    {
      "epoch": 19.68722073279714,
      "grad_norm": 6.21875,
      "learning_rate": 3.72774520616348e-08,
      "loss": 0.9523,
      "num_input_tokens_seen": 76687528,
      "step": 132180
    },
    {
      "epoch": 19.6879654453381,
      "grad_norm": 9.9375,
      "learning_rate": 3.7100279841412177e-08,
      "loss": 1.0707,
      "num_input_tokens_seen": 76690184,
      "step": 132185
    },
    {
      "epoch": 19.688710157879058,
      "grad_norm": 6.0,
      "learning_rate": 3.6923529344007336e-08,
      "loss": 0.8053,
      "num_input_tokens_seen": 76693128,
      "step": 132190
    },
    {
      "epoch": 19.68945487042002,
      "grad_norm": 9.25,
      "learning_rate": 3.674720057240955e-08,
      "loss": 1.2075,
      "num_input_tokens_seen": 76696136,
      "step": 132195
    },
    {
      "epoch": 19.69019958296098,
      "grad_norm": 10.0,
      "learning_rate": 3.657129352959698e-08,
      "loss": 1.0929,
      "num_input_tokens_seen": 76699080,
      "step": 132200
    },
    {
      "epoch": 19.690944295501936,
      "grad_norm": 9.625,
      "learning_rate": 3.63958082185395e-08,
      "loss": 1.1587,
      "num_input_tokens_seen": 76701544,
      "step": 132205
    },
    {
      "epoch": 19.691689008042896,
      "grad_norm": 5.3125,
      "learning_rate": 3.622074464220415e-08,
      "loss": 0.8699,
      "num_input_tokens_seen": 76704648,
      "step": 132210
    },
    {
      "epoch": 19.692433720583853,
      "grad_norm": 14.3125,
      "learning_rate": 3.604610280354692e-08,
      "loss": 1.0941,
      "num_input_tokens_seen": 76707912,
      "step": 132215
    },
    {
      "epoch": 19.693178433124814,
      "grad_norm": 13.75,
      "learning_rate": 3.587188270551822e-08,
      "loss": 0.9946,
      "num_input_tokens_seen": 76710408,
      "step": 132220
    },
    {
      "epoch": 19.693923145665774,
      "grad_norm": 9.4375,
      "learning_rate": 3.569808435106292e-08,
      "loss": 1.2501,
      "num_input_tokens_seen": 76713320,
      "step": 132225
    },
    {
      "epoch": 19.69466785820673,
      "grad_norm": 24.625,
      "learning_rate": 3.552470774311478e-08,
      "loss": 1.2889,
      "num_input_tokens_seen": 76716040,
      "step": 132230
    },
    {
      "epoch": 19.695412570747692,
      "grad_norm": 7.9375,
      "learning_rate": 3.535175288460479e-08,
      "loss": 0.7767,
      "num_input_tokens_seen": 76718920,
      "step": 132235
    },
    {
      "epoch": 19.69615728328865,
      "grad_norm": 8.5625,
      "learning_rate": 3.5179219778452846e-08,
      "loss": 1.3838,
      "num_input_tokens_seen": 76722024,
      "step": 132240
    },
    {
      "epoch": 19.69690199582961,
      "grad_norm": 13.0625,
      "learning_rate": 3.500710842757604e-08,
      "loss": 0.8832,
      "num_input_tokens_seen": 76724936,
      "step": 132245
    },
    {
      "epoch": 19.69764670837057,
      "grad_norm": 8.9375,
      "learning_rate": 3.483541883487762e-08,
      "loss": 1.3302,
      "num_input_tokens_seen": 76727720,
      "step": 132250
    },
    {
      "epoch": 19.698391420911527,
      "grad_norm": 12.5,
      "learning_rate": 3.466415100326359e-08,
      "loss": 1.1608,
      "num_input_tokens_seen": 76731496,
      "step": 132255
    },
    {
      "epoch": 19.699136133452487,
      "grad_norm": 18.375,
      "learning_rate": 3.449330493562608e-08,
      "loss": 1.1113,
      "num_input_tokens_seen": 76734696,
      "step": 132260
    },
    {
      "epoch": 19.699880845993448,
      "grad_norm": 10.625,
      "learning_rate": 3.4322880634851674e-08,
      "loss": 0.991,
      "num_input_tokens_seen": 76737544,
      "step": 132265
    },
    {
      "epoch": 19.700625558534405,
      "grad_norm": 10.75,
      "learning_rate": 3.415287810381584e-08,
      "loss": 1.0988,
      "num_input_tokens_seen": 76740328,
      "step": 132270
    },
    {
      "epoch": 19.701370271075366,
      "grad_norm": 10.6875,
      "learning_rate": 3.3983297345391296e-08,
      "loss": 0.9639,
      "num_input_tokens_seen": 76743176,
      "step": 132275
    },
    {
      "epoch": 19.702114983616323,
      "grad_norm": 8.8125,
      "learning_rate": 3.381413836244796e-08,
      "loss": 0.8466,
      "num_input_tokens_seen": 76746248,
      "step": 132280
    },
    {
      "epoch": 19.702859696157283,
      "grad_norm": 5.875,
      "learning_rate": 3.36454011578391e-08,
      "loss": 0.9378,
      "num_input_tokens_seen": 76749224,
      "step": 132285
    },
    {
      "epoch": 19.703604408698244,
      "grad_norm": 8.625,
      "learning_rate": 3.347708573441521e-08,
      "loss": 1.2356,
      "num_input_tokens_seen": 76751976,
      "step": 132290
    },
    {
      "epoch": 19.7043491212392,
      "grad_norm": 12.8125,
      "learning_rate": 3.330919209502126e-08,
      "loss": 1.3053,
      "num_input_tokens_seen": 76754664,
      "step": 132295
    },
    {
      "epoch": 19.70509383378016,
      "grad_norm": 9.25,
      "learning_rate": 3.314172024249662e-08,
      "loss": 1.2174,
      "num_input_tokens_seen": 76757576,
      "step": 132300
    },
    {
      "epoch": 19.70583854632112,
      "grad_norm": 10.0,
      "learning_rate": 3.297467017966405e-08,
      "loss": 1.1356,
      "num_input_tokens_seen": 76760168,
      "step": 132305
    },
    {
      "epoch": 19.70658325886208,
      "grad_norm": 4.8125,
      "learning_rate": 3.280804190935183e-08,
      "loss": 0.8498,
      "num_input_tokens_seen": 76762888,
      "step": 132310
    },
    {
      "epoch": 19.70732797140304,
      "grad_norm": 5.5,
      "learning_rate": 3.264183543436883e-08,
      "loss": 0.8979,
      "num_input_tokens_seen": 76765416,
      "step": 132315
    },
    {
      "epoch": 19.708072683943996,
      "grad_norm": 18.625,
      "learning_rate": 3.2476050757529466e-08,
      "loss": 1.6762,
      "num_input_tokens_seen": 76768264,
      "step": 132320
    },
    {
      "epoch": 19.708817396484957,
      "grad_norm": 16.25,
      "learning_rate": 3.231068788162872e-08,
      "loss": 1.3169,
      "num_input_tokens_seen": 76771496,
      "step": 132325
    },
    {
      "epoch": 19.709562109025917,
      "grad_norm": 8.6875,
      "learning_rate": 3.214574680946436e-08,
      "loss": 1.354,
      "num_input_tokens_seen": 76774568,
      "step": 132330
    },
    {
      "epoch": 19.710306821566874,
      "grad_norm": 8.8125,
      "learning_rate": 3.198122754382305e-08,
      "loss": 1.0787,
      "num_input_tokens_seen": 76777384,
      "step": 132335
    },
    {
      "epoch": 19.711051534107835,
      "grad_norm": 9.8125,
      "learning_rate": 3.181713008748033e-08,
      "loss": 1.1589,
      "num_input_tokens_seen": 76780200,
      "step": 132340
    },
    {
      "epoch": 19.711796246648795,
      "grad_norm": 12.5,
      "learning_rate": 3.1653454443211774e-08,
      "loss": 1.2494,
      "num_input_tokens_seen": 76782984,
      "step": 132345
    },
    {
      "epoch": 19.712540959189752,
      "grad_norm": 14.5625,
      "learning_rate": 3.1490200613779056e-08,
      "loss": 1.255,
      "num_input_tokens_seen": 76785832,
      "step": 132350
    },
    {
      "epoch": 19.713285671730713,
      "grad_norm": 10.3125,
      "learning_rate": 3.132736860194385e-08,
      "loss": 1.2075,
      "num_input_tokens_seen": 76788840,
      "step": 132355
    },
    {
      "epoch": 19.71403038427167,
      "grad_norm": 10.75,
      "learning_rate": 3.116495841045675e-08,
      "loss": 0.7096,
      "num_input_tokens_seen": 76791528,
      "step": 132360
    },
    {
      "epoch": 19.71477509681263,
      "grad_norm": 13.375,
      "learning_rate": 3.1002970042059984e-08,
      "loss": 1.2045,
      "num_input_tokens_seen": 76794664,
      "step": 132365
    },
    {
      "epoch": 19.71551980935359,
      "grad_norm": 22.25,
      "learning_rate": 3.084140349949027e-08,
      "loss": 1.1539,
      "num_input_tokens_seen": 76797384,
      "step": 132370
    },
    {
      "epoch": 19.716264521894548,
      "grad_norm": 21.25,
      "learning_rate": 3.0680258785478756e-08,
      "loss": 0.9816,
      "num_input_tokens_seen": 76800328,
      "step": 132375
    },
    {
      "epoch": 19.71700923443551,
      "grad_norm": 10.75,
      "learning_rate": 3.051953590274548e-08,
      "loss": 1.4158,
      "num_input_tokens_seen": 76802952,
      "step": 132380
    },
    {
      "epoch": 19.717753946976465,
      "grad_norm": 11.25,
      "learning_rate": 3.035923485400771e-08,
      "loss": 1.122,
      "num_input_tokens_seen": 76805960,
      "step": 132385
    },
    {
      "epoch": 19.718498659517426,
      "grad_norm": 15.9375,
      "learning_rate": 3.0199355641971626e-08,
      "loss": 1.3177,
      "num_input_tokens_seen": 76808904,
      "step": 132390
    },
    {
      "epoch": 19.719243372058386,
      "grad_norm": 16.25,
      "learning_rate": 3.003989826934062e-08,
      "loss": 0.9414,
      "num_input_tokens_seen": 76811720,
      "step": 132395
    },
    {
      "epoch": 19.719988084599343,
      "grad_norm": 6.90625,
      "learning_rate": 2.9880862738804196e-08,
      "loss": 0.9497,
      "num_input_tokens_seen": 76814472,
      "step": 132400
    },
    {
      "epoch": 19.720732797140304,
      "grad_norm": 8.0,
      "learning_rate": 2.9722249053054653e-08,
      "loss": 0.9077,
      "num_input_tokens_seen": 76817160,
      "step": 132405
    },
    {
      "epoch": 19.721477509681264,
      "grad_norm": 12.375,
      "learning_rate": 2.9564057214767627e-08,
      "loss": 0.9591,
      "num_input_tokens_seen": 76819944,
      "step": 132410
    },
    {
      "epoch": 19.72222222222222,
      "grad_norm": 9.0625,
      "learning_rate": 2.9406287226618756e-08,
      "loss": 1.0767,
      "num_input_tokens_seen": 76823016,
      "step": 132415
    },
    {
      "epoch": 19.722966934763182,
      "grad_norm": 18.25,
      "learning_rate": 2.92489390912698e-08,
      "loss": 0.9981,
      "num_input_tokens_seen": 76826024,
      "step": 132420
    },
    {
      "epoch": 19.72371164730414,
      "grad_norm": 10.4375,
      "learning_rate": 2.909201281138252e-08,
      "loss": 0.6388,
      "num_input_tokens_seen": 76828808,
      "step": 132425
    },
    {
      "epoch": 19.7244563598451,
      "grad_norm": 18.125,
      "learning_rate": 2.8935508389607568e-08,
      "loss": 1.0994,
      "num_input_tokens_seen": 76831816,
      "step": 132430
    },
    {
      "epoch": 19.72520107238606,
      "grad_norm": 16.375,
      "learning_rate": 2.8779425828584506e-08,
      "loss": 1.0156,
      "num_input_tokens_seen": 76834632,
      "step": 132435
    },
    {
      "epoch": 19.725945784927017,
      "grad_norm": 4.84375,
      "learning_rate": 2.862376513095566e-08,
      "loss": 0.913,
      "num_input_tokens_seen": 76837512,
      "step": 132440
    },
    {
      "epoch": 19.726690497467978,
      "grad_norm": 7.6875,
      "learning_rate": 2.846852629934671e-08,
      "loss": 0.8148,
      "num_input_tokens_seen": 76840168,
      "step": 132445
    },
    {
      "epoch": 19.727435210008938,
      "grad_norm": 4.6875,
      "learning_rate": 2.831370933638333e-08,
      "loss": 0.877,
      "num_input_tokens_seen": 76843176,
      "step": 132450
    },
    {
      "epoch": 19.728179922549895,
      "grad_norm": 12.25,
      "learning_rate": 2.8159314244680103e-08,
      "loss": 1.0586,
      "num_input_tokens_seen": 76846216,
      "step": 132455
    },
    {
      "epoch": 19.728924635090856,
      "grad_norm": 12.0625,
      "learning_rate": 2.800534102684327e-08,
      "loss": 0.9552,
      "num_input_tokens_seen": 76849128,
      "step": 132460
    },
    {
      "epoch": 19.729669347631813,
      "grad_norm": 4.9375,
      "learning_rate": 2.7851789685476304e-08,
      "loss": 0.9538,
      "num_input_tokens_seen": 76852008,
      "step": 132465
    },
    {
      "epoch": 19.730414060172773,
      "grad_norm": 8.1875,
      "learning_rate": 2.7698660223174355e-08,
      "loss": 1.005,
      "num_input_tokens_seen": 76854920,
      "step": 132470
    },
    {
      "epoch": 19.731158772713734,
      "grad_norm": 10.875,
      "learning_rate": 2.7545952642521466e-08,
      "loss": 1.0863,
      "num_input_tokens_seen": 76857896,
      "step": 132475
    },
    {
      "epoch": 19.73190348525469,
      "grad_norm": 12.8125,
      "learning_rate": 2.7393666946098906e-08,
      "loss": 0.9885,
      "num_input_tokens_seen": 76860808,
      "step": 132480
    },
    {
      "epoch": 19.73264819779565,
      "grad_norm": 17.375,
      "learning_rate": 2.7241803136479616e-08,
      "loss": 1.2476,
      "num_input_tokens_seen": 76863656,
      "step": 132485
    },
    {
      "epoch": 19.73339291033661,
      "grad_norm": 11.0625,
      "learning_rate": 2.7090361216230987e-08,
      "loss": 1.331,
      "num_input_tokens_seen": 76866600,
      "step": 132490
    },
    {
      "epoch": 19.73413762287757,
      "grad_norm": 11.75,
      "learning_rate": 2.693934118790653e-08,
      "loss": 1.0411,
      "num_input_tokens_seen": 76869608,
      "step": 132495
    },
    {
      "epoch": 19.73488233541853,
      "grad_norm": 8.625,
      "learning_rate": 2.678874305405976e-08,
      "loss": 0.8763,
      "num_input_tokens_seen": 76872264,
      "step": 132500
    },
    {
      "epoch": 19.735627047959486,
      "grad_norm": 10.4375,
      "learning_rate": 2.663856681723864e-08,
      "loss": 0.8403,
      "num_input_tokens_seen": 76875208,
      "step": 132505
    },
    {
      "epoch": 19.736371760500447,
      "grad_norm": 8.5625,
      "learning_rate": 2.6488812479974478e-08,
      "loss": 0.9194,
      "num_input_tokens_seen": 76877960,
      "step": 132510
    },
    {
      "epoch": 19.737116473041407,
      "grad_norm": 13.0,
      "learning_rate": 2.6339480044801355e-08,
      "loss": 1.2241,
      "num_input_tokens_seen": 76880808,
      "step": 132515
    },
    {
      "epoch": 19.737861185582364,
      "grad_norm": 8.625,
      "learning_rate": 2.6190569514239484e-08,
      "loss": 1.1775,
      "num_input_tokens_seen": 76883688,
      "step": 132520
    },
    {
      "epoch": 19.738605898123325,
      "grad_norm": 14.625,
      "learning_rate": 2.6042080890806285e-08,
      "loss": 0.8817,
      "num_input_tokens_seen": 76886536,
      "step": 132525
    },
    {
      "epoch": 19.73935061066428,
      "grad_norm": 8.5,
      "learning_rate": 2.5894014177010872e-08,
      "loss": 0.8368,
      "num_input_tokens_seen": 76889608,
      "step": 132530
    },
    {
      "epoch": 19.740095323205242,
      "grad_norm": 8.1875,
      "learning_rate": 2.5746369375351242e-08,
      "loss": 1.2719,
      "num_input_tokens_seen": 76892904,
      "step": 132535
    },
    {
      "epoch": 19.740840035746203,
      "grad_norm": 5.875,
      "learning_rate": 2.55991464883254e-08,
      "loss": 1.2529,
      "num_input_tokens_seen": 76895784,
      "step": 132540
    },
    {
      "epoch": 19.74158474828716,
      "grad_norm": 8.3125,
      "learning_rate": 2.545234551842024e-08,
      "loss": 1.0195,
      "num_input_tokens_seen": 76898792,
      "step": 132545
    },
    {
      "epoch": 19.74232946082812,
      "grad_norm": 13.4375,
      "learning_rate": 2.530596646811434e-08,
      "loss": 1.1731,
      "num_input_tokens_seen": 76901768,
      "step": 132550
    },
    {
      "epoch": 19.74307417336908,
      "grad_norm": 10.5625,
      "learning_rate": 2.5160009339877944e-08,
      "loss": 1.1348,
      "num_input_tokens_seen": 76904904,
      "step": 132555
    },
    {
      "epoch": 19.743818885910038,
      "grad_norm": 7.6875,
      "learning_rate": 2.5014474136184075e-08,
      "loss": 1.2729,
      "num_input_tokens_seen": 76908072,
      "step": 132560
    },
    {
      "epoch": 19.744563598451,
      "grad_norm": 7.53125,
      "learning_rate": 2.486936085948355e-08,
      "loss": 0.8444,
      "num_input_tokens_seen": 76910664,
      "step": 132565
    },
    {
      "epoch": 19.74530831099196,
      "grad_norm": 18.5,
      "learning_rate": 2.4724669512232734e-08,
      "loss": 1.2989,
      "num_input_tokens_seen": 76913608,
      "step": 132570
    },
    {
      "epoch": 19.746053023532916,
      "grad_norm": 6.15625,
      "learning_rate": 2.45804000968769e-08,
      "loss": 1.0747,
      "num_input_tokens_seen": 76916584,
      "step": 132575
    },
    {
      "epoch": 19.746797736073876,
      "grad_norm": 4.75,
      "learning_rate": 2.4436552615850205e-08,
      "loss": 1.2379,
      "num_input_tokens_seen": 76919720,
      "step": 132580
    },
    {
      "epoch": 19.747542448614833,
      "grad_norm": 10.0,
      "learning_rate": 2.4293127071584043e-08,
      "loss": 0.9043,
      "num_input_tokens_seen": 76922632,
      "step": 132585
    },
    {
      "epoch": 19.748287161155794,
      "grad_norm": 4.71875,
      "learning_rate": 2.4150123466498698e-08,
      "loss": 1.2114,
      "num_input_tokens_seen": 76925640,
      "step": 132590
    },
    {
      "epoch": 19.749031873696755,
      "grad_norm": 12.5625,
      "learning_rate": 2.400754180301723e-08,
      "loss": 1.1245,
      "num_input_tokens_seen": 76928456,
      "step": 132595
    },
    {
      "epoch": 19.74977658623771,
      "grad_norm": 8.6875,
      "learning_rate": 2.38653820835405e-08,
      "loss": 1.3116,
      "num_input_tokens_seen": 76931496,
      "step": 132600
    },
    {
      "epoch": 19.750521298778672,
      "grad_norm": 12.25,
      "learning_rate": 2.3723644310474914e-08,
      "loss": 0.8196,
      "num_input_tokens_seen": 76934312,
      "step": 132605
    },
    {
      "epoch": 19.75126601131963,
      "grad_norm": 13.6875,
      "learning_rate": 2.3582328486213e-08,
      "loss": 1.3771,
      "num_input_tokens_seen": 76937128,
      "step": 132610
    },
    {
      "epoch": 19.75201072386059,
      "grad_norm": 13.75,
      "learning_rate": 2.3441434613141743e-08,
      "loss": 1.0889,
      "num_input_tokens_seen": 76940072,
      "step": 132615
    },
    {
      "epoch": 19.75275543640155,
      "grad_norm": 12.1875,
      "learning_rate": 2.3300962693645345e-08,
      "loss": 1.0492,
      "num_input_tokens_seen": 76943144,
      "step": 132620
    },
    {
      "epoch": 19.753500148942507,
      "grad_norm": 16.375,
      "learning_rate": 2.3160912730091357e-08,
      "loss": 0.9635,
      "num_input_tokens_seen": 76945864,
      "step": 132625
    },
    {
      "epoch": 19.754244861483468,
      "grad_norm": 11.875,
      "learning_rate": 2.30212847248501e-08,
      "loss": 0.8636,
      "num_input_tokens_seen": 76948968,
      "step": 132630
    },
    {
      "epoch": 19.754989574024428,
      "grad_norm": 9.8125,
      "learning_rate": 2.288207868027803e-08,
      "loss": 0.9867,
      "num_input_tokens_seen": 76951848,
      "step": 132635
    },
    {
      "epoch": 19.755734286565385,
      "grad_norm": 6.1875,
      "learning_rate": 2.2743294598726038e-08,
      "loss": 1.0059,
      "num_input_tokens_seen": 76954952,
      "step": 132640
    },
    {
      "epoch": 19.756478999106346,
      "grad_norm": 7.25,
      "learning_rate": 2.260493248254225e-08,
      "loss": 1.1505,
      "num_input_tokens_seen": 76957960,
      "step": 132645
    },
    {
      "epoch": 19.757223711647303,
      "grad_norm": 13.875,
      "learning_rate": 2.246699233406091e-08,
      "loss": 0.9563,
      "num_input_tokens_seen": 76961032,
      "step": 132650
    },
    {
      "epoch": 19.757968424188263,
      "grad_norm": 18.625,
      "learning_rate": 2.2329474155616258e-08,
      "loss": 1.2204,
      "num_input_tokens_seen": 76964008,
      "step": 132655
    },
    {
      "epoch": 19.758713136729224,
      "grad_norm": 15.4375,
      "learning_rate": 2.2192377949525888e-08,
      "loss": 1.1067,
      "num_input_tokens_seen": 76966856,
      "step": 132660
    },
    {
      "epoch": 19.75945784927018,
      "grad_norm": 11.1875,
      "learning_rate": 2.205570371811294e-08,
      "loss": 1.4185,
      "num_input_tokens_seen": 76969704,
      "step": 132665
    },
    {
      "epoch": 19.76020256181114,
      "grad_norm": 8.75,
      "learning_rate": 2.1919451463678353e-08,
      "loss": 0.9613,
      "num_input_tokens_seen": 76972840,
      "step": 132670
    },
    {
      "epoch": 19.7609472743521,
      "grad_norm": 19.5,
      "learning_rate": 2.178362118853139e-08,
      "loss": 1.2676,
      "num_input_tokens_seen": 76975880,
      "step": 132675
    },
    {
      "epoch": 19.76169198689306,
      "grad_norm": 9.1875,
      "learning_rate": 2.1648212894961884e-08,
      "loss": 1.0604,
      "num_input_tokens_seen": 76978664,
      "step": 132680
    },
    {
      "epoch": 19.76243669943402,
      "grad_norm": 17.625,
      "learning_rate": 2.1513226585256897e-08,
      "loss": 1.172,
      "num_input_tokens_seen": 76981352,
      "step": 132685
    },
    {
      "epoch": 19.763181411974976,
      "grad_norm": 11.9375,
      "learning_rate": 2.137866226170071e-08,
      "loss": 1.301,
      "num_input_tokens_seen": 76984072,
      "step": 132690
    },
    {
      "epoch": 19.763926124515937,
      "grad_norm": 9.6875,
      "learning_rate": 2.1244519926566507e-08,
      "loss": 1.0464,
      "num_input_tokens_seen": 76986856,
      "step": 132695
    },
    {
      "epoch": 19.764670837056897,
      "grad_norm": 8.25,
      "learning_rate": 2.111079958211637e-08,
      "loss": 1.2686,
      "num_input_tokens_seen": 76989864,
      "step": 132700
    },
    {
      "epoch": 19.765415549597854,
      "grad_norm": 14.5,
      "learning_rate": 2.0977501230612374e-08,
      "loss": 1.1517,
      "num_input_tokens_seen": 76992840,
      "step": 132705
    },
    {
      "epoch": 19.766160262138815,
      "grad_norm": 8.75,
      "learning_rate": 2.0844624874305497e-08,
      "loss": 0.8165,
      "num_input_tokens_seen": 76995592,
      "step": 132710
    },
    {
      "epoch": 19.766904974679775,
      "grad_norm": 12.3125,
      "learning_rate": 2.0712170515443942e-08,
      "loss": 1.0402,
      "num_input_tokens_seen": 76998664,
      "step": 132715
    },
    {
      "epoch": 19.767649687220732,
      "grad_norm": 9.1875,
      "learning_rate": 2.0580138156259256e-08,
      "loss": 0.9844,
      "num_input_tokens_seen": 77001576,
      "step": 132720
    },
    {
      "epoch": 19.768394399761693,
      "grad_norm": 8.1875,
      "learning_rate": 2.0448527798985762e-08,
      "loss": 1.1898,
      "num_input_tokens_seen": 77004392,
      "step": 132725
    },
    {
      "epoch": 19.76913911230265,
      "grad_norm": 5.90625,
      "learning_rate": 2.0317339445849458e-08,
      "loss": 0.9371,
      "num_input_tokens_seen": 77007400,
      "step": 132730
    },
    {
      "epoch": 19.76988382484361,
      "grad_norm": 10.125,
      "learning_rate": 2.0186573099059693e-08,
      "loss": 0.7292,
      "num_input_tokens_seen": 77010376,
      "step": 132735
    },
    {
      "epoch": 19.77062853738457,
      "grad_norm": 10.5,
      "learning_rate": 2.005622876083135e-08,
      "loss": 0.9848,
      "num_input_tokens_seen": 77013288,
      "step": 132740
    },
    {
      "epoch": 19.771373249925528,
      "grad_norm": 5.40625,
      "learning_rate": 1.9926306433362683e-08,
      "loss": 1.1949,
      "num_input_tokens_seen": 77016296,
      "step": 132745
    },
    {
      "epoch": 19.77211796246649,
      "grad_norm": 13.0625,
      "learning_rate": 1.979680611885193e-08,
      "loss": 1.186,
      "num_input_tokens_seen": 77019048,
      "step": 132750
    },
    {
      "epoch": 19.772862675007445,
      "grad_norm": 7.96875,
      "learning_rate": 1.9667727819486225e-08,
      "loss": 1.1335,
      "num_input_tokens_seen": 77022216,
      "step": 132755
    },
    {
      "epoch": 19.773607387548406,
      "grad_norm": 6.875,
      "learning_rate": 1.9539071537444387e-08,
      "loss": 1.0254,
      "num_input_tokens_seen": 77025128,
      "step": 132760
    },
    {
      "epoch": 19.774352100089367,
      "grad_norm": 5.78125,
      "learning_rate": 1.9410837274899674e-08,
      "loss": 1.0264,
      "num_input_tokens_seen": 77028072,
      "step": 132765
    },
    {
      "epoch": 19.775096812630323,
      "grad_norm": 9.3125,
      "learning_rate": 1.9283025034019797e-08,
      "loss": 1.2652,
      "num_input_tokens_seen": 77031048,
      "step": 132770
    },
    {
      "epoch": 19.775841525171284,
      "grad_norm": 4.53125,
      "learning_rate": 1.9155634816966916e-08,
      "loss": 1.4259,
      "num_input_tokens_seen": 77034440,
      "step": 132775
    },
    {
      "epoch": 19.776586237712245,
      "grad_norm": 12.1875,
      "learning_rate": 1.902866662588654e-08,
      "loss": 1.0365,
      "num_input_tokens_seen": 77037480,
      "step": 132780
    },
    {
      "epoch": 19.7773309502532,
      "grad_norm": 4.46875,
      "learning_rate": 1.890212046292972e-08,
      "loss": 0.9895,
      "num_input_tokens_seen": 77040488,
      "step": 132785
    },
    {
      "epoch": 19.778075662794162,
      "grad_norm": 9.6875,
      "learning_rate": 1.877599633023086e-08,
      "loss": 1.2104,
      "num_input_tokens_seen": 77043496,
      "step": 132790
    },
    {
      "epoch": 19.77882037533512,
      "grad_norm": 11.8125,
      "learning_rate": 1.865029422992437e-08,
      "loss": 1.4247,
      "num_input_tokens_seen": 77046408,
      "step": 132795
    },
    {
      "epoch": 19.77956508787608,
      "grad_norm": 10.4375,
      "learning_rate": 1.8525014164127996e-08,
      "loss": 1.3006,
      "num_input_tokens_seen": 77049224,
      "step": 132800
    },
    {
      "epoch": 19.78030980041704,
      "grad_norm": 16.125,
      "learning_rate": 1.8400156134962265e-08,
      "loss": 0.9729,
      "num_input_tokens_seen": 77052168,
      "step": 132805
    },
    {
      "epoch": 19.781054512957997,
      "grad_norm": 13.1875,
      "learning_rate": 1.82757201445366e-08,
      "loss": 1.2915,
      "num_input_tokens_seen": 77054888,
      "step": 132810
    },
    {
      "epoch": 19.781799225498958,
      "grad_norm": 13.75,
      "learning_rate": 1.8151706194952102e-08,
      "loss": 1.0743,
      "num_input_tokens_seen": 77057800,
      "step": 132815
    },
    {
      "epoch": 19.782543938039918,
      "grad_norm": 8.875,
      "learning_rate": 1.8028114288301535e-08,
      "loss": 1.25,
      "num_input_tokens_seen": 77060744,
      "step": 132820
    },
    {
      "epoch": 19.783288650580875,
      "grad_norm": 13.375,
      "learning_rate": 1.7904944426677673e-08,
      "loss": 1.0364,
      "num_input_tokens_seen": 77063784,
      "step": 132825
    },
    {
      "epoch": 19.784033363121836,
      "grad_norm": 6.875,
      "learning_rate": 1.7782196612159406e-08,
      "loss": 0.8196,
      "num_input_tokens_seen": 77066792,
      "step": 132830
    },
    {
      "epoch": 19.784778075662793,
      "grad_norm": 9.375,
      "learning_rate": 1.7659870846820083e-08,
      "loss": 1.0406,
      "num_input_tokens_seen": 77069640,
      "step": 132835
    },
    {
      "epoch": 19.785522788203753,
      "grad_norm": 8.5625,
      "learning_rate": 1.7537967132727483e-08,
      "loss": 1.0633,
      "num_input_tokens_seen": 77072552,
      "step": 132840
    },
    {
      "epoch": 19.786267500744714,
      "grad_norm": 6.625,
      "learning_rate": 1.7416485471938304e-08,
      "loss": 1.0094,
      "num_input_tokens_seen": 77075464,
      "step": 132845
    },
    {
      "epoch": 19.78701221328567,
      "grad_norm": 9.9375,
      "learning_rate": 1.7295425866506453e-08,
      "loss": 0.9367,
      "num_input_tokens_seen": 77078472,
      "step": 132850
    },
    {
      "epoch": 19.78775692582663,
      "grad_norm": 7.25,
      "learning_rate": 1.7174788318477518e-08,
      "loss": 1.4226,
      "num_input_tokens_seen": 77081416,
      "step": 132855
    },
    {
      "epoch": 19.788501638367592,
      "grad_norm": 7.375,
      "learning_rate": 1.705457282988876e-08,
      "loss": 1.2323,
      "num_input_tokens_seen": 77084360,
      "step": 132860
    },
    {
      "epoch": 19.78924635090855,
      "grad_norm": 9.9375,
      "learning_rate": 1.6934779402771884e-08,
      "loss": 1.0623,
      "num_input_tokens_seen": 77087336,
      "step": 132865
    },
    {
      "epoch": 19.78999106344951,
      "grad_norm": 13.6875,
      "learning_rate": 1.681540803915027e-08,
      "loss": 1.2029,
      "num_input_tokens_seen": 77089992,
      "step": 132870
    },
    {
      "epoch": 19.790735775990466,
      "grad_norm": 15.875,
      "learning_rate": 1.669645874103898e-08,
      "loss": 1.6219,
      "num_input_tokens_seen": 77092456,
      "step": 132875
    },
    {
      "epoch": 19.791480488531427,
      "grad_norm": 11.0625,
      "learning_rate": 1.6577931510450283e-08,
      "loss": 1.1706,
      "num_input_tokens_seen": 77095400,
      "step": 132880
    },
    {
      "epoch": 19.792225201072387,
      "grad_norm": 5.84375,
      "learning_rate": 1.6459826349385365e-08,
      "loss": 1.0172,
      "num_input_tokens_seen": 77098184,
      "step": 132885
    },
    {
      "epoch": 19.792969913613344,
      "grad_norm": 11.0,
      "learning_rate": 1.6342143259839848e-08,
      "loss": 1.1907,
      "num_input_tokens_seen": 77100712,
      "step": 132890
    },
    {
      "epoch": 19.793714626154305,
      "grad_norm": 12.25,
      "learning_rate": 1.622488224380103e-08,
      "loss": 0.933,
      "num_input_tokens_seen": 77103592,
      "step": 132895
    },
    {
      "epoch": 19.794459338695262,
      "grad_norm": 10.5,
      "learning_rate": 1.6108043303250664e-08,
      "loss": 0.7824,
      "num_input_tokens_seen": 77106472,
      "step": 132900
    },
    {
      "epoch": 19.795204051236222,
      "grad_norm": 9.875,
      "learning_rate": 1.5991626440162165e-08,
      "loss": 1.1099,
      "num_input_tokens_seen": 77109416,
      "step": 132905
    },
    {
      "epoch": 19.795948763777183,
      "grad_norm": 13.625,
      "learning_rate": 1.587563165650341e-08,
      "loss": 1.0694,
      "num_input_tokens_seen": 77112200,
      "step": 132910
    },
    {
      "epoch": 19.79669347631814,
      "grad_norm": 7.09375,
      "learning_rate": 1.5760058954233935e-08,
      "loss": 1.2844,
      "num_input_tokens_seen": 77114696,
      "step": 132915
    },
    {
      "epoch": 19.7974381888591,
      "grad_norm": 15.0,
      "learning_rate": 1.564490833530219e-08,
      "loss": 1.174,
      "num_input_tokens_seen": 77117832,
      "step": 132920
    },
    {
      "epoch": 19.79818290140006,
      "grad_norm": 7.625,
      "learning_rate": 1.5530179801659382e-08,
      "loss": 1.0102,
      "num_input_tokens_seen": 77120584,
      "step": 132925
    },
    {
      "epoch": 19.798927613941018,
      "grad_norm": 12.6875,
      "learning_rate": 1.5415873355240086e-08,
      "loss": 1.017,
      "num_input_tokens_seen": 77123272,
      "step": 132930
    },
    {
      "epoch": 19.79967232648198,
      "grad_norm": 10.75,
      "learning_rate": 1.5301988997978857e-08,
      "loss": 1.2174,
      "num_input_tokens_seen": 77126056,
      "step": 132935
    },
    {
      "epoch": 19.800417039022935,
      "grad_norm": 6.90625,
      "learning_rate": 1.5188526731793608e-08,
      "loss": 0.7013,
      "num_input_tokens_seen": 77128744,
      "step": 132940
    },
    {
      "epoch": 19.801161751563896,
      "grad_norm": 11.625,
      "learning_rate": 1.50754865586078e-08,
      "loss": 1.1291,
      "num_input_tokens_seen": 77131912,
      "step": 132945
    },
    {
      "epoch": 19.801906464104857,
      "grad_norm": 12.1875,
      "learning_rate": 1.4962868480325465e-08,
      "loss": 1.0517,
      "num_input_tokens_seen": 77134856,
      "step": 132950
    },
    {
      "epoch": 19.802651176645814,
      "grad_norm": 9.6875,
      "learning_rate": 1.4850672498853413e-08,
      "loss": 0.856,
      "num_input_tokens_seen": 77137768,
      "step": 132955
    },
    {
      "epoch": 19.803395889186774,
      "grad_norm": 14.0625,
      "learning_rate": 1.4738898616084573e-08,
      "loss": 1.1329,
      "num_input_tokens_seen": 77140648,
      "step": 132960
    },
    {
      "epoch": 19.804140601727735,
      "grad_norm": 2.859375,
      "learning_rate": 1.4627546833909101e-08,
      "loss": 1.032,
      "num_input_tokens_seen": 77143656,
      "step": 132965
    },
    {
      "epoch": 19.80488531426869,
      "grad_norm": 14.125,
      "learning_rate": 1.4516617154206048e-08,
      "loss": 0.895,
      "num_input_tokens_seen": 77146536,
      "step": 132970
    },
    {
      "epoch": 19.805630026809652,
      "grad_norm": 8.0625,
      "learning_rate": 1.440610957885169e-08,
      "loss": 0.912,
      "num_input_tokens_seen": 77149288,
      "step": 132975
    },
    {
      "epoch": 19.80637473935061,
      "grad_norm": 11.5,
      "learning_rate": 1.4296024109711203e-08,
      "loss": 1.2091,
      "num_input_tokens_seen": 77152200,
      "step": 132980
    },
    {
      "epoch": 19.80711945189157,
      "grad_norm": 11.8125,
      "learning_rate": 1.4186360748644212e-08,
      "loss": 1.2594,
      "num_input_tokens_seen": 77155080,
      "step": 132985
    },
    {
      "epoch": 19.80786416443253,
      "grad_norm": 4.90625,
      "learning_rate": 1.4077119497507562e-08,
      "loss": 1.0442,
      "num_input_tokens_seen": 77157896,
      "step": 132990
    },
    {
      "epoch": 19.808608876973487,
      "grad_norm": 7.5625,
      "learning_rate": 1.3968300358138675e-08,
      "loss": 1.1687,
      "num_input_tokens_seen": 77160616,
      "step": 132995
    },
    {
      "epoch": 19.809353589514448,
      "grad_norm": 14.125,
      "learning_rate": 1.3859903332383296e-08,
      "loss": 1.1099,
      "num_input_tokens_seen": 77163528,
      "step": 133000
    },
    {
      "epoch": 19.81009830205541,
      "grad_norm": 18.875,
      "learning_rate": 1.3751928422070515e-08,
      "loss": 1.0956,
      "num_input_tokens_seen": 77166184,
      "step": 133005
    },
    {
      "epoch": 19.810843014596365,
      "grad_norm": 11.875,
      "learning_rate": 1.3644375629023875e-08,
      "loss": 0.8759,
      "num_input_tokens_seen": 77168968,
      "step": 133010
    },
    {
      "epoch": 19.811587727137326,
      "grad_norm": 9.875,
      "learning_rate": 1.3537244955061368e-08,
      "loss": 0.907,
      "num_input_tokens_seen": 77172104,
      "step": 133015
    },
    {
      "epoch": 19.812332439678283,
      "grad_norm": 11.0625,
      "learning_rate": 1.343053640198988e-08,
      "loss": 1.1248,
      "num_input_tokens_seen": 77174824,
      "step": 133020
    },
    {
      "epoch": 19.813077152219243,
      "grad_norm": 6.5625,
      "learning_rate": 1.3324249971613523e-08,
      "loss": 0.8155,
      "num_input_tokens_seen": 77177512,
      "step": 133025
    },
    {
      "epoch": 19.813821864760204,
      "grad_norm": 11.9375,
      "learning_rate": 1.3218385665730859e-08,
      "loss": 1.1016,
      "num_input_tokens_seen": 77180520,
      "step": 133030
    },
    {
      "epoch": 19.81456657730116,
      "grad_norm": 12.875,
      "learning_rate": 1.3112943486129347e-08,
      "loss": 1.0566,
      "num_input_tokens_seen": 77183400,
      "step": 133035
    },
    {
      "epoch": 19.81531128984212,
      "grad_norm": 7.0,
      "learning_rate": 1.3007923434585345e-08,
      "loss": 0.9244,
      "num_input_tokens_seen": 77186664,
      "step": 133040
    },
    {
      "epoch": 19.816056002383082,
      "grad_norm": 22.625,
      "learning_rate": 1.290332551288076e-08,
      "loss": 1.4205,
      "num_input_tokens_seen": 77189832,
      "step": 133045
    },
    {
      "epoch": 19.81680071492404,
      "grad_norm": 13.5625,
      "learning_rate": 1.2799149722775294e-08,
      "loss": 1.1546,
      "num_input_tokens_seen": 77192712,
      "step": 133050
    },
    {
      "epoch": 19.817545427465,
      "grad_norm": 6.46875,
      "learning_rate": 1.2695396066034205e-08,
      "loss": 0.7413,
      "num_input_tokens_seen": 77195464,
      "step": 133055
    },
    {
      "epoch": 19.818290140005956,
      "grad_norm": 15.0625,
      "learning_rate": 1.2592064544408866e-08,
      "loss": 0.6859,
      "num_input_tokens_seen": 77198280,
      "step": 133060
    },
    {
      "epoch": 19.819034852546917,
      "grad_norm": 8.375,
      "learning_rate": 1.248915515964233e-08,
      "loss": 1.0116,
      "num_input_tokens_seen": 77201416,
      "step": 133065
    },
    {
      "epoch": 19.819779565087877,
      "grad_norm": 8.6875,
      "learning_rate": 1.2386667913477645e-08,
      "loss": 1.046,
      "num_input_tokens_seen": 77204424,
      "step": 133070
    },
    {
      "epoch": 19.820524277628834,
      "grad_norm": 11.875,
      "learning_rate": 1.228460280764121e-08,
      "loss": 1.3391,
      "num_input_tokens_seen": 77207240,
      "step": 133075
    },
    {
      "epoch": 19.821268990169795,
      "grad_norm": 9.4375,
      "learning_rate": 1.2182959843862196e-08,
      "loss": 1.2553,
      "num_input_tokens_seen": 77210056,
      "step": 133080
    },
    {
      "epoch": 19.822013702710755,
      "grad_norm": 12.9375,
      "learning_rate": 1.2081739023855899e-08,
      "loss": 1.0759,
      "num_input_tokens_seen": 77212712,
      "step": 133085
    },
    {
      "epoch": 19.822758415251712,
      "grad_norm": 18.25,
      "learning_rate": 1.198094034933206e-08,
      "loss": 1.2632,
      "num_input_tokens_seen": 77215400,
      "step": 133090
    },
    {
      "epoch": 19.823503127792673,
      "grad_norm": 18.125,
      "learning_rate": 1.1880563821992096e-08,
      "loss": 1.078,
      "num_input_tokens_seen": 77218344,
      "step": 133095
    },
    {
      "epoch": 19.82424784033363,
      "grad_norm": 7.125,
      "learning_rate": 1.1780609443534652e-08,
      "loss": 1.1998,
      "num_input_tokens_seen": 77220840,
      "step": 133100
    },
    {
      "epoch": 19.82499255287459,
      "grad_norm": 16.25,
      "learning_rate": 1.1681077215644487e-08,
      "loss": 1.329,
      "num_input_tokens_seen": 77223560,
      "step": 133105
    },
    {
      "epoch": 19.82573726541555,
      "grad_norm": 10.5625,
      "learning_rate": 1.1581967140009142e-08,
      "loss": 1.0376,
      "num_input_tokens_seen": 77226152,
      "step": 133110
    },
    {
      "epoch": 19.826481977956508,
      "grad_norm": 10.375,
      "learning_rate": 1.1483279218296728e-08,
      "loss": 0.7812,
      "num_input_tokens_seen": 77229000,
      "step": 133115
    },
    {
      "epoch": 19.82722669049747,
      "grad_norm": 4.53125,
      "learning_rate": 1.1385013452178128e-08,
      "loss": 1.0362,
      "num_input_tokens_seen": 77231880,
      "step": 133120
    },
    {
      "epoch": 19.827971403038426,
      "grad_norm": 20.125,
      "learning_rate": 1.1287169843313127e-08,
      "loss": 1.2086,
      "num_input_tokens_seen": 77234856,
      "step": 133125
    },
    {
      "epoch": 19.828716115579386,
      "grad_norm": 9.625,
      "learning_rate": 1.1189748393353184e-08,
      "loss": 1.0457,
      "num_input_tokens_seen": 77238120,
      "step": 133130
    },
    {
      "epoch": 19.829460828120347,
      "grad_norm": 15.25,
      "learning_rate": 1.10927491039442e-08,
      "loss": 0.8824,
      "num_input_tokens_seen": 77240872,
      "step": 133135
    },
    {
      "epoch": 19.830205540661304,
      "grad_norm": 15.9375,
      "learning_rate": 1.0996171976726532e-08,
      "loss": 0.8953,
      "num_input_tokens_seen": 77243720,
      "step": 133140
    },
    {
      "epoch": 19.830950253202264,
      "grad_norm": 18.625,
      "learning_rate": 1.0900017013329434e-08,
      "loss": 0.8333,
      "num_input_tokens_seen": 77248008,
      "step": 133145
    },
    {
      "epoch": 19.831694965743225,
      "grad_norm": 13.75,
      "learning_rate": 1.0804284215379379e-08,
      "loss": 1.0891,
      "num_input_tokens_seen": 77250600,
      "step": 133150
    },
    {
      "epoch": 19.83243967828418,
      "grad_norm": 12.75,
      "learning_rate": 1.070897358449452e-08,
      "loss": 1.1447,
      "num_input_tokens_seen": 77253448,
      "step": 133155
    },
    {
      "epoch": 19.833184390825142,
      "grad_norm": 9.125,
      "learning_rate": 1.0614085122281902e-08,
      "loss": 1.2016,
      "num_input_tokens_seen": 77256488,
      "step": 133160
    },
    {
      "epoch": 19.8339291033661,
      "grad_norm": 6.375,
      "learning_rate": 1.0519618830348577e-08,
      "loss": 1.2255,
      "num_input_tokens_seen": 77259368,
      "step": 133165
    },
    {
      "epoch": 19.83467381590706,
      "grad_norm": 14.5625,
      "learning_rate": 1.0425574710284936e-08,
      "loss": 0.8306,
      "num_input_tokens_seen": 77262120,
      "step": 133170
    },
    {
      "epoch": 19.83541852844802,
      "grad_norm": 10.8125,
      "learning_rate": 1.033195276368415e-08,
      "loss": 0.837,
      "num_input_tokens_seen": 77264552,
      "step": 133175
    },
    {
      "epoch": 19.836163240988977,
      "grad_norm": 6.5625,
      "learning_rate": 1.0238752992128287e-08,
      "loss": 0.814,
      "num_input_tokens_seen": 77267400,
      "step": 133180
    },
    {
      "epoch": 19.836907953529938,
      "grad_norm": 7.65625,
      "learning_rate": 1.0145975397188311e-08,
      "loss": 0.911,
      "num_input_tokens_seen": 77270280,
      "step": 133185
    },
    {
      "epoch": 19.8376526660709,
      "grad_norm": 10.125,
      "learning_rate": 1.0053619980435191e-08,
      "loss": 1.0728,
      "num_input_tokens_seen": 77273224,
      "step": 133190
    },
    {
      "epoch": 19.838397378611855,
      "grad_norm": 17.125,
      "learning_rate": 9.961686743426012e-09,
      "loss": 1.3187,
      "num_input_tokens_seen": 77276104,
      "step": 133195
    },
    {
      "epoch": 19.839142091152816,
      "grad_norm": 11.125,
      "learning_rate": 9.870175687715089e-09,
      "loss": 0.9749,
      "num_input_tokens_seen": 77278632,
      "step": 133200
    },
    {
      "epoch": 19.839886803693773,
      "grad_norm": 11.0625,
      "learning_rate": 9.77908681485118e-09,
      "loss": 0.998,
      "num_input_tokens_seen": 77281320,
      "step": 133205
    },
    {
      "epoch": 19.840631516234733,
      "grad_norm": 6.65625,
      "learning_rate": 9.688420126369168e-09,
      "loss": 0.9084,
      "num_input_tokens_seen": 77284040,
      "step": 133210
    },
    {
      "epoch": 19.841376228775694,
      "grad_norm": 13.375,
      "learning_rate": 9.598175623801165e-09,
      "loss": 1.2457,
      "num_input_tokens_seen": 77286760,
      "step": 133215
    },
    {
      "epoch": 19.84212094131665,
      "grad_norm": 9.0625,
      "learning_rate": 9.508353308673723e-09,
      "loss": 1.0928,
      "num_input_tokens_seen": 77289576,
      "step": 133220
    },
    {
      "epoch": 19.84286565385761,
      "grad_norm": 16.0,
      "learning_rate": 9.4189531825023e-09,
      "loss": 1.1419,
      "num_input_tokens_seen": 77292776,
      "step": 133225
    },
    {
      "epoch": 19.843610366398572,
      "grad_norm": 11.5,
      "learning_rate": 9.329975246799571e-09,
      "loss": 1.1526,
      "num_input_tokens_seen": 77295496,
      "step": 133230
    },
    {
      "epoch": 19.84435507893953,
      "grad_norm": 13.3125,
      "learning_rate": 9.241419503069892e-09,
      "loss": 1.2145,
      "num_input_tokens_seen": 77298472,
      "step": 133235
    },
    {
      "epoch": 19.84509979148049,
      "grad_norm": 12.875,
      "learning_rate": 9.153285952803736e-09,
      "loss": 1.0259,
      "num_input_tokens_seen": 77301480,
      "step": 133240
    },
    {
      "epoch": 19.845844504021446,
      "grad_norm": 9.8125,
      "learning_rate": 9.065574597494352e-09,
      "loss": 0.9449,
      "num_input_tokens_seen": 77304200,
      "step": 133245
    },
    {
      "epoch": 19.846589216562407,
      "grad_norm": 20.625,
      "learning_rate": 8.978285438621115e-09,
      "loss": 1.2037,
      "num_input_tokens_seen": 77306888,
      "step": 133250
    },
    {
      "epoch": 19.847333929103367,
      "grad_norm": 10.5625,
      "learning_rate": 8.891418477660617e-09,
      "loss": 1.1092,
      "num_input_tokens_seen": 77309480,
      "step": 133255
    },
    {
      "epoch": 19.848078641644324,
      "grad_norm": 15.8125,
      "learning_rate": 8.804973716081132e-09,
      "loss": 0.9912,
      "num_input_tokens_seen": 77312328,
      "step": 133260
    },
    {
      "epoch": 19.848823354185285,
      "grad_norm": 12.375,
      "learning_rate": 8.718951155339827e-09,
      "loss": 1.2185,
      "num_input_tokens_seen": 77315656,
      "step": 133265
    },
    {
      "epoch": 19.849568066726242,
      "grad_norm": 7.1875,
      "learning_rate": 8.633350796893869e-09,
      "loss": 1.0689,
      "num_input_tokens_seen": 77318600,
      "step": 133270
    },
    {
      "epoch": 19.850312779267203,
      "grad_norm": 4.625,
      "learning_rate": 8.548172642186547e-09,
      "loss": 1.0216,
      "num_input_tokens_seen": 77321288,
      "step": 133275
    },
    {
      "epoch": 19.851057491808163,
      "grad_norm": 12.625,
      "learning_rate": 8.463416692658377e-09,
      "loss": 1.3895,
      "num_input_tokens_seen": 77324200,
      "step": 133280
    },
    {
      "epoch": 19.85180220434912,
      "grad_norm": 6.59375,
      "learning_rate": 8.37908294973877e-09,
      "loss": 0.9414,
      "num_input_tokens_seen": 77327496,
      "step": 133285
    },
    {
      "epoch": 19.85254691689008,
      "grad_norm": 15.8125,
      "learning_rate": 8.295171414854363e-09,
      "loss": 1.2589,
      "num_input_tokens_seen": 77330248,
      "step": 133290
    },
    {
      "epoch": 19.85329162943104,
      "grad_norm": 6.0,
      "learning_rate": 8.211682089423467e-09,
      "loss": 1.0567,
      "num_input_tokens_seen": 77333000,
      "step": 133295
    },
    {
      "epoch": 19.854036341971998,
      "grad_norm": 10.5625,
      "learning_rate": 8.128614974856064e-09,
      "loss": 1.2151,
      "num_input_tokens_seen": 77335944,
      "step": 133300
    },
    {
      "epoch": 19.85478105451296,
      "grad_norm": 12.625,
      "learning_rate": 8.04597007255381e-09,
      "loss": 1.0235,
      "num_input_tokens_seen": 77338632,
      "step": 133305
    },
    {
      "epoch": 19.855525767053916,
      "grad_norm": 11.375,
      "learning_rate": 7.963747383915587e-09,
      "loss": 0.9876,
      "num_input_tokens_seen": 77341480,
      "step": 133310
    },
    {
      "epoch": 19.856270479594876,
      "grad_norm": 5.09375,
      "learning_rate": 7.881946910329175e-09,
      "loss": 1.2392,
      "num_input_tokens_seen": 77344328,
      "step": 133315
    },
    {
      "epoch": 19.857015192135837,
      "grad_norm": 9.5,
      "learning_rate": 7.800568653174023e-09,
      "loss": 1.1978,
      "num_input_tokens_seen": 77347432,
      "step": 133320
    },
    {
      "epoch": 19.857759904676794,
      "grad_norm": 14.875,
      "learning_rate": 7.719612613829586e-09,
      "loss": 1.348,
      "num_input_tokens_seen": 77350312,
      "step": 133325
    },
    {
      "epoch": 19.858504617217754,
      "grad_norm": 9.0625,
      "learning_rate": 7.639078793661436e-09,
      "loss": 1.1099,
      "num_input_tokens_seen": 77353352,
      "step": 133330
    },
    {
      "epoch": 19.859249329758715,
      "grad_norm": 12.1875,
      "learning_rate": 7.558967194029598e-09,
      "loss": 1.3435,
      "num_input_tokens_seen": 77356104,
      "step": 133335
    },
    {
      "epoch": 19.85999404229967,
      "grad_norm": 17.5,
      "learning_rate": 7.479277816285768e-09,
      "loss": 1.9636,
      "num_input_tokens_seen": 77358888,
      "step": 133340
    },
    {
      "epoch": 19.860738754840632,
      "grad_norm": 15.375,
      "learning_rate": 7.400010661781642e-09,
      "loss": 0.9468,
      "num_input_tokens_seen": 77361640,
      "step": 133345
    },
    {
      "epoch": 19.86148346738159,
      "grad_norm": 5.46875,
      "learning_rate": 7.321165731849488e-09,
      "loss": 0.9261,
      "num_input_tokens_seen": 77364616,
      "step": 133350
    },
    {
      "epoch": 19.86222817992255,
      "grad_norm": 13.3125,
      "learning_rate": 7.242743027827126e-09,
      "loss": 1.175,
      "num_input_tokens_seen": 77367272,
      "step": 133355
    },
    {
      "epoch": 19.86297289246351,
      "grad_norm": 11.6875,
      "learning_rate": 7.1647425510384944e-09,
      "loss": 0.9266,
      "num_input_tokens_seen": 77370152,
      "step": 133360
    },
    {
      "epoch": 19.863717605004467,
      "grad_norm": 10.125,
      "learning_rate": 7.087164302796434e-09,
      "loss": 1.0109,
      "num_input_tokens_seen": 77373064,
      "step": 133365
    },
    {
      "epoch": 19.864462317545428,
      "grad_norm": 12.0,
      "learning_rate": 7.010008284416558e-09,
      "loss": 1.2359,
      "num_input_tokens_seen": 77375880,
      "step": 133370
    },
    {
      "epoch": 19.86520703008639,
      "grad_norm": 7.75,
      "learning_rate": 6.933274497200604e-09,
      "loss": 0.7761,
      "num_input_tokens_seen": 77378888,
      "step": 133375
    },
    {
      "epoch": 19.865951742627345,
      "grad_norm": 8.25,
      "learning_rate": 6.856962942447531e-09,
      "loss": 0.672,
      "num_input_tokens_seen": 77381896,
      "step": 133380
    },
    {
      "epoch": 19.866696455168306,
      "grad_norm": 6.5,
      "learning_rate": 6.781073621442424e-09,
      "loss": 0.9661,
      "num_input_tokens_seen": 77384680,
      "step": 133385
    },
    {
      "epoch": 19.867441167709263,
      "grad_norm": 12.6875,
      "learning_rate": 6.70560653546759e-09,
      "loss": 1.2118,
      "num_input_tokens_seen": 77387912,
      "step": 133390
    },
    {
      "epoch": 19.868185880250223,
      "grad_norm": 6.75,
      "learning_rate": 6.6305616857997855e-09,
      "loss": 0.8107,
      "num_input_tokens_seen": 77390824,
      "step": 133395
    },
    {
      "epoch": 19.868930592791184,
      "grad_norm": 12.5,
      "learning_rate": 6.55593907370744e-09,
      "loss": 1.2215,
      "num_input_tokens_seen": 77393544,
      "step": 133400
    },
    {
      "epoch": 19.86967530533214,
      "grad_norm": 6.125,
      "learning_rate": 6.481738700450657e-09,
      "loss": 0.9328,
      "num_input_tokens_seen": 77396520,
      "step": 133405
    },
    {
      "epoch": 19.8704200178731,
      "grad_norm": 9.375,
      "learning_rate": 6.4079605672839886e-09,
      "loss": 0.957,
      "num_input_tokens_seen": 77399560,
      "step": 133410
    },
    {
      "epoch": 19.87116473041406,
      "grad_norm": 13.6875,
      "learning_rate": 6.334604675450884e-09,
      "loss": 1.4095,
      "num_input_tokens_seen": 77402216,
      "step": 133415
    },
    {
      "epoch": 19.87190944295502,
      "grad_norm": 14.6875,
      "learning_rate": 6.2616710261920176e-09,
      "loss": 1.2901,
      "num_input_tokens_seen": 77405192,
      "step": 133420
    },
    {
      "epoch": 19.87265415549598,
      "grad_norm": 10.9375,
      "learning_rate": 6.189159620739737e-09,
      "loss": 1.0382,
      "num_input_tokens_seen": 77407944,
      "step": 133425
    },
    {
      "epoch": 19.873398868036936,
      "grad_norm": 13.4375,
      "learning_rate": 6.117070460318064e-09,
      "loss": 0.9866,
      "num_input_tokens_seen": 77410824,
      "step": 133430
    },
    {
      "epoch": 19.874143580577897,
      "grad_norm": 10.375,
      "learning_rate": 6.045403546148243e-09,
      "loss": 1.2355,
      "num_input_tokens_seen": 77413896,
      "step": 133435
    },
    {
      "epoch": 19.874888293118858,
      "grad_norm": 13.6875,
      "learning_rate": 5.974158879434866e-09,
      "loss": 1.2639,
      "num_input_tokens_seen": 77416904,
      "step": 133440
    },
    {
      "epoch": 19.875633005659815,
      "grad_norm": 16.875,
      "learning_rate": 5.903336461388076e-09,
      "loss": 1.235,
      "num_input_tokens_seen": 77419720,
      "step": 133445
    },
    {
      "epoch": 19.876377718200775,
      "grad_norm": 17.125,
      "learning_rate": 5.832936293201363e-09,
      "loss": 1.4297,
      "num_input_tokens_seen": 77422792,
      "step": 133450
    },
    {
      "epoch": 19.877122430741732,
      "grad_norm": 5.5625,
      "learning_rate": 5.762958376065441e-09,
      "loss": 0.8432,
      "num_input_tokens_seen": 77425480,
      "step": 133455
    },
    {
      "epoch": 19.877867143282693,
      "grad_norm": 4.09375,
      "learning_rate": 5.693402711159923e-09,
      "loss": 0.984,
      "num_input_tokens_seen": 77428840,
      "step": 133460
    },
    {
      "epoch": 19.878611855823653,
      "grad_norm": 10.125,
      "learning_rate": 5.6242692996616445e-09,
      "loss": 1.019,
      "num_input_tokens_seen": 77431784,
      "step": 133465
    },
    {
      "epoch": 19.87935656836461,
      "grad_norm": 14.25,
      "learning_rate": 5.555558142736339e-09,
      "loss": 1.2587,
      "num_input_tokens_seen": 77434536,
      "step": 133470
    },
    {
      "epoch": 19.88010128090557,
      "grad_norm": 11.875,
      "learning_rate": 5.487269241549742e-09,
      "loss": 1.118,
      "num_input_tokens_seen": 77437288,
      "step": 133475
    },
    {
      "epoch": 19.88084599344653,
      "grad_norm": 14.6875,
      "learning_rate": 5.419402597250934e-09,
      "loss": 1.1513,
      "num_input_tokens_seen": 77439976,
      "step": 133480
    },
    {
      "epoch": 19.881590705987488,
      "grad_norm": 11.9375,
      "learning_rate": 5.351958210986219e-09,
      "loss": 1.2288,
      "num_input_tokens_seen": 77442760,
      "step": 133485
    },
    {
      "epoch": 19.88233541852845,
      "grad_norm": 11.8125,
      "learning_rate": 5.284936083899128e-09,
      "loss": 1.1419,
      "num_input_tokens_seen": 77445512,
      "step": 133490
    },
    {
      "epoch": 19.883080131069406,
      "grad_norm": 10.125,
      "learning_rate": 5.218336217119313e-09,
      "loss": 1.2063,
      "num_input_tokens_seen": 77448392,
      "step": 133495
    },
    {
      "epoch": 19.883824843610366,
      "grad_norm": 12.75,
      "learning_rate": 5.152158611770874e-09,
      "loss": 1.1632,
      "num_input_tokens_seen": 77451304,
      "step": 133500
    },
    {
      "epoch": 19.884569556151327,
      "grad_norm": 12.9375,
      "learning_rate": 5.086403268975137e-09,
      "loss": 1.0235,
      "num_input_tokens_seen": 77454376,
      "step": 133505
    },
    {
      "epoch": 19.885314268692284,
      "grad_norm": 13.625,
      "learning_rate": 5.0210701898395494e-09,
      "loss": 1.2788,
      "num_input_tokens_seen": 77457160,
      "step": 133510
    },
    {
      "epoch": 19.886058981233244,
      "grad_norm": 20.25,
      "learning_rate": 4.956159375468783e-09,
      "loss": 1.3363,
      "num_input_tokens_seen": 77459816,
      "step": 133515
    },
    {
      "epoch": 19.886803693774205,
      "grad_norm": 14.6875,
      "learning_rate": 4.891670826959183e-09,
      "loss": 1.1848,
      "num_input_tokens_seen": 77462920,
      "step": 133520
    },
    {
      "epoch": 19.88754840631516,
      "grad_norm": 11.5625,
      "learning_rate": 4.8276045454043185e-09,
      "loss": 1.3558,
      "num_input_tokens_seen": 77465960,
      "step": 133525
    },
    {
      "epoch": 19.888293118856122,
      "grad_norm": 7.90625,
      "learning_rate": 4.763960531878331e-09,
      "loss": 1.1887,
      "num_input_tokens_seen": 77468424,
      "step": 133530
    },
    {
      "epoch": 19.88903783139708,
      "grad_norm": 16.5,
      "learning_rate": 4.700738787466463e-09,
      "loss": 1.1835,
      "num_input_tokens_seen": 77471432,
      "step": 133535
    },
    {
      "epoch": 19.88978254393804,
      "grad_norm": 10.5,
      "learning_rate": 4.637939313226203e-09,
      "loss": 1.0308,
      "num_input_tokens_seen": 77474024,
      "step": 133540
    },
    {
      "epoch": 19.890527256479,
      "grad_norm": 16.25,
      "learning_rate": 4.575562110228915e-09,
      "loss": 1.1134,
      "num_input_tokens_seen": 77476808,
      "step": 133545
    },
    {
      "epoch": 19.891271969019957,
      "grad_norm": 19.625,
      "learning_rate": 4.513607179520985e-09,
      "loss": 1.4294,
      "num_input_tokens_seen": 77479912,
      "step": 133550
    },
    {
      "epoch": 19.892016681560918,
      "grad_norm": 7.28125,
      "learning_rate": 4.452074522148797e-09,
      "loss": 1.0228,
      "num_input_tokens_seen": 77482568,
      "step": 133555
    },
    {
      "epoch": 19.89276139410188,
      "grad_norm": 10.125,
      "learning_rate": 4.390964139158737e-09,
      "loss": 1.1544,
      "num_input_tokens_seen": 77485544,
      "step": 133560
    },
    {
      "epoch": 19.893506106642835,
      "grad_norm": 13.625,
      "learning_rate": 4.330276031577762e-09,
      "loss": 0.9784,
      "num_input_tokens_seen": 77488456,
      "step": 133565
    },
    {
      "epoch": 19.894250819183796,
      "grad_norm": 6.125,
      "learning_rate": 4.270010200430052e-09,
      "loss": 1.0686,
      "num_input_tokens_seen": 77491688,
      "step": 133570
    },
    {
      "epoch": 19.894995531724753,
      "grad_norm": 9.0,
      "learning_rate": 4.210166646737013e-09,
      "loss": 1.3547,
      "num_input_tokens_seen": 77494600,
      "step": 133575
    },
    {
      "epoch": 19.895740244265713,
      "grad_norm": 18.5,
      "learning_rate": 4.150745371508946e-09,
      "loss": 1.102,
      "num_input_tokens_seen": 77497192,
      "step": 133580
    },
    {
      "epoch": 19.896484956806674,
      "grad_norm": 12.5625,
      "learning_rate": 4.0917463757506045e-09,
      "loss": 1.021,
      "num_input_tokens_seen": 77499976,
      "step": 133585
    },
    {
      "epoch": 19.89722966934763,
      "grad_norm": 9.0,
      "learning_rate": 4.033169660458413e-09,
      "loss": 1.0937,
      "num_input_tokens_seen": 77502696,
      "step": 133590
    },
    {
      "epoch": 19.89797438188859,
      "grad_norm": 12.0625,
      "learning_rate": 3.975015226617695e-09,
      "loss": 0.9521,
      "num_input_tokens_seen": 77505704,
      "step": 133595
    },
    {
      "epoch": 19.898719094429552,
      "grad_norm": 6.21875,
      "learning_rate": 3.917283075216549e-09,
      "loss": 0.8055,
      "num_input_tokens_seen": 77508520,
      "step": 133600
    },
    {
      "epoch": 19.89946380697051,
      "grad_norm": 14.625,
      "learning_rate": 3.8599732072264195e-09,
      "loss": 0.7366,
      "num_input_tokens_seen": 77511496,
      "step": 133605
    },
    {
      "epoch": 19.90020851951147,
      "grad_norm": 12.25,
      "learning_rate": 3.803085623618752e-09,
      "loss": 0.8967,
      "num_input_tokens_seen": 77514312,
      "step": 133610
    },
    {
      "epoch": 19.900953232052427,
      "grad_norm": 14.9375,
      "learning_rate": 3.746620325351113e-09,
      "loss": 0.7822,
      "num_input_tokens_seen": 77517320,
      "step": 133615
    },
    {
      "epoch": 19.901697944593387,
      "grad_norm": 10.9375,
      "learning_rate": 3.690577313381072e-09,
      "loss": 1.3094,
      "num_input_tokens_seen": 77520392,
      "step": 133620
    },
    {
      "epoch": 19.902442657134348,
      "grad_norm": 16.25,
      "learning_rate": 3.6349565886523163e-09,
      "loss": 1.3246,
      "num_input_tokens_seen": 77523016,
      "step": 133625
    },
    {
      "epoch": 19.903187369675305,
      "grad_norm": 5.5625,
      "learning_rate": 3.579758152105761e-09,
      "loss": 0.9726,
      "num_input_tokens_seen": 77526024,
      "step": 133630
    },
    {
      "epoch": 19.903932082216265,
      "grad_norm": 7.90625,
      "learning_rate": 3.524982004676769e-09,
      "loss": 1.0574,
      "num_input_tokens_seen": 77529160,
      "step": 133635
    },
    {
      "epoch": 19.904676794757222,
      "grad_norm": 16.25,
      "learning_rate": 3.4706281472840495e-09,
      "loss": 1.5609,
      "num_input_tokens_seen": 77532232,
      "step": 133640
    },
    {
      "epoch": 19.905421507298183,
      "grad_norm": 17.625,
      "learning_rate": 3.4166965808518637e-09,
      "loss": 0.9751,
      "num_input_tokens_seen": 77535304,
      "step": 133645
    },
    {
      "epoch": 19.906166219839143,
      "grad_norm": 9.125,
      "learning_rate": 3.363187306287818e-09,
      "loss": 1.6894,
      "num_input_tokens_seen": 77538280,
      "step": 133650
    },
    {
      "epoch": 19.9069109323801,
      "grad_norm": 9.0625,
      "learning_rate": 3.310100324499521e-09,
      "loss": 1.0308,
      "num_input_tokens_seen": 77540936,
      "step": 133655
    },
    {
      "epoch": 19.90765564492106,
      "grad_norm": 10.5,
      "learning_rate": 3.2574356363807013e-09,
      "loss": 1.1325,
      "num_input_tokens_seen": 77543592,
      "step": 133660
    },
    {
      "epoch": 19.90840035746202,
      "grad_norm": 11.25,
      "learning_rate": 3.2051932428195375e-09,
      "loss": 1.1861,
      "num_input_tokens_seen": 77546344,
      "step": 133665
    },
    {
      "epoch": 19.909145070002978,
      "grad_norm": 11.625,
      "learning_rate": 3.153373144704208e-09,
      "loss": 0.8408,
      "num_input_tokens_seen": 77549000,
      "step": 133670
    },
    {
      "epoch": 19.90988978254394,
      "grad_norm": 11.75,
      "learning_rate": 3.1019753429062383e-09,
      "loss": 1.1945,
      "num_input_tokens_seen": 77551880,
      "step": 133675
    },
    {
      "epoch": 19.910634495084896,
      "grad_norm": 10.1875,
      "learning_rate": 3.050999838294377e-09,
      "loss": 1.2965,
      "num_input_tokens_seen": 77554696,
      "step": 133680
    },
    {
      "epoch": 19.911379207625856,
      "grad_norm": 18.75,
      "learning_rate": 3.000446631729048e-09,
      "loss": 0.8872,
      "num_input_tokens_seen": 77557704,
      "step": 133685
    },
    {
      "epoch": 19.912123920166817,
      "grad_norm": 9.875,
      "learning_rate": 2.9503157240651226e-09,
      "loss": 0.8766,
      "num_input_tokens_seen": 77560648,
      "step": 133690
    },
    {
      "epoch": 19.912868632707774,
      "grad_norm": 14.75,
      "learning_rate": 2.900607116151921e-09,
      "loss": 0.8815,
      "num_input_tokens_seen": 77563688,
      "step": 133695
    },
    {
      "epoch": 19.913613345248734,
      "grad_norm": 4.46875,
      "learning_rate": 2.8513208088248867e-09,
      "loss": 1.1396,
      "num_input_tokens_seen": 77566600,
      "step": 133700
    },
    {
      "epoch": 19.914358057789695,
      "grad_norm": 25.5,
      "learning_rate": 2.802456802919462e-09,
      "loss": 1.1323,
      "num_input_tokens_seen": 77569224,
      "step": 133705
    },
    {
      "epoch": 19.915102770330652,
      "grad_norm": 11.5,
      "learning_rate": 2.7540150992627633e-09,
      "loss": 1.0865,
      "num_input_tokens_seen": 77572008,
      "step": 133710
    },
    {
      "epoch": 19.915847482871612,
      "grad_norm": 11.625,
      "learning_rate": 2.705995698668029e-09,
      "loss": 1.2465,
      "num_input_tokens_seen": 77574792,
      "step": 133715
    },
    {
      "epoch": 19.91659219541257,
      "grad_norm": 10.9375,
      "learning_rate": 2.658398601951273e-09,
      "loss": 1.0402,
      "num_input_tokens_seen": 77577704,
      "step": 133720
    },
    {
      "epoch": 19.91733690795353,
      "grad_norm": 7.96875,
      "learning_rate": 2.6112238099146315e-09,
      "loss": 1.2612,
      "num_input_tokens_seen": 77580200,
      "step": 133725
    },
    {
      "epoch": 19.91808162049449,
      "grad_norm": 10.1875,
      "learning_rate": 2.564471323354689e-09,
      "loss": 1.1812,
      "num_input_tokens_seen": 77582984,
      "step": 133730
    },
    {
      "epoch": 19.918826333035447,
      "grad_norm": 6.5,
      "learning_rate": 2.5181411430597045e-09,
      "loss": 1.1356,
      "num_input_tokens_seen": 77585992,
      "step": 133735
    },
    {
      "epoch": 19.919571045576408,
      "grad_norm": 14.375,
      "learning_rate": 2.472233269817936e-09,
      "loss": 1.2987,
      "num_input_tokens_seen": 77589000,
      "step": 133740
    },
    {
      "epoch": 19.92031575811737,
      "grad_norm": 7.9375,
      "learning_rate": 2.4267477043982134e-09,
      "loss": 0.8899,
      "num_input_tokens_seen": 77591976,
      "step": 133745
    },
    {
      "epoch": 19.921060470658325,
      "grad_norm": 8.5625,
      "learning_rate": 2.3816844475749167e-09,
      "loss": 1.2266,
      "num_input_tokens_seen": 77594824,
      "step": 133750
    },
    {
      "epoch": 19.921805183199286,
      "grad_norm": 7.03125,
      "learning_rate": 2.337043500102998e-09,
      "loss": 1.0282,
      "num_input_tokens_seen": 77597992,
      "step": 133755
    },
    {
      "epoch": 19.922549895740243,
      "grad_norm": 9.0625,
      "learning_rate": 2.2928248627429595e-09,
      "loss": 1.1513,
      "num_input_tokens_seen": 77600712,
      "step": 133760
    },
    {
      "epoch": 19.923294608281203,
      "grad_norm": 16.125,
      "learning_rate": 2.249028536238651e-09,
      "loss": 1.2465,
      "num_input_tokens_seen": 77603720,
      "step": 133765
    },
    {
      "epoch": 19.924039320822164,
      "grad_norm": 13.75,
      "learning_rate": 2.205654521331146e-09,
      "loss": 1.5052,
      "num_input_tokens_seen": 77606408,
      "step": 133770
    },
    {
      "epoch": 19.92478403336312,
      "grad_norm": 16.5,
      "learning_rate": 2.162702818753193e-09,
      "loss": 1.1893,
      "num_input_tokens_seen": 77608968,
      "step": 133775
    },
    {
      "epoch": 19.92552874590408,
      "grad_norm": 12.375,
      "learning_rate": 2.120173429226435e-09,
      "loss": 1.0279,
      "num_input_tokens_seen": 77611624,
      "step": 133780
    },
    {
      "epoch": 19.92627345844504,
      "grad_norm": 10.3125,
      "learning_rate": 2.0780663534752944e-09,
      "loss": 0.8569,
      "num_input_tokens_seen": 77615080,
      "step": 133785
    },
    {
      "epoch": 19.927018170986,
      "grad_norm": 9.9375,
      "learning_rate": 2.036381592207537e-09,
      "loss": 1.0484,
      "num_input_tokens_seen": 77618120,
      "step": 133790
    },
    {
      "epoch": 19.92776288352696,
      "grad_norm": 8.0625,
      "learning_rate": 1.9951191461281547e-09,
      "loss": 1.2358,
      "num_input_tokens_seen": 77621096,
      "step": 133795
    },
    {
      "epoch": 19.928507596067917,
      "grad_norm": 6.9375,
      "learning_rate": 1.9542790159365887e-09,
      "loss": 1.133,
      "num_input_tokens_seen": 77623848,
      "step": 133800
    },
    {
      "epoch": 19.929252308608877,
      "grad_norm": 10.0625,
      "learning_rate": 1.913861202318401e-09,
      "loss": 1.0892,
      "num_input_tokens_seen": 77626696,
      "step": 133805
    },
    {
      "epoch": 19.929997021149838,
      "grad_norm": 14.25,
      "learning_rate": 1.873865705959155e-09,
      "loss": 1.367,
      "num_input_tokens_seen": 77629704,
      "step": 133810
    },
    {
      "epoch": 19.930741733690795,
      "grad_norm": 9.6875,
      "learning_rate": 1.83429252753331e-09,
      "loss": 0.6918,
      "num_input_tokens_seen": 77632520,
      "step": 133815
    },
    {
      "epoch": 19.931486446231755,
      "grad_norm": 11.625,
      "learning_rate": 1.7951416677097766e-09,
      "loss": 0.8134,
      "num_input_tokens_seen": 77635432,
      "step": 133820
    },
    {
      "epoch": 19.932231158772712,
      "grad_norm": 4.4375,
      "learning_rate": 1.7564131271519123e-09,
      "loss": 0.9577,
      "num_input_tokens_seen": 77638376,
      "step": 133825
    },
    {
      "epoch": 19.932975871313673,
      "grad_norm": 9.5625,
      "learning_rate": 1.7181069065119736e-09,
      "loss": 1.3368,
      "num_input_tokens_seen": 77641448,
      "step": 133830
    },
    {
      "epoch": 19.933720583854633,
      "grad_norm": 5.0625,
      "learning_rate": 1.6802230064366653e-09,
      "loss": 0.7517,
      "num_input_tokens_seen": 77644296,
      "step": 133835
    },
    {
      "epoch": 19.93446529639559,
      "grad_norm": 6.53125,
      "learning_rate": 1.642761427567141e-09,
      "loss": 1.0607,
      "num_input_tokens_seen": 77647176,
      "step": 133840
    },
    {
      "epoch": 19.93521000893655,
      "grad_norm": 7.59375,
      "learning_rate": 1.605722170536228e-09,
      "loss": 1.2234,
      "num_input_tokens_seen": 77650248,
      "step": 133845
    },
    {
      "epoch": 19.93595472147751,
      "grad_norm": 6.34375,
      "learning_rate": 1.5691052359684266e-09,
      "loss": 1.1492,
      "num_input_tokens_seen": 77652872,
      "step": 133850
    },
    {
      "epoch": 19.93669943401847,
      "grad_norm": 20.875,
      "learning_rate": 1.5329106244854618e-09,
      "loss": 0.9146,
      "num_input_tokens_seen": 77655880,
      "step": 133855
    },
    {
      "epoch": 19.93744414655943,
      "grad_norm": 12.125,
      "learning_rate": 1.4971383366951807e-09,
      "loss": 1.0771,
      "num_input_tokens_seen": 77658920,
      "step": 133860
    },
    {
      "epoch": 19.938188859100386,
      "grad_norm": 8.5,
      "learning_rate": 1.4617883732026549e-09,
      "loss": 0.8881,
      "num_input_tokens_seen": 77662120,
      "step": 133865
    },
    {
      "epoch": 19.938933571641346,
      "grad_norm": 9.625,
      "learning_rate": 1.4268607346074048e-09,
      "loss": 1.1152,
      "num_input_tokens_seen": 77664904,
      "step": 133870
    },
    {
      "epoch": 19.939678284182307,
      "grad_norm": 8.8125,
      "learning_rate": 1.3923554214978485e-09,
      "loss": 1.2008,
      "num_input_tokens_seen": 77667560,
      "step": 133875
    },
    {
      "epoch": 19.940422996723264,
      "grad_norm": 13.625,
      "learning_rate": 1.3582724344568532e-09,
      "loss": 1.1305,
      "num_input_tokens_seen": 77670568,
      "step": 133880
    },
    {
      "epoch": 19.941167709264224,
      "grad_norm": 7.0,
      "learning_rate": 1.3246117740589592e-09,
      "loss": 1.1305,
      "num_input_tokens_seen": 77673288,
      "step": 133885
    },
    {
      "epoch": 19.941912421805185,
      "grad_norm": 7.34375,
      "learning_rate": 1.2913734408759314e-09,
      "loss": 1.2134,
      "num_input_tokens_seen": 77676264,
      "step": 133890
    },
    {
      "epoch": 19.942657134346142,
      "grad_norm": 10.625,
      "learning_rate": 1.258557435465657e-09,
      "loss": 1.3439,
      "num_input_tokens_seen": 77679144,
      "step": 133895
    },
    {
      "epoch": 19.943401846887102,
      "grad_norm": 11.0625,
      "learning_rate": 1.226163758386023e-09,
      "loss": 1.2257,
      "num_input_tokens_seen": 77682408,
      "step": 133900
    },
    {
      "epoch": 19.94414655942806,
      "grad_norm": 6.5,
      "learning_rate": 1.1941924101838142e-09,
      "loss": 0.919,
      "num_input_tokens_seen": 77685384,
      "step": 133905
    },
    {
      "epoch": 19.94489127196902,
      "grad_norm": 8.3125,
      "learning_rate": 1.1626433913947132e-09,
      "loss": 1.1846,
      "num_input_tokens_seen": 77688488,
      "step": 133910
    },
    {
      "epoch": 19.94563598450998,
      "grad_norm": 3.640625,
      "learning_rate": 1.1315167025571782e-09,
      "loss": 0.9784,
      "num_input_tokens_seen": 77691144,
      "step": 133915
    },
    {
      "epoch": 19.946380697050937,
      "grad_norm": 8.75,
      "learning_rate": 1.1008123441957896e-09,
      "loss": 1.0064,
      "num_input_tokens_seen": 77694024,
      "step": 133920
    },
    {
      "epoch": 19.947125409591898,
      "grad_norm": 15.5,
      "learning_rate": 1.070530316826801e-09,
      "loss": 1.1326,
      "num_input_tokens_seen": 77697064,
      "step": 133925
    },
    {
      "epoch": 19.947870122132855,
      "grad_norm": 10.25,
      "learning_rate": 1.0406706209636908e-09,
      "loss": 0.9314,
      "num_input_tokens_seen": 77699848,
      "step": 133930
    },
    {
      "epoch": 19.948614834673815,
      "grad_norm": 7.375,
      "learning_rate": 1.01123325711161e-09,
      "loss": 0.9558,
      "num_input_tokens_seen": 77702792,
      "step": 133935
    },
    {
      "epoch": 19.949359547214776,
      "grad_norm": 10.8125,
      "learning_rate": 9.82218225767384e-10,
      "loss": 1.0307,
      "num_input_tokens_seen": 77705608,
      "step": 133940
    },
    {
      "epoch": 19.950104259755733,
      "grad_norm": 13.3125,
      "learning_rate": 9.536255274195105e-10,
      "loss": 0.9638,
      "num_input_tokens_seen": 77708552,
      "step": 133945
    },
    {
      "epoch": 19.950848972296694,
      "grad_norm": 9.0,
      "learning_rate": 9.254551625509367e-10,
      "loss": 0.8743,
      "num_input_tokens_seen": 77711496,
      "step": 133950
    },
    {
      "epoch": 19.951593684837654,
      "grad_norm": 10.6875,
      "learning_rate": 8.977071316418339e-10,
      "loss": 1.2893,
      "num_input_tokens_seen": 77714440,
      "step": 133955
    },
    {
      "epoch": 19.95233839737861,
      "grad_norm": 12.5625,
      "learning_rate": 8.703814351557205e-10,
      "loss": 1.2879,
      "num_input_tokens_seen": 77717320,
      "step": 133960
    },
    {
      "epoch": 19.95308310991957,
      "grad_norm": 8.875,
      "learning_rate": 8.434780735561143e-10,
      "loss": 1.1796,
      "num_input_tokens_seen": 77720520,
      "step": 133965
    },
    {
      "epoch": 19.95382782246053,
      "grad_norm": 5.3125,
      "learning_rate": 8.169970473009825e-10,
      "loss": 1.173,
      "num_input_tokens_seen": 77723784,
      "step": 133970
    },
    {
      "epoch": 19.95457253500149,
      "grad_norm": 9.125,
      "learning_rate": 7.909383568316386e-10,
      "loss": 1.118,
      "num_input_tokens_seen": 77726760,
      "step": 133975
    },
    {
      "epoch": 19.95531724754245,
      "grad_norm": 3.71875,
      "learning_rate": 7.65302002592172e-10,
      "loss": 0.8277,
      "num_input_tokens_seen": 77729512,
      "step": 133980
    },
    {
      "epoch": 19.956061960083407,
      "grad_norm": 3.65625,
      "learning_rate": 7.400879850155695e-10,
      "loss": 1.0269,
      "num_input_tokens_seen": 77732296,
      "step": 133985
    },
    {
      "epoch": 19.956806672624367,
      "grad_norm": 4.90625,
      "learning_rate": 7.152963045264915e-10,
      "loss": 1.2763,
      "num_input_tokens_seen": 77735272,
      "step": 133990
    },
    {
      "epoch": 19.957551385165328,
      "grad_norm": 11.875,
      "learning_rate": 6.909269615440472e-10,
      "loss": 0.8682,
      "num_input_tokens_seen": 77738248,
      "step": 133995
    },
    {
      "epoch": 19.958296097706285,
      "grad_norm": 10.25,
      "learning_rate": 6.669799564817947e-10,
      "loss": 0.9691,
      "num_input_tokens_seen": 77741032,
      "step": 134000
    },
    {
      "epoch": 19.959040810247245,
      "grad_norm": 8.3125,
      "learning_rate": 6.434552897421897e-10,
      "loss": 1.4131,
      "num_input_tokens_seen": 77743944,
      "step": 134005
    },
    {
      "epoch": 19.959785522788202,
      "grad_norm": 12.3125,
      "learning_rate": 6.203529617221371e-10,
      "loss": 0.992,
      "num_input_tokens_seen": 77746728,
      "step": 134010
    },
    {
      "epoch": 19.960530235329163,
      "grad_norm": 14.875,
      "learning_rate": 5.976729728129904e-10,
      "loss": 1.4366,
      "num_input_tokens_seen": 77749544,
      "step": 134015
    },
    {
      "epoch": 19.961274947870123,
      "grad_norm": 8.4375,
      "learning_rate": 5.754153233977766e-10,
      "loss": 1.0424,
      "num_input_tokens_seen": 77752552,
      "step": 134020
    },
    {
      "epoch": 19.96201966041108,
      "grad_norm": 10.5625,
      "learning_rate": 5.535800138539715e-10,
      "loss": 1.1697,
      "num_input_tokens_seen": 77755240,
      "step": 134025
    },
    {
      "epoch": 19.96276437295204,
      "grad_norm": 13.1875,
      "learning_rate": 5.321670445479488e-10,
      "loss": 1.1104,
      "num_input_tokens_seen": 77758280,
      "step": 134030
    },
    {
      "epoch": 19.963509085493,
      "grad_norm": 9.6875,
      "learning_rate": 5.111764158433063e-10,
      "loss": 1.046,
      "num_input_tokens_seen": 77761192,
      "step": 134035
    },
    {
      "epoch": 19.96425379803396,
      "grad_norm": 9.3125,
      "learning_rate": 4.906081280953157e-10,
      "loss": 1.0187,
      "num_input_tokens_seen": 77764040,
      "step": 134040
    },
    {
      "epoch": 19.96499851057492,
      "grad_norm": 9.0,
      "learning_rate": 4.704621816481458e-10,
      "loss": 1.2579,
      "num_input_tokens_seen": 77766920,
      "step": 134045
    },
    {
      "epoch": 19.965743223115876,
      "grad_norm": 12.6875,
      "learning_rate": 4.5073857684319043e-10,
      "loss": 1.1122,
      "num_input_tokens_seen": 77770056,
      "step": 134050
    },
    {
      "epoch": 19.966487935656836,
      "grad_norm": 5.3125,
      "learning_rate": 4.3143731401629194e-10,
      "loss": 0.8484,
      "num_input_tokens_seen": 77772872,
      "step": 134055
    },
    {
      "epoch": 19.967232648197797,
      "grad_norm": 16.0,
      "learning_rate": 4.1255839349219063e-10,
      "loss": 1.3152,
      "num_input_tokens_seen": 77775464,
      "step": 134060
    },
    {
      "epoch": 19.967977360738754,
      "grad_norm": 7.46875,
      "learning_rate": 3.9410181559007553e-10,
      "loss": 0.9175,
      "num_input_tokens_seen": 77778248,
      "step": 134065
    },
    {
      "epoch": 19.968722073279714,
      "grad_norm": 15.25,
      "learning_rate": 3.760675806180336e-10,
      "loss": 0.9621,
      "num_input_tokens_seen": 77781064,
      "step": 134070
    },
    {
      "epoch": 19.969466785820675,
      "grad_norm": 10.25,
      "learning_rate": 3.5845568888692726e-10,
      "loss": 1.2478,
      "num_input_tokens_seen": 77783624,
      "step": 134075
    },
    {
      "epoch": 19.970211498361632,
      "grad_norm": 13.875,
      "learning_rate": 3.412661406881901e-10,
      "loss": 1.0506,
      "num_input_tokens_seen": 77786632,
      "step": 134080
    },
    {
      "epoch": 19.970956210902592,
      "grad_norm": 9.6875,
      "learning_rate": 3.244989363188067e-10,
      "loss": 0.9655,
      "num_input_tokens_seen": 77789416,
      "step": 134085
    },
    {
      "epoch": 19.97170092344355,
      "grad_norm": 11.5,
      "learning_rate": 3.0815407605633283e-10,
      "loss": 1.0778,
      "num_input_tokens_seen": 77792168,
      "step": 134090
    },
    {
      "epoch": 19.97244563598451,
      "grad_norm": 12.75,
      "learning_rate": 2.922315601783243e-10,
      "loss": 1.2756,
      "num_input_tokens_seen": 77794984,
      "step": 134095
    },
    {
      "epoch": 19.97319034852547,
      "grad_norm": 13.6875,
      "learning_rate": 2.7673138895678574e-10,
      "loss": 1.0633,
      "num_input_tokens_seen": 77798184,
      "step": 134100
    },
    {
      "epoch": 19.973935061066427,
      "grad_norm": 14.5,
      "learning_rate": 2.61653562649844e-10,
      "loss": 0.9808,
      "num_input_tokens_seen": 77801480,
      "step": 134105
    },
    {
      "epoch": 19.974679773607388,
      "grad_norm": 14.5625,
      "learning_rate": 2.469980815128503e-10,
      "loss": 1.5422,
      "num_input_tokens_seen": 77804328,
      "step": 134110
    },
    {
      "epoch": 19.97542448614835,
      "grad_norm": 9.5625,
      "learning_rate": 2.3276494579560492e-10,
      "loss": 1.001,
      "num_input_tokens_seen": 77807176,
      "step": 134115
    },
    {
      "epoch": 19.976169198689306,
      "grad_norm": 9.9375,
      "learning_rate": 2.1895415573680578e-10,
      "loss": 0.9748,
      "num_input_tokens_seen": 77810024,
      "step": 134120
    },
    {
      "epoch": 19.976913911230266,
      "grad_norm": 7.125,
      "learning_rate": 2.055657115695997e-10,
      "loss": 0.7791,
      "num_input_tokens_seen": 77812744,
      "step": 134125
    },
    {
      "epoch": 19.977658623771223,
      "grad_norm": 14.4375,
      "learning_rate": 1.925996135215824e-10,
      "loss": 1.2182,
      "num_input_tokens_seen": 77816104,
      "step": 134130
    },
    {
      "epoch": 19.978403336312184,
      "grad_norm": 18.375,
      "learning_rate": 1.8005586181202295e-10,
      "loss": 1.0819,
      "num_input_tokens_seen": 77818984,
      "step": 134135
    },
    {
      "epoch": 19.979148048853144,
      "grad_norm": 17.375,
      "learning_rate": 1.6793445664908813e-10,
      "loss": 1.0835,
      "num_input_tokens_seen": 77822408,
      "step": 134140
    },
    {
      "epoch": 19.9798927613941,
      "grad_norm": 9.375,
      "learning_rate": 1.5623539824372035e-10,
      "loss": 0.9533,
      "num_input_tokens_seen": 77825064,
      "step": 134145
    },
    {
      "epoch": 19.98063747393506,
      "grad_norm": 14.5625,
      "learning_rate": 1.4495868678743307e-10,
      "loss": 1.1994,
      "num_input_tokens_seen": 77827944,
      "step": 134150
    },
    {
      "epoch": 19.98138218647602,
      "grad_norm": 7.78125,
      "learning_rate": 1.3410432247173977e-10,
      "loss": 0.9808,
      "num_input_tokens_seen": 77830984,
      "step": 134155
    },
    {
      "epoch": 19.98212689901698,
      "grad_norm": 10.0625,
      "learning_rate": 1.2367230548537834e-10,
      "loss": 1.0304,
      "num_input_tokens_seen": 77833864,
      "step": 134160
    },
    {
      "epoch": 19.98287161155794,
      "grad_norm": 6.28125,
      "learning_rate": 1.1366263599765781e-10,
      "loss": 0.8022,
      "num_input_tokens_seen": 77836840,
      "step": 134165
    },
    {
      "epoch": 19.983616324098897,
      "grad_norm": 8.75,
      "learning_rate": 1.040753141834383e-10,
      "loss": 1.2953,
      "num_input_tokens_seen": 77839624,
      "step": 134170
    },
    {
      "epoch": 19.984361036639857,
      "grad_norm": 10.0625,
      "learning_rate": 9.491034020092659e-11,
      "loss": 0.8786,
      "num_input_tokens_seen": 77842536,
      "step": 134175
    },
    {
      "epoch": 19.985105749180818,
      "grad_norm": 4.59375,
      "learning_rate": 8.616771420555391e-11,
      "loss": 0.994,
      "num_input_tokens_seen": 77845608,
      "step": 134180
    },
    {
      "epoch": 19.985850461721775,
      "grad_norm": 8.875,
      "learning_rate": 7.784743634720038e-11,
      "loss": 1.1429,
      "num_input_tokens_seen": 77848296,
      "step": 134185
    },
    {
      "epoch": 19.986595174262735,
      "grad_norm": 9.6875,
      "learning_rate": 6.994950676186829e-11,
      "loss": 0.9081,
      "num_input_tokens_seen": 77851240,
      "step": 134190
    },
    {
      "epoch": 19.987339886803692,
      "grad_norm": 12.4375,
      "learning_rate": 6.247392558833553e-11,
      "loss": 0.9079,
      "num_input_tokens_seen": 77854152,
      "step": 134195
    },
    {
      "epoch": 19.988084599344653,
      "grad_norm": 11.125,
      "learning_rate": 5.542069295150221e-11,
      "loss": 1.0028,
      "num_input_tokens_seen": 77856872,
      "step": 134200
    },
    {
      "epoch": 19.988829311885613,
      "grad_norm": 8.1875,
      "learning_rate": 4.878980896794172e-11,
      "loss": 1.1039,
      "num_input_tokens_seen": 77859720,
      "step": 134205
    },
    {
      "epoch": 19.98957402442657,
      "grad_norm": 6.3125,
      "learning_rate": 4.258127375145193e-11,
      "loss": 0.9086,
      "num_input_tokens_seen": 77862824,
      "step": 134210
    },
    {
      "epoch": 19.99031873696753,
      "grad_norm": 7.1875,
      "learning_rate": 3.679508740472848e-11,
      "loss": 0.9499,
      "num_input_tokens_seen": 77865416,
      "step": 134215
    },
    {
      "epoch": 19.99106344950849,
      "grad_norm": 5.96875,
      "learning_rate": 3.143125003046699e-11,
      "loss": 0.8804,
      "num_input_tokens_seen": 77868232,
      "step": 134220
    },
    {
      "epoch": 19.99180816204945,
      "grad_norm": 14.125,
      "learning_rate": 2.648976171470974e-11,
      "loss": 1.3833,
      "num_input_tokens_seen": 77871208,
      "step": 134225
    },
    {
      "epoch": 19.99255287459041,
      "grad_norm": 7.40625,
      "learning_rate": 2.197062254349902e-11,
      "loss": 1.0161,
      "num_input_tokens_seen": 77874248,
      "step": 134230
    },
    {
      "epoch": 19.993297587131366,
      "grad_norm": 13.125,
      "learning_rate": 1.7873832591774885e-11,
      "loss": 1.1008,
      "num_input_tokens_seen": 77877256,
      "step": 134235
    },
    {
      "epoch": 19.994042299672326,
      "grad_norm": 11.5625,
      "learning_rate": 1.4199391928926276e-11,
      "loss": 1.083,
      "num_input_tokens_seen": 77880264,
      "step": 134240
    },
    {
      "epoch": 19.994787012213287,
      "grad_norm": 8.6875,
      "learning_rate": 1.0947300618791013e-11,
      "loss": 1.1765,
      "num_input_tokens_seen": 77882920,
      "step": 134245
    },
    {
      "epoch": 19.995531724754244,
      "grad_norm": 10.3125,
      "learning_rate": 8.11755871410469e-12,
      "loss": 0.6322,
      "num_input_tokens_seen": 77885832,
      "step": 134250
    },
    {
      "epoch": 19.996276437295204,
      "grad_norm": 23.25,
      "learning_rate": 5.710166262051786e-12,
      "loss": 1.0896,
      "num_input_tokens_seen": 77889288,
      "step": 134255
    },
    {
      "epoch": 19.997021149836165,
      "grad_norm": 7.59375,
      "learning_rate": 3.725123307041223e-12,
      "loss": 0.8047,
      "num_input_tokens_seen": 77891880,
      "step": 134260
    },
    {
      "epoch": 19.997765862377122,
      "grad_norm": 17.5,
      "learning_rate": 2.162429879604133e-12,
      "loss": 1.3143,
      "num_input_tokens_seen": 77894952,
      "step": 134265
    },
    {
      "epoch": 19.998510574918082,
      "grad_norm": 6.25,
      "learning_rate": 1.022086004720535e-12,
      "loss": 0.9782,
      "num_input_tokens_seen": 77897512,
      "step": 134270
    },
    {
      "epoch": 19.99925528745904,
      "grad_norm": 10.875,
      "learning_rate": 3.040917045948888e-13,
      "loss": 1.3613,
      "num_input_tokens_seen": 77900488,
      "step": 134275
    },
    {
      "epoch": 20.0,
      "grad_norm": 5.03125,
      "learning_rate": 8.446993104982426e-15,
      "loss": 0.822,
      "num_input_tokens_seen": 77902976,
      "step": 134280
    },
    {
      "epoch": 20.0,
      "eval_loss": 1.0852864980697632,
      "eval_runtime": 43.8654,
      "eval_samples_per_second": 68.026,
      "eval_steps_per_second": 17.007,
      "num_input_tokens_seen": 77902976,
      "step": 134280
    },
    {
      "epoch": 20.0,
      "num_input_tokens_seen": 77902976,
      "step": 134280,
      "total_flos": 3.5080602449577247e+18,
      "train_loss": 1.0952728209393985,
      "train_runtime": 26067.2795,
      "train_samples_per_second": 20.603,
      "train_steps_per_second": 5.151
    }
  ],
  "logging_steps": 5,
  "max_steps": 134280,
  "num_input_tokens_seen": 77902976,
  "num_train_epochs": 20,
  "save_steps": 6714,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3.5080602449577247e+18,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}